Abstract
The Oslo Epigenetics Symposium 2012 held in Oslo, Norway, brought together ten speakers from several European countries and the USA for an evening public lecture and a full day of presentations on emerging topics in the field of epigenetics, gene regulation and organization of the cell nucleus.
The conference was opened with a plenary evening public lecture by Tim Spector (Kings College London, London, UK). The historic Rococo room at the Grand Hotel Oslo was fully packed and Spector engaged the audience by asking: if we could, would we change our genes? Spector presented his recent book “Identically different. Why you can change your genes” and used 20 years of research on monozygotic twins to explain how epigenetics link our genes to the environment. Furthermore, he explained how cohort studies show that grandparent‘s lifestyle may affect their grandchildren‘s health and longevity owing to epigenetic changes in gene activity.
Axel Imhof (Ludwig Maximilians University Munich, Munich, Germany) opened the second day of the meeting with insights on how ‘RNA shapes chromatin‘. Formation of higher order chromatin structure determines DNA accessibility and activity of genome domains. Imhof and colleagues have identified an RNA-dependent mechanism maintaining an open chromatin structure Citation[1]. In vitro reconstitution of chromatin in Drosophila embryo extracts, which are rich in histones and assembly factors, requires Df31 that specifically binds RNA and localizes to euchromatin. Df31 can tether ssRNAs to chromatin. Such RNAs are stably linked to chromatin and mainly contain snoRNAs. Thus, Df31 may form a RNA–chromatin network resulting in open chromatin domains enabling binding of transcription factors (TFs) Citation[1].
Mapping initiatives reveal that most regulators of gene activity localize to only a small subset of their potential binding sites in the genome Citation[2,3]. Now, however, the ability to write DNA at a rate of 200 megabases per week opens new doors to testing functional elements. Using novel assays based on high-throughput DNA synthesis and sequencing, Tarjei Mikkelsen (Broad Institute of MIT and Harvard, MA, USA) asked how these binding sites are selected by TFs. Together with his colleagues, he synthesized PPARG motifs shown to be bound by PPARG in the genome, and PPARG motifs from regions with no PPARG binding. Hundreds of sequences cloned into reporter constructs can be transfected into cells as a library of 27,000 plasmids. Subsequent chromatin immunoprecipitation (ChIP)-sequencing shows that PPARG binding is stronger at sites effectively bound by PPARG compared with unbound PPARG-binding sites, suggesting other factors mediate the binding affinity of TFs. A similar approach can be undertaken to examine sequences flanking TF-binding motifs and systematically dissect gene regulatory elements Citation[4].
Dysregulation of gene expression by DNA demethylation in immune cells plays a key role in hematological malignancies, autoimmune diseases and immunodeficiencies Citation[5]. The question is how demethylation is achieved in these disease contexts. Using a model system of Epstein–Barr virus-mediated B-cell transformation, Esteban Ballestar (Bellvitge Biomedical Research Institute, Barcelona, Spain) showed that transformation is associated with promoter hypomethylation without any evidence of marked methylation, 5-hydroxymethylation or involvement of enzymes eliciting active demethylation. Hypomethylation occurs at sites with low abundance of DNA methyltransferases and only half of the hypomethylated genes are upregulated. Therefore, hypomethylation does not seem to be an active process in B-cell transformation and is likely linked to inefficient DNA methylation at active regions.
5-hydroxymethyl cytosine (5hmC) is a DNA modification shown to affect transcription. Current methods of 5hmC detection lack sensitivity and seem to be poorly suited for genome-wide mapping Citation[6]. John Arne Dahl (Oslo University Hospital, Oslo, Norway) presented a new development in 5hmC detection. The assay relies on two proteins, β-glucosyltransferase, which converts 5-hmC to β-glucosyl-5-hmC, and JBP1, which recognizes and binds β-glucosyl-5-hmC. JBP1 can be coupled to magnetic beads to pull down DNA enriched in 5-hmC Citation[7], which in turn can be sequenced. The group is also developing a new assay termed HyLo, to identify 5hmC at single-base resolution.
