Abstract: Meiosis is essential for fertility of sexual eukaryotes and its core structures and progression are conserved across kingdoms. Nevertheless, meiotic proteins are often less conserved in primary sequence than we might expect, and sometimes show evidence of having experienced directional selection. Why? What challenges does meiosis face that might cause it to evolve adaptively and how does this alter the system? We study two important factors that can challenge the stability of meiosis and drive evolutionary responses: whole genome duplication and temperature. My group seeks to understand how meiosis evolves in response to challenges, that is, what its evolutionary potential is within the constraints of being an essential and complex structural progression. We use Arabidopsis arenosa, which occurs naturally as an autotetraploid and a diploid, and where both cytotypes have colonized a range of habitats. In a genome scan for adaptation to whole genome duplication, we found that eight interacting meiotic proteins critical for axis formation and synapsis show strong evidence of having been under selection in the tetraploid arenosas. This is associated with a reduction in crossover number, and a greater tendency for terminal localization of chiasmata. More recently, we found that two of the same genes under selection in tetraploids, also show strong evidence of having been under selection in a diploid A. arenosa lineage. This lineage colonized a warmer lowland habitat (the ancestral form is found in cooler mountain environments), we have evidence that this lineage evolved greater temperature tolerance of meiosis, and also found that they, too have fewer and more distal crossovers in lab conditions. Distinct alleles of the same genes were under selection after both whole genome duplication and habitat colonization, and thus have twice come under selection for apparently distinct reasons; does this suggest evolutionary constraint on the system? Does it indicate that the same processes are challenged by distinct stresses?
The finding that the genes that came under selection in both lineages are known to interact also highlights the possible need for co-evolution of interacting partners in meiotic evolution, which may be more broadly relevant to the evolution of proteins that participate in large complexes.