How much can history constrain adaptive evolution? A real‐time evolutionary approach of inversion polymorphisms in Drosophila subobscura

Abstract

Chromosomal inversions are present in a wide range of animals and plants, having an important role in adaptation and speciation. Although empirical evidence of their adaptive value is abundant, the role of different processes underlying evolution of chromosomal polymorphisms is not fully understood. History and selection are likely to shape inversion polymorphism variation to an extent yet largely unknown. Here, we perform a real-time evolution study addressing the role of historical constraints and selection in the evolution of these polymorphisms. We founded laboratory populations of Drosophila subobscura derived from three locations along the European cline and followed the evolutionary dynamics of inversion polymorphisms throughout the first 40 generations. At the beginning, populations were highly differentiated and remained so throughout generations. We report evidence of positive selection for some inversions, variable between foundations. Signs of negative selection were more frequent, in particular for most cold-climate standard inversions across the three foundations. We found that previously observed convergence at the phenotypic level in these populations was not associated with convergence in inversion frequencies. In conclusion, our study shows that selection has shaped the evolutionary dynamics of inversion frequencies, but doing so within the constraints imposed by previous history. Both history and selection are therefore fundamental to predict the evolutionary potential of different populations to respond to global environmental changes.

Predictable phenotypic, but not karyotypic, evolution of populations with contrasting initial history

“…In particular, the higher frequency of the so called ‘cold-adapted’ inversions in northern populations and ‘warm-adapted’ inversions in populations of more southern locations 32 leads to the expectation that populations sampled from different locations along this latitudinal cline might adapt differently when placed in a novel laboratory environment featuring a constant, mild temperature, such as that imposed in our lab (18 °C). We found a clear initial impact of history at the start of the experiment, with quick convergence during laboratory adaptation for several life-history, physiological, and morphological traits 26, 27 but not for chromosomal inversion frequencies 35 .…”

“…Recently, we have addressed the impact of historical effects on the evolutionary dynamics of initially differentiated D. subobscura populations in a novel laboratory environment 26, 27, 35 . In contrast to previous studies where we had sampled populations from nearby locations in Portugal 29 , we took advantage of the well-known European latitudinal cline in this species (see above), sampling wild populations from contrasting latitudes, which we know are genetically differentiated 26 .…”

Evolution of mating behavior between two populations adapting to common environmental conditions

“…Interestingly, these populations also did not show convergence at the inversion frequency level (Fragata et al. 2014b). Perhaps the lack of convergence observed for mating behavior relates with effects of inversions that maintained differences between populations.…”

“…These populations presented initial high differentiation both in life-history traits and inversion frequencies (Fragata et al. 2014a,b). The degree of early phenotypic differentiation among such populations was much higher than that shown by populations derived from neighboring locations (Fragata et al.…”

History, chance and selection during phenotypic and genomic experimental evolution: replaying the tape of life at different levels

“…Interestingly, and in contrast to what we found for phenotypic traits, history played an important role in the evolutionary dynamics of inversions. Though we obtained clear evidence for the role of selection, after 40 generations of laboratory evolution, populations remained differentiated at the inversion frequencies level (Fragata et al, 2014b). Genome-wide analyses on several generations of these populations are underway to clarify how much convergence occurred at the genome level, and its association with the genetic content of chromosomal inversions.…”