Previously, we have shown that loss of the histone 3 lysine 27 (H3K27) monomethyltransferases ARABIDOPSIS TRITHORAX-RELATED 5 (ATXR5) and ATXR6 (ATXR6) results in the overreplication of heterochromatin. Here we show that the overreplication results in DNA damage and extensive chromocenter remodeling into unique structures we have named "overreplication-associated centers" (RACs). RACs have a highly ordered structure with an outer layer of condensed heterochromatin, an inner layer enriched in the histone variant H2AX, and a low-density core containing foci of phosphorylated H2AX (a marker of double-strand breaks) and the DNA-repair enzyme RAD51. atxr5,6 mutants are strongly affected by mutations in DNA repair, such as ATM and ATR. Because of its dense packaging and repetitive DNA sequence, heterochromatin is a challenging environment in which to repair DNA damage. Previous work in animals has shown that heterochromatic breaks are translocated out of the heterochromatic domain for repair. Our results show that atxr5,6 mutants use a variation on this strategy for repairing heterochromatic DNA damage. Rather than being moved to adjacent euchromatic regions, as in animals, heterochromatin undergoes large-scale remodeling to create a compartment with low chromatin density.

The relationship between epigenetic marks on chromatin and the regulation of DNA replication is poorly understood. Mutations of the H3K27 methyltransferase genes, Arabidopsis trithorax-related protein5 (ATXR5) and ATXR6, result in re-replication (repeated origin firing within the same cell cycle). Here we show that mutations that reduce DNA methylation act to suppress the re-replication phenotype of atxr5 atxr6 mutants. This suggests that DNA methylation, a mark enriched at the same heterochromatic regions that re-replicate in atxr5/6 mutants, is required for aberrant re-replication. In contrast, RNA sequencing analyses suggest that ATXR5/6 and DNA methylation cooperatively transcriptionally silence transposable elements (TEs). Hence our results suggest a complex relationship between ATXR5/6 and DNA methylation in the regulation of DNA replication and transcription of TEs.

Transcription initiation has long been considered a primary regulatory step in gene expression. Recent work, however, shows that downstream events, such as transcription elongation, can also play important roles.^1-3 A well-characterized example from animals is promoter-proximal pausing, where transcriptionally engaged Pol II accumulates 30-50 bp downstream of the transcription start site (TSS) and is thought to enable rapid gene activation.² Plants do not make widespread use of promoter-proximal pausing; however, in a phenomenon known as 3' pausing, a significant increase in Pol II is observed near the transcript end site (TES) of many genes.^4-6 Previous work has shown that 3' pausing is promoted by the BORDER (BDR) family of negative transcription elongation factors. Here we show that BDR proteins play key roles in gene repression. Consistent with BDR proteins acting to slow or pause elongating Pol II, BDR-repressed genes are characterized by high levels of Pol II occupancy, yet low levels of mRNA. The BDR proteins physically interact with FPA,⁷ one of approximately two dozen genes collectively referred to as the autonomous floral-promotion pathway,⁸ which are necessary for the repression of the flowering time gene FLOWERING LOCUS C (FLC).^9-11 In early-flowering strains, FLC expression is repressed by repressive histone modifications, such as histone H3 lysine 27 trimethylation (H3K27me3), thereby allowing the plants to flower early. These results suggest that the repression of transcription elongation by BDR proteins may allow for the temporary pausing of transcription or facilitate the long-term repression of genes by repressive histone modifications.

An evolutionary response to selection requires genetic variation; however, even if it exists, then the genetic details of the variation can constrain adaptation. In the simplest case, unlinked loci and uncorrelated phenotypes respond directly to multivariate selection and permit unrestricted paths to adaptive peaks. By contrast, 'antagonistic' pleiotropic loci may constrain adaptation by affecting variation of many traits and limiting the direction of trait correlations to vectors that are not favoured by selection. However, certain pleiotropic configurations may improve the conditions for adaptive evolution. Here, we present evidence that the Arabidopsis thaliana gene FRI (FRIGIDA) exhibits 'adaptive' pleiotropy, producing trait correlations along an axis that results in two adaptive strategies. Derived, low expression FRI alleles confer a 'drought escape' strategy owing to fast growth, low water use efficiency and early flowering. By contrast, a dehydration avoidance strategy is conferred by the ancestral phenotype of late flowering, slow growth and efficient water use during photosynthesis. The dehydration avoidant phenotype was recovered when genotypes with null FRI alleles were transformed with functional alleles. Our findings indicate that the well-documented effects of FRI on phenology result from differences in physiology, not only a simple developmental switch.

Constitutive heterochromatin in Arabidopsis thaliana is marked by repressive chromatin modifications, including DNA methylation, histone H3 dimethylation at Lys9 (H3K9me2) and monomethylation at Lys27 (H3K27me1). The enzymes catalyzing DNA methylation and H3K9me2 have been identified; alterations in these proteins lead to reactivation of silenced heterochromatic elements. The enzymes responsible for heterochromatic H3K27me1, in contrast, remain unknown. Here we show that the divergent SET-domain proteins ARABIDOPSIS TRITHORAX-RELATED PROTEIN 5 (ATXR5) and ATXR6 have H3K27 monomethyltransferase activity, and atxr5 atxr6 double mutants have reduced H3K27me1 in vivo and show partial heterochromatin decondensation. Mutations in atxr5 and atxr6 also lead to transcriptional activation of repressed heterochromatic elements. Notably, H3K9me2 and DNA methylation are unaffected in double mutants. These results indicate that ATXR5 and ATXR6 form a new class of H3K27 methyltransferases and that H3K27me1 represents a previously uncharacterized pathway required for transcriptional repression in Arabidopsis.

Eukaryotic genomes are regulated by epigenetic marks that act to modulate transcriptional control as well as to regulate DNA replication and repair. In Arabidopsis thaliana, mutation of the ATXR5 and ATXR6 histone methyltransferases causes reduction in histone H3 lysine 27 monomethylation, transcriptional upregulation of transposons, and a genome instability defect in which there is an accumulation of excess DNA corresponding to pericentromeric heterochromatin. We designed a forward genetic screen to identify suppressors of the atxr5/6 phenotype that uncovered loss-of-function mutations in two components of the TREX-2 complex (AtTHP1, AtSAC3B), a SUMO-interacting E3 ubiquitin ligase (AtSTUbL2) and a methyl-binding domain protein (AtMBD9). Additionally, using a reverse genetic approach, we show that a mutation in a plant homolog of the tumor suppressor gene BRCA1 enhances the atxr5/6 phenotype. Through characterization of these mutations, our results suggest models for the production atxr5 atxr6-induced extra DNA involving conflicts between the replicative and transcriptional processes in the cell, and suggest that the atxr5 atxr6 transcriptional defects may be the cause of the genome instability defects in the mutants. These findings highlight the critical intersection of transcriptional silencing and DNA replication in the maintenance of genome stability of heterochromatin.