Cleavage-mediated Activation of Chk1 during Apoptosis*

The Chk1 kinase is highly conserved from yeast to humans and is well known to function in the cell cycle checkpoint induced by genotoxic or replication stress. The activation of Chk1 is achieved by ATR-dependent phosphorylation with the aid of additional factors. Robust genotoxic insults induce apoptosis instead of the cell cycle checkpoint, and some of the components in the ATR-Chk1 pathway are cleaved by active caspases, although it has been unclear whether the attenuation of the ATR-Chk1 pathway has some role in apoptosis induction. Here we show that Chk1 is activated by caspase-dependent cleavage when the cells undergo apoptosis. Treatment of chicken DT40 cells with various genotoxic agents, UV light, etoposide, or camptothecin induced Chk1 cleavage, which was inhibited by a pan-caspase inhibitor, benzyloxycarbonyl-VAD-fluoromethyl ketone. The cleavage of Chk1 was similarly observed in human Jurkat cells treated with a non-genotoxic apoptosis inducer, staurosporine. We have determined the cleavage site(s), Asp-299 in chicken and Asp-299 and Asp-351 in human cells. We further show that a truncated form of human Chk1 mimicking the N-terminal cleavage fragment (residues 1–299) possesses strikingly elevated kinase activity. Moreover, the ectopic expression of Chk1-(1–299) in human U2OS cells induces abnormal nuclear morphology with localized chromatin condensation and phosphorylation of histone H2AX. These results suggest that Chk1 is activated by caspase-mediated cleavage during apoptosis and might be implicated in enhancing apoptotic reactions rather than attenuating the ATR-Chk1 pathway.

The Chk1 kinase is highly conserved from yeast to humans and is well known to function in the cell cycle checkpoint induced by genotoxic or replication stress. The activation of Chk1 is achieved by ATR-dependent phosphorylation with the aid of additional factors. Robust genotoxic insults induce apoptosis instead of the cell cycle checkpoint, and some of the components in the ATR-Chk1 pathway are cleaved by active caspases, although it has been unclear whether the attenuation of the ATR-Chk1 pathway has some role in apoptosis induction. Here we show that Chk1 is activated by caspase-dependent cleavage when the cells undergo apoptosis. Treatment of chicken DT40 cells with various genotoxic agents, UV light, etoposide, or camptothecin induced Chk1 cleavage, which was inhibited by a pan-caspase inhibitor, benzyloxycarbonyl-VADfluoromethyl ketone. The cleavage of Chk1 was similarly observed in human Jurkat cells treated with a non-genotoxic apoptosis inducer, staurosporine. We have determined the cleavage site(s), Asp-299 in chicken and Asp-299 and Asp-351 in human cells. We further show that a truncated form of human Chk1 mimicking the N-terminal cleavage fragment (residues 1-299) possesses strikingly elevated kinase activity. Moreover, the ectopic expression of Chk1-(1-299) in human U2OS cells induces abnormal nuclear morphology with localized chromatin condensation and phosphorylation of histone H2AX. These results suggest that Chk1 is activated by caspase-mediated cleavage during apoptosis and might be implicated in enhancing apoptotic reactions rather than attenuating the ATR-Chk1 pathway.
Cell cycle checkpoint machinery is activated in response to DNA damage or inhibition of DNA synthesis and induces cell cycle arrest for enabling cells to repair the insults that may cause genomic instability (1,2). The checkpoint kinase Chk1 is a highly conserved serine-threonine kinase that functions as a major effector in the G 2 /M-phase DNA damage checkpoint and S-phase replication checkpoint (3). A major upstream kinase of Chk1 is ATR (ataxia-telangiectasia mutated and Rad3-related), which belongs to phosphoinositide 3-kinase-re-lated protein kinase family and forms a complex with ATRinteracting protein (ATRIP) 2 (4 -6). The ATR-Chk1 pathway also controls the timing of firing on early and late replication origins during normal S-phase (7).
The ATR-dependent activation of Chk1 requires additional factors including the Rad17-replication factor C complex, the Rad9-Hus1-Rad1 complex, TopBP1, and Claspin (8 -10). ATR-ATRIP and the Rad9-Hus1-Rad1 clamp are independently loaded onto single-stranded DNA tracts and recessed DNA ends generated by stalled DNA replication forks, with the aid of replication protein A and the clamp loader Rad17-replication factor C, respectively. TopBP1 contains BRCA1 C-terminal (BRCT) domains I through VIII and directly activates the ATR-ATRIP complex via the ATR-activating domain intervening between BRCT VI and VII (11). It has been shown recently that the BRCT I-II region of TopBP1 interacts with Rad9, and this interaction is necessary for the binding of ATR-ATRIP to TopBP1 (12,13).
