Lysine 27 of histone H3.3 is a fine modulator of developmental gene expression and stands as an epigenetic checkpoint for lignin biosynthesis in Arabidopsis.

Chromatin is a dynamic platform within which gene expression is controlled by epigenetic modifications, notably targeting amino acid residues of histone H3. Among them is lysine 27 of H3 (H3K27), the trimethylation of which by the Polycomb Repressive Complex 2 (PRC2) is instrumental in regulating spatiotemporal patterns of key developmental genes. H3K27 is also subjected to acetylation and is found at sites of active transcription. Most information on the function of histone residues and their associated modifications in plants was obtained from studies of loss-of-function mutants for the complexes that modify them. To decrypt the genuine function of H3K27, we expressed a non-modifiable variant of H3 at residue K27 (H3.3K27A ) in Arabidopsis, and developed a multi-scale approach combining in-depth phenotypical and cytological analyses, with transcriptomics and metabolomics. We uncovered that the H3.3K27A variant causes severe developmental defects, part of them are reminiscent of PRC2 mutants, part of them are new. They include early flowering, increased callus formation and short stems with thicker xylem cell layer. This latest phenotype correlates with mis-regulation of phenylpropanoid biosynthesis. Overall, our results reveal novel roles of H3K27 in plant cell fates and metabolic pathways, and highlight an epigenetic control point for elongation and lignin composition of the stem.

Chromatin Manipulation and Editing: Challenges, New Technologies and Their Use in Plants.

An ongoing challenge in functional epigenomics is to develop tools for precise manipulation of epigenetic marks. These tools would allow moving from correlation-based to causal-based findings, a necessary step to reach conclusions on mechanistic principles. In this review, we describe and discuss the advantages and limits of tools and technologies developed to impact epigenetic marks, and which could be employed to study their direct effect on nuclear and chromatin structure, on transcription, and their further genuine role in plant cell fate and development. On one hand, epigenome-wide approaches include drug inhibitors for chromatin modifiers or readers, nanobodies against histone marks or lines expressing modified histones or mutant chromatin effectors. On the other hand, locus-specific approaches consist in targeting precise regions on the chromatin, with engineered proteins able to modify epigenetic marks. Early systems use effectors in fusion with protein domains that recognize a specific DNA sequence (Zinc Finger or TALEs), while the more recent dCas9 approach operates through RNA-DNA interaction, thereby providing more flexibility and modularity for tool designs. Current developments of "second generation", chimeric dCas9 systems, aiming at better targeting efficiency and modifier capacity have recently been tested in plants and provided promising results. Finally, recent proof-of-concept studies forecast even finer tools, such as inducible/switchable systems, that will allow temporal analyses of the molecular events that follow a change in a specific chromatin mark.

GIP1 and GIP2 Contribute to the Maintenance of Genome Stability at the Nuclear Periphery.

The maintenance of genetic information is important in eukaryotes notably through mechanisms occurring at the nuclear periphery where inner nuclear membrane proteins and nuclear pore-associated components are key factors regulating the DNA damage response (DDR). However, this aspect of DDR regulation is still poorly documented in plants. We addressed here how genomic stability is impaired in the gamma-tubulin complex component 3-interacting protein (gip1gip2) double mutants showing defective nuclear shaping. Using neutral comet assays for DNA double-strand breaks (DSBs) detection, we showed that GIP1 and GIP2 act redundantly to maintain genome stability. At the cellular level, γ-H2AX foci in gip1gip2 were more abundant and heterogeneous in their size compared to wild-type (WT) in root meristematic nuclei, indicative of constitutive DNA damage. This was linked to a constitutive activation of the DDR in the gip1gip2 mutant, with more emphasis on the homologous recombination (HR) repair pathway. In addition, we noticed the presence of numerous RAD51 foci which did not colocalize with γ-H2AX foci. The expression of GIP1-GFP in the double mutant rescued the cellular response to DNA damage, leading to the systematic colocalization of RAD51 and γ-H2AX foci. Interestingly, a significant proportion of RAD51 foci colocalized with GIP1-GFP at the nuclear periphery. Altogether, our data suggest that GIPs may partly contribute to the spatio-temporal recruitment of RAD51 at the nuclear periphery.