Introduction

Exposure of cells to endogenous or exogenous genotoxic agents generates at different stages of the cell cycle a variety of structurally different DNA lesions, including single-strand breaks (SSBs), double-strand breaks (DSBs), base modifications and crosslinking. In eukaryotic cells, these lesions activate a network of signalling pathways or checkpoints that coordinately delay cell cycle progression and activate DNA repair mechanisms (for a review, see Walworth, 2000). Incorrect activation of these surveillance systems leads to genomic instability and, ultimately, to cancer predisposition (McDonald and El-Deiry, 2001).

Chk2, the mammalian orthologue of yeast Rad53/Cds1 kinase, participates in multiple cell cycle phase arrest in response to DNA damage. The activation of Chk2 involves an initial phosphorylation step on Thr68 (Brown et al., 1999; Matsuoka et al., 2000; Ward et al., 2001) by Ataxia Telangiectasia Mutated (ATM), the kinase that functions in DNA damage signalling by targeting several effector molecules (for a review, see Abraham, 2001). Following this initial event, Chk2 undergoes a cascade of phosphorylation events (Schwarz et al., 2003) that eventually lead to its activation. Interestingly, the sustained activation of Chk2 requires the functional expression of the Mre11/Rad50/Nbs1 (MRN) complex (Buscemi et al., 2001; Girard et al., 2002) crucial for DNA damage sensing and repair, and checkpoint activation (D'Amours and Jackson, 2002). Mutations of NBS1 and MRE11 genes are involved in Nijmegen breakage syndrome (NBS) and AT-like disease (ATLD), both sharing phenotypic similarities with AT, including chromosome instability, hypersensitivity to genotoxic agents and cell cycle checkpoint defects (Lavin and Khanna, 1999; Abraham, 2001; Taylor, 2001). In addition to being a target of ATM (Gatei et al., 2000; Zhao et al., 2000), the MRN complex is a critical stimulator of ATM kinase, as underscored by the markedly defective radiation-induced autophosphorylation of ATM Ser1981 and phosphorylation of ATM substrates, for example, p53-Ser15 and Chk2-Thr68 in NBS and ATLD cells (Uziel et al., 2003). Additional biochemical studies have substantiated the critical role of MRN in the stimulation of ATM activity (Lee and Paull, 2004).

Activated Chk2 contributes to the transcriptional induction by p53 of the Cdk inhibitor p21^waf1, thus allowing G1 arrest (Chehab et al., 2000; Hirao et al., 2000; Shieh et al., 2000; Takai et al., 2002; Vaziri et al., 2003). The mechanism by which Chk2 regulates p53 remains elusive, particularly since the claimed phosphorylation of p53-Ser20 by Chk2 has been recently questioned (Ahn et al., 2003; Jallepalli et al., 2003). Through the phosphorylation of Cdc25A–Ser123 (Falck et al., 2001) and Cdc25C–Ser216 (Brown et al., 1999), Chk2 enforces the S- and G2-M phase arrest, respectively. In addition to regulating cell cycle checkpoints, Chk2 appears to regulate a number of proapoptotic targets of p53 such as Bax and Noxa (Hirao et al., 2002; Takai et al., 2002) and this activity has been recently demonstrated by genome-wide comparisons in Drosophila embryos wild type or deficient for MNK (homologue of Chk2) (Brodsky et al., 2004). Mutations of CHK2 have been found associated with the Li–Fraumeni-like syndrome, with familial breast cancer and sporadic tumors (for a review, see Bartek and Lukas, 2003), underlying a function for Chk2 as a tumor suppressor.

As the nature of the signal required to activate Chk2 has still not been fully defined, we have assessed the ATM–Chk2 pathway in relation to the initial yield of SSBs and DSBs caused by different types and doses of genotoxic agents. Our study demonstrates that, while being unresponsive to SSBs, ATM and Chk2 kinases are differentially sensitive to DSBs, with the former being activated by <8 breaks per cell, as reported already by Bakkenist and Kastan (2003), and the latter by >19 breaks per cell. Together with other observations, these findings suggest that cells can repair DNA lesions present below a certain threshold level, without activating the Chk2-dependent cell cycle checkpoint arrest.

Results

DNA SSBs and DSBs evaluation after genotoxic treatments

The initial yield of genomic SSBs and DSBs generated in cells by genotoxic agents was determined in exponentially growing cells by the alkaline elution technique (AET) (Kohn et al., 1981). As explained below, DSBs were, in addition, quantitated by the analysis of phosphorylated H2AX (-H2AX) foci (Rogakou et al., 1999). In LCL-N cells, SSBs determined by AET increased with the dose of IR, but even at the lowest dose tested (0.25 Gy) their yield was rather marked, accounting for a fragmentation index (FI) of 40% (Figure 1a). DSBs, on the other hand, became detectable by AET at doses of IR >2 Gy (FI: 8% at 2 Gy) (Figure 1a), and their yield increased in a dose-dependent manner. As expected, massive amounts of SSBs and DSBs (FI: 100%) were seen after 50 Gy. Extensive SSBs were seen in LCL-N cells exposed to 20 and 100 M H₂O₂ (FI: 50 and 100%, respectively); however, DSBs were seen only in response to the latter concentration (FI: 40–60%) (Figure 1a). The yields of SSBs and DSBs in irradiated AT52 and NBS cells were close to those of LCL-N receiving similar treatment (data not shown).

Figure 1.

