Where did it come from?

Nearly four years later, the origins of COVID-19 remain maddeningly elusive

In the early morning hours, a ghostly group of figures in white fanned out across the Huanan seafood market in Wuhan, China.

Clad in hazmat suits, the researchers from the Chinese Centre for Disease Control and Prevention had been dispatched to the sprawling downtown market to hunt for the cause of a strange new virus that was infecting the city’s residents.

Closed that morning by Chinese authorities, the market was a mélange of alleys with tightly packed stalls selling all manner of seafood, meat and miscellany. Straddling a main road near a busy railway station, the market was also the place to go if you fancied such animals for pets or even food as snakes, porcupines, hog badgers and wild boar.

It had recently come to the attention of authorities as the site of multiple infections of this “pneumonia of unknown origins” and Chinese scientists were hoping that analyzing samples from the environment in and around the market could yield some clues about where the illness came from.

In their sweeps beginning on Jan. 1, 2020, the researchers targeted a wide swath of locations — doors, floors, wildlife stalls, public toilets, trash cans, sewage drains and even dead and stray animals. They collected nearly 1,000 environmental samples and more than 400 animal samples during several visits over the next three months.

But it would be another three years — three years during which the COVID-19 pandemic swept the planet and became the largest public health emergency in a century, infecting more than 750 million people, killing an estimated 15 million, and upending the lives of countless more — before the genetic data culled from those market samples would be made public.

It was a delay that prevented scientists early in the pandemic from gaining insight into how the virus may have first infected humans, information that could have aided officials in their efforts to clamp down on the early spread. The delay allowed conspiracy theories to fester and was a decision that Dr. Maria Van Kerkhove, technical lead for the COVID-19 response at the World Health Organization, calls “inexcusable.”

It wasn’t until March 2023 that a French researcher stumbled on the Chinese data and blew the search for the origins of COVID wide open.

The missing data was just one puzzle piece in what has become a complicated and contentious search for the virus’s origins — an endeavour made all the more difficult by a lack of transparency on the part of Chinese authorities and a desire by some U.S. lawmakers and pundits to exploit every turn for political or financial gain.

As we approach the fourth anniversary of the first diagnosis of COVID on the planet, we still don’t know conclusively where the virus came from. While a scientific consensus has emerged among many of the world’s top virologists about how it most likely spread to humans, there has been no smoking gun found.

The truth is, experts say, when it comes to unravelling the origins of COVID-19, there may never be a smoking gun. There may just be the

The lab-leak theory has

continued to thrive, thanks in part to social media and the

theory’s value as a political wedge issue

best theory of what happened, based on the evidence uncovered.

That can be a frustrating answer, in a time when many of us expect a certainty from science similar to movies and television. The legacy of that uncertainty has spread into almost every aspect of our lives: our politics, our schools, our workplaces, our relationships.

Many cannot let go of then-U.S. President Donald Trump’s comments early in the pandemic that he was confident the virus originated in the Wuhan Institute of Virology, one of China’s top biotech labs studying coronaviruses.

“I can’t tell you that. I’m not allowed to tell you that,” Trump told reporters in April 2020 when asked what evidence he had seen.

It would be a theme he returned to numerous times in the ensuing years, helping fuel conspiracy theories and division on Capitol Hill. Indeed, the four-year search for truth has become inextricably entangled with politics on both sides of the Pacific to the point where ordinary people, those who will suffer most in the event of another pandemic, often find it hard to distinguish between legitimate scientific evidence and theories from the dark corners of the internet.

This is the story of how scientists are working to unravel the mystery of COVID’s origins, and what they have found so far.

A French data find

One Saturday afternoon in early March of this year, evolutionary biologist Florence Débarre was sitting on the couch in her Paris apartment, poking around a global platform used by scientists to share data on emerging viruses.

She was trying to determine when a sample collected from gloves found at the Huanan market in January 2020 had been genetically sequenced.

To her astonishment, her search of genetic sequences from the market yielded dozens more results than she had seen on the site before.

At first, she thought what she was looking at on the platform, known as GISAID (Global Initiative on Sharing All Influenza Data), were mere placeholders. But a few days later, she went back and realized the magnitude of what she had found: massive files containing millions of short genetic sequences culled from swabs taken at the market in early 2020. Somehow, these troves of raw genetic sequences — until now unseen by the rest of the world — had quietly materialized online and sat there since January 2023 until Débarre came upon them.