Stephan Beck (University College London, London, UK) asked: ‘can genome-wide association studies (GWAS) be enhanced by epigenomics data‘. Over 2000 SNPs significant for disease are now identified. Yet, GWAS do not explain much of the disease phenotypes and effect sizes of GWAS variants remain modest. There is, therefore, an opportunity to improve the significance of GWAS with epigenome-wide association studies Citation[8]. ENCODE and other large data sets can tremendously help in this process. Beck gave an outlook on how epigenome-wide association studies may be used to better understand phenotypic plasticity in health and disease. Following up, Tim Spector (Kings College London) emphasized the need for larger sample sizes, genetic–epigenetic analyses and longitudinal studies to establish the role of epigenetic variants in disease Citation[9]. Spector discussed how large methylome mapping studies in twins can provide insights into epigenetic heritability using DNA methylation patterns as a dynamic quantitative trait.
Polycomb group (PcG) proteins are epigenetic repressors important for embryonic development and whose misregulation can cause cancer. Recruitment of polycomb proteins in Drosophila depends on Polycomb response elements (PREs), which can be distant from the promoter they regulate. PcG proteins exist in all cells in Drosophila embryos but only repress the homeobox gene Ubx in cells where the gene was previously active. Therefore, there is a system that distinguishes cells in which PcG targets should be repressed from cells where they should not. Yuri Schwartz (Umeå University, Umeå, Sweden) explained that Polycomb repressive complexes are recruited to PREs through interactions with different DNA binding proteins, which are all individually weak recruiters. PHO and its related protein PHOL are the only PcG proteins that possess sequence-specific DNA binding activity towards many PREs Citation[10]. PHO/PHOL DNA binding activity depends on interaction with SFMBT. This crosstalk provides evidence of a combinatorial action for polycomb recruitment to chromatin.
PRC2 proteins are conserved in plants and Marcel Lafos (Heinrich-Heine University, Düsseldorf, Germany) has mapped PRC2-mediated H3K27 methylation (H3K27me3) during differentiation of Arabidopsis meristeme stem cells to young leaves Citation[11]. A dynamic regulation of H3K27me3 during plant development correlates with changes in gene expression. A large fraction of miRNA genes is found to be differentially marked by H3K27me3 in meristems and leaves. H3K27me3 enrichment on miRNA genes also occurs in Drosophila and humans, and reveals an alternative mechanism of polycomb regulation through repression of miRNA-regulated translation control Citation[12,13].
Rein Aasland (University of Bergen, Bergen, Norway) addressed the issue of how genome-wide epigenetic landscapes are interpreted. It has been found that hundreds of proteins with histone recognition modules have an active role in chromatin regulation. Using the bromodomain/PHD finger of the acetyltransferase P300 as an example, Aasland showed that bromodomain/PHD finger can modulate acetyltransferase activity of P300 through binding of the histone acetyltransferase domain. Aasland also presented a new histone recognition module for methylated H3K4, the Cys-Trp domain Citation[14]. Interestingly, Cys-Trp domain-containing proteins often have other functional domains, underpinning a general observation for many histone recognition modules proteins.
Kerstin Bystricky (University of Toulouse, Toulouse, France) uses hormone regulation of ERα target genes in breast cancer cells as a model to characterize how chromatin responds to external stimuli. Upon hormone-induced activation, the histone variant H2A.Z-mediated interaction between the promoter and downstream enhancer of the CCND1 gene is released owing to TIP48-mediated acetylation and eviction of H2A.Z, allowing ERα binding. She further showed that in a triple ER-negative breast cancer cells, CCND1 is transcriptionally stimulated by depletion of H2A.Z Citation[15]. The data provide new insights on hormonal gene activation through rapid chromatin remodeling. Bystricky also presented a new genetic tag, the ANCHOR tag system, for visualization of gene loci in living cells; we anticipate this DNA tag will open up new avenues in the field of chromatin regulation.
Wendy Bickmore (University of Edinburgh, Edinburgh, UK) discussed long-range gene control, using the regulation of the Hoxd locus during mouse limb patterning as an example. Chromatin looping in this region can be visualized by FISH Citation[16] and chromosome conformation capture data. However, other long-range interactions detected by chromosome conformation capture around the Hoxd locus cannot be observed by FISH. Bickmore thus asked whether other chromatin conformation models may account for these 3D structures; for instance, can large modules set up a confined nuclear space as a landing platform for binding of proteins that do not require direct interaction with DNA elements? Bickmore also presented evidence for a novel enhancer mark, H4K16ac, co-occurring with H3K4me1 on P300-independent active enhancers. This new mark adds another flavor to enhancer activity.
