Genetic mapping of adaptive wing size variation in Drosophila simulans

Abstract

Many ecologically important traits exhibit latitudinal variation. Body size clines have been described repeatedly in insects across multiple continents, suggesting that similar selective forces are shaping these geographical gradients. It is unknown whether these parallel clinal patterns are controlled by the same or different genetic mechanism(s). We present here, quantitative trait loci (QTL) analysis of wing size variation in Drosophila simulans. Our results show that much of the wing size variation is controlled by a QTL on Chr 3L with relatively minor contribution from other chromosome arms. Comparative analysis of the genomic positions of the QTL indicates that the major QTL on Chr 3 are distinct in D. simulans and D. melanogaster, whereas the QTL on Chr 2R might overlap between species. Our results suggest that parallel evolution of wing size clines could be driven by non-identical genetic mechanisms but in both cases involve a major QTL as well as smaller effects of other genomic regions.

Figure 1

Genetic architecture of wing size variation in D. simulans males. Chart shows the average wing centroid size of the parental lines, F1's and the backcross progeny. The symbol ‘S' on the horizontal axis represents a chromosome derived from L20, or the small line; ‘B' represents a chromosome derived from L71, or the big line; ‘Y' represents the Y chromosome. The order of the letters represents chromosome number. For example, SY:SS:SB=males carrying an X chromosome from the small line, two copies of chromosome 2 from the small line, one copy of chromosome 3 from the small line and the other copy from the big line. Error bars represent s.e. values of the mean wing centroid size based on 15–45 individuals.

Figure 2

QTL analysis of wing size variation in D. simulans. QTL analysis using 47 markers covering chromosomes 2 and 3 revealed one major peak on Chr 3L and another on Chr 2R. The percentage phenotypic variance explained by a QTL is indicated above each peak. The y-axis is the logarithm (to the base 10) of odds (LOD) score. The horizontal bar at LOD=2.5 represents the experimental-wise significance threshold at P<0.05 with 10 000 permutations. Triangles on the x-axis indicate marker positions. The horizontal bar underneath each significant QTL represents a one-LOD support interval, with a black dot indicating the peak marker within each interval. The lower graph indicates the additive components along the chromosomes.

Figure 3

Absence of epistatic interaction between QTLs on Chr 3L and Chr 2R. Mean wing centroid size in mm is plotted against the four genotypic combinations. ‘L318=1' on the x-axis denotes the presence of the L71 allele at the L318 locus; likewise for marker L219. Error bars represent s.e. values of the mean wing centroid size.

Figure 4

Comparative mapping of wing size QTLs in D. simulans and D. melanogaster. The four homologous chromosome arms (Muller elements B, C, D and E) are aligned using orthologous markers. A major chromosomal inversion between the two species is evident on Chr 3R, but the linear order of the gene markers used is conserved on other chromosome arms. Solid bars indicate locations of wing size QTL regions. A circle represents the approximate position of each QTL peak in either species. The D. melanogaster QTL information is based on Calboli et al. (2003). The star symbol indicates the location of Dca, the candidate gene for body size variation in McKechnie et al. (2010).