Quantitation of DNA strand breaks by alkaline elution and -H2AX nuclear foci in LCL-N cells exposed to IR or H₂O₂. SSBs and DSBs (a) were detected by AET on DNA extracted from cells immediately after irradiation on ice, or exposure to H₂O₂. The yield of these lesions is expressed as an FI, relative to untreated cells, as described in Materials and methods. (b) Immunofluorescence analysis and quantitation of -H2AX nuclear foci before (0) and after treatment with the indicated doses of IR, 4NQO, H₂O₂. Cells were allowed to recover for 5 min or 1 h prior to harvesting. Foci were scored by fluorescence microscopy and the results are shown at the bottom. (c) Quantitation of -H2AX foci (s.d.) in the ATM-deficient cell line AT52, scored 10 min and 1 h post-IR. (d) Quantitation of -H2AX foci (s.d.) in HCT116 and HCT116-Chk2-/- cells. Foci were scored 5 min and 1 h post-IR

Full figure and legend (226K)

Although the AET enables the quantification of both SSBs and DSBs from the same sample, because of its sensitivity limit it cannot detect DSBs induced by <2 Gy IR, estimated to generate <60 breaks per diploid cell (Rogakou et al., 1999). As the average number of -H2AX foci per cell well compares with the number of induced DSBs (Rogakou et al., 1999; Petersen et al., 2001; Furuta et al., 2003; Jakob et al., 2003; Rothkamm and Lobrich, 2003), we quantitated DSBs in exponentially growing cells by enumerating -H2AX foci. In LCL-N, the basal number of foci per cells accounted for 2.80.9, but 5 min after irradiation with 0.25, 0.5 and 1 Gy their number rose to 7.80.4, 13.60.8 and 19.91, respectively (Figure 1b). When analysed 1 h post-irradiation (post-IR), the foci decreased to 41 (0.25 Gy), 101 (0.5 Gy) and 122.1 (1 Gy) (Figure 1b), evidence that a fraction of DNA lesions had undergone repair (Nunez et al., 1995). Concordant with other studies (Rogakou et al., 1999), the foci per cell showed, in all samples, a Poisson distribution (data not shown). LCL-N cells treated with 20 and 100 M H₂O₂ presented 6.90.5 and 21.41 -H2AX foci, respectively, but these values were markedly reduced 1 h after recovery (Figure 1b), due to a repair activity. By contrast, no foci were seen in response to 4-NQO, a UV-mimetic compound generating base modifications and secondary SSBs, but not DSBs (Brosh et al., 1999; Mirzayans et al., 1999), thus supporting the specificity of -H2AX for DSBs. The damaging activity of 4-NQO was confirmed by the accumulation of p53 protein (data not shown).

To examine in more detail the DNA damage response in relation to ATM and Chk2, we monitored the radiation-induced -H2AX foci in the ATM- and Chk2-deficient cell lines AT52 and HCT116-Chk2-/-, respectively. As in ATM-deficient cells the IR-induced foci are slightly delayed, though quantitatively similar to wild-type cells (Kuhne et al., 2004; Stiff et al., 2004), the initial foci in AT52 cells were analysed 10 min (rather than 5 min) after irradiation. At this time point, AT52 exhibited a similar number of (but less brighter) foci as LCL-N cells treated in the same way (Figure 1c). In contrast to LCL-N, however, the foci in AT52 at 1 h post-IR slightly decreased only in response to 1 Gy, but not to lower doses, suggesting that ATM deficiency somewhat delays early DNA repair events, at least as monitored by the foci assay. In both HCT116 parental and HCT116-Chk2-/- cells, -H2AX foci detected 5 min post-IR (Figure 1d), which were numerically similar to those in LCL-N cells receiving an equal treatment, showed an almost 40% reduction at 1 h, a finding consistent with an apparently normal early DSB repair activity in these cells, regardless of Chk2 expression.

Collectively, these data demonstrate that IR and H₂O₂ generate a dose-dependent number of DSBs, detectable even at the lowest doses tested (0.25 Gy and 20 M H₂O₂), that are always accompanied by a preponderant amount of SSBs. Furthermore, in normal cells a first wave of DNA repair activity occurs within 1 h of the damage, resulting in almost 50% reduction of DSB-associated -H2AX foci. This rapid repair, dependent on ATM but not Chk2, can almost restore the basal conditions in cells treated with 0.25 Gy or 20 M H₂O₂.

IR dose-dependent activation of ATM

Exposure of cells to radiation triggers ATM kinase activity, enabling it to phosphorylate several substrates involved in multiple cell cycle checkpoints (Shiloh, 2003). We evaluated in LCL-N cells the in vitro and in vivo activities of ATM to determine its response to the initial amount of genomic damage. In vitro kinase assays were performed on cells harvested 30 min and 3 h after irradiation. At 30 min, the basal activity of ATM increased in an IR-dose-dependent manner starting from 0.25 Gy and reaching a maximum of 3-fold increase in response to 2 Gy or above (Figure 2a). However, at 3 h no ATM activity was seen after 0.25 or 0.5 Gy, whereas after 4 Gy this activity was still elevated (Figure 2a), hence suggesting that doses of IR between 0.25 and 0.5 Gy elicit a transient activation of ATM. To verify this possibility, we analysed the autophosphorylation of ATM Ser1981, an event reflecting the activation of ATM by DNA damage (Bakkenist and Kastan, 2003). Time-dependent analysis after 0.25 and 4 Gy (Figure 2b) showed that the ATM pS1981 signal, whereas after 0.25 Gy had increased 2.5-fold at 30 min to drop to basal level at 3 h, after 4.0 Gy the signal increase at 30 min was fivefold and only slightly dropped at 3 h (3.5-fold). Taken together with the in vitro kinase assays, these results indicate that the total activity of ATM is dependent on the yield of DNA lesion.

Figure 2.

IR dose-dependent activation of ATM. The in vitro catalytic activity of ATM immunoprecipitated from LCL-N cells 30 min and 3 h post-IR was tested against the PHAS-I recombinant substrate. (a) Following autoradiography, the reactions on gels were immunoblotted for ATM, to verify the amount of immunoprecipitated protein per sample. (b) The time-dependent autophosphorylation of ATM-Ser1981 in LCL-N cells in response to low-dose (0.25 Gy) and high-dose (4.0 Gy) IR was examined by immunoblotting using a phosphospecific antibody. -Actin normalized the protein loading per lane. The histogram at the bottom was obtained from the densitometric analysis of ATM phospho-Ser1981 signals after normalization for -actin (results are averages of two independent experiments)

Full figure and legend (57K)

As an indication of the ATM activity in vivo, we assessed the phosphorylation of its target residues Nbs1–Ser343, Chk2–Thr68 and p53–Ser15 (Shiloh, 2003). Nbs1–Ser343 and Chk2–Thr68 phosphoresidues, examined on cytospin preparations by immunofluorescence staining, showed in LCL-N a localization in nuclear foci, whose basal number increased starting from 0.25 Gy (Figure 3a and b). In AT cells, the basal number of phosphorylated Nbs1-S343 and Chk2-Thr68 foci did not increase even after 4 Gy IR (Figure 3a and b), concordant with an ATM dependence of these events. The specificity of Chk2–Thr68 phosphorylation was confirmed on immunoblots of Chk2 immunoprecipitated from untreated and treated cells (Figure 3c). In LCL-N cells, the levels of p53-P-Ser15 at 30 min rose modestly after 0.25–1 Gy, but substantially after 2–4 Gy (Figure 3d). The defective phosphorylation of p53-Ser15 in irradiated AT cells verified the ATM dependence of this event (Figure 3a and data not shown).

