In retrospect, a news story published by the science journal Nature in 2015 makes for chilling reading. It reported that scientists in the US had transferred a small piece of the genome from a coronavirus found in bats in China into a variant of Sars-CoV, the virus responsible for the outbreak of the lethal respiratory infection Sars between 2002 and 2003. The added genetic material encoded a so-called spike protein that enabled the Sars-like virus, which had been adapted to infect mice, to infect cells of the human respiratory tract too. “Our work”, wrote the scientists, led by epidemiologist Ralph Baric, “suggests a potential risk of Sars-CoV re-emergence from viruses currently circulating in bat populations.”
“The virus in wild bats would need to evolve to pose any threat to humans,” Nature wrote—“a change that may never happen, although it cannot be ruled out.”
It is now generally accepted that Sars-CoV-2, the virus that caused the Covid pandemic, is a bat coronavirus, closely related to Sars-CoV and able to infect humans—a so-called zoonotic virus, capable of jumping host species—because of its spike protein. Had more attention been paid to the risk that Baric and colleagues identified, might the catastrophe have been avoided?
On the other hand, was it wise for researchers to create in the lab a new way for Sars-like coronaviruses to infect humans? Baric’s experiments provoked alarm from some experts. “If the virus escaped”, virologist Simon Wain-Hobson told Nature back then, “nobody could predict the trajectory.” The work had proceeded despite a 2014 US government moratorium on federal funding of such “gain-of-function” (GOF) research. In this research, potentially pathogenic viruses are given new capabilities by lab-based cultivation or genetic engineering in a way that might increase their infectiousness or virulence in humans, with the aim of identifying and averting the risks posed by such agents of disease. The US National Institutes of Health (NIH) decided, after a review, that Baric’s work was not dangerous enough to fall under the moratorium.
But there is another story one could tell about that controversial research. One of Baric’s co-authors was Zhengli Shi of the Wuhan Institute of Virology (WIV), where Sars-CoV-2 was first isolated after people in the city began turning up in hospital with a mysterious respiratory disease in late December 2019. Working at the WIV using the same methods developed by Baric, Shi later went on to engineer other Sars-like bat viruses. That work was partly funded with money from an NIH grant to the EcoHealth Alliance, a US nonprofit organisation led by British zoologist Peter Daszak, which was given exemption from the NIH moratorium, seemingly on the grounds that wild bat coronaviruses were not known to infect humans and that the specific experiments at the WIV did not meet the definition of gain-of-function.
You probably see where this is going. The WIV and Shi’s lab in particular, as well as Daszak’s involvement and the NIH funding of that work, are now at the centre of suggestions that Sars-CoV-2 was not, after all, a natural zoonotic virus that spilled over to humans, but rather was produced in lab GOF experiments and leaked out into the human population because of poor biosafety measures.
This lab-leak idea has become another politically polarised pandemic controversy, perhaps even more explosive than the furious arguments over masks, lockdowns and vaccines. Because if Chinese scientists manufactured the Covid virus and were responsible for it leaking out into the world, what then should be the international response?
There are several strands to the lab-leak theory. Can it be coincidence, some ask, that GOF work on bat coronaviruses was going on in the very city where the outbreak began? Others have argued that the Sars-CoV-2 virus itself shows telltale signs that it had been engineered. But all such “evidence” so far has been circumstantial—often little more than innuendo—and some of it proved plain wrong. Careful investigations of how Covid began have uncovered a wealth of evidence supporting the view that Sars-CoV-2 is wholly natural in origin. By contrast, lab-leak origin theories require the assumption that an awful lot of people are lying. Indeed, some theories invoke a cover-up that implicates not just the Chinese government and scientists but also the EcoHealth Alliance, the NIH and perhaps even the US government and its chief medical adviser, Anthony Fauci. At the extreme end you find the stuff of classic conspiracy theories.
Yet the lab-leak theory is not going away, for—in its more respectable forms—it is not just a baroque fantasy of cranks, political agitators and xenophobes (although they flock to it). Some respected scientists still harbour these suspicions, and a preliminary report last June by the World Health Organisation—a follow-up to an earlier WHO investigation between 2020 and 2021—did not rule out the idea. So while for some the lab-leak idea seems to be a politically motivated excuse for China-bashing, it should not be dismissed out of hand. The better we understand where Sars-CoV-2 came from, the more likely we are to prevent another such virus from killing millions again.
