Wuhan, China“I get a shudder every time I drive past here,” says the taxi driver as he pulls over near the Huanan Seafood Wholesale Market in Wuhan, China. The wet market—where live poultry and mammals are sold and slaughtered on site—is where most of the early COVID-19 patients worked or visited before they were hospitalized in December 2019. “Huanan has become synonymous to the virus,” he says.
The market has been off limits since January 1, 2020, frozen in time behind high blue walls. Peering through the gap in the gate, I see rows of abandoned stalls under a beige plastic arched roof, an empty blue chair, a fish net, a few cooler boxes, and an overflowing rubbish bin. A red sign advertises Wuchang beam, a signature fish species of the Yangtze River, famed for its tender and aromatic belly. Another sign reads “fresh and live.”
As the world enters the third year of the COVID-19 pandemic, the precise origins of the virus that causes the disease, SARS-CoV-2, and the role of the Huanan market in kickstarting the outbreak remain hotly contested. Most experts agree that COVID-19 emerged because of zoonotic spillover—which happens when a virus jumps from wild animals to humans. But it’s been unclear exactly when and where the virus made that leap. Pinpointing where spillover took place is not just a matter of academic interest. “It has important policy implications” that will help avert risky practices and hopefully prevent the next pandemic, says Roger Frutos, a virologist at the University of Montpellier in France.
This week, after a comprehensive review of the available data, an international team published two studies in the journal Science that together conclude the Huanan market was the epicenter of the pandemic. Their work suggests that two closely related versions of SARS-CoV-2 jumped from an animal to a human during two separate events, most likely within the market.
The studies support previous the suspicion that “the animals in the market were the key element in early transmission,” says Dominic Dwyer, an epidemiologist at the University of Sydney, Australia. He was not involved in either study but was a member of the World Health Organization (WHO) team that visited Wuhan last year to probe the origins of COVID-19. “It’s clear the viruses were circulating in the market and then exploded out of it.”
But scientists still debate whether it was an infected animal or an infected person that brought the virus to the market in the first place. What’s more, scientists have yet to find an animal infected with a progenitor of SARS-CoV-2, so the conclusion that spillover took place at the market rests entirely on circumstantial evidence. “We shouldn’t base important statements on correlation,” says Virginie Courtier, an evolutionary biologist at the University of Paris, France.
It’s clear that Huanan sold live wildlife susceptible to SARS-CoV-2 in late 2019. But spillover could have happened to anybody who handled the animals across the supply chain, including the wildlife farms where wild animals are raised for sale, says Ronald Rosenberg, a virologist in Fort Collins, Colorado, who recently retired from the U.S. Centers for Disease Control and Prevention. According to a 2017 report commissioned by the Chinese Academy of Engineering, hundreds of millions of wild animals from hundreds of species are farmed for sale in the country. The report estimates the industry, which employs about 10 million people, is worth $76 billion; only a quarter of this amount came from animals raised for consumption.
And in the long run, experts warn that closing wet markets or banning wildlife consumption, as China has promised to do, is not going be enough to avert disease outbreaks. “We have a much bigger problem,” says David Redding, a conservation biologist at the Institute of Zoology in London, U.K.
As agriculture expands and intensifies around the world, different animal species—both domesticated and wild—that don’t normally come together will increasingly mix and exchange pathogens, creating a viral melting pot. This, says Redding, is the root cause of not only SARS and COVID-19 but other infectious diseases that have been emerging with increasing frequency in recent decades.
The troubling history of wet markets
Even before COVID-19, Huanan and other wet markets were notorious public-health threats because of their potential as breeding grounds for new pathogens. Most notably, a Chinese marketplace was found to be ground zero for the SARS outbreak in 2002, and other wet markets have been linked to several outbreaks of bird flu in recent years.
In fact, researchers in China were already conducting routine surveys of live animals sold at the Huanan market. In response to outbreaks of a deadly tick-borne disease, a team of scientists had been investigating four markets in Wuhan—including Huanan—every month between May 2017 and November 2019. That data, published in Scientific Reports last year, has since become a valuable resource for scientists trying to understand the origins of SARS-CoV-2.
The team revealed that these markets housed nearly 48,000 caged wild animals belonging to 38 species, almost all of which were sold alive and stacked in cramped, unhygienic conditions perfect for virus mingling and transmission. All of the wildlife trade the scientists surveyed at these four markets was illegal. Many vendors sold protected species, and none held the required certificates stating the source of the animals or that they were free of disease.
This survey was one of the core pieces of evidence used in the new Science papers, along with a study jointly conducted by the WHO and China, which published in March last year, and a leaked report from the Chinese CDC dated January 22, 2020, that described some of the environmental samples collected at Huanan in early 2020.