Figure 3.

IR-induced phosphorylation of p53, Nbs1 and Chk2. Nbs1–Ser343 (a) and Chk2–Thr68 (b) phosphoresidues were detected by immunofluorescence staining of cytospin preparations with phosphospecific antibodies in LCL-N and AT cells before (0) and 30 min after IR. The numbers of nuclear foci per cell in representative experiments are depicted in the histograms. The Thr68 phosphorylation (c) was detected on Chk2 immunoprecipitated from LCL-N cells 30 min after irradiation by immunoblot with a phosphospecific antibody (top). Blots were reprobed with an anti-Chk2 antibody, to verify the amount of immunoprecipitated Chk2 per lane (bottom). The phosphorylation of p53–Ser15 phosphorylation was analysed by immunoblot (d) in LCL-N and AT cells 30 min after irradiation. Reprobing for -actin verified lanes for protein content

Full figure and legend (263K)

Collectively, these findings demonstrate that, while ATM kinase is triggered by doses of IR as low as 0.25 Gy, its total activity is greatly enhanced by doses of IR >1 Gy, and thus in response to a substantial number of DSBs.

Chk2 phosphorylation and activation by IR

As the phosphorylation and activation of Chk2 are largely dependent on ATM (Brown et al., 1999; Matsuoka et al., 2000; Buscemi et al., 2001), we examined these events as function of the amount of DNA damage. IR dose-dependence studies showed that the phosphorylation-related upward shift of Chk2, just detectable 3 h after 0.5 Gy, was maximal after 2 Gy (Figure 4a) and involved most of the Chk2 molecules. A similar Chk2 mobility shift was seen 1 h after irradiation (data not shown). In vitro kinase assays on Chk2 immunoprecipitated from LCL-N cells to determine the correlation between mobility shift and enzymatic activity showed an enhancement of the basal activity of Chk2 towards Cdc25C substrate only in response to 1 Gy, but not to lower doses of IR(Figure 4b).

Figure 4.

Chk2 mobility shift and catalytic activity after IR. Immunoblots for Chk2 (a) were performed on exponentially growing LCL-N cells 3 h after exposure with escalating doses of IR. Note the progressive, phosphorylation-related upward shift of Chk2 becoming apparent after IR doses above 1 Gy. In vitro Chk2 kinase assays (b) were performed on the same cell preparations using a GST-Cdc25C recombinant substrate. The reactions were resolved by gel electrophoresis and, after autoradiography, immunoblotted for Chk2 to verify the amount of immunoprecipitated protein per sample (Chk2 mobility shift is not detectable using this electrophoresis condition). The autophosphorylation of Chk2–Thr387 (c) was detected on Western blot with a phosphospecific antibody on Chk2 immunoprecipitated from cells 45 min post-IR. The blot was reprobed with an anti-Chk2 to verify the amount of immunoprecipitated Chk2 per sample. The histogram below depicts the IR dose-dependent increase in Chk2–Thr387 phosphorylation, obtained by the densitometric analysis (s.d.) of three independent IP/WB experiments

Full figure and legend (98K)

As a measure of Chk2 activity in vivo, the autophosphorylation of Chk2-Thr387 (Schwarz et al., 2003) was evaluated on Chk2 immunoprecipitated from LCL-N, AT and NBS cells and immunoblotted with a phosphospecific antibody. In LCL-N (Figure 4c), a phospho-Thr387 signal became detectable after 0.5 Gy, but not 0.25 Gy, and markedly increased with the dose of IR (up to 14-fold after 4 Gy). Concordant with the ATM and Nbs1 dependence of this event, neither AT nor NBS cells showed a radiation-induced Chk2–Thr387 phosphorylation signal (Figure 4c). As an additional indicator of Chk2 activity in vivo, we evaluated the CDK inhibitor p21^waf1, whose radiation-induced expression depends on Chk2 (Hirao et al., 2000; Takai et al., 2002). In LCL-N cells, p21^waf1 protein was markedly induced at 3 h only in response to doses >1 Gy, whereas in AT this induction was defective (Figure 5a).

Figure 5.

IR dose-dependent p21^waf1 accumulation and S-phase checkpoint arrest. p21^waf1 protein level in LCL-N cells 3 h after treatment with the indicated doses of IR (a, top). Time-course analysis of p21^waf1 protein accumulation in normal and AT cells after 0.25 and 4 Gy (a, bottom). (b) The S-phase arrest was evaluated on cells radiolabelled with [¹⁴C]thymidine, irradiated and after 60 min pulse labelled for 15 min with [³H]thymidine, as described in Materials and methods. The ratios of [³H] to [¹⁴C] c.p.m. (corrected for those c.p.m. that were the results of channel crossover) were a measure of DNA synthesis. Data are averages (s.d.) of three independent experiments

Full figure and legend (76K)

As Chk2 mediates DNA synthesis arrest in response to DSBs via phosphorylation of Cdc25A (Falck et al., 2001), we analysed the IR dose-dependent occurrence of this event. In LCL-N cells (Figure 5b), the inhibition of DNA synthesis was apparent after 1 Gy, but not at lower doses of IR, and, in accordance with the ATM and Nbs1 dependence of this event, no DNA synthesis arrest was seen in AT52, whereas in NBS cells a partial inhibition was seen after 4 Gy.

Together, these findings demonstrate that Chk2 becomes activated only in response to a threshold number of DSBs, below which it remains inactive despite undergoing phosphorylation on Thr68 by ATM.

