Spotlight on Biosecurity: How ‘gain-of-function’ slipped through the net

As the global Covid-19 pandemic swept across the globe in 2020 devastating lives and national economies, a small community of scientists in a fringe corner of virology found themselves pausing for breath.

It was a moment that could, in theory, vindicate a controversial scientific research method they had been enthusiastically developing. It was known as “gain-of-function”.

As one critical scientist told me, the method might as well have been described as a form of accelerated evolution, albeit one purposefully directed at juicing viruses to their maximum lethality and contagiousness.

Those favouring the method justified the inherent risk on the basis that if scientists knew how viruses were likely to mutate or what damage they could inflict on humans, they could pre-develop potential cures and vaccines to counter unexpected outbreaks and thus save lives. They claimed the motivation for the research was noble.

But tinkering with nature made many other scientists uneasy. They worried an accidental lab leak of a new synthetic pathogen developed in this way could destabilise the planet.

The North Carolina Connection 

Among those knee-deep in the research was Ralph Baric, a memorably moustachioed microbiologist in his late 60s at the University of North Carolina in Chapel Hill. Dr. Baric had gained a reputation as a tough talking coronavirus specialist whose interest in the field predated the original SARS outbreak of 2002. Back then nobody considered the spike-based virus to be of pandemic potential.

But Baric had batted on regardless.

By the time SARS – a coronavirus – broke out, Baric’s work began to look inspired. A rising reputation quickly catapulted the UNC resident into the top tier of virologists working on pandemic mitigations. In time Baric found himself collaborating with scientists at the Wuhan Institute of Virology in China, including Institute director Shi Zhengli – known locally as the “batwoman” – as well as, more mundanely, the School of Life Sciences in Kingston, Surrey.

By 2015, the group’s collective work had proven that bat coronaviruses, similar to those behind the Covid-19 outbreak, could be mutated to become highly infectious to humans. Much of this work – as has now come to light – was being done in biosafety level 3 (BSL3) labs, one level below the highest security biosafety level 4 categorisation.

It is still disputed whether Baric’s work, which qualified for US government funding at the time, counted as gain-of-function activity.

Even so, those who considered Baric an excellent scientist had grown concerned that the experimental methods he favoured in the study could one day lead to a global pandemic.

It was this context that raised concerns among a select group of scientists In the early days of the Covid-19 outbreak in Wuhan in January 2020, that there may be more to the source of the outbreak than a zoonosis spillover from a natural source.

The Wuhan Institute’s main BSL4 lab, where much of Baric’s collaborative coronavirus work had been carried out, was located about 10km from the Wuhan wet market linked to the first known cases of Covid. Given the proximity, the possibility of an accidental lab leak seemed plausible.

Adding credence to the theory was a private email written on January 31, 2020, by Kristian Anderson, a virologist at Scripps Research Institute to Dr Anthony Fauci, the director of the National Institute of Allergy and Infectious Diseases and coordinator of America’s pandemic response. Since made public it read:

“The unusual features of the virus make up a really small part of the genome (<0.1%) so one has to look really closely at all the sequences to see that some of the features (potentially) look engineered.”

But Andersen’s stance on the issue was not to last.

Before the outbreak was officially declared a pandemic on March 11, a group of 27 well-connected scientists, including the Wellcome Trust’s Jeremy Farrar, clubbed together on February 19 to pen a letter decrying Covid’s connections to the Wuhan Institute.

Also included as a signatory was Peter Daszak, president of the EcoHealth Alliance, an environmental health non-profit that draws funding from US Government funding agencies and non-governmental sources and which has funded both the North Carolina and Wuhan-based work.

Thanks to the Lancet letter, attention soon turned away from gain-of-function research over to the likely zoonotic origins of Covid for most of 2020.

Andersen’s own thinking fell in line with that of the group shortly after. He has published numerous works in support of the zoonosis origin theory since then, most recently in February 2022.

The North Carolina News & Observer, meanwhile, dubbed Ralph Baric “North Carolinian of the Month” for his significant contribution to viral research, arguing that his UNC laboratory had laid the groundwork for Covid-19 treatments, including the drug Remdesivir.

It took another year before concerns over conflicts of interest — including that the group may have purposefully colluded to discourage researchers and journalists from probing their connection to the Wuhan Institute of Virology, gain-of-function research and a potential connection to the pandemic — began to be officially recognised.

