NEW YORK (GenomeWeb News) – An individual’s epigenome can shift over time — and the direction and magnitude of that shift seems to be at least partly heritable, new research suggests.
In a paper appearing in today’s issue of the Journal of the American Medical Association, an international team of researchers used two prospective studies to assess changes in hundreds of individuals’ epigenetics — specifically their total DNA methylation — over a decade or more. In each case they found changes in many individuals’ total DNA methylation over time. But the direction and extent of that change varied between individuals while families tended to share methylation trends, suggesting some epigenetic traits are inherited.
Epigenetics, in general, refers to processes that can lead to differential expression of an identical stretch of DNA at different times and/or in different environments. This can be a consequence of processes such as DNA methylation, histone modification, chromatin folding or other changes that modify DNA or DNA-protein interactions.
These epigenetic markers can vary from one individual to the next and from tissue to tissue. It is thought that epigenetics can also reflect environmental exposures and other changes in the body. Even so, just how each person’s epigenome changes with time, if at all, remains poorly understood.
To understand this better, senior author Andrew Feinberg, an oncologist, molecular biologist, and geneticist at Johns Hopkins University, led a team of American, Icelandic, and Swedish researchers studying two groups of people: one in Iceland and another in the United States.
In the first study, the team focused on 111 individuals between the ages of 70 and 82 years old who were participating in Iceland’s Age, Gene/Environment Susceptibility, or AGES-Reykjavik, Study, an offshoot of the Reykjavik Heart Study. This included 61 participants for whom large DNA samples were available, along with 50 participants who had survived various types of cancer.
The researchers isolated DNA from blood samples taken from each individual in 1991 and again more than a decade later (between 2002 and 2005). They then compared each person’s total DNA methylation at each sample time using a luminometric methylation assay.
Some 63 percent of participants had a five percent or greater change in methylation between sampling times. About half of these had increased methylation over time, while the total methylation decreased in the other half. Roughly 30 percent had a ten percent or greater overall change in methylation. Eight percent showed larger changes of 20 percent or more.
In an effort to verify these methylation changes — and look at whether epigenetic patterns are heritable — the researchers assessed total methylation in samples from 126 individuals between the ages of five and 72 years old from 21 families in Utah. They sampled the individuals first between 1982 and 1985 and again after an average of 16 years, between 1997 and 2005.
The team also assessed the methylation of 807 genes at 1505 CpG loci in a subset of 41 individuals who showed the greatest loss or gain in DNA methylation using the Illumina GoldenGate Methylation Solution.
Again, methylation patterns varied from one individual to the next. But there were also distinct patterns within families. Related individuals tended to cluster based on whether total methylation went up or down over time. In some cases the extent of total methylation change was similar between members of the same family.
In the most extreme example, the researchers found five related individuals with methylation that decreased by more than 30 percent, particularly for genes related to immunological functions.
The results are surprising given previous research by others indicating that DNA methylation levels are stable over time. For their part, Feinberg and his co-workers argued that previous analyses have overlooked methylation changes by averaging methylation across individuals, since some individuals’ methylation increases while others’ decreases.
“[O]ur data suggest considerable inter-individual age variation, with differences accruing over time within individuals that would be missed by group averaging,” the authors wrote.
Because the inappropriate loss or gain of methylation could lead to problems in the cell, including cancer, the researchers noted, more work is necessary to understand the consequences of familial and individual methylation changes, if any.
“We still haven’t concretely figured out what this means for health and disease, but as an epidemiologist, I think this is very interesting, since epigenetic changes could be an important link between environment, aging, and genetic risk for disease,” Johns Hopkins researcher Daniele Fallin, an author on the paper, said in a statement.
“Inappropriate methylation levels can contribute to disease — too much might turn necessary genes off, too little might turn genes on at the wrong time or in the wrong cell,” author Vilmundur Gudnason, a cardiovascular researcher at the University of Iceland and director of the Icelandic Association’s Heart Preventive Clinic and Research Institute, added in a statement.