Map of DNA methylation across tissues adds another layer to understanding of disease

Genome wide association studies (GWAS) offer great promise toward understanding the genetic causes of disease. By comparing the genomes of many different people, researchers hope to identify genetic changes that are associated with common conditions such as cancer, diabetes, and heart disease.

However, most diseases aren’t caused by variations in the coding regions of genes directly—instead these variants exert their influence by regulating how nearby genes are expressed. Scientists can identify expression quantitative trait loci (eQTLs), or genetic variants that are linked to differences in gene expression between individuals, and yet that still doesn’t paint a complete picture of how these are in turn linked to complex traits in different types of tissue.

In a new study published this week in Nature Genetics, researchers from the University of Chicago added another layer to our understanding of how genetic variants influence human traits. Epigenetic changes control gene activity without altering DNA sequences themselves; one common type of epigenetic change is methylation, in which a handful of atoms are attached to DNA and silence it, turning off expression. The team, led by Meritxell Oliva, PhD, a former postdoctoral researcher at UChicago now working at AbbVie, analyzed nearly 1,000 human tissue samples, and catalogued genetic variants that may affect methylation patterns (mQTLs, or methylation quantitative trait loci). The hope is that this additional data will provide more insight into the impact of genetic variations on disease.

“We produced this map to find links between positions of the genome that are associated with complex traits and diseases and identify those that are also associated to DNA methylation levels in this set of tissues,” Oliva said. “By doing that, our aim is to try to understand mechanisms that are causal to complex traits and the diseases.”

Most studies on mQTLs thus far have used blood samples, mainly because blood is easy to collect. But gene expression and patterns of methylation can vary across different tissues, so, sampling blood doesn’t tell the whole story. The research team worked with samples of nine different types of human tissues, chosen because they are common sites of cancer (breast, colon, lung, kidney, prostate), or because they age uniquely (muscle, ovary, and testis). The samples came from the GTEx Project, a publicly funded database housing human eQTL data, which can give a more extensive view of when and where genetic variation has an impact, instead of just what can be detected in the blood.

“DNA methylation can be highly context dependent, which means if we get the data from one tissue type or one cell type, it could potentially have different genetic effects than in other tissues and cell types,” said Lin Chen, PhD, Associate Professor of Public Health Sciences, and a senior author of the study. “For complex diseases and traits, many of those genetic effects are quite a context dependent, so, this tissue specific information is very valuable.”

The result of this analysis provides a useful tool for researchers hoping to understand the effects of genetic variation on methylation levels, which in turn affect complex traits and the development of disease.

“The past decade of work has pointed us to many locations of interest in the genome, and for every single one of those locations, the downstream goal is to understand the biology of how those differences end up impacting people's health,” said Brandon Pierce, PhD, Associate Professor of Public Health Sciences, and another senior author of the study.

“We can see that these effects on DNA methylation are fairly abundant in the genome, and many of them appear to line up with effects on health and disease. This tool provides another layer of biological explanation for the effects of genetic variation on health.”

The study, “DNA methylation QTL mapping across diverse human tissues provides molecular links between genetic variation and complex traits,” was funded by the National Institutes of Health. Additional authors include Kathryn Demanelis from the University of Pittsburgh and Yihao Lu, Meytal Chernoff, Farzana Jasmine, Habibul Ahsan, and Muhammad G. Kibriya from UChicago.