On June 26 2021, the Lancet finally published an update to the original letter noting that “some readers have questioned the validity of this disclosure, particularly as it relates to one of the authors, Peter Daszak.”

Baric himself admitted in an Italian documentary in November 2020 that it was standard practice for researchers to add genetic signatures to the sequences of viruses they were engineering, and that it was possible to “engineer a virus without leaving any trace.”

Andersen, Daszak, and Farrar did not responded to requests for comment.

Ralph Baric’s representatives, meanwhile, refuted allegations that any of his post 2014 research qualified as gain-of-function work.

From Cotton Buds to Lethal Pathogens

When Ron Fouchier, an Erasmus scientist in the Netherlands, first used gain-of-function techniques to make the H5N1 flu more infectious and transmissible to humans in November 2011, the findings set off alarms across the global virology community.

One of the key concerns was that Fouchier had used an extremely simple and replicable process to create the highly lethal pathogen. He had taken flu samples and used them to infect ferrets many times over, cherry-picking specimens from the sickest ferrets to infect the healthy ones next in line. It was the simplicity of the process that had spooked many of his colleagues, who felt the publication of the work itself posed risks. Others, meanwhile, believed there was little benefit to be had from the work at all.

“The promise of gain-of-function research was that it would help us predict the next pandemic strain of flu,” said Simon Wain-Hobson, a recently retired virologist from the Pasteur Institute when interviewed by me in June 2021. This, he added, had turned out to be “science fiction”.

Wain-Hobson went on to become one of the world’s most vocal critics of the gain-of-function research method, publishing a number of critiques in academic journals.

He told me he had confronted Fouchier directly at a scientific meeting on the subject in Hanover in 2014, asking Fouchier why he thought the process would deliver demonstrable gains? Fouchier’s reply was that it might take 30 years. Far from good enough in Wain-Hobson’s opinion.

Neither Fouchier nor the Erasmus Institute replied to queries for comment.

A Half-Hearted Moratorium?

In time Wain-Hobson and a handful of other peers, among them Richard Ebright, a molecular biologist and biodefense expert at Rutgers University in Piscataway, felt concerned enough to advocate at the highest political levels for a moratorium.

In 2014, just as Baric and his team veered ever closer to their breakthrough, their efforts seemed to pay off. The Obama administration agreed on the back of mounting scrutiny to impose a moratorium on government funding of gain-of-function research from October 17 onwards.

In theory, the moratorium would halt all related research immediately. In practice, work which had already qualified for funding continued to be waived through by the NIH and NIAID or transferred to areas not covered by the suspension.

Just four days after the moratorium was finalised, Sherrie Settle, however, the director of proposal management at the University of North Carolina at Chapel Hill, received a letter from Chernay Mason, the Grants Management Specialist at NIAID/NIH/DHHS. The letter referenced Ralph Baric’s work on bat viruses, highlighting that his grant may include gain-of-function research that is subject to the recently-announced US government funding pause.

But it also caveated that: “As your grant is currently funded, this pause is voluntary”.

Greenlighted by the Mason letter, Baric’s gain-of-function work with Shi Zhengli in Wuhan continued and in effect, was outsourced to China. The research also drew further funding and support from Ecohealth Alliance, which itself was being funded by grants from the US Defence Threat Reduction Agency. It was this arrangement that eventually led to the soon-to-be highly scrutinised paper: “A Sars-like cluster of circulating bat coronaviruses shows potential for human emergence.”

While I was at the FT, a spokesman for UNC told me that a committee at the NIH would have been responsible for the final decision on what constituted gain-of-function work, adding that the NIH findings indicated the work conducted for the 2015 paper did not qualify as gain-of-function research.

According to my former colleague Kiran Stacey, who was responsible for reaching out to NIH, the committee process and its membership were confidential and remain so today.

In highlighting that Baric’s team had managed to create a chimeric virus that had been adapted to grow in mice and to mimic human disease, the 2015 paper soon reopened all the old sores over whether bioengineering virus variants with possible pandemic potential was really worth the risks.

But even as scientists debated whether Baric’s work had flown too close to the sun, Shi Zhengli and her team continued on the work in China. In November 2017 – the same year that the Institute had achieved its BSL4 certification – the team published findings based on a novel coronavirus they had created that stopped just short of being transmissible to humans.

