Editorial Article
Wednesday September 08, 2010
by Jeffrey M. Perkel
Within each human genome is a record of that individual's genetic past, and a glimpse into their future. Yet it is only a glimpse. A few high-profile disease loci notwithstanding, most human traits are not written solely in triplets of As, Cs, Gs, and Ts; other factors have a say as well.
One of those factors is epigenetics, a set of chemical events—DNA methylation, histone modification, and the transcription of small and large non-coding RNAs—that combine to tweak an organism's phenotype without altering its genomic sequence. If the genome is a blueprint, epigenetic events are like transparent informational overlays that enrich that genome like a multicolor molecular atlas.
Given that those overlays have been implicated in such processes as cancer, stem cell biology, development, and addiction, it is hardly surprising that epigenetics is an intensively studied biological discipline. And it is one for which there exists a robust—and evolving—set of molecular tools.
"Epigenetics is one of the more exciting things going on in biology right now," says Shannon Dempsey, epigenetics market segment manager at Sigma-Aldrich. "The fact that we've begun to uncover this beautifully complicated system of regulation that exists beyond just the order and composition of bases in DNA, is amazingly interesting to me."
Researchers have devised several strategies to track CpG methylation, a chemical alteration that is generally associated with diminished gene expression, on a number of informational levels.
To obtain a global view of methylation, researchers have traditionally hydrolyzed genomic DNA and resolved the resulting base mixture using thin-layer chromatography, high-pressure liquid chromatography, or mass spectrometry. Researchers still do that of course. But there's an easier alternative. Sigma-Aldrich's Imprint® Methylated DNA Quantification Kit provides similar data in an ELISA format. According to Dempsey, the Imprint kit "allows [researchers] to pretty quickly measure global methylation shifts."
Epigeneticists can get a finer-grained analysis by digesting DNA with both methylation-sensitive (e.g. HpaII) and methylation-resistant (e.g. MspI) restriction enzymes and amplifying what's left, the presence or absence of a specific amplicon being diagnostic for the region's methylation status. Zymo Research's soon-to-be-released OneStep pMethyl-PCR kit is one commercial form of this assay.
For nucleotide-level resolution, researchers can use bisulfite conversion. Bisulfite chemistry transforms unmethylated cytosine residues to uracil, while leaving methylated cytosines alone. Following conversion, the DNA can then be sequenced (bisulfite sequencing), or analyzed by PCR. In one variation of the latter approach, researchers use two different primer sets and ask which of the two can amplify the treated DNA, a technique called methylation-specific PCR. Alternatively, the methylation-selective reagent can be a TaqMan-like probe that is complementary to the methylated site. If that probe is built using Exiqon's locked nucleic acid chemistry, says Peter Mouritzen, the company's director of life science product development, the assay becomes nearly binary. "The introduction of LNA bases in a fully substituted 8-mer allows you to distinguish one base pair by [a melting temperature of] 27°C versus 8-9°C with DNA probes," Mouritzen says. "That's enormous, and what it translates to is you get an almost digital readout—either it's there or it’s not."
Roche Applied Scienceoffers an alternative approach on its LightCycler® 480 real-time PCR instrument, called high-resolution melting. As Larson Manifold, LightCycler marketing manager, explains, bisulfite converted DNA has a different G/C composition than the starting DNA, which in turn translates into a difference in the number of hydrogen bonds holding the two DNA strands together; A:T base pairs have two hydrogen bonds, while G:C pairs have three. As a result, the two samples will differ in their melting profiles. "We have seen sensitivity down to 1-5% changes in the methylated state of samples," using this approach, says Manifold; that is, it can detect the difference between three and five methyl residues in a 100-bp fragment.
Bisulfite conversion kits are available from a number of vendors, including Sigma-Aldrich, EMD Millipore, Active Motif, and Zymo Research. Although bisulfite conversion is really the only game in town for those who want nucleotide-level resolution, the epigenetics community does not love it. Laborious, technically challenging, and prone to artifacts due to over- or under-conversion, bisulfite chemistry is, says Michael Sturges, epigenetics product manager at EMD Millipore, "a necessary evil."
"Bisulfite is very harsh," agrees Sriharsa Pradhan, research division head of RNA biology at New England Biolabs, "it can damage DNA like nobody's business. You also don't know the percent conversion efficiency," meaning it is difficult for researchers to be confident they are accurately reading the methylation status of a given piece of DNA.
Commercial kits certainly simplify the process. Zymo's EZ DNA Methylation-Direct™ Kit, for instance, promises conversion efficiencies of up to 99.5% from input samples of as little as 10 intact cells (as opposed to purified DNA) in just four hours (compared to 16 hours for some protocols), according to Scientist James Yen.
Yet alternatives are coming. New England BioLabs will release a trio of restriction enzymes this October that includes MspJ1, which digests many methylated or hydroxymethylated DNA sequences into precise 32-bp fragments with methylC or hydroxymethylC at the center. Researchers can thus identify methylated bases, without chemical treatment, simply by digesting genomic DNA, purifying the 32-bp products, and sequencing the resulting pool. "We give the customers a choice not to do bisulfite conversion and to sequence the DNA directly," Pradhan explains.
