An overactivated ATR/CHK1 pathway is responsible for the prolonged G2 accumulation in irradiated AT cells.

Induction of checkpoint responses in G1, S, and G2 phases of the cell cycle after exposure of cells to ionizing radiation (IR) is essential for maintaining genomic integrity. Ataxia telangiectasia mutated (ATM) plays a key role in initiating this response in all three phases of the cell cycle. However, cells lacking functional ATM exhibit a prolonged G2 arrest after IR, suggesting regulation by an ATM-independent checkpoint response. The mechanism for this ataxia telangiectasia (AT)-independent G2-checkpoint response remains unknown. We report here that the G2 checkpoint in irradiated human AT cells derives from an overactivation of the ATR/CHK1 pathway. Chk1 small interfering RNA abolishes the IR-induced prolonged G2 checkpoint and radiosensitizes AT cells to killing. These results link the activation of ATR/CHK1 with the prolonged G2 arrest in AT cells and show that activation of this G2 checkpoint contributes to the survival of AT cells.

In response to ionizing radiation (IR), 1 proliferating cells slow their progress through the cell cycle by activating the DNA damage-induced checkpoints, G 1 , S, and G 2 phase checkpoints, believed to promote DNA repair and to benefit genomic integrity (1)(2)(3)(4). ATM, the protein product of the gene mutated in AT cells, is one such central signal kinase responding to IR (5,6). Many proteins serving as checkpoint effectors are phosphorylated and activated by ATM kinase. As such, p53, CHK2, BRCA1, and Rad17 are involved in G 1 and G 2 /M cell cycle arrests (7)(8)(9)(10)(11)(12)(13), while CHK2 and NBS/SMC1 are involved in the S checkpoint (14 -16). AT cells deficient in ATM function show an abnormality in all checkpoint responses including G 1 , S, and G 2 phase checkpoints and are sensitive to IR (17)(18)(19)(20), indicating the importance of the ATM-dependent checkpoint regulation in maintaining genetic integrity. Although multicheckpoint responses are deficient in AT cells, AT cells do show an IR-induced delayed S checkpoint (21) and a prolonged G 2 accumulation (19,(22)(23)(24), indicating activation of the ATM-independent checkpoint pathways. The mechanism by which the ATM-independent pathway regulates the IR-induced G 2 checkpoint in AT cells, however, remains unclear.
In addition to ATM, ATR is another important kinase that regulates the multi checkpoints after DNA damage (25). Expression of a dominant-negative ATR sensitizes mammalian cells to many different types of DNA damage and diminishes the IR-induced G 2 /M checkpoint (26,27), emphasizing the important roles of ATR in IR-induced checkpoint activation. The downstream substrate of ATR regulating the checkpoint is CHK1 (28 -31). CHK1 is involved in IR-induced S and G 2 checkpoints in mammalian cells (21,30,32). Although ATM and ATR have overlapping roles, they do have distinctive roles in the signaling pathways (25). While Atm Ϫ/Ϫ is viable in mammalian cells (33), Atr Ϫ/Ϫ and Chk1 Ϫ/Ϫ both are lethal in mammalian cells (30,34,35), indicating that the ATR/CHK1 pathway plays an essential and ATM-independent role in mammalian cells. Caffeine, a nonspecific inhibitor of ATM and ATR (36) sensitizes AT cells to IR-induced killing (37), suggesting that the target of caffeine in AT cells might be ATR which is critical for AT cell survival.
We show here that the prolonged G 2 checkpoint in irradiated human AT cells without ATM function correlates with the overactivated ATR/CHK1 pathway following IR. Like caffeine (a nonspecific inhibitor of ATR) or UCN-01 (a nonspecific inhibitor of CHK1), Chk1 siRNA also abolishes the IR-induced prolonged G 2 checkpoint and radiosensitize AT cells to killing. These results clearly demonstrate that an overactivated ATR/ CHK1 pathway is responsible for the IR-induced prolonged G 2 checkpoint in AT cells and that this checkpoint is important for maintaining AT cell survival.

