Abstract
Hybridization of closely related plant species is frequently connected to endosperm arrest and seed failure, for reasons that remain to be identified. In this study, we investigated the molecular events accompanying seed failure in hybrids of the closely related species pair Capsella rubella and C. grandiflora. Mapping of QTL for the underlying cause of hybrid incompatibility in Capsella identified three QTL that were close to pericentromeric regions. We investigated whether there are specific changes in heterochromatin associated with interspecific hybridizations and found a strong reduction of chromatin condensation in the endosperm, connected with a strong loss of CHG and CHH methylation and random loss of a single chromosome. Consistent with reduced DNA methylation in the hybrid endosperm, we found a disproportionate deregulation of genes located close to pericentromeric regions, suggesting that reduced DNA methylation allows access of transcription factors to targets located in heterochromatic regions. Since the identified QTL were also associated with pericentromeric regions, we propose that relaxation of heterochromatin in response to interspecies hybridization exposes and activates loci leading to hybrid seed failure.
Publication types
-
Research Support, Non-U.S. Gov't
MeSH terms
-
Capsella / classification
-
Capsella / genetics*
-
Centromere / genetics
-
Chromatin / genetics*
-
Chromatin / metabolism
-
Chromosome Aberrations
-
DNA Methylation
-
Endosperm / genetics*
-
Gene Expression Regulation, Plant
-
Genes, Plant / genetics
-
Heterochromatin / genetics
-
Heterochromatin / metabolism
-
Hybridization, Genetic*
-
Quantitative Trait Loci / genetics
-
Seeds / genetics*
-
Species Specificity
Substances
-
Chromatin
-
Heterochromatin
Grants and funding
This research was supported by grants from the Swedish Research Council VR (to CK, grant #2017-04119), a grant from the Knut and Alice Wallenberg Foundation (to CK, grant #2018-0206), and support from the Göran Gustafsson Foundation for Research in Natural Sciences and Medicine (to CK). The work of BL and TS was supported by grants from the Science for Life Laboratory and the Swedish Research Council (grant #621-2010-5508 and #621-2013-4320). Computational work was enabled by resources provided by the Swedish National Infrastructure for Computing (SNIC) at UPPMAX partially funded by the Swedish Research Council through grant agreement no. 2016-07213. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.