Misinterpretations of study lead to conspiracy theories

SciCheck Digest

A study showed a type of lab mouse is highly susceptible to a coronavirus derived from pangolins, a scaly, cat-sized mammal. This doesn’t mean the virus is dangerous to humans. The virus is related to the one that causes COVID-19 but did not descend from it, contrary to claims that it is a “mutant COVID-19 strain.” Nor did scientists “craft” the virus.

Full Story

Biologists sometimes work with lab mice engineered to have human-like tissues, cells or genes. Researchers studying viruses may use mice that have been genetically modified to have human receptors on their cells that allow entry of viruses that infect humans.

While these “humanized” mice can give insights into viruses and what treatments or vaccines might work against them, the mice are not that similar to humans. A virus that kills a humanized mouse will not necessarily be dangerous to people.

A recent study of a version of a pangolin virus in one of these modified mouse models has been misrepresented. Pangolins are mammals prized in Asia for their meat and unusual scales. A preliminary version of the study was posted Jan. 4 as a preprint that has not been peer-reviewed. The researchers posted an updated preprint on Jan. 21 in response to widespread misinterpretations of their work.

“Chinese lab crafts mutant COVID-19 strain with 100% kill rate in ‘humanized’ mice: ‘Surprisingly’ rapid death,” said the headline of a Jan. 16 New York Post story.

“Chinese scientists ‘create’ a mutant coronavirus strain that attacks the BRAIN and has a 100% kill rate in mice – as they admit there’s a ‘risk it spills over to humans,’” read the headline of a Daily Mail story published the same day. Various versions of these claims have spread widely on social media.

In reality, the researchers looked at GX_P2V, a virus found in pangolins. When they infected four mice modified to produce certain human receptors, the virus killed the mice. But co-author Lihua Song, a researcher at Beijing University of Chemical Technology, clarified that the study did not mean the virus was dangerous to people.

The mice used in the experiments “are unique and do not exist in nature,” Song wrote in a Jan. 17 comment on the preprint server where the study was posted. “The outcomes from these tests cannot be applicable to humans.”

There are many different types of coronaviruses. The coronavirus used in the study is in the same family as SARS-CoV-2, the coronavirus that causes COVID-19, but it is incorrect to say that the virus in the study is a “mutant COVID-19 strain,” as it is not descended from SARS-CoV-2.

Nor did the authors of the study “craft” or “create” the version of GX_P2V they used to infect the mice. The virus used in the study was not engineered by scientists and had been previously described. The researchers explained in the updated preprint that the new findings do not alter their fundamental impression that the virus is relatively weak, or attenuated.

Mouse Characteristics Explain Why Virus Was Lethal

The researchers hypothesized in the updated preprint that GX_P2V had proven so lethal in the mice because they had been engineered to be unusually susceptible to infection in their brains. They noted that humans or normal mice, for that matter, would not be expected to be similarly susceptible.

To enter cells, viruses need to glom onto specific receptors. GX_P2V, like SARS-CoV-2, enters cells using ACE2 receptors. Many different types of animals have ACE2 receptors. To better understand viral infection in humans, researchers sometimes engineer mice to produce human ACE2 receptors.

In a previous study in mice with human ACE2 receptors, GX_P2V had limited ability to sicken the mice. But the mice used in the new study had been engineered to produce large quantities of human ACE2 across multiple tissues, including the brain, according to the preprint. ACE2 levels are lower in human brains, the researchers wrote.

The mice in the study “are cranking out massive levels of ACE2 on pretty much every cell in the body so they are getting infected with much higher levels of virus in more organs than a human would,” said Angela Rasmussen, a virologist at the University of Saskatchewan’s Vaccine and Infectious Disease Organization, in a Jan. 19 thread on X, formerly known as Twitter. The animals died because they had been engineered “to support massive virus growth,” she said.

Researchers Did Not ‘Craft’ New Virus

GX_P2V was originally isolated from a pangolin — which had been seized in an anti-smuggling operation — in 2017 by a different group of researchers. The virus was described in a 2020 paper published in Nature. The researchers grew it in cells, but the version this yielded — dubbed GX_P2V(short_3UTR) — was slightly different from the version originally isolated from the pangolin.

Rasmussen explained on X that it is common for an RNA virus like GX_P2V to change as it grows in cell culture. “It’s not unexpected and usually is attenuating,” she said, meaning the viruses become less virulent.