Chk2 response to hydrogen peroxide

The relationship between the DNA damage spectra and Chk2 responsiveness was further examined in cells exposed to sublethal concentrations of H₂O₂, an oxidant which primarily generates SSBs (Birnboim and Sandhu, 1997; Guidarelli et al., 1997; Figure 1a). In LCL-N, AT and NBS cells exposed to 20 M H₂O₂, the electrophoretic mobility of Chk2 remained unchanged for up to 2 h, relative to untreated cells (Figure 6a). Conversely, a time-dependent Chk2 mobility delay in response to 100 M H₂O₂ was seen in LCL-N, but not in AT or NBS, cells (Figure 6a). The extent of this phosphorylation-related shift, abolished by phosphatase treatment (data not shown), was similar to that seen after treatment with 4 Gy. The basal in vitro catalytic activity of Chk2 was unaffected by 20 M H₂O₂ in LCL-N, AT or NBS cells, whereas after treatment with 100 M H₂O₂ this activity rose in LCL-N cells only (Figure 6b). Both concentrations of H₂O₂ had an effect on ATM activity, as determined by in vitro kinase assays on LCL-N cells (Figure 6c), but the response was more marked after 100 M H₂O₂. Regarding p21^waf1, this protein was induced in LCL-N cells after treatment with 100 M, but not 20 M H₂O₂ (Figure 6d), and the ATM dependence of this event was underscored by the defective p21^waf1induction in AT cells (Figure 6d).

Figure 6.

Chk2 mobility shift and activation after H₂O₂ treatments. Cells were harvested before, 1 and 2 h after incubation with H₂O₂. (a) The immunoblot shows a phosphorylation-related upward shift of Chk2 only in normal LCL-N cells treated with 100 M H₂O₂. (b) Chk2 was immunoprecipitated and tested for kinase activity in vitro (see legend to Figure 2). The relative levels of enzymatic activity per lane, determined by densitometric analysis, are depicted. The effect of H₂O₂ on ATM activity (c) was examined in in vitro kinase assays on ATM immunoprecipitated from LCL-N cells harvested 1 h after treatment with the oxidant. A lysate from irradiated cells was included as a positive control. (d) Time-course analysis of p21^waf1 protein accumulation in normal and AT cells after 20 and 100 M H₂O₂

Full figure and legend (95K)

Together with the above data (Figure 1), these results demonstrate a DNA lesion-dependent activation of Chk2, whatever the nature of genotoxic agent, for example, IR or oxidative stress. Furthermore, Chk2 appears unresponsive to SSBs and nucleotide base modifications.

Discussion

Upon sensing genomic damage, cells elicit an integrated network of signalling pathways that transiently prevent cell cycle progression and allow DNA repair (Walworth, 2000). Among the different types of DNA lesions generated by cellular stresses or genotoxic agents, DSBs are the most harmful as they can lead to chromosomal breaks, rearrangements and tumor development (Khanna and Jackson, 2001). The ATM kinase is crucial for the initiation of signalling pathways in mammalian cells in response to DSBs (Shiloh, 2003). One of its effector targets is Chk2, a kinase that enforces checkpoint arrest at multiple cell cycle phases (Bartek and Lukas, 2003). While many underlying biochemical and functional features of Chk2 have been described in recent years, the precise nature of the DNA lesions that evoke the phosphorylative activation of Chk2 remain debated. Thus, whereas in lower organisms such as Xenopus or Caenorhabditis elegans Chk2 demonstrates a DSB response (Guo and Dunphy, 2000; MacQueen and Villeneuve, 2001), in mammalian cells Chk2 is responsive to UV or hydroxyurea (Brown et al., 1999; Tominaga et al., 1999), two agents that, while primarily causing pyrimidine dimers and stalled replication forks, respectively, in certain circumstances can induce DSBs (Golos and Malec, 1991). The involvement of Chk2 in DSBs response is also suggested by its ATM- and Nbs1-dependent activation (Buscemi et al., 2001). AT and NBS cells, which carry defects in these proteins, are hypersensitive to DSB-inducing agents (e.g. -radiation) (Huo et al., 1994; Kraakman-van der Zwet et al., 1999). Moreover, AT and NBS patients show an immunodeficiency due to inappropriate resolution of DSBs associated with V(D)J and class switch recombination intermediates (Petersen et al., 2001).

In this study, we have examined the relationship between the spectrum and yield of inflicted DNA lesions and induction of the ATM-Chk2 response to determine the initial threshold damage capable of triggering this pathway. In order to quantitate the genotoxic damage, the nuclear DNA was extracted from cells immediately after the appropriate treatment and analysed by AET, which detects SSBs and DSBs (Kohn et al., 1981). As the current methods for measuring DNA breaks, that is, comet assay and pulsed-field gel electrophoresis, all underestimate yields due to sensitivity limits and nonrandom distribution of generated lesions (Prise et al., 2001), we additionally assessed the -H2AX nuclear foci which rapidly accumulate at the sites of damage and which, in several studies, have been used as sensitive markers for DSBs (Rogakou et al., 1999; Petersen et al., 2001; Furuta et al., 2003; Jakob et al., 2003; Rothkamm and Lobrich, 2003). Combining the AET and the -H2AX foci results, we have precisely estimated the initial yield of genomic SSBs and DSBs generated by various treatments. Accordingly, we have shown that IR generates an extensive number of SSBs, even at the lowest dose tested, along with a more limited number of DSBs, accounting for 8 and 19 in response to 0.25 and 1.0 Gy, respectively.

As reactive oxygen species produced during mitochondrial respiration and peroxisomal metabolism are the major intracellular sources of DNA damage (Shackelford et al., 2000), we analysed the effect of increased intracellular free radicals by H₂O₂, an oxidant that primarily generates SSBs (Birnboim and Sandhu, 1997; Guidarelli et al., 1997). In normal cells, treatment with 20 and 100 M H₂O₂ generated, besides extended amounts of SSBs, a number of DSBs comparable with 0.25 and 1–2 Gy IR, respectively.

