Summary Genetic information encoded in chromosomal DNA is challenged by intrinsic and exogenous sources of DNA damage. DNA double-strand breaks (DSBs) are extremely dangerous DNA lesions. RAD51 plays a central role in homologous DSB repair, by facilitating the recombination of damaged DNA with intact DNA in eukaryotes. RAD51 accumulates at sites containing DNA damage to form nuclear foci. However, the mechanism of RAD51 accumulation at sites of DNA damage is still unclear. Post-translational modifications of proteins, such as phosphorylation, acetylation and ubiquitylation play a role in the regulation of protein localization and dynamics. Recently, the covalent binding of small ubiquitin-like modifier (SUMO) proteins to target proteins, termed SUMOylation, at sites containing DNA damage has been shown to play a role in the regulation of the DNA-damage response. Here, we show that the SUMOylation E2 ligase UBC9, and E3 ligases PIAS1 and PIAS4, are required for RAD51 accretion at sites containing DNA damage in human cells. Moreover, we identified a SUMO-interacting motif (SIM) in RAD51, which is necessary for accumulation of RAD51 at sites of DNA damage. These findings suggest that the SUMO–SIM system plays an important role in DNA repair, through the regulation of RAD51 dynamics.

Summary In response to UV light, single-stranded DNA intermediates coated with replication protein A (RPA) are generated, which trigger the ATR-Chk1 checkpoint pathway. Recruitment and/or activation of several checkpoint proteins at the damaged sites is important for the subsequent cell cycle arrest. Surprisingly, upon UV irradiation, Rad9 and RPA only minimally accumulate at DNA lesions in G2 phase, suggesting that only a few single-stranded DNA intermediates are generated. Also, little phosphorylated Chk1 is observed in G2 phase after UV-irradiation, and UV light fails to elicit efficient accumulation of typical DNA damage response proteins at sites of damage in this phase. By contrast, p38 MAPK is phosphorylated in G2 phase cells after UV damage. Interestingly, despite the lack of an obvious activation of the ATR-Chk1 pathway, only the combined inhibition of the ATR- and p38-dependent pathways results in a complete abrogation of the UV-induced G2/M arrest. This suggests that UV light induces less hazardous lesions in G2 phase or that lesions created in this phase are less efficiently processed, resulting in a low activation of the ATR-Chk1 pathway. UV-induced G2 checkpoint activation in this situation therefore relies on signalling via the p38 MAPK and ATR-Chk1 signalling cascades.

The cell cycle checkpoint kinase Chk1 is phosphorylated and activated by ATR in response to DNA damage and is crucial for initiating the DNA damage response. A number of factors act in concert with ATR to facilitate Chk1 phosphorylation, including Rad17-RFC, the Rad9-Rad1-Hus1 complex, TopBP1 and Claspin. Rad17 is required for loading of Rad9-Rad1-Hus1 (9-1-1) onto sites of DNA damage. Although phosphorylation of Rad17 by ATR is required for checkpoint function, how this affects 9-1-1 regulation remains unclear. We report that exposure of cells to DNA damage or replication stress results in Rad17-dependent immobilisation of Rad9 into nuclear foci. Furthermore, expression of mutant Rad17 that cannot be phosphorylated by ATR (Rad17 AA ), or downregulation of ATR, results in a decreased number of cells that display Rad9 foci. Photobleaching experiments reveal an increase in the dynamic behaviour of Rad9 within remaining foci in the absence of ATR or following expression of Rad17 AA . Together, these data suggest a model in which Rad17 and ATR collaborate in regulating Rad9 localisation and association at sites of DNA damage.

Live cell studies of DNA repair mechanisms are greatly enhanced by new developments in real-time visualization of repair factors in living cells. Combined with recent advances in local sub-nuclear DNA damage induction procedures these methods have yielded detailed information on the dynamics of damage recognition and repair. Here we analyze and discuss the various types of DNA damage induced in cells by three different local damage induction methods: pulsed 800 nm laser irradiation, Hoechst 33342 treatment combined with 405 nm laser irradiation and UV-C (266 nm) laser irradiation. A wide variety of damage was detected with the first two methods, including pyrimidine dimers and single- and double-strand breaks. However, many aspects of the cellular response to presensitization by Hoechst 33342 and subsequent 405 nm irradiation were aberrant from those to every other DNA damaging method described here or in the literature. Whereas, application of low-dose 266 nm laser irradiation induced only UV-specific DNA photo-lesions allowing the study of the UV-C-induced DNA damage response in a user-defined area in cultured cells.