53BP1 and NFBD1/MDC1-Nbs1 Function in Parallel Interacting Pathways Activating Ataxia-Telangiectasia Mutated (ATM) in Response to DNA Damage

[CANCER RESEARCH 63, 8586–8591, December 15, 2003] Advances in Brief

Tamara A. Mochan,1,2 Monica Venere,1,2 Richard A. DiTullio, Jr.,1,2 and Thanos D. Halazonetis1,3 1The Wistar Institute, Philadelphia, Pennsylvania; 2Cell and Molecular Biology Graduate Group, Biomedical Graduate Studies, and 3Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, Pennsylvania

Abstract this residue with alanine compromises the ability of Nbs1 to facilitate phosphorylation of ATM substrates (15–20). 53BP1 and NFBD1/MDC1 are recruited rapidly to sites of DNA double- Although it is clear that the adaptors/mediators described above strand breaks (DSBs), where they are hypothesized to function down- participate in DNA DSB signaling, the phenotype observed when their stream of the ataxia-telangiectasia mutated (ATM) checkpoint kinase as “mediators” of DNA DSB signaling. To test this hypothesis, we suppressed function is suppressed is also consistent with a function upstream of 53BP1 and NFBD1/MDC1 expression by small interference RNA and ATM because decreased ATM activation will lead to decreased phos- monitored ATM autophosphorylation at Ser1981 as a marker for ATM phorylation of ATM substrates. In fact, Nbs1 was recently shown to activation. Suppression of NFBD1/MDC1 led to decreased ATM activa- function upstream of ATM in addition to its well-established role tion and phosphorylation of ATM substrates. This phenotype was identi- downstream of ATM (21, 22). For 53BP1 and NFBD1/MDC1, there cal to that observed in cells with defective Nbs1 function and is consistent is no experimental evidence to indicate that these proteins function with recent observations identifying NFBD1/MDC1 as a component of the exclusively downstream of ATM, as suggested by the term mediator. Mre11–Rad50–Nbs1 protein complex. In cells with wild-type Nbs1, sup- Here we reevaluate their role in ATM activation, using a recently pression of 53BP1 expression had no effect on ATM activation but was described phosphospecific antibody that can monitor ATM activation associated with increased recruitment of NFBD1/MDC1 and Nbs1 to sites with high sensitivity (23). Our experiments were designed under the of DNA breaks, suggesting that decreased 53BP1 function might be com- assumption that the so-called adaptors/mediators may be components pensated for by increased NFBD1/MDC1 and Nbs1 activity. Indeed, in cells with mutant Nbs1, suppression of 53BP1 led to decreased ATM of independent pathways leading to ATM activation, in which case activation and phosphorylation of ATM substrates. We conclude that inhibition of multiple adaptors/mediators would be required to inhibit DNA DSBs activate ATM through at least two independent pathways ATM activation. involving 53BP1 and NFBD1/MDC1-Nbs1, respectively. Materials and Methods