The following month, the NIH lifted the moratorium on gain-of-function research, allowing government funds to flow once again into research focused on manipulating influenza, SARS, and MERS in a bid to figure out how it might become more infectious or deadly.

Human Error is the Norm Not the Exception in Biosafety Labs

Even in the most professionally operated and trustworthy hands, laboratory accidents are common occurrences. Ralph Baric’s own lab, like most others, was not immune to accidents. The latest of these occurred in April 2020, and involved a mouse infected with a mouse-adapted SARS-Cov-2 strain biting a laboratory researcher.

According to UNC’s spokesman no infection resulted from that potential exposure because the laboratory scientist was wearing double gloves and the mouse bite did not penetrate beyond the inside glove.

But the incident is a good example of the risks facing technicians and scientists handling highly infectious pathogens in secure facilities across the world.

“These incidents happen even in very well trained, very well equipped, and very oversighted labs, so it does call for even more safety precautions when it comes to doing this very dangerous kind of research,” said Filippa Lentzos, a social scientist researching biological agent threats at King’s College London.

When SARS re-emerged in China in 2004, that outbreak was attributed to the exposure of researchers to the virus while conducting research in a lab. The 2007 outbreak of foot and mouth disease in Britain was connected to a malfunction at a laboratory facility in Pirbright in Wales. These are just some of the many multiples of lab-related outbreaks that have occurred in recent decades.

In a bid to quantify the risks, Lynn Klotz, a senior science fellow at the Center for Arms Control and Non-Proliferation, published an influential piece in the Bulletin of Atomic Scientists in 2019 assessing the scale of human error in BSL3 biocontainment labs. He noted that during the 2009-2015 time period, the Federal Select Agent Program received 749 incident reports of lab related incidents, 79.3 per cent of which involved human error.

Meanwhile, research conducted for the 2017 meeting of the Biological Weapons Convention (BWC) estimated a 20 per cent probability for a release of an airborne transmissible pathogen of Avian flu from at least one of the 10 labs researching it over a 10-year period.

This is the risk critics of gain-of-function work are most worried about: lab accidents happen all the time, even in the highest security facilities.

From Gain-of-Function to Bioweapons Risk

Academics worry about lab leaks and accidental spillovers. But government security officials worry about rogue actors using published academic work to develop biological weapons off their own back for nefarious purposes.

This paranoia exists for good reason.

The publication of Fouchier’s findings meant that anyone with access to the internet could, in theory, figure out how to make a lethal pathogen with bioweapon potential with nothing more than a makeshift lab, some cotton buds, and a shipment of ferrets.

It was the moment, according to RP Eddy (whose consulting group Ergo has been providing pandemic-related intelligence to the Biden administration and other government agencies), that the scientific and public policy community realised bioweapons development risk had moved beyond nation state capability.

“The garage lab is now a much more powerful place than it used to be because of the ease largely of CRISPR gene-editing technology,” Eddy told me while I was still at the FT. “Next thing you know, you’ve got an airborne H5N1 human andromeda strain, in a BSL3 lab which is like a closet with an air pump.”

Right now, Eddy added, there really isn’t an agreed to and followed set of standards for how BSL3 and BSL4 labs should secure themselves.

Despite the risks, in the years following Fouchier’s experiment, gain-of-function research drew support from the highest levels of US academia and government. With funding secured from bodies like the NIH, the NIAID, and the DoD, the research method, mostly conducted in lesser security BSL2 and BSL3 labs, proliferated in labs across America and Europe.

As gain-of-function research gained ground in academic settings, and findings were increasingly publicised, fears grew the technology’s dual-use nature — the fact that it could aid both the creation of bioweapons or prophylactics — could be easily exploited by military elements of rogue states, or even by terrorist networks.

“By definition, gain-of-function was classed as biological warfare in the textbooks from the 1980s,” said Paul Dabrowa, a biohacker, who has advised the Australian government on bio-weapon capability. “It was not supposed to happen. But somehow, from the mid noughties onwards, the idea that it could be researched in a civilian capacity with no fallout gained ground.”