In February 2022, Chinese researchers had released a pre-print study (meaning not yet peer-reviewed or published in an academic journal) analyzing the market samples and found some of them had genetic material of both SARSCoV-2 and humans. But the researchers claimed that none of the swabs testing positive for SARSCoV-2 had DNA from the animals found, which, they argued, “highly suggests” the virus made its way to the market through humans. They also floated the idea that the virus could have travelled to the market on frozen food.

Best practices would dictate that the genomic data of the samples, including those containing animal traces, that the Chinese researchers relied upon would have been posted at the time of their preprint. In this case, it wasn’t, and it’s not clear why.

“I do think it was very unusual,” said computational biologist Alex Crits-Christoph, a member of the international team of researchers who analyzed the data Débarre had stumbled upon. “Typically you deposit your sequencing data along with your pre-print so that others can have a look at the data too. That’s certainly the only way for it to be reproducible.”

Michael Worobey, a Canadian evolutionary biologist at the University of Arizona and also part of the team of international researchers, is more blunt in his assessment.

“They either knew all of that animal data was in those sequences from 2020 onward and didn’t say anything until they apparently accidentally shared data prematurely on GISAID, or they really weren’t aware of those different species until two years into the pandemic,” he said.

“I have a hard time thinking that they were so incompetent that they really didn’t understand the data back in January and February of 2020 that they were sitting on.”

The coronavirus that caused the SARS epidemic in 2002 was traced to horseshoe bats in China’s Yunnan province and is believed to have spread to humans through mammals sold at wet markets. Could animals sold at the Huanan market have passed the virus on to humans, similar to the first SARS epidemic? And could the genetic data contained in the samples posted to GISAID indicate a possible culprit?

“I was excited, because they were data I had been awaiting for a year,” said Débarre, a senior researcher at the French National Centre for Scientific Research.

She immediately contacted colleagues in other European countries, Australia and North America.

That night, Crits-Christoph set to work analyzing the sequences. Within a few hours, he had a prime suspect.

When Débarre opened her computer the next day, she was elated.

There was the unmistakable genetic signature of a raccoon dog, a mammal related to the fox and sold in China for fur and meat. Raccoon dogs are also highly susceptible to SARS-CoV-2 infection and are known to shed the virus efficiently.

“The feeling was intense,” Débarre recalled.

But while the samples confirmed that raccoon dogs were present at the market, Débarre stresses that the data do not prove that any of these animals were infected, nor that the virus entered the market in an animal.

What the data did show, however, was that the genetic fingerprints of animals known to be susceptible to SARS-CoV-2, including raccoon dogs, but also Siberian weasels, Amur hedgehogs and hoary bamboo rats, appeared in the live wildlife stalls in the southwest corner of the market where a lot of the virus was detected.

While there was no human case of COVID detected in the wildlife stall where a high proportion of samples were positive, Débarre says it is possible someone there was infected at some point.

“To me, so far the data are all consistent with a natural origin,” she said.

It was a kind of “missing link” moment researchers around the world had been waiting for. It was the evidence that confirmed a transmission from an infected animal at the market to a human was possible.

Then, just as surreptitiously as the data was posted on GISAID, it disappeared.

Worried that the data might be gone for good, the international team of researchers posted the results of their preliminary analysis in a report online.

Worobey contacted the Chinese researcher who had shared the data on GISAID to tell them of the raccoon dog finding and to ask if they would be interested in collaborating. He says he never heard back. He also reached out to the three senior researchers on the Chinese team and says they all declined to collaborate.

It wasn’t until April that the formal, peer-reviewed results of Chinese researchers’ own analysis of the market data was published in the journal Nature. This time the researchers confirmed the presence at the market of raccoon dogs and other animals. But the authors said that they did not have sufficient evidence to support the hypothesis that the virus spilled over from market animals to humans, nor that any of the animals were infected.

George Gao, former China CDC director and one of the authors of the study, did not respond to multiple interview requests from the Star, but told the prestigious journal Science earlier this year that the sequences were “nothing new” and that it was known there was “illegal animal dealing” at the market.

The WHO’s Van Kerkhove slammed the Chinese researchers’ decision to sit on the data for three years, saying in an editorial in Science that the information should have been shared immediately.

