Introducing FOXM1 into cultured skin cells from older adults improved DNA repair and reduced the expression of genes involved in inflammation and cellular senescence
Introducing FOXM1 into cultured skin cells from older adults improved DNA repair and reduced the expression of genes involved in inflammation and cellular senescence
On November 16, Aging Cell published a study by Elsa Logarinho's group at the University of Porto. The authors traced a sequence of events in skin cells from donors of different ages: as FOXM1, a regulator of DNA repair genes, becomes less active with age, DNA breaks accumulate and cells activate a senescence program that includes inflammatory signaling.
The authors consider DNA damage a major cause of aging, but improving repair is difficult because it involves many pathways. They suggest that this probably requires a “master regulator of DNA repair capable of coordinating multiple repair factors simultaneously.” They chose FOXM1 over two other regulators of DNA repair genes: excess E2F1 triggers senescence in human cells, while suppressing E2F4 could disrupt cell division control. Logarinho's group showed that FOXM1 declines with age in human skin cells in 2018. Then, in 2022, they periodically activated an introduced gene encoding an active form of FOXM1 in mice, extending lifespan in both mice with progeria (a disease of accelerated aging) and old mice. The authors say the new study helps explain the mechanism behind this effect.
The proportion of nuclei containing protein foci that mark DNA double-strand breaks increased with donor age. Suppressing FOXM1 in young cells increased this proportion and slowed repair: after treatment with etoposide, a drug that causes DNA breaks, the foci disappeared 1,7 times more slowly. Introducing FOXM1 into old cells reduced the proportion of these nuclei and accelerated repair by a factor of 1,3. Expression of 63 DNA repair genes fell when FOXM1 was suppressed and rose when it was introduced. The authors identify KIF2C, a protein that helps repair DNA breaks, as a FOXM1 target: the same group used a compound to enhance KIF2C activity and extended median lifespan in mice with progeria from 15,7 to 17,1 weeks.
A 2012 study described a link between DNA breaks and DNA packaging: in senescent cells, DNA damage signaling causes the enzyme G9a to be degraded. G9a adds the H3K9me2 mark to histones, the proteins that package DNA. This mark is characteristic of inactive DNA regions near the nuclear envelope. Cells from older donors had less G9a and less H3K9me2, and the mark became sparser near the nuclear envelope with age. Introducing FOXM1 partially restored both the enzyme and the mark. In the authors' model, repair reduces DNA breaks, allowing G9a to escape degradation. A G9a inhibitor weakened the restoration of the mark, indicating that the enzyme is responsible for restoring it.
Mapping H3K9me2 showed that suppressing or introducing FOXM1 altered the mark at 13–14 thousand DNA regions, most of which resembled enhancers (DNA sequences that increase the activity of nearby genes). Previous studies in mouse cells found that enhancers carrying this mark are inactive. Regions that lost the mark were enriched for binding sites for the protein AP-1. According to a 2020 study, AP-1 is the first to bind enhancers in senescent cells (cells that stop dividing and release inflammatory signals). The authors propose that the loss of H3K9me2 makes it easier for AP-1 to bind these enhancers. Expression of gene sets involved in inflammation and senescence increased when FOXM1 was suppressed and decreased when it was introduced. At a cluster of enhancers for inflammatory CXCL genes, suppressing FOXM1 reduced H3K9me2 and increased a mark of enhancer activity; introducing FOXM1 produced the opposite changes.
The authors call H3K9me2 a “gatekeeper of the senescence program.” To test this role, they treated young cells with a G9a inhibitor for three days. The mark decreased, and signs of senescence appeared: levels of p16 and p21, proteins that halt cell division, increased.
FOXM1 partially restored the mark in old cells that had not become fully senescent. Cells that had become fully senescent through telomere attrition (shortening of the chromosome ends) or treatment with nutlin-3a did not respond to FOXM1: both H3K9me2 and G9a remained low.