Recently, Oberhofer et al (2018) published a paper examining the mechanisms of homing endonuclease gene drives in The Proceedings of the National Academy of Sciences. They used a novel nuclease-encoding cassette containing four multiplexed gRNAs targeting genes required for viability or female fertility in Drosophila melanogaster.
Gene drives transmit alleles through a population in a super-Mendelian fashion using homing endonuclease genes (HEGs). The HEG encodes a site-specific endonuclease that creates a double stranded DNA break in susceptible heterozygotic progeny and is used as a template to repair that break through homologous recombination, effectively creating a homozygote containing the HEG. HEGs have been proposed as a tool for population suppression in target organisms by introducing a fitness cost with the HEG. These HEGs would target recessive genes in only germline cells. However, targeted endonucleases are sensitive to both genetic variability at the target site and mutations introduced through error-prone nonhomologous end joining (NHEJ), which prevent targeting and cleavage by gRNA/Cas9 complexes. Oberhofer et al. examine transmission dynamics of multiplexed gRNA HEG cassettes to gain insight into the mechanics of CRISPR/Cas9 gene drives targeting four different loci simultaneously.
The two recessive target genes for each construct are yellow-g (yg) and deformed (Dfd), which produce collapsed egg and embryonic lethality phenotypes, respectively. The HEG constructs contain Cas9 under the control of the nanos germline promoter and a 3xP3-td-tomato dominant marker, both flanked by two gRNAs under the control of a U6 promoter and targeting different loci of either yg or dfd, all of which is contained within ~1 kb homology arms to the corresponding genes. They crossed 25 male and female HEG heterozygote flies (HEG/+) with flies lacking the target loci (Df) on one allele and carrying a version of the TM3,Sb balancer chromosome with the dominant Sb marker on the other (Df/TM3,Sb). The resulting phenotypes indicate whether homing and cleavage occur in the progeny using either the td-tomato marker or loss-of-function and absence of td-tomato phenotypes, respectively.
All but the female Dfd-HEG crosses produced viable eggs, presumably because of maternal carryover of Cas9/gRNA complexes from the zygote to the developing embryo, cleaving the functional alleles and preventing embryogenesis. The remaining crosses had average homing rates ranging from 19-33% with a maximum of 83% and cleavage rates ranging from 89-100%. Maternal carryover-dependent cleavage of paternal alleles in somatic tissue will prevent the spread of suppression HEGs but Oberhofer et al suggest that more stringent germline promoters or degron-incorporated Cas9 might ameliorate this issue. In viable, but sterile progeny from the yg-HEG/+ crosses, all flies lacked the td-tomato marker and had diverse Cas9-mediated deletions and incomplete homing as indicated through sequencing the target region. Drive experiments seeding 12 successive generations with either yg-HEG or dfd-HEG progeny showed that there was an initial increase in the HEG allele but that it was quickly eliminated from the population due to fitness costs.
Maintaining suppressive gene drives requires resistance-proof constructs that are adaptable to heterogenous populations. Multiplexing gRNAs to target different loci yields promising results in mathematical models but has only been tested once before this publication. The experiments of Oberhofer et al explore the temporal and spatial dynamics of homology-directed repair and are a breakthrough in our understanding of gene drive mechanics.