Non-coding changes cause sex-specific wing size differences between closely related species of Nasonia

Abstract

The genetic basis of morphological differences among species is still poorly understood. We investigated the genetic basis of sex-specific differences in wing size between two closely related species of Nasonia by positional cloning a major male-specific locus, wing-size1 (ws1). Male wing size increases by 45% through cell size and cell number changes when the ws1 allele from N. giraulti is backcrossed into a N. vitripennis genetic background. A positional cloning approach was used to fine-scale map the ws1 locus to a 13.5 kilobase region. This region falls between prospero (a transcription factor involved in neurogenesis) and the master sex-determining gene doublesex. It contains the 5'-UTR and cis-regulatory domain of doublesex, and no coding sequence. Wing size reduction correlates with an increase in doublesex expression level that is specific to developing male wings. Our results indicate that non-coding changes are responsible for recent divergence in sex-specific morphology between two closely related species. We have not yet resolved whether wing size evolution at the ws1 locus is caused by regulatory alterations of dsx or prospero, or by another mechanism. This study demonstrates the feasibility of efficient positional cloning of quantitative trait loci (QTL) involved in a broad array of phenotypic differences among Nasonia species.

Figure 1. Wing size differences due to ws1.

Wings of N. giraulti (ws1_gG), N. vitripennis (ws1_vV) and giraulti ws1 in vitripennis background (ws1_gV_40kb). Wing area ± S.D. relative to ws1_vV males is shown (see also Table 1). Scale bar: 100 µm.

Figure 2. Positional cloning in Nasonia using linked lethals.

The diagram illustrates a screen for fine-scale recombinants (marked with *) between ws1_g and a lethal located 0.6cM away. Because of the lethal (l), the only live haploid males with ws1_g (large wings, w) are recombinant. The blue (b) marker is used as a second phenotypic marker and as a way to identify recombinants on the other side of ws1. Gray bar: introgression (N. giraulti) sequence. White bar: N. vitripennis sequence. Proportions shown are based on estimates of recombination rates between the markers (Figure 3) and assuming no double recombinants.

Figure 3. Positional cloning: ws1 maps to the doublesex locus.

(A) Recombination map of ws1 and flanking phenotypic markers used for positional cloning. (B) Genome map of the region around ws1 including gene annotations. (C) Genotype and phenotype of recombinants near ws1. Black: N. giraulti, White: N. vitripennis. (D) Map of sequence features in the 13.5kb ws1 locus. Triangles: Insertions, with size given in base pairs (bp). Letters: Microsatellite repeats, with size given in bp. Blue: dsx 5′-UTR.

Figure 4. Change in doublesex expression due to ws1.

Expression level change was estimated by quantitative RT–PCR. Note that the vitripennis dsx protein-coding region is present in both genotypes (i.e., the giraulti region of ws1_gV_40kb does not include the dsx coding sequence). Mean expression ratios ± standard errors are shown. Expression ratios greater than 1 indicate higher male dsx (dsx^M) transcript level in ws1_vV than in ws1_gV_40kb. Sample size indicates number of independent biological replicates.