Institutions like the NIH and NHAID understood the dangers and were charged with regulating any dual-use activity. Even so, it was not unusual for labs across the world to be closely associated with military personnel or funding, since any offensive development was strictly prohibited under the 1972 Biological Weapons Convention.

North Carolina Chapel Hill’s Division of Infectious Diseases faculty were among many in the country to benefit from funding provided by the Defense Advanced Research Projects Agency (DARPA) for programmes related to biodefense and emerging diseases.

“The vast majority of BL4 activity in the United States, at least historically, was done by the US military. And it was peaceful,” said Eddy.

“If you go to BL4 labs of most nations, there’s often a military liaison or component to the work. If you went in 1995, a plurality, if not majority, of BL4 labs would have a US military relationship.”

Outsourcing to China

Despite the challenges with the Biological Weapon Convention’s enforceability, the impetus to collaborate on even sensitive gain-of-function work on a cross-border level grew ever stronger in the mid-to-late 2000s as globalisation forces took hold. This applied to research potentially considered dual-use too.

Chinese scientists had by now become world experts in coronaviruses in the wake of the SARS 2002 outbreak. Not collaborating with them would have been considered detrimental to any serious work on the topic. A culture of active cross-border cooperation and mutual gatherings evolved.

This is how Baric came to present his gain-of-function work at a Chinese Academy of Sciences conference in 2015, in which researchers working for the People’s Liberation Army just so happened to be in attendance.

At the time, Baric insists, none of this was beyond the norm. “The National Academy of Sciences has held meetings with the Chinese Academy of Sciences to discuss research on emerging pathogens, biosafety and biosecurity and a dozen US scientists attend such meetings,” he emailed me via a representative while I was at the Financial Times. “To my knowledge, the national academies of both nations have met regularly ever since President Nixon normalised relations with China.”

According to Richard Muller, physicist and emeritus professor of physics at the University of California, Berkeley, who has been consulting on the investigation being led by the State Department into Covid’s origins, the benefits of multi-jurisdiction cooperation and multi-authored research were always going to be hugely alluring.

Academia, like so many other industries, stood to benefit from both internationalising and outsourcing its activities.

“Work can be done in China, it’s less expensive than if it’s done here, there’s not as much oversight there as there is here. And the Chinese scientists are very good,” Muller told me. “They are people you want to collaborate with, they are leaders in the field. And just to avoid collaborating with leaders in the field puts you at a real disadvantage.”

In his mind, the West has become dependent on China for everything from certain medicines to facemasks, to rare earth metals. “This is a new kind of mercantilism, and it needs a name,” he said.

But outsourcing academic work to foreign jurisdictions also opened the door to similar constraints that affect conventional supply chains: an inability to easily monitor or ensure that operations on the ground were really keeping up with Western best practices.

In the case of potentially sensitive scientific work in China there was also the added risk that the Chinese Communist Party could make questionable demands on civilian researchers in the name of national security.

Baric’s collaborators at the Wuhan Institute of Virology seemed acutely aware of the risks. Some tried to flag them to the international community very early on.

On December 9 2011, when the NIH entertained a panel on the intersection of science and security with the intention of giving attendees a greater understanding of dual-use research, Dr Yuan Zhiming, a microbiologist at the Wuhan Institute and a close collaborator of Shi Zhengli, warned that “in China there is no regulation on the identification of dual-use research, and there’s no regulation on the classification of research and classification of information”.

The video below had only been watched 200 or so times when it was first brought to my attention in May, 2021.

Is the Virus the Whistleblower?

The race to isolate the virus responsible for a spate of mysteriously deadly pneumonia cases which had started to grip Wuhan in December began with a phone call to Shi Zhengli, who was in Shanghai at the time of the outbreak. In the years preceding, Zhengli had gained a reputation as China’s foremost specialist on bat coronaviruses, to the extent that she became known as the “Batwoman” on account of her frequent escapades into bat caves looking for infected specimens.

On December 30, 2019, the Wuhan Center for Disease Control, positioned just 300 metres from the Wuhan wet market, had detected what they believed could be a virus of pandemic potential and wanted Zhengli to investigate.

Zhengli later told Scientific American Magazine she thought the CDC had got it wrong because she had never expected this kind of thing to happen somewhere like Wuhan. But she also admitted to wondering that if coronaviruses were proven to be the driver of the outbreak “could they have come from our lab?”

