In their current pre-publication Hammond et al investigate the emergence of resistance to a homing-based gene drive designed to disrupt a gene essential for female Anopheles gambiae fertility.
In recent years, there has been much theoretical design and modeling work concerning the use of CRISPR/Cas9 as a gene drive system. Gene drives have great potential for pest control, and could be used for the purposes of modifying or suppressing populations.
The authors originally developed a gene drive system to disrupt a gene involved in female fertility and this was the first empirical study on CRISPR/Cas9-based gene drives for the purposes of population suppression
Initially, heterozygous individuals for the gene drive transmitted the drive to >99% of their offspring and females had an approximately 90% reduction in fertility. However, after long-term monitoring the frequency of the gene drive decreased to less than 20% by G12 and multiple nuclease homing resistant alleles had evolved. Interestingly, the resistant alleles differed in their ability to restore fertility.
By mapping sequences against the Anopheles gambiae genome, the authors observed deletions with a wide range of sizes from G2 to G12, with the population of indels becoming less diverse. In later generations, the 4 most abundant resistant alleles were those with in-frame deletions or insertions. The same deletions ere observed in mutiple haplotypes.
To test if mutant alleles conferred resistance to homing and if the in-frame indels did not disrupt target gene function, Hammond et al. crossed heterozygous females from G20 with wild type males. Most crosses generated viable progeny at rates significantly higher than previously observed (~99%). The offspring also showed non-biased inheritance of the gene drive construct (~51%), consistent with Mendelian segregation.
Hammond et al. also determined if decreased homing was due to mutations in the target site or to underlying mutations in the coding sequences of Cas9 or the gRNAs, resulted in transmission rates significantly higher than expected by Mendelian inheritance, but lower than previously reported. The authors speculate that the reduced gene drive activity could be attributed to maternally-loaded Cas9, leading to indels in zygotes of heterozygous mothers. This indel formation is significant because zygotes carrying a wild-type allele could be resistant to further cutting, that would be required for the propagation of the gene drive. This is a possible avenue for resistance formation and an embryonic end-joining rate of 79.6% of wild type alleles being converted to cleavage-resistant alleles.
The resistance to gene drive could be a major limitation of this technology and warrants thorough investigation. The results of Hammond et al agree with the current literature that indicates the propensity of a pest to develop resistance to drive may likely be species and construct specific.
Hammond et al. also stress the likely importance of population structure and genetic diversity in natural populations as a likely limitation of current drive architecture and this will warrant design alterations in order to achieve successful drive.
This paper should be of special interest to gene drive developers. Key recommendations in this paper are to have target sites with low tolerance for sequence variation, and to select for sites with a lower propensity for the production of indels that can confer resistance to drive and restore target gene function. Perhaps most interesting is the authors’ observation of embryonic end joining and their emphasis on the importance of using tightly regulated, early germline promoters to drive expression of the nuclease to reduce maternal effects.