Seasonal changes in recombination characteristics in a natural population of Drosophila melanogaster

Abstract

Environmental seasonality is a potent evolutionary force, capable of maintaining polymorphism, promoting phenotypic plasticity and causing bet-hedging. In Drosophila, environmental seasonality has been reported to affect life-history traits, tolerance to abiotic stressors and immunity. Oscillations in frequencies of alleles underlying fitness-related traits were also documented alongside SNPs across the genome. Here, we test for seasonal changes in two recombination characteristics, crossover rate and crossover interference, in a natural D. melanogaster population from India using morphological markers of the three major chromosomes. We show that winter flies, collected after the dry season, have significantly higher desiccation tolerance than their autumn counterparts. This difference proved to hold also for hybrids with three independent marker stocks, suggesting its genetic rather than plastic nature. Significant between-season changes are documented for crossover rate (in 9 of 13 studied intervals) and crossover interference (in four of eight studied pairs of intervals); both single and double crossovers were usually more frequent in the winter cohort. The winter flies also display weaker plasticity of both recombination characteristics to desiccation. We ascribe the observed differences to indirect selection on recombination caused by directional selection on desiccation tolerance. Our findings suggest that changes in recombination characteristics can arise even after a short period of seasonal adaptation (~8-10 generations).

Fig. 1. The location of the studied markers on the genetic map (A) and the overall scheme of experiment (B).

A The markers’ positions are stated in map units (cM). B Only crosses with the marker stock for one autosome are shown as an example.

Fig. 2. The effect of season on desiccation tolerance in the autumn and the winter parental isofemale lines and their hybrids with three marker stocks.

The effect is assessed using nested ANOVA, N = 70 per season (seven lines per season, ten replicates per line). The asterisks show the significance of the F-statistics: **p < 0.01; ***p < 0.001.

Fig. 3. The effect of season on crossover rates in hybrids reared in normal conditions.

The effect is assessed using the t-test for independent samples, N = 7 per season. The asterisks show the FDR-corrected significance of the t-statistics: *p_FDR < 0.1; **p_FDR < 0.01; ***p_FDR < 0.001.

Fig. 4. The effect of desiccation stress on crossover rates in the autumn (A) and the winter (B) hybrids.

The effect is assessed using the t-test for paired samples, N = 7 per season. The asterisks show the FDR-corrected significance of the t-statistics: *p_FDR < 0.1; **p_FDR < 0.01; *** p_FDR < 0.001.