EXPERIMENTAL PROCEDURES
Cell Lines, Chemical Treatment, and Irradiation-Both GM847 (ATM ϩ/ϩ ) and AT5BIVA (ATM Ϫ/Ϫ ) cells are transformed human fibroblasts. These cells were adapted to growing in Dulbecco's modified Eagle's medium supplemented with 10% iron-supplemented calf serum (Sigma). The incubations were at 37°C in an atmosphere of 5% CO 2 and 95% air. Caffeine (Sigma) or UCN-01 (NCI) was added to the culture 30 min before the cells were exposed to x-rays (310 kV, 10 mA, 2-mm aluminum filter) and remained in the culture after IR until the cells were collected.
One-and Two-parameter Flow Cytometry Assay-For propidium iodide (PI) one-parameter assay, cells were collected at different times following IR and stained with PI solution as described previously (38). For bromodeoxyuridine (BrdUrd, Sigma) and PI, a two-parameter assay, we followed the method described by McKay et al. (39) with a minor modification. Thirty M BrdUrd was added to the growth medium immediately following IR and was maintained in the medium until cells were collected (24 h following IR). The FITC-conjugated anti-BrdUrd antibody was purchased from Dako Co. After cell collection, the cells were incubated with the antibody according to the manufacturer's in-structions. The cells were assessed by a flow cytometer (Coulter Epics Elite) for PI (DNA content) and FITC (DNA synthesis) measurements.
ATR Kinase Activity Assay-ATR activity was examined with a chromatin-bound extract prepared as described previously (40). Briefly, cells were collected and washed in cold phosphate-buffered saline. Proteins were then extracted with cold 0.1% Triton X-100 in CSK buffer (10 mM PIPES, pH 6.8, 100 mM NaCl, 300 mM sucrose, 1 mM MgCl 2 , 1 mM EGTA, 1 mM dithiotheitol, 1 mM phenylmethylsulfonyl fluoride) for 20 min at 4°C. The sample was then pelleted by low speed centrifugation at 3,000 rpm for 5 min at 4°C. The supernatant was named fraction 1. These pellets were then re-extracted by incubating in CSK buffer and were collected by centrifugation at 3,000 rpm for 10 min at 4°C. This supernatant was named fraction 2. The final pellet fraction (containing chromatin-bound proteins) was solubilized in radioimmunopreciptation assay (RIPA) buffer (150 mM NaCl, 40 mM MOPS, pH 7.2, 1 mM EDTA, 1% Nonidet P-40, 1% sodium deoxycholate, 0.1% SDS) and was named fraction 3. For kinase assay, 500 g of fraction 3 was mixed with 2 g of ATR antibody (sc-1887, Santa Cruz Biotechnology, Inc.) in the presence of 20 l of a 50% (v/v) protein G-Sepharose slurry (Invitrogen) in 500 l of Buffer A (0.5% Nonidet P-40, 1 mM Na 3 VO 4 , 5 mM NaF, 0.2 mM phenylmethylsulfonyl fluoride in phosphate-buffered saline buffer) and gently rotated overnight at 4°C. Immune complexes were washed twice with Buffer A, then twice with Buffer B (10 mM HEPES, pH 8.0, 10 mM MgCl 2 , 10 mM MnCl 2 , 1 mM dithiothreitol). The kinase immunoprecipitate was incubated at 30°C for 30 min with 1 g of PHAS-1 (Stratagene) in 25 l of Buffer B containing 10 Ci of [␥-32 P]ATP. Samples were analyzed by 12% SDS-PAGE and the kinase activities determined by the incorporation of 32 P into PHAS-1 protein using a PhosphorImager.
CHK1 Kinase Activity Assay-Cell extracts were prepared for this purpose by using the NE-PER TM kit (Pierce) according to the manufacturer's instructions. The nuclear extracts (250 g) were then mixed with 1 g of CHK1 antibody (sc-7898, Santa Cruz Biotechnology, Inc.) in the presence of 10 l of a 50% (v/v) protein A-Sepharose slurry (Repligen). The following procedures are similar to those described previously (41), except that kinase buffer without NaCl was used.