Introduction Cell Lines. Cell lines used were obtained from the following sources: U2OS and HeLa cells were from American Type Culture Collection (Manas- DNA double-strand break (DSB) signaling in higher eukaryotes is sas, VA); AT5BI was from Coriell (Camden, NJ); normal human dermal mediated by protein kinases, such as ataxia-telangiectasia mutated fibroblasts (NHDFs) were from Clonetics (San Diego, CA); NBS-ILB1 LXIN (ATM) and Chk2, as well as by many other proteins whose functions [Nijmegen breakage syndrome (NBS) cells], NBS-ILB1 NBS1 are less well-defined (1–3). Several proteins in the latter category, (NBSϩnbs1wt), and NBS-ILB1 S343A (NBSϩnbs1A343) were from P. Con- such as 53BP1, NFBD1/MDC1, and BRCA1, contain COOH-termi- cannon (Virginia Mason Research Center, Seattle, WA; Refs. 15, 24); and nal BRCT repeats. The proteins 53BP1 and NFBD1/MDC1 have been XPC-EN12 [xeroderma pigmentosum complementation group C (XPC) cells], referred to as “mediators,” an ill-defined term for proteins that are an XPC human fibroblast cell line immortalized with SV40 and EBNA, was thought to function downstream of ATM in transducing the DNA from E. Friedberg (University of Texas Southwestern Medical School, Dallas, damage signal (4–7). Although their function has not been well TX). defined, 53BP1 and NFBD1/MDC1 may recruit substrates to ATM Cell Transfection. Flag-tagged Chk2 mutants were expressed in U20S cells by transient transfection using calcium phosphate, as described previously (4–14). (25). Sixteen h after transfection, the cells were washed, and fresh medium was Nbs1 is another protein that fits the definition of mediator, because added. Cells were analyzed after an additional 24 h. suppression of its function also compromises DNA DSB signaling, Small Interfering RNA (siRNA) Transfection. All siRNA transfections leading to decreased phosphorylation of ATM substrates (15–18). The were performed with Oligofectamine reagent (Invitrogen, Carlsbad, CA). Cells role of Nbs1 in ATM signaling was described well before the term seeded in 60-mm plates were incubated with mixtures of Oligofectamine and mediator was introduced, and Nbs1 had been referred to as an “adap- 400 pmol control (luciferase) or one of the following siRNAs (Dharmacon, tor,” a term describing its suggested role in recruiting substrates to Lafayette, CO): atm (SMART pool), 53bp1 (GAACGAGGAGACG- ATM. The evidence that Nbs1 functions downstream of ATM is GUAAUAdTdT), nbs1 (GCAGUUCAGUCCAAGAAGCdTdT), or nfbd1/ compelling; ATM phosphorylates Nbs1 on Ser343, and substitution of mdc1 (UCCUGAGACCUCCUAAGGUUUdTdT). Cells were analyzed 72 h after siRNA treatment. Immunofluorescence Analysis. NHDF and XPC cells were seeded in Received 10/26/03; revised 11/4/03; accepted 11/5/03. 100-mm plates. Immunofluorescence microscopy and image acquisition were The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with performed as described previously (26) with antibodies specific for 53BP1 18 U.S.C. Section 1734 solely to indicate this fact. (26), NFBD1/MDC1 (a gift from J. Chen, Mayo Clinic, Rochester, MN; Ref. Grant Support: National Cancer Institute Grant CA76367 (to T. D. H.) and Training 6), or Nbs1 (Calbiochem, La Jolla, CA). Grants CA09171 (to T. A. M.) and CA09677 (to M. V.). Cell Extracts and Immunoblotting. Whole-cell extracts were prepared Requests for reprints: Thanos D. Halazonetis, the Wistar Institute, 3601 Spruce Street, Philadelphia, PA 19104-4268. Fax: (215) 573-9271; E-mail: halazonetis@wistar. and analyzed by immunoblotting as described previously (25) with antibodies upenn.edu. specific for Chk2, Chk2 phosphorylated at Thr68, Chk2 phosphorylated at 8586

