Scientists use genomic tools to create maps of DNA methylation

Much of the field of stem cell biology and development remains
uncharted territory. Just as famous explorers and astronomers mapped
out landmasses and constellations, researchers are working fervently to
chart the molecular landscapes within stem cells — especially embryonic
stem cells. A clearer understanding of the cells’ unique properties,
particularly their ability to give rise to nearly any type of cell,
could unlock fundamental questions about biology and may even spur
novel ways to treat disease.

A team of researchers at the
Broad Institute of Harvard and MIT has helped break new ground in stem cell research
through work described in two recent Nature
papers. The most recently published study, appearing in the July 6
advance online issue, involves an effort to map regions of cells’
genomes marked by DNA methylation — one of several so-called
‘epigenetic’ modifications.

If DNA is the blueprint of a living
organism, epigenetic marks, often in the form of chemical tags called
methyl groups, are the gatekeepers to that blueprint. When affixed to
DNA or to its protein scaffold (called “chromatin”), methyl groups can
enable genes to be switched on or off, orchestrating signals that allow
cells in the body, which share the same DNA, to assume different forms
and functions.

In work published last year, Broad Institute
researchers applied genomic tools to map the methylation of chromatin
proteins called histones across the genomes of several types of cells,
including embryonic stem cells. To complete that “epigenomic” picture,
they decided to expand their work to include DNA methylation. “We used
some of the latest genomic technologies,” said co-first author Alex
Meissner, an assistant professor in Harvard’s new Department of Stem Cell and Regenerative Biology, “to address a question many have wondered about: what’s the
role of DNA methylation in cell development and differentiation?”

A long road towards DNA methylation maps

Researchers
from the Whitehead Institute, Harvard University, and Harvard Medical
School came together at the Broad Institute to analyze DNA methylation
throughout the genomes of embryonic stem cells, as well as more
developmentally mature cells.

Meissner said the new DNA methylation maps are the result of a
long-term effort to address fundamental questions about how epigenetic
factors influence cell development.

Researchers are able to
create these maps using a technique known as bisulphite DNA sequencing.
Although epigenetic information generally cannot be read from the As,
Gs, Cs, and Ts that make up the DNA code, it turns out that a special
chemical, sodium bisulphite, actually makes it possible to detect
epigenetic modifications. Just as fine powders help detectives identify
otherwise invisible fingerprints, sodium bisulphite helps scientists
visualize the spots in a cell’s genome that harbor methyl groups and
the spots that do not. The technique offers detailed views of DNA
methylation and can now be implemented on a large-scale due to advances
in high-throughput sequencing technologies.

With these
advanced technologies, the scientists created DNA methylation maps of
embryonic stem cells, as well as cell types derived from them,
signifying the first such maps of mammalian cells. Several findings
stood out from careful analyses of the maps, the most notable of which
was the correlation between DNA methylation and histone methylation.
Just as a topographic map of steep terrain and a political map of
countries and borders may show similar patterns, chromatin and
methylation maps can be used individually, or, more effectively,
together to see a clearer picture of the molecular landscape. “In the
past, these two types of epigenetic marks were rarely studied
together,” said Tarjei Mikkelsen, a graduate student at the Broad
Institute and co-first author of the latest Nature
paper. “By examining them as a whole, we now have one of the first
integrated pictures of epigenetic changes during cellular development.”

By perusing the maps, the researchers, led by Broad director
Eric Lander, were also able to pick out specific sites within the
genome where methylation fluctuates as cells develop, such as when
embryonic stem cells mature into neural cells. Peering more closely at
these dynamic changes, they identified certain sites associated with
developmental genes that become overly methylated. 

“Hypermethylation
can be a sign that nearby genes are inaccessible, permanently shut off.
And it is something that’s commonly observed in the genomes of tumor
cells,” said Meissner. “These maps as well as the approaches used to
create them may help shed light on the role of DNA methylation in human
cancers.”

Increasing the efficiency of reprogramming

DNA methylation was also at the core of the scientists’ earlier paper, published in July 3 print issue of Nature.
That paper described several molecular hurdles that impede a powerful
technique in stem cell research — a recently described laboratory
procedure that can nudge adult cells into a more primitive, stem-cell
like state. This cellular “reprogramming” is now the focus of intense
interest as a potential way to artificially derive embryonic stem cells
from readily available adult tissues, such as skin. The method, though,
can be slow and inefficient, with most cells failing to be reprogrammed.

Epigenetic
marks such as DNA methylation are thought to act like a kind of memory
storage for cells, helping cells “remember” their identities by keeping
certain genes turned off and others on. If that’s true, then
reprogramming likely requires those memories to be reset or wiped
clean, allowing cells to assume new identities. During the course of
their research, the scientists discovered that some epigenetic
information, particularly DNA methylation, is especially difficult to
expunge, hindering the reprogramming process. They then showed that
treating incompletely reprogrammed cells with a drug that temporarily
inhibits DNA methylation could greatly increase the efficiency of the
process.

“The same genes that are slow to respond to
reprogramming are the genes we see hypermethylated early on in
development,” Mikkelsen said  “Improving the low efficiency of the
reprogramming process required circumventing this mechanism without
disabling it permanently.”

Having a map of these mechanisms
could help researchers orient themselves in genomic space and develop
additional methods to steer cells safely through the entire
reprogramming process. Perhaps, like explorers before them, researchers
today will also come across new discoveries along the way that will
fill in more of the epigenetic maps.

Other Broad researchers
who contributed to these papers include Eric Lander, Andreas Gnirke,
Xiaolan Zhang, Bradley Bernstein, Andrey Sivachenko, Hongcang Gu, Chad
Nusbaum, and David Jaffe. Researchers from the Whitehead Institute,
Massachusetts General Hospital, and Harvard Medical School also
contributed to the research.