The mammals of New Zealand have long posed a threat to native species. The Predator Free 2050 program is an effort to rid the island of these invaders – including using the tools of CRISPR-based genome editing to create a gene drive to jumpstart extinctions.
It’s a very bad idea.
In the 1993 film Jurassic Park, mathematician Ian Malcolm listens to arrogant dinosaur daddy John Hammond describe the island’s supposedly all-female populations of the giant reptiles:
“John, the kind of control you’re attempting simply is… it’s not possible. If there is one thing the history of evolution has taught us it’s that life will not be contained. Life breaks free, it expands to new territories and crashes through barriers, painfully, maybe even dangerously, but, … there it is … I’m simply saying that life, uh… finds a way.”
The wise Dr. Malcolm may prove prescient when it comes to using gene drive technology to get rid of pesky species.
Today reptiles, albeit smaller ones than dinosaurs, are among the threatened natives of New Zealand. Prior to the arrival of people, only bats and marine species represented class Mammalia, except for a few archaic types a few million years ago. Then the Māori people introduced Polynesian rats and dogs in about 1250 CE, and Europeans five centuries later contributed mice, pigs, more rats (ship stowaways), possums, weasels, stoats, and ferrets. Native birds, reptiles, invertebrates, snails, insects, and even the forest canopies began to lose out in the competition for natural resources and to predation.
The New Zealand government painted the newcomers as pests, interlopers, invaders. “Introduced predators: the bad guys,” states one pamphlet.
In a simpler and perhaps more violent time, pests might have been shot, drowned, or poisoned. But a 2003 paper from Austin Burt, a “selfish gene” proponent from Imperial College, London, proposed the concept of a gene drive.
Building on natural DNA repair
A gene drive harnesses one of the ways that cells repair DNA, called “homing,” that snips out one copy of a gene and replaces it with a copy of whatever corresponding gene variant (allele) is on the paired chromosome. It would be like cutting out a word in this sentence and replacing it with a copy of the word below it. If done to a gene that affects fertility in a fertilized ovum – aka the germline – the intervention can lead, within a few generations, to mass sterility and a plummeting population – a “gene drive” towards extinction.
A gene drive skews Mendelian inheritance. Instead of one of a pair of genes coming from the father and one from the mother, both copies are from one parent. In the language of genetics, the intervention can turn a heterozygote (2 different copies of a gene) into a homozygote (2 identical copies). Nature does this in several ways, but the tools of CRISPR-Cas9, first described in 2012, offer a faster route to a gene drive, and can target several genes at once.
Visions of vanquishing the mosquitoes that carry the malaria parasite or zika virus dampened initial scrutiny of gene drives. In 2016, the National Academies of Sciences, Engineering, and Medicine (NASEM) released a 200+ page report that discussed reasons to proceed with caution, but endorsed continued laboratory experimentation as well as limited field trials of gene drives.
Today, a short paper in Science responds to the NASEM report with “Guiding principles for the sponsors and supporters of gene drive research.” I’ll return to the new recommendations after a trip down biotech memory lane – what distinguishes this blog from the clonal regurgitations of aggregated science news.
The triple-headed purple monster precedent of Asilomar
In February 1975, a who’s who of molecular biologists had convened at Asilomar, on California’s Monterey peninsula, to explore the implications of combining genes of two species, starting with insertion of a bacterial gene into a cancer-causing virus.
The 150 scientists discussed fail-safe measures to control recombinant organisms. The Asilomar conference begat guidelines for “physical containment” via specialized hoods and airflow systems and “biological containment” to weaken organisms so that they couldn’t survive outside the lab.
Despite initial concerns, recombinant DNA technology turned out to be safer than expected, and it spread to industry fast and in diverse ways. A handful of important drugs, starting with human insulin, became safer and more abundant thanks to recombinant DNA techniques. In the agricultural arena, we’ve been eating GMO foods for decades, although the containment hasn’t exactly worked, as the example of canola growing along the roadways of North Dakota illustrates.
In 1985 geneticists met again to assess the safety, feasibility, and value of another huge project: sequencing “the” human genome. I doubt any of them could have foreseen a time when we would carry our genome sequences on our smartphones.
Back then, researchers packed a room at the Cold Spring Harbor Laboratory on New York’s Long Island. At first those against outnumbered those for 5:1, ticking off their fears: shifting research from inquiry-based experimentation to data dumps, comparing the sequencing effort to climbing Mt. Everest just because it’s there, and diverting funds to fight HIV/AIDs. Finally, the National Academy of Sciences jumped in to debate both sides, and in 1988, Congress authorized the National Institutes of Health (NIH) and the Department of Energy to start sequencing. Foreshadowing of gene drives?
On the reproductive front, the first test-tube baby, Louise Joy Brown, was discussed as if she were a space alien until her ordinariness became apparent, and today more than 5 million folks have been born beginning with in vitro fertilization. Similarly, one of the first families to speak to the media about their use of preimplantation genetic diagnosis (PGD) to select an embryo who would one day provide stem cells to save his sister was vilified – PGD is now a common adjunct to IVF to select the healthiest embryos.
But a gene drive doesn’t provide information, drugs, improved cabbages, or babies. It has the potential to tilt the biosphere.
The changeability of DNA
When the inventors of a new biotechnology pull a 180 on applications of their brainchild, it’s time to take notice. That’s what Kevin Esvelt from MIT and Neil Gemmell from the University of Otago, Dunedin, New Zealand, did in their Perspective in the November 16 issue of PLOS Biology, “Conservation demands safe gene drive.” They shout out a warning.
