Tagging of DNA-damage-associated proteins by ubiquitin is key to coordinating the DNA-damage response. The ubiquitin-related protein SUMO is revealed as a crucial regulator of ubiquitylation in DNA repair.
Ubiquitylation — the attachment of ubiquitin groups to cellular proteins — was initially characterized by its role in promoting protein destruction. However, we now know that the consequences of ubiquitylation are diverse, and that it affects many cellular systems. The ubiquitin modification comes in many flavours (addition of a single ubiquitin molecule, for example, or of polyubiquitin chains that differ in the position of the linkage between ubiquitin molecules), and the various types of ubiquitylation can alter the fate of target proteins in different ways. In addition, the cell has ubiquitin-related modifiers, such as the SUMO proteins, that also alter protein fate or function after conjugation1. One process that has been inextricably linked to ubiquitylation is the cellular response to DNA damage. Although studies2,3 had suggested a link between the DNA-damage response and the SUMO pathway, proof that SUMOylation is important for DNA repair had remained elusive. In this issue, two groups, Morris et al.4 (page 886) and Galanty et al.5 (page 935), now provide good evidence that SUMO functions together with ubiquitin to coordinate DNA repair.
DNA double-strand breaks (DSBs) result in the recruitment and activation of the protein kinases ATM, ATR and DNA-PK, which phosphorylate target proteins, such as the variant histone H2AX. The phosphorylated proteins then promote the recruitment of other DNA-repair proteins to DSBs6, including MDC1 (mediator of the DNA-damage checkpoint), 53BP1 and the E3 ubiquitin ligases RNF8, RNF168 and BRCA1 (ref. 6), which catalyse ubiquitylation events7 at DSBs. (Conjugation of ubiquitin or related modifiers to target proteins requires an E1 activating enzyme, an E2 conjugating enzyme and an E3 ligase.)
To investigate the involvement of the SUMO pathway in the DNA-damage response, Morris et al.4 and Galanty et al.5 analysed the subcellular localization of SUMO-pathway components in mammalian cells. Both groups4,5 report that the E1 SUMO-activating enzyme SAE1, the E2 SUMO-conjugating enzyme UBC9, and the three forms of vertebrate SUMO protein, SUMO1 and the closely related SUMO2 and SUMO3 (SUMO2/3), are recruited to DSBs.
The authors4,5 used RNA interference and fluorescence microscopy to show that the SUMO E3 ligases PIAS1 and PIAS4 are responsible for SUMOylation events at DSBs. Depletion of PIAS1 impaired accumulation of SUMO2 and SUMO3 (but not SUMO1) at DSBs, whereas depletion of PIAS4 impaired recruitment of SUMO1 and SUMO2/3. Furthermore, recruitment of 53BP1 to DSBs depended on PIAS4, whereas recruitment of BRCA1 depended on both PIAS1 and PIAS4. Is SUMOylation necessary for DSB repair? The answer is, emphatically, yes — cells lacking PIAS1 or PIAS4 showed defects in DSB repair and were also highly sensitive to DSBs caused by ionizing radiation.
What are the targets of the SUMO pathway during the DNA-damage response? Prompted by a study showing interaction between UBC9 and BRCA1 in the nematode worm Caenorhabditis elegans2, both groups4,5 independently showed that BRCA1 is SUMOylated during the DNA-damage response in a PIAS1- and PIAS4-dependent manner (Fig. 1). Depletion of PIAS1 and PIAS4 impaired recruitment of BRCA1 to DSBs4,5, significantly impaired ubiquitylation at DSBs, and reduced ubiquitylation of the histones H2A and H2AX; the latter process has been shown to require the ligase activities of RNF8, RNF168 and BRCA1 (ref. 7). Galanty et al.5 also showed that 53BP1 is SUMOylated and that this affects its retention at DSBs.
RNF8 and RNF168 catalyse the formation of lysine-63-linked ubiquitin chains, whereas BRCA1 catalyses formation of lysine-6-linked ubiquitin chains8 (Fig. 1). Morris et al.4 exploited this difference in ubiquitin-chain linkage to pinpoint the effects of PIAS proteins on BRCA1 activity. They showed that over-expression of BRCA1 increased ubiquitylation events in cells; these events were reduced following PIAS1/4 depletion. Co-localization of lysine-6-linked ubiquitin chains with DSBs was also impaired in BRCA1-, PIAS1- or PIAS4-depleted cells. Furthermore, mutation of the two consensus SUMO-conjugation sites in BRCA1 reduced SUMO1 association and BRCA1-dependent ubiquitylation. Thus, the authors propose that BRCA1 is a SUMO-regulated ubiquitin ligase (Fig. 1).
These findings raise several immediate questions. Are the activities of RNF8, RNF168 and/or other E3 ubiquitin ligases also regulated by SUMOylation? Certainly, such a scenario is possible for RNF8. The current studies4,5 found that, although recruitment of RNF8 to DSBs was unaffected by PIAS1/4 depletion, RNF8 could not ubiquitylate DSBs, suggesting that it may be inactive in the absence of PIAS1/4. How does SUMOylation stimulate the E3 ubiquitin-ligase activity of BRCA1? Previous studies9 have shown that DNA damage promotes association between BRCA1 and its E2 conjugating enzyme to form an active E3 ubiquitin ligase. It is therefore tempting to speculate that SUMOylation induces a conformational change in BRCA1 that enhances its binding to an E2 conjugating enzyme. It is clear from the current studies that SUMOylation functions at multiple levels during the DNA-damage response and this will provide fertile ground for future research. The discovery that the SUMO pathway is important for ubiquitylation at DSBs raises the possibility that SUMOylation may activate other ubiquitylation events in the cell.
Welchman, R. L., Gordon, C. & Mayer, R. J. Nature Rev. Mol. Cell Biol. 6, 599–609 (2005).
Boulton, S. J. et al. Curr. Biol. 14, 33–39 (2004).
Golebiowski, F. et al. Sci. Signal. 2, ra24 (2009).
Morris, J. R. et al. Nature 462, 886–890 (2009).
Galanty, Y. et al. Nature 462, 935–939 (2009).
Harper, J. W. & Elledge, S. J. T. Mol. Cell 28, 739–745 (2007).
Panier, S. & Durocher, D. DNA Repair 8, 436–443 (2009).
Morris, J. R. & Solomon, E. Hum. Mol. Genet. 13, 807–817 (2004).
Polanowska, J., Martin, J. S., Garcia-Muse, T., Petalcorin, M. I. & Boulton, S. J. EMBO J. 25, 2178–2188 (2006).
About this article
Increased cellular accumulation and distribution of amrubicin contribute to its activity in anthracycline-resistant cancer cells
Cancer Chemotherapy and Pharmacology (2012)