Ser33/35 (Cell Signaling Technology, Beverly, MA), ATM (MAT3-mouse MDC1, expression of one of these proteins was suppressed by siRNA ATM #3; a gift from Y. Shiloh, Tel Aviv University, Ramat Aviv, Israel), in NHDFs and recruitment of the other protein to sites of DNA DSBs ATM phosphorylated at Ser1981 (a gift from M. Kastan, St. Jude Children’s was monitored by immunofluorescence. In nonirradiated cells, neither Research Hospital, Memphis, TN; Ref. 23), Nbs1 (Calbiochem, La Jolla, CA), 53BP1 nor NFBD1/MDC1 localized to discreet nuclear foci. After 957 53BP1 (26), SMC1, SMC1 phosphorylated at Ser (Bethyl Laboratories, exposure to relatively high doses of ionizing radiation [5–9Gyof Montgomery, TX), NFBD1/MDC1 (a gift from S. Jackson, University of ionizing radiation (IR)] both proteins colocalized to IR-induced foci Cambridge, Cambridge, UK; Ref. 5), or Flag epitope (M5; Sigma, St. Louis, MO). and suppression of the expression of either of them by siRNA did not affect recruitment of the other protein to the sites of DNA DSBs (data not shown). Exposure of NHDFs to a relatively low dose of IR (1 Gy) Results led to recruitment of 53BP1, but not NFBD1/MDC1, to IR-induced 53BP1 Suppresses NFBD1/MDC1 Recruitment to Sites of DNA foci. At this dose, suppression of 53BP1 expression by siRNA facil- DSBs. One property shared by various adaptors/mediators is recruit- itated recruitment of NFBD1/MDC1 to IR-induced foci, whereas ment to sites of DNA DSBs, a term used to describe not only the DSB suppression of NFBD1/MDC1 expression by siRNA had no effect on lesion itself, but also long stretches of chromatin adjacent to the DSB 53BP1-induced focus formation (Fig. 1A). Similar results were ob- lesion that undergo DNA damage-induced modifications, such as served with fibroblasts derived from an individual with XPC, in which phosphorylation of histone H2AX (27, 28). Interestingly, recruitment suppression of 53BP1 expression led to increased localization of of adaptors/mediators to sites of DNA DSBs is hierarchical. For NFBD1/MDC1 and of Nbs1 to sites of DNA DSBs (Fig. 1B). The example, recruitment of Nbs1 is dependent on NFBD1/MDC1, a increased localization of Nbs1 to sites of DNA DSBs in response to finding consistent with these two proteins associating with each other suppression of 53BP1 expression was accompanied by increased as part of a larger protein complex that also contains Mre11 and phosphorylation at Ser343 (Fig. 1C), suggesting that in the absence of Rad50 (5). Recruitment dependency of one adaptor/mediator on an- 53BP1 a greater pool of Nbs1 participates in ATM signaling and other could indicate that the two adaptors/mediators function in the becomes phosphorylated by ATM. We conclude that 53BP1 and same pathway. We were particularly interested in the recruitment NFBD1/MDC1-Nbs1 are recruited to sites of DNA DSBs independ- dependency between 53BP1 and NFBD1/MDC1 because recruitment ently of each other, although there is some cross-talk because sup- of 53BP1 to sites of DNA DSBs has been reported to be NFBD1/ pression of 53BP1 expression in cells exposed to low doses of IR is MDC1-dependent by one laboratory (7) and NFBD1/MDC1-inde- accompanied by increased recruitment of NFBD1/MDC1 and Nbs1 to pendent by another (5). sites of DNA DSBs. If 53BP1 and NFBD1/MDC1 perform similar To study the recruitment dependency between 53BP1 and NFBD1/ functions in DNA DSB signaling, then their independent recruitment