Their analyses show that most of the environmental samples that tested positive for SARS-CoV-2 were concentrated in the southwestern section of Huanan market, where vendors illegally sold live animal species susceptible to the virus—including raccoon dogs, red foxes, and bamboo rats—in the weeks leading up to the early outbreaks. A single stall in that part of the market—stall #29—yielded five positive samples, and four of them were from items apparently associated with the wildlife trade: a metal cage, a machine for removing hair or feathers, and two carts for transporting animals.
The most likely explanation is that the virus in those samples was shed by an infected animal, indicting a spillover event in the market, says Edward Holmes, a virologist at the University of Sydney in Australia and a co-author of both Science papers.
But scientists hunting for COVID-19’s origins are frustrated that animals infected with a progenitor of SARS-CoV-2 remain elusive. When Huanan was suspected to be the source of the early outbreak, vendors illegally selling live wildlife disappeared with their animals. Several research groups raced to track down the infected wildlife without success.
Between January 7 and 18, 2020, Tian Junhua of the Wuhan CDC and his colleagues collected samples from wild animals near the city, including 15 raccoon dogs from farms that supplied Huanan and hundreds of bats; they found no trace of SARS-CoV-2. Scientists who were part of the WHO-China joint study also failed to find the virus when they analyzed more than 600 samples from farms in Hubei Province that supplied the Huanan market. Their analysis of nearly 2,000 samples belonging to more than two dozen species of wildlife from the southern provinces of Yunnan, Guangdong, and Guangxi that are home to numerous coronavirus-carrying bats also failed to yield samples of SARS-CoV-2.
Rosenberg isn’t surprised the researchers came up empty handed. He says tracking down the source of a new zoonotic disease is like looking for a needle in a haystack, and it’s especially challenging—if not impossible—when the infection of an animal population might be only fleeting.
A cloudy genomic picture
Without homing in on infected wildlife, one way to prove spillover at Huanan is to find the animal virus in the market that started the pandemic. If COVID-19 emerged when a virus jumped from an animal to a human at Huanan, then it should be possible to find in the market a progenitor of SARS-CoV-2 that is different from the human counterparts. But such a virus hasn’t been found.
In a study posted in February that has not yet been peer reviewed, a team led by George Gao, director of the Chinese CDC who just stepped down this week, unveiled a complete analysis of the 800 environmental samples from Huanan, including sewage wells, the floor, walls, freezers, and animal cages. Two thirds of the 64 positive samples came from the southwestern section of the market where live wildlife were sold. Four of those positive samples—none from Stall 29 which Holmes and his colleagues believe was where spillover took place—yielded complete genomic sequences that are identical to human SARS-CoV-2. Therefore, the Chinese team concluded that the viruses were shed by humans, rather than animals, bolstering the case that the market was an amplifier of the virus, not the source.
Many scientists are skeptical. It’s hard to verify the conclusion since the data Gao’s team used are not publicly available, says Andrew Rambaut, an evolutionary biologist at the University of Edinburgh in U.K. and a co-author of both Science papers. He notes that positive samples from the market environment contain genetic materials from animals, and he would want to know which species they were, the quantity of genetic material, and how that might correlate with the location of the wildlife stalls.
Moreover, authors on the two new Science papers think they’ve found genetic clues to spillover happening at the market.
Joel Wertheim of the University of California, San Diego, led a team that scrutinized nearly 800 early viral sequences that were sampled from around the globe before February 2020. They found two forms of SARS-CoV-2, dubbed A and B, that differ by just two units of genetic code. Crucially, these forms gave rise to two gigantic explosions of genomic diversity in the early pandemic. Weitherm calls this “a telltale sign” that two closely related versions of the virus jumped from animals to humans on two separate occasions.
Both lineage A and B turned up in some of the Huanan environmental samples that the Chinese CDC team analyzed, notes Michael Worobey, an evolutionary biologist at the University of Arizona in Tucson and also a co-author of both Science papers. If the two-jumps theory holds water, then it’s most likely both jumps took place at the market—a hypothesis he regards as more probable than a scenario in which two individuals infected from outside Huanan brought the viruses to the same market.
“It’s a cannonball through the whole of this idea that the Huanan market was just an amplifying event,” says Worobey.
But other work suggests that a third virus lineage that emerged in October at the latest is the one that gave rise to all SARS-CoV-2 genomes. In a study published in Bioinformatics in March, a team led by Sudhir Kumar of Temple University in Philadelphia analyzed more than a million genomic sequences of SARS-CoV-2 collected from around the world and concluded that there was only one common ancestor, indicating just one jump.
These studies used different methods to infer viral evolution based on observed genomic sequences, and both approaches have uncertainties, so several scientists approached by National Geographic say that their conclusions should be taken with a grain of salt.
