Posted by Jef Akst

[Entry posted at 3rd February 2010 10:00 PM GMT]

 

With the Human Genome Project largely complete, scientists are turning to variation in the epigenome and beginning to map chemical modifications of DNA that affect gene expression. Two recent studies that provide the first comprehensive maps of human DNA methylation -- one of the most commonly studied epigenetic modifications -- and a new initiative that aims to generate 1,000 more are a testament to this new focus in genetics research.

Image: Wikimedia commons, National Human Genome Research Institute

In the first direct comparison of the DNA methylation patterns at two different stages of differentiation in a single cell line, published online today in Genome Research, stem cell systems biologist Jeanne Loring of the Scripps Research Institute and her colleagues detailed how methylation changes over the course of development.

The changes were "much more subtle and clever than I would have imagined," Loring said. "Some [regions] remain methylated, some become demethylated, some become methylated," and these changes can occur in "certain genes, certain parts of genes, and certain things that aren't genes," she added. "I can't imagine that any of that is not important." Overall, the amount of methylation in the differentiated cells was lower than in hESCs, but in the genomic regions that differed between the two cell types, differentiated cells had higher levels of methylation.

"This paper does provide a more comprehensive characterization of methylation during differentiation" than previous studies, Yuan Gao of Virginia Commonwealth University, who was not involved in the research, wrote in an email to The Scientist. Such information may make it "possible to use methylation markers to quantify or characterize the "undifferentiatiedness" of a cell and its pluripotency."

This paper follows close on the heels of a similar epigenetic mapping project, published by molecular biologist Joseph Ecker of The Salk Institute for Biological Studies, La Jolla, California, and colleagues in Nature last October. That study, based on hESCs and fetal fibroblasts of a different cellular origin, presented the first genome-wide, single-base-resolution methylation maps. "It's is great that [Loring's] study pretty much confirms all of our findings," Ecker wrote in an email to The Scientist, "which is the way science should work!"

Loring believes these two papers are only the first of many. With the recent technological advancements in the field, "it's now feasible and relatively affordable" to create these maps, she said, which "open the door to understanding gene regulation and epigenetic regulation [in different] cell types." The amount of data generated by these studies is enormous, added Loring, and she hopes that other scientists will take advantage of this plethora of information. "The data are all there," she said. "All they need is for somebody to look at them."

Such information could be "used as a quality control tool" in induced pluripotent stem (iPS) cell production or research, Gao wrote in his email. In addition, "beyond how stem cells are programmed, understanding how genes change in response to their environment is of obvious importance," Ecker added. For example, some studies "show that the changes in the epigenome are affected by things like diet and obesity/hunger." But to understand such effects, he said, "first you need to map these marks."

Loring is already working on the next map -- that of a young neuron. In addition, a recently announced initiative dubbed the International Human Epigenome Consortium aims to map 1,000 reference epigenomes in the next 10 years.

"Epigenomes are changeable, programmable and will feed us the bottom line on how the genome works," Rob Martienssen of Cold Spring Harbor Laboratory in New York told Nature yesterday.