Fig. 1. Suppression of 53BP1 expression enhances recruitment of NFBD1/MDC1 and Nbs1 to sites of DNA breaks and the ataxia-telangiectasia mutated (ATM)-dependent phosphorylation of Nbs1. A, 53BP1 or NFBD1/MDC1 expression was suppressed by small interfering RNA (siRNA) in normal human dermal fibroblasts (NHDFs). Seventy-two h later the cells were exposed to 1 Gy of ionizing radiation and recruitment of 53BP1 and NFBD1/MDC1 (NFBD1) to sites of DNA double-strand breaks was monitored by immunofluorescence (IF). ctl-i, control siRNA; nfbd1-i, nfbd1/mdc1 siRNA; 53bp1-i, 53bp1 siRNA. B, 53BP1 expression was suppressed by siRNA in fibroblasts from a patient with xeroderma pigmentosum complementation group C (XPC). Recruitment of MDFC1/NFBD1 and Nbs1 to sites of DNA breaks after irradiation was monitored by immunofluorescence. The cells marked 1, 2, and 3 are also shown in higher magnification (bottom) to better visualize the Nbs1 foci. C, 53BP1 expression was suppressed by siRNA in HeLa cells. Phosphorylation of Nbs1 at Ser343 (pS343) after irradiation was monitored by immunoblotting. 8587

