Gene drive, a means of genetically engineering entire populations, is encountering resistance—and not just from people who question the wisdom of creating genetically modified organisms. No, gene drive is being opposed by the very organisms that the technology would suppress, or somehow alter to suit human ends. More precisely, gene drives, when used in laboratory settings, have shown that they not only instigate the spread of desirable genes through populations of insects, they also give rise to resistance alleles.
When used to modify populations of mosquitoes, beetles, and fruit flies, the most promising gene-drive technology, CRISPR/Cas9-based gene drive, has delivered mixed results. In some instances, this technology appears to be undone through naturally occurring genetic variation, which can alter target sequences and negate CRISPR/Cas9’s homing capabilities. In other instances, CRISPR/Cas9 is defeated even if it manages to create a double-strand break where needed. This double-strand break, which is supposed to trigger a mechanism called homology-directed repair, may instead instigate nonhomologous end-joining.
The problem is, only homology-directed repair inserts homologous DNA, that is, DNA copied from a homologous gene. It is this mechanism that puts the “drive” in gene drive, provided the homologous DNA has been engineered to contain code for the CRISPR/Cas9 endonuclease, the endonuclease’s guide RNA, and a payload gene, all between code for flanking segments. The guide RNA is constructed such that it directs the endonuclease to create a double-strand break at a selected chromosome location, in a native allele. Ultimately, and by design, the DNA segments on either side of the double-strand break align with the flanking segments of the CRISPR/Cas9-containing construct that constitutes the homologous DNA.
By taking advantage of homology-directed repair, CRISPR/Cas9-based gene drives may instigate a mutagenic chain reaction, a cascade of genetic changes whereby cells heterozygous for the drive allele become homozygous. At the population level, this chain reaction can drive an allele through a population at super-Mendelian rates of inheritance.
But deployment to a CRISPR/Cas9-based gene drive may not always work out that way. Much depends on the relative rates of competing processes, such as the spread of drive genes and the spread of resistance alleles, and when these processes are initiated.
A study examining how these processes could defeat gene drive was recently completed by scientists based at Cornell University. These scientists, led by Philipp W. Messer, Ph.D., published their findings June 20 in PLOS Genetics, in an article entitled, “Novel CRISPR/Cas9 gene drive constructs reveal insights into mechanisms of resistance allele formation and drive efficiency in genetically diverse populations.”
The researchers tested two different CRISPR gene-drive constructs, one based on nanos and one based on vasa promoters, in the model fruit fly, Drosophila melanogaster, to investigate the rise of resistance. “We observed resistance allele formation at high rates both prior to fertilization in the germline and post-fertilization in the embryo due to maternally deposited Cas9,” the authors indicated.
Further analysis showed that in insects with genetically diverse backgrounds, as found in wild populations, there was considerable variation in terms of how efficiently the offspring converted, and how often resistance genes arose.
“Resistance alleles were mostly of type 2 that disrupted the target gene, but they could also be type 1, which preserved the function of the target gene,” noted Jackson Champer, Ph.D., a researcher in Dr. Messer’s laboratory and the first author of the current study. “We also showed that rates of resistance are dependent on the genetic background of insects, so a gene drive could perform much better or worse in one insect than another of the same species by testing their performance in fly lines gathered from all over the world.”
The authors of the PLOS Genetics article asserted that their findings have important implications for the feasibility of gene-drive strategies in the wild.
“Our experimental demonstration that resistance alleles form at high rates certainly indicates that gene drives are not currently ready for deployment. However, this doesn’t necessarily mean that they will never work,” commented Dr. Champer. “Instead, we are simply presented with a new obstacle that can potentially be overcome. There are several possibly strategies for reducing resistance allele formation rates, and we have already started pursuing some in the lab.”
Work undertaken in Dr. Messer’s lab suggests that a second gRNA could significantly improve gene-drive performance, but not as much as an “idealized expectation would indicate,” said Dr. Champer—and not enough to ready gene drives for deployment in the wild. “Overall, it will certainly be very difficult to sufficiently reduce resistance allele formation,” Dr. Champer continued, “but we are optimistic that it may well be possible with a major scientific effort.
Dr. Champer also distinguished between two basic types of gene drive strategies: population modification (spreading a payload gene through a population) and population suppression (trying to reduce or eliminate the population of a species). Of these two strategies, population suppression may prove to be more challenging.
“Population suppression is much more vulnerable to resistance alleles than population modification,” Dr. Champer advised. “Even a few surviving resistant insects can quickly allow the population to recover.”
In contrast, in population modification, a few resistant insects “could eventually take over,” but the modified insects “may still persist at high abundance for a long time with positive effects.”
Dr. Champer emphasized that population suppression approaches do have an advantage: they would mostly be affected by the less common type 1 resistance alleles, with type 2 resistance alleles having less of an impact in stopping them. He also allowed, however, that if one were to devise a clever strategy, one could do something similar for a population modification approach.
“On average, it would be harder to conduct successful population suppression compared to successful population modification, for this and other reasons not directly related to our manuscript,” he explained. “However, it partially depends on the specifics of the drive systems that you are comparing, so it’s not possible to make an absolute statement about this.”
In any case, the current study makes the case that new gene-drive approaches will be necessary to reduce the formation of resistance alleles, particularly in genetically diverse natural populations. The development of such approaches will require a better understanding of the mechanisms by which wild-type alleles are converted to drive alleles or resistance alleles.
As new gene-drive approaches are developed, they will move closer to deployment in the field. Anticipated deployments include those intended for mice on islands off the coast of Massachusetts to prevent the spread of Lyme disease, and in tree snakes in Guam to control these invasive species.