CDC2 Phosphorylation and Kinase Activity Assay-The CDC2 (also called CDK1) phosphorylation and CDC2 kinase assay are similar to previous reports (40). Cell extracts were prepared using the NE-PER TM kit (Pierce) for the kinase assay according to the manufacturer's instructions. The nuclear extracts (250 g) were then mixed with 1 g of CDC2 antibody (sc-54, Santa Cruz Biotechnology, Inc.) in the presence of 10 l of a 50% (v/v) protein A-Sepharose slurry (Repligen) for the measurement of CDC2 activity as described before (40), except that kinase buffer without NaCl was used.
Colony-forming Assay-Cellular sensitivity to radiation was determined by the loss of colony-forming ability as described previously (38).
Transfection of Chk1 siRNA-The Chk1 siRNA was designed to specifically target the sequence of 127-147 from the start codon region of the human Chk1 mRNA (5Ј-AAGCGUGCCGUAGACUGUCCA-3Ј) (32). The siRNA was synthesized by Dharmacon Research Inc. The scramble duplex RNA (Dharmacon Research Inc.) was used as the control RNA. The RNAs were delivered to the cells by OLIGO-FECTAMINE TM (Invitrogen) according to the manufacturer's instructions. The cells were analyzed 36 h posttransfection.

TABLE I
Cell cycle distribution in ATM Ϫ/Ϫ and ATM ϩ/ϩ cells UCN-01 (100 nM) was added to the cell culture 30 min before IR. After the human fibroblast cells (ATM ϩ/ϩ and ATM Ϫ/Ϫ ) were irradiated (6 Gy), 30 M BrdUrd was immediately added to the culture medium. The cells were collected 24 h after IR as described under "Experimental Procedures." The amount of BrdUrd incorporation (FITC signal) was proportional to DNA content (PI signal) (Fig. 1B). The percentage of cell cycle distribution is derived from the results shown in Fig. 1 to IR (19,(22)(23)(24), suggesting that an overactivated G 2 checkpoint exists in irradiated AT cells. To test this hypothesis, we first examined whether this G 2 accumulation in irradiated AT cells was a reversible process. We treated AT cells with 4 mM caffeine (a nonspecific inhibitor of ATM and ATR) or with 100 nM UCN-01 (a nonspecific inhibitor of CHK1). We observed that both caffeine (data not shown) and UCN-01 clearly reduced the G 2 accumulation in AT cells (Fig. 1A), explaining that the inhibitor blocked the IR-induced G 2 checkpoint. To exclude the possibility that the inhibitor held the irradiated cells in S phase resulting in fewer cells entering G 2 phase, we observed BrdUrd and PI, the double-labeled signals in both IR and the inhibitortreated cells (Fig. 1B). The results ( Fig. 1 and Table I) showed that there is not much difference of S phase ration between the cells treated and those not treated with the inhibitor. This indicates that the prolonged G 2 accumulation in irradiated AT cells reflects a cellular active G 2 checkpoint that is sensitive to caffeine or to UCN-01.

Abolishing the Prolonged G 2 Accumulation with Kinase Inhibitors Sensitizes AT Cells to IR-induced
Killing-It is known that AT cells are very sensitive to IR-induced killing. However, it remains unknown whether the prolonged G 2 accumulation observed after irradiation contributes to the survival of AT cells. Checkpoint activation facilitates cell DNA repair; thus, prolonged G 2 accumulation should play a protective role for AT cell survival. Previous work provides hints that this might indeed be the case (37). To test this hypothesis, we examined the radiosensitivity in irradiated AT cells under conditions in which the IR-induced G 2 arrest was abolished. After the prolonged G 2 accumulation was abolished by caffeine or UCN-01, AT cells became much more sensitive to IR-induced killing (Fig.  2), suggesting that this prolonged G 2 accumulation in irradiated AT cells is important for cell survival.
A Highly Activated ATR/CHK1 Pathway Exists in Irradiated AT Cells-To test whether the caffeine-sensitive response in AT cells is regulated by the ATR pathway, we measured ATR activity in the wild-type and in AT cells. There was no difference in ATR activity between irradiated and control samples from both cytoplasmic and nuclear extracts (data not shown). However, ATR activity of the chromatin-bound fraction was higher in irradiated (12 h) than in non-irradiated cells for both cell lines (Fig. 3A), suggesting that this pool of ATR contains the protein activated in response to DNA damage. When compared with the non-irradiated controls, the level of ATR activity increased in irradiated AT cells more than in irradiated wild-type cells (Fig. 3B), indicating an overactivated ATR in irradiated AT cells.
