SCOT NICHOLLSResearchers have made an extensive map of several types of methylation in the brains of mice and humans. The work, described in Science today (July 4), shows that epigenetic modification varies greatly over the course of development but is remarkably consistent between individuals and between mice and humans.
“This is the very first full-scale profiling into DNA methylation in the developing brain,” said Paolo Sassone-Corsi, a molecular biologist at the University of California, Irvine, who was not involved in the research. “It basically is a wealth of amazing information for a large number of researchers.”
Study author Margarita Behrens, a neuroscientist at the Salk Institute in San Diego, said she started the project because she wanted to understand the role of epigenetic changes in the brain as mental illnesses took hold in humans. She chose to look at DNA methylation—a process in which methyl groups are added to nucleotides—in the frontal cortex because the brain region is associated with executive function and can show abnormalities in people with mental disorders.
Using whole genome bisulfite sequencing to map methylation of cytosine nucleotides in the frontal cortices of mice, she and colleagues noticed something unusual. Most methylation in adult mammals happens where cytosine and guanine nucleotides sit next to each other on a DNA molecule. This common type of modification is called CG methylation.
Cytosine can also be methylated in locations where it sits next to another nucleotide, but in most regions of the body this non-CG methylation has seldom been observed in differentiated cells. Behrens’s infant mice lacked non-CG methylation in their frontal cortices, but it rapidly accumulated as they aged and their brains developed. “When we found it in the brain we said, ‘What is this doing there?’” Behrens recalled.
The researchers found similar patterns of non-CG methylation accumulation in the human brain and noticed that the modifications accumulated especially rapidly in the first two years of life, a time of rapid synapse formation. Non-CG methylation continued to build through adolescence before decreasing slightly in older subjects. The researchers also noted that non-CG methylation of genes in the same subjects was associated with lower levels of gene expression.
Behrens speculates that the non-CG methylation could help determine the character of neurons during development. “The fine tuning of the expression of the transciptome could be related to the non-CG methylation,” she said.
The researchers also mapped hydroxymethylation, in which a hydroxymethyl group modifies DNA, in mouse and human frontal cortexes. They found that hydroxymethylation primarily occurs at CG sites. Intriguingly, they noted that certain sites with high levels of hydroxymethylation in fetuses were often demethylated in adults.
The patterns of methylation were highly conserved between individuals who were tested and even between mice and humans. “The conservation between individuals is a strong clue [the methylation] may have a meaning,” said Eran Mukamel, a neuroscientist at the Salk Insitute and coauthor of the paper.
Now that she’s mapped the frontal cortex, Behrens hopes to go back to her original goal of understanding mental disease, using her map as a reference.
“It gives us a landscape of the normal pattern of development,” said Mukamel. “It will be interesting for people who want to look at disruptions that occur in psychiatric illness, learning, and memory.”
R. Lister, “Global epigenomic reconfiguration during mammalian brain development,” Science, doi:10.1126/science.1237905, 2013.