Epigenetic Modifications and the Regulation of Gene Expression
During these processes, the chemical structure of 2'-deoxycytidine (dC) is heavily altered. The epigenetic deactivation of a gene is directed by the methylation of the 5 position of dC to methyl-dC (mdC), which in cells is achieved enzymatically by the DNA methyltransferase (DNMT) enzymes . The subsequent demethylation of mdC to dC yields a reactivated gene, which can in turn be expressed again (fig. 1). Although the process of silencing and reactivating a gene through methylation and demethylation is long known, the chemically sophisticated removal of the 5-methyl group is not yet completely understood. In 2009, hydroxymethyl-dC (hmdC) was found in genomic DNA and soon after, in 2011, the further oxidised species formyl-dC (fdC) and carboxy-dC (cadC) were also discovered . These non-canonical bases are generated through a stepwise oxidation from mdC to hmdC, fdC and cadC by the Ten eleven translocation (Tet) enzymes and are believed to play an important role in the aforementioned active demethylation (fig. 1).
Although a large part of epigenetic research focuses on the biological outcome of these modifications, most studies are driven in large part by a vital synthetic contribution which is coupled to biological experiments. Herein we describe the methodologies used in our group in order to answer the fundamental questions of this highly exciting field of research.
Quantification of Epigenetic Modifications
In order to begin elucidating the function of these modifications, it is important to first determine their abundance in different tissues and cell types as well as at different developmental stages respectively. Stable-isotope dilution methodologies in combination with high-performance liquid chromatography–mass spectrometry (HPLC-MS) allow for a highly accurate and sensitive means of quantification .
Following the isolation of DNA from a relevant tissue or cell type, it is enzymatically digested to the nucleoside level (fig. 2). Subsequently, a known amount of a nucleoside-isotopologue is used to spike the digest and provide an internal standard with the same chromatographic and ionization properties of the natural nucleoside.
Because of the mass difference between the two however, two signals are detected during the MS analysis. This enables the quantification of the modification by direct comparison of its integral to that of its isotopologue. With respect to the epigenetic modifications, one investigation showed that hmdC is widely distributed and can be found in all tissues in different levels. What is particularly interesting is the discovery that the human brain contains up to 1.2% of hmdC relative to dG, which complements all dC modifications .
Synthesis of Modified DNA for Investigations of Biological Systems
Synthesis of Building
Blocks for Solid-phase Synthesis
The complex function and properties of these modifications within a biological system can only be determined by expanding the research from the individual monomeric nucleosides. Exploration of their properties within the nucleic acid itself brings us closer to understanding nature’s design with regards to this second layer of information in our genetic code. Recent sequencing methods for the genome wide localisation of hmdC and fdC in genomic DNA inspire the design
and synthesis of nature-mimicking, sequence specific oligodeoxyribonucleotides (ODNs). For this, it is essential to be able to have a reliable and direct method of accessing ODNs with the modifications inserted at specific loci. It is therefore necessary to construct phosphoramidite building blocks compatible with conventional solid-phase DNA synthesis (fig. 3B). Knowledge of the reactivity of certain functionalities in the modifications, eg fdC’s vulnerability to both nucleophilic attack and/or oxidation, enables the design of phosphoramidites with protecting groups which are able to mask this reactivity prior to their incorporation into ODNs. These need to be both stable enough to withstand the synthesis but at the same time labile enough to allow facile deprotection of the completed strand using established protocols for ODNs. Building blocks for the epigenetic bases which were developed in our group are now commercially available (fig. 3B) .
At an early stage, these ODNs can be used for physical organic studies in order to elucidate how the modifications change the structure and stability of DNA. Furthermore, using modern MS and proteomics they can be used in biochemical experiments in order to decipher both the mechanism as well as to reveal the protagonist enzymes of the well orchestrated dynamic methylation.