International scientists were quick to point out several errors in the Nature paper and questioned how a prestigious journal could have allowed such glaring mistakes to slip by. For example, the researchers claimed the samples contained genetic traces of animals that could not possibly have been at the market, including a panda, a finding universally derided by the scientific community and one that even Gao told Vanity Fair was laughable, without explaining how the error could have happened. (Killing a panda, which is a protected species and a symbol of Chinese pride, is punishable by death in China.)

The theory of a leak from a Wuhan lab won’t go away, and maybe never will. But virologists are amassing increasing evidence to explain how the virus emerged

“This means that the Chinese study either misidentified the DNA samples they studied (incorrect assignments), or that their swabs were contaminated in the lab when they were being analyzed,” Alice Hughes, a conservation biologist at the University of Hong Kong, wrote in the British weekly The Spectator. “Either way, this does not inspire confidence in the study.”

In an email to the Star, a spokesperson for the company that publishes Nature pointed to a line in the study saying the exact list of species is “not definitive” and that more analysis with “other methods will be required to provide more information regarding the wildlife species present at the market.”

Despite the Nature article’s shortcomings, it nonetheless added to the growing body of peer-reviewed research establishing that animals capable of spreading SARS-CoV-2 were present at the market.

The first COVID patients

It was not just the presence of animals at the market that has many virologists leaning toward a zoonotic transmission origin.

In November 2021, Worobey, who has made it his life’s work to figure out how viruses evolve, published a report pointing out that the patients with the earliest known COVID cases either worked at the market, had some connection to it or lived nearby.

The next year, Worobey and more than a dozen international colleagues published a study expanding on his earlier findings and made the striking finding that the homes of early patients who had no connection to the market whatsoever were geographically clustered, almost like a bullseye, around the market and resided even closer to it than cases with established links.

“This finding is really only consistent with establishment in a human population at the market and then it spreads outwards from there,” said Angela Rasmussen, a virologist at the University of Saskatchewan and a co-author of the study.

The researchers also found that positive SARS-CoV-2 environmental samples had a higher density in the southwest side of the market where live animals were known to have been sold.

But, says Rasmussen, because there are no samples available that were collected from the animals themselves, “it will not be possible to ever say definitively that a certain animal was the intermediate species. We can only try to narrow it down as best we can.”

Another study by a who’s who of global experts in virology, evolutionary biology and microbiology, including Worobey, also published in July 2022, not only lends further credence to a zoonotic transmission at Huanan market, but also highlights the improbable logistics required for the virus to make its way there from the Wuhan Institute of Virology.

The scientists submit that two distinct viral lineages of SARS-CoV-2, dubbed lineage A and B, infected people in late November or early December 2019.

Using a molecular clock analysis, which is based on the rate at which a virus evolves over time, the scientists concluded that the two lineages of SARS-CoV-2 evolved separately. That means they did not evolve from each other and would have had to have made their own jumps from animals into humans at different times, likely within weeks of one another. The scientists believe lineage B was the first to jump to humans in late November 2019 before it went on to become the most common variant in the world during the pandemic. Lineage A likely jumped into humans from an animal reservoir a week or two later.

SARS-CoV-2 is a generalist virus, able to infect and evolve in multiple mammals, including humans and many of the kinds of live animals sold at the Huanan market.

“So it’s not a strange thing for the virus, if it’s in animals that are in contact with humans, to jump multiple times,” said Worobey, who coled the molecular clock analysis study. “What would be strange is if it didn’t.”

Many scientists, including Rasmussen, argue that the peer-reviewed evidence we have to date simply doesn’t support a lab-leak hypothesis.

“There’s only one piece of evidence that supports the lab leak,” she said. “That is that the pandemic started in Wuhan and there happens to be a lab there.”

For the lab-leak theory to hold water, she says, a worker at the Wuhan Institute of Virology’s BSL-4 (biosafety level 4, the highest level of safety for the most dangerous microbes) lab would have to have been infected with lineage B at work, travel more than 20 kilometres to the Huanan market without infecting anyone else along the way, and then infect people at the market. The same thing would then have to happen again a few weeks later for lineage A.

“Is it possible? Yes. Is it probable or likely? Not really. Talk about an incredible coincidence,” Rasmussen said. “That would be nuts.”

U.S. intelligence on COVID

Earlier this year, a U.S. bill declassifying intelligence looking into the origins of COVID — hotly anticipated by representatives on both sides of the political spectrum — was signed into law by President Joe Biden.