The Chk1 kinase is activated by phosphorylation of at least two serine residues, Ser-317 and Ser-345, located in a Ser/Thr-Gln (SQ/TQ)-rich domain (14). The SQ/TQ domain is mapped between a highly conserved N-terminal kinase domain (residues 1-265) and a C-terminal domain with ill-defined function (15). The crystal structure of human Chk1 kinase domain (residues 1-289) revealed an open kinase conformation, and the truncated Chk1 without the C-terminal half exhibited more than 20-fold higher activity than full-length Chk1 (16). Katsuragi and Sagata (17) identified an autoinhibitory region (AIR) in the C-terminal ϳ85 amino acids of Xenopus Chk1, which largely overlaps with a bipartite and unusually long nuclear localization signal (NLS). The authors further suggested that the phosphorylation of Ser-317 and Ser-345 induces a conformational change of the AIR and reverses the autoinhibition.
The phosphorylation of Ser-345, but not Ser-317, is also required for proteolytic degradation of Chk1 following the treatment of the anti-cancer drug camptothecin (18). The Chk1 degradation is mediated by an ubiquitin/proteasome system containing Cul1 or Cul4A and was suggested to function in limiting the duration of Chk1 signaling induced by low-intensity replication stress. Furthermore, Claspin, a mediator protein required for Chk1 activation, is also degraded by an ubiquitin/ proteasome system in an SCF ␤TrCP -dependent manner (19,20).
The proteasome-dependent down-regulation of Claspin appears to be implicated in cell cycle-dependent fluctuation of its cellular level as well as recovery from genotoxic stress.
Robust genotoxic insults induce apoptosis rather than the cell cycle checkpoint. The implication of Chk1 in apoptosis remains to be elucidated. In the literature, some components in the ATR-Chk1 pathway are known to be cleaved by activated caspases during apoptosis. Human Rad9 is cleaved by caspase-3 after exposure to DNA-damaging agents or staurosporine (21). Cleavage-resistant Rad9 generated by site-directed mutagenesis appeared to protect the cells from death induced by DNAdamaging agents, suggesting a positive role of Rad9 cleavage in promoting apoptosis. Clarke et al. (22) have reported that Claspin is also cleaved by caspase-7 during the initiation of apoptosis, and a smaller C-terminal fragment has a dominant inhibitory effect on Chk1 phosphorylation. More recently, another group has shown that Claspin is cleaved into multiple fragments by caspase-3 and -7 and also degraded via the proteasome (23). The authors suggest the possibility that the down-regulation of Claspin by two different pathways promotes apoptosis. However, the biological roles of caspase-mediated cleavage of ATR-Chk1 components during apoptosis remain to be fully understood.
Here, we demonstrate that Chk1 is cleaved into two fragments by caspase during apoptosis in chicken DT40 and human Jurkat cells. The cleavage site(s) is Asp-299 in chicken and Asp-299 and Asp-351 in human cells, which are located between the N-terminal kinase domain and the C-terminal domain containing the AIR and NLS. We further show that a truncated form of human Chk1 mimicking the N-terminal cleavage fragment (residues 1-299) possesses ϳ8-fold higher kinase activity compared with full-length Chk1. Moreover, ectopic expression of Chk1-(1-299) in human U2OS cells induces locally condensed nuclear morphology as well as H2AX phosphorylation, suggesting a possible role of Chk1 cleavage in promoting apoptosis.
Human B-lymphocyte line Jurkat was cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum and gentamicin (Sigma) in a 5% CO 2 incubator at 37°C. To induce apoptosis, cells were treated with 1 M staurosporine (Sigma). Human osteosarcoma U2OS and kidney cell line HEK293T were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and gentamicin as described above.
The cells treated with various agents were collected by centrifugation and fixed in 70% ethanol at Ϫ20°C. After washing with FACS buffer (phosphate-buffered saline containing 0.1% bovine serum albumin and 2 mM EDTA), cells were incubated with FACS buffer containing 50 g/ml RNase A for 30 min at 37°C and stained with 25 g/ml propidium iodide. The samples were analyzed using FACSCalibur and Cell Quest Pro software (BD Biosciences).
Cloning of Chicken Chk1 cDNA and Construction of Expression Plasmids-Total RNA was isolated from wild-type DT40 cells using an RNeasy mini kit (Qiagen) and reverse-transcribed to cDNAs by ImProm-II reverse transcription system (Promega). A full-length chicken Chk1 (chChk1) cDNA was amplified by PCR, subcloned into the pEF6/Myc-His plasmid (Invitrogen), and verified by DNA sequencing (pEF6/chChk1-Myc-His). In addition, pEF6/chChk1-Myc-His was digested with KpnI and PmeI and subcloned into the pCMV-Myc plasmid (Clontech) to generate the expression construct of chChk1 tagged with Myc and Myc-His at the N and C termini, respectively (pCMV-Myc/chChk1-Myc-His).
Transfection and Immunoblotting-Exponentially growing DT40 or Jurkat cells (5 ϫ 10 6 ) were washed with phosphatebuffered saline and resuspended in 500 l of growth medium. Twenty g (DT40) or 50 g (Jurkat) of expression plasmids were added to the cell suspension and electroporated using a Gene Pulser apparatus (Bio-Rad) at 250 V and 950 microfarads. After incubation for 18 h, cells were treated with or without apoptosis-inducing agents and lysed in lysis buffer (10 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, and 1 mM phenylmethylsulfonyl fluoride). The cell lysates were clarified by centrifugation and used for immunoblot analysis. The antibodies used in this study are as follows: monoclonal anti-Chