Could scientists use the bat coronavirus RaTG13 to engineer SARS-CoV-2, the virus that causes COVID-19, in a lab?

One of the biggest outstanding questions of the COVID-19 pandemic is the origin of SARS-CoV-2, the coronavirus responsible for the disease. One idea, put forward by some news sites like Fox News and government officials, is that SARS-CoV-2 was man-made. As very few living people have witnessed a pandemic of this scale before, it is natural to wonder whether this could happen without human intervention. Furthermore, the presence of a research laboratory in Wuhan, China, the epicenter of the pandemic and known for its research on coronaviruses, seemed a troubling coincidence. It is important for such a hypothesis to be considered, investigated, and evaluated along with other hypotheses about the origin of SARS-CoV-2.

Although now refuted, various arguments were put forward to demonstrate that SARS-CoV-2 was man-made, such as the presence of genetic sequences from HIV or the presence of small artificial genetic sequences called pShuttles. However, according to virology research, the most likely origin of SARS-CoV-2 is zoonotic, meaning the virus jumped from an animal host to humans^[1,2,3]. The animal host for SARS-CoV-2 has not been identified yet. Health Feedback previously summarized the evidence for and against different hypotheses about the origins of SARS-CoV-2, including that it arose in nature or was man-made.

However, a different form of the hypothesis that SARS-CoV-2 was man-made also circulated on social media and in various articles. This hypothesis involves the claim that researchers used other coronaviruses as templates to engineer SARS-CoV-2. Specifically, this hypothesis suggests that scientists modified a bat coronavirus to produce SARS-CoV-2 based on the genetic similarities between SARS-CoV-2 and some bat coronaviruses.

For instance, a preprint published by virologist Li-Meng Yan claimed that two other bat coronaviruses, ZC45 and ZXC21, provided the genetic backbone for scientists to artificially create SARS-CoV-2. In support of this claim, Yan pointed to the 100% identity shared in some parts of the genomes of SARS-CoV-2 and the bat coronaviruses. As Health Feedback previously explained, virologists refuted this hypothesis and highlighted that the genetic sequences of ZC45 and ZXC21 are very different from that of SARS-CoV-2.

This hypothesis was discussed again by Fox News in a segment aired on 28 February 2021. Steve Hilton noted that a bat coronavirus, RaTG13, is currently the closest known virus to SARS-CoV-2 and shares 96.2% genetic identity with SARS-CoV-2^[4]. Fox News claimed that the 4% of genetic difference “is in the exact places where gain-of-function techniques would be used to make the virus more contagious. In technical language: the Spike receptor binding domain and furin cleavage site”.

The Spike receptor binding domain (RBD) is part of the Spike protein located on the surface of SARS-CoV-2, which enables the virus to latch onto its target cells and infect them^[5]. The molecular structure and composition of the Spike RBD is crucial because it dictates the infectivity of the virus^[6]. The emergence of several SARS-CoV-2 variants in the course of the pandemic in the UK and South Africa is tied to specific changes in the Spike RBD.

The furin cleavage site is a specific molecular structure within the RBD that improves the virus’ ability to penetrate the target cells, making it more infectious^[7].

The so-called gain-of-function experiments that Fox News referred to are genetic techniques used in microbiology to alter specific functions in viral or bacterial strains that were not present in the original strain, such as increasing yields or reducing replication abilities. According to the U.S. Department of Health and Human Services, a gain-of-function study “improves the ability of a pathogen to cause disease”.

In essence, Fox News suggested that researchers replaced the original Spike RBD from RaTG13 with a new, man-made RBD equipped with a furin cleavage site, resulting in that 4% genetic difference and giving birth to the highly-infectious SARS-CoV-2.

However, this hypothesis is scientifically unsound for several reasons that we address in this article. Specifically, we explain why the hypothesis that other coronaviruses served as a template for engineering a new virus is unlikely and how it conflicts with what we know about these viruses and the current state of genetic engineering technology.