While the appearance of -H2AX foci is an early response to DSBs, their kinetics of disappearance closely resembles DSB repair (Rothkamm and Lobrich, 2003). By analysing -H2AX 1 h after irradiation with 0.25–1 Gy, we have shown that Chk2-deficient cells undergo a similar reduction in foci as parental cells, suggesting a dispensable role for Chk2, at least in the early phases of DNA repair following low doses of IR. However, in view of the functional link between Chk2 and Brca1 in DSB repair, demonstrated in studies using 2 Gy of IR and a different experimental system (Zhang et al., 2004), these findings would suggest that Chk2 might function in subpathways of the general DSB repair process.

To determine the relationship between the initial yield of DNA lesions and induction of ATM-dependent responses, we examined the activity of ATM as well as the phosphorylation of the target substrates Nbs1–Ser343, p53–Ser15 and Chk2–Thr68 (Abraham, 2001; Shiloh, 2003), the latter two involved in cell cycle arrest. We have shown, by in vitro kinase assays and analysis of the autophosphorylation status of ATM-Ser1981, that the ATM activity is induced even after 0.25 Gy, a dose that generates about eight DSBs per cell, underscoring the capacity of ATM to sense and respond to subtle DNA lesions. This finding agrees with recent results showing that as little as 0.1 Gy, estimated to generate four DSBs per diploid cell, can trigger the autophosphorylation of ATM on Ser1981 (Bakkenist and Kastan, 2003).

The activation of ATM by low doses of IR (0.25 Gy) was evidenced by the phosphorylation of the targets Nbs1–Ser343 and Chk2–Thr68. Noteworthy, in spite of the phosphorylation on Thr68 at 0.25 Gy, only doses of IR >1 Gy (causing >19 DSBs per cell) associated with additional modifications of Chk2, such as enhanced mobility shift and autophosphorylation on Thr387, were able to activate Chk2. Very recently, Ser33/35 and Ser516 of Chk2 have been shown to be sites of in vivo phosphorylation and autophosphorylation, respectively (Mochan et al., 2003; Wu and Chen, 2003), and, quite interestingly, both of these events occur in response to radiation doses higher than those necessary for Thr68 phosphorylation, altogether underscoring the multistep nature of Chk2 activation in vivo (Schwarz et al., 2003). With regard to the DNA damage dose dependency of Chk2 activation, it can be hypothesized that the cascade of phosphorylation steps, initiated by ATM, may depend on additional Chk2-targeting kinases such as Plk1 or Plk3 (Bahassi el et al., 2002; Tsvetkov et al., 2003), whose activity might be triggered by a higher threshold of DSBs than required to activate ATM.

Despite the marked yield of SSBs generated in normal cells by 20 M H₂O₂ or 0.25–0.5 Gy, neither the mobility shift nor the catalytic activity of Chk2 was affected, underscoring the unresponsiveness of Chk2 to SSBs. Whether the ATR/Chk1 pathway (Plumb et al., 1999) is involved in SSBs responses remains unknown.

Chk2 mediates checkpoint arrest at multiple cell cycle phases (Bartek and Lukas, 2003). To get a further insight into the functional activity of Chk2 in relation to the amount of DNA damage, we analysed the S-phase checkpoint. We have shown in normal cells a dose-dependent suppression of DNA replication in response to >1 Gy of IR, but not to lower doses (e.g. 0.25–0.5 Gy), which prevalently generate SSBs without activating Chk2. These results thus provide, on one hand, a correlation between Chk2 activation and enforcement of an S-phase checkpoint arrest; on the other hand, the evidence that the S-phase checkpoint is unresponsive to SSBs.

In summary, we have shown that the ATM-dependent Chk2 pathway is fully activated by DSBs, but not SSBs, and only when these lesions are present above a threshold level (>19 DSBs per cell), suggesting that limiting numbers of DSBs may be silently repaired without enforcing cell cycle checkpoint arrest, as recently found in yeast (Leroy et al., 2001). This would be consistent with our data evidencing an apparently normal early DNA repair-associated disappearance of -H2AX foci, regardless of Chk2 expression, at least after low-dose IR. Given the delay with which Chk2 is activated (relative to rapid activation of ATM), we cannot exclude that DSBs resistant to a first wave of repair activity (Nunez et al., 1995; Wang et al., 2001) could function as an activating signal for Chk2 pathway.

Materials and methods

Cell lines and treatments

The lymphoblastoid cell lines (LCLs) were established by Epstein–Barr virus (EBV) immortalization of peripheral blood from normal individuals (LCL-N), from an AT patient (AT52RM, also denoted AT52) (Delia et al., 2000) and from an NBS patient (GM07078, Coriell Cell Repository, Camden, NJ, USA). LCL cells were cultured at a concentration of 4–5 10⁵ cells/ml in RPMI 1640 medium (BioWittaker, Walkersville, MD, USA) supplemented with 15% heat-inactivated fetal calf serum (FCS), penicillin (100 U/ml) and streptomycin (100 g/ml). HCT116- and HCT116-Chk2-/--deficient cells (Jallepalli et al., 2003) were grown in McCoys medium supplemented with 10% FCS. Cells were -irradiated using a ¹³⁷Cs source emitting a dose rate of 8 Gy/min or exposed for 1 h to hydrogen peroxide (Sigma, Italy).

Alkaline elution assay for DNA SSB and DSB

Cells were labelled with 0.08 Ci/ml 2-[¹⁴C]thymidine for 24 h at 37°C. After removal of the radiolabelled thymidine, cells were incubated for a 24 h prior to treatment with hydrogen peroxide or irradiation (the latter performed on ice). For the filter elution procedure (Kohn et al., 1981; Capranico et al., 1987; Perego et al., 2000), the cells after genotoxic treatments were layered on 2 m pore polycarbonate filtres, lysed with a solution containing 2% SDS, 0.1 M glycine, 25 mM disodium EDTA (pH 10 for SSB, pH 9.6 for DSB) and 0.5 mg/ml proteinase K, and the DNA eluted with a solution containing 0.1% SDS, 20 mM EDTA (acid form) adjusted to pH 12.15 (for SSB) or 9.6 (for DSB) with tetrapropylammonium hydroxide. Fractions eluted during 15 h were collected and radioactivity measured by liquid scintillation. To allow a comparison between samples, the amount of DNA damage was converted into a FI based on the formula (1-r/r₀, where r and r₀ are the fractions of [¹⁴C]DNA remaining on filtres of treated and untreated cells, respectively), calculated after 12 h of elution.

