A defect in DNA double strand break processing in cells from unaffected parents of retinoblastoma patients and other apparently normal humans

Cells from unaffected parents of retinoblastoma (RB) patients were previously shown to be hypersensitive to radiation induced G(1) arrest and cell killing [1]. The hypersensitivity was similar to that reported for cells from ATM heterozygotes. The latter was consistent with a mild DNA DSB rejoining defect which we demonstrated using a gamma-H2AX focus assay after low dose-rate (LDR) irradiation of non-cycling G(0) cells [2,3]. Since neither parent carried the mutant RB allele of the RB heterozygous probands, these results suggested the possibility of an enhanced germline mutation rate, perhaps resulting from some mild defect in genome maintenance. We therefore examined levels of gamma-H2AX foci for cells from these RB parents in this G(0) LDR assay, which reflects the non-homologous end joining (NHEJ) capacity of cells and in a G(2)/M assay, which reflects additional contributions from other G(2)-related damage processing systems. For several of the cell strains parallel radiosensitivity comparisons were made for cell killing and for G(2) chromosomal radiosensitivities. G(0) cells from the RB parents were clearly hypersensitive both in the LDR gamma-H2AX assay, and for cell killing. In addition, cultured fibroblasts from 6 of 15 apparently normal individuals in this study (and one of six in a previous study) were also hypersensitive in the same assays. In the G(2)/M gamma-H2AX assay, the relative sensitivities were similar to those seen in the low dose-rate G(0) assay and tracked with chromosomal radiosensitivity, but some differences were observed.     

DNA repair defect in AT cells and their hypersensitivity to low-dose-rate radiation

Ataxia telangiectasia (AT) and normal cells immortalized with the human telomerase gene were irradiated in non-proliferative conditions with high- (2 Gy/min) or low-dose-rate (0.3 mGy/min) radiation. While normal cells showed a higher resistance after irradiation at a low dose rate than a high dose rate, AT cells showed virtually the same survival after low- and high-dose-rate irradiation. Although the frequency of micronuclei induced by low-dose-rate radiation was greatly reduced in normal cells, it was not reduced significantly in AT cells. The number of gamma-H2AX foci increased in proportion to the dose in both AT and normal cells after high-dose-rate irradiation. Although few gamma-H2AX foci were observed after low-dose-rate irradiation in normal cells, significant and dose-dependent numbers of gamma-H2AX foci were observed in AT cells even after low-dose-rate irradiation, indicating that DNA damage was not completely repaired during low-dose-rate irradiation. Significant phosphorylation of ATM proteins was detected in normal cells after low-dose-rate irradiation, suggesting that the activation of ATM plays an important role in the repair of DNA damage during low-dose-rate irradiation. In conclusion, AT cells may not be able to repair some fraction of DNA damage and are severely affected by low-dose-rate radiation.     

Comparison of the induction and disappearance of DNA double strand breaks and gamma-H2AX foci after irradiation of chromosomes in G1-phase or in condensed metaphase cells

The induction and disappearance of DNA double strand breaks (DSBs) after irradiation of G1 and mitotic cells were compared with the gamma-H2AX foci assay and a gel electrophoresis assay. This is to determine whether cell cycle related changes in chromatin structure might influence the gamma-H2AX assay which depends on extensive phosphorylation and dephosphorylation of the H2AX histone variant surrounding DSBs. The disappearance of gamma-H2AX foci after irradiation was much slower for mitotic than for G1 cells. On the other hand, no difference was seen for the gel electrophoresis assay. Our data may suggest the limited accessibility of dephosphorylation enzyme in irradiated metaphase cells or trapped gamma-H2AX in condensed chromatin.     

Focus formation of DNA repair proteins in normal and repair-deficient cells irradiated with high-LET ions

