Researchers have identified a possible mechanism by which DNA regions that don't encode proteins can still determine phenotypic traits such as a person's height or susceptibility to a particular disease, researchers report online in Science today.

Image: Wikipedia

The scientists found that certain chromatin modifications often considered to be epigenetic -- meaning, regulated by factors other than genetic sequence -- are in fact determined by a person's DNA.

Moreover, they found that this chromatin variation is associated with distinct single nucleotide polymorphisms, suggesting that the variation may serve as a platform to enable these SNPs -- often found in non-coding regions of DNA -- to influence phenotype.

"This is quite novel," said Emmanouil Dermitzakis, a geneticist at the University of Geneva Medical School, who was not involved in the study. "Epigenetics has been used as a term that is orthogonal to genetics. This study clearly shows it's not."

Genome-wide association studies have linked single nucleotide polymorphisms (SNPs) to particular diseases or characteristics, but how SNPs relate to phenotype has been unclear. Because SNPs often occur in non-coding regions of the genome, researchers have generally thought that what links these regions to phenotype are processes that control how genes are regulated, but such a relationship has never been demonstrated. "What this paper really does is show this to be the case," said Vishwanath Iyer of the University of Texas at Austin, one of study's three lead authors.

The researchers examined two processes involved in gene regulation: chromatin structure -- in particular, whether or not chromatin is open, allowing genes to be transcribed -- and transcription factor binding. Both of these features can be regulated epigenetically -- that is, by factors such as DNA methylation and histone modifications.

They used high-throughput sequencing to analyze cell lines taken from six individuals whose genomes had originally been sequenced within the 1000 Genomes Project -- two parents and their daughter from Utah, of northern and western European ancestry, and two parents and their daughter from Ibadan, Nigeria, of Yoruban ancestry. This pool of six people allowed them to compare these factors across both alleles of genes carried by single individuals, as well as examine familial heritability in the two daughters. "It's the first time someone has looked at chromatin structure throughout the genome in related individuals," Iyer said.

In 10 percent of the sites they examined, they found that chromatin sites that tended to be open in the parents also tended to be open in their daughters, suggesting chromatin structure was heritable. Similarly, many differences in transcription factor binding were maintained across individuals.

But these differences could still be the result of heritable changes in epigenetics, not genetics. So the authors also looked at these traits at at both alleles in single genes in the individuals' genomes. They found that, in a subset of sites, chromatin structure and transcription factor binding occurred differently at each allele, which suggested it was the genetic code at each allele that produced the difference. "We find that it is actually the sequence that matters," said Iyer.

If a non-coding SNP leads to a difference in chromatin structure and/or transcription factor binding, that effect can indeed be inherited, thereby providing a potential mechanism for transmission of phenotypes like disease susceptibility, Iyer said. "What we have is a way to link polymorphisms to phenotypes via things like chromatin structure and transcription factor binding." He stressed that not all forms of chromatin-level regulation were dependent on the genome sequence. "We are looking at one transcription factor and one measure of chromatin structure," he said; other kinds of chromatin variation, however, may be due to mechanisms such as DNA methylation, which are independent of the sequence.

"The paper in itself is not really surprising," said Maxwell Lee, a geneticist at the National Cancer Institute, but the researchers examined variation at many more loci than previous studies have looked at, and also were able to examine variation both at the level of different chromosomes as well as different individuals.

The study was done in the context of the ENCODE Consortium, a large-scale project conducted by the National Human Genome Research Institute that aims to understand all the functional elements in the human genome. The next step, said Iyer, is to examine this type of variation in a much larger set of people to determine when it is functional, and exactly how it drives phenotype.


Related stories:

Epigenetic suicide note
[August 2009]

Burning chromatin at both ends
[March 2009]

An epigenetic inheritance
[January 19th 2009]