Analysis of endogenous Chk2 in HeLa cells revealed phospho-Ser33/35 reactivity when the cells were exposed to relatively high doses (9 Gy) of IR, but no reactivity when the cells were exposed to low-dose (1 Gy) IR or UV light (Fig. 2B). In contrast, phosphorylation of Thr68 was evident after exposure of the cells to 1 or 9 Gy of IR and UV light. Analysis of primary fibroblasts from a patient with ataxia-telangiectasia revealed that phosphorylation of Ser33/35, like phosphorylation of Thr68, is ATM- dependent (Fig. 2C). The ATM-dependence of the Ser33/35 phosphorylation could be further demonstrated by siRNA experiments; suppression of ATM protein expression by siRNA abolished Chk2 Ser33/35 phosphorylation in irradiated NHDFs and HeLa cells (Fig. 2D). Phosphorylation of Thr68 was not suppressed by the ATM-specific siRNA, arguing that small residual levels of ATM are sufficient for phosphorylation of Chk2 Thr68. We conclude that phosphorylation of Chk2 at Ser33/35 may be a more sensitive indicator of decreased ATM activity than Chk2 phosphorylation at Thr68. These results may also explain why Chk2 phosphorylation at Thr68 is evident in cells exposed to both 1 and 9 Gy of IR, whereas phosphorylation at Ser33/35 is evident only in cells exposed to 9 Gy of IR (Fig. 2B). Defective Chk2 Ser33/35 Phosphorylation and ATM Activation in NBS Cells. NFBD1/MDC1 associates with Nbs1 (5), raising the possibility that loss of NFBD1/MDC1 and Nbs1 function may have similar phenotypes. Therefore, as a first step in exploring whether NFBD1/ MDC1 is required for ATM activation, we examined cells from patients with NBS, in which the nbs1 gene is mutated. Previous studies indicate that NBS cells exposed to IR show decreased phosphorylation of ATM substrates. Expression of wild-type Nbs1 reverses these defects, but NBS cells expressing Nbs1, in which Ser343 has been substituted with Ala, still Fig. 2. Ataxia-telangiectasia mutated (ATM)-dependent phosphorylation of Chk2 at show at least a partial defect in phosphorylation of ATM substrates Ser33/35 in response to DNA double-strand breaks. A, specificity of the Chk2 Ser33/35 phosphospecific antibody. U2OS cells were transiently transfected with plasmids express- (15–18). A recent report further indicates that ATM activation is defec- ing FLAG-tagged Chk2 in which Thr26, Ser28, Ser33, Ser35, and Ser50 were substituted tive in NBS cells and that the defect can be reconstituted with wild-type 68 with alanine (5A) or in which Thr was substituted with alanine (A68). Extracts from 343 these cells were prepared 1 h after irradiation and immunoblotted with phosphospecific Nbs1 (21). However, the ability of the Nbs1 Ala mutant to reconstitute antibodies for Ser33/35 and Thr68. The same extracts were also blotted with antibodies that ATM activation was not examined. recognize the FLAG tag. B, phosphorylation of Chk2 at Ser33/35 is evident after exposure Using NBS cells that were infected with a control retrovirus or of cells to high, but not low, doses of ionizing radiation (IR). HeLa cells were exposed to IR (1 or 9 Gy) or UV light (50 J/m2). Extracts were prepared 10 min or 1 h later and infected with retroviruses expressing wild-type Nbs1 or Nbs1, in immunoblotted with phosphospecific antibodies for Chk2 and ATM, as indicated. C, which Ser343 was substituted with alanine, we observed a delay in the 33/35 phosphorylation of Chk2 at Ser is ATM-dependent. Primary fibroblasts from a patient electrophoretic mobility shift of Chk2 after irradiation in the control with ataxia-telangiectasia (AT5BI) and U2OS osteosarcoma cells, as controls, were exposed to IR. Extracts from these cells were immunoblotted with phosphospecific NBS cells and, to a lesser extent, in the NBS cells expressing the antibodies for Chk2, as indicated. D, suppression of ATM protein levels by small Ala343 mutant (Fig. 3A). Chk2 phosphorylation at Thr68 was evident 33/35 68 interfering RNA abrogates phosphorylation of Chk2 at Ser , but not at Thr . Normal 33/35 human dermal fibroblasts (NHDF) and HeLa cells were transfected with control siRNA in all cells, but Chk2 phosphorylation at Ser was absent in the 343 (ctl-i) or siRNA directed against atm (atm-i). Seventy-two h later the cells were exposed control NBS cells and the NBS cells expressing the Ala mutant to IR. Cell extracts prepared 10 min after irradiation were immunoblotted with phos- (Fig. 3A). We also examined phosphorylation of SMC1 at Ser957. phospecific antibodies for Chk2, as indicated. Consistent with previous reports (16, 17), phosphorylation of SMC1 at Ser957 was diminished in the control NBS cells and the NBS cells to sites of DNA DSBs raises the possibility that suppression of the expressing the Ala343 mutant (Fig. 3B). function of one of them may have little effect on ATM signaling. This Interestingly, the effect of loss of Nbs1 function on Chk2 phospho- may be particularly true when 53BP1 expression is suppressed be- rylation at Thr68 and Ser33/35 exactly paralleled the effect of partial cause this is accompanied by increased recruitment of NFBD1/MDC1 loss of ATM function by siRNA in NHDFs and HeLa cells (Fig. 2D). to sites of DNA DSBs. We therefore examined whether ATM activation, as monitored by ATM-Dependent Phosphorylation of Chk2 at Ser33/35. As a first autophosphorylation at Ser1981, was defective in irradiated NBS cells. step in studying whether ATM activation is dependent on 53BP1 The control NBS cells showed almost complete absence of ATM and/or NFBD1/MDC1, we attempted to identify novel phosphoryla- phosphorylation when exposed to 1 Gy of IR and modest ATM tion events mediated by activated ATM. In response to DNA DSBs, phosphorylation when exposed to 9 Gy of IR, whereas the NBS cells 68 ATM phosphorylates Chk2 at Thr (1–3). However, the NH2 termi- reconstituted with wild-type Nbs1 or, interestingly, with the Nbs1 nus of Chk2 contains multiple other potential ATM phosphorylation Ala343 mutant showed robust ATM phosphorylation after exposure to sites. We used a phosphospecific antibody that recognizes Chk2 both doses of IR (Fig. 3C). Consistent with a recent report (21), these phosphorylated at both Ser33 and Ser35 to probe whether these two findings suggest that Nbs1 is required for ATM activation. Further- residues are phosphorylated in response to DNA damage. The phos- more, the Nbs1 Ala343 mutant dissociates the function of Nbs1 re- phospecific antibody used for this purpose had a clearly distinct and quired for ATM activation from its adaptor/mediator function down- nonoverlapping specificity from the previously characterized Chk2 stream of ATM. Thr68 phosphospecific antibody, as revealed by analysis of irradiated Defective Chk2 Ser33/35 Phosphorylation and ATM Activation U2OS cells transiently expressing FLAG-tagged Chk2 mutant pro- after Suppression of NFBD1/MDC1. Because NFBD1/MDC1 is teins (Fig. 2A). thought to function in the same complex as Nbs1, we next sought to 8588