To investigate the repair of clustered lesions within the DNA/chromatin, the focus formation and persistence of foci of the phosphorylated histone protein H2AX and the repair protein MRE11 were studied in normal cells and in cells lacking DNA-PKcs (M059J) or ATM (GM2052D) after irradiation with high-LET nitrogen ions or low-LET photons. There was a rapid formation of MRE11 and gamma-H2AX foci, and 0.5 h after high-LET irradiation, the number of foci in normal cells correlated well with the number of particle hits per cell nucleus. After 8 h of repair, there were significantly more gamma-H2AX foci than MRE11 foci remaining in the normal cells, independent of radiation quality. The difficulty in repairing clustered breaks was detected as slower rejoining of DSBs (measured by DNA fragmentation analysis), as quantification of the amount of gamma-H2AX over time, and as a larger fraction of repair foci remaining after 24 h in cells irradiated with high- LET ions. These data indicate that clustered lesions are repaired by a pathway involving the same proteins that repair sparsely distributed breaks. Further, for both low- and high- LET radiation, no reduction of the initial number of gamma-H2AX and MRE11 foci was detected in M059J cells up to 21 h after irradiation, which was in accordance with a complete absence of DSB rejoining in these cells. In the GM2052D cells there was also a higher level of foci remaining after 21 h; however, this was not accompanied by unrejoined DSBs, indicating that these foci not only represent DSBs but also may be a sign of persistent problems even when breaks are rejoined.     

Phosphorylation of histone H2AX in radiation-induced micronuclei

DNA double-strand breaks are thought to precede the formation of most radiation-induced micronuclei. Phosphorylation of the histone H2AX is an early indicator of DNA double-strand breaks. Here we studied the phosphorylation status of the histone H2AX in micronuclei after exposure of cultured cells to ionizing radiation or treatment with colchicine. In human astrocytoma SF268 cells, after exposure to gamma radiation, the proportion of gamma-H2AX-positive to gamma-H2AX-negative micronuclei increases. The majority of the gamma-H2AX-positive micronuclei are centromere-negative. The number of gamma-H2AX-positive micronuclei continues to increase even 24 h postirradiation when most gamma-H2AX foci in the main nucleus have disappeared. In contrast, in normal human fibroblasts (BJ), the proportion of gamma-H2AX-positive to gamma-H2AX-negative micronuclei remains constant, and the majority of the centromere-negative cells are gamma-H2AX-negative. Treatment of both cell lines with colchicine results in mostly centromere-positive, gamma-H2AX-negative micronuclei. Immunostaining revealed co-localization of MDC1 and ATM with gamma-H2AX foci in both main nuclei and micronuclei; however, other repair proteins, such as Rad50, 53BP1 and Rad17, that co-localized with gamma-H2AX foci in the main nuclei were not found in the micronuclei. Combination of the micronucleus assay with gamma-H2AX immunostaining provides new insights into the mechanisms of the formation and fate of micronuclei.     

Replication protein A and gamma-H2AX foci assembly is triggered by cellular response to DNA double-strand breaks

Human replication protein A (RPA p34), a crucial component of diverse DNA excision repair pathways, is implicated in DNA double-strand break (DSB) repair. To evaluate its role in DSB repair, the intranuclear dynamics of RPA was investigated after DNA damage and replication blockage in human cells. Using two different agents [ionizing radiation (IR) and hydroxyurea (HU)] to generate DSBs, we found that RPA relocated into distinct nuclear foci and colocalized with a well-known DSB binding factor, gamma-H2AX, at the sites of DNA damage in a time-dependent manner. Colocalization of RPA and gamma-H2AX foci peaked at 2 h after IR treatment and subsequently declined with increasing postrecovery times. The time course of RPA and gamma-H2AX foci association correlated well with the DSB repair activity detected by a neutral comet assay. A phosphatidylinositol-3 (PI-3) kinase inhibitor, wortmannin, completely abolished both RPA and gamma-H2AX foci formation triggered by IR. Additionally, radiosensitive ataxia telangiectasia (AT) cells harboring mutations in ATM gene product were found to be deficient in RPA and gamma-H2AX colocalization after IR. Transfection of AT cells with ATM cDNA fully restored the association of RPA foci with gamma-H2AX illustrating the requirement of ATM gene product for this process. The exact coincidence of RPA and gamma-H2AX in response to HU specifically in S-phase cells supports their role in DNA replication checkpoint control. Depletion of RPA by small interfering RNA (SiRNA) substantially elevated the frequencies of IR-induced micronuclei (MN) and apoptosis in human cells suggestive of a role for RPA in DSB repair. We propose that RPA in association with gamma-H2AX contributes to both DNA damage checkpoint control and repair in response to strand breaks and stalled replication forks in human cells.     

