Issue Date: June 6, 2011
New Base On The Block
Like most folks with a basic knowledge of genetics, you probably believe that life’s blueprint is based on just five components. Namely, DNA’s four bases—adenine, cytosine, guanine, and thymine—and the periodically methylated cytosine bases that help silence genes so that a heart cell doesn’t produce toenails and puberty doesn’t hit a toddler.
It turns out that you, along with a battery of geneticists, biochemists, and textbook writers, may have missed an important genetic base. As Yoda says to Luke Skywalker in “The Empire Strikes Back,” “There is another.”
The sixth base in life’s blueprint is 5-hydroxymethylcytosine (5-hmC), a cytosine decorated with a methylene and a hydroxyl group. For decades, scientists had observed 5-hmC in bacterial viruses, but it was generally thought to be an experimental artifact or random DNA damage in mammalian cells. “5-hmC was pretty much ignored—until two years ago,” explains Chuan He, a chemist at the University of Chicago. “That’s when two labs independently showed to everybody’s huge surprise that 5-hmC was found in genomic DNA from both brain and stem cells.”
Two back-to-back articles published in Science in May 2009 “pretty much changed everything,” says Thomas Carell, a biochemist at Ludwig Maximilian University in Munich. Suddenly, a new genetic base seemed to be playing a role in brain research and developmental biology, two of life sciences’ most interesting subject areas (DOI: 10.1126/science.1170116 and 10.1126/science.1169786).
“Since then, basically every week or two a paper on 5-hmC in an outstanding journal is published,” Carell says. He’s not exaggerating. More than 50 articles have been published just this year in Nature and several other prominent journals, Cell, Proceedings of the National Academy of Sciences USA, Journal of the American Chemical Society, and Angewandte Chemie International Edition—many of them just in the past month or two.
Now the race is on to figure out 5-hmC’s exact function in a cell. Results so far hint that 5-hmC plays a global role in human development, memory, and cancer. Because the modified cytosine is often found inside genes, researchers believe there’s a high probability that the 5-hmC modification, just like methylcytosine’s, helps supervise which genes are expressed in one cell type and locked in deep storage in another. But even as researchers invent new ways to profile the presence of 5-hmC, scientists say a lack of analytical techniques to study the modification hampers the new field, and they call for more chemists to help develop methods to probe this tiny but important component of the genetic code.
The two groups that spawned this field in 2009 were not competitors—they didn’t even know each other—and they had little in common except for an interest in the rather extensive field of DNA methylation research.
In fact, the groups’ simultaneous discovery of 5-hmC was just pure “coincidence,” says Nathaniel Heintz, a molecular biologist at New York’s Rockefeller University. Heintz and his postdoc Skirmantas Kriaucionis, now at Oxford University, were trying to figure out whether the nuclei of meganeurons, called Purkinje cells, were unusually bulky because of substantial DNA methylation. While trying to isolate methylated DNA, Skirmantas noticed “an extra spot on the chromatogram,” Heintz says. The two scientists excitedly discussed possible biological implications and how to prove the spot wasn’t an artifact.
Two hundred miles away at Harvard University Medical School in Boston, Anjana Rao and her graduate student Mamta Tahiliani, now at New York University, were looking for elusive enzymes that demethylate DNA. In the first moments after the DNA from a sperm and an egg combine, there seems to be a rapid and extensive loss of methyl groups from cytosine residues in the new genome. This stripped genome results in an embryonic stem cell that has the power to turn into any one of the thousands of different cells in our body, explains Rao’s collaborator Suneet Agarwal, who is also at Harvard Medical School.
Rao’s team did not find the proposed demethylase that carries out this process—nobody has to date—but Rao and Tahiliani did discover a family of enzymes that build 5-hmC, starting from methylcytosine.
“I knew nothing about the Heintz study until the day after I presented our work” at a National Institutes of Health workshop, Rao says. “Someone who heard my talk was traveling home on the train and ran into another faculty member who was reviewing the Heintz manuscript. I imagine the conversation went something like this: ‘I heard a talk about a new base called 5-hydroxymethylcytosine.’ Response: ‘That’s strange. I’m reviewing a manuscript about the same new base.’ ” Rao says she contacted editors at Sci ence and requested that the 5-hmC papers be reviewed together.
That the new base exists in several cell types and that enzymes actively produce 5-hmC persuaded many in the field that this modification is an intentional mark and not simply a transient intermediate or random bit of DNA damage, Carell says.
