Abstract
Headline-grabbing attention has been given to DNA demethylation pathways as new epigenetic mechanisms, with reviews and hypotheses outnumbering research papers. As candidate proteins for DNA demethylation include well-known DNA repair enzymes, it was timely to join epigenetics and DNA repair experts at the first international meeting on DNA Demethylation, Repair and Beyond. New mechanistic insights were presented for known players orchestrating the symphony on cytosine – ‘the symphony on C‘ (TET1, 2, 3; GADD45; AID; and TDG), while new instruments and classical themes were pulled into the amalgamation. What may appear as just an unintentional cacophony of random oxidative lesions and abasic sites in a bed of chromatin noise may turn out to be a gene-expressing regulatory melody.
Active DNA demethylation had, for a long time, been suspected to occur but recent discoveries of hydroxymethylated cytosine and additional cytosine DNA modifications in mammals (the 6th, 7th and 8th base) have fueled the broad interest in understanding DNA methylation removal Citation[1] and how such modifications affect chromatin and with it, transcription. Since base modifications are considered deleterious and efficiently removed by DNA repair, it is important to understand how DNA repair is controlled in the context of DNA methylation. Therefore, the first ever meeting on DNA demethylation and DNA repair, organized by Christof Niehrs and colleagues (IMB, Mainz, Germany), was very timely in combining the fields of epigenetics and DNA repair, aiming to accelerate the understanding of the C-dynamics in our genome.
With the choice of the two keynote speakers in Anjana Rao (La Jolla Institute for Allergy & Immunology, CA, USA) and Azim Surani (University of Cambridge, Cambridge, UK) the tone was set for an intriguing meeting. Rao provided a historical perspective on the (re)discovery of 5-hydroxymethylcytosine (5hmC) and the enzymes involved in the oxidation of 5mC (TET enzymes) Citation[2]. Her seminal work, based on the modified DNA base J found in Trypanosoma, uncovered a DNA base modification pathway that was long thought to exist only in bacteriophages Citation[3], while Surani provided further insight into the dynamic DNA methylome at different developmental stages and highlighted the role of various DNA repair factors (especially those from the base excision repair pathway) Citation[4].
DNA repair itself, with a focus on nucleotide excision repair (NER), was introduced by Jan Hoeijmakers and Wim Vermeulen (Erasmus Medical Center, Rotterdam, The Netherlands). The Janus-like behavior of too much NER (possibly via transcription-coupled repair), leading to protection against cancer at the expense of aging, highlights the direct implications for various clinical phenotypes. Niehrs had previously demonstrated that GADD45a, a key player in NER, is intimately linked to DNA demethylation Citation[5] and provided some mechanistic insight into the recruitment of GADD45 to H3K4Me3-marked chromatin via ING1b.
Extending on the interplay between DNA repair and chromatin, Penny Jeggo (University of Sussex, Sussex, UK) demonstrated that inhibition of DNA damage response protein DNA-PKCs, possibly via histone modifications, shifted repair from nonhomologous end joining to homologous recombination Citation[6]. Analogously, the H3K9me3 reader protein SPOC1, in conjunction with histone chaperons and methyltransferases (Andreas Mund, Heinrich-Pette-Institute, Hamburg, Germany), reduced nonhomologous end-joining repair activity at sites of DNA damage. This could establish a feedback loop of chromatin altering DNA repair, which alters DNA demethylation. Protein ubiquitination was also shown to play an important role in DNA repair pathway choice, as presented by Vermeulen and Helle Ulrich (Cancer Research UK London Research Institute, London, UK), on histones and PCNA Citation[7], respectively.
Interdependence between DNA repair and RNA POLII transcription has long been known, exemplified by transcription-coupled repair of NER, hinting at a role for DNA demethylation. Indeed, depletion of XPA/XPC (NER pathway) prevented DNA demethylation (Jean-Marc Egly, IGBMC, Illkirch-Graffenstaden, France) Citation[8]. Previous work by George Reid (IMB) and Egly, among others, demonstrated DNA de-/re-methylation cycles to occur within minutes Citation[9,10], implicating a highly dynamic demethylation mechanism.
Active DNA demethylation can proceed via a number of different pathways, including modifying the methyl group (e.g., hydroxymethylation) or the nucleoside (i.e., deamination), or removing the nucleotide (glycosylation). Indeed, the term ‘demethylation‘ has to be well defined Citation[1]. If taken in its most restricted form, only complete removal of the 5-methyl from cytosine, either via chemical removal or base substitution is acceptable (such as whole-base substitution, via ROS and DEMETER as demonstrated in plants [presented by Rafael Ariza, Cordoba, Spain]). Alternatively, the modification of the 5-methyl group (such as oxidation to 5hmC, 5-formylcytosine (5fC) or 5-carboxylcytosine (5caC), as performed by TET proteins) is sufficient for 5mCpG binding proteins to lose their biological function, and hence ‘hiding‘ of 5-methyl becomes equivalent to demethylation. Indeed, a third of the presentations were centered around TET proteins, ranging from evolutionary context to developmental regulation, reprogramming and biochemical molecular mechanisms.