Low-dose hyper-radiosensitivity is not caused by a failure to recognize DNA double-strand breaks

One of the earliest cellular responses to radiation-induced DNA damage is the phosphorylation of the histone variant H2AX (gamma-H2AX). gamma-H2AX facilitates the local concentration and focus formation of numerous repair-related proteins within the vicinity of DNA DSBs. Previously, we have shown that low-dose hyper-radiosensitivity (HRS), the excessive sensitivity of mammalian cells to very low doses of ionizing radiation, is a response specific to G(2)-phase cells and is attributed to evasion of an ATM-dependent G(2)-phase cell cycle checkpoint. To further define the mechanism of low-dose hyper-radiosensitivity, we investigated the relationship between the recognition of radiation-induced DNA double-strand breaks as defined by gamma-H2AX staining and the incidence of HRS in three pairs of isogenic cell lines with known differences in radiosensitivity and DNA repair functionality (disparate RAS, ATM or DNA-PKcs status). Marked differences between the six cell lines in cell survival were observed after high-dose exposures (>1 Gy) reflective of the DNA repair capabilities of the individual six cell lines. In contrast, the absence of functional ATM or DNA-PK activity did not affect cell survival outcome below 0.2 Gy, supporting the concept that HRS is a measure of radiation sensitivity in the absence of fully functional repair. No relationship was evident between the initial numbers of DNA DSBs scored immediately after either low- or high-dose radiation exposure with cell survival for any of the cell lines, indicating that the prevalence of HRS is not related to recognition of DNA DSBs. However, residual DNA DSB damage as indicated by the persistence of gamma-H2AX foci 4 h after exposure was significantly correlated with cell survival after exposure to 2 Gy. This observation suggests that the persistence of gamma-H2AX foci could be adopted as a surrogate assay of cellular radiosensitivity to predict clinical radiation responsiveness.     

Qualitative and quantitative analysis of phosphorylated ATM foci induced by low-dose ionizing radiation

We examined the formation of phosphorylated ataxia telangiectasia mutated (ATM) foci in exponentially growing normal human diploid cells exposed to low doses of X rays. Phosphorylated ATM foci were detected immediately after irradiation, and the number of foci decreased as the time after irradiation increased. The kinetics of phosphorylated ATM foci was comparable to that of phosphorylated histone H2AX. We found that there were fewer spontaneous phosphorylated ATM foci than that phosphorylated histone H2AX foci. Notably, significant numbers of phosphorylated histone H2AX foci, but not phosphorylated ATM foci, were detected in the S-phase cells. The induction of foci showed a linear dose-response relationship with doses ranging for 10 mGy to 1 Gy, and the average number of phosphorylated ATM foci per gray was approximately 50. The average size of the foci was comparable for the cells irradiated with 20 mGy and 1 Gy, and there was no significant difference in the kinetics of disappearance of foci, indicating that DNA double-strand breaks are similarly recognized by DNA damage checkpoints and are repaired irrespective of the dose.     

Phosphorylation of histone H2AX at M phase in human cells without DNA damage response

A variant of histone H2A, H2AX, is phosphorylated on Ser139 in response to DNA double-strand breaks (DSBs), and clusters of the phosphorylated form of H2AX (gamma-H2AX) in nuclei of DSB-induced cells show foci at breakage sites. Here, we show phosphorylation of H2AX in a cell cycle-dependent manner without any detectable DNA damage response. Western blot and immunocytochemical analyses with the anti-gamma-H2AX antibody revealed that H2AX is phosphorylated at M phase in HeLa cells. In ataxia-telangiectasia cells lacking ATM kinase activity, gamma-H2AX was scarcely detectable in the mitotic chromosomes, suggesting involvement of ATM in M-phase phosphorylation of H2AX. Single-cell gel electrophoresis assay and Western blot analysis with the anti-phospho-p53 (Ser15) antibody indicated that H2AX in human M-phase cells is phosphorylated independently of DSB and DNA damage signaling. Even in the absence of DNA damage, phosphorylation of H2AX in normal cell cycle progression may contribute to maintenance of genomic integrity.     