Having received the first samples on December 30, it took Zhengli and her team 10 days to isolate the SARS-CoV-2 sequence before sharing the findings with the US CDC and the WHO on January 10, 2020. The race to identify the possible origins of SARS-CoV-2 began in earnest soon after. Key to the identification process would be analysis of the databases of the Wuhan Institute of Virology, administered by Zhengli, to see if any of the viruses held there shared any similar features.

But these, it was later discovered, had been mysteriously deleted in September 2019, just weeks before the first cases of Covid appeared in the community. Why they had been deleted remains a mystery to this day.

When asked about the missing files by the BBC in December 2020, Zhengli said access to them had been restricted “to prevent cyber security attacks”.

But Zhengli also attested that once the full viral genome had been sequenced, she had crosschecked the sequence against her own lab’s records and concluded none of them matched.

Under the pre-existing collaborative framework, however, many samples held by the Wuhan Institute had already been shared with foreign researchers.

It was thanks to these broader datasets that on January 29, 2020, Greek researchers led by Sotiroios Tsiodras identified that RaTG13 was the virus that most closely resembled SARS-CoV-2, based on information that had been previously shared by the Wuhan Institute internationally. RaTG13 had been discovered in a bat cave in Yunnan in 2012 following the illnesses of six miners, all of whom had suffered a severe pneumonia, with four dying thereafter.

But RaTG13 — at a 96.3 per cent match — was still considered too different to represent a credible source of Covid-19. According to Fang Chi-tai, a professor at Taiwan University’s College of Public Health, it would have needed at least four amino acid mutations to have morphed into Covid, something that was unlikely to happen without laboratory intervention.

On February 3, 2020, Zhengli and her team confirmed the findings, publishing a paper that noted “RaTG13 is the closest relative of 2019-nCoV and they form a distinct lineage from other SARSr-CoVs.”

For Richard Muller, however, the lack of transparency surrounding the Chinese response in the early days of the outbreak was indicative of a potential cover-up. He told me he believes Covid 19 may not just have leaked from a lab but have been purposefully engineered to be more infectious and transmissible to humans using gain-of-function methods. His assertion is supported by an international team of scientists and data analysts led by Dr Steven Quay, who have made a case that the genetic features of the virus indicate it was likely man-made.

“If we tried to do it here someone would blow the whistle and there would be a congressional investigation and a big scandal and newspapers would write it up. In China there are no whistleblowers,” Muller said. “I like to call the virus the whistleblower which escaped China and carried with it a message saying, “look at how I was made.”

The lack of transparency is also concerning to Andy Weber, the former Assistant Secretary of Defense for Nuclear, Chemical, and Biological Defense Programs under the Obama administration. In the context of the behavioural protocols usually demanded by the Biological Weapon Convention, it is hard to understand the motivation for it. “Their [the Chinese] natural modus operandi is to not be transparent,” Weber told me in 2021. “Do they just not want to be blamed? Or are they really covering up something that they know?”

To ascertain the facts, Weber said you’d have to have the scientists themselves open up and provide their laboratory records and information on the earliest cases of Covid. But the more time passes, the harder that becomes and the stranger it is to justify not sharing the information in the databases.

“The genetic sequencing databases that the laboratory had which they closed access to…there’s a lot of information in those that might help determine if they had a virus that matches the SARS-CoV2 genetic sequence,” Weber said.

A WHO fact-finding trip, led in part by Peter Daszak, occurred in January 2021, but much like Matthew Meselson fruitless trip to Russia in 1986 (as recounted in part one of this two part series), it was largely criticised for failing to ascertain these facts and for being subject to conflicts of interest. Many have called for another visit with much broader access. But some, like Muller, suspect the Chinese will never agree to more inspections if a secret biological weapons programme is really being conducted.

“The gain-of-function used in Wuhan was beyond the gain-of-function you would use to look for upcoming diseases,” Muller said. “The only thing that’s possible at this point is to have open and transparent inspections. And you have to visit all parts of the laboratory,” he said.

In an unexpected move on May 21, 2021, Ralph Baric himself — along with 17 other top scientists — argued in a letter in the Lancet that more investigation was still needed to determine the origin of the pandemic and that a “proper investigation should be transparent, objective, data-driven, inclusive of broad expertise, subject too independent oversight, and responsibly managed to minimise the impact of conflicts of interest.”

