Novel CRISPR/Cas9 gene drive constructs reveal insights into mechanisms of resistance allele formation and drive efficiency in genetically diverse populations

Abstract

A functioning gene drive system could fundamentally change our strategies for the control of vector-borne diseases by facilitating rapid dissemination of transgenes that prevent pathogen transmission or reduce vector capacity. CRISPR/Cas9 gene drive promises such a mechanism, which works by converting cells that are heterozygous for the drive construct into homozygotes, thereby enabling super-Mendelian inheritance. Although CRISPR gene drive activity has already been demonstrated, a key obstacle for current systems is their propensity to generate resistance alleles, which cannot be converted to drive alleles. In this study, we developed two CRISPR gene drive constructs based on the nanos and vasa promoters that allowed us to illuminate the different mechanisms by which resistance alleles are formed in the model organism Drosophila melanogaster. 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. Assessment of drive activity in genetically diverse backgrounds further revealed substantial differences in conversion efficiency and resistance rates. Our results demonstrate that the evolution of resistance will likely impose a severe limitation to the effectiveness of current CRISPR gene drive approaches, especially when applied to diverse natural populations.

Fig 1. Gene drive constructs.

(A) Our nanos-based gene drive construct consists of a Cas9 gene driven by the nanos promoter and followed by a nanos 3’ UTR. It also contains a dsRed fluorescent marker driven by the 3xP3 promoter, and a gRNA gene driven by the U6:3 promoter. Regions corresponding to the yellow gene flank the gene drive components on both ends. (B) Our vasa-based gene drive construct has a similar design, but with vasa elements replacing the nanos ones and flanked by yellow promoter sequences.

Fig 2. Mechanisms and rates of resistance allele formation.

In a heterozygous female with genotype D/+, expression of Cas9 in a germline cell can produce one of three outcomes: (i) successful conversion of the wild type allele into a drive allele by HDR, (ii) formation of a resistance allele when HDR is incomplete or cleavage is repaired by NHEJ, or (iii) continuing presence of the wild type allele if no cleavage occurred or was perfectly repaired. For our nanos construct in the w¹¹¹⁸ line, we observed successful germline conversion in D/+ females at a rate of approximately 60%, leaving 80% of gametes with gene drive alleles. Almost all remaining gametes (20%) contained resistance alleles, with only less than 1% of gametes carrying a wild type allele. (iv) Note that some of these resistance alleles could have formed during later stages of meiosis or in the gamete when persistent Cas9 cleaved the target while there was no template available for HDR. (v) We also observed the formation of resistance alleles in early female embryos after fertilization by a wild type male, suggesting post-fertilization activity of maternal Cas9 after which cleavage was repaired by NHEJ despite the presence of a template available for HDR. In those embryos that originated from a D/+ or D/D mother, we observed such post-fertilization formation of a resistance allele on the paternal chromosome in approximately 30% of embryos. (vi) In embryos that originated from D/r2 mothers, we observed this in only 20% of cases, consistent with presumably lower Cas9 levels in the eggs that would also be found in the r2 gametes from D/+ mothers in the figure. (vii) Formation of resistance alleles was also observed in embryos that did not receive any copy of a drive allele, although formation rates may be lower in this case. (viii) A small number of embryos that inherited wild type alleles from both parents may even have experience double cleavage to form two resistance alleles. Note that any formation of resistance alleles in the embryo may result in mosaicism of adult individuals, as we frequently observed in our crosses. Tables 1–3 in S1 Dataset provide the calculations used for inference of these rates from the phenotypes of progeny in our crosses.