The question of Covid’s origin was always likely to be fraught. As the pandemic hit in early 2020, US president Donald Trump found anti-Chinese sentiment politically expedient. He insisted on calling Sars-CoV-2 the “Chinese virus”, and it wasn’t long before his administration began hinting darkly that they had strong evidence (which has never materialised) of it being engineered in China. The secretary of state, Mike Pompeo, claimed there was “a pile of evidence a hundred feet high” that it was human-made. One former official implied that Trump’s State Department wasn’t so much investigating whether Sars-CoV-2 was a Chinese bioweapon that had escaped as actively constructing the argument that it was. Media sympathetic to the right-wing agenda have given the idea legs. Much of that advocacy drew on hearsay—as when, in May 2021, the Wall Street Journal reported that three Chinese researchers had fallen mysteriously ill in November 2019, before anyone else had caught the virus (there is no evidence they had Covid).
From the outset, then, the lab-leak theory became associated with an aggressively Trumpist agenda to defame China and deflect from domestic US failures in pandemic management. NIH funding of the EcoHealth Alliance’s partnership with the WIV for researching bat coronaviruses was withdrawn in April 2020, apparently on Trump’s command.
The issue created deep discord at a time when the US, China and other nations urgently needed to be fostering diplomatic harmony. Some worried that a breakdown of scientific collaboration could hamper efforts to control this and future pandemics. Epidemiologist Ray Yip told Nature that the anti-China rhetoric probably led to a less cooperative mindset in Beijing. “I think there was a shift in China’s attitude when they began to feel they were being humiliated or blamed for this pandemic”, he said.
In May 2020, the WHO announced its first investigation into the origin of Sars-CoV-2. From its initial findings released in March 2021, the inquiry concluded that a lab leak was “extremely unlikely”—even so Tedros Adhanom Ghebreyesus, the WHO’s director general, said that it could not be ruled out. That caveat was echoed by the expert signatories of a letter published in May 2021 in the journal Science, implying that the WHO investigation had taken too much on trust from the Chinese authorities. “If the only information you’re allowing to be weighed is provided by the very people who have everything to lose by revealing such evidence, that just doesn’t come close to passing the sniff test”, microbiologist David Relman told the Washington Post.
Such well-intentioned scepticism was, however, hard to keep proportionate while conspiracy theories were abounding. Some scientists claimed that the virus was linked to the WIV based on little more than guilt-by-proximity—ignoring the fact that most major cities in China had similar labs studying coronaviruses because of the 2002 Sars outbreak, meaning there would be such a “coincidence” if the pandemic had started in almost any big Chinese city. Some US politicians and media sources openly alleged that the Chinese government, and perhaps the US government too, was covering up evidence of a lab leak. In summer 2021, Republican senator Rand Paul suggested that Shi’s lab might have created the virus, while accusing Fauci of having lied to Congress about the kind of work the US supported at the WIV. Fauci (who stepped down from his institutional and governmental roles at the end of 2022) has been a target of the more unhinged lab-leak proponents: Elon Musk recently accused him of funding research that killed millions of people.
Virologist Stuart Neil of King’s College London says that, despite such allegations, “there is no evidence that Sars-CoV-2 is the product of laboratory engineering or any sort of experimental fiddling”. The leading theory now backed by most scientists is that the virus arose in wild bats and found its way into animals (perhaps via a pangolin or a civet cat) sold at the Huanan Seafood Wholesale Market in Wuhan. The bat populations most likely to have harboured the virus are in southern China (probably Yunnan province), far from Wuhan—but that is no big mystery, for the market was, before being shut down in January 2020, partly supplied by wildlife farms in that region. Such long-distance transmission of a viral pathogen is nothing unusual, says Robert Garry of Tulane University Medical Center in New Orleans—it happened with the Ebola virus in Africa, for example. Zoonotic transmission from wild animals sold at Chinese markets is also how Sars is thought to have entered the human population.