Detection of nuclear foci

Lymphoblastoid cells were maintained in exponentially growing conditions at the concentration of 4–5 10⁵ cells/ml and were replenished with fresh medium 24 h before treatments. Following treatments, cells were deposited onto glass slides using a cytocentrifuge (Shandon), air dried and fixed in ice-cold methanol (20 min at -20°C) and then in acetone (2 min at -20°C). After washing with PBS and blocking for 1 h with 3% BSA, the slides were incubated for 2 h with 1 : 400 dilution of an anti-phospho-Ser139 H2AX antibody (clone JBW301, Upstate Biotechnology) and thereafter with an F(ab)₂ fragment of a FITC-conjugated secondary antibody for 1 h (Jackson Laboratories). After three washes in PBS, slides were counterstained with DAPI and mounted with an antifade solution. Nbs1 and Chk2 nuclear foci were detected as described (Ward et al., 2001), using phosphospecific antibodies against Nbs1–Ser343 (Gatei et al., 2000) and Ckh2–Thr68 (Ward et al., 2001). Images were collected using a Zeiss Axioskop fluorescence microscope and digital imaging. Nuclear foci were enumerated by two operators, on experiments performed three independent times and on duplicate slides.

Immunoblot analysis

Western blots were performed as described (Buscemi et al., 2001). The membranes were incubated with either a mouse monoclonal anti-Chk2 antibody made by us (clone 44D4/21) or a rabbit anti-Chk2 (Tominaga et al., 1999), p53–pSer15 (clone 16G8, Signal Transduction), p21^waf1 (Pharmingen BD, San Jose, CA, USA), -actin (Sigma, Italy) and ATM (clone 4D2 made by us; ref. 15), and subsequently with a peroxidase-conjugated secondary antibody. The immunoreactive bands were visualized by ECL Super Signal (Pierce, Rockford, IL, USA) on autoradiographic films, scanned and quantitated by ImageQuant software (Molecular Dynamics, Sunnyvale, CA, USA). The phosphospecific rabbit antibodies to Chk2–Thr68 and Chk2–Thr387 (Cell Signalling Tech., Beverly, MA, USA) were tested on Western blots of Chk2 immunoprecipitates, as described (Ward et al., 2001). The phosphospecific rabbit antibody to ATM Ser1981 was from Rockland Immunochemicals (Gilbertsville, PA, USA).

Radioresistant DNA synthesis assay

Radiation-induced DNA synthesis (RDS) inhibition, which reflects the function of the S-phase checkpoint, was evaluated as described (Xu et al., 2001). Briefly, cells were labelled for 24 h with 10 nCi/ml [¹⁴C]thymidine (Amersham Biosciences, UK), then washed and incubated for 6 h in nonradioactive medium. Thereafter, the cells were irradiated and immediately seeded in quadruplicates in 96-well plates and incubated for 60 min, thereafter pulse labelled for 15 min with 2 Ci [³H]thymidine, harvested and radioactivity analysed by scintillation counting. The resulting ratios of [³H] to [¹⁴C] c.p.m. (corrected for those c.p.m. that were the results of channel crossover) were a measure of DNA synthesis.

Immunoprecipitations and in vitro kinase reactions

For Chk2 assays, cells were lysed on ice for 30 min in a buffer containing 50 mM Tris–HCl, pH 7.4, 0.2% Triton X-100, 0.3% NP-40, 150 mM NaCl, 1 mM EDTA plus protease and phosphatase inhibitors (1 mM PMSF, 1 g/ml pepstatin, 2 g/ml leupeptin, 2 g/ml aprotinin, 25 mM NaF and 1 mM Na₃VO₄). Lysates were clarified by centrifugation, precleared with 10 l Sepharose–Protein A (Sigma) for 1 h and immunoprecipitated with 5 g of anti-Chk2 antibody and 10 l of Sepharose–Protein A at 4°C for 2 h. Kinase reactions were performed at 30°C for 30 min in 20 l reaction mixtures containing 50 mM HEPES, pH 8.0, 10 mM MgCl₂, 2.5 mM EDTA, 1 mM dithiothreitol, 10 M -glycerophosphate, 1 mM NaF, 0.1 mM Na₃VO₄, 0.1 mM PMSF, 10 M ATP, 20 Ci [-³²P]ATP and 1 g of GST-Cdc25C fragment as a substrate (Tominaga et al., 1999). The reaction products were separated on SDS–PAGE, autoradiographed and immunoblotted for Chk2 to verify the amount of immunoprecipitated protein per sample. ATM was extracted from cells lysed in 50 mM Tris–HCl, pH 7.4, 1% Tween-20, 0.2% NP-40, 150 mM NaCl, 1 mM EDTA plus protease and phosphatase inhibitors, using the monoclonal anti-ATM antibody clone 4D2, and kinase reactions performed in a buffer containing 20 mM HEPES, pH 8.0, 50 mM NaCl, 10 mM MgCl₂, 10 mM MnCl₂, 1 mM dithiothreitol, 20 M ATP, 10 Ci [-³²P]ATP and 1 g PHAS1 substrate.