Back in 2014, Esvelt and his colleagues had suggested using “self-propagating CRISPR-based drive systems for conservation.” They also discussed variations on the theme, including a “daisy drive” system that sets up a series of interventions, like a series of locks on a bank vault, and the “trojan female” technique that sneaks male infertility mutations into mitochondrial DNA.
Today’s second thoughts about deploying gene drives were perhaps already lurking in the minds of people familiar with the nature of DNA, as Jurassic Park’s mathematician intuited. DNA changes! That’s why it’s the genetic material and why the idea that we aren’t still evolving is absurd.
A gene swapped into a rat or a possum’s genome to squelch fertility can change. Such spontaneous mutation happens because of the nature of the DNA molecule. Each of the 4 types of DNA bases exists, when unlinked, fleetingly, in a slightly alternate form. If a DNA replication fork should happen down the old double helix and catch a clinging base in its rare form, a base pair can be replaced with a different one – creating a new allele. It’s simply the chemistry of life.
A gene drive also assumes that one allele is predominant in a population, and that isn’t necessarily the case. What if the harnessed repair mechanism lassos another variant of that gene, a rarer one? Different outcome.
The inherent changeability of DNA alerted the scientists at Asilomar and Cold Spring Harbor. We can never predict all risks, about anything, and surprises have consequences. Who would have thought we’d all have to haul off our boots when checking in at the airport thanks to a lone shoe bomber?
DNA also flits from cell to cell, aboard elements called transposons or, more colorfully, jumping genes. That’s how bacteria share sets of antibiotic resistance genes. What if a CRISPR gene drive harpoons something other than its intended target? Goodbye beloved kiwi birds rather than the weasels that eat their eggs? What if a targeted species “hitches a ride to other islands and continents before it eliminates the local population and extinguishes itself?” Drs. Esvelt and Gemmell write.
The bottom line: gene drives may create the equivalent of the very thing they are being deployed to fight: invasive species. Write Drs. Esvelt and Gemmell of their former approval of gene drives for conservation, “We now believe that inclusion was a mistake: such drive systems lack control mechanisms and are consequently highly invasive.”
And so also on November 16, Dr. Esvelt, with Charleston Noble, Ben Adlam, George Church, and Martin Nowak from Harvard, published “Current CRISPR gene drive systems are likely to be highly invasive in wild populations” in bioRxiv. Their paper warns against even limited field tests because of “mitigating factors,” including scenarios as yet unimagined. They did a mathematical analysis to counter recent reports that downplayed the potential ecological danger of a gene drive by claiming that natural resistances will emerge to block the spread to untargeted wild populations. Sound familiar? “Contrary to the National Academy report on gene drive, our results suggest that standard drive systems should not be developed nor field-tested in regions harboring the host organism,” they conclude.
The “guiding principles for the sponsors and supporters of gene drive research” published in today’s Science, from Claudia Emerson, Stephanie James, Katherine Littler, and Filippo Randazzo, are déjà vu all over again for those of us who recall Asilomar circa 1975. Perhaps the principles are attempting to prevent the public outcry at town hall meetings and destruction of some GM crops (most notably ice minus bacteria on plants) that accompanied the entry and acceptance of recombinant organisms.
According to the principles, gene drive experiments should
• have goals of social value and the public good
• take biosafety measures, comply with regulations, and conduct ecological risk assessment
• have transparency and accountability, with sharing of data
• engage the public
Dr. Emerson and her colleagues make a good case for the need to find new ways to limit the spread of vector-borne infectious diseases like malaria and zika. Let’s hope that gene drive technology goes the successful way of recombinant DNA technology and not the way of GMO escapees in agriculture or in the hands of bioterrorists.
Let’s listen to Dr. Malcolm.
[Editor's note: Kevin Esvelt of MIT commented on this article on PLOS Blogs. He wrote:
Respectfully, this somewhat mischaracterizes our point.
We think it unwise to build gene drive systems capable of spreading indefinitely beyond the target population.
Because standard self-propagating gene drive systems can spread indefinitely, we think they should only be developed and used for a handful of applications such as malaria eradication, for which the target population includes every Anopheles gambiae s.l. mosquito in Africa.
In contrast, we feel that self-propagating gene drive should not be used for invasive species control because there is always a native population that could be affected.
Instead, we should focus on developing locally-confined drive systems that cannot spread indefinitely. Local drive systems could enable each community to make decisions about its own environment without necessarily affecting people far away. There are several forms that have been modeled or are under development, including Trojan female, killer-rescue, daisy drive, and threshold drive, and hopefully still better ones will be invented.
A final note: there is essentially no risk that transposons, a natural and nearly ubiquitous form of gene drive, will cause a CRISPR-based drive system to spread in another species. The reason is that CRISPR is highly specific and the target DNA sequences would not be present in the genome, so the system would not function – exactly the same way that laboratory genome editing fails when there is a strain-specific mutation in the CRISPR-targeted sequence.
Life usually does find a way eventually; the question is how long it will take. We have a remarkable opportunity to address many serious ecological problems using nature’s own language. With care, humility, and collective scrutiny – as obtained through open research and broadly inclusive societal discussions – we have a chance to do so wisely. Sometimes, that means walking away from an exciting idea.]
A version of this article was originally published on PLOS Blog's website as An Argument Against Gene Drives to Extinguish New Zealand Mammals: Life Finds a Way and has been republished here with permission from the author.