Study Points to Methylation-based ‘Clock’ for Determining Tissue Age

Posted on Oct 21, 2013

Study Points to Methylation-based ‘Clock’ for Determining Tissue Age

NEW YORK (GenomeWeb News) – Methylation marks across the human genome may make up an “epigenetic clock” for gauging the chronological age of various tissues in the human body, according to a study published online last night in Genome Biology.

A University of California at Los Angeles researcher used a computational method to fish out more than 350 age-informative cytosine methylation markers from thousands of healthy samples collected across the human lifespan and profiled using microarrays for prior studies.

Findings from the study suggest the resulting methylation-based aging predictor can accurately determine chronological age across multiple tissue types. The work also offered insights into how tissues age with time and revealed differences in aging profiles between tissues and in tumor samples. More work is needed to untangle the nature of the relationship between age and the methylation profiles described in the study.

“The big question is whether the biological clock controls a process that leads to aging,” the study’s author, Steve Horvath, a human genetics and biostatistics researcher affiliated with UCLA’s David Geffen School of Medicine and the UCLA Fielding School of Public Health, said in a statement.

“If so, the clock will become an important biomarker for studying new therapeutic approaches to keeping us young,” he added.

In an effort to explore previously proposed ties between aging and epigenetics, Horvath brought together cytosine methylation profiles that had been ascertained for 7,844 samples using Illumina 27K or 450K arrays.

The sample represented 51 non-cancerous human tissue and/or cell types and came from 82 different datasets, Horvath noted.

With a training set that included 39 of the datasets, he used a so-called elastic net regression model to whittle down to a set of 352 cytosine methylation marks that appeared promising for predicting the age of multiple tissues.

This methylation-based epigenetic clock was subsequently validated in healthy samples from dozens more studies before being used to assess aging patterns at specific cell stages or in tissue types.

After establishing these methylation-based clocks in normal tissues, for instance, Horvath turned his attention to 5,826 cancer samples from 32 different DNA methylation datasets. That arm of the analysis indicated that the ability to predict age using the methylation clock tends to break down somewhat in tumor samples.

Generally speaking, though, cancers have DNA methylation “ages” beyond their years and appear far older than corresponding non-cancerous tissue. That was especially true for tumors containing relatively modest somatic mutation burdens.

It was also the case for some of the breast cancer samples examined in the study, though even normal breast tissue appeared to acquire more aged methylation marks than other tissues of the same chronological age.

“Healthy breast tissue is about two to three years older than the rest of a woman’s body,” Horvath noted. “If a woman has breast cancer, the healthy tissue next to the tumor is an average of 12 years older than the rest of her body.”

Breast tissue was one of the tissue types that showed less precise calibration on the DNA methylation-based aging clock, Horvath noted, particularly in the cancerous cases. But for samples from women without cancer, the epigenetic clock came up with age estimates that were within around seven-and-a-half years of individuals’ chronological age, on average.

Induced pluripotent stem cells had “young” or “reset” epigenetic clocks, according to the study, reverting to methylation patterns reminiscent of those found in embryonic stem cells.

At the other extreme, Horvath saw that methylation marks did not seem to coincide with age in samples from individuals with premature-aging conditions such as Werner syndrome or Hutchinson-Gilford progeria.

Based on findings so far, Horvath argued that “[t]his novel epigenetic clock can be used to address a host of questions in developmental biology, cancer, and aging research.” UCLA has reportedly filed for a provisional patent related to the work presented in the Genome Biology study.

Going forward, Horvath hopes to delve into the methylation clock’s role in aging, if any, looking at the aging and/or cancer risk consequences of interfering with cells’ ability to introduce age-related methylation changes.

Future follow-up work may also include studies aimed at determining whether a similar system exists in mice. Results so far suggest the DNA methylation method can be used to determine the age of chimpanzee tissues.

READ MORE