Rao‘s insightful use of Corpinus cinereous (a fungus) for identifying ancestral TET proteins Citation[3] and technological advancements in base modification analysis Citation[11], provides further evidence on the widespread nature of DNA modification in eukaryotes, with the intriguing hypothesis that the TET proteins can move horizontally via transposable elements. Analysis of the TET family protein expression during developmental reprogramming (Guoliang Xu, Chinese Academy of Sciences, Shanghai, China, and Surani), as well as determining the abundance of 5hmC and its further oxidations (5fC and 5caC; addressed by many), is beginning to reveal the finer molecular details of this base modification. The finding that TET proteins have a pivotal role during meiosis (Yi Zhang, Harvard Medical School, MA, USA), possibly via direct transcriptional regulation Citation[12], further indicates that DNA (de)methylation is progressing via different pathways, depending on the microenvironment. The detailed analysis by Jörn Walter (Saarland University. Saarbrücken, Germany) on the type and distribution of modifications during early stages of a developing embryo provided further molecular details Citation[13] and introduced the concept of 5hmC being a way to induce passive DNA demethylation, as no copy mechanism is yet available for 5hmC. Furthermore, Xu‘s finding on TET3 localisation in oocyte expression in early development provides a mechanism for the extensive 5hmC presence in the paternal nucleus Citation[14].
Primo Schär (University of Basel, Basel, Switzerland) extended on his seminal finding that the base excision repair protein, TDG, is required for proper development Citation[15]. A DNA glycosylase, originally thought to act on deaminated 5mC (i.e., deoxythymidine) in the context of a CpG, was now shown to efficiently remove 5caC and 5fC, while not affecting 5mC or 5hmC. TDG‘s presence during times of DNA demethylation also supports the demethylation pathway depending on deamination of cytosine via DNA deaminases (e.g., activation-induced deaminase [AID]). Biochemical Citation[16] and genetic Citation[17] analyses had identified AID‘s role in demethylation, and is now demonstrated to proceed not only at the site of the lesion but efficiently over distances of up to 2 kb (Petersen-Mahrt, IFOM-Fondazione Istituto FIRC di Oncologia Molecolare, Milan, Italy). Although AID was implicated in deaminating 5hmC for further processing, it was shown to be a biochemical impossibility (Petersen-Mahrt, Schär). Therefore, it is more likely that TET-modified bases are ‘restored‘ to cytosine via TDG or a carboxylase, as suggested by Xu and Thomas Carell (Ludwig Maximilians University, Munich, Germany) Citation[18].
Aside from serving as a conduit for DNA demethylation, the TET-dependent cytosine modifications could also serve as a new binding platform to establish chromatin modifications. As aptly demonstrated by Francois Fuks (Free University of Brussels, Brussels, Belgium), TET2 and TET3 proteins recruited OGT (for protein modification) mainly to transcription start sites. Although this recruitment was not associated with 5mC/5hmC changes, nor was TET activity affected, there seemed to be an effect on transcription via H3K4Me3. To firmly establish a causal role of effectors, such as TET proteins, the targeting of an effector to a single endogenous side in the genome (epigenetic editing) provides a promising approach (Marianne Rots, University Medical Center Groningen, Groningen, The Netherlands) Citation[19].
It will be a long and interesting journey to discover all the intricacies of the various modification pathways and how they influence growth, development and disease, especially considering that TET1 knockout mice are viable but seem deficient in learning and memory (presented by Xu).
The data from the Reik (Babraham Institute, Cambridge, UK) laboratory presented another layer of complexity in this growing field. Returning to the importance of why there is plasticity and reprogramming in our genomes (introduced by Surani), external stimuli, via signal transduction pathways, were analyzed in their ability to activate key reprogramming factors. By inhibiting two kinase cascades, novel understanding on the kinetics and regulation of DNA methylation were uncovered, which led to understanding the important correlation between embryonic stem cell culture in vitro and reprogramming/development in vivo.
Good conferences do not provide definitive conclusions but provide fertile ground for better questions. Important questions were raised during the meeting: which DNA demethylation pathway will be important, when and where, and how are they activated? At what stage of development and for what purpose does each system function? How much redundancy is there and how much interdependency? In terms of molecular mechanisms, involvement of polycomb proteins (Nicole Francis, Harvard University, MA, USA) has only been touched upon, but is likely to play an (in)direct role in DNA demethylation and DNA repair. Alterations in DNA methylation can change how DNA damage response, via p53 binding (Kristina Kirschner, Cambridge Research Institute, Li Ka Shing Centre, Cambridge, UK), can alter a cellular response depending on the cell state. Identification of pharmaceuticals for DNA demethylation may provide new tools for transcription and DNA repair (Bernd Epe [Johannes Gutenberg University, Mainz, Germany] and Reid).
