Most scientists studying the origins of COVID-19 have concluded that the SARS-CoV-2 virus probably evolved naturally and infected humans via incidental contact with a wild or domesticated animal. But a few persistent voices, including respected microbiologist and biosafety advocate Richard Ebright, PhD, continue to highlight circumstantial evidence suggesting that SARS-CoV-2 escaped from a biohazard laboratory in Wuhan, China. Such an escape might have occurred via accidental infection of a lab worker who came into contact with the isolated virus, an infected lab animal, or animal waste.

For now, we cannot definitively rule out either origin story—a lab accident or a natural animal-to-human transmission. Determining the origins of this pandemic is unlikely to help current efforts to treat and cure the disease. It could, however, be an important determinant for efforts to prevent future outbreaks.

Escape from a lab?

Most arguments in favor of the lab accident theory are based on geography. Wuhan, the site of COVID-19’s first reported cases in late 2019, is also home to two of China’s most advanced biological laboratories. The Wuhan Institute of Virology (WIV) has come under special scrutiny because it hosts China’s only biosafety level 4 lab facilities.

Ebright, a former Howard Hughes Medical Institute investigator and current laboratory director at Rutgers University, is the most prominent scientist espousing the possibility of a lab escape. He raised biosafety concerns about WIV when it opened in 2017, as did Tim Trevan, founder of CHROME Biosafety and Biosecurity Consulting. Ebright has long opposed the expansion of preventative research into dangerous pathogens, and he views the possible lab escape of SARS-CoV-2 as strong support for that position.

Corona virus close up
Source: Radoslav Zilinsky/Getty Images

When asked why he continues pushing the lab accident theory when many of his peers have moved on, Ebright states, “There currently is no scientific evidence and no forensic evidence—absolutely none—that enables a choice between the natural-accident and laboratory-accident scenarios.”

Jamie Metzl, a foreign policy expert and author of Hacking Darwin, is another vocal supporter of the lab theory. “Suggesting that an outbreak of a deadly bat coronavirus coincidentally occurred near the only level 4 virology institute in all of China—which happened to be studying the closest known relative of that exact virus—strains credulity,” he wrote in a July op-ed for the Wall Street Journal.

Group of Greater horseshoe bat (Rhinolophus ferrumequinum)
Horseshoe bats. [AttilaBarsan/Getty Images]

The “closest known relative” to which Metzl refers is a virus known as RaTG13, a coronavirus sampled from horseshoe bats in China’s Yunnan province in 2013. RaTG13 bears 96.2% overall genetic sequence identity to SARS-CoV-2 across the 29,000 bases of the coronavirus genome. But in the region encoding the receptor-binding domain of the virus’s signature spike protein, the sequence identity drops to just 85%.

Virus evolution experts have stated that the sequence divergence between the two viruses represents an evolutionary distance of at least 20 years. Joseph Petrosino, PhD, a metagenomics expert at Baylor College of Medicine, says that “the observation that the RaTG13 backbone component is still many mutations away from SARS-CoV-2 suggests other intermediates likely exist.” Others caution that the evolutionary tree is not sufficiently complete to allow for definitive analysis.

Shi Zhengli, PhD, a well-known Chinese researcher who leads the WIV lab studying RaTG13 and other bat coronaviruses, did not respond to GEN’s requests for comment. However, she recently told Science that her lab was only studying the genome sequence of RaTG13, insisting that they did not isolate the virus itself. She has repeatedly denied that SARS-CoV-2 ever existed in any WIV lab before the current outbreak, and has highlighted WIV’s adherence to safety protocols.

However, accidents do occur, even in labs with excellent safety standards. Ebright cites a USA TODAY investigation that concluded that there were more than 1,100 lab accidents involving dangerous pathogens or toxins in the U.S. alone between 2008 and 2012. The SARS virus escaped from research labs in the early 2000s and caused minor outbreaks in Beijing, Taiwan, and Singapore, prompting the World Health Organization (WHO) to tighten biosafety recommendations for SARS-related coronaviruses. Petrosino says that labs studying these pathogens now have protocols in place to reduce accidental exposures and respond swiftly when they do occur.

