Late one Saturday evening in November 2013, a researcher at the high-security Influenza Research Institute in Madison, Wisconsin, did the one thing scientists working there feared most. Handling a needle loaded with a potentially deadly flu virus, the person accidentally pricked their finger, drawing blood.
Realising the danger, the scientist immediately sprayed their finger with disinfectant and ran it under water. They were then advised to squeeze the wound in an attempt to draw out any infected blood.
The researcher has not been identified, but their story is detailed in contemporary laboratory safety reports which have been seen by the Financial Times. They were told to put on new gloves, take a shower and isolate. Meanwhile their family was told to move to a hotel for a week so the scientist could quarantine alone at home.
The flu strain involved was not a regular seasonal virus. The contents of the syringe had been artificially created in the Wisconsin laboratory, by splicing together a mutated version of the H5N1 avian flu with a more regular human version. And it was not the first accident to occur in the laboratory. Just a week earlier, reports seen by the FT show a scientist had spilled liquid containing the H5N1 virus.
H5N1 is known to be incredibly dangerous: 60 per cent of humans who become infected with it, die. The only positive is that it is unable to spread easily between humans. Yet, the flu virus created in the Wisconsin laboratory replicated quickly enough to spread between ferrets via respiratory droplets in the air. If the same were true for humans, the research team warned, it could trigger a global pandemic.
Up to 2013, this type of experiment — where a pathogen is enhanced to increase transmissibility or its ability to cause disease — had rarely been conducted on such potentially dangerous viruses. Yoshihiro Kawaoka, the head of the Wisconsin laboratory, had only revealed publicly that he could do such complex “gain-of-function” work on viruses two years earlier, much to the alarm of some of his peers.
The justification given for the work is that by manipulating the genetic make-up of certain viruses and isolating individual characteristics, scientists can work out what makes them most deadly, and how to identify future threats. Many working in the field say that drug companies would have found it much harder to create vaccines and treatments against Covid-19, for example, without gain-of-function work on Sars viruses, which helped explain how they infect human cells. And they insist it can be done safely.
The University of Wisconsin-Madison, said: “Since the onset of the work in Dr Kawaoka’s lab — the Influenza Research Institute — the university has adopted, created and implemented systems and processes to help us meet the highest standards of biosafety and biosecurity.”
But letters between the university and the National Institutes of Health, which funded the programme and was overseeing it, show government officials were highly concerned by safety protocols following the accidents at the Kawaoka laboratory.
In the letters, released to the FT under the Freedom of Information Act, NIH officials identified several problems with the laboratory’s practices, including the use of a needle in the first place, and allowing the researcher to quarantine at home.
There were no infections caused as a result of the incidents at Wisconsin, and both researchers were fine. But a small number of politicians and intelligence officials are wondering whether such an accident on the other side of the world might have triggered the Covid-19 pandemic.
While most scientists believe the virus first infected humans via animals, some — including one unnamed US intelligence agency — now believe it is more likely that the pandemic originated with exactly this kind of research being carried out at the Wuhan Institute of Virology, a world leader in gain-of-function work on coronaviruses.
Between 2015 and 2020, the Wuhan lab was given around $600,000 of US taxpayer money via a third-party organisation called EcoHealth Alliance, run by the British scientist Peter Daszak. The federal government has spent far more on such work at home.
An FT analysis of publicly available figures suggests the US government has spent over $30m in the past 15 years on domestic research which could plausibly be classified by the US government as gain-of-function research on potential pandemic-causing pathogens. The dispute over what happened at Wuhan means this US-based work is now under greater scrutiny than ever before.
“I believe this is the most dangerous scientific field in the world,” says Richard Ebright, professor of chemical biology at Rutgers University. “It does not matter whether Covid-19 actually leaked from Wuhan. Just knowing it could have should be enough for us to change our approach.”
It is a view echoed by others in the field. “It is only a matter of time before one of these pathogens escapes somewhere,” says Alina Chan, a molecular biologist at MIT’s Broad Institute. “Doing this research in a densely-packed city, as some researchers do, is like throwing a match into a forest in the middle of a drought.”
Kawaoka was one of the first scientists to show that the genetic make-up of avian flu could be manipulated to spread among mammals. A decade ago, he and a team of researchers managed to mutate the H5N1 virus to make it more like H1N1 — more commonly known as swine flu — which spreads so quickly among humans it caused a pandemic in 2009.
Their rationale was “to prepare for potential pandemics caused by influenza viruses”. And the work Kawaoka and his team conducted was heavily funded by the US government. Starting in 2006, his work on bird flu received funding of around $500,000 annually, and from 2009 an additional $600,000 a year was made available for research into why the 1918 flu virus spread so quickly.
