Detecting cytosine methylation at the single DNA molecule level

Although the genomes of many organisms (humans, plants, invertebrates and vertebrates) have been sequenced and many of the genes identified, our understanding of the regulation of the genes is limited due to lack of analysis technology. DNA is composed of four nucleic acid bases, adenine, guanine, cytosine and thymine. Some of these nucleic acid bases can be modified by enzymes and as a result have an additional methyl group; here we will investigate new technologies for the detection of methylcytosine and unmethylated cytosine within single DNA molecules. The methylation of cytosine nucleic acid bases is associated with gene silencing. In humans DNA methylation is considered to play a critical role in development and is aberrant in many diseases, but as yet the complete role remains unclear. There are numerous techniques for the detection of methylated cytosine in DNA, but the current methodologies do not yet provide a simple, fast, reliable cheap approach. A major problem is the need to evaluate DNA from cell samples that will contain the same DNA sequence but which are heterogeneous with respect to the cytosine residues that are methylated. So an average is often obtained. Those techniques that do allow single DNA strands to be evaluated are highly laborious and limited. Here we will develop a new approach for detecting sequences containing methylated cytosines at the single molecule level. There are currently other groups working in the field of DNA sequencing of single molecules, but these methods are slow and the DNA is investigated as a single strand. We will interrogate double-stranded DNA and this will allow us to detect methylated or unmethylated cytosine molecules on each strand, called hemi-methylation. Our approach is to create an artificial form of DNA, an oligonucleotide, that associates and wraps within the major groove of double-stranded DNA molecule at specific sequences. This artificial form of DNA when associated is called a triplex and the molecules synthesised will also contain a fluorophore. When the DNA sample has been treated with these triplex forming oligonucleotides the helix will contain fluorophores at different points along it. We will inject the DNA sample into a small channel that will result in unravelling and straightening of the strand so that it is then threaded into an optical interrogation channel. The fluorophores will be excited with light, which in the presence of nanostructures within the nanochannel will result in fluorescence intensity changes. The change in intensity will provide a code that indicates the methylation status of the different cytosine containing sequences (unmethylated, hemi-methylated, doubly methylated). A simple technique to detect the methylated and unmethylated cytosines within DNA sequences will be important for a wide academic, clinical and industrial research community, since this will allow a greater understanding of gene regulation. There are many research areas where cytosine methylation is considered to play a significant role in humans, such as diet related disease, inflammatory diseases, embryonic development to name a few, or in plants for understanding the effect of environmental stress. But as noted above, cytosine methylation is important for many organisms, and a technique that allows for the analysis of the patterns of methylation within genes has the potential to be commercially valuable in the longer-term. First a better understanding of DNA methylation is required, but it is possible that a form of the approach proposed here will yield a diagnostic tool.

There are a number of approaches that have been tried, tested, dropped and adopted, but it is perhaps true to say that of the current assays at the disposal to researchers in the field no single assay fits all needs. To map methylcytosines in DNA, the gold standard was considered to be treatment with bisulphite which results in the conversion of unmethylated cytosine to uracil and thus the resulting DNA strand can be amplified and cloned. The resulting strands can thus be detected by using microarrays, but this has limited the sequences that have been interrogated. Recently the combination of bisulphite treatment with high-throughput sequencing has allowed for the mapping of methylated cytosines in the genome of the plant arabidopsis thaliana. But the use of such a method for organisms with larger genomes is impractical because so many samples must be sequenced to study tissue and disease specific variability. The problem has been even more complex with the recent finding that 5-hydroxymethylcytosine is present and the bisulphite followed by sequencing methods does not resolve this. Single DNA molecule analysis is considered by many as the potential way forward. Here we will investigate a new way to interrogate individual double-stranded DNA sequences. Novel oligonucleotide probes that associate with specific DNA sequences as triplexes will be synthesised; these probes will be designed to associate to sequences containing methylated, hemi-methylated and unmethylated cytosine. . Probes will also be developed during the later stages of the programme for the detection of sequences containing 5-hydroxymethylcytosine.. These probes will be labelled with different fluorophores. The fluorescently labelled probes will be interrogated in fluidic channels using optical methods.

Research in the field of epigenetics, notably DNA methylation, is severely hindered by a lack of simple, cheap, fast technologies for reliable analysis. There is enormous diversity of DNA methylation analysis techniques, and many of these techniques yield results that cannot be compared easily. Because DNA is collected from a number of cells either the result is obtained as an average methylation level across many DNA molecules or as a pattern along short sequences of individual DNA molecules. The level of information obtained is severely hindered by speed, cost and effort needed. In addition the information is also compromised by the potential flaws in some of the techniques used including polymerase bias, frequency of mutations of methylcytosine to thymine (and hence masking of information by the bisulphite conversion when used in combination with sequencing methods), and the more recent finding of 5-hydroxymethylcytosine, which is not detected by the bisulphite-sequencing approach. New techniques for the analysis of DNA methylation will significantly 'openup' the opportunities for researchers working in the field of epigenetics, as well as opening up opportunities for companies supplying these research tools. Eventually, an understanding of DNA methylation will provide significant opportunities in the fields of diagnostics, an area where the UK has current industrial growth. The UK has an ageing population and a society with a poor diet; many of the diseases associated with individuals with these profiles are considered to be a result of aberrant DNA methylation profiles. Diagnostics in the future are likely to contain assays for DNA methylation analysis. DNA methylation is also important for plants, invertebrates and other vertebrates. Understanding the impact of the environment on these organisms on DNA methylation will be important for a wide range of agricultural activities. This will be particularly important in understanding the role of stress, the environment, the length of a day, on various crops. But it is also important for understanding the interplay of temperature, nutritional, brain and reproductive networks in mammals, insects and fish. An understanding of these factors will be important for efficient food production in the future.