Genomic basis of parallel adaptation varies with divergence in Arabidopsis and its relatives

Abstract

Parallel adaptation provides valuable insight into the predictability of evolutionary change through replicated natural experiments. A steadily increasing number of studies have demonstrated genomic parallelism, yet the magnitude of this parallelism varies depending on whether populations, species, or genera are compared. This led us to hypothesize that the magnitude of genomic parallelism scales with genetic divergence between lineages, but whether this is the case and the underlying evolutionary processes remain unknown. Here, we resequenced seven parallel lineages of two Arabidopsis species, which repeatedly adapted to challenging alpine environments. By combining genome-wide divergence scans with model-based approaches, we detected a suite of 151 genes that show parallel signatures of positive selection associated with alpine colonization, involved in response to cold, high radiation, short season, herbivores, and pathogens. We complemented these parallel candidates with published gene lists from five additional alpine Brassicaceae and tested our hypothesis on a broad scale spanning ∼0.02 to 18 My of divergence. Indeed, we found quantitatively variable genomic parallelism whose extent significantly decreased with increasing divergence between the compared lineages. We further modeled parallel evolution over the Arabidopsis candidate genes and showed that a decreasing probability of repeated selection on the same standing or introgressed alleles drives the observed pattern of divergence-dependent parallelism. We therefore conclude that genetic divergence between populations, species, and genera, affecting the pool of shared variants, is an important factor in the predictability of genome evolution.

Parallel adaptation in autopolyploid Arabidopsis arenosa is dominated by repeated recruitment of shared alleles

“…2a and Supplementary Data 4 ) and we hereafter refer to this as ‘significant parallelism’. Such a fraction of parallel gene candidates (0.02–0.04 out of all candidates from that particular population pair) is in line with other naturally adapting systems of comparable divergence 33 , 38 – 40 . The parallel differentiation candidates were significantly enriched ( p < 0.05) for GO terms, such as regulation of ion transmembrane transport, voltage-gated calcium and potassium channel activity or plasma membrane (Supplementary Data 4 ).…”

“…We chose to place selected sites at eight locations at equal distance (default value recommended by authors https://github.com/kristinmlee/dmc ) from each other along the particular gene. Such a density (one site per ~500 bp on average, as the mean length of serpentine adaptation candidate is ~4,000 bp) is in fact well within the range of the LD decay of 150–800 bp, which was estimated in A. arenosa 40 . For the calculation of co-ancestry decay, we also considered a 25 kbp upstream and downstream region from each gene.…”

“…We used the custom script ( https://github.com/mbohutinska/ScanTools_ProtEvol ) based on ScanTools pipeline that has been successfully applied in our previous analyses of autotetraploid A. arenosa 8 . The window size of 1 kbp was selected to properly account for the average genome-wide linkage disequilibrium (LD) decay of genotypic correlations (150–800 bp) previously estimated in autotetraploid A. arenosa 40 . We used windows of fixed length and thus with homogeneous position on genome across all population pairs (in contrast to windows defined by number of SNPs and thus varying in exact position and length) to facilitate comparisons of selection candidates across distinct population pairs.…”

“…49 . Further, we considered the case significantly non-neutral only if the MCL difference between the selected parallel model and the corresponding neutral model was higher than the maximum of the distribution of the differences from the simulated neutral data in A. arenosa inferred in Bohutínská et al 40 (i.e. MCL difference >21, a conservative estimate).…”

Highly Replicated Evolution of Parapatric Ecotypes

“…These results from our current work in combination with strong divergent selection between ecotypes ( Melo et al 2014 ; Richards and Ortiz-Barrientos 2016 ; Richards et al 2016 ; Walter et al 2016 ; Walter, Wilkinson, et al 2018 ; Wilkinson et al 2021 ), strong trait–nvironment association in the system ( Walter, Aguirre, et al 2018 ; James et al 2020 ) and adaptation across replicate populations occurring mainly via mutations in different genes ( Roda, Liu, et al 2013 ; James et al 2020 ; Wilkinson et al 2021 ), implies that selection has independently driven the parallel evolution of coastal populations in this system. This makes S. lautus one of the clearest examples of the parallel evolution of ecotypes discovered yet, adding to the increasing number of potential cases of parallel evolution in plants ( Foster et al 2007 ; Trucchi et al 2017 ; Cai et al 2019 ; Bohut í nsk á et al 2021 ). It also positions the species as a powerful system of replicated parapatric divergence to study the origin of adaptation and reproductive isolation.…”