Josh Rogin - Chaos Under Heaven

In May, Chinese CDC officials declared on Chinese state media that they had ruled out the possibility that the seafood market was the origin of the virus, completely abandoning the original official story. But Beijing now claimed that neither the market nor the lab was the source. At the time of this writing, that is still their official position.

As for the “bat woman” herself, Shi didn’t think the lab accident theory was so crazy. In her March interview, she described frantically searching her own lab’s records after learning of the coronavirus outbreak in Wuhan. “Could they have come from our lab?” she recalled asking herself.

Shi said that she was relieved when she didn’t find the new coronavirus in her files. “That really took a load off my mind,” she said. “I had not slept a wink in days.” Of course, if she had found the virus, she likely would not have been able to admit it, given that the Chinese government was going around the world insisting the lab had not been involved in the outbreak.

Playing with Fire

A key argument of those Chinese and American scientists disputing the lab accident theory is that Chinese researchers had performed their work out in the open and had disclosed the coronavirus research they were performing. This argument was used to attack anyone who didn’t believe the Chinese scientists’ firm denials their labs could possibly have been responsible for the outbreak. But one senior administration official told me that many officials in various parts of the US government, especially the NSC and the State Department, came to believe that these researchers had not been as forthcoming as had been claimed. They had evidence that Chinese labs were performing gain-of-function research on a much larger scale than was publicly disclosed, meaning they were taking more risks in more labs than anyone outside China was aware of. This insight, in turn, fed into the lab accident hypothesis in a new and troubling way.

A little-noticed study was released in early July 2020 by a group of Chinese researchers in Beijing, including several affiliated with the Academy of Military Medical Science. These scientists said that they had created a new model for studying SARS-CoV-2 by creating mice with humanlike lung characteristics by using the CRISPR gene-editing technology to give the mice lung cells with the human ACE2 receptor—the cell receptor that allowed coronaviruses to so easily infect human lungs. The paper noted that “the lack of vaccine and antivirals has brought an urgent need for an animal model,” and recorded that—although none of the animals died during the study—the mice displayed signs of infection similar to those seen in human victims of COVID-19. The researchers therefore asserted that their mice model was “a useful tool” for research into the virus, the disease it causes, and its potential cures.

Buried in the Chinese report was a huge clue that these scientists were hiding information about their past work on the issue, the senior official told me. The Beijing scientists had studied some mice that were thirty weeks old at the time they were infected with SARS-CoV-2, according to the paper; as would be expected, these older mice suffered worse symptoms from COVID-19 than their baby mice counterparts. But in order for the mice to be thirty weeks old at the time of infection, the mouse model the scientists said they created would have had to been developed months before the SARS-CoV-2 virus was ever publicly identified.

Senior US officials who were briefed on this study were puzzled. Creating mice models to use for coronavirus testing was a fairly common practice, even mice that were “humanized” by giving them the ACE2 receptor. But these Chinese scientists didn’t report that they had been developing such mouse models for studying SARS-like coronaviruses before the SARS-CoV-2 outbreak. Considering that human ACE2 mouse models had been in use in coronavirus research for many years, the Chinese scientists may have had such models already in use. But their lack of such an explanation suggested they weren’t being transparent about their prior research. “It’s plausible they would have these mice already kicking around, but it’s not what they say. It may be true, but it’s not what they say,” the senior administration official told me. “So what’s going on here? Why wouldn’t they have just said, we are using these mice for other things we are looking at?”

After consultations with experts, some US officials came to believe that this Beijing lab was likely conducting coronavirus experiments on mice fitted with ACE2 receptors well before the coronavirus outbreak—research that they hadn’t disclosed and continued not to admit to. That, by itself, did not help to explain how SARS-CoV-2 originated. But it did make clear to US officials that there was a lot of risky coronavirus research going on in Chinese labs that the rest of the world was simply not aware of. “This was just a peek under a curtain of an entire galaxy of activity, including labs in Beijing and Wuhan playing around with coronaviruses in ACE2 mice in unsafe labs,” the senior administration official said. “It suggests we are getting a peek at a body of activity that isn’t understood in the West or even has precedent here.”

The Beijing study further reinforced the suspicions of many people inside the US government that the pandemic resulted in part due to the actions of humans, specifically Chinese researchers. The virus itself may not have been engineered, but the animal hosts that were being used to test it were engineered, which could explain how the virus might have evolved over a short period of time from something found in nature to something so deadly to humans that it would cause the worst pandemic in modern history. “If SARS-CoV-2 arose from gain-of-function research in a lab, it is entirely plausible, even likely, that an animal model involved in that work would have been humanized ACE2 mice,” said Richard H. Ebright, a microbiologist and biosafety expert at Rutgers University. “There is clear plausibility, there is not evidence. To get evidence would require either a confession or an investigation.”

There was, at least, evidence that the Wuhan scientists were using mice with the same engineered mutations as the Beijing scientists. According to Ebright, EcoHealth Alliance and its subcontractor WIV explicitly had proposed using humanized ACE2 mice in the proposal for a grant they received from the NIH in 2014, which was granted through the EcoHealth Alliance, run by Peter Daszak. In June 2020, Shi’s lab released its own study of SARS-CoV-2 using humanized ACE2 mice, which Shi said she obtained from the lab of Ralph S. Baric, a researcher at the University of North Carolina who had worked with the WIV on bat coronaviruses and gain-of-function research for several years.

If the Wuhan researchers were using humanized mice models with the ACE2 receptor to study coronaviruses for years before the 2020 outbreak, it would raise the distinct possibility that they had included in their studies one virus in particular: BtCoV/4991, also known as RaTG13—the closest known relative to SARS-CoV-2. It is not hard to imagine how a combination of gene-edited hosts and gain-of-function research methods could have resulted in a novel coronavirus that was uniquely adapted to infect human cells through the human ACE2. But the WIV never published any research on its work with RaTG13, which struck many scientists as odd, considering it was found in the same general location as where the miners got sick.

Consider, too, that during much of the period the WIV scientists did most of their coronavirus research in labs that did not have the most stringent safety standards. The WIV had seventeen BSL-2 labs and only two BSL-3 labs, information the institute later tried to erase off its website. Their BSL-4 lab, the highest level of biosafety, wasn’t open for business until 2017.