References

1. Abraham RT. (2001). Genes Dev., 15, 2177–2196. | Article | PubMed | ISI | ChemPort |
2. Ahn J, Urist M and Prives C. (2003). J. Biol. Chem., 278, 20480–20489. | Article | PubMed | ISI | ChemPort |
3. Bahassi el M, Conn CW, Myer DL, Hennigan RF, McGowan CH, Sanchez Y and Stambrook PJ. (2002). Oncogene, 21, 6633–6640. | Article | PubMed | ChemPort |
4. Bakkenist CJ and Kastan MB. (2003). Nature, 421, 499–506. | Article | PubMed | ISI | ChemPort |
5. Bartek J and Lukas J. (2003). Cancer Cell, 3, 421–429. | Article | PubMed | ISI | ChemPort |
6. Birnboim HC and Sandhu JK. (1997). J. Cell. Biochem., 66, 219–228. | Article | PubMed | ISI | ChemPort |
7. Brodsky MH, Weinert BT, Tsang G, Rong YS, McGinnis NM, Golic KG, Rio DC and Rubin GM. (2004). Mol. Cell. Biol., 24, 1219–1231. | Article | PubMed | ISI | ChemPort |
8. Brosh Jr RM, Balajee AS, Selzer RR, Sunesen M, Proietti De Santis L and Bohr VA. (1999). Mol. Biol. Cell, 10, 3583–3594. | PubMed |
9. Brown AL, Lee CH, Schwarz JK, Mitiku N, Piwnica-Worms H and Chung JH. (1999). Proc. Natl. Acad. Sci. USA, 96, 3745–3750. | Article | PubMed | ChemPort |
10. Buscemi G, Savio C, Zannini L, Micciche' F, Masnada D, Nakanishi M, Tauchi H, Komatsu K, Mizutani S, Khanna KK, Chen P, Concannon P, Chessa L and Delia D. (2001). Mol. Cell. Biol., 21, 5214–5222. | Article | PubMed | ISI | ChemPort |
11. Capranico G, Riva A, Tinelli S, Dasdia T and Zunino F. (1987). Cancer Res., 47, 3752–3756. | PubMed |
12. Chehab NH, Malikzay A, Appel M and Halazonetis TD. (2000). Genes Dev., 14, 278–288. | PubMed | ISI | ChemPort |
13. D'Amours D and Jackson SP. (2002). Nat. Rev., 3, 317–327.
14. Delia D, Mizutani S, Panigone S, Tagliabue E, Fontanella E, Asada M, Yamada T, Taya Y, Prudente S, Saviozzi S, Frati L, Pierotti MA and Chessa L. (2000). Br. J. Cancer, 82, 1938–1945. | Article | PubMed | ISI | ChemPort |
15. Falck J, Mailand N, Syljuasen RG, Bartek J and Lukas J. (2001). Nature, 410, 842–847. | Article | PubMed | ISI | ChemPort |
16. Furuta T, Takemura H, Liao ZY, Aune GJ, Redon C, Sedelnikova OA, Pilch DR, Rogakou EP, Celeste A, Chen HT, Nussenzweig A, Aladjem MI, Bonner WM and Pommier Y. (2003). J. Biol. Chem., 278, 20303–20312. | Article | PubMed | ISI | ChemPort |
17. Gatei M, Young D, Cerosaletti KM, Desai-Mehta A, Spring K, Kozlov S, Lavin MF, Gatti RA, Concannon P and Khanna KK. (2000). Nat. Genet., 25, 115–119. | Article | PubMed | ISI | ChemPort |
18. Girard P-M, Riballo E, Begg AC, Waugh A and Jeggo PA. (2002). Oncogene, 21, 4191–4199. | Article | PubMed | ISI | ChemPort |
19. Golos B and Malec J. (1991). Neoplasma, 38, 559–564. | PubMed |
20. Guidarelli A, Cattabeni F and Cantoni O. (1997). Free Radic. Res., 26, 537–547. | PubMed |
21. Guo Z and Dunphy WG. (2000). Mol. Biol. Cell, 11, 1535–1546. | PubMed | ISI | ChemPort |
22. Hirao A, Cheung A, Duncan G, Girard PM, Elia AJ, Wakeham A, Okada H, Sarkissian T, Wong JA, Sakai T, Destanchina E, Bristow RG, Suda T, Lowe SW, Jeggo PA, Elledge SJ and Mak TW. (2002). Mol. Cell Biol., 22, 6521–6532. | Article | PubMed | ISI | ChemPort |
23. Hirao A, Kong Y-Y, Matsuoka S, Wakeham A, Ruland J, Yoshida H, Liu D, Elledge SJ and Mak TW. (2000). Science, 287, 1824–1827. | Article | PubMed | ISI | ChemPort |
24. Huo YK, Wang Z, Hong JH, Chessa L, McBride WH, Perlman SL and Gatti RA. (1994). Cancer Res., 54, 2544–2547. | PubMed |
25. Jakob B, Scholz M and Taucher-Scholz G. (2003). Radiat. Res., 159, 676–684. | PubMed |
26. Jallepalli PV, Lengauer C, Vogelstein B and Bunz F. (2003). J. Biol. Chem., 278, 20475–20479. | Article | PubMed | ISI | ChemPort |
27. Khanna KK and Jackson SP. (2001). Nat. Genet., 27, 247–254. | Article | PubMed | ISI | ChemPort |
28. Kohn KW, Ewig RAG, Erickson LC and Zwelling LA. (1981). Measurement of DNA Strand-breaks and Cross Links by Alkaline Elution in DNA Repair: A Laboratory Manual of Research Procedures Friedberg EC and Hanawalt PC (eds) Marcel-Dekker: New York, pp. 370–401.
29. Kraakman-van der Zwet M, Overkamp WJ, Friedl AA, Klein B, Verhaegh GW, Jaspers NG, Midro AT, Eckardt-Schupp F, Lohman PH and Zdzienicka MZ. (1999). Mutat. Res., 434, 17–27. | Article | PubMed | ISI | ChemPort |
30. Kuhne M, Riballo E, Rief N, Rothkamm K, Jeggo PA and Lobrich M. (2004). Cancer Res., 64, 500–508. | Article | PubMed | ISI |
31. Lavin MF and Khanna KK. (1999). Int. J. Radiat. Biol., 75, 1201–1214. | Article | PubMed | ISI | ChemPort |
32. Lee JH and Paull TT. (2004). Science, 304, 93–96. | Article | PubMed | ISI | ChemPort |
33. Leroy C, Mann C and Marsolier MC. (2001). EMBO J., 20, 2896–2906. | Article | PubMed | ISI | ChemPort |