Fig. 3. Nbs1 functions both as an activator and adaptor of ataxia-telangiectasia mutated (ATM). A and B, phosphorylation of Chk2 at Ser33/35 and SMC1 at Ser957 after irradiation requires wild-type Nbs1. NBS cells reconstituted with wild- type Nbs1 (NBS ϩ nbs1wt) or a Nbs1 mutant protein with substitution of Ser343 with alanine (NBS ϩ nbs1A343) and nonreconstituted control cells (NBS) were exposed to ionizing radiation (IR). Extracts prepared at several time points after irradiation were immunoblotted with phosphospecific antibodies for Chk2 and SMC1, as indicated. The results shown in A and B represent independent experiments. C, phosphorylation of ATM at Ser1981 after irradiation requires full-length Nbs1, but not phosphorylation of Nbs1 at Ser343. The NBS cells described above were exposed to IR, and ATM autophosphorylation was monitored by immunoblotting. Immunoblots from two independent experiments are shown.

examine whether it also plays a role in ATM activation. We first type, which is similar to that observed in NBS cells, was also observed suppressed NFBD1/MDC1 expression by siRNA in NHDFs. When when Nbs1 expression was suppressed by siRNA in NHDFs (Fig. 4A). these primary cells were exposed to 9 Gy of IR, suppression of The effect of suppressing NFBD1/MDC1 on ATM autophospho- NFBD1/MDC1 led to diminished phosphorylation of Chk2 at Ser33/35, rylation at Ser1981 was examined in two cell lines in which we could but phosphorylation at Thr68 was unaffected (Fig. 4A). This pheno- achieve very efficient suppression of NFBD1/MDC1 protein levels with siRNA. In HeLa cells exposed to 1 Gy of IR, suppression of NFBD1/MDC1 led to decreased ATM phosphorylation at Ser1981 (Fig. 4B). Similarly, in an immortalized fibroblast cell line derived from a patient with XPC, which we were studying in the context of another research program, suppression of NFBD1/MDC1 led to decreased ATM autophosphorylation at Ser1981 after exposure of the cells to either 1 or 9 Gy of IR (Fig. 4C). Defective Chk2 Phosphorylation and ATM Activation after Suppression of 53BP1 in NBS cells. Like NFBD1/MDC1, 53BP1 has COOH-terminal BRCT repeats and has been implicated in ATM signaling pathways. Suppression of 53BP1 function by siRNA or gene knockout leads to modest cell cycle checkpoint defects and to decreased ATM-dependent substrate phosphorylation at sites of DNA DSBs, as detected by immunofluorescence. However, with the excep- tion of one report, suppression of 53BP1 function has no effect on Chk2 phosphorylation (4, 11–14). We have also seen no effect on Chk2 phosphorylation at Thr68 or even at Ser33/35 when 53BP1 levels were suppressed by siRNA in NHDFs, HeLa, or U2OS osteosarcoma cells (data not shown). One possible explanation for this finding is that 53BP1 and the NFBD1/MDC1–Nbs1 complex have redundant functions. Accordingly, the effects of suppressing 53BP1 function on DNA damage signaling might be stronger in cells with defective NFBD1/MDC1-Nbs1 function. Indeed, in the control NBS cells, suppression of 53BP1 led to decreased phosphorylation of Chk2 at Thr68, decreased phosphorylation of SMC1 at Ser957, as well as decreased autophosphorylation of ATM at Ser1981 (Fig. 5A). In the Fig. 4. Suppression of NFBD1/MDC1 compromises ataxia-telangiectasia mutated NBS cells reconstituted with wild-type Nbs1, suppression of 53BP1 (ATM) autophosphorylation in response to DNA double-strand breaks. A, suppression of expression had no effect on the electrophoretic mobility supershift of NFBD1/MDC1 (NFBD-1) or Nbs1 expression by small interfering RNA (siRNA) in 33/35 normal human dermal fibroblasts (NHDF) leads to reduced Chk2 phosphorylation at Chk2 after irradiation or on the phosphorylation of Chk2 at Ser , Ser33/35 after irradiation but has little effect on phosphorylation of Chk2 Thr68. NHDFs SMC1 at Ser957 and ATM at Ser1981 (Fig. 5B). were transfected with control siRNA (ctl-i) or siRNA directed against nfbd1/mdc1 (nfbd1-i)ornbs1 (nbs1-i). Seventy-two h later the cells were exposed to ionizing radiation (IR). Cell extracts prepared 10 min after irradiation were immunoblotted with phos- Discussion phospecific antibodies for Chk2 as indicated. B and C, suppression of NFBD1/MDC1 expression by siRNA in HeLa cells and fibroblasts from a patient with xeroderma pigmentosum complementation group C (XPC) compromises autophosphorylation of The major obstacle toward establishing whether the adaptors/me- ATM at Ser1981 after irradiation. The cells were treated as described above for A. diators 53BP1 and NFBD1/MDC1 function upstream or downstream 8589