Last July, a detailed study of the early epidemiology of Covid-19 offered compelling support for this picture. A team of scientists led by biologist Michael Worobey of the University of Arizona reported that most of the earliest human cases centred around the Huanan market. In the section where live wild animals were sold, samples containing the virus were taken from carts, drains and a metal animal cage. It was already known that 27 of the 41 people initially hospitalised in Wuhan had had direct exposure to the market, while 55 of 168 of the first known cases were also associated with it. Worobey and colleagues also found that the residential addresses of 155 of those first cases pointed to the market as the epicentre of infection. “There are no other epidemiological links to any other place in the city,” said Kristian Andersen of the Scripps Research Institute in California, a co-author of the study.
A parallel study of the Sars-CoV-2 genome suggested that there were at least two events, probably sometime in November 2019, when the virus jumped between species. Many experts feel that the conclusion is now rather clear. “The science is largely pointing in one direction, and that is that the pandemic began naturally,” virologist Angela Rasmussen of the University of Saskatchewan told the BBC last November.
The lab-leak origin theory requires an awful lot of people to be lying
What, then, sustains the lab-leak theory? Among scientists, the idea that Sars-CoV-2 was engineered has tended to rely on a different kind of evidence: claims that the biochemical nature of the virus bears signs of the kind of artificial tampering that might be expected in GOF experiments. One early line of argument concerned a part of the spike protein that attaches to human cells in the first stage of attack. Called the furin cleavage site (FCS), it acts like a kind of switch that flips the virus into infectious mode, and is thought to be a key factor in making Sars-CoV-2 highly transmissible. FCSs are found in other coronaviruses in the wild, but the original Sars virus does not have one—if somebody was trying to modify it, this would be the obvious change to make. In May 2021, Nobel laureate virologist David Baltimore was quoted as saying that the FCS looked like a “smoking gun” for an engineered origin—an imprudent suggestion that he later withdrew, in view of the fact that the FCS could have hopped to Sars-CoV-2 from some other virus in the wild.
The somewhat unexpected FCS was noticed as soon as the genetic sequence of Sars-CoV-2 was known in early 2020. Fauci discussed what this might mean for the idea of an engineered origin on 1st February, in a confidential teleconference and subsequent emails with NIH director Francis Collins, head of the Wellcome Trust Jeremy Farrar, the UK government chief scientific adviser Patrick Vallance, and other experts. Although conspiracy theorists have asserted that these scientists then tried to bury the issue, emails released last November show a conventional story. The participants took seriously the idea that the FCS might imply a lab origin, sought expert advice, notified the WHO, and concluded that a lab origin was “largely incompatible with the data” but that more study was needed. In other words it was science as normal, conducted with open minds and commendable rapidity.
There was another flurry of lab-leak rumours last October, when a trio of scientists (none of them virologists) released a preprint—a paper not yet peer-reviewed—claiming that a certain kind of genetic sequence called a restriction site recurs in the Sars-CoV-2 genome at suspiciously regular intervals. This, the authors argued, suggests that the viral genome might have been artificially constructed by joining together short segments of roughly equal size. Several experts initially took the idea seriously. Systems biologist Francois Balloux of University College London told the Economist that “I couldn’t identify any fatal flaw in the reasoning and methodology”—only to later withdraw that view. Others pointed out that there were statistical reasons why a sequence like this may arise naturally. Andersen called the pattern “random noise”, while virologist Edward Holmes, who was part of the team that first announced the genome sequence of Sars-CoV-2, called the preprint’s claims “nonsense”.
Such false alarms ought to recommend caution about what a natural or engineered virus “must” look like, or what nature can and can’t produce. “If you’ve been a virologist for any length of time, nothing surprises you about what a virus can do,” says Neil. There is a frighteningly vast reservoir of wild coronaviruses; we know next to nothing about the overwhelming majority of them. The genomes of such viruses are immensely diverse, precisely because they are in a sense “randomly engineered” by natural selection in the wild. If two coronaviruses infect the same host, they can exchange their genetic material. This “recombination” helps the virus to increase its diversity and thereby its potential to spread: occasionally it will, by chance, make a much more effective virus. It is not at all hard to see how the FCS could have arisen this way. Recombination also explains why it is hard to identify a natural source for a virus like Sars-CoV-2: every segment of the genome will probably have come from a different progenitor.