Not long after, the Office of the Director of National Intelligence released a report summarizing what U.S. intelligence agencies apparently knew about what was going on at the Wuhan Institute of

Virology. The report said the first human infection with SARS-CoV-2 was “most likely” caused by an exposure to an infected animal and that there was no indication that the Wuhan Institute of Virology had pre-pandemic holdings of the virus or a close progenitor. Most of the agencies, the report said, had concluded that the virus was not genetically engineered.

The lab-leak theory has nonetheless continued to thrive, thanks in part to social media and the theory’s value as a political wedge issue.

It has also been helped along by the FBI, which, in contrast to many other U.S. intelligence agencies, says the virus “most likely” originated from a potential “lab incident” in Wuhan.

“The Chinese government, seems to me, has been doing its best to try and thwart and obfuscate the work here, the work that we’re doing, and that’s unfortunate for everybody,” FBI director Christopher Wray told Fox News in late February 2023.

The FBI declined to answer the Star’s question as to what led it to make such a conclusion.

Also this year, the Wall Street Journal revealed that the U.S. Department of Energy had determined with “low confidence” that based on classified intelligence the virus was most likely the result of a lab leak. CNN reported that this determination was based, in part, on intelligence about research being done at the Wuhan CDC, a few kilometres away from the Huanan market, unlike the Wuhan Institute of Virology.

China has rebuffed these assertions, saying it has been “open and transparent” in its efforts to track the cause of the pandemic.

Shi Zhengli, director of the Centre for Emerging Infectious Diseases at the Wuhan Institute of Virology, did not respond to several interview requests from the Star.

The WHO’s Van Kerkhove says any U.S. intelligence reports are simply hearsay without scientific evidence that can be publicly scrutinized. And without more data from Chinese authorities, a lab leak — or a breach in biosafety, as she calls it — can’t be ruled out.

Science is a process of trial and error, hypothesis refinement and following evidence under the watchful eyes of (oftentimes critical) peers. The intelligence community, on the other hand, operates behind closed doors. In many ways, the two mindsets are understandably incompatible.

In 2021, the WHO’s director-general established the Scientific Advisory Group for the Origins on Novel Pathogens (SAGO) to help guide the organization’s search for the origin of emerging viruses. In its preliminary report last year, it said access to evidence of all laboratory activities involving coronaviruses would have to be granted.

To date, Van Kerkhove says that hasn’t happened.

“We are told the work is done, look elsewhere. We are told that the animals have been tested and to date, not a single animal has tested positive in China … it’s just not possible,” she said, adding she believes Chinese scientists likely conducted studies tracing the animals sold at the Huanan market back to the farms where they originated, as well as serology tests of market vendors and maybe even animal farm handlers.

“I believe it was done because the China CDC knew what to do. They’ve done it before,” said Van Kerkhove, referencing the country’s experience with avian influenza and the first SARS outbreak in 2002 and 2003. “There are records that exist in these labs.”

The lack of transparency from Chinese authorities has fuelled skepticism in some scientific quarters.

Indeed, a tension has emerged between scientists who have spent years digging into COVID’s origins using what they say is the best evidence that has emerged to date and those who are holding out for a higher standard of proof. The problem is that, four years on, such standards may never be met.

David Relman, a microbiologist at Stanford University, argues that just because we don’t have good data about what research was being done at the Wuhan Institute of Virology or the Wuhan CDC doesn’t mean there is evidence that nothing dangerous was going on.

Relman was one of 18 scientists, including Worobey and Bloom, who signed a May 2021 letter in Science saying that in the absence of good data, both natural spillover and lab-leak hypotheses should be taken seriously.

But while Worobey says the scientific evidence that has emerged to date points to a zoonotic transmission as more likely, Relman says we don’t have a full picture about what was going on in the Wuhan laboratories yet to make any conclusions.

“What were the viruses on which they were working? What were the experimental designs? What were the starting points? What about all the experiments that didn’t quite work out, or at least didn’t seem to have worked out?” he says.

Still, Rasmussen and many of her colleagues wonder if any amount of evidence will be sufficient for those who doubt a zoonotic transmission at the Huanan market.

There may simply be a limit, she and others say, to where the scientific sleuthing over COVID’s origins will take us, given all the challenges there have been.

“Sometimes a lack of evidence when you’ve been repeatedly looking for it, might be evidence that the evidence you’re looking for doesn’t exist. And that’s one of the challenges.”

The lack of transparency from Chinese authorities has fuelled skepticism in some scientific quarters