The lab escape theory has also found favor among U.S. government officials, including Senator Tom Cotton and Secretary of State Mike Pompeo, who claim that Chinese government actions in early 2020 were part of a cover-up.

China was slow to report initial findings about COVID-19 to international organizations like the WHO, and the Associated Press repeatedly reported that Chinese authorities seized tight control over data and investigations related to the virus and its origins. But journalist Bob Woodward confirmed in a February interview with President Trump that the Chinese premier communicated transparently about how deadly COVID-19 could be. In an April op-ed for the Wall Street Journal, Cotton also cited allegations that Chinese labs studying the novel coronavirus in late December and early January received orders to destroy their samples, although he failed to note that Chinese investigators submitted the novel coronavirus draft genome to a U.S. database as soon as it was sequenced on January 5.

The Office of the Director of National Intelligence confirmed last April that the U.S. intelligence community was conducting an official investigation into the possibility of a lab escape. “There is great risk associated with the absence of transparency, and the Chinese government has yet to sufficiently share data or samples with the international community,” said a Department of State spokesperson in a statement to GEN. “We still don’t have the answers we need about a virus that has left 700,000 dead and counting. For the world to have those answers, Beijing must provide open and transparent access to full information needed to allow for a complete understanding of the origins of the virus.”

The WHO is also planning to conduct its own independent investigation into the origins of COVID-19. Neither U.S. intelligence nor the WHO have thus far released any investigation results that might provide evidence for or against the lab accident theory.

The incidental transmission case

Most scientists remain skeptical of the lab escape hypothesis, concluding instead that SARS-CoV-2 likely evolved in nature and infected humans outside a lab. There are two main reasons. First, the lack of evidence that SARS-CoV-2 was in a Wuhan laboratory is significant, because such evidence should have been available to outsiders unless lab employees engaged in elaborate secrecy protocols. Second, SARS-CoV-2 is able to jump between multiple species, and past epidemics like SARS and avian flu show that animal-transmitted diseases pose a serious pandemic threat.

Massa Shoura, PhD, a pathology researcher at Stanford University, was part of a team that performed a comprehensive search of the Sequence Read Archive (SRA), a public database used by researchers to deposit newly identified DNA and RNA sequences.

“If you publish your entire dataset, a very tiny bit might have signatures of what’s being sequenced in the lab from other sources, because we share machines and equipment,” says Shoura. If SARS-CoV-2 were being studied in a laboratory that deposited any sequence data in the SRA, she says fragments of the virus’s RNA sequence would be present in the database because of cross-contamination—even if the researchers had not yet chosen to label it or publish the full SARS-CoV-2 sequence. However, her team’s search revealed no such traces in the SRA.

“Given that a search of the public database did not yield any sequence that matched SARS-CoV-2 with 100% identity, we can say that this sequence—and by extension, this virus—did not exist in the labs that have deposited data in the SRA,” says Shoura. A lab studying SARS-CoV-2 would need to have operated in complete secrecy, never releasing sequence data from any projects.

Shoura finds that scenario unlikely. Labs at both the Wuhan Center for Disease Control and WIV—including Shi’s bat coronavirus lab—regularly upload sequences of interest to the repository, and it is unclear why they would have chosen to hide the SARS-CoV-2 sequences prior to any escape.

A pangolin roaming the bush for food
Pangolian. [CarlFourie/Getty Images]

Although no conclusive evidence shows any traces of the SARS-CoV-2 either in the wild or in a laboratory before December 2019, multiple known animal viruses share significant similarities with SARS-CoV-2. Several studies have zeroed in on the genomic sequences of coronaviruses in pangolins (a scaly mammal native to multiple regions of Asia) and present a working hypothesis for how SARS-CoV-2 may have emerged from this pool of potential precursors.


