Read our COVID-19 research and news.

Genome editing has created cows that grow no horns, the better to avoid painful dehorning.

Genome editing has created cows that grow no horns, the better to avoid painful dehorning.

Cornell Alliance for Science

Proposed U.S. biotech rules raise industry hopes and anxieties

Makers of genetically engineered (GE) products have long been on the lookout for changes in U.S. regulations. The system that divvies up the safety review of these products between the Food and Drug Administration (FDA), Environmental Protection Agency (EPA), and U.S. Department of Agriculture (USDA) is in the middle of a years-long overhaul, in part to accommodate modern gene-editing technologies such as CRISPR. And a set of new proposals released last week offers hints about what those agencies still hope to change.

Products made through genetic engineering have often faced an inconsistent and counterintuitive path to market. Some have worried that potentially risky products could fall through the regulatory cracks and avoid proper review. (USDA’s decision last April not to regulate a CRISPR-edited nonbrowning mushroom, for example, shocked some genetically modified [GM] organism opponents.) Meanwhile, developers of these products have argued that new, precise techniques to edit a few letters of genetic code shouldn’t be held to the same strict standards as older approaches that make less predictable or more widespread genetic changes.

Earlier this month, the White House released an update to the overarching system of biotech regulation, known as the Coordinated Framework for Regulation of Biotechnology. But it’s still up to individual agencies to clarify how they intend to classify and evaluate various GE products. In an apparent effort to get plans on the table before a change of administration, USDA and FDA put out draft proposals on 18 January addressing several categories of GE products. Reactions from the companies behind those products have been mixed:

FDA could give way to EPA on GE mosquitos

Controversy continues around proposed field trials for GM mosquitoes developed by the U.K. biotech firm Oxitec, but the agency responsible for determining their safety might change. The company created a strain of Aedes aegypti—the mosquito species that transmits the Zika and chikungunya viruses—designed to spread a lethal gene and dramatically cut down populations. The company underwent a nearly 5-year process with FDA’s Center for Veterinary Medicine before its proposed test release of mosquitoes in Florida was cleared last August.

That FDA office oversees technologies for sterilizing and controlling animal populations, but giving it responsibility for gauging the environmental impact of a mosquito raised eyebrows on both sides of the debate. “Without relevant expertise, not surprisingly the FDA has been ill-equipped to review the application expeditiously,” wrote attorney John Cohrssen and physician and former FDA official Henry Miller in a Forbes op-ed last January.

In a draft guidance released last week, FDA explains that products intended to reduce mosquito populations should be considered pesticides. That means that, if finalized, the guidance routes future GE mosquito strains to EPA for review. (Designer mosquitoes intended to reduce disease transmission, such as those that spread the insect parasite Wolbachia, would still be considered “new animal drugs,” and would fall to FDA.)

Oxitec now plans to seek approval for a new release site after residents in the Florida community of Key Haven voted against the proposed release last fall. FDA could still be the one to make that determination before the new guidance is finalized, says Jack Bobo, chief of communications at Intrexon in Washington, D.C., which acquired Oxitec in 2015.

It’s not yet clear how the EPA requirements would differ from those of FDA, he says. “We will talk to EPA in advance of any field trial to see if there’s data they would have required that perhaps FDA didn’t, but it’s also possible that we’ve gathered a lot of data that EPA wouldn’t have requested,” Bobo says. “I’m not sure that it matters very much to us who regulates, as long as it’s a transparent, predictable, and science-based regulatory process.”

A broad net for GE animals

A separate FDA draft guidance on GM DNA in other animals, meanwhile, seems to sweep a whole range of technologies under the agency’s purview. It specifies that any intentional change to an animal’s genes meets the definition of a “new animal drug” because it’s meant to “affect the structure or function of the body of the animal.” That definition could subject a wide range GM animal products—from a pig engineered to resist flu to milk from a genetically altered hornless cow—to the full safety and efficacy review process required of a new drug.

“I understand where somebody could look at that and kind of freak out a bit,” says Todd Kuiken, an environmental scientist who studies biotech regulation at North Carolina State University in Raleigh. Many had expected some products of genetic engineering to be exempt from that complex and costly premarket review process.

That group includes Scott Fahrenkrug, founder and chief scientific officer of Recombinetics in Saint Paul. The company aims to eliminate the painful process of dehorning dairy cows by instead endowing them with a stretch of DNA from a hornless cattle breed. The company has planned to market milk from those cows under an FDA designation—“generally recognized as safe”—that sidesteps the need for a full premarket review. (The idea is that milk has already been shown to be safe, and adding a gene from another breed won’t change that.)

Now, the new guidance means the Recombinetics cows might need to go through premarket review, even though their genome sequence is identical to hornless dairy cattle that could be produced through conventional breeding, Fahrenkrug says. “Regulating identical products … that can’t be practically differentiated from each other seems doomed to fail,” he says, and FDA’s proposal “would put the U.S. at a huge competitive disadvantage with other countries with less [regulatory] burden.”

Kuiken suggests that FDA’s document is more of a “trial balloon”—a first attempt to collect safety data on a broad range of gene-editing techniques. “I don’t read that as, they’re going to regulate every single genetically engineered animal,” he says. In fact, among the questions FDA is putting out for public comment: whether there’s good evidence that certain categories of genome-altered animals pose no risk to consumers or the environment.

Exemptions for GE plants, with caveats

Finally, a proposed rule from USDA’s Animal and Plant Health Inspection Service (APHIS) puts forth definitions of GE plants that could let certain long-regulated products off the hook. Early techniques to transfer plant genes relied on the bacterial vector Agrobacterium, which triggers premarket review by APHIS because it’s classified as a “plant pest.” Newer approaches like CRISPR, however, don’t rely on that vector. USDA aims in its proposal to move away from regulating GE plants based on the process that produced them, and instead to evaluate the product itself for risk as a plant pest or noxious weed.

The proposed rule exempts certain products from the definition of GE, including plants containing inserted DNA from a sexually compatible species, and plants with DNA changes that could also be achieved through older chemical or radiation-based methods.

Those allowances were welcome news to some industry groups, including the American Seed Trade Association, which released a statement praising the policy.

But there’s a troubling caveat in the document, says Antony Evans, CEO of TAXA Biotechnologies in San Francisco, California, which is developing ornamental plants through synthetic biology. Among the GE plants deemed subject to premarket regulation are any plants with traits that haven’t already been evaluated by APHIS for risk as a plant pest or noxious weed.

Well-understood traits are inevitably those developed and licensed by large firms, he says. And smaller companies like his could still be required to run multimillion-dollar safety assessments, even for plants that would seem to pose no risk of acting like weeds. His company, for example, is developing fragrant mosses that contain genes from other plants, such as patchouli. Evans says he plans to ship out a new line of moss in March. “Should I continue to develop new flavors?” he wonders. “Are we just going to have stop selling [the moss], or are we going to get some sort of grandfathering clause?”

Any of these policies could change significantly before the documents are finalized; the agencies are now accepting public comment. And many researchers are unsure how the nebulous deregulatory agenda endorsed by the new administration of President Donald Trump might influence the final rules. Fahrenkrug notes that Trump has said he plans to cut corporate regulations by 75% or more, “so we are taking a wait-and-see approach on this,” he says.