The idea sounds appealingly simple: Quickly spread a gene through a population of animals in order to prevent it from transmitting disease, or, more directly, to kill a destructive species such as an agricultural pest. But a workshop hosted yesterday by the National Academies of Sciences, Engineering, and Medicine (NAS) in Washington, D.C., made abundantly clear that a lot of uncertainty—scientific and regulatory—still exists for the so-called gene-drive technology at the heart of such concepts. And as result, field applications of gene drives are “still years off,” says Austin Burt, a population geneticist at Imperial College London who spoke at the meeting.
Over the past 3 years, a technology called CRISPR-Cas9 has revolutionized scientists’ ability to make precise changes in the DNA of a wide range of organisms. By being cheap, relatively easy to use, and effective in almost every species tested, this genome editing method is putting another technology, called gene drive, within reach for many organisms. Because gene drive shifts biases inheritance to favor certain versions of genes, a genetic alteration introduced into a few members of a population spreads rapidly throughout the entire population. If that alteration inhibits reproduction or survival in some way, gene drive can drive that population extinct in theory. In other uses, a desired trait could be driven through a population.
Last year, Harvard University biologists proposed CRISPR-Cas9 gene drive systems be used in conservation to get rid of invasive species or to improve the genetic makeup of endangered ones. A few scientists immediately called for increased regulation of this technology because, once released, a gene drive could be hard to stop or reverse. And in July, geneticists showed that one gene drive system was almost 100% effective in spreading a mutated pigmentation gene through a population of lab fruit flies, fueling fears about the power of gene drive.
NAS formed a committee to evaluate the technology, and yesterday hosted the second of four informational workshops in preparation of a report about the science, ethics, and governance of gene drive research. Researchers debated whether existing regulatory and ethical frameworks are sufficient to guide the development of this technology, and reported that much more needs to be learned about the ecological effects of gene drive, the specificity of gene drive targets, and the ability of researchers to effectively spread a genetic change through a population, or a species. “It’s pretty clear we know so little about these systems,” says Zach Adelman, a molecular geneticist at Virginia Polytechnic Institute and State University in Blacksburg. But given the early stage of gene drives, NAS should have time to sort this all out, Burt says.
Burt and others also noted a number of reasons why gene drives may not be as useful, or scary, as some think:
- Gene drive works only in sexually reproducing species, and the genetic change spreads further with each successive generation. So changing or eliminating a population is practical only if the species has a short generation time—like Drosophila, or mosquitos. With many vertebrates, it would take decades for an introduced gene mutation or trait to spread wide enough to make a difference.
- It has not yet been demonstrated that a CRISPR-Cas9 gene-driven change persists across many generations. The paper where gene drive was so effective only reported on one generation. For mosquitos, in which researchers want to knock out populations near people or introduce a parasite-resistant gene, modeling efforts indicate that drive would have to persist 20 generations to spread completely, Burt says.
- There are few, if any, organisms so well characterized, say biologists, that they can predict the ecological effect of a gene-driven change or a disappearing population. We will “have the ability not just to change the genome but [also] to change the balance of species in a community,” says Allison Snow, a plant biologist from Ohio State University, Columbus. “There’s a lot of potential for ignorance, human error, or intent to cause harm.”
- Before gene drive can be applied to wild populations instead of well-characterized laboratory ones, the CRISPR-Cas9 genome–editing technology needs to become even more precise. As Shengdar Tsai, a CRISPR researcher at Massachusetts General Hospital in Boston, pointed out, his team’s analysis of the method in human cells uncovered about two dozen so-called off-target effects—places where the DNA not meant to be changed was. The sites identified confirm that the sites affected by CRISPR-Cas9 can be difficult to predict.
- Instead of using gene drive to make malaria-carrying mosquitos extinct, a less ecologically worrisome strategy would be to change the insect’s genome so it would not transmit the malaria parasite to humans. But researchers don’t know enough about the mosquito immune system to target a specific gene for this type of gene drive yet, Burt says.
- An effective “fail-safe” strategy that would cause gene drive to peter out after a specified number of generations or because researchers decide they needed to stop a gene’s spread still needs to be developed. One of the more promising ones—to undo the genetic change with another gene drive effort—may still be problematic if it’s gene drive itself that goes awry.
- Hybridization between closely related animal species needs to be better understood before gene drives are unleashed. Successful mating between two species results in so-called gene flow, which could allow a gene-driven mutation to hop into an unintended species. This could be useful for malaria control—a gene drive given to one parasite-carrying mosquito species could spread it to the other seven that carry the human pathogen. But in other scenarios, a species-hopping gene drive could lead to the demise of the wrong species. At the meeting, Nora Besansky, a malaria researcher at the University of Notre Dame in South Bend, Indiana reported that the avalanche of whole-genome data is suggesting that hybridization between animal species, though not as common as in plants, occurs much more often than expected.
- Although some researchers argue that current regulations, such as those regarding recombinant DNA, and institutional review boards are sufficient for overseeing gene drive efforts, others call gene drive fundamentally different, if only because the point of gene drive is to have a mutation spread. Political boundaries may be breached. “There is no international governance yet,” says David Wirth, who studies international treaties related to biotechnology at Boston College. “It’s useful to have harmonized standards.”
- As researchers from Thailand, Africa, and Guatemala pointed out during the meeting, getting local people to embrace gene-drive control strategies takes a lot of legwork and time. “It’s not as cut and dry that once we get the technology, it will be used,” says Todd Kuiken, who follows gene drive at the Woodrow Wilson International Center for Scholars, a think tank in Washington, D.C.
In December, NAS will convene a summit on another CRISPR-Cas9–enabled technology: editing human embryos. With that technology, the work is already being done—a Chinese team engineered human embryos and several teams in the United Kingdom want to, for example—so the summit in a sense will be an effort to rein in a horse that’s already out of the gate. That’s not yet the case with gene drive. For that reason, NAS’s scrutiny “is certainly coming at the right time,” Kuiken says.