When DNA is under attack, the repair machinery moves in quickly. There's just one small problem — the damaged DNA is wrapped up as chromatin, which first needs to be unravelled (or 'remodelled'). A sensible solution would be to link the DNA repair machinery with chromatin-remodelling factors, but such a link has been elusive.

Reporting in Molecular Cell, Marc Tini and colleagues now describe the interaction of an enzyme involved in DNA repair — thymine DNA glycosylase (TDG) — with the transcriptional co-activators CBP or p300. Strikingly, the resulting complex is still active in DNA repair and transcription, suggesting that it functions in both processes.

Tini and co-workers detected the CBP–TDG complex by co-immunoprecipitation, GST (glutathione-S-transferase)-fusion-protein interaction assays, and gel filtration of fractionated HeLa cell extracts. They also showed colocalization of the proteins in vivo. CBP alone (tagged with a fluorescent protein) had a granular distribution in the nucleus and TDG alone showed a diffuse nuclear staining, whereas co-expression resulted in a distinctive macrogranular pattern for both proteins.

The authors next studied whether formation of the CBP–TDG complex affects the function of either protein. The complex showed high affinity for a G/T mispaired duplex oligonucleotide — a natural substrate for TDG — and was able to cleave it (an initial step in the repair process). Conversely, CBP retained its chromatin-remodelling (histone acetylase; HAT) activity in the CBP–TDG complex. Indeed, cotransfection with TDG led to a dose-dependent increase in the expression of a GAL reporter gene fused to CBP, suggesting that TDG stimulates the transcriptional activity of CBP.

As CBP can acetylate both histones and non-histone proteins, Tini and colleagues wondered whether TDG might be acetylated by CBP. They found that it is, both in vitro and in vivo, and that acetylation leads to the release of CBP from the complex. Although acetylation did not affect the ability of TDG to bind or cleave mispaired DNA, it did prevent TDG from binding another component of the repair machinery, the apurinic/apyrimidinc endonuclease APE1. This suggests CBP-mediated acetylation could be involved in regulating DNA repair.

This is, say the authors, “the first example of a repair enzyme involved in detection of primary DNA lesions that interacts directly with... CBP/p300 and represents a new class of HAT substrate”. So, by providing a chromatin-modifying activity at sites of G/T repair, the CBP–TDG complex could regulate the access of other components of the repair machinery to the DNA. Finally, these findings imply that mutations in cbp or p300 — which have been found in various tumours — could deregulate TDG-coupled DNA repair and so contribute to the genetic instability that is often associated with cancer.