When Gregor Mendel tracked pea-plant characteristics over successive generations in the nineteenth century1, his landmark study revealed key insights into the fundamental mechanisms governing genetic inheritance. Mendel observed consistent patterns of inheritance that corresponded to each descendant receiving one of the two maternal copies of a gene affecting the characteristic and one of the two paternal copies of this gene. In this typical scenario of genetic inheritance, both maternal copies of a gene have an equal probability of being inherited, as do both paternal copies.
However, inheritance does not always proceed so fairly, and in some cases the odds of a particular copy of a gene being transmitted to the next generation can be heavily skewed. One natural example is that of ‘jumping genes’, which are inherited in a non-Mendelian pattern2. Genetic-engineering approaches are being developed to manipulate the inheritance pattern of a gene copy such that it will spread through a population more rapidly than would be expected by normal Mendelian inheritance, generating what is called a gene drive and leading to super-Mendelian inheritance3,4. This process generates what is called a gene drive. So far, gene drives have been mainly engineered in insects. Writing in Nature, Grunwald et al.5 report a method for generating a gene drive in mice, offering an option to use this approach in mammals.
Gene drives developed in insects might provide a way to alter mosquito populations to decrease the probability that they transmit diseases such as malaria or dengue fever3,4. For example, a gene drive that affects mosquito fertility could be used to specifically eliminate a species of malaria-transmitting mosquito4, allowing its ecological niche to be filled by other mosquito species that cannot harbour the malaria-causing parasite. Alternatively, gene drives can be designed6 to confer widespread, species-specific resistance to infection by this parasite, for instance by using a gene drive to spread sequences that encode antimalarial antibodies so that mosquitoes are no longer infected by the parasite7.
The technology needed for gene drives has been greatly accelerated in insects by harnessing a gene-editing technique called CRISPR3,4,6. This system relies on the insects being engineered to express the enzyme Cas9 and a guide RNA that provides gene-targeting specificity. Cas9 generates a cut in a genomic DNA sequence that matches the guide RNA sequence (Fig. 1). If the guided cut generates a double-stranded DNA break in one copy of a gene, this break can be repaired by a process called homologous recombination, in which the undamaged chromosome containing a sequence that matches that in the region of the DNA break is used as a repair template.
A DNA sequence needed for the gene drive, called a cassette, which encodes CRISPR machinery, can be engineered and inserted into a chosen site in a host chromosome. The cassette encodes components needed to initiate a targeted Cas9-mediated DNA break on the sister chromosome. Successful repair of this break by homologous recombination using the chromosome that contains the cassette results in both the maternal and paternal sister chromosomes having identical copies of this cassette (a state called homozygosity). The cassette can be engineered to deliver additional DNA sequences, and such gene editing results in cells that are homozygous for any desired gene on the cassette. Achieving this effect consistently in the reproductive cells (germ cells) would ensure that all offspring receive the cassette, rather than just half the offspring as expected by Mendelian patterns of inheritance. If a gene drive works efficiently in rapidly reproducing populations such as insects, it would be predicted that an entire population could be manipulated to carry the desired gene on the cassette.
Gene drives have flourished in mosquito studies that have adapted the genetic-engineering tools developed in the fruit fly Drosophila melanogaster. Gene drives engineered in mosquitoes can be stably transmitted over many generations through a process that uses a form of high-fidelity homologous recombination that is remarkably efficient in the mosquito reproductive tissues (the germ line)4,6. However, it has been difficult to apply these approaches in mammals, which have evolved independently from insects for more than 700 million years.
But now, Grunwald and colleagues have developed a CRISPR-based gene drive for mice. They engineered animals to express Cas9 and a cassette they called CopyCat, which encoded a guide RNA that targets a sequence in the gene Tyr (Fig. 1). CopyCat was inserted into the Tyr sequence at a position that ensured that the guide RNA wouldn’t target the copy of Tyr in which the cassette was inserted.
Tyr encodes an enzyme called tyrosinase, which affects mouse coat colour. This enabled the frequency of genetic modification of the gene to be tracked over generations by monitoring coat colour and using DNA-sequence analysis to assess the transmission of the CopyCat cassette. The authors tested the effect of different genetic elements called promoters that affect Cas9 expression patterns. If Cas9 was expressed ubiquitously and continuously, the Cas9-mediated cut site in Tyr had a high level of DNA damage, which arose from a DNA-repair process called non-homologous end joining (NHEJ). When the authors limited Cas9 expression to the male germ line, they also observed high rates of DNA damage caused by NHEJ. However, the gene drive worked successfully when Cas9 was expressed specifically in the female germ line, and, in this context, the Cas9 cuts of the Tyr sequence were repaired by homologous recombination. The transmission rates of the CopyCat element to the next generation in female mice were greater than the 50% transmission that would be expected for standard Mendelian inheritance. The maximum efficiency of this CRISPR editing was a 72% success rate in copying the CopyCat cassette.
The reason for the sex-specific differences in homologous recombination and NHEJ that the authors observed is unknown. But it could be a major impediment for using mammalian gene drives because NHEJ damages the guide-RNA recognition site and therefore blocks the ability to transmit the gene drive. Male and female germ-cell development is substantially different, so further investigation will be needed to learn whether efficient homologous recombination occurs in the male germ line when the timing or pattern of Cas9 expression is altered. Nevertheless, Grunwald and colleagues’ work is an important proof-of-concept that will surely be followed by modifications that might lead to improvements in future mammalian gene drives.
If gene drives become efficient in mammals, one possible way in which they might be used is to tackle pests or disease-causing agents. The eradication of invasive rodents from islands can bring about a dramatic recovery of native ecosystems, but achieving this eradication using current pest-control methods requires Herculean efforts8. A mammalian gene drive might provide a powerful alternative. However, eradication is not the desired outcome if a disease-harbouring species is native to a region but has a key role in supporting ecosystem balance9,10. Native species can harbour organisms, such as the bacterium that causes plague, that are responsible for deadly human diseases. A gene drive engineered to express an antibody to block an infectious agent would protect people from animal-transmitted disease and maintain native species that are essential9,10 to the ecosystem.
Another possible application of mammalian gene drives is to speed the generation of animal models of disease, because it can be challenging to breed a mouse that has specific combinations of mutations in several genes.
Because gene drives have the potential to alter an entire species, appropriate regulation of this technology is a major concern. Only the most intractable and major health challenges should be considered for possible interventions using gene drives. Any proposed genetic change should be tested to minimize the chances of unintended consequences to the species or the ecosystem. This challenge is particularly daunting for highly mobile species such as the mosquito, which can fly long distances and across national boundaries. Certainly, the use of a gene drive for mosquito-borne diseases such as malaria warrants international efforts that proceed using careful planning and monitoring, and with the engagement of local communities. Nevertheless, it should be remembered that even the best-planned efforts can have unexpected outcomes. A mammalian gene drive might offer a more attractive test case than an insect one for pest eradication or infectious-disease control, because wild mammalian populations can be more easily restricted to a geographic region than can insect populations.
More than 150 years after Mendel’s work illuminated one way in which genetic inheritance can be governed, a powerful tool has emerged to manipulate inheritance in mammals. It seems certain that the promise of continual improvements in gene drives will be matched with even more discussion of how to move forward. The development of this technique to generate a mammalian gene drive is another milestone in this exciting area of research.
Nature 566, 43-45 (2019)