The virus used in the study did not result from engineering or any kind of intentional manipulation, but rather “occurred in the normal course of isolating this virus through classic virological techniques,” she said.

Song, the preprint co-author, told FactCheck.org via email that he and his colleagues were in fact trying to investigate whether the GX_P2V variant could itself be used as a vaccine to protect broadly against SARS-CoV-2 strains. There has been interest in developing vaccines that would protect against a greater variety of SARS-CoV-2 variants, including future variants.

Song said that the GX_P2V variant appeared promising as a vaccine candidate because in prior research it had been identified as “highly attenuated across various animal species,” meaning that it did not significantly sicken the animals.

Song said the researchers gave the mice the virus primarily to see what kind of immune response it generated. He said the discovery that the virus was lethal to the modified mice “was unforeseen” and presented new ideas for how the modified mice and the virus could be used in research.

Researchers could vaccinate the mice with a prospective COVID-19 vaccine and then expose them to the GX_P2V variant, he suggested. This could help them assess whether the vaccine could provide broad protection. Song also said the model is unique because the virus replicated and killed the mice without causing the inflammation that comes with a SARS-CoV-2 infection. He said researchers could use the model to test how well antiviral drugs suppress viral replication.

In the original preprint, the researchers had written that the work “underscores a spillover risk of GX_P2V into humans.” They removed this statement in the subsequent version.

Song said the original phrasing was based on the thought that the experiment “corroborated earlier findings that GX_P2V can indeed utilize human ACE2 for infections, which led me to suggest a potential, albeit theoretical, risk of transmission into humans.”

But, he said, the phrasing “unintentionally misguided readers into believing there was a potential risk of the virus spilling over into human brains and causing 100% fatality, which is not accurate.”

Furthermore, he explained, scientists who read the preprint pointed out there currently isn’t empirical evidence indicating spillover risk to humans. “While it has been demonstrated in prior research that this virus can bind to the human ACE2 receptor, assessing spillover risks involves a broader evaluation than just the receptor interaction,” he said.

Misrepresentations Morph Into Conspiracy Theories

As we’ve said, the authors of the preprint did not create a new virus or cause a virus to become more harmful.

Despite these facts, posts spun unsubstantiated theories about a new disease. Some posts mashed up claims about the mouse study with references to Disease X — a placeholder name for potential future pandemic threats that has been co-opted to support conspiracy theories.

“There’s gonna be a new covid -19 stran coming soon called dease-x It 100% kills you with 8 days !!!” read one post, mixing up multiple unsupported claims.

Other posts referred to the work as gain-of-function research. “So the gain of function research is not just a gain of function for the coronavirus, it’s a gain of function for the totalitarian Empire virus,” read one post, arguing that the creation of new viruses “consolidates the power of State.”

Song denied that his study constituted gain-of-function research in a comment on the preprint. “There have been some folks trying to misinterpret our work as gain-of-function research,” he said. “Let me be clear – that is not the case.”

There are various definitions of gain-of-function research, which most broadly just refers to research in which an organism gains some new ability. More narrow definitions attempt to focus on a subset of gain-of-function research that could be risky.

The U.S. government defines one such subset, called enhanced potential pandemic pathogen research, as work that is “reasonably anticipated to create, transfer or use potential pandemic pathogens resulting from the enhancement of a pathogen’s transmissibility and/or virulence in humans.”

Rasmussen agreed on X that Song’s study was not gain-of-function research, referring to the latter definition. This was both because the GX_P2V variant was not engineered or produced intentionally, and because the virus “didn’t cause much disease in hamsters” and couldn’t be “reasonably anticipated” to cause severe disease.

She said that it is valid to have varying opinions of risks posed by virus research but objected to fearmongering and unsupported criticisms.

“The reason why this was so deadly in these particular mice is because they are engineered to support massive virus growth,” Rasmussen said. “There is a gain of function in the mice—high levels of human ACE2 everywhere—not the virus.”

Editor’s note: SciCheck’s articles providing accurate health information and correcting health misinformation are made possible by a grant from the Robert Wood Johnson Foundation. The foundation has no control over FactCheck.org’s editorial decisions, and the views expressed in our articles do not necessarily reflect the views of the foundation.