SARS-like virus engineered in the lab gives insight into possible disease threat

Researchers have assessed the threat of a SARS-like virus, found naturally in bats, by ‘reverse engineering’ the virus in the lab. The team engineered two versions of the WIV1 virus, finding that human airway cells were easily infected in the lab. The authors, publishing in PNAS, conclude that WIV1 poses a threat to emerge as a human disease. They also note that their method of engineering viruses could be used for gain-of-function research, the subject of a current moratorium in the US, but that such experiments would provide useful data for understanding potential emerging diseases.


Dr. Michael Imperiale, Professor, Department of Microbiology and Immunology, University of Michigan Medical School (webpage):

Expertise: research in human polyomarvirus, Dr. Imperiale has previously written on the risks and benefits of gain of function research which involves modifying the ability of an infectious disease agent to cause disease to improve understanding

“I think this is an important study because it addresses the potential ease with which a virus might be able to jump from bats into humans without further mutation. It seems that the results indicate that the threat is greater than zero. I’m not sure how one quantifies it, though.

“I am not concerned about the engineered virus, i.e., the virus in which they moved the WIV1 spike protein into the mouse-adapted SARS background. This appears to have allowed them to answer the important question of whether the WIV1 spike protein can mediate mammalian infection without additional mutation. As such, it means that we need to be alert to the possibility that the virus can make a direct jump.

“As it pertains to gain of function research in general, we must first ask, is this an important question that needs to be addressed? Is it helpful to know how easy it might be for a new coronavirus to jump into humans? I think the answer is undoubtedly yes because it influences how public health officials approach surveillance, and it indicates that we must keep working diligently towards new vaccines and therapeutics because the threat is real. Second, are there other ways to address this question? My sense is that the authors’ approach is the best approach in this case. If the work is performed safely and responsibly, one can make the argument they do, namely that there may be greater risk if the biological questions remain unanswered.”


Dr. Stanley Perlman, Professor, Departments of Microbiology and Pediatrics, University of Iowa Carver College of Medicine (website):

Expertise: pathogenesis of murine and human coronaviruses

“The authors in this study show that WIV1-CoV, a SARS-like bat coronavirus grows as well in human cells as a human strain of SARS-CoV. However, the virus grew less well in mice than the human strain, suggesting that additional genetic changes were required for widespread infection of humans.

“I think that the methods are appropriate and mostly justify the conclusions of the authors. However, the wording in the title is somewhat misleading and conflicts with the conclusions drawn in the manuscript. In the text, the authors conclude that WIV1 will require significant mutation before it robustly infected humans. By extension, this virus is not likely to cause widespread outbreaks without extensive changes.”

“In their discussion, the authors address the important and complex issue of gain of function studies in which pathogens are altered in ways that may increase their transmissibility or pathogenicity. I agree that understanding factors that might increase transmissibility is important but it is also possible that this could create pathogens with increased transmissibility or virulence. That being said, this has not occurred. In the influenza A virus studies conducted in ferrets a few years ago, viruses that demonstrated increased transmissibility became attenuated. In conclusion, i think that caution is warranted and careful consideration by oversight panels is warranted, at least until we better understand the risks.”


Dr. Susan R. Weiss & Stephen A. Goldstein, Perelman School of Medicine, University of Pennsylvania (webpage):

Dr. Weiss is a Professor of Microbiology. Stephen A. Goldstein is a student Dr. Weiss’ laboratory.

Expertise: coronavirus pathogenesis, organ specific virus- host interactions, Middle East Respiratory Syndrome

“We’ve known that a large pool of SARS-like coronaviruses is circulating in Chinese bats, but until recently it has been unclear whether these posed any threat to human health. The authors who recently published a similar paper in Nature Medicine, have now shown that at least two of these viruses replicate in human lung cells equivalently to SARS itself. The study is both scientifically interesting and informative for pandemic preparedness.

“The public health significance of the work lies largely in that it identifies protective antibodies that can now be screened against other, related spike proteins important for infection and be included in a currently sparse toolkit for responding to SARS-like coronavirus outbreaks and epidemics.

“The authors’ conclusions are solidly supported by the data, and they are clearly aware of and discuss the inherent caveat of using mouse models. With respect to human health, we can take away from this study that WIV1 replicates equally well as SARS in human cells, and can be inhibited by antibodies that block SARS infection. We cannot infer much about the potential pathogenicity of WIV1 in humans, and would not discount it as a pandemic threat based on this study.

“Whether WIV1 will spill over depends largely on the frequency of contact between infected bats and humans and/or the potential to spread to an intermediate host with more contact with humans. Of course, to cause any significant outbreak WIV1 also has to be transmissible in humans, which we have no way of evaluating.

