A high-resolution crossover landscape in Drosophila santomea reveals rapid and concerted evolution of multiple properties of crossing over control.

Abstract

Crossing over is a fundamental process in sexually reproducing species, ensuring proper chromosome segregation during gamete formation and generating new allelic combinations that enhance adaptation. Despite its essential role, genes involved in crossing over evolve rapidly and there is extensive variation in the rate and genomic distribution of crossovers across species. Considering this rapid evolution, identifying differences between very closely related species is crucial for understanding the molecular basis of natural variation in crossing over control. Here, we present a genome-wide, high-resolution crossover map for Drosophila santomea and compare it with those of its sister species D. yakuba and the more distantly related D. melanogaster. Upon examining 784 individual meiotic products based on an experimental design that captures intraspecific variation in crossing over control, we identified 2,288 crossovers genome-wide. Our analyses reveal striking differences in crossover patterns between D. santomea and D. yakuba despite their recent split only 400,000 years ago and sharing a significant amount of ancestral polymorphism. The D. santomea X chromosome shows a major reduction in genetic length compared to D. yakuba (62.7 cM vs. 93.8 cM), while autosomes show a slight increase (262.6 vs. 245.6 cM), resulting in overall genetic maps of 324.2 cM for D. santomea and 339.3 cM for D. yakuba. All D. santomea autosomal arms show a significant reduction of the centromere effect relative to D. yakuba, more closely resembling D. melanogaster autosomes. At the same time, estimates of crossover interference indicate weaker intensity across all autosomal arms in D. santomea compared to D. yakuba, while the X chromosome exhibits considerably stronger interference. These findings suggest a link between the intensity of crossover interference and the centromere effect. We propose that stronger crossover interference is associated with a smaller crossover-competent region-determined by the combined centromere and telomere effects-to prevent the deleterious consequences of multiple crossovers occurring too close together. Finally, we examined whether the D. santomea X chromosome exhibits the crossover-associated meiotic drive mechanism (MDCO) reported in D. yakuba, in which chromatids with crossovers are preferentially included in oocytes. Tetrad analysis of the D. santomea X chromosome revealed no evidence of an active MDCO, potentially explaining the reduced crossover rates observed on this chromosome relative to D. yakuba even though the numbers of meiosis I crossovers may be similar in both species.