Jeffrey Perkel has been a scientific writer and editor since 2000. He holds a PhD in Cell and Molecular Biology from the University of Pennsylvania, and did postdoctoral work at the University of Pennsylvania and at Harvard Medical School.
The last few years have seen an explosion of interest in probing DNA methylation across the genome. Based on microarrays or more recently next-generation DNA sequencing, such experiments provide a 30,000-foot overview of epigenetic status, identifying regions worthy of further study.
But what about when you’re ready to validate and home in on those regions? It’s generally impractical to apply genome-wide scans to large numbers of samples, in part for economic reasons but also because there’s no reason to survey a genome’s worth of DNA to interrogate a handful of sites.
Fortunately, there are plenty of tools to probe these genetic loci more efficiently.
Methylation-specific PCR
One of the more common approaches for assessing the methylation status of a handful of loci is methylation-specific PCR.
Methylation-specific PCR is based on bisulfite modification. When methylated DNA is treated with sodium bisulfite, unmethylated cytosine residues are converted to uracil, but methylated residues are not. By carefully designing PCR primers to distinguish between the two, researchers can quantify the methylation state of a given location based on the presence or absence of an amplification product.
“If the DNA was methylated, you can create primers that will anneal to that converted sequence and give a PCR product,” explains Kyle Hondorp, product manager at epigenetics firm Active Motif. “If the DNA was not methylated, you will not get amplification.”
Active Motif offers tools to enrich methylated DNA prior to PCR. The company’s MethylCollector™ Ultra kit uses methyl-CpG-binding domain proteins (MBDs) to pull down methylated sequences, and its MeDIP kit uses an anti-methyl antibody to do the same thing.
According to Hondorp, although the two approaches are theoretically identical, they do have subtle differences. Anti-methyl antibodies, for instance, can theoretically pick up any methylation event—not only traditional mammalian CpG methylation but also methylcytosines in the context of CpA and CpT dinucleotides. MBDs, she notes, do not recognize those modifications. On the other hand, MeDIP antibodies only recognize methylation in the context of single-stranded DNA, whereas MBDs can handle double-stranded DNA.
Restriction enzyme digestion
An alternative analysis approach relies on enzymes that distinguish methylated and unmethylated sequences.
Restriction enzymes like HpaII and MspI (both of which cut C^CGG) and AccII (CG^CG) are differentially sensitive to DNA methylation status, cutting methylated and unmethylated sequences with varying efficiency. The results of those digestion reagents can be detected by qPCR or PCR across the putative site or using, for instance, the so-called HELP (HpaII tiny fragment Enrichment by Ligation-mediated PCR) assay. These methods provide near-nucleotide resolution without the difficulties and DNA damage induced by bisulfite conversion.
Mass spectrometry
Agena Bioscience, which acquired Sequenom’s life-science research business, has developed a methylation assay called EpiTYPER® based on the MassARRAY® Analyzer 4, a MALDI time-of-flight mass spectrometer.
The assay begins with bisulfite conversion. The DNA is then cleaned up and PCR amplified to isolate approximately 500-bp regions of interest. Next, the amplication products are transcribed into RNA in vitro and treated with RNAse A, which cleaves after uracil residues. Because unmethylated cytosines will convert to uracil during bisulfite treatment, and methylated cytosines will not, RNAse A treatment should produce different fragmentation patterns depending on the methylation state of the DNA.
To read the results, the fragmented RNA products are separated and quantified with the MassARRAY Analyzer.
According to Robin Everts, manager for Agena’s Field Applications Support & Services Laboratory, the assay enables the assessment of multiple CpG sites per PCR fragment—possibly 20 or more sites per fragment. “This is a validation tool,” Everts says. “If you do genome-wide methylation screens, and you find regions that are hyper- or hypo-methylated, our method is a great way to screen a limited number of amplicons on a large number of samples easily.”
Hydroxymethylcytosine
Tools that distinguish cytosine from 5-methylcytosine (5-mC) generally are incapable of differentiating between 5-mC and a more recently discovered modification, 5-hydroxylmethylcytosine (5-hmC). But some reagents are available.
Active Motif’s hMeDIP kit uses an anti-5-hmC antibody specifically to enrich DNA containing that modification. Alternatively, the company’s Hydroxymethyl Collector™ kit uses a beta-glucosyltransferase enzyme to couple a glucose moiety to 5-hmC residues. The glucose is chemically modified to contain an azide group, which is used in a subsequent step to biotinylate 5-hmC-containing sequences. Finally, the biotinylated DNA is pulled down using magnetic streptavidin beads.
To cut 5-hmC-containing DNA, New England Biolabs offers a restriction enzyme called AbaSI. This enzyme specifically digests 5-hmC-containing DNA that has been glucosylated with a beta-glucosyltransferase, cutting 11 to 13 bases away. “AbaS1 gives near base-level resolution including [of] low-occupancy regions,” explains Sriharsa Pradhan, a senior scientist in the genome biology department at New England Biolabs. The enzyme, he says, was used to map 5-hmC distribution in the mouse embryonic stem cell genome. “You get a window of 13 nucleotides, but it tells you exactly where the 5-hydroxymethyl site is.”
NEB’s EpiMark® 5-hmC and 5-mC Analysis Kit takes an alternate tack. Here, DNA is glucosylated to tag 5-hmC residues and then split into three reactions. One is treated with the restriction enzyme MspI, which cuts methylated and unmethylated (but not glucosylated hydroxymethylated) DNA; another is treated with HpaII, which cannot cut methylated (or glucosylated) DNA; and the last is left untreated. PCR amplification of the resulting products reveals which sites were hydroxymethylated, which were methylated and which were unmodified, enabling “an accurate percentage of different methylated species” to be determined, Pradhan says.
Targeted sequencing
For those interested in scanning a somewhat wider swath of the genome, there’s targeted next-gen DNA sequencing.
Tools such as Roche NimbleGen's SeqCap Epi kits and Agilent Technologies' SureSelectXT MethylSeq use pools of oligonucleotide probes to capture potentially methylated regions of interest regardless of methylation status. Agilent's SureSelectXT MethylSeq catalog product, for instance, pulls down some 84 megabases of genomic DNA, but custom designs targeting as little as 200 kb or so are possible, says Kyeong-Soo Jeong, a research and development scientist at Agilent Technologies. (The smaller designs tend to be less efficient at capture than larger ones, he warns.) The captured DNA is then bisulfite converted and sequenced to identify methylated residues.
This approach offers a possible advantage over other enrichment strategies, says Jeong. Antibody- and protein-based enrichment methods tend to favor relatively densely methylated regions, he says. “Our technique doesn’t depend on the methylation status or CpG density [of the DNA],” he says. “We just pull down whatever we are targeting. And that’s why we get very uniform results.”
Thanks to the tools described here, you can get uniform methylation results, too, regardless of the number of sites you’re interested in.
Image: Christoph Bock, Max Planck Institute for Informatics. (Wikimedia Commons)