The bat coronavirus RaTG13 couldn’t be a template for SARS-CoV-2; the genetic gap between the two is too wide to be bridged by engineering

Health Feedback reached out to several experts in virology, who unanimously rejected the hypothesis that the bat coronavirus RaTG13 served as a template to engineer SARS-CoV-2.

Firstly, David Robertson, a professor of viral genomics at the University of Glasgow, explained that it is inaccurate to claim that the 4% of genetic difference between the two viruses are entirely confined to the Spike RBD.

“The replacement of RaTG13 RBD with SARS-CoV-2 RBD would still be a relatively divergent virus from SARS-CoV-2. This is because there are other mutations in RaTG13’s genome that make it distinct from SARS-CoV-2,” he said.

In other words, there are genetic differences in several parts of the genome^[4,8]. Therefore, simply swapping out the original RBD of RaTG13 for a new man-made RBD would not produce the genomic sequence of SARS-CoV-2, since differences in the virus’ genetic code would still remain in other parts of its genome.

Robert Garry, a professor of microbiology at the University of Tulane, concurred: “While 96% sounds close, in evolutionary terms, it is quite distant, and it would take decades of evolution for the genome of RaTG13 to resemble that of SARS-CoV-2. The difference is about 1,200 bases or 400 amino acids. Gain-of-function research cannot close that gap.” He added, “This would require a virus much closer than RaTG13, at least 99% similar or more likely 99.9% similar”.

The reviewers also highlighted another limitation of this hypothesis. In order for scientists to genetically engineer SARS-CoV-2 from RaTG13, they must have known which Spike RBD to use to replace RaTG13’s original one. However, researchers didn’t discover a Spike RBD similar to the one of SARS-CoV-2 before the pandemic. Given this knowledge gap, there is no plausible scenario to explain how scientists engineered the exact genetic sequence seen in the SARS-CoV-2 Spike protein RBD, since it was unknown to scientists until after the outbreak.

As Stanley Perlman, a professor of microbiology at the University of Iowa, said, “it would not be known in advance what sequence should be used to replace the RaTG13 Spike protein.” Susan Weiss, a professor of microbiology at the University of Pennsylvania countered the hypothesis with the question: “Where would they get the RBD from to insert into RaTG13?”

After the SARS outbreak in 2003, which was caused by another coronavirus, SARS-CoV-1, researchers identified a set of key amino acids within the Spike RBD important for SARS-CoV-1 infectiousness^[9,10]. To improve the infectiousness of a coronavirus, the best engineering strategy would have been to use the amino acid sequences discovered in SARS-CoV-1, as these are known to be efficient and can then be refined to produce an even better molecular design for SARS-CoV-2.

Surprisingly, the current SARS-CoV-2 Spike RBD doesn’t contain this optimal set of amino acids recognized in SARS-CoV-1^[1], yet it is nonetheless able to bind to its target human cells with an affinity even higher than SARS-CoV-1^[11,12]. This finding undermines the claim that SARS-CoV-2 is the result of modifying RaTG13 to give it an enhanced Spike RBD.

There is neither evidence nor a plausible scenario for scientists to engineer SARS-CoV-2 through gain-of-function experiments

The scientists who evaluated this claim also considered a hypothetical scenario in which scientists produced SARS-CoV-2 through gain-of-function experiments, but using a template other than RaTG13. They unanimously deemed it very unlikely that SARS-CoV-2 originated from such gain-of-function experiments.

Commenting on this, Kristian Andersen, a professor of immunology at Scripps Research, said that “there is no way gain-of-function could have created SARS-CoV-2 from RaTG13”.

Robertson stated that “It’s extremely unlikely SARS-CoV-2 was generated by gain-of-function experiments […] Moreover all of the properties of SARS-CoV-2 can be explained by natural processes, such as a mutation and recombination that are well-documented in coronaviruses”.

Furthermore, researchers highlighted that coronavirus manipulation is so difficult that it is extremely unlikely that scientists engineered SARS-CoV-2 through gain-of-function experiments. Weiss said that “it is not easy to design viruses to behave the way you might predict—I believe that SARS-CoV-2 was selected in nature and not designed”. Likewise, Perlman stated, “Finally the whole process of reverse genetics for coronaviruses is difficult, even for experts. So my answer is that it is very unlikely (impossible) for this scenario to be the explanation.”