But while the precise location of the spillover remains murky, most virologists and infectious-disease experts agree that China’s wildlife trade played a central role in igniting the pandemic wildfire. There is, says Frutos, an urgent need to look at risks across the entire supply chain.
The future of wildlife farming
On an overcast December afternoon last year, I entered the bustling Qiyimen market in central Wuhan—one of the four markets that researchers found illegally selling live wild animals. The air was filled with a fishy scent and the sounds of vendors hawking their goods. The locals strolled past rows of stalls with fresh produce ranging from lotus roots and swimming eels to skinned rabbits dangling from hooks.
When asked about yewei (“wild taste”), the vendors became stone-faced. “Don’t mention that word,” one told me. “It’s not allowed anymore.”
China has banned the consumption of wild animals since February 2020. But many scientists say that it isn’t enough to prevent pandemics because numerous undomesticated species are still being raised for fur, leather, traditional Chinese medicine, zoos, pets, wildlife parks, and research.
“It’s literally wildlife farming on steroids,” says Peter Li, an expert on wildlife policy at the University of Houston-Downtown in Texas.
Until recently, little was known about public-health risks posed by wildlife farming. In February, an international team of scientists led by researchers in China reported the discovery of more than one hundred new types of virus in samples taken from nearly 2,000 farmed wild animals belonging to 18 species—many of which were sold at the Huanan market. Two dozen of those viruses were deemed high risk to humans. Many of those viruses, including new species of coronaviruses and influenza, have the potential to readily jump between species.
Some of the animals the team sampled were sick—often with coughs, runny noses, or diarrhea—but many infected individuals didn’t show obvious signs of disease. As in humans, viruses can silently circulate in wild animals, says Holmes, hopping from one species to another and constantly mutating. This gets easier when multiple species are raised on the same farm and housed together in crowded, poorly ventilated structures.
Compounding the problem, few farmers adopt basic biosafety measures, such as disinfecting before and after going into a barn or wearing gloves when handling dead animals. It’s incredibly dangerous to handle animals full of viruses without any protection, says Holmes. “Even the slackest biosafety standards in a lab are a thousand times better.”
Furthermore, animals are transported across the country on a regular basis without quarantine, potentially facilitating the spread of emerging diseases, says Li. Of greatest concern is that novel pathogens have ample opportunities to infect farmed animals because many wildlife farms are near forested areas and farmers often capture wild animals to replenish their stock.
Considering the situation, it’s highly likely that transient viral outbreaks plague wildlife farms every now and then, says Frutos. In most cases, infected workers may develop only mild or no symptoms, and the viruses die out after limited transmission. In a computer simulation Wertheim’s team published in Science last year, 95 percent of viruses causing spillovers in sparsely populated rural settings would go extinct. “You’d need a fair number of people and a fair number of contacts to sustain the transmission,” he says.
As spillovers happen more frequently, though, the odds of sustained human-to-human transmission increase—particularly as more people move between the farms and urban settings like wet markets.
A country at the crossroads
For COVID-19, the first infected person could have been a middleman who collected animals from multiple farms and delivered them to markets and warehouses in the city, a common practice in China and Southeast Asia, says Frutos. This person could have then infected some of the vendors at at Huanan and other locations in Wuhan, a theory that is consistent with Gao’s finding that SARS-CoV-2 was present in the environment of multiple markets and associated warehouses in the city.
Crucially, says Frutos, such scenarios do not require persistent widespread infection in animals because the progenitor virus might have acquired pandemic potential only after evolution and selection in humans.
This hypothetical scenario underscores a growing realization that people might be infected with new pathogens more frequently than we previously recognized. “Most of those infections would go unnoticed or misdiagnosed,” Rosenberg says.
In two separate studies, scientists at the Wuhan Institute of Virology and their colleagues found that up to 4 percent of people living close to bats and working closely with wildlife in southern China were infected with potentially dangerous animal viruses, including coronaviruses; the infection rate was 9 percent among butchers, though none developed severe symptoms.
Such frequent spillover events were likely the sparks that ultimately caused the SARS outbreaks and the COVID-19 pandemic, says Frutos.
Scientists admit that they don’t have all the answers and may never know for sure where COVID-19 came from, not unlike many other emerging infectious diseases. But we know enough to know that wildlife farming in China continues to pose a grave threat to global public health, says Holmes.
In addition to banning the consumption of wild animals, Holmes says China should step up biosafety measures across the wildlife supply chain by regularly surveying farmed wildlife and their handlers for infection. Such surveillance, he adds, should pay particular attention to animals such as minks and raccoon dogs, which are known to be susceptible to coronavirus infection and will continue to be farmed in large quantities in China for their fur.
China is by no means unique; many other nations practice wildlife farming and trade in wet markets. Still, many scientists say the country is now at a crossroads. If it fails to put its house in order, says Li, “COVID-19 is unlikely to be the last pandemic coming out of China.”