Sex chromosomes have evolved independently many times across eukaryotes. Despite a considerable body of literature on sex chromosome evolution, the causes and consequences of variation in their formation, degeneration, and turnover remain poorly understood. Chromosomal rearrangements are thought to play an important role in these processes by promoting or extending the suppression of recombination on sex chromosomes. Sex chromosome variation may also contribute to barriers to gene flow, limiting introgression among species. Comparative approaches in groups with sexual system variation can be valuable for understanding these questions. Rumex is a diverse genus of flowering plants harboring significant sexual system and karyotypic variation, including hermaphroditic and dioecious clades with XY (and XYY) sex chromosomes. Previous disagreement in the phylogenetic relationships among key species have rendered the history of sex chromosome evolution uncertain. Resolving this history is important for investigating the interplay of chromosomal rearrangements, introgression, and sex chromosome evolution in the genus. Here, we use new transcriptome assemblies from 11 species representing major clades in the genus, along with a whole-genome assembly generated for a key hermaphroditic species. Using phylogenomic approaches, we find evidence for the independent evolution of sex chromosomes across two major clades, and introgression from unsampled lineages likely predating the formation of sex chromosomes in the genus. Comparative genomic approaches revealed high rates of chromosomal rearrangement, especially in dioecious species, with evidence for a complex origin of the sex chromosomes through multiple chromosomal fusions. However, we found no evidence of elevated rates of fusion on the sex chromosomes in comparison with autosomes, providing no support for an adaptive hypothesis of sex chromosome expansion due to sexually antagonistic selection. Overall, our results highlight a complex history of karyotypic evolution in Rumex, raising questions about the role that chromosomal rearrangements might play in the evolution of large heteromorphic sex chromosomes.

Samples for live tissue were sourced from the USDA GRIN network (USDA Agricultural Research Service), the Southwest China Wildlife Germplasm Genobank (http://www.genobank.org/), and the collections from Spencer Barrett’s lab, totaling 10 species. Plants were grown under glasshouse conditions between 2018 and 2020. Because many Rumex species are perennials, we collected leaf, bud, and pollen tissue opportunistically based on availability. Leaf and bud tissue were collected directly into LN2 using sterilized forceps. Pollen was collected using keif boxes (WACKY WILLYS,Inc. BC, Canada) and either frozen in LN2 or germinated prior to sequencing using the medium developed in Adhikari & Campbell (1998). 

We used Trinity v2.11.0 (Grabherr et al. 2011) with each of the 10 species to create a transcriptome assembly for the sequencing libraries (representing a single tissue/individual combination). These assemblies were combined for each species and then reduced to representative species-wide transcriptomes using the traa2cds.pl script from EvidentialGene (Gilbert 2013) and retaining the primary transcript for each gene model.