Specific ATM-mediated phosphorylation dependent on radiation quality

Whalen, M. K., Gurai, S. K., Zahed-Kargaran, H. and Pluth, J. M. Specific ATM-Mediated Phosphorylation Dependent on Radiation Quality. Radiat. Res. 170, 353-364 (2008).To determine whether the physical differences between high- and low-LET radiation are reflected in the biological responses of exposed cells, we detailed phospho-protein profiles of three proteins functional in radiation repair and signal transduction. Detailing gamma-H2AX, pATF2 Ser(490/498) and pSMC1 Ser(957) kinetics after X-ray and iron-ion exposure also provides a window into understanding the underlying cellular responses. Phosphorylated forms of these proteins have been documented to co-localize at sites of double-strand breaks (DSBs) after low-LET radiation exposures, and two of these phosphorylations, pATF2 and pSMC1, are specifically dependent on ATM. Flow cytometry-based methods were used to quantify total levels of each phospho-protein at various times after irradiation. As expected, we observed a greater induction and persistence in gamma-H2AX after iron-ion (high-LET) exposure compared to X-ray (low-LET) exposure. In contrast, pATF2 and pSMC1 showed markedly lower induction levels after iron-ion exposure compared to equivalent doses of X rays. Quantification of pATF2 and pSMC1 foci revealed fewer cells containing foci and fewer foci per cell after iron-ion compared to X-ray exposure. These findings suggest that ATM responds to DSBs induced by high-LET radiation differently from DSBs induced by low-LET radiation.     

gamma-H2AX dephosphorylation by protein phosphatase 2A facilitates DNA double-strand break repair

Phosphorylated histone H2AX (gamma-H2AX) forms foci over large chromatin domains surrounding double-stranded DNA breaks (DSB). These foci recruit DSB repair proteins and dissolve during or after repair is completed. How gamma-H2AX is removed from chromatin remains unknown. Here, we show that protein phosphatase 2A (PP2A) is involved in removing gamma-H2AX foci. The PP2A catalytic subunit [PP2A(C)] and gamma-H2AX coimmunoprecipitate and colocalize in DNA damage foci and PP2A dephosphorylates gamma-H2AX in vitro. The recruitment of PP2A(C) to DNA damage foci is H2AX dependent. When PP2A(C) is inhibited or silenced by RNA interference, gamma-H2AX foci persist, DNA repair is inefficient, and cells are hypersensitive to DNA damage. The effect of PP2A on gamma-H2AX levels is independent of ATM, ATR, or DNA-PK activity.     

Elimination of radiation-induced gamma-H2AX foci in mammalian nucleus can occur by histone exchange

Double-strand breaks in mammalian DNA lead to rapid phosphorylation of C-terminal serines in histone H2AX (gamma-H2AX) and formation of large nuclear gamma-H2AX foci. After DNA repair these foci disappear, but molecular mechanism of elimination of gamma-H2AX foci remains unclear. H2AX protein can be phosphorylated and dephosphorylated in vitro in the absence of chromatin. Here, we compared global exchange of GFP-H2AX with kinetics of formation and elimination of radiation-induced gamma-H2AX foci. Maximal number of gamma-H2AX foci is observed one hour after irradiation, when approximately 20% of GFP-H2AX is exchanged suggesting that formation of the foci mostly occurs by in situ H2AX phosphorylation. However, slow elimination of gamma-H2AX foci is weakly affected by an inhibitor of protein phosphatases calyculin A which is known as an agent suppressing dephosphorylation of gamma-H2AX. This indicates that elimination of gamma-H2AX foci may be independent of dephosphorylation of H2AX which can occur after its removal from the foci by exchange.     

Radiation-induced H2AX phosphorylation and neural precursor apoptosis in the developing brain of mice

We showed that gamma irradiation of the developing mouse brain with 2 Gy induced a massive apoptosis of neural precursors but not of neurons within 24 h. Successive phosphorylation and dephosphorylation of histone H2AX have been linked to DNA breaks and repair. Similar numbers of nuclear foci of phosphorylated H2AX (gamma-H2AX) were found 1 h postirradiation in neural precursors and in neurons, suggesting that differences in radiosensitivity were not related to variations in the numbers of DNA double-strand breaks induced by radiation. Surviving neural precursors like neurons totally lost gamma-H2AX within 24 h after irradiation, but they had a slower kinetics of loss of gamma-H2AX foci. This suggests that the DNA repair machinery processed damage more slowly in these neural precursors in relation to their greater radiosensitivity. We also found a bright and diffuse gamma-H2AX staining of nuclei of cells at an early stage of apoptosis, whereas cells at later stages of apoptosis were unstained. This was probably related to phosphorylation and subsequent degradation of H2AX in the course of DNA fragmentation during apoptosis. Detection of gamma-H2AX-bright nuclei may thus be a useful marker of neural cells at an early stage of apoptosis.     