In 2012, our group developed a synthesis for a 15N2-labelled cadC building block which was used for an isotope tracing experiment shown in figure 3A in order to investigate the active demethylation of mdC to dC through the decarboxylation of cadC . In this study, the 15N2-cadC phosphoramidite (fig. 3) was incorporated into a synthetic 30mer bearing a sequence of a promotor segment that is known to be the subject of active demethylation. This ODN, which also contained a biotin tag for affinity purification, was incubated with nuclear extracts of mouse embryonic stem cells (mESC) and then re-isolated with streptavidin-bound magnetic beads (fig. 3A). After digestion of the isolated ODN to the nucleoside level, the mixture was analysed using HPLC-MS. To our delight, 15N2-dC was indeed detected, which showed that the nuclear extract of mESC has the capability to decarboxylate cadC to dC. These findings support our proposed hypothesis of an oxidation-dependent active demethylation pathway for mdC which is under heavy investigation in our group.
Identification of the “Writers, Readers and Erasers” of Epigenetic Modifications
As a way of determining the enzymes responsible for these pathways, control and modified biotinylated ODNs, differing only in their dC context can be used as “baits” for proteins during incubation with cell lysates and tissues. As we are investigating the silencing and activation of specific genes during cell reprograming at the early stages of development, we are deeply interested in mESC. Similar to our mechanistic investigations, streptavidin-bound magnetic beads are used to remove the bait from the lysate whilst simultaneously “fishing” out the proteins which have bound to the respective strands. Using a range of quantitative MS proteomic techniques eg stable isotope labelling with amino acids in cell culture (SILAC) , it is then possible to determine the proteins associated with each strand (fig. 3B). This allows us to distinguish which proteins are associated with the modifications and which are repelled by them. In collaboration with Prof. Vermeulen from the University of Nijmegen, the Netherlands, using SILAC/MS we were able to discover dynamic readers of hmdC . Overall, specific proteins were found that bind hmdC, fdC and cadC suggesting that these bases trigger important biological responses. These investigations bring us closer to discovering the assembly of protein “writers, readers and erasers” of the modifications and the dynamic mechanism through which they orchestrate gene expression.
To summarize, our group recruits synthetic chemists as well as biologists and biochemists in order to answer the fundamental questions in epigenetics. This enables us to develop methodologies to describe and investigate the interface between biology and chemistry in this field. Through our pursuit in unravelling the mechanisms and the key enzymes coordinating the events of gene activation and silencing, we aim to continue developing methods useful to both academia and industry.
 J. A. Law, S. E. Jacobsen, Establishing, maintaining and modifying DNA methylation patterns in plants and animals, Nat. Rev. Genet. 11, 204-220 (2010) – DOI: 10.1038/nrg2719
 T. Pfaffeneder et al, The Discovery of 5-Formylcytosine in Embryonic Stem Cell DNA, Angew. Chem. Int. Ed. 50, 7008-7012 (2011) – DOI: 10.1002/anie.201103899
 D. Globisch et al, Tissue Distribution of 5-Hydroxymethylcytosine and Search for Active Demethylation Intermediates, PLoS One 5, e15367 (2010) - DOI: 10.1371/journal.pone.0015367
 M. Wagner et al., Age-Dependent Levels of 5-Methyl-, 5-Hydroxymethyl-, and 5 Formylcytosine in Human and Mouse Brain Tissues, Angew. Chem. Int. Ed. – DOI: 10.1002/anie.201502722
 A.S. Schroder et al., Synthesis of a DNA Promoter Segment Containing All Four Epigenetic Nucleosides: 5-Methyl-, 5-Hydroxymethyl-, 5-Formyl-, and 5-Carboxy-2′-Deoxycytidine, Angew. Chem. Int. Ed. 53, 315-318 (2014) – DOI: 10.1002/anie.201308469
 S. Schiesser et al, Mechanism and Stem-Cell Activity of 5-Carboxycytosine Decarboxylation Determined by Isotope Tracing, Angew. Chem. Int. Ed. 51, 6516–6520 (2012) – DOI: 10.1002/anie.201202583
 C.G. Spruijt et al, Dynamic Readers for 5-(Hydroxy)Methylcytosine and Its Oxidized Derivatives, Cell 152, 1146 (2013) – DOI: 10.1016/j.cell.2013.02.004
Thomas Carell, Iacovos Michaelides, Rene Rahimoff
Ludwig-Maximilians-University Munich, Department for Chemistry and Pharmacy, Munich
Department for Chemistry and Pharmacy