Human skin cells lost some distinctive features of their DNA methylation patterns during prolonged growth in culture: preprint

Human skin cells lost some distinctive features of their DNA methylation patterns during prolonged growth in culture: preprint

Researchers at Tel Aviv University and Reichman University reanalyzed publicly available data on fibroblasts, connective tissue cells found in the skin. In a preprint posted on November 3, they compared two routes to the point at which cells stop dividing: months of division and DNA damage caused by X-rays.

A cell's type depends in part on which genes are active. Gene activity is associated with the methylation pattern, the distribution of chemical marks on DNA. One hypothesis of aging proposes that repairing DNA breaks draws proteins that regulate gene activity away from their usual locations. Errors occur when the previous regulatory state is restored after repair, and the cell gradually loses features characteristic of its type. David Sinclair's group tested this model in mice by repeatedly inducing DNA breaks.

The authors of the new analysis had previously observed that distinctive methylation patterns became less pronounced in diseased tissues. In the liver, this change increased with fibrosis, the buildup of scar tissue. However, tissue damage also involves inflammation, cell death, and division of surviving cells. To distinguish these processes, the researchers selected cultures of a single cell type exposed to different conditions.

Each of the 24 sets of cultures included a control group of cells without the exposure being studied. Some cells divided for months until they exhausted their capacity to divide; others stopped dividing soon after irradiation. This shared endpoint allowed the researchers to compare how methylation patterns changed along two routes to the same state.

They first selected DNA regions whose methylation distinguished fibroblasts from twenty other cell types. During prolonged growth, the overall difference across these regions increased. Most of the selected features turned out to be shared by cells of other types capable of continuous growth in the laboratory.

The authors interpreted these shared features as a mark of cell division. For a separate analysis, they selected several hundred regions whose methylation patterns were specific to fibroblasts. During prolonged growth, the two groups of regions changed in opposite directions: differences from other cell types increased in the larger group of shared regions and diminished in the selected group. The combined measure therefore concealed the loss of distinctive fibroblast features.

The researchers calculated how far the methylation patterns at the selected regions had shifted toward the average pattern of other cell types. Regions that initially distinguished fibroblasts more strongly contributed more to the result. For comparison, they used other DNA regions with the same initial frequency of methylation marks. Changes in these comparison regions provided a measure of background change, which they subtracted from the result. After this adjustment, cells that had exhausted their capacity to divide had lost 24,3% of the initial difference between fibroblasts and other cell types at the selected regions. After X-ray exposure, they detected no loss beyond the background change in methylation.

When testing whether a distinctive methylation pattern has been preserved or restored, the authors suggest comparing treated cells with control cells that have undergone the same number of divisions:

An intervention that reduces damage usually also changes how many times cells divide.