Modest transcriptomic response to polyploidization in allohexaploid wheat synthetics

Abstract

Bread wheat is a recent allohexaploid (genomic constitution AABBDD) that emerged through a hybridization between tetraploid Triticum turgidum (AABB) and diploid Aegilops tauschii (DD) <10,000 years ago. The hexaploidization can be re-created artificially, producing synthetic wheat that has been used to study immediate genomic responses to polyploidization. Here we produced several synthetic wheats from alternative parental genotypes and reciprocal crosses, and examined transcriptomes from two different tissues and successive generations. We did not detect a massive reprogramming in gene expression, with only [~]1% of expressed genes showing significant differences compared to their lower-ploidy parents. Most of the differential expression is located on the D subgenome, without consistency in the direction of the expression change. Leaves and developing endosperm show distinct patterns of homoeologous expression bias, and almost non-overlapping sets of differentially expressed genes, implying that the polyploidization-triggered reprogramming is not effectuated through permanent (epi)genetic changes. While 0-3 families of transposable elements (TEs) became upregulated in wheat synthetics, we did not detect any significant association between TEs and the expression of nearby genes. We conclude that the modest tissue-specific and partially genotype-specific transcriptomic response to polyploidization is likely caused by rare incompatibilities of parental regulomes, and we discuss the pitfalls of transcriptomic comparisons across ploidy levels that can inflate the de-regulation signal.