Introgression of a synthetic sex ratio distortion transgene into different genetic backgrounds of Anopheles coluzzii

Abstract

The development of genetically modified mosquitoes (GMM) and their subsequent field release offers innovative approaches for vector control of malaria. A non-gene drive self-limiting male-bias Ag(PMB)1 strain has been developed in a 47-year-old laboratory G3 strain of Anopheles gambiae s.l. When Ag(PMB)1 males are crossed to wild-type females, expression of the endonuclease I-PpoI during spermatogenesis causes the meiotic cleavage of the X chromosome in sperm cells, leading to fertile offspring with a 95% male bias. However, World Health Organization states that the functionality of the transgene could differ when inserted in different genetic backgrounds of Anopheles coluzzii which is currently a predominant species in several West-African countries and thus a likely recipient for a potential release of self-limiting GMMs. In this study, we introgressed the transgene from the donor Ag(PMB)1 by six serial backcrosses into two recipient colonies of An. coluzzii that had been isolated in Mali and Burkina Faso. Scans of informative Single Nucleotide Polymorphism (SNP) markers and whole-genome sequencing analysis revealed a nearly complete introgression of chromosomes 3 and X, but a remarkable genomic divergence in a large region of chromosome 2 between the later backcrossed (BC6) transgenic offspring and the recipient paternal strains. These findings suggested to extend the backcrossing breeding strategy beyond BC6 generation and increasing the introgression efficiency of critical regions that have ecological and epidemiological implications through the targeted selection of specific markers. Disregarding differential introgression efficiency, we concluded that the phenotype of the sex ratio distorter is stable in the BC6 introgressed An. coluzzii strains.

Keywords: anopheles; genomic divergence; introgression; malaria vector; sex-ratio distorter.

FIGURE 1

Introgression of the sex ratio distorter from the donor Ag(PMB)1 to the recipient An. Coluzzii wild‐type strains, Mali‐NIH (ML) and BF_Ac(WT) (BF). (a) Patterns of genetic admixture between An. Gambiae and An. Coluzzii for three parental strains and G3 control, using DIS genotyping assay (Lee et al., 2013). First mosquito generation used for starting the introgression process was displayed for each strain, Ag(PMB)1, ML_1st generation and BF_1st generation. Columns represent 15 discriminative SNPs for An. Coluzzi and An. Gambiae is located in islands of genomic divergence in chromosome X (7 SNPs spanning 4.4 Mbp), 2L (5 SNPs spanning 2.2 Mbp) and 3L (3 SNPs spanning 117 kbp). 100 individual mosquitoes are represented by coloured horizontal lines, with individuals stacked vertically. (b) Schematic diagram of the introgression of [3xP3‐DsRed]β2‐eGFP::I‐PpoI‐124L construct from the donor Ag(PMB)1 transgenic line developed using G3 laboratory strain (Galizi et al., 2014), into two genetic background of An. Coluzzii laboratory strains by six serial backcrosses

FIGURE 2

Following the serial introgression of the sex ratio distortion transgene (TG) from Ag(PMB)1 into two An. Coluzzzii genetic backgrounds, Mali‐NIH (ML) and BF_Ac(WT) (BF) by DIS assay. Genotypic characterization of the donor parental strain Ag(PMB)1, two wild‐type An.Coluzzii recipient colonies, ML and BF at generation 1, 2, and 7, transgenic F1 and BC1 generations produced during introgression processes, the backcrossed transgenic BC6 progenies, using the divergence island SNPs. (DIS assay, Lee et al., 2013). Columns represent 15 discriminative SNPs for An. Coluzzi and An. Gambiae. For each population, 100 individual mosquitoes are represented by coloured horizontal lines, with individuals stacked vertically.

FIGURE 3

The genomewide F_ST differences on the chromosomes 2, 3 and X between parental and introgressed transgenic strains in two genetic backgrounds (ML and BF). Smoothed F_ST‐values for (a) the wildtype recipient strain ML at generation 1 (ML_1 gen) and (b) the wildtype recipient strain BF at generation 1 (BF_1 gen) versus the transgenic donor strain Ag(PMB)1, the backcrossed transgenic BC6 progenies and the recipient wildtype strains at generation 7 used as an internal control. The 1 kb sliding window‐based approach was used to generate the F_ST‐values for each chromosome arm. The 85% threshold for each F_ST distribution is indicated with a dashed red line. Six most common inversions of chromosome 2 (2Rj, 2Rb, 2Rc, Ru, 2Rd and 2La) are shown by the coloured‐shaded areas. The three islands of divergence, the transgene integration site (TG) and Kdr locus were also reported at the top of the figure.

FIGURE 4

Distribution of the 2La chromosome inversion genotypes (homozygotes 2L+^a/2L+^a, heterozygotes 2La/2L+^a and homozygotes 2La/2La) assayed by PCR (White et al., 2007) during the serial backcrossing of transgene from Ag(PMB)1 into two An. Coluzzii genetic backgrounds (ML and BF). Karyotype frequencies (%) of the 2La chromosome inversion in the two wildtype recipient strains ML (a) and BF (b) at generation 1, 2 and 7, the transgenic donor strain Ag(PMB)1, G3 control, transgenic F1 and BC1 generations produced during introgression processes and the backcrossed transgenic BC6 progenies. The increment of 2La/2La frequency in the transgenic mosquitoes introgressed with Burkina Faso genetic background at generation number 22 (TG_BF_BC22) is also displayed in the striped area.