For more than half a century, scientists have dreamed of harnessing an odd quirk of nature — "selfish genes", which bypass the normal 50/50 laws of inheritance and force their way into offspring — to engineer entire species. A few years ago, the advent of the CRISPR-Cas9 gene editing technology turned this science fictional concept into a dazzling potential reality, called a gene drive. But after all the hype, and fear of the technology's misuse, scientists are now questioning whether gene drives will work at all.
Image: Getty Images
Gene drive is a molecular technology that forces an edited gene to be passed along into all of an organism's offspring, overriding nature's 50/50 inheritance mix. The first human-engineered gene drive was only demonstrated in fruit flies in 2015, but scientists were soon talking about using gene drives to exterminate invasive pests or kill off throngs of malarial mosquitoes.
But soon after, other researchers demonstrated that as an infertility mutation in female mosquitoes was successfully passed on to offspring over many generations, resistance emerged, allowing some mosquitoes to avoid inheriting the mutation. Just as bacteria can develop resistance to antibiotics, wild populations can develop resistance to modifications aimed at destroying them. Gene drive, dead.
Now, in a new paper out today in PLOS Genetics, scientists at Cornell show that, at least in fruit flies, many more flies than expected seemed to possess a natural genetic resistance to gene drive. The paper offers even stronger evidence that engineering large populations of wild species isn't as simple as splicing open a genome and inserting some gene drive DNA.
In New Zealand, the government is mulling using gene drives to wipe out invasive pests. On Nantucket and Martha's Vineyard, one scientist wants to use it to eradicate Lyme disease. In Guam, they want to control tree snakes. But not so fast, scientists are saying.
"These resistance rates were so high that a gene drive would not spread in a population," Phillip Messer, a co-author on the study, told Gizmodo. "Our take home is that resistance is clearly a bigger problem than we had initially thought. This technology could still work, but it's not as simple as the first papers suggested."
The Cornell paper appeared alongside an opinion piece with a headline that suggested a provocative notion: Until now, the conversation about gene drives has existed in a reality-free bubble.
"This 'resistance' outcome would easily thwart virtually any intended application of a gene drive, and it poses a serious challenge to the many hoped-for applications of this technology," its' authors wrote.
Resistance isn't the only hurdle to putting gene drives to practical use. For one, so far, synthetic gene drives have only been demonstrated to work in insects and yeast. Safety is a big concern. And based on the outcry such science has already seen from environmental groups, it's safe to say there will be a fair number of regulatory and political obstacles, too.
But resistance may very well be the biggest problem, and it's a problem that has been downplayed until recently.
"People are starting to dig more into the nuances of this stuff and we're getting into the nitty gritty of what needs to be addressed," Gabriel Zenter, an Indiana University biologist, told Gizmodo.
In the new research, scientists for the first time gave some hint of the mechanisms that may be responsible for resistance. Certain flies, even though they were all members of the same species, just seemed to be better equipped genetically to fight back against a drive. They also found that resistance developed both before fertilisation in the germline, and within an embryo. And resistance could crop up within a single generation. This means that were a gene drive deployed in the wild, it is hard to say how effective it would really be.
"You don't know whats lurking around in the genome that could influence a gene drive positively or negatively," said Zenter, who was not associated with the study. "People didn't anticipate things like the genetic background issue. I think we're kind of coming towards a more mature understanding of the hurdles that will need surmounted."
At least a few research groups already are working on a way around those hurdles. In another paper out this year, researchers proposed a way to redesign gene drives in order to work around potential immunity, hypothesising that a more complex architecture would make it difficult for a mutation to occur in a short period of time. Instead of just including instructions for a gene drive to cut a piece of DNA in one place, their architecture it cuts in multiple places, meaning it would require multiple mutations to overwrite the drive. They also suggested a second method that harnesses a species' survival programming, targeting areas of the genome that are essential to a species' fitness, and which are less likely to mutate in the first place.
In a pre-print paper, Messer's lab has already experimented with the first scenario. "It works, but not as well as we had hoped," he said.
In the end, he said, a working gene drive will probably be much more complex than anyone imagines, incorporating several different strategies into the architecture to override resistance.
Charleston Noble, a Harvard PhD candidate studying gene drives, is more optimistic. After all, he points out, mosquito species have shown to be naturally less likely to develop resistance than fruit flies. Not every species might be so tricky to manipulate, and in some cases you may not need to alter an entire population to bring about the desired change.
And Kevin Esvelt, a synthetic biologist at MIT, said the experiments only confirmed what scientists have long known.
"These elegant experiments conclusively show that there is no reason to build a gene drive system that only cleaves a single site," he told Gizmodo. "I'm not so sure it amounts to popping a 'bubble' in the field, or that this is any kind of new reality."
In the realm of synthetic biology, it has become a well-worn cliche that "life finds a way". In the end, though, there is something to it. Engineering nature will require more than the flip of a simple genetic switch.