Both grants came from the National Institute for Allergy and Infectious Diseases, led by Anthony Fauci, the virologist who has come to international prominence during the Covid pandemic and is now US president Joe Biden’s chief medical adviser.
Fauci declined to speak to the FT for this article, though in an interview earlier this year he defended the idea of genetically manipulating viruses to observe their effects on human cells. “You need to adapt the virus to be able to use it as a tool to ask questions,” he said.
Kawaoka’s government backing was not sufficient to protect him from a heavy backlash when he sought to publish the results of his gain-of-function work. When he submitted his results to the journal Nature in 2011, his peers were sufficiently alarmed that the National Science Advisory Board for Biosecurity — a government-convened panel of experts — recommended delaying publication until he redacted parts of his method, in case anyone tried to copy it.
The NSABB was even more worried about similar work being carried out in the Netherlands by Ron Fouchier, a scientist at the Erasmus Institute in Rotterdam. Fouchier had also been breeding more transmissible flu strains using ferrets, and reportedly joked to a conference in 2011 that his work was so dangerous it was “really, really stupid”.
Both scientists eventually published their papers with redactions in 2012, and the government funding continued. In 2013, Kawaoka’s team was given a further grant to manipulate the genes of both influenza and Ebola viruses to see whether any previously uncharacterised genes helped contribute to their spread. NIAID funded that work by between $300,000 and $600,000 a year until 2017.
Ralph Baric, who worked closely with Kawaoka and ran his own laboratory at the University of North Carolina, Chapel Hill, was working on a similar project. Baric had managed in 2005 to genetically engineer mice so their immune systems closely mimicked those of humans. He used the mice to test genetically-manipulated coronaviruses in an attempt to work out which genes helped the virus replicate. From 2013 until 2017, NIAID gave Baric a total of $2.3m to help fund this work.
Accidents in the lab
In 2014 the Obama White House began asking questions about exactly what kind of research the administration was helping to fund after a string of accidents at high-security laboratories in the US.
In June, the US Centers for Disease Control and Prevention announced 75 staff might have been inadvertently exposed to live anthrax bacteria. A month later, employees at NIH stumbled across a set of long-forgotten vials, only to find that two contained live samples of smallpox. One month later the CDC admitted it had sent samples of regular flu virus to an external lab which had accidentally been contaminated with the H5N1 strain.
The accidents did not involve gain-of-function research. But they were worrying enough to persuade officials to impose a moratorium on government funding for any such research on Sars, Mers or flu viruses — considered by many experts to be the riskiest science around.
The moratorium threatened to break the close relationship between US health agencies and this small subset of the scientific community. But in the end the money continued to flow. Under the terms of the moratorium, any project that had already been approved could continue to access public funds — including the work being carried out by both Baric and Kawaoka.
Not all of these experiments would necessarily have qualified as gain-of-function work under the terms of the US moratorium. They all involved manipulating flu, Sars or Mers viruses, often making those viruses more dangerous in the process. But to count as gain-of-function work, the scientists involved had to “reasonably anticipate” that would be the outcome, a woolly definition which helped NIH officials grant exemptions to 13 out of the 21 studies originally put on pause.
David Relman, a professor of medicine at Stanford University, calls this debate “semantic”, pointing out that many of these experiments fall into a simpler category, which he terms “exceptionally risky research”.
The relationship between NIH and some of these scientists goes beyond just gain-of-function work. Public funding data analysed by the FT show that across his career teams led by Kawaoka have received more than $63m in total government grants. For Baric, the figure is more than $105m.
Critics believe the sums involved indicate a close relationship between a handful of respiratory virus researchers and the US government.
“These people are a group that could more accurately be described as a tightly knit club of establishment insiders with a powerful grip on their research community and its funders,” says Peter Hale, founder of the Foundation for Vaccine Research in the US, a pressure group. “They are terrific at writing grant proposals but less so at justifying their excessively risky experiments. And whenever trouble appears, they circle the wagons.”
Kawaoka rejects this claim. “This is insulting to the scientific community, including the independent funding agencies and independent scientists that evaluate research proposals, and the researchers who work hard to develop competitive proposals to compete for limited funding.”
‘Batwoman’ and the road to Wuhan
In 2014 Baric, nicknamed the “coronavirus hunter”, met Shi Zhengli — a scientist whose willingness to hunt down coronavirus strains in bat caves had earned her the moniker “Batwoman”.
