Final Report Summary - MEDEA (Mechanisms of Epigenetic regulation in Development, Evolution and Adaptation)
To elucidate the molecular pathways and their complex interplay in the regulation of genomic imprinting at the MEA locus, we identified several novel factors regulating imprinting using a genetic screen for loss of imprinting of transgenic reporter gene where the MEA Imprinting Control Region (ICR), which is not methylated (Wöhrmann et al., 2012), drives imprinted expression of a GUS reporter gene. The cloning of these EMS-induced mutants using our newly established methodology (Lindner et al., 2012) turned out to be challenging, however, likely due to genome instabilities caused by these mutations. The identification of two imprinting regulators is currently ongoing using alternative strategies. Furthermore, quantitative genetic methods identified several paternal modifiers of imprinting (Pires et al., 2016). We could also show that DNA methylation and histone H3K27me3 methylation cooperate rather than exclude each other, at least as some imprinted loci (Schmidt et al., 2013). To molecularly test a model for MEA imprinting that involves higher order chromatin structure, i.e. chromatin looping (Wöhrmann et al., 2012), we successfully established Chromosome Conformation Capture (3C)-based technologies in plants. However, the highly streamlined, gene-dense Arabidopsis genome turned out to be unsuitable for the identification of enhancer-promoter interactions or chromatin loops by 4C or Hi-C, while this methodology allows a detailed characterization of the three dimensional organization of the genome (Grob et al., 2013) and led to the identification of a novel, evolutionary conserved nuclear structure, the KNOT (Grob et al., 2014). Finally, the biochemically characterization of MEA-containing complexes and their targets is still ongoing and, so far, has led to the development of novel, biochemical tools allowing the in vivo labeling and observation of proteins.
To investigate the role of epigenetic regulation in plant evolution and adaptation, we analyzed the epigenetic response to selection in experimental populations of Arabidopsis. While we could not demonstrate an epigenetic memory of stress in epigenetically highly uniform populations (Grob et al., in preparation), we could clearly show that standing, epigenetic variation is subject to selection and find phenotypic changes that are heritable for at least 2-3 generation in the absence of selection in three independent, replicated selection experiments (Schmid et al., in preparation). We developed new bioinformatics tools based on linear models to compare multiple methylomes with each other. To our knowledge, this is the first example showing selection of phenotypic traits in the absence of genetic variation that is fully supported by whole-genome molecular analyses. Finally, we have now firmly established the epigenetic nature of the switch in pollinator syndrome in Mimulus (Hirsch et al. 2012), using genetic crossing experiments and field studies (Baumberger & Grossniklaus, in prepration). To further elucidate the relationship between phenotypic change and epigenetic regulation we established a refernce transcrptome (Hirsch et al., in preparation) and identified several transcription factors correlated with phenotypic changes, which we are currently characterizing further at the molecular and functional level (Steinbach et al., in preparation).