Archives
We next sought to determine the kinase responsible for IR
We next sought to determine the kinase responsible for IR-induced phosphorylation of 53BP1. Because the sites under investigation all lie in a consensus sequence for ATM, ATR and DNA-PK, that are all activated by IR, the involvement of each of these kinases was investigated. Preincubation of leukotriene receptor antagonists with the NU7441, a specific inhibitor of DNA-PK [29] had no effect on IR-induced phosphorylation of 53BP1 (data not shown). There are no specific inhibitors of ATR currently available. However, somatic cells have been made in which one allele of ATR is disrupted and the remaining allele is flanked by flox recombination sequences and can therefore be removed by viral transduction of the CRE recombinase [30]. Ablation of ATR in this manner (Fig. 4B) had no effect on IR-induced phoshorylation of 53BP1 (data not shown). In contrast, the KU55933, a specific inhibitor of ATM [31] severely reduced phosphorylation 53BP1 at Thr302, Ser831, Ser166, Ser176/Ser178 and Ser452 (Fig. 3B) and similar results were obtained in cells lacking ATM, but not in cells lacking DNA-PK (data not shown). As reported previously, IR-induced phosphorylation of p53 at Ser15 and, to a lesser extent, phosphorylation of SMC1 at Ser966 were inhibited by KU55933 (Fig. 3B). Therefore, ATM phosphorylates the novel 53BP1 phosphorylation sites identified in this study, in response to double-strand breaks. Most studies on 53BP1 function concentrate on its role in responding to DSBs and little data has been presented to implicate 53BP1 in cellular responses to other types of DNA lesion. 53BP1 forms nuclear foci in human cells in response to IR but not in response to UV or replication stress [23]. This is consistent with the notion that 53BP1 responds specifically to DBSs. We examined the effect of UV-irradiation of 53BP1 phosphorylation. Surprisingly, 53BP1 became phoshorylated rapidly at Thr302, Ser831, Ser166, Ser176/Ser178 and Ser452 in response to UV light (Fig. 4A). UV-induced phosphorylation of 53BP1 was apparent 15min post-irradiation and increased over time, reaching a maximum at approximately 60min. Similar results were obtained in U2OS, HCT116 cells and in HEK293 cells (data not shown). Although ATM is responsible for IR-induced phosphorylation of 53BP1 in response to DSBs, neither ATM nor DNA-PK is activated by UV light and so these kinases are unlikely to mediate UV-induced phoshorylation of 53BP1. Consistent with this, preincubation of cells with KU55933 (Fig. 4A) or with NU7441 (data not shown) had no effect on UV-induced phosphorylation of Thr302, Ser831, Ser166, Ser176/Ser178 and Ser452. Because ATR is activated by UV light, the involvement of this kinase in regulation of 53BP1 by UV was investigated. HCT116 ATR−/flox, or HCT116 parent cells, were infected with the CRE recombinase for 36h to maximally deplete ATR (Fig. 4B; [30]). Cells were then exposed to UV light and allowed to recover. As shown in Fig. 4B, no phosphorylation of 53BP1 was observed in cells lacking ATR. Infection of HCT116 parent cells with CRE had no effect on UV-induced phosphorylation of 53BP1. In addition, phosphorylation of 53BP1 in ATR−/flox cells that were not infected with CRE was similar to that observed in wild-type cells (data not shown). These results indicate that, surprisingly, ATR regulates 53BP1 and suggest that 53BP1 may play a role in responses to UV light-induced DNA damage. In summary, we have identified several novel DNA damage-induced sites of phosphorylation in 53BP1 by a combination of mass spectrometric methods and bioinformatics analysis of conserved S/T–Q motifs. Phosphorylation of these sites was subsequently studied with phospho-specific antibodies; this revealed that IR-induced phosphorylation of 53BP1 at these new sites is catalysed by ATM. Surprisingly, 53BP1 was phosphorylated in response to UV damage and this did not require ATM but was dependent on ATR instead. This raises the possibility that 53BP1 is involved in responding to UV-induced DNA damage and this will be interesting to investigate. At present, the functional consequences of DNA damage-induced phosphorylation of the novel sites in 53BP1 described above are not clear; this is compounded by the fact that the function of the region that these residues lie in – that is, outwith the conserved Tudor and BRCT domains – is unclear. Almost all of the 53BP1 phosphorylation sites identified in this study are highly conserved between species and are likely to modulate 53BP1 function. Several of these new sites lie close together, for example Ser166 and Ser176/178 lie in a small patch of 15 residues of almost complete sequence identity. It will be interesting to test the function of this region of 53BP1. It was reported previously that ATM-phosphorylated 53BP1 interacts with hPTIP after treatment of cells with IR [32], [33]. However, mutation of the novel sites identified in this study, singly or in combination, did not affect the DNA damage-inducible interaction of hPTIP and 53BP1 (data not shown). It will be interesting to examine, however, whether mutation of these sites affects the ability of 53BP1 to complement the DNA damage signalling and DNA repair defects seen in cells from 53BP1−/− mice, for example, and to search for proteins that can interact with these phosphorylated residues. Interestingly, the Chen laboratory recently reported that mutation of all 15 conserved S/T–Q motifs in 53BP1 to alanine was unable to rescue the increase in γ-H2AX foci seen in 53BP1 null MEFs, whereas wild-type 53BP1 efficiently rescued this increase [34]. However, these researchers did not test whether that any of these 15 residues were phosphorylated. In this study, we showed that at least some of these residues are phosphorylated after DNA damage. Although it is possible that 15 mutations in one protein could affect the conformation of the protein in a non-specific manner, these results could mean that phosphorylation of one or more of these sites, several of which were shown to be phosphorylated after DNA damage in this study, are important for 53BP1 function.