Then cancer scientists discovered that one of the enzymes making 5-hmC “is mutated in an unprecedented number of blood cancers,” and excitement about 5-hmC just snowballed, Agarwal says. Dozens of laboratories around the world began to focus on 5-hmC, and suddenly it was observed in a variety of additional cell types.
For example, brain cell DNA has the most 5-hmC; some 0.3 to 0.7% of the bases in brain DNA are 5-hmC. Heart and kidney cells possess a medium amount of 5-hmC, while liver, testes, and stem cells have the lowest levels. This variation was somewhat unexpected because the more abundant epigenetic marker, methylated cytosine, is found at fairly constant levels, about 4.5% of genomic DNA bases in most cell types, Carell explains.
Levels of 5-hmC also seem to vary with time. For example, He found that levels of 5-hmC in a newborn mouse brain triple in the first three weeks of life. This fact, coupled with the particularly high abundance of 5-hmC in the brain, makes some researchers suspect that 5-hmC may have a role in cognition and memory, says Hon gjun Song, a neurobiologist at Johns Hopkins University School of Medicine.
But even as scientists detect 5-hmC in the genomes of an increasing variety of cell types, nobody knows exactly what the modification does. “It’s still a guess,” Song says.
Part of the problem is that nobody has detected a protein that can bind 5-hmC, although “many groups—including my own” are in hot pursuit, Song explains. Finding binding proteins could illuminate which cell signaling networks 5-hmC activates or represses.
Another problem is that because the field is so new, the methods and instrumentation available to study these modifications are limited. “We need chemists to come up with better analytical technology,” Song says. For example, existing high-throughput-sequencing platforms simply can’t distinguish precisely where the 5-hmC modification is on the genome. A solution to this problem may come from a technique called single molecule real-time sequencing. It has the appropriate discrimination powers, but the technology is in the early stages of development and cannot yet do the necessary processing of whole genomes in a high-throughput manner (Nat. Meth., DOI: 10.1038/nmeth.1459).
The field would not have advanced as much as it has if there weren’t some analytical methods that can profile the basic abundance of 5-hmC in a genome. Antibodies specific to 5-hmC are used in one detection strategy. However, antibodies may not be sensitive enough to identify lone 5-hmC modifications, Agarwal says. Multiple 5-hmCs might need to be in close association to yield a hit with antibodies, he says.
Last December, He reported a sensitive technique that can detect lone 5-hmCs. The methodology begins with an azide-functionalized glucose molecule that is transferred to the hydroxyl group of 5-hmC via a β-glucosyltransferase (Nat. Biotechnol., DOI: 10.1038/nbt.1732). Using a “click” cycloaddition, the azide is connected to a biotin tag that can then be used for purification after the DNA is chopped up. The resulting DNA fragments can be sequenced, providing information about where the modification is located. Because even a small fragment of DNA has multiple cytosines, determining the exact placement of a given modification remains elusive. Working with Peng Jin at Emory University, He used the technique to find that in elderly mice, segments of the genome typically associated with neurodegenerative disease had more 5-hmC than those of young, healthy mice.
Last month Rao and Agarwal reported two new methods for profiling 5-hmC (Nature, DOI: 10.1038/nature10102). In the first technique, an enzyme transfers a glucose molecule onto 5-hmC, but then periodate oxidation modifies the sugar so that it can be pulled out of the sample. In the second technique, 5-hmC is first converted to cytosine 5-methylenesulfonate by reaction with sodium bisulfite. The researchers then extract their 5-hmC target via immunoprecipitation. Using the techniques, the team showed that 5-hmC is enriched in “promoter” areas of the genome, where transcription machinery lands to start reading DNA. This suggests that 5-hmC may play a role in regulating transcription, Agarwal says.
Although there are many implied roles for 5-hmC, a recent report from Song provided a more conclusive occupation for the modification, Carell says (Cell, DOI: 10.1016/j.cell.2011.03.022). In it, the team provides data that suggest 5-hmC is a step on the pathway by which methylcytosine becomes unmethylated—the process needed in a fertilized egg to produce a pluripotent embryonic stem cell.
Naysayers might be tempted to conclude from this paper that the 5-hmC modification is just an intermediate along the pathway from methylcytosine to cytosine, and that all the buzz about 5-hmC’s role as a new, independent epigenetic marker is just hot air. But Song and all the researchers C&EN interviewed argue otherwise. They believe there’s enough evidence to suppose that 5-hmC probably has several roles in a cell and that serving as an intermediate is just one of many 5-hmC functions—some of which haven’t yet been found because the field is so new. Now they’ve just got to do the hard work to prove it.
- Chemical & Engineering News
- ISSN 0009-2347
- Copyright © American Chemical Society