Complex H2AX phosphorylation patterns by multiple kinases including ATM and DNA-PK in human cells exposed to ionizing radiation and treated with kinase inhibitors

In eukaryotic cells, DNA double strand breaks (DSBs) cause the prompt phosphorylation of serine 139 at the carboxy terminus of histone H2AX to generate gamma-H2AX, detectable by Western blotting or immunofluorescence. The consensus sequence at the phosphorylation site implicates the phosphatidylinositol 3-like family of protein kinases in H2AX phosphorylation. It remains open whether ATM (ataxia telangiectasia mutated) is the major H2AX kinase, or whether other members of the family, such as DNA-PK (DNA dependent protein kinase) or ATR (ATM and Rad3 related), contribute in a functionally complementary manner. To address this question, we measured global H2AX phosphorylation in cell lysates and foci formation in individual cells of either wild type or mutant (ATM or DNA-PK) genetic background. Normal global phosphorylation kinetics is observed after irradiation in cells defective either in ATM or DNA-PK alone, suggesting a complementary contribution to H2AX phosphorylation. This is further supported by the observation that initial H2AX phosphorylation is delayed when both kinases are inhibited by wortmannin, as well as when ATM is inhibited by caffeine in DNA-PK deficient cells. However, robust residual global phosphorylation is detectable under all conditions of genetic or chemical inhibition suggesting the function of additional kinases, such as ATR. Treatment with wortmannin, caffeine, or UCN-01 produces a strong DNA-PK dependent late global hyperphosphorylation of H2AX, uncoupled from DNA DSB rejoining and compatible with an inhibition of late steps in DNA DSB processing. Evaluation of gamma-H2AX foci formation confirms the major conclusions made on the basis of global H2AX phosphorylation, but also points to differences particularly several hours after exposure to IR. The results in aggregate implicate DNA-PK, ATM and possibly other kinases in H2AX phosphorylation. The functional significance and the mechanisms of coordination in space and time of these multiple inputs require further investigation.     

Elevated sodium chloride concentrations enhance the bystander effects induced by low dose alpha-particle irradiation

Previous studies have shown that high NaCl can be genotoxic, either alone or combined with irradiation. However, little is known about the relationship between environmental NaCl at elevated conditions and radiation-induced bystander effects (RIBE). RIBE, which has been considered as non-targeted bystander responses, has been demonstrated to occur widely in various cell lines. In the present study, RIBE under the elevated NaCl culture condition was assessed in AG 1522 cells by both the induction of gamma-H2AX, a reliable marker of DNA double-strand break (DSB) for the early process (<1h post irradiation), and the generation of micronuclei (MN), a sensitive marker for relative long process of RIBE. Our results showed that in the absence of irradiation, NaCl at elevated concentration such as 8.0, 9.0 and 10.0g/L did not significantly increase the frequency of gamma-H2AX foci-positive cells and the number of foci per positive cell comparing with that NaCl at a normal concentration (6.8g/L). However, with 0.2cGy alpha-particle irradiation, the induced fraction of gamma-H2AX foci-positive cells and the number of induced gamma-H2AX foci per positive cell were significantly increased in both irradiated and adjacent non-irradiated regions. Similarly, the induction of MN by 0.2cGy alpha-particle irradiation also increased with the elevated NaCl concentrations. With N(G)-methyl-l-arginine, an inhibitor of nitric oxide synthase, the induced fraction of foci-positive cells was effectively inhibited both in 0.2cGy alpha-particle irradiated and adjacent non-irradiated regions under either normal or elevated NaCl conditions. These results suggested that the cultures with elevated NaCl medium magnified the damage effects induced by the low dose alpha-particle irradiation and nitric oxide generated by irradiation was also very important in this process.     

Megabase chromatin domains involved in DNA double-strand breaks in vivo.

The loss of chromosomal integrity from DNA double-strand breaks introduced into mammalian cells by ionizing radiation results in the specific phosphorylation of histone H2AX on serine residue 139, yielding a specific modified form named gamma-H2AX. An antibody prepared to the unique region of human gamma-H2AX shows that H2AX homologues are phosphorylated not only in irradiated mammalian cells but also in irradiated cells from other species, including Xenopus laevis, Drosophila melanogaster, and Saccharomyces cerevisiae. The antibody reveals that gamma-H2AX appears as discrete nuclear foci within 1 min after exposure of cells to ionizing radiation. The numbers of these foci are comparable to the numbers of induced DNA double-strand breaks. When DNA double-strand breaks are introduced into specific partial nuclear volumes of cells by means of a pulsed microbeam laser, gamma-H2AX foci form at these sites. In mitotic cells from cultures exposed to nonlethal amounts of ionizing radiation, gamma-H2AX foci form band-like structures on chromosome arms and on the end of broken arms. These results offer direct visual confirmation that gamma-H2AX forms en masse at chromosomal sites of DNA double-strand breaks. The results further suggest the possible existence of units of higher order chromatin structure involved in monitoring DNA integrity.     

