Cells have many mechanisms for faithfully repairing DNA after damage, but what about restoring the epigenetic marks that are also crucial for proper function? New research on nucleotide excision repair (NER) after UV damage shows that new histones are incorporated at the damage site instead of recycled histones that would already bear the correct modifying marks. This might have implications for epigenetic stability.

Almouzni and colleagues developed an assay for the incorporation of histone variant H3.1 at sites of UV damage, which involves visualizing local concentrations of epitope-tagged histone in transiently transfected cells that are subject to local UV irradiation. Even at low UV doses they found that new histones were recruited to damage sites about half an hour after irradiation. This effect was not limited to S phase, which would imply that the histones were being incorporated at replication forks, and was also not found in NER-deficient cells.

Having shown that new H3.1 is incorporated at UV-damaged sites in association with NER, the authors investigated whether the histone chaperone chromatin assembly factor 1 (CAF1), which associates with H3.1, was involved. Using RNAi, they showed that CAF1 is necessary after the repair process, which is consistent with a direct role in the incorporation of new histones.

These results raise many questions about the maintenance of epigenetic marks in the face of DNA damage. Are epigenetic marks restored after damage, and, if so, how? Neighbouring regions of chromatin could function as a template to guide the correct histone-modifying enzymes to the new unmarked histones. As the results do not discount some recycling of old histones, an alternative is semi-conservative replication, analogous to that of DNA, in which new histones are used in equal proportion to old histones, from which they obtain their correct marks.

On the other hand, it is possible that the histone modifications are not restored. Instead, the presence of new histones could form a memory of the damage. This could serve an adaptive role in the recovery process or, instead, be maladaptive, and help to explain the occurrence of radiation-induced genomic instability in the progeny of the damaged cells.

More mechanistic studies are now needed to distinguish between these possibilities and explain how epigenetics interacts with damage repair. The experiments also need to be extended to other histone variants to test their generality.