Suspicions that the virus was leaked from a lab have surely been fed by the habitual belligerence, secrecy and misinformation coming from China’s leaders. The Chinese propaganda machine retaliated to the accusations from Trump and his followers by encouraging far-fetched rumours that the virus in fact originated in the US, and its pandemic narrative now seeks to build on suggestions that the virus came from beyond the country’s borders. Several studies by Chinese authors have insisted there is no evidence of close relatives of Sars-CoV-2 in bats in China—which to many outside scientists is utterly implausible. Although similar viruses have indeed been reported in animals in Laos, Cambodia and other parts of southeast Asia, China remains the most probable source. In fact, in November last year, a study of bats in Yunnan showed these creatures to be a veritable reservoir of viruses, including several capable of infecting humans—one of which is closely related to both Sars-CoV and Sars-CoV-2. Transmission between bat species, and probably to other animals, is surely happening there all the time. “The idea that the pandemic didn’t originate in China is inconsistent with so many other things,” Jesse Bloom, a virologist at the Fred Hutchinson Cancer Center in Seattle, said last August.
From the very beginning of the pandemic, the authorities in Beijing moved quickly to control what Chinese scientists could say. In March 2020 they issued a directive that said all studies related to the pandemic (including those by private companies) should be vetted by government research units before publication. Such restrictions seemed motivated partly by a reluctance to disclose details of how wild animals are supplied to China’s meat markets. The WHO investigation was not given access to the wildlife farms that supplied the Wuhan markets (and which apparently no longer exist), and its report was evasive on the issue of what was on sale at the Huanan market. Such reticence might be due to official embarrassment at the failure to enforce bans on the illegal sale of protected wild species there. The problem is that, supposed lab leaks aside, China already has too much that it might want to cover up.
Alina Chan, a scientific adviser at the Broad Institute and author with journalist Matt Ridley of a book that argues the case for a lab-leak origin, points out that the countries that exert the most leverage on organisations like the WHO—such as China and the US—are also the ones most invested in high-risk research, such as in GOF experiments. “I’m not even saying that the Covid-19/Sars-CoV-2 virus came from a lab,” she has said. “I’m saying that the bar set for launching a credible investigation into [a lab origin] is set so high that I can’t see how you could launch one unless you already had incontrovertible proof for it.”
Supposed lab leaks aside, China already has too much that it might want to cover up
But casting aspersions on a secretive state is easy, and often unfalsifiable. Although some are calling for an independent investigation of the WIV, says Jane Qiu, a freelance science reporter based in China who has had unparalleled journalistic access to Shi and the institute’s work, it is likely that “no investigation would satisfy them that a lab leak didn’t cause the pandemic, because they can always argue WIV have hidden something or have destroyed all the evidence.” Accusations that “they’re hiding something” easily morph into anti-Chinese bias. “Lab-leak theories are often bolstered by racist tropes that suggest that epidemiological, genetic, or other scientific data have been purposefully withheld”, Garry wrote last November.
With such wildly different motivations at play, the debate about the origin of Covid-19 quickly curdled; it is now poisonous. Scientists and journalists pointing out the shortcomings of the lab-leak theory have been subjected to torrents of abuse, sometimes from other scientists. Female researchers and reporters have often borne the brunt. Qiu in particular has been attacked, partly on the basis of her nationality (even though she has regularly reported stories critical of the Chinese government). “Because people don’t trust the Chinese government, for very good reasons, they think it follows that they can’t trust anybody who is Chinese,” she says.
The problem is that the lab-leak theory is both inherently possible but also endlessly attractive to contrarians and bad-faith actors: it is rich ground for “deep state” conspiracies, xenophobia, misogyny and anti-authority campaigns against figures like Fauci. “It’s really a perfect storm of everything,” says Qiu.
As evidence of a natural origin has accumulated, Rasmussen told the World Service, “a lot of the people who really favoured the so-called lab-leak theory have become increasingly desperate—really there’s no other word for it—to keep the lab-leak theory on the table.” Now it has got to the stage “where the personal attacks have actually moved from being accusations of conflicts of interest and speculation about conspiracies to saying that we should be physically targeted.” Last November, an anonymous post on an online message board encouraged readers to kill six Covid scientists, including Rasmussen, along with their family members. “I did not go into academic virology thinking I was going to need security,” Rasmussen said.