When asked what had motivated Baric to sign the letter, a spokesman from UNC representing coronavirus specialist highlighted the inadequacies of the January WHO inspection, and that more investigation and transparency are necessary. He said, “a rigorous investigation would have reviewed the biosafety level under which bat coronavirus research was conducted at WIV,” adding that “it would have included detailed information on the training procedures with records, the safety procedures with records and strategies that were in place to prevent inadvertent or accidental escape.”

Nonetheless, the professional opinion of Dr. Baric on the likely origins of Covid remained the same, his spokesman said. “The genetic structure of the Covid-19 virus points to the virus originating in natural wildlife populations, most likely bats, that passed from animals to humans,” and that “the natural origin hypothesis is consistent from the most likely scenario presented in the WHO report”.

On May 26, 2021, US President Joe Biden ordered US intelligence agencies to investigate the possibility of a Wuhan lab-leak related origin to the pandemic as well as a zoonotic one.

“That is the entirety of the US intelligence community giving one shared estimate,” said RP Eddy, noting that this is very different from the State-Department led investigation that has been run to date by David Asher.

Its conclusion, released in August 2021, however, failed to deliver a definitive verdict.

A one-page summary noted the community remained split on the issue, with one agency believing it emerged through a lab leak, four believing it came from animals and three others unable to decide.

“China’s co-operation most likely would be needed to reach a conclusive assessment of the origins of Covid-19,” the summary said. “Beijing, however, continues to hinder the global investigation, resist sharing information and blame other countries, including the United States.”

An imperfect control mechanism

The report’s findings are likely to have influenced the broader international community’s resistance to invoking the BWC agreement, which China signed in 1984, when asking for further transparency on the matter.

“What is legitimate is drawing attention to Chinese military involvement in the Wuhan Institute of Virology, both in terms of leadership, in terms of co-authorship on publications, in terms of the funding going into all those things, which is a lot of stuff we don’t know about, and the Chinese are not being very open or transparent about,” said Filippa Lentzos.

“There are very legitimate questions around the dual-use of what they’re doing in this facility, but again, it’s not implying they’re necessarily doing anything in contravention of the BWC.”

State signatories of the BWC are supposed to meet every five years under the auspices of the UN to review the text of the convention. It is presumed, under that framework, additional understandings and agreements can be set out and published to ensure continuous evolution of the agreement.

Ever since the Americans pulled out of a collective agreement to update terms to include inspections in 2001, a secondary protocol for annual meetings, known as intersessional cycles, was put in place to help build confidence. These include a meeting for experts, such as biological scientists, and a separate political meeting or States Parties.

Because of the pandemic, however, the five-year review that was due to take place in 2021 was delayed until 2022, while 2020’s intersessional sessions was postponed until August 30 2021. According to Lentzos, this is unfortunate because the states could have used the pandemic to galvanise their work and initiate new transparency activities.

As it stands, only Kazakhstan submitted proposals at the 2021 meeting to help further transparency, enforcement and confidence in the BWC and which referenced the Covid origin uncertainties.

The Kazakh proposals specifically called for a new agency, the “International Agency for Biological Safety (IABS)”, to be created to address five existing gaps in the BWC. These were worded diplomatically as constituting the following:

Signatories nonetheless remain cautious about joining the call for further transparency from China due to fears their own biodefense activities might come under the spotlight or be weaponised in disinformation campaigns against them too.

According to Lentzos, one does not have to look far to see that British and American labs are doing a lot of biological defense work themselves. “American labs are doing, I think, way too much biodefense work that could easily be viewed as crossing the line towards more offensive kinds of applications with their work,” she said.

In an echo of the disinformation campaigns that plagued the aftermath of the Russian visits in the 1990s to public and private sector US labs, Chinese media have already been touting a counterconspiracy theory that Covid-19 may have been released by US military personnel during the 2019 Military World Games, which took place in Wuhan from October 18-27.

Vladimir Putin, meanwhile, has turned the tables on the West in more glaringly by arguing the biodefence work being done in biolabs in and around Ukraine is merely a cover for an offensive programme designed to target Russia.

For as long as all sides resist transparency, however, disproving such claims concretely will remain a difficult task.