In a first for the field of epigenomics, scientists at the Baylor College of Medicine (BCM) have demonstrated the feasibility of studying systemic differences in DNA methylation at the population level through a new method called target-capture bisulfite sequencing. According to the authors of a new study, these differences between individuals had remained undetectable using Illumina’s methylation platform, which has been the predominant tool for population-wide DNA methylation studies over the past decade.
“The EWAS [epigenome-wide association studies] field relied almost exclusively on Illumina arrays, which were designed without consideration of inter-individual variation in DNA methylation and generally target CpGs that show little,” the study authors noted.
Using their new method, the BCM investigators assessed differences in CpG island methylation at 4,086 selected regions in the human genome called CoRSIVs (correlated regions of systemic interindividual variation), in tissues from 188 donors who were part of the NIH Gene-Tissue Expression (GTEx) program. Results of this large-scale assessment of human methylation quantitative trait loci (mQTL)—genetic variants that may affect CpG island DNA methylation patterns—were published in an article in the journal Genome Biology (“Systemic interindividual epigenetic variation in humans is associated with transposable elements and under strong genetic control”). The findings could help clarify how epigenetic differences affect diseases.
“It’s as if there’s been this massive and very expensive fishing expedition for the last 10 years, but everyone’s been fishing in the wrong place. We hope that the new tool we’ve developed will accelerate progress in understanding epigenetic causality of disease,” said Robert Waterland, PhD, professor of pediatrics-nutrition at Baylor’s USDA/ARS Children’s Nutrition Research Center, and co-senior author of the study.
CoRSIVs can be assayed in DNA isolated from blood and do not reflect differences in cellular composition among individuals. They have been associated with diverse phenotypes and diseases such as thyroid function, cognition, cleft palate, schizophrenia, childhood obesity, and autism spectrum disorder. Waterland’s team had conducted an earlier screen on 10 individuals for CoRSIVs.
“Three years ago, we reported nearly 10,000 such regions in the human genome and proposed that studying them could be a novel way to uncover epigenetic causes of disease,” said Waterland.
In the new study, the team assessed CoRSIVs in a larger group to characterize associations among inter-individual genetic, epigenetic, and transcriptional variations, with a view to understanding how DNA methylation at CoRSIVs is affected by genetics. More than 200 studies of human mQTL have been published to date, nearly all using Illumina’s methylation arrays.
“In addition to validating CoRSIVs as systemic epigenetic variants, assessing correlations with gene expression, and characterizing associations with transposable elements, we discovered that CoRSIVs exhibit much stronger mQTL than previously observed,” the authors noted. “Because inter-individual variation is essential not just for mQTL detection but also for epigenetic epidemiology, our results have important implications for the EWAS field.”
For the past 20 years, genome-wide association studies (GWAS) have identified genetic variations that reflect a range of phenotypes and diseases. However, DNA variants discovered to date still fail to explain most human diseases and phenotypes. This has spurred interest in the epigenome, as epigenetic differences constitute another layer of stable molecular information associated with the genome during development. The epigenome determines which regions of the genome are turned on or off in specific cells.
“Many people know that each person has a unique DNA sequence or genome. Less well known is that every cell in the body likewise has a unique level of molecular individuality called its epigenome,” said Waterland. “Epigenetic differences between people can affect their risk of diseases.”
Over the past decade, more than 1,000 published EWAS studies have searched for links between epigenetic modifications (such as DNA methylation) and disease. The new BCM study suggests that the tool that has been the workhorse for most of these studies—Illumina’s methylation platform that assesses hundreds of thousands of CpG sites throughout the genome—may not be ideal for population-level epigenetics.
Waterland’s team focused on CpG sites, where DNA methylation differs significantly among individuals but is consistent across the different tissues of each person (CoRSIVs). They reasoned that these sites would be most useful for population-level studies because DNA isolated from blood samples can be used to assess epigenetic causes of disease in internal organs. The researchers then compared methylation at CoRSIVs with those detected in the largest EWAS study conducted earlier.
“What we found was somewhat of a shock,” said Chathura Gunasekara, PhD, a data analyst in Waterland’s lab and first author of the study. “Compared to the most powerful previous study including 33,000 people, our much smaller study that focused on CoRSIVs discovered 72 times more mQTL.”
The team discovered that nearly 95% CpG sites assessed on commercial arrays do not show appreciable methylation differences among individuals and therefore cannot detect mQTL. The Illumina methylation platform assesses a stable subset of CpG sites in the human genome. This results in one quantitative value for each site and simplifies data sharing and integration across multiple studies and populations. However, in the context of population epigenetics, this is a major shortcoming, because most CpGs included do not show appreciable differences among individuals, a parameter called population variance.
“Population variance is essential not only for mQTL detection, but also for detecting associations between DNA methylation and risk of disease,” said Cristian Coarfa, PhD, associate professor of molecular and cellular biology at BCM and co-senior author of the study. “Compared to what the field has been doing, we anticipate that focusing on CoRSIVs will make epigenome-wide association studies about 70 times more powerful.”
The authors suggest that improving the coverage of CoRSIVs would enhance the utility of Illumina’s array for studies in population epigenetics.
