New Epigenetic Clock Can Profile the Biological Age of Single Cells

3D rendering of single human cell concept

Over the past decade, “epigenetic clocks” have revolutionized the study of aging. These machine learning models, based on DNA methylation data, can help evaluate the age of any biological tissue across the entire life course.

Epigenetic clocks are exciting because they capture progression through aging and reveal cases of accelerated and decelerated aging. Epigenetic clocks could also identify molecular targets of interventions such as treatments that slow down the aging process or support rejuvenation during cell reprogramming.

However, one problem is that methylation profiles of single cells are sparse and binary. Thus, until recently, all existing epigenetic clocks relied on average measurements derived from samples containing multiple cells. This necessarily obscured any differences that may exist among individual cells.

Vadim N. Gladyshev, PhD, director of the Center for Redox Medicine at Brigham and Women’s Hospital, Alexandre Trapp, of the Division of Genetics, and a colleague have developed scAge, an epigenetic clock that allows predictions of biological age at the single-cell level. In Nature Aging, they describe validating it in mice and explain the implications of scAge for deeper research into aging.

How scAge Works

Methylation with age occurs at specific DNA sequences involving CG dinucleotides, called CpG sites. In any given cell, bulk sequencing methods currently assay only a small fraction of CpGs at one time.

scAge is unaffected by which CpGs are used to calculate the age of each cell. First, an algorithm selects age-related CpGs that are in both a cell and a reference dataset. Next, it computes the range of ages for that cell. It assigns the most likely age as the epigenetic age of the cell.

This method pinpoints the chronological age of tissues while also uncovering the epigenetic heterogeneity of individual cells.

Proof of Concept

In murine studies, scAge revealed that individual cells within organisms do age—at different rates. Specific intriguing findings were that:

  • scAge predicted age accurately for single hepatocytes and embryonic fibroblasts whether it was trained on a liver reference dataset or a multi-tissue reference dataset
  • For muscle stem cells, the epigenetic age was lower compared with the chronological age, as has been found in previous studies

The “Ground Zero” Hypothesis of Aging

Recent research suggests that cells may both age and get rejuvenated. The most notable example of cell rejuvenation is the conversion of somatic cells to induced pluripotent stem cells. Dr. Gladyshev recently proposed in a Trends in Molecular Medicine article that early embryogenesis may also be associated with the decrease in biological age. He described a model of “ground zero”: the mid-embryonic state characterized by the lowest biological age where both organismal life and aging begin. In a separate Science Advances study, his lab then demonstrated, by applying bulk epigenetic clocks, that the biological age is indeed decreased during early embryogenesis, followed by an increase.

scAge might help uncover the mechanisms underlying these processes, which is already of interest to scientists investigating cellular rejuvenation therapy (“aging reversal”). Indeed, scAge analyses revealed that, during embryonic development (specifically, at a stage called gastrulation), certain cells, which give rise to the embryo, showed a strongly significant decrease in epigenetic age—that is, they were rejuvenated. On the other hand, cells from extraembryonic tissues did not show this effect.

A New Direction for Research

Being the first single-cell aging clock, scAge should have other profound applications for somatic, germline and cancer cells, as it may permit mapping of “young” and “old” cells within heterogeneous tissues.

Leave a Reply