When researchers at Boston University (BU) in Massachusetts inserted a gene from the Omicron variant of SARS-CoV-2 into a strain of the virus from the beginning of the pandemic, they were trying to understand why Omicron causes mild disease.
But the experiments, described in a 14 October preprint1, have ignited a red-hot controversy over what constitutes truly risky SARS-CoV-2 research — especially now that much of the world’s population has some immune protection from the virus and COVID-19 treatments are available.
At issue is whether — and when — researchers modifying SARS-CoV-2 or other deadly pathogens need to keep regulators and funding agencies such as the US National Institutes of Health (NIH) informed about their work, even if the agencies didn’t fund the experiments in question. Studies that make pathogens more transmissible or virulent are sometimes called ‘gain of function’ research.
The controversy sparked by the BU study highlights “the lack of clarity that people have on exactly what sorts of experiments have benefits that outweigh risks, and who decides how it’s all reviewed”, says Jesse Bloom, an evolutionary virologist at the Fred Hutchinson Cancer Center in Seattle, Washington.
“Some guidance is really needed,” says Pei-Yong Shi, a virologist at the University of Texas Medical Branch at Galveston, whose team is seeking permission from the NIH to study whether SARS-CoV-2 can develop resistance to antiviral drugs the group is developing.
The brouhaha over the BU research started after a team led by Mohsan Saeed, a virologist at BU’s School of Medicine, posted a preprint1 on bioRxiv showing that the properties of Omicron’s spike protein — the part of the virus that allows it to infect human cells — might not explain the clinical mildness of the COVID-19 cases it causes. Saeed’s team had created a new strain of SARS-CoV-2 by putting the spike protein from the Omicron BA.1 lineage into the backbone of a viral strain isolated in the early days of the pandemic.
Unlike BA.1, which usually causes mild, non-fatal disease, this strain caused severe disease in mice engineered to be susceptible to SARS-CoV-2 infection. Eight of the ten mice exposed to the strain died or had to be killed as a result of weight loss and other consequences of the infection. However, that wasn’t quite as lethal as the unaltered ancestral SARS-CoV-2 strain, which killed all six mice that were infected in the study.
This research is valuable because it suggests that the factors that make certain strains of SARS-CoV-2 deadly might lie outside the spike protein, says David Ho, a virologist at Columbia University in New York City. “But it raises concerns that we have an Omicron virus that’s evasive to many antibodies and yet is more pathogenic than the current version of Omicron.”
The work had been approved by a BU biosafety committee, as well as a Boston city public-health board, and was conducted in a biocontainment facility deemed safe for work with SARS-CoV-2. But it is unclear whether the BU study has run afoul of any rules governing risky pathogen research. Under current guidelines, any research funded by the US Department of Health and Human Services (HHS) — of which the NIH is part — that can be “reasonably anticipated” to make a potential pandemic pathogen (PPP) more virulent or transmissible should undergo extra review.
Saeed’s team acknowledged grants from the National Institute of Allergy and Infectious Diseases (NIAID) and other branches of the NIH in the preprint. But in a statement this week, BU said that the experiments “were carried out with funds from Boston University”, which it said means that they are exempt from the additional review. NIAID’s support was acknowledged “because it was used to help develop the tools and platforms that were used in this research; they did not fund this research directly”, said the university.
On the spectrum of coronavirus research, the experiments are relatively low-risk, Bloom says. The hybrid virus is derived from two strains that have been out-competed by successive variants, so it would be unlikely to spread widely if it ever escaped. Shi points out that the virus the researchers created is less pathogenic than the ancestral strain, which laboratories around the world continue to work with.
“This type of work needs to be reviewed carefully, and it needs to undergo risk–benefit assessments. But I would not put this in sort of the category of the most alarming types of coronavirus studies,” says Bloom. “It seems exceedingly unlikely that this virus would have pandemic potential.”
In a statement, the NIH said that it did not fund the specific experiments reported in the preprint, and it is looking into whether the research still fell under its oversight.
Shi says that in his experience, regular communication between researchers, funders and local biosafety committees can prevent problems and misunderstandings of the kind surrounding the BU study. After such discussions, his team created similar strains to study variants’ ability to evade vaccines that are made with a weakened form of SARS-CoV-2.
When Luis Martinez-Sobrido and Chengjin Ye, virologists at the Texas Biomedical Research Institute in San Antonio, wanted to conduct experiments nearly identical to those described by Saeed’s team, they contacted NIAID, which was supporting the researchers through an existing grant.
NIAID and the researchers’ institutional biosafety committee both gave the green light to the work — with the proviso that if any of the changes significantly enhanced the pathogenicity of the strain in animals or its capacity to replicate in cells, the researchers would halt the work and quickly inform the funder. Martinez says his obligations are clear.
Ho’s lab, which also receives NIH funding, has been one of the world leaders in studying SARS-CoV-2 during the pandemic. Ho says it wasn’t always clear what research was subject to review and what wasn’t, and he found himself frequently checking in with officials. When his team reported2 privately funded work showing that SARS-CoV-2 can evolve resistance to a component of the antiviral treatment Paxlovid, NIAID officials got in touch to confirm that the experiments didn’t fall under its oversight.
In another instance, Ho’s team was growing the virus in the presence of monoclonal antibody drugs, to study its ability to evolve resistance. The studies identified a host of antibody-dodging mutations that would later emerge in Omicron offshoots, including a sublineage called BQ.1 that is likely to drive an infection wave later this year.
But Ho says he scaled back the research and decided not to publish the findings, because of his concerns about how officials at NIAID would perceive the work if it were made public. The agency didn’t fund those experiments, but supported related work characterizing SARS-CoV-2 variants. “There’s a lot of valuable information that could have been shared, but because of these concerns, that was held back,” Ho says.
The discussion around the BU preprint comes amid a years-long effort to revise the US government’s funding guidelines for research involving enhanced PPPs (ePPPs). In February, the NIH asked the US National Science Advisory Board for Biosecurity (NSABB) to revisit its current policy, which was set in 2017. The NSABB released draft recommendations in September, and plans to release its final report late this year or early next. One recommendation calls for a significant expansion in the pathogens that could fall under the policy.
Marc Lipsitch, an epidemiologist at the Harvard T.H. Chan School of Public Health in Boston, says that the draft recommendations provide more clarity, but do not address the fundamental concerns that the BU study raises. The final policy should cover any ePPP research done at any US institution — not just research funded by HHS — and should allow for the additional review step to occur if potential for an ePPP to be created becomes apparent, even after the project is funded, he says.
Researchers hope that the update will provide clearer direction on which SARS-CoV-2 research needs NIH approval, and how the agency conducts its extra review. As Shi and his team develop COVID-19 antivirals, he would like to study how readily the virus can evolve mutations to evade drugs, and whether mutations linked to existing drugs can foil new ones. But he says that he has not yet received clear guidance from the NIH on what experiments he can and cannot do.
In some cases, discussions seem to be driven by publicity surrounding experiments such as the BU study, instead of by considerations of the potential risks and benefits of such work, says Bloom. The latest controversy highlights the disconnect between how scientists and the public perceive the risk of research into certain pathogens, he adds. “It’s important for scientists to recognize it’s the general public that’s funding all this research. And there are good reasons that people want more transparency and understanding.”