Diverse genome organization following 13 independent mesopolyploid events in Brassicaceae contrasts with convergent patterns of gene retention

Abstract

SummaryHybridization and polyploidy followed by genome-wide diploidization significantly impacted the diversification of land plants. The ancient At-α whole-genome duplication (WGD) preceded the diversification of crucifers (Brassicaceae). Some genera and tribes also experienced younger, mesopolyploid WGDs concealed by subsequent genome diploidization. Here we tested if multiple base chromosome numbers originated due to genome diploidization after independent mesopolyploid WGDs and how diploidization impacted post-polyploid gene retention. Sixteen species representing ten Brassicaceae tribes were analyzed by comparative chromosome painting and/or whole-transcriptome analysis of gene age distributions and phylogenetic analyses of gene duplications. Overall, we found evidence for at least 13 independent mesopolyploidies followed by different degrees of diploidization across the Brassicaceae. New mesotetraploid events were uncovered for tribes Anastaticeae, Iberideae and Schizopetaleae, and mesohexaploid WGDs for Cochlearieae and Physarieae. In contrast, we found convergent patterns of gene retention and loss among these independent WGDs. Our combined analyses of Brassicaceae genomic data indicate that the extant chromosome number variation in many plant groups, and especially polybasic but monophyletic taxa, can result from clade-specific genome duplications followed by diploidization. Our observation of parallel gene retention and loss across multiple independent WGDs provides one of the first multi-species tests that post-polyploid genome evolution is predictable.Significance statementOur data show that multiple base chromosome numbers in some Brassicaceae clades originated due to genome diploidization following multiple independent whole-genome duplications (WGD). The parallel gene retention/loss across independent WGDs and diploidizations provides one of the first tests that post-polyploid genome evolution is predictable.