The debate on the origins of Sars-CoV-2 has joined the roster of arguments in the age of political polarisation on which nothing can be said that will not outrage someone. But whatever the merits or flaws of the lab-leak theory, the possibility that a pandemic could begin this way—by escape of a virus engineered in GOF experiments—must be taken seriously. Covid-19 has rightly prompted reconsiderations of such studies: under what criteria might they be warranted, and are current safety procedures adequate? One of the charges thrown at the WIV, for instance, is that some of its research was being conducted with insufficient biosecurity measures. “It’s no longer abstract to think about the destruction that the spread of a new virus can cause,” epidemiologist Marc Lipsitch, of the Harvard TH Chan School of Public Health, told Nature.
There was plenty of unease about GOF experiments even before Covid-19. A 2014 study by US and Japanese scientists that produced a virulent and transmissible avian virus with genetic similarities to that which caused the 1918 “Spanish flu” was denounced as “absolutely crazy” by former UK government chief scientific adviser Robert May, and prompted Wain-Hobson to say: “If society, the intelligent layperson, understood what was going on, they would say ‘What the F are you doing?’”
All the same, Neil is uncomfortable linking the debate about GOF experiments too closely to the lab-leak hypothesis. “We have a bunch of people with no idea what they’re talking about trying to push this debate for political or ideological reasons, and trying to present people in my profession as cowboys,” he says. Some of this research is vital for understanding and combating viral threats, he argues. But zoologist Matthew Cobb, whose 2022 book The Genetic Age considers the hazards of genetically engineered viruses, points out that “none of the supposed insights provided by GOF studies were of any use to us” in fighting Covid-19.
The earlier flu work, along with a few lab accidents, was part of what motivated the 2014 moratorium in the US. Other studies continue to be controversial. There was alarm last year, for example, at plans by a team at the US National Institute of Allergy and Infectious Diseases to engineer a strain of the monkeypox virus so that it might cause more serious illness, and to test it on mice as part of a strategy to develop defences against it. The strict rules on manipulation of “potential pandemic pathogens” don’t apply to this study because monkeypox is not classed as such a pathogen, and the experiments will just use natural mutations rather than creating new ones. Part of the problem with regulation is that there is no clear definition of what is and is not a “gain of function”.
Neil welcomes a review of the guidelines: “We have to update these things constantly,” he argues. “It’s absolutely legitimate that people should have fears about the proliferation of labs [conducting this kind of research] around the world.” The pandemic itself delayed a planned 2020 review of guidelines from the US National Science Advisory Board for Biosecurity on risky pathogen research; after issuing draft recommendations in September 2022, the final report is scheduled for early 2023. Draft legislation prepared by Republicans to place another moratorium on funding of GOF research, after the previous one was lifted in 2017, has worried some researchers. The challenge, most agree, is to find the right balance between permitting research that could be vital to tackling infectious disease, and preventing risks that do not have adequate potential payoffs. The Covid pandemic has clearly sharpened the debate, for example by prompting the formation of an international task force called the Pathogens Project to make recommendations for future biosafety regulations.
Why make such a fuss about the origins of Covid? Some say that it is here anyway and we must deal with it regardless of where it came from. But that is too simplistic. We might question the motives of those hoping to make political hay from Chinese embarrassment, or even demanding reparations. But if China’s (or anyone’s) biosafety procedures were found to be inadequate, the implications would be global. And if, conversely, the source is indeed natural, we should want to know what and where it was and to understand the routes of zoonotic transmission in order to reduce the chance of it happening again.
It typically takes a long time—years or decades—to understand where new viruses have emerged from. There are still gaps in what we know about the origins of the Ebola virus and the coronaviruses that cause Sars and Mers. But identifying possible shortcomings in biosafety measures isn’t contingent on proving that they failed for Sars-CoV-2. And identifying potential transmission routes of zoonotic viruses into human populations can guide policy even if we’re not totally sure whether they applied to this virus. An international agreement for open cooperation on future investigations of this sort would be a gain for all. There may come a point where we have to accept that we know all we can about the possible origins of Covid, and ask instead how we can put what we have learnt to best use.