“WIV1 is a naturally circulating virus in wild bats, so constructing it in the lab does not pose any novel risk. Any lab-constructed chimeric virus (containing genetic material from more than one virus) is likely to be less well-adapted to humans than a virus that has been shaped by naturally occurring evolutionary pressures. The inherent unknowns contain some risk, but what we do know about virus biology does not suggest that these experiments introduce any outsized risk. Instead, we have gained new information about the diverse threats posed by naturally circulating coronaviruses, and about our ability to mitigate their potential impact.”


Dr. Michael Buchmeier, Professor, Division of Infectious Disease, Department of Medicine, University of California, Irvine (webpage):

Expertise: coronaviruses, viral pathogenesis, biology of zoonotic and emerging viruses, biodefense. Dr. Buchmeier is a member of the NIH RAC subcommittee on biosafety, and a former member of the RAC.

“Two papers have been published about these groups of viruses in the recent past including one very similar to the current study, which was done by the same research team. I don’t think the results in this paper justify the conclusion of the authors that WIV1 is an emerging threat. There are multiple factors that may contribute to pathogenesis of a virus like WIV1. While the experiments in cultured human airway epithelial cells are convincing, caution must be expressed in extrapolating that finding to the ominous statement that the virus is “poised for human emergence”.

“The authors are not really talking about human infection. They are talking about receptor binding, saying that WIV1, a SARS-like bat coronavirus binds to a human protein receptor, ACE2 and that this might facilitate infection of human cells. However their mouse studies show the expression of human ACE2 in the mouse is not sufficient to increase the pathogenic potential of the WIV1 virus in mice.

“This kind of research, reconstructing viruses in order to understand them, is quite important and this group has done excellent work in that field. However, in this particular study, the analysis presented is quite superficial. The authors do acknowledge the shortcomings in the discussion but do not emphasize it strongly enough to balance the more declarative title.

“What I would accept from this paper is that there are SARS-like coronavirus strains out there that show specificity to human receptor proteins; something that we already knew based on the paper published earlier by Ge et al., in Nature.

“The authors present a case for allowing gain of function (GOF) research in the discussion. The 1918 flu virus became a hot button issue when researchers made changes to the virus that made it more virulent. The moratorium that ensued on this type of GOF research was as much about a breach of etiquette on how the gene sequences were supposed to be used as a fear of recreating the 1918 flu epidemic. From my perspective the gain of function experiments have a place in research on the pathobiology of influenza, and the moratorium was necessary to allow a chance to stop and recalibrate how the regulations applied, so regulators did have some control over what was going on. I’m not as concerned about GOF studies as I am about the issues surrounding the intentional creation of viruses that may not exist in Nature by genetic engineering.”


Dr. Amesh Adalja, Senior Associate, Center for Health Security, Adjunct Instructor, Division of Infectious Diseases, Department of Medicine, University of Pittsburgh Medical Center and School of Medicine (webpage): 

Expertise: infectious diseases, critical care, biosecurity

“This research is very important as it is focused on a viral group that is a major concern for those preparing for the next emerging infectious disease outbreak: the coronaviruses, of which SARS and MERS are infamous members. Because of the trajectory of SARS and MERS, it has become vital to assess if another member of this diverse viral family will spill into humans and spark a pandemic. The methods used here involved finding another coronavirus that uses the same receptor protein for human infection as SARS, assessing its ability to infect human cells, and its susceptibility to SARS vaccines and antibodies. Both the original virus WIV1-CoV and a chimeric SARS-WIV1 containing genetic material from both SARS and WIV1 virus were tested. The authors found that both viruses were able to infect human cells and evade an experimental SARS vaccine. However these viruses were less severe in a mouse model. It is important to remember that these are experimental studies and may not be fully extrapolatable to human settings, but are notable in their own right.

“Because WIV1 has the capacity to infect human cells and coronaviruses as such are a viral group which has several members that can infect humans, WIV1 merits closer attention as potential coronavirus that could spill into humans in the future.

“Any time one is working with a potential pathogen, biosafety norms and procedures should be followed. This experiment was approved by the relevant university regulatory committee and appears to have been conducted appropriately. The risk of gain-of-function research, to me, is usually outweighed by the benefits of understanding pandemic potential, informing preparedness and response, and the greater scientific knowledge acquired.”



Declared interests (see GENeS register of interests policy):

Dr. Susan R. Weiss:  I collaborate with the senior author Dr. Ralph Baric on a project unrelated to the current study and we were co-authors of a mouse hepatitis virus study published last year.

No further interests declared



“SARS-like WIV1-CoV poised for human emergence” by Vineet Menachery et al. published in Proceedings of the National Academy of Sciences on Monday 14 March 2016.

Please feel free to leave your comments below, but be aware that by doing so you agree to our Terms & Conditions.