Why scientist Eddie Holmes changed his mind on the origins of COVID-19

Among the public, virologist Eddie Holmes is perhaps best known for his strident and repeated argument SARS-CoV-2, the virus that causes COVID-19, did not come from a lab.

“There’s no evidence in any of the science. There’s absolutely nothing,” he said last year. “The siren has definitely sounded on the lab leak theory.”

In the early, frantic days of the pandemic, Holmes was worried they were looking at the terrifying consequences of a lab accident. “We all thought, ‘Oh f---’,” he tells The Age.

Virologist Eddie Holmes, in Sydney in March 2022.Credit: The New York Times

Holmes, a University of Sydney professor and researcher who studies how viruses evolve, was the first person to publish a genetic sequence of SARS-CoV-2. He spent the next few weeks emailing, exchanging Slack messages and huddling in late-night teleconferences with other scientists, discussing their discoveries and anxieties.

They soon spotted two unusual features in the virus’s genetic code; taken together they suggested the virus had been engineered – allowed to infect cells in a lab until it developed lethal mutations, turning mouse into monster.

“I just can’t figure out how this gets accomplished in nature,” one of the scientists wrote in early 2020.

And they knew the Wuhan Institute of Virology was working on coronaviruses using low-level biosecurity practices.

Surely, no one would be mad enough to do what they were suggesting, wrote one researcher. Another replied: “Wild west…”

Six weeks later, the researchers published a paper in Nature Medicine setting out an entirely different scenario: they were now convinced the virus had leapt from a wild animal into a human.

“We do not believe that any type of laboratory-based scenario is plausible,” they wrote.

The authors titled their paper The proximal origin of SARS-CoV-2. It became famous at first: a scientific rebuke to conspiracy theorists and Donald Trump. Then it became infamous.

Last month, two of Holmes’ co-authors were hauled before a Republican-controlled congressional investigation in the United States, where the scientists were accused of collusion in a cover-up. Their internal emails and messages were released and dissected. They have received death threats. Holmes, who won the Prime Minister’s Prize for Science in 2021 for his coronavirus work, was front-page news.

The final indignity: US Congressional Republicans released an investigation titled The Proximal Origins of a Cover-up.

All this heat and light comes down to a single question: why, within a few frantic weeks, did Eddie Holmes change his mind?

A protester sits in a congressional hearing into the origins of the pandemic, in Washington in April.Credit: Bloomberg

Holmes did not set out to become a public figure. Unlike other coronavirus researchers, he has made little effort at growing a social media following. He continues to do what he has always done: publish highly cited papers on viruses (24 this year already).

Under such concerted attack, some might grow jaded or bitter, or respond with anger. Holmes seems more exasperated. “It’s surreal,” he says. “It’s just a bloody scientific paper.”

It helps, perhaps, that Holmes does not think this is about him. “We are pawns in a game,” he says. The goal? To checkmate Anthony Fauci.

In America’s hyper-polarised discourse, Fauci – the former director of the National Institute of Allergy and Infectious Diseases – has become a lightning-rod for Republican fury about lockdowns, mask mandates and school closures.

On February 1, 2020, Fauci held a teleconference with several researchers – including Holmes – to discuss their growing fears that SARS-CoV-2 might have leaked from a lab.

US Republicans now claim Fauci pressured or bribed the scientists to drop those fears and had them write their paper to quash any discussion about a lab leak (“Complete garbage,” says Holmes).

Holmes declined a request to give evidence at the Republicans’ inquiry.

“We were found guilty before we even went to testify,” he says. “I’m not going to fly halfway around the world to be shouted at for a few hours.”

While the committee does not have Holmes, they have subpoenaed the emails he traded with US government bureaucrats and Slack messages.

Extensive coverage of the issue in The Australian newspaper and around the world drew an extraordinary response from the usually austere Australian Academy of Science, which released an open letter supporting Holmes.

“It is easy to sow doubt – to take sentences from here and there in email streams and compare early thinking with later conclusions – and presume any change is due to some unspecified pressure rather than a change in the weight or direction of evidence, or even argument,” the academy’s letter reads, noting similar tactics were used by the tobacco industry to attack evidence about the harms of smoking.

“The current level of discourse … reflects a worrying pattern of deliberate undermining of public trust in science.”

All of which brings us back to the core question: why did Eddie Holmes change his mind?

Reading the frantic messages between scientists is like cracking a time-capsule from the anxiety-soaked early days of the pandemic: we knew nothing, understood nothing, and even our worst nightmares seemed plausible.

Viral infection is a process with multiple steps. The virus first must lock tightly onto a cell, and then leverage its way inside.

The outside of our cells are studded with proteins. One, known as ACE2, listens for signals from the body to raise or lower blood pressure; our throat and lung cells are particularly rich in it.

