Pervasive hybridizations in the history of wheat relatives

Abstract

Bread wheat and durum wheat derive from an intricate evolutionary history of three genomes, namely A, B and D, present in both extent diploid and polyploid species. Despite its importance for wheat research, no consensus on the phylogeny of the wheat clade has emerged so far, possibly because of hybridizations and gene flows that make phylogeny reconstruction challenging. Recently, it has been proposed that the D genome originated from an ancient hybridization event between the A and B genomes1. However, the study only relied on four diploid wheat relatives when 13 species are accessible. Using transcriptome data from all diploid species and a new methodological approach, we provide the first comprehensive phylogenomic analysis of this group. Our analysis reveals that most species belong to the D-genome lineage and descend from the previously detected hybridization event, but with a more complex scenario and with a different parent than previously thought. If we confirmed that one parent was the A genome, we found that the second was not the B genome but the ancestor of Aegilops mutica (T genome), an overlooked wild species. We also unravel evidence of other massive gene flow events that could explain long-standing controversies in the classification of wheat relatives. We anticipate that these results will strongly affect future wheat research by providing a robust evolutionary framework and refocusing interest on understudied species. The new method we proposed should also be pivotal for further methodological developments to reconstruct species relationship with multiple hybridizations.

Domestication and crop evolution of wheat and barley: Genes, genomics, and future directions

“…These three ancestors, and other grasses within the Triticeae, are related to one another by descent and through ancestral hybridization (Marcussen et al ). Further, each of the ancestors has apparently undergone ancient polyploidization events, followed by subsequent reversion to diploid states (Pont and Salse ; Glemin et al ).…”

Homoploid hybridization signals due to ancestral subdivision: a case study on the D lineage in wheat

“…Compared with barley, rye could be considered as a closer outgroup to determine the evolutionary relationships among the three diploid donors of bread wheat (Glemin et al 2018). To assess this feasibility, we re-examined their relationships based on genome sequences.…”

“…Third, restricted chromosomal recombination would be expected in hybrids involving distantly related genotypes (Ungerer et al 1998; Buerkle et al 2007) as well as genotypes with different chromosomal structures (Barb et al 2014). However, large parental blocks from either the A or B genome were not detected on any of the D chromosomes (Marcussen et al 2014; Glemin et al 2018). Restricted recombination does not even exist in either of the chromosomal segments corresponding to the relative 4/5 translocation (supplementary fig.1).…”

“…Based on analyses of the evolutionary dynamics of transposon elements and mutations, El Baidouri et al (2017) also reached the conclusion that the D lineage was derived from a complex history of multiple rounds of HH between ancient A and S (progenitor of the B subgenome) lineages as well as other relatives. Based on the analysis of transcriptome data, Glemin et al (2018) concluded that pervasive HH events were involved in the evolution of these diploid wheat lineages, and that most of those belonging to the D lineage were derived from HHS between the A and B lineages.…”

“…The key evidence used in arguing for HHS of the D lineage is that, based on shared variations (gene trees, gene content, transposon elements (TE) and single nucleotide polymorphism (SNP)), the A and B lineages are more closely related to D individually than to each other (Marcussen et al 2014; El Baidouri et al 2017; Glemin et al 2018). It is believed that, under the condition of random mating, shared variations concordant with the species tree should have the highest probability, while the other two discordant with the species tree should be equal in a bifurcating speciation.…”