vation has been suggested to be dependent on the yeast Mre11– Rad50–Xrs2 complex, which is the ortholog of the human Mre11– Rad50–Nbs1 complex (29). The fact that 53BP1 and NFBD1/MDC1 are independently recruited to sites of DNA DSBs and the siRNA phenotypes suggests that these two proteins activate ATM through independent pathways. The apparent functional independence of the 53BP1 and NFBD1/MDC1 pathways is further supported by the observation that suppression of 53BP1 led to increased recruitment of NFBD1/MDC1 and Nbs1 to sites of DNA DSBs, as well as to increased phosphorylation of Nbs1 at Ser343. This also indicates than once one pathway, e.g., the 53BP1 pathway, is established to activate ATM, then the other redundant pathways are inhibited. As shown in Fig. 5C, we consider that the Mre11–Rad50–Nbs1 complex functions within the context of the NFBD1/MDC1 pathway because recruitment of Nbs1 to sites of DNA DSBs is dependent on NFBD1/MDC1 and because NFBD1/MDC1 was identified as the fourth component of a purified Mre11–Rad50– Nbs1–NFBD1/MDC1 complex (5). Our observations do not allow us to establish whether the 53BP1 and NFBD1/MDC1 pathways are the only pathways that can activate ATM. It is possible that additional pathways may also activate ATM, which would explain why suppression of 53BP1 in NBS cells did not lead to complete loss of ATM autophosphorylation at Ser1981 after irradiation. Our studies do not address the molecular mechanism by which 53BP1 and NFBD1/MDC1 activate ATM. In the case of NFBD1/ MDC1, the mechanism may involve recruitment of the Mre11– Rad50–Nbs1 complex to sites of DNA DSBs, which may then recruit and activate ATM. In support of this mechanism, in human cells Fig. 5. Suppression of 53BP1 in NBS cells compromises ataxia-telangiectasia mutated retention of ATM on chromatin after DNA damage is dependent on (ATM) autophosphorylation in response to DNA double-strand breaks (DSBs). A, NBS the Mre11–Rad50–Nbs1 complex (21, 22), and in budding yeast Xrs2 cells were transfected with control small interfering RNA (siRNA; ctl-i) or siRNA directed against 53bp1 (53bp1-i). Seventy-two h later, the cells were exposed to ionizing recruits Tel1 to sites of DNA DSBs (30). A similar mechanism may radiation (IR). Cell extracts were prepared 10 or 30 min after irradiation and immuno- underlie activation of ATM by 53BP1 because 53BP1 and ATM blotted with phosphospecific antibodies for Chk2, SMC1, and ATM, as indicated. B, NBS cells reconstituted with wild-type Nbs1 were transfected with control siRNA (ctl-i)or coimmunoprecipitate in irradiated, but not in control, cells (12). siRNA directed against 53bp1 (53bp1-i). Seventy-two h later, the cells were exposed to Irrespective of the mechanism, our findings are not inconsistent with IR. Cell extracts were prepared 30 min after irradiation and immunoblotted with phos- a recent model proposing that ATM activation is initiated by DNA phospecific antibodies for Chk2, SMC1, and ATM, as indicated. C, diagram showing ATM activation by parallel independent pathways involving 53BP1 and NFBD1/MDC1, DSB-initiated changes in chromatin structure that lead to autophos- respectively. NFBD1/MDC1 is shown in the same pathway with the Mre11–Rad50–Nbs1 phorylation of ATM at Ser1981 and dissociation of inactive ATM complex, based on its identification as a new subunit of the Mre11–Rad50–Nbs1 com- dimers into active ATM monomers (23). The DNA DSB-initiated plex. Arrows from ATM to 53BP1, NFBD1/MDC1, and Nbs1 indicate that these proteins are ATM substrates. Phosphorylation of Nbs1 is important for its function as an adaptor. change in chromatin structure may be the stimulus for recruitment of The diagram also shows that at low doses of IR the 53BP1 pathway inhibits the 53BP1 or NFBD1/MDC1 to megadaltons of DNA flanking the DSB recruitment of NFBD1/MDC1 and Nbs1 to sites of DNA DSBs as well as the ATM- dependent phosphorylation of Nbs1 at Ser343. and, subsequently, for ATM recruitment and activation. of ATM had been the lack of a sensitive assay to monitor ATM Acknowledgments activation. This obstacle was overcome recently by development of a We thank Michael Kastan for the gift of the ATM phosphospecific antibody phosphospecific antibody that monitors ATM autophosphorylation at before its publication; Patrick Concannon for the gift of control NBS cells and Ser1981 (23). On the basis of our results obtained with this antibody, NBS cells reconstituted with wild-type Nbs1 and Nbs1 Ala343; Errol Friedberg we propose that 53BP1 and NFBD1/MDC1 activate ATM through for the gift of XPC cells; Steve Jackson, Junjie Chen, and Yossi Shiloh for independent pathways that may, however, interact with each other, antibodies; and Yossi Shiloh and Matthew Weitzman for exchange of unpub- such that decreased function of the 53BP1 pathway may be compen- lished results. sated for by increased function of the NFBD1/MDC1 pathway (Fig. 5C). References Placing 53BP1 and NFBD1/MDC1 upstream of ATM is consistent 1. Shiloh, Y. ATM and related protein kinases: safeguarding genome integrity. Nat. Rev. with previous observations. First, both NFBD1/MDC1 and 53BP1 are Cancer, 3: 155–168, 2003. recruited to sites of DNA DSBs independent of ATM, as shown by 2. Melo, J., and Toczyski, D. A unified view of the DNA-damage checkpoint. Curr. analysis of cells from patients with ataxia-telangiectasia (5, 7, 26). Opin. Cell Biol., 14: 237–245, 2002. 3. Iliakis, G., Wang, Y., Guan, J., and Wang, H. DNA damage checkpoint control in Second, NFBD1/MDC1 is required for recruitment of Nbs1 to sites of cells exposed to ionizing radiation. Oncogene, 22: 5834–5847, 2003. DNA DSBs (5, 7), and the Mre11-Rad50-Nbs1 complex is required 4. Wang, B., Matsuoka, S., Carpenter, P. B., and Elledge, S. J. 53BP1, a mediator of the for ATM activation in human cells (21, 22). Assuming that the DNA damage checkpoint. Science (Wash. DC), 298: 1435–1438, 2002. 5. Goldberg, M., Stucki, M., Falck, J., D’Amours, D., Rahman, D., Pappin, D., Bartek, Mre11–Rad50–Nbs1 complex activates ATM at sites of DNA DSBs, J., and Jackson, S. P. MDC1 is required for the intra-S-phase DNA damage check- then NFBD1/MDC1 would be expected to be required for ATM point. Nature (Lond.), 421: 952–956, 2003. 6. Lou, Z., Minter-Dykhouse, K., Wu, X., and Chen, J. MDC1 is coupled to activated activation. Third, budding yeast Tel1, the ortholog of human ATM, is CHK2 in mammalian DNA damage response pathways. Nature (Lond.), 421: 957– not thought to be directly activated by DNA DSBs. Rather, its acti- 961, 2003. 8590