Association of a genome with nuclear lamins occurs through large chromatin domains and often correlates with gene repression in these domains. In order to elucidate the relationship between lamin binding and gene activity, Philippe Collas (University of Oslo, Oslo, Norway) reported the association of lamin A/C with thousands of promoters in human cells. The bound genes are mainly repressed, but repression correlates with coenrichment in repressive histone marks rather than lamin binding per se. Lamins can associate with distinct subregions on promoters, with distinct transcriptional outputs, and structural perturbations in the lamina reveal an uncoupling of lamin binding from promoter inactivity. This supports the existence of restricted lamin binding sites on promoters with distinct position-dependent transcriptional outcomes.
Financial & competing interests disclosure
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.
References
- Schubert T , PuschMC, DiermeierS et al. Protein and snoRNAs maintain accessible higher-order structures of chromatin. Mol. Cell 48 , 434–444 (2012).
- Bernstein BE , StamatoyannopoulosJA, CostelloJF et al. The NIH Roadmap Epigenomics Mapping Consortium. Nat. Biotechnol. 28 , 1045–1048 (2010).
- Mikkelsen TS , XuZ, ZhangX et al. Comparative epigenomic analysis of murine and human adipogenesis. Cell 143 , 156–169 (2010).
- Melnikov A , MuruganA, ZhangX et al. Systematic dissection and optimization of inducible enhancers in human cells using a massively parallel reporter assay. Nat. Biotechnol. 30 , 271–277 (2012).
- Rodriguez-Cortez VC , HernandoH, de la Rica L, Vento R, Ballestar E. Epigenomic deregulation in the immune system. Epigenomics3 , 697–713 (2011).
- Robertson AB , DahlJA, OuglandR, KlunglandA. Pull-down of 5-hydroxymethylcytosine DNA using JBP1-coated magnetic beads. Nat. Protoc.7 , 340–350 (2012).
- Robertson AB , DahlJA, VagboCB, TripathiP, KrokanHE, KlunglandA. A novel method for the efficient and selective identification of 5-hydroxymethylcytosine in genomic DNA. Nucleic Acids Res.39 , e55 (2011).
- Rakyan VK , DownTA, BaldingDJ, BeckS. Epigenome-wide association studies for common human diseases. Nat. Rev. Genet.12 , 529–541 (2011).
- Bell JT , SpectorTD. DNA methylation studies using twins: what are they telling us? Genome Biol.13 , 172 (2012).
- Park SY , SchwartzYB, KahnTG, AskerD, PirrottaV. Regulation of polycomb group genes Psc and Su(z)2 in Drosophila melanogaster. Mech. Dev.128 , 536–547 (2012).
- Lafos M , KrollP, HohenstattML, ThorpeFL, ClarenzO, SchubertD. Dynamic regulation of H3K27 trimethylation during Arabidopsis differentiation. PLoS Genet.7 , e1002040 (2011).
- Kuchen S , ReschW, YamaneA et al. Regulation of microRNA expression and abundance during lymphopoiesis. Immunity 32 , 828–839 (2010).
- Enderle D , BeiselC, StadlerMB, GerstungM, AthriP, ParoR. Polycomb preferentially targets stalled promoters of coding and noncoding transcripts. Genome Res.21 , 216–226 (2011).
- Hoppmann V , ThorstensenT, KristiansenPE et al. The CW domain, a new histone recognition module in chromatin proteins. EMBO J. 30 , 1939–1952 (2011).
- Dalvai M , BellucciL, FleuryL, LavigneAC, MoutahirF, BystrickyK. H2A.Z-dependent crosstalk between enhancer and promoter regulates Cyclin D1 expression. Oncogene doi:10.1038/onc.2012.442 (2012) (Epub ahead of print).
- Williamson I , EskelandR, LetticeLA et al. Anterior-posterior differences in HoxD chromatin topology in limb development. Development 139 , 3157–3167 (2012).