Quantitative detection of (125)IdU-induced DNA double-strand breaks with gamma-H2AX antibody

When mammalian cells are exposed to ionizing radiation and other agents that introduce DSBs into DNA, histone H2AX molecules in megabase chromatin regions adjacent to the breaks become phosphorylated within minutes on a specific serine residue. An antibody to this phosphoserine motif of human H2AX (gamma-H2AX) demonstrates that gamma-H2AX molecules appear in discrete nuclear foci. To establish the quantitative relationship between the number of these foci and the number of DSBs, we took advantage of the ability of (125)I, when incorporated into DNA, to generate one DNA DSB per radioactive disintegration. SF-268 and HT-1080 cell cultures were grown in the presence of (125)IdU and processed immunocytochemically to determine the number of gamma-H2AX foci. The numbers of (125)IdU disintegrations per cell were measured by exposing the same immunocytochemically processed samples to a radiation-sensitive screen with known standards. Under appropriate conditions, the data yielded a direct correlation between the number of (125)I decays and the number of foci per cell, consistent with the assumptions that each (125)I decay yields a DNA DSB and each DNA DSB yields a visible gamma-H2AX focus. Based on these findings, we conclude that gamma-H2AX antibody may form the basis of a sensitive quantitative method for the detection of DNA DSBs in eukaryotic cells.     

Explanation for excessive DNA single-strand breaks and endogenous repair foci in pluripotent mouse embryonic stem cells

Pluripotent mouse embryonic stem cells (mES cells) exhibit ∼ 100 large γH2AX repair foci in the absence of measurable numbers of DNA double-strand breaks. Many of these cells also show excessive numbers of DNA single-strand breaks (> 10,000 per cell) when analyzed using the alkaline comet assay. To understand the reasons for these unexpected observations, various methods for detecting DNA strand breaks were applied to wild-type mES cells and to mES cells lacking H2AX, ATM, or DNA-PK_cs. H2AX phosphorylation and expression of other repair complexes were measured using flow and image analysis of antibody-stained cells. Results indicate that high numbers of endogenous γH2AX foci and single-strand breaks in pluripotent mES cells do not require ATM or DNA-PK kinase activity and appear to be associated with global chromatin decondensation rather than pre-existing DNA damage. This will limit applications of γH2AX foci analysis in mES cells to relatively high levels of initial or residual DNA damage. Excessive numbers of single-strand breaks in the alkaline comet assay can be explained by the vulnerability of replicating chromatin in mES cells to osmotic shock. This suggests that caution is needed in interpreting results with the alkaline comet assay when applied to certain cell types or after treatment with agents that make chromatin vulnerable to osmotic changes. Differentiation of mES cells caused a reduction in histone acetylation, γH2AX foci intensity, and DNA single-strand breakage, providing a link between chromatin structural organization, excessive γH2AX foci, and sensitivity of replicating mES cell chromatin to osmotic shock.     

Nuclear retention of ATM at sites of DNA double strand breaks.

The ATM protein kinase mediates a rapid induction of cellular responses to DNA double strand breaks (DSBs). ATM kinase activity is enhanced immediately after exposure of cells to DSB-inducing agents, but no changes in its amount or subcellular location following that activation have been reported. We speculated that some of the ATM molecules associate with sites of DSBs, while the rest of the nuclear ATM pool remains in the nucleoplasm, masking detection of the damage-associated ATM fraction. Using detergent extraction to remove nucleoplasmic proteins, we show here that immediately following induction of DSBs, a fraction of the ATM pool becomes resistant to extraction and is detected in nuclear aggregates. Colocalization of the retained ATM with the phosphorylated form of histone H2AX (gamma-H2AX) and with foci of the Nbs1 protein suggests that ATM associates with sites of DSBs. The striking correlation between the appearance of retained ATM and of gamma-H2AX, and the rapid association of a fraction of ATM with gamma-H2AX foci, are consistent with a major role for ATM in the early detection of DSBs and subsequent induction of cellular responses.