Also available are methyl-DNA enrichment kits. These come in two flavors: MeDIP, or methylated DNA immunoprecipitation, relies on antibodies to 5-methylcytosine; the alternative approach uses methyl-DNA binding proteins to accomplish the same thing. Commercial enrichment kits include the MethylCollector Ultra kit (Active Motif), MethylMiner (Life Technologies), EMD Millipore's soon-to-be-released CpG MethylQuest™ DNA Isolation kit, and Zymo Research's Methylated-DNA IP Kit. The purified DNA can then by analyzed by next-gen sequencing or on specially designed microarrays, such as Roche NimbleGen's 2.1M Deluxe Promoter Array.
Researchers interested in the protein side of chromatin must use an entirely different set of tools.
One classic tool is ChIP, or chromatin immunoprecipitation. In ChIP, DNA-protein complexes are frozen in place with fixative. The DNA is then fragmented, immunoprecipitated with antibodies to proteins of interest, such as specific histone modifications, transcription factors, or remodeling enzymes, and then uncrosslinked to release the enriched DNA. That pool can then be analyzed either by PCR, microarrays ("ChIP-chip"), or next-gen sequencing ("ChIP-seq").
ChIP reagents and kits are widely available from the likes of Sigma-Aldrich (Imprint® ChIP Kit), EMD Millipore (Magna ChIP™ Chromatin Immunoprecipitation kits, Magna ChIP2™ DNA microarray kits, and ChIP Ab+ validated antibody/primer sets), Roche NimbleGen (ChIP-chip DNA microarray kits), Agilent Technologies (Human ENCODE ChIP-on-chip Microarray), and Illumina (ChIP-Seq DNA Sample Prep Kit). All of these kits simplify and streamline the process over traditional home-brew protocols. For instance, Life Technologies’ MAGnify™ Chromatin Immunoprecipitation System uses an optimized pull-down protocol to shorten the assay from an overnight incubation to an hour-and-a-half, explains Christofer Cunning, senior manager of market development. More importantly, Cunning says, researchers can reduce their input material from millions of cells down to "as little as 10,000 cells per experiment."
Active Motif offers a wide selection of histone-analysis tools, including histone purification kits, recombinant histones (which can be used to create "designer chromatin," according to Product Manager Kyle Hondorp), histone modifying enzymes and antibodies, and ELISAs for global quantification of specific histone modifications.
One new tool in the company's portfolio is the MODified™ Histone Peptide Array, an array of 384 distinct histone N-terminal tails (in duplicate) per microscope slide, each bearing a distinct set of up to four modified residues. According to Hondorp, the array can be used to study protein-protein interactions, enzyme and antibody specificity, and the like. "People using these arrays can find if there's an issue with cross-reactivity with an antibody, and they can also study neighboring effects—that is, if a neighboring modification interferes with or prevents protein or enzyme binding [at a second site]."
Active Motif's toolkit also contains a reagent for addressing a newly discovered niche in epigenetic biology. In late 2009, two teams reported the discovery of a new epigenetic modification, 5-hydroxymethylcytosine (5hmC), in eukaryotic DNA. [1, 2] What this base does is not yet known, but whatever its function, the normal slate of methylation techniques (such as MeDIP, bisulfite sequencing, and restriction enzyme analysis) won't be much help: They cannot distinguish 5hmC from standard 5-methylcytosine (5mC).
Active Motif, though, offers a reagent that can—an antibody specific to 5hmC. Richard Meehan and Donncha Dunican, project leaders for chromosomes and gene expression at the MRC Human Genetics Unit, Edinburgh, Scotland, whose group demonstrated that neither bisulfite sequencing nor restriction enzymes could resolve 5mC and 5hmC [3], have used that antibody in their own research. Meehan says it "works incredibly well."
The only alternative for 5hmC researchers is DNA degradation, for instance with Zymo Research's DNA Degradase™ and Degradase Plus™ enzymes, followed by chromatographic or mass spectrometric analysis.
The discovery of 5hmC—what Yen calls "the sixth base"—induced something of a seismic shift in the epigenetics landscape, Meehan says. "It means we are put in a place of ignorance again," he says, "because we don't know how much functional overlap there is between [5mC] and [5hmC], and we have to go back and reexamine all the old [methylation] results."
At the pace of current tool development, don't expect that ignorance to last for long.
References
[1] S. Kriaucionis and N. Heintz, "The nuclear DNA base 5-hydroxymethylcytosine is present in Purkinje neurons and the brain," Science, 324:929-30, 2009.
[2] M. Tahiliani et al., "Conversion of 5-methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by MLL partner TET1," Science, 324:930-35, 2009.
[3] C. Nestor et al., "Enzymatic approaches and bisulfite sequencing cannot distinguish between 5-methylcytosine and 5-hydroxymethlcytosine in DNA," BioTechniques, 48:317-9, 2010.