At the largest zebrafish facility in the country, Kevin Bishop, NHGRI Zebrafish Core staff member, holds up a tank of zebrafish to observe their behavior and physiology. Source: NHGRI

New computer modelling has predicted that releasing even a small number of edited organisms into an ecosystem as part of a gene drive will lead to the gene edits becoming more invasive and wide-spread than intended. A preprint published in biorxiv.org and a second paper in PLOS Biology report that gene drives are likely to spread further in the natural population than intended and that more research is needed before they transition from the lab and into practice.

Gene drives involve genetically modifying a small number of organisms, likely with CRISPR, to alter their gene inheritance rules, before releasing them into the wild-type population. Research has shown that through genetic modifications, it is possible to increase the chance of a particular gene from being passed on to future generations from 50% to almost 100%. This would allow researchers to eradicate undesirable organisms from an ecosystem, such as pests that destroy food crops or mosquitoes capable of malaria transmission.

Currently, gene drives are only taking place in a laboratory environment, but this could change in the near future. In particular, many places in the world are considering using gene drives to wipe out mosquitoes carrying diseases like malaria and Zika over the next few years.

The scientific community has raised several concerns around gene drives. The first is that there is a chance that wild organisms will evolve resistance to the changes and thus the gene edits will be unable to spread sufficiently far to achieve the desired outcome. The second concern is the opposite; some researchers are worried that the gene edits will be invasive in nature, and will spread well beyond the intended parameters of the drive. If this were the case, it would be very difficult to predict what changes would take place within the ecosystem and how damaging it could be.

A preprint that appears to confirm the latter worry was released yesterday by Harvard University researchers, Charleston Noble, Ben Adlam, PhD, George Church, PhD, and Martin Nowak, PhD, and MIT researcher, Kevin Esvelt, PhD. The team used two computer models to investigate the effects of gene drives in an ecosystem, one of which modelled large populations and one which focused on smaller populations between which individuals could migrate. To ensure that the models were as accurate as possible, they incorporated data from previously published gene drive experiments into their research.

The models demonstrated that in both scenarios, small numbers of edited organisms would become invasive and spread throughout the population. Even in modelled cases where the gene drive was ineffective (gene transmission was low) and resistance to the edits was established, the team found that the gene drive would still overwhelm the population.

While this data is not encouraging for future gene drives, there are certain points that are important to note. For one, the models were using a type of gene drive called a replacement or alteration drive, which swaps out an undesirable gene for a more preferable one or simply removes the genes entirely. These drives have been considered for use but it is more likely that suppression drives, which target genes needed for an organism’s survival, will be the more widely used form of the practice. Lead author Charleston Noble has said that he wants to continue this research to also examine these types of drives as well, but for now, there is no available data.

The models also didn’t consider the physical factors that could limit a gene drive’s impact on an ecosystem, such as any variations in local ecology, the distance and feasibility of travel between different populations, and the ease of mating between wild-type and edited organisms. The team have said that this was an intentional omission so that the models could be kept as simple as possible and to avoid introducing a containment bias that may not be accurate. 

Nonetheless, the concerns raised by these models have been echoed in the second paper, published by Dr. Esvelt and Neil Gemmell, PhD, from the University of Otago in New Zealand. In this paper, they predicted the effect that a gene drive could have on a relatively isolated ecosystem, such as that of New Zealand, and questioned whether or not the science was ready to deal with such a scenario. Above all, they argue that more research is needed in the field before we start to use gene drives in a practical setting.

The PLOS Biology paper concludes by asking: “Are New Zealand communities prepared to guide the development and oversee the testing of these systems? Are similar conversations happening internationally? We hope so, because this conversation should not be confined to scientists, regulators, politicians, or any single nation, no matter how strong its legislative frameworks, environmental risk management, and biosecurity networks. If we have learned anything from the spread of invasive species, it is that ecosystems are connected in myriad ways and that a handful of organisms introduced in 1 country may have ramifications well beyond its own borders.”

“Our results have numerous implications for future gene drive research,” the authors on the Biorxiv preprint wrote. “First, researchers interested in studying self-propagating gene drives may wish to refrain from constructing systems that are capable of invading wild populations. Invasion can be avoided by employing intrinsic molecular confinement mechanisms such as synthetic site targeting or split drive, as recommended by the National Academies.

“Second, contrary to the National Academies’ recommendation of a staged testing strategy, the predicted invasiveness of current CRISPR-based drive systems may preclude field trials, possibly even on ostensibly isolated islands. The development of ‘local’, intrinsically self-exhausting gene drive systems, sensitive methods of monitoring population genetics, and strategies for countering self-propagating drive systems and restoring populations to wild-type should be a correspondingly high priority.”