Epistasis modifies the dominance of loci causing hybrid male sterility in the Drosophila pseudoobscura species group

Abstract

Speciation, the evolution of reproductive isolation between populations, serves as the driving force for generating biodiversity. Postzygotic barriers to gene flow, such as F(1) hybrid sterility and inviability, play important roles in the establishment and maintenance of biological species. F(1) hybrid incompatibilities in taxa that obey Haldane's rule, the observation that the heterogametic sex suffers greater hybrid fitness problems than the homogametic sex, are thought to often result from interactions between recessive-acting X-linked loci and dominant-acting autosomal loci. Because they play such prominent roles in producing hybrid incompatibilities, we examine the dominance and nature of epistasis between alleles derived from Drosophila persimilis that confer hybrid male sterility in the genetic background of its sister species, D. pseudoobscura bogotana. We show that epistasis elevates the apparent dominance of individually recessive-acting QTL such that they can contribute to F(1) hybrid sterility. These results have important implications for assumptions underlying theoretical models of hybrid incompatibilities and may offer a possible explanation for why, to date, identification of dominant-acting autosomal "speciation genes" has been challenging.

Figure 1. Epistasis between hybrid incompatibility QTL

Epistasis necessarily occurs between incompatibility-conferring loci from different species (a). However, epistasis may also occur between multiple loci within a single species (b) and/or between multiple loci from both species (c). Open rectangles indicate incompatibility loci. Dashed arrows indicate epistatic interactions between loci. Note that not all possible interactions are depicted.

Figure 2

A. Crossing scheme for single introgression. D. persimilis males were crossed to D. p. bogotana females. F₁ females were backcrossed to D. p. bogotana males for 10 generations until the genetic background was devoid of D. persimilis alleles except for the QTL of interest. Three introgression lines were created for each autosomal QTL (Q2, Q3, and Q4). Males homozygous for each QTL introgression were generated by self-crossing within each line. B. Example of cross for multiple introgressions. Example of a cross to generate male flies heterozygous for both the 2^nd - and the 3^rd-chromosome QTL.

Figure 3. Effect on male sterility of single QTL introgressions

Percent of males sterile for each genotype at the three QTL is given. White rectangles represent homozygous D. p. bogotana genotype. Grey rectangles represent heterozygous D. persimilis / D. p. bogotana genotype. Black rectangles represent homozygous D. persimilis genotype.

Figure 4. Effect on male sterility of multiple QTL introgressions

Percent of males sterile for each genotype at the three QTL is given. White rectangles represent homozygous D. p. bogotana genotype. Grey rectangles represent heterozygous D. persimilis / D. p. bogotana genotype. Black rectangles represent homozygous D. persimilis genotype.