34. MacQueen AJ and Villeneuve AM. (2001). Genes Dev., 15, 1674–1687. | Article | PubMed | ISI | ChemPort |
35. Matsuoka S, Rotman G, Ogawa A, Shiloh Y, Tamai K and Elledge SJ. (2000). Proc. Natl. Acad. Sci. USA, 97, 10389–10394. | Article | PubMed | ChemPort |
36. McDonald III ER and El-Deiry WS. (2001). Ann. Med., 33, 113–122. | PubMed | ISI | ChemPort |
37. Mirzayans R, Bashir S, Murray D and Paterson MC. (1999). Carcinogenesis, 20, 941–946. | Article | PubMed |
38. Mochan TA, Venere M, DiTullio Jr RA and Halazonetis TD. (2003). Cancer Res., 63, 8586–8591. | PubMed | ISI | ChemPort |
39. Nunez MI, Villalobos M, Olea N, Valenzuela MT, Pedraza V, McMillan TJ and Ruiz de Almodovar JM. (1995). Br. J. Cancer, 71, 311–316. | PubMed | ISI | ChemPort |
40. Perego P, Gatti L, Carenini N, Dal Bo L and Zunino F. (2000). Int. J. Cancer, 87, 343–348. | Article | PubMed |
41. Petersen S, Casellas R, Reina-San-Martin B, Chen HT, Difilippantonio MJ, Wilson PC, Hanitsch L, Celeste A, Muramatsu M, Pilch DR, Redon C, Ried T, Bonner WM, Honjo T, Nussenzweig MC and Nussenzweig A. (2001). Nature, 414, 660–665. | Article | PubMed | ISI | ChemPort |
42. Plumb MA, Smith GC, Cunniffe SM, Jackson SP and O'Neill P. (1999). Int. J. Radiat. Biol., 75, 553–561. | Article | PubMed | ISI | ChemPort |
43. Prise KM, Pinto M, Newman HC and Michael BD. (2001). Radiat. Res., 156, 572–576. | PubMed |
44. Rogakou EP, Boon C, Redon C and Bonner WM. (1999). J. Cell. Biol., 146, 905–916. | Article | PubMed | ISI | ChemPort |
45. Rothkamm K and Lobrich M. (2003). Proc. Natl. Acad. Sci. USA, 100, 5057–5062. | Article | PubMed | ChemPort |
46. Schwarz JK, Lovly CM and Piwnica-Worms H. (2003). Mol. Cancer Res., 1, 598–609. | PubMed | ISI | ChemPort |
47. Shackelford RE, Kaufmann WK and Paules RS. (2000). Free Radic. Biol. Med., 28, 1387–1404. | Article | PubMed | ISI | ChemPort |
48. Shieh SY, Ahn J, Tamai K, Taya Y and Prives C. (2000). Genes Dev., 14, 289–300. | PubMed | ISI | ChemPort |
49. Shiloh Y. (2003). Nat. Rev. Cancer, 3, 155–168. | Article | PubMed | ISI | ChemPort |
50. Stiff T, O'Driscoll M, Rief N, Iwabuchi K, Lobrich M and Jeggo PA. (2004). Cancer Res., 64, 2390–2396. | Article | PubMed | ISI | ChemPort |
51. Takai H, Naka K, Okada Y, Watanabe M, Harada N, Saito S, Anderson CW, Appella E, Nakanishi M, Suzuki H, Nagashima K, Sawa H, Ikeda K and Motoyama N. (2002). EMBO J., 21, 5195–5205. | Article | PubMed | ISI | ChemPort |
52. Taylor AM. (2001). Best Pract. Res. Clin. Haematol., 14, 631–644. | Article | PubMed | ISI | ChemPort |
53. Tominaga K, Morisaki H, Kaneko Y, Fujimoto A, Tanaka T, Ohtsubo M, Hirai M, Okayama H, Ikeda K and Nakanishi M. (1999). J. Biol. Chem., 274, 31463–31467. | Article | PubMed | ISI | ChemPort |
54. Tsvetkov L, Xu X, Li J and Stern DF. (2003). J. Biol. Chem., 278, 8468–8475. | Article | PubMed | ISI | ChemPort |
55. Uziel T, Lerenthal Y, Moyal L, Andegeko Y, Mittelman L and Shiloh Y. (2003). EMBO J., 22, 5612–5621. | Article | PubMed | ISI | ChemPort |
56. Vaziri C, Saxena S, Jeon Y, Lee C, Murata K, Machida Y, Wagle N, Hwang DS and Dutta AA. (2003). Mol. Cell, 11, 997–1008. | Article | PubMed | ISI | ChemPort |
57. Walworth NC. (2000). Curr. Opin. Cell. Biol., 12, 697–704. | Article | PubMed | ISI | ChemPort |
58. Wang H, Zeng ZC, Bui TA, Sonoda E, Takata M, Takeda S and Iliakis G. (2001). Oncogene, 20, 2212–2224. | Article | PubMed | ChemPort |
59. Ward IM, Wu X and Chen J. (2001). J. Biol. Chem., 276, 47755–47758. | Article | PubMed | ISI | ChemPort |
60. Wu X and Chen J. (2003). J. Biol. Chem., 278, 36163–36168. | Article | PubMed | ISI | ChemPort |
61. Xu B, Kim St and Kastan MB. (2001). Mol. Cell. Biol., 21, 3445–3450. | Article | PubMed | ISI | ChemPort |
62. Zhang J, Willers H, Feng Z, Ghosh JC, Kim S, Weaver DT, Chung JH, Powell SN and Xia F. (2004). Mol. Cell. Biol., 24, 708–718. | Article | PubMed | ISI | ChemPort |
63. Zhao S, Weng YC, Yuan SS, Lin YT, Hsu HC, Lin SC, Gerbino E, Song MH, Zdzienicka MZ, Gatti RA, Shay JW, Ziv Y, Shiloh Y and Lee EY. (2000). Nature, 405, 473–477. | Article | PubMed | ISI | ChemPort |

Acknowledgements

We thank Drs B Vogelstein and F Bunz, The Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Johns Hopkins University, Baltimore, Maryland, for kindly providing the HCT116-Chk2-/- cells. This work was financially supported by the Italian Telethon Foundation grant GP0205/01, the Italian Association for Cancer Research (AIRC), the National Research Council (CNR, grant CU03.00416) and the Italian Ministry of Health (Ricerca Finalizzata). GB is recipient of a fellowship of the Italian Foundation for Cancer Research (FIRC).