A coronavirus found in Malayan pangolins shared 98% similarity with SARS-CoV-2 in the crucial receptor-binding motif, the region where RaTG13 had the most sequence divergence. According to Petrosino, the Baylor metagenomics expert and the principle investigator on one of the relevant studies, “the preponderance of evidence supports recombination events in pangolins and bats that led to the virus we have today.”

This connection does not imply that a pangolin or bat was the host that immediately preceded the virus’s jump to humans. “It is well known that coronaviruses can manage to use different mechanisms to achieve the jump to another species,” says Leo Poon, PhD, head of the division of public health laboratory science at the University of Hong Kong.

SARS-CoV-2 has already demonstrated a high level of flexibility. Scientists have documented infections in multiple species, including humans, cats, dogs, and minks. This ability to travel within the animal kingdom could help explain how the virus made its way to Wuhan even though horseshoe bats hibernate in winter and do not have major colonies near the city.

“There are so many viruses in those bat caves that are exchanging parts. Why would we expect to find the exact same virus?” asks William Haseltine, PhD, a veteran infectious disease expert and president of ACCESS Health International. “There are at least five different plagues of this virus circulating right now. I don’t see that as a mystery. I see that as expected.”

Haseltine says his own experiences researching and observing animal-borne viruses such as HIV, Ebola, MERS, SARS, and avian flu have shaped his theories about COVID-19’s origins. “We know that a number of pandemics originate in animals and get into humans. It happens because of the way we live now,” he says. “We provide a brand new ecological niche for them.”

As urbanization and climate change cause ever more frequent clashes between human societies and wild ecosystems, Haseltine and other scientists warn that the chances of animal viruses making the cross-species jump to humans are continually growing.

Moving forward

For now, the world will continue fighting COVID-19 without knowing where it came from. “My gut feeling is that it will take a long time for scientists to address [these] questions,” says Poon, the Hong Kong public health expert. “Believe it or not, although we all believe SARS in 2003 was caused by a bat virus, we still cannot find SARS-CoV in bats in the last 17 years.”

Nevertheless, scientists say the answer is well worth pursuing in the long run. “Whether it originated from a wild exposure, food market, or lab is really inconsequential for how we treat and try to put an end to the pandemic. But it’s important from an origin story perspective, so we can try to stop these things from happening in the future,” says Petrosino.

All of the scientists who spoke to GEN brought up the need for increased preventative measures. Ebright emphasizes tightening biosecurity standards; if SARS-CoV-2 escaped from a lab, understanding how it did so would reveal relevant gaps in lab safety protocols. He says a credible forensic investigation to obtain this information will require an independent organization to gain access to facilities, samples, and records at Wuhan labs, and would involve comprehensive sampling and interviews with all personnel including construction workers and janitorial staff.

There is also widespread support for preemptive virus surveillance programs. If SARS-CoV-2 jumped from animals to humans outside a laboratory, pinpointing the setting in which that transmission occurred could provide a focus for surveillance and other preventative measures.

In late August and early September, the U.S. announced renewed funding for programs that will develop research centers across the globe aimed at identifying animal viruses that could spill over into human populations. It’s a start, but Shoura says it’s not enough.

“There’s no shortage of people who are willing to go to caves and deal with bats and isolate viruses and work on them. There’s no shortage of labs that are interested in dedicating time and resources,” she says. “The shortage is in the resources and the diversity of the people who receive these resources.”

Petrosino offers a hopeful note, pointing out that researchers all over the world have been conducting rigorous surveillance of SARS-CoV-2 since it emerged, sequencing patient and animal samples in order to understand transmission, mutation, and trends in infectivity. The same strategies and technologies would provide the foundation of a robust global zoonotic surveillance problem. “We’re not as far away as you might think,” he says.

This site uses Akismet to reduce spam. Learn how your comment data is processed.