Conclusion

In summary, the weight of the scientific evidence indicates that the bat coronavirus RaTG13 couldn’t have served as a backbone for engineering SARS-CoV-2. Despite some similarities, there are too many genetic differences scattered across their genomes for RaTG13 to serve as a template for SARS-CoV-2.

In addition, even if one decided to create a coronavirus capable of causing a pandemic, there is no plausible scientific rationale justifying the choice of RaTG13 as a backbone or the design of the SARS-CoV-2 Spike RBD. Scientific experts in virology found it very unlikely that gain-of-function experiments could create a virus similar to SARS-CoV-2. The simplest explanation—the hypothesis that requires the least “ifs” and “maybes”—is that SARS-CoV-2 is the product of natural selection in the wild and was transmitted from animals to humans in a process that has occurred repeatedly throughout history.

Kristian Andersen, Professor, Scripps Research:

Fox News is confusing “Gain of Function Research” and “Basic Research”. The bat research performed at the Wuhan Institute of Virology [of which] EcoHealth was a part, was basic research – and in fact, was instrumental in our ability to respond quickly when SARS-CoV-2 emerged.

There’s absolutely no validity to the claim [that gain-of-function] experiments could have created SARS-CoV-2 – there is no way gain-of -function could have created SARS-CoV-2 from RaTG13.

David Robertson, Professor, University of Glasgow:

• Is it likely that SARS-CoV-2 is derived from RaTG13 by artificially replacing the Spike RBD?

No, the replacement of RaTG13’s RBD with SARS-CoV-2’s RBD would still be a relatively divergent virus from SARS-CoV-2. This is because there are other mutations in RaTG13’s genome that make it distinct from SARS-CoV-2. This was quantified here by Wang et al.^[8]. You’d have to change these other parts of RaTG13’s genome to arrive at SARS-CoV-2’s sequence.

• Is it likely that SARS-CoV-2 is the result of gain-of-function experiments?

It’s extremely unlikely SARS-CoV-2 was generated by gain-of-function experiments. We didn’t know anything about this new coronavirus before it emerged in 2019 so how could it have been designed? What would have been the template for this? It’s just incredibly implausible that some random experiments could have generated a virus with such dramatic properties but so unlike anything we’d observed before.

Moreover all of the properties of SARS-CoV-2 can be explained by natural processes such as a mutation and recombination that are well documented in coronaviruses. We discussed the probable source of the SARS-CoV-2 progenitor here in a study by Lytras et al.^[2].

Susan Weiss, Professor, University of Pennsylvania:

• Is it likely that SARS-CoV-2 is derived from RaTG13 by artificially replacing the Spike RBD?

NO. The RBD is not the only difference between the two genomes. They would have to make many other changes. Where would they get the RBD from to insert into RaTG13? Why would they suppose that changing the RBD of some random bat virus would produce a virus lethal for humans? These viruses are naturally selected to do what they do—just as variants now are selected to optimize stability and spread.

• Is it likely that SARS-CoV-2 is the result of gain-of-function experiments?

I do not think so for similar reasons as above. No diabolic genius could figure out how to design a virus to behave this way. Where would you start—why RaTG13? It makes no sense to me.

I remembered two types of engineering we attempted many years ago with murine coronavirus. In one set of experiments, we constructed viruses with chimeric spike proteins (S1 and S2 from closely related strains and some other combinations). These viruses replicated well in cell culture, but were quite dead in animals. The point being: we tried to engineer coronaviruses from closely related strains and it did not give us pathogenic viruses.

In another case, we exchanged a furin site for a sequence from a closely related strain that did not have a furin site. The new virus replicated well in cultures, but did not cause disease in mice. On the other hand, when we let the virus be selected for persistence in mouse glial cells, we obtained variants with a mutant cleavage site—which arose through natural selection —and they were quite pathogenic in the brain but had lost liver tropism [meaning that the virus could not infect cells from the liver anymore]. My point being: it is not easy to design viruses to behave the way you might predict—I believe that SARS-CoV-2 was selected in nature and not designed.