The emerging theme from this meeting was that the interconnections between development, cell cycle, DNA repair, chromatin, RNA transcription, evolution, signal transduction and protein trafficking is forcing us to combine forces, learn new notes, develop new instruments and listen more intently to all symphonies. As became obvious from this IMB meeting: in order to fully elucidate the mechanisms and functions of the different melodies on C, we need to become multidisciplinary experts (proper biologists) – again.
Acknowledgements
The authors would like to thank all presenters (oral and poster) for clear presentations and attendees for stimulating discussion, C Niehrs for providing such a stimulating environment and the authors are looking forward to the next meeting at the IMB on chromatin dynamics and stem cells.
Financial & competing interests disclosure
Funding to MG Rots includes UMCG-Rosalind Franklin Fellowship and National Dutch Scientific Research Organisation; SK Petersen-Mahrt is supported by Fondazione Istituto FIRC di Oncologia Molecolare, Milan, Italy. The authors have no other 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 apart from those disclosed.
No writing assistance was utilized in the production of this manuscript.
Additional information
Funding
References
- Franchini DM , SchmitzKM, Petersen-MahrtSK. 5-methylcytosine DNA demethylation: more than losing a methyl group. Annu. Rev. Genet.46 , 419–441 (2012).
- Tahiliani M , KohKP, ShenY et al. Conversion of 5-methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by MLL partner TET1. Science 324(5929) , 930–935 (2009).
- Iyer LM , TahilianiM, RaoA, AravindL. Prediction of novel families of enzymes involved in oxidative and other complex modifications of bases in nucleic acids. Cell Cycle8(11) , 1698–1710 (2009).
- Hajkova P , JeffriesSJ, LeeC, MillerN, JacksonSP, SuraniMA. Genome-wide reprogramming in the mouse germ line entails the base excision repair pathway. Science329(5987) , 78–82 (2010).
- Barreto G , SchäferA, MarholdJ et al. Gadd45a promotes epigenetic gene activation by repair-mediated DNA demethylation. Nature 445(7128) , 671–675 (2007).
- Shibata A , ConradS, BirrauxJ et al. Factors determining DNA double-strand break repair pathway choice in G2 phase. EMBO J. 30(6) , 1079–1092 (2011).
- Ulrich HD . Ubiquitin, SUMO, and phosphate: how a trio of posttranslational modifiers governs protein fate. Mol. Cell.47(3) , 335–337 (2012).
- Le May N , FradinD, IltisI, BougneresP, EglyJM. XPG and XPF endonucleases trigger chromatin looping and DNA demethylation for accurate expression of activated genes. Mol. Cell.47(4) , 622–632 (2012).
- Kangaspeska S , StrideB, MetivierR et al. Transient cyclical methylation of promoter DNA. Nature 452(7183) , 112–115 (2008).
- Le May N , Mota-FernandesD, Vélez-CruzR, IltisI, BiardD, EglyJM. NER factors are recruited to active promoters and facilitate chromatin modification for transcription in the absence of exogenous genotoxic attack. Mol. Cell.38(1) , 54–66 (2010).
- Pastor WA , PapeUJ, HuangY et al. Genome-wide mapping of 5-hydroxymethylcytosine in embryonic stem cells. Nature 473(7347) , 394–397 (2011).
- Yamaguchi S , HongK, LiuR et al. Tet1 controls meiosis by regulating meiotic gene expression. Nature doi:10.1038/nature11709 (2012) (Epub ahead of print).
- Arand J , SpielerD, KariusT et al. In vivo control of CpG and non-CpG DNA methylation by DNA methyltransferases. PLoS Genet.8(6) , e1002750 (2012).
- Gu TP , GuoF, YangH et al. The role of Tet3 DNA dioxygenase in epigenetic reprogramming by oocytes. Nature 477(7366) , 606–610 (2011).
- Cortazar D , KunzC, SelfridgeJ et al. Embryonic lethal phenotype reveals a function of TDG in maintaining epigenetic stability. Nature 470(7334) , 419–423 (2011).
- Morgan HD , DeanW, CokerHA, ReikW, Petersen-MahrtSK. Activation-induced cytidine deaminase deaminates 5-methylcytosine in DNA and is expressed in pluripotent tissues: implications for epigenetic reprogramming. J. Biol. Chem.279(50) , 52353–52360 (2004).
- Popp C , DeanW, FengS et al. Genome-wide erasure of DNA methylation in mouse primordial germ cells is affected by AID deficiency. Nature 463(7284) , 1101–1105 (2010).
- Schiesser S , HacknerB, PfaffenederT et al. Mechanism and stem-cell activity of 5-carboxycytosine decarboxylation determined by isotope tracing. Angew. Chem. Int. Ed. Engl. 51(26) , 6516–6520 (2012).
- De Groote ML , VerschurePJ, RotsMG. Epigenetic editing: targeted rewriting of epigenetic marks to modulate expression of selected target genes. Nucleic Acids Res.40(21) , 10596–10613 (2012).