7. Stewart, G. S., Wang, B., Bignell, C. R., Taylor, A. M., and Elledge, S. J. MDC1 is 18. Girard, P. M., Riballo, E., Begg, A. C., Waugh, A., and Jeggo, P. A. Nbs1 promotes a mediator of the mammalian DNA damage checkpoint. Nature (Lond.), 421: 961– ATM dependent phosphorylation events including those required for G1/S arrest. 966, 2003. Oncogene, 21: 4191–4199, 2002. 8. Shang, Y. L., Bodero, A. J., and Chen, P. L. NFBD1, a novel nuclear protein with 19. Lim, D. S., Kim, S. T., Xu, B., Maser, R. S., Lin, J., Petrini, J. H., and Kastan, M. B. signature motifs of FHA and BRCT, and an internal 41-amino acid repeat sequence, ATM phosphorylates p95/nbs1 in an S-phase checkpoint pathway. Nature (Lond.), is an early participant in DNA damage response. J. Biol. Chem., 278: 6323–6329, 404: 613–617, 2000. 2003. 20. Gatei, M., Young, D., Cerosaletti, K. M., Desai-Mehta, A., Spring, K., Kozlov, S., 9. Xu, X., and Stern, D. F. NFBD1/KIAA0170 is a chromatin-associated protein Lavin, M. F., Gatti, R. A., Concannon, P., and Khanna, K. ATM-dependent phosphorylation of nibrin in response to radiation exposure. Nat. Genet., 25: 115–119, involved in DNA damage signaling pathways. J. Biol. Chem., 278: 8795–8803, 2003. 2000. 10. Peng, A., and Chen, P. L. NFBD1, like 53BP1, is an early and redundant transducer 21. Uziel, T., Lerenthal, Y., Moyal, L., Andegeko, Y., Mittelman, L., and Shiloh, Y. mediating Chk2 phosphorylation in response to DNA damage. J. Biol. Chem., 278: Requirement of the MRN complex for ATM activation by DNA damage. EMBO J., 8873–8876, 2003. 22: 5612–5621, 2003. 11. Fernandez-Capetillo, O., Chen, H. T., Celeste, A., Ward, I., Romanienko, P. J., 22. Carson, C. T., Schwartz, R. A., Stracker, T. H., Lilley, C. E., Lee, D. V., and Morales, J. C., Naka, K., Xia, Z., Camerini-Otero, R. D., Motoyama, N., Carpenter, Weitzman, M. D. The Mre11 complex is required for ATM activation and the G2/M P. B., Bonner, W. M., Chen, J., and Nussenzweig, A. DNA damage-induced G2-M checkpoint. EMBO J., in press. checkpoint activation by histone H2AX and 53BP1. Nat. Cell Biol., 4: 993–997, 23. Bakkenist, C. J., and Kastan, M. B. DNA damage activates ATM through intermo- 2002. lecular autophosphorylation and dimer dissociation. Nature (Lond.), 421: 499–506, 12. DiTullio, R. A., Jr., Mochan, T. A., Venere, M., Bartkova, J., Sehested, M., Bartek, 2003. J., and Halazonetis, T. D. 53BP1 functions in an ATM-dependent checkpoint pathway 24. Desai-Mehta, A., Cerosaletti, K. M., and Concannon, P. Distinct functional domains that is constitutively activated in human cancer. Nat. Cell Biol., 4: 998–1002, 2002. of nibrin mediate Mre11 binding, focus formation, and nuclear localization. Mol. 13. Ward, I. M., Minn, K., van Deursen, J., and Chen, J. p53 Binding protein 53BP1 is Cell. Biol., 21: 2184–2191, 2001. required for DNA damage responses and tumor suppression in mice. Mol. Cell. Biol., 25. Chehab, N. H., Malikzay, A., Appel, M., and Halazonetis, T. D. Chk2/hCds1 23: 2556–2563, 2003. functions as a DNA damage checkpoint in G(1) by stabilizing p53. Genes Dev., 14: 14. Morales, J. C., Xia, Z., Lu, T., Aldrich, M. B., Wang, B., Rosales, C., Kellems, R. E., 278–288, 2000. Hittelman, W. N., Elledge, S. J., and Carpenter, P. B. Role for the BRCA1 C-terminal 26. Schultz, L. B., Chehab, N. H., Malikzay, A., and Halazonetis, T. D. p53 binding repeats (BRCT) protein 53BP1 in maintaining genomic stability. J. Biol. Chem., 278: protein 1 (53BP1) is an early participant in the cellular response to DNA double- strand breaks. J. Cell Biol., 151: 1381–1390, 2000. 14971–14977, 2003. 27. Rogakou, E. P., Boon, C., Redon, C., and Bonner, W. M. Megabase chromatin 15. Buscemi, G., Savio, C., Zannini, L., Micciche, F., Masnada, D., Nakanishi, M., domains involved in DNA double-strand breaks in vivo. J. Cell Biol., 146: 905–916, Tauchi, H., Komatsu, K., Mizutani, S., Khanna, K., Chen, P., Concannon, P., Chessa, 1999. L., and Delia, D. Chk2 activation dependence on Nbs1 after DNA damage. Mol. Cell. 28. Kao, G. D., McKenna, W. G., Guenther, M. G., Muschel, R. J., Lazar, M. A., and Biol., 21: 5214–5222, 2001. Yen, T. J. Histone deacetylase 4 interacts with 53BP1 to mediate the DNA damage 16. Kim, S. T., Xu, B., and Kastan, M. B. Involvement of the cohesin protein, Smc1, in response. J. Cell Biol., 160: 1017–1027, 2003. Atm-dependent and independent responses to DNA damage. Genes Dev., 16: 560– 29. Usui, T., Ogawa, H., and Petrini, J. H. A DNA damage response pathway controlled 570, 2002. by Tel1 and the Mre11 complex. Mol. Cell., 7: 1255–1266, 2001. 17. Yazdi, P. T., Wang, Y., Zhao, S., Patel, N., Lee, E. Y., and Qin, J. SMC1 is a 30. Nakada, D., Matsumoto, K., and Sugimoto, K. ATM-related Tel1 associates with downstream effector in the ATM/NBS1 branch of the human S-phase checkpoint. double-strand breaks through an Xrs2-dependent mechanism. Genes Dev., 17: 1957– Genes Dev., 16: 571–582, 2002. 1962, 2003.

8591

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 2003 American Association for Cancer Research. 53BP1 and NFBD1/MDC1-Nbs1 Function in Parallel Interacting Pathways Activating Ataxia-Telangiectasia Mutated (ATM) in Response to DNA Damage

Tamara A. Mochan, Monica Venere, Richard A. DiTullio, Jr., et al.

Cancer Res 2003;63:8586-8591.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/63/24/8586

Cited articles This article cites 29 articles, 16 of which you can access for free at: http://cancerres.aacrjournals.org/content/63/24/8586.full#ref-list-1

Citing articles This article has been cited by 48 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/63/24/8586.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/63/24/8586. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.