Baric provided his mice for Shi to experiment on in Wuhan and in 2015, the pair published a paper describing how they had spliced the Sars virus together with another coronavirus to create a “chimeric” virus which could replicate quickly in human cells. The findings were so alarming that the authors added a cautionary note to their paper: “Scientific review panels may deem similar studies building chimeric viruses based on circulating strains too risky to pursue.”
Also in 2015, Baric presented his work at a Chinese Academy of Sciences conference, in which researchers working for the People’s Liberation Army were also in attendance — something he insists was par for the course.
A year later, his team published a paper warning that one of the virus strains they had been working on was “poised for human emergence”. And later in 2016, Shi and Daszak helped co-author a paper revealing in the footnotes that some of their joint work was being carried out at biosafety level two — roughly equivalent to that of a dentist’s surgery.
“This was unnecessarily risky research,” says Relman. “They may not have known what properties the chimeric viruses they were creating would have, but they did know they were playing with the WIV1 bat coronavirus, which is already pretty well adapted to human cells.”
EcoHealth Alliance, which funded the Shi research has previously said: “Biosafety regulations are determined country-by-country and while EcoHealth Alliance would welcome global standardised rules dictating biosafety regulations that all countries would agree to, those do not currently exist.”
In December 2017, the Trump administration ended the moratorium on government funding of gain-of-function research — a decision which those involved in the process say was guided by Fauci and Francis Collins, the outgoing head of the NIH.
Under the post-moratorium guidelines, which the two men helped write, a committee at the US health department would review bids from scientists for government money for gain-of-function work. But the committee, whose membership is secret and whose minutes are not published, would not be able to block any contract awards, only advise the NIH on them.
These guidelines have attracted criticism from many in the scientific world, including those who continue to advise the US government.
Marc Lipsitch is director of science at the CDC’s Center for Forecasting and Outbreak Analytics. But before he recently took up his government job he told the FT of his concerns over the gain-of-function review process. “We have a group of people allegedly addressing a series of criteria. But who they are, how the criteria are assessed and what the substance of their deliberations is, are all completely unknowable to the scientific community and the general public.”
Another person with concerns is Rocco Casagrande, a scientific investigator who the Obama administration commissioned to look at the risks and benefits of this kind of scientific work. His report, which was meant to inform the guidelines that followed, warned that a strain of flu to which the population had little immunity and caused more than 5 per cent of sufferers to die “would pose more of a risk of a global pandemic than any wild type strain [previously] identified”.
“Some of this research is important because you need to understand how pandemics evolve,” says Casagrande. “The problem occurs when you are creating strains that would not evolve in nature, or when you create them with huge selective pressure, like passing them through several generations of mammals [as Kawaoka did]. Those are arguably not worth the risk.”
The NIH has defended the guidelines which it said were arrived at through a rigorous process that included “external risk benefit assessment and ethics analyses”. It added: “NIH will continue to work across the federal government to regularly reassess this policy to ensure it upholds the highest safety standards for research involving [gain-of-function work on potential pandemic-causing pathogens] while pursuing the science needed to ensure the US is prepared for the next pandemic.”
Looking for answers
US officials still cannot say for sure what caused Covid-19 to jump to humans. In a report declassified in October, intelligence agencies could not agree on its origin. Four agencies said they think it passed from bats through an as-yet-unidentified mammal into people and three said they were unable to decide.
One agency, however, was more categorical. The unnamed agency said it believed the virus escaped from the Wuhan Institute of Virology, pointing to the lax conditions under which some work was happening. “It is plausible that researchers may have unwittingly exposed themselves to the virus without sequencing it during experiments or sampling activities, possibly resulting in asymptomatic or mild infection,” the report said.
NIH funding of the Wuhan project has now stopped, having been put on hold by the Trump administration. The longer-lasting impact of the controversy might instead be on the millions of dollars the institute continues to spend on gain-of-function work in the US.
The Government Accountability Office — which audits spending — is preparing a report on which gain-of-function projects the US has been supporting and where. Several members of Congress are now calling for a new moratorium on all public funding for such research. “We are dealing with something akin to nuclear warheads here,” says Roger Marshall, the Republican senator for Kansas.
The University of North Carolina said the Baric lab identified remdesivir and molnupinivir as two different broad spectrum anti-coronavirus drugs. But some critics argue that the most damning fact about gain-of-function work on viruses which can cause pandemics is not what it has done, but what it has not.
“The justification was that in the event of a pandemic, their cutting-edge work was supposed to help us come up with a vaccine,” says Hale. “But when a pandemic hit, it was the people working on mRNA and other platforms who saved us. All they needed was the sequence.
“Gain-of-function work carries all the risks,” adds Hale, “but apparently few of the benefits.”