Robert Garry, Professor, University of Tulane:

RatG13 could not have served as the backbone of SARS-CoV-2. While 96% sounds close, in evolutionary terms, it is quite distant, and it would take decades of evolution for the genome of RaTG13 to resemble that of SARS-CoV-2. The difference is about 1,200 bases or 400 amino acids. Gain-of-function research cannot close that gap. Passage in cell culture or transgenic animals would never create the changes. Replacing the RaTG13 RBD with the RBD of another virus (such as the pangolin coronavirus) certainly would not close the gap. It still leaves you with a virus that is still 96% different from SARS-CoV-2. It is highly unlikely, in fact,near impossible—that SARS-CoV-2 is the result of gain-of-function research. This would require a virus much closer than RaTG13, at least 99% similar or more likely 99.9% similar.

Stanley Perlman, Professor, University of Iowa:

• Is it likely that SARS-CoV-2 is derived from RaTG13 by artificially replacing the Spike RBD?

RaTG13 is 4% different from SARS-CoV-2. This is equivalent to 1,200 nucleotides.
I do not know what fraction of the differences between SARS-CoV-2 and RaTG13 are in the S protein; there are certainly some outside of the [Spike RBD] protein.

In addition, it would not be known in advance what sequence should be used to replace the RaTG13 S protein. Finally, the whole process of reverse genetics for coronaviruses is difficult,
even for experts. So my answer is that it is very unlikely (impossible) for this scenario to be the explanation. It is much more likely that nature did this.

• Is it likely that SARS-CoV-2 is the result of gain-of-function experiments?

This possibility implies that a virus was already in hand that was known to infect human cells. There is no evidence for this. Even if this putative virus existed, it would not be known in advance how to modify it to enhance transmission and virulence. Passage through tissue culture cells generally results in virus attenuation. In fact, that is how the poliovirus vaccine was developed from wild type polio.

READ MORE

Christian Stevens from the Benhur Lee lab at the Mount Sinai School of Medicine has provided a comprehensive explanation of the multiple scientific studies examining the origin of the coronavirus.

REFERENCES

• 1 – Andersen et al. (2020) The proximal origin of SARS-CoV-2. Nature.
• 2 – Lytras et al. (2020) Exploring the natural origins of SARS-CoV-2. bioRXiv (preprint).
• 3 – Boni et al. (2020) Evolutionary origins of the SARS-CoV-2 sarbecovirus lineage responsible for the COVID-19 pandemic. Nature microbiology.
• 4 – Zhou et al. (2020) A pneumonia outbreak associated with a new coronavirus of probable bat origin. Nature.
• 5 – Tai et al. (2020) Characterization of the receptor-binding domain (RBD) of 2019 novel coronavirus: implication for development of RBD protein as a viral attachment inhibitor and vaccine. Cellular & Molecular Immunology.
• 6 – Li et al. (2020) The impact of mutations in SARS-CoV-2 Spike on viral infectivity and antigenicity. Cell.
• 7 – Johnson et al. (2021) Loss of furin cleavage site attenuates SARS-CoV-2 pathogenesis. Nature.
• 8 – Wang et al. (2020) Synonymous mutations and the molecular evolution of SARS-CoV-2 origins. Virus evolution.
• 9 – Wan et al. (2020) Receptor Recognition by the Novel Coronavirus from Wuhan: an Analysis Based on Decade-Long Structural Studies of SARS Coronavirus. Journal of Virology.
• 10 – Wu et al. (2012) Mechanisms of Host Receptor Adaptation by Severe Acute Respiratory Syndrome Coronavirus. Journal of biological chemistry.
• 11 – Wrapp et al. (2020) Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation. Science.
• 12 – Nguyen et al. (2020) Does SARS-CoV-2 Bind to Human ACE2 More Strongly Than Does SARS-CoV? The Journal of physical chemistry B.