SARS-CoV-2’s spike protein has a “receptor binding domain” that is perfectly shaped to slide onto the human ACE2 receptor and grip tight, like a lock sliding onto a key. It “didn’t look like anything we had seen before”, Holmes says.

The virus must then fuse with the cell. It does this by opening up its spike protein, like a yawning mouth, pulling it toward the cell. The spike is locked closed by a small protein loop, that needs to be snipped at just the right time by furin, a human enzyme.

Holmes and his colleagues had never seen a sarbecovirus, a genus of coronavirus that includes SARS and SARS-CoV-2 with a furin cleavage site, where the furin snips open the spike protein.

On February 4, 2020, Holmes emailed around a summary of the group’s thinking. The summary noted it was “possible” the furin cleavage site could have been acquired as part of lab work on a virus; the summary notes similar work being done in Wuhan.

January 2020: A worker removes a giant salamander that escaped at the live seafood market in Wuhan, China.Credit: Getty

But by February 8 that year, emails show a complete about-face. “Personally, with the pangolin virus possessing 6/6 key sites in the receptor binding domain, I am in favour of the natural evolution theory,” Holmes wrote.

Why? Three key pieces of evidence convinced him, he says.

First, a search of the scientific literature revealed furin cleavage sites were commonly found in RNA viruses, including coronaviruses. They “come and go in their evolution”, Holmes said; the coronavirus that causes Middle East respiratory syndrome has one.

Second, the furin cleavage site was clearly “atypical” and not the type a scientist would insert as part of an experiment.

And most crucially, data came in from a Hong Kong team that was sequencing coronaviruses found in pangolins seized by Chinese customs. The coronaviruses had receptor-binding domains nearly identical to those seen in SARS-CoV-2.

“It was there in pangolins, which really showed it was natural,” Holmes says. “And if the receptor binding domain is natural why not the furin cleavage site?”

The Age asked four scientists on whether Holmes’ conduct was a normal part of the scientific practice or a cover-up. Should the scientists have disclosed their earlier fears about a lab leak?

“No. Absolutely not. Not the prior talks, no. Certainly, not in a scientific paper,” says Professor Sujatha Raman, who holds chair in science communication at the Australian National University.

Scientists should debate hypotheses, Raman says, but a paper just needs to reflect the conclusion, not the debate.

Professor Dominic Dwyer, a member of the World Health Organisation’s mission to China to investigate the source of the pandemic, described Holmes’ conduct as “absolutely reasonable, and typical of good scientific debate”.

Other researchers disagree. Professor David Relman, a microbiologist at Stanford University in California, says the evidence simply wasn’t strong enough to support the hypothesis in the scientific paper published in March 2020. The paper, Relman says, rested on the “faulty assumption” the Chinese had published all the sequences for viruses they held in their labs.

“Some of these authors continue to show a lack of objectivity and a continued interest in overstating what we know and don’t know,” he tells The Age.

Richard Ebright, a molecular biologist at Rutgers University in New Jersey, who has become among the most-outspoken scientific critics of a natural origin, goes even further. He accuses the researchers of “scientific fraud”.

The released emails “make it absolutely clear the authors knew the conclusions of the paper were unsound,” he says. “There is nothing – absolutely nothing – that resembles ‘a normal part of the scientific process’ in the process revealed by the authors’ correspondence.”

Did Proximal Origins stymie debate on the origins of COVID-19? While the paper calls a lab leak implausible, in the very next sentence it notes “more scientific data could swing the balance of evidence to favour one hypothesis over another”. Indeed, the acres of newsprint devoted to the topic would suggest little debate has been stoppered.

Flinders Street Station during a COVID-19 lockdown in August 2020.Credit: AP

Holmes, again, is incredulous.“This is not the 14th century. I’m not the Pope,” he says. “If you don’t believe it, write your own paper.”

Some did. Several scientists, including Relman and Canadian evolutionary biologist Professor Michael Worobey, wrote a call for the lab leak hypothesis to be taken seriously.

Worobey answered his own challenge. In a paper published in Science, he showed the first cases clustered around the Huanan Seafood Market, not the Wuhan Institute of Virology. The market was selling animals known to be able to be infected with COVID-19, and that environmental detections of the virus clustered on stalls selling wildlife.

Worobey and his co-authors further demonstrated two distinct viral lineages were present at the market; in a second paper, they showed it was most likely both had jumped into humans separately.

Finally, in March, Worobey published data from genetic swabs taken from the market at the start of the pandemic. Some swabs tested positive for both the virus and animals known to carry it. But, crucially, not humans.

The Morning Edition newsletter is our guide to the day’s most important and interesting stories, analysis and insights. Sign up here.