Close cousins of SARS-CoV-2 found in a cave in Laos yield new clues about pandemic’s origins

When SARS-CoV-2 was discovered in early 2020, some researchers initially suspected it might have been engineered in a laboratory and accidentally escaped, in part because a region of the viral surface protein that latches onto human cellular receptors had an uncanny fit. “It looked kind of strange, right?” says Edward Holmes, an evolutionary biologist at the University of Sydney. “People were saying, well, maybe that’s been selected in a lab.”

Now, however, the discovery of three coronaviruses in bats living in limestone caves in Laos adds substantial weight to existing evidence that the virus was not engineered, Holmes says. The three viruses, reported in a preprint on Research Square on 17 September, are the closest relatives to SARS-CoV-2 found to date, and they are the first discovered that are nearly identical in that key region. “The core, functional part of SARS-CoV-2 has a natural origin,” says Linfa Wang, a bat coronavirus researcher at the Duke-NUS Medical School in Singapore. “It’s proven.”

Some scientists say the finding says nothing about the so-called “lab-leak” scenario, in which a natural, unaltered virus first infected humans doing field or lab work. “The Laos finding does not move the needle at all,” says Gilles Demaneuf, a data scientist who is part of an internet-based group called DRASTIC that has defended the lab-origin hypothesis from what it sees as unwarranted attacks. “It is perfectly compatible with both hypotheses.”

In evolutionary time, several decades separate these bat viruses—dubbed BANAL, because researchers found them in bat anal swabs—from SARS-CoV-2, so the new viruses could not have sparked the pandemic. But the study further expands the family tree of SARS-CoV-2 and raises new questions about how it may have arisen. And the BANAL viruses could well pose a threat to humans themselves, Wang cautions. “This virus could be SARS-CoV-3,” he says.

For the study, a team of researchers from the National University of Laos working with colleagues from the Pasteur Institute, which has a branch in Laos, sampled 645 bats from four different sites. In the karstic terrain of the Feuang district, they found bats from three different species of the Rhinolophus genus that were infected with viruses up to 96.8% identical in genetic sequence to SARS-CoV-2.

The closest previous isolate, RaTG13, came from yet another bat species, R. sinicus, living in a cave in Mojiang, in China’s Yunnan province. Its similarity to SARS-CoV-2 is 96.2%—only slightly lower than the BANAL viruses—but the number obscures a profound distinction between the new isolates and RaTG13.

SARS-CoV-2 uses its surface protein, spike, to dock onto human cellular receptors known as angiotensin-converting enzyme 2 (ACE2) and initiate an infection. A small portion on the tip of spike called the receptor-binding domain (RBD) plays a lead role in this process. Compared with the RBD on RaTG13, those of the BANAL isolates are far closer in sequence to the one on SARS-CoV-2. What’s more, crystallography of BANAL-236—the one isolate grown in culture, a difficult laboratory feat—showed its structure was identical to the RBD of SARS-CoV-2. And in test tube studies, BANAL-236 readily infected cells that had the human ACE2 receptor. “When we saw that the RBD was very close to SARS-CoV-2 it was very exciting,” says the lead researcher on the project, virologist Marc Eloit of the Pasteur Institute.

The BANAL isolates join a growing list of bat coronaviruses related to SARS-CoV-2, which have also been found in Cambodia, Thailand, and Japan. All are in Rhinolophus, or horseshoe, bats, which do not migrate far but often infect other species that share roosts. That allows coronaviruses to hop, skip, and jump around Southeast Asian countries and China, Eloit and other researchers contend. “This tells us that … they’re very common in Rhinolophous across that part of southern China and Southeast Asia,” Holmes says.

The BANAL isolates, like all close relatives of SARS-CoV-2, are missing a signature component of the pandemic virus known as the furin cleavage site. The human virus’ spike protein has a region near its middle that, when cleaved by furin, a human enzyme, markedly enhances infectivity and virulence.

Furin cleavage sites have been found in bat coronaviruses far away from SARS-CoV-2 on the family tree. But coronaviruses do not need a furin cleavage site to sicken animals. Eloit’s team plans to put BANAL-236 into “humanized” mice and possibly monkeys to see whether it can cause disease. That should offer clues to whether it presents a threat to humans. The studies, which Eloit notes will be done by vaccinated researchers working under strict biosafety rules, may also show the BANAL isolates are not pathogenic, in which case the new virus might even protect from SARS-CoV-2, given its similarity. “This is either like SARS-CoV-3 or a free, live vaccine,” Eloit says.

But the absence of the furin cleavage site in BANAL-236 and other SARS-CoV-2 relatives raises the question of how and when the pandemic virus’ progenitor could have picked up its site. Wang and many other researchers think this may have happened when it swapped genetic material with other more distant coronaviruses that do have a furin cleavage site. “Recombination happens in bats with a high frequency,” Wang says.

Others contend it’s more likely that recombination occurred in some other animal species coinfected with a bat coronavirus and a different one that has a furin cleavage site, which could include anything from canids to rodents to felines. “I doubt it’s gone straight from bats to humans, because that rarely happens,” Holmes says. It could also have happened in humans, Eloit says. Four other coronaviruses circulate in people—they all cause mild disease—and some have furin cleavage sites.

It’s also possible that the furin cleavage site arose without recombination. “This type of virus replicates very fast in humans, particularly if they don’t have any previous immunity, and accumulates a lot of mutations that could generate a furin site,” Eloit says. And while it was accruing these mutations, the virus could have “circulated silently” from one person to another, giving it time to move undetected from a rural area to Wuhan, China, where the pandemic erupted at the beginning of 2020. “I would be very interested in finding any data regarding serological screening in Wuhan before the epidemic,” he says.

The intriguing harvest from Laos makes it even stranger that a similar but vastly larger hunt for SARS-CoV-2 relatives in neighboring China, also reported this month, came up empty.

The study, published on 20 September by Wu Zhiqiang of the Chinese Academy of Medical Sciences and Peking Union Medical College and his colleagues, found no SARS-CoV-2–related viruses in 13,064 bats collected at 703 sites around China between 2016 and 2021. SARS-CoV-2–related viruses “might not actively circulate among bats in China,” the researchers concluded. Wu did not reply to an email from Science requesting an interview.

Scientists are impressed by the size and geographic scope of the study and the fact that it was published at all—very few bat coronavirus studies have come out of China during the pandemic. “I look at it as positive because they tell you that they’re doing something,” Wang says. But he suspects sampling bias may explain why the researchers didn’t find any SARS-CoV-2–related viruses. A single visit to a site often doesn’t yield any interesting isolates, he says: “If you go to a cave 10 times, you’d be lucky to get one.”

Holmes is “very skeptical” about the study, especially because his own group and others have reported finding viruses related to SARS-CoV-2—including RaTG13—in some of the sites the new study sampled. “It’s very hard to know what’s going on in China,” Holmes says. “Anything that relates to the origins this virus is going to be carefully vetted. It’s not clear what we’re allowed to see.”

source: sciencemag.org