The main downstream target of ATR for regulating the checkpoint is CHK1. CHK1 is an important G 2 checkpoint regulator in mammalian cells exposed to IR (32,38,40). Al-though CHK2 is also implicated in IR-induced G 2 checkpoint (42), the CHK2-regulated response depends on ATM kinase (10,42,43). Therefore, the prolonged G 2 accumulation in irradiated AT cells (ATM Ϫ/Ϫ ) is not likely to be regulated by this kinase. We next examined whether the CHK1 pathway was involved in the prolonged G 2 accumulation in irradiated AT cells. Although the phosphorylation of CHK1 in cells following IR is hard to detect by using one-dimensional gel electrophoresis (30), we observed more phosphorylated CHK1 in irradiated AT cells than the wild-type control cells by increasing the radiation dose (20 Gy) and by using the whole cell lyses (Fig.   FIG. 4. Chk1 siRNA abolishes the G 2 checkpoint response and sensitizes AT (ATM ؊/؊ ) cells to IR. A, the levels of CHK1 expression were measured with the extracts from either Chk1 siRNA or control RNA treated human fibroblast cells (ATM ϩ/ϩ and ATM Ϫ/Ϫ ) cells. Proliferating cell nuclear antigen (PCNA) signal was detected by the antibody (sc-56, Santa Cruz Biotechnology, Inc.) as the internal control. B, the treatments of Chk1 siRNA are as described under "Experimental Procedures." The cells were treated with Chk1 siRNA for 36 h then were irradiated (2 Gy). The cells were collected at 12 h after IR. The preparation and measurement of flow cytometric profiles of cell cycle distribution are the same as described in the legend to Fig. 1. C, as described under "Experimental Procedures," the cells were treated with Chk1 siRNA for 36 h and then were irradiated (2 Gy). After IR, the cells were trypsinized and plated to new dishes for colony formation with the medium. Data shown are the average from three independent experiments. 3A). Higher CHK1 kinase activities in both irradiated wildtype and AT cells as compared with the non-irradiated controls (Fig. 3A) were observed. The increased ratio of CHK1 activity in irradiated AT cells is much higher than that in irradiated wild-type control cells at 12 h following IR (6 Gy) (Fig. 3, A and  B), indicating that CHK1 is more activated in irradiated AT cells.
The G 2 checkpoint is believed to be mediated by an inhibition of the CDC25 phosphatase that activates the CDC2 kinase by removing inhibitory phosphates (Thr 14 and Tyr 15 ), thus allowing entry into mitosis (20,32,44). CHK1 could regulate IRinduced G 2 checkpoint by phosphorylating CDC25A, which results in CDC25A degradation (32). To examine whether activation of the CHK1 kinase was associated with CDC25A protein level changes in irradiated AT cells, we examined the CDC25A levels. The results (Fig. 3A) showed that less CDC25A protein was observed in AT cells than in the wild-type control cells following IR (6 Gy). We next measured CDC2 activity. The results are consistent with those of CHK1 activation (Fig. 3) and of G 2 arrest (Fig. 1). The CDC2 activities decreased in both wild-type and AT cells at 12 h after IR (6 Gy), but the changes were more apparent in AT cells (Fig. 3, A and B). These observations suggest that CDC25A and CDC2 are the downstream effectors of CHK1 in the regulation of the G 2 arrest in AT cells.
Chk1 siRNA Abolishes the Prolonged G 2 Accumulation and Radiosensitizes the Cells to IR-To confirm that the ATR/CHK1 pathway is responsible for the prolonged G 2 accumulation in irradiated AT cells, we examined the effects of Chk1 siRNA on this checkpoint response. The Chk1 siRNA specifically inhibited CHK1 expression in the transfected cells (Fig. 4A) and reduced the prolonged G 2 accumulation in irradiated AT cells (Fig. 4B). By using BrdUrd and PI, the double-labeled method, we observed the effects of Chk1 siRNA on BrdUrd incorporation. Similar with the UCN-01 results, Chk1 siRNA did not hold the irradiated cells in S phase (data not shown), which probably is also because of the role of CHK1 in abolishing the S checkpoint (21,32). These results provide direct evidence that the ATR/CHK1 pathway plays a key role in the prolonged G 2 arrest of irradiated AT cells.
To study the relationship between G 2 checkpoint response and radiosensitivity, we examined the radiosensitivity of AT cells after abolishing their G 2 checkpoint response by Chk1 siRNA. Chk1 siRNA radiosensitized both wild-type and AT cells, but the sensitization in AT cells is larger than that in wild-type cells (Fig. 4C), indicating that this component of the checkpoint response is more critical in AT cells than in wildtype cells. DISCUSSION The AT phenotype of prolonged G 2 accumulation following IR was reported by different groups several years ago (19,22,23), but the underlying mechanism remained unknown (24). Our results indicate for the first time that the prolonged G 2 accumulation in irradiated AT cells is regulated by the ATR/CHK1 pathway.
Although ATM and ATR are two of the most important DNA damage signal transducers in mammalian cells (4,17,25,(45)(46)(47)(48)(49), it was generally believed that ATM mainly responded to DNA double strand breaks (DSBs) induced by IR or chemical agents, and ATR mainly responded to other types of DNA damage induced by UV or chemical agents. Additional evidence now demonstrates that besides ATM, ATR is also a very important checkpoint regulator in IR-irradiated cells (21,26,40). Our results show that both ATR and its substrate, CHK1, are overactivated in AT cells following IR. Furthermore, the IRinduced, prolonged G 2 accumulation in AT cells is abolished by blocking the ATR/CHK1 pathway, indicating that the overac-tivated ATR/CHK1 pathway is responsible for the prolonged G 2 accumulation in irradiated AT cells.
Following IR, activation of the ATM pathway is observed almost immediately in mammalian cells (14,21,24,42,50). Activation of the ATR/CHK1 pathway is observed about 1 h later and reaches a maximum level at about the 3-h time point (21). This observation suggests that ATM and ATR regulate different pathways in response to the induction of DNA DSBs. ATM could be activated from a trans-acting process immediately by changes in the structure of chromatin induced by DSBs (50). Alternately, ATM could be activated from combination of a trans-acting (changes in the structure of chromatin) and a cis-acting (DNA-binding) process that either through ATM directly binding to DNA DSBs (51,52) or by other DNA-binding protein formed complexes with ATM (53), thus playing a role in the initiating stage of multicheckpoints following IR. The fact of ATM-regulated S and G 2 -M checkpoints in a dose (IR)-dependent manner is supported more by the model of combined trans-and cis-acting processes. ATR is also a DNA binding protein (54,55). In the absence of ATM, ATR may have a greater opportunity to interact directly with the damaged DNA induced by IR and cause the observed overactivation. However, we cannot exclude the possibility that activated ATM inhibits ATR activity and AT cells without ATM showing an overactivated ATR/CHK1 pathway following IR. These hypotheses require rigorous testing.
Two major DNA DSB repair pathways, non-homologous end joining (NHEJ) and homologous recombination repair (HRR), exist in mammalian cells. NHEJ is a very fast process and HRR is a relatively slow process. HRR is thought to occur mainly during S and G 2 phase (56), suggesting that it benefits from a checkpoint that holds the cells in these phases of the cell cycle. The function of ATM is linked with HRR (57), suggesting that the ATM-dependent checkpoint facilitates HRR. Our previous data suggest that NHEJ is a process independent of checkpoints but that HRR is a checkpoint-utilizing process in vertebrate cells (58,59). When the IR-induced prolonged G 2 accumulation is abolished, ATM Ϫ/Ϫ cells became much more sensitive to killing by IR, suggesting that the ATR/CHK1-dependent checkpoint enhances cell survival by facilitating HRR as well.
In summary, we show here that IR-induced prolonged G 2 accumulation in irradiated AT cells reflects an ATM-independent checkpoint regulated by the ATR/CHK1 pathway. We also show that AT cells become more sensitive to IR-induced killing when this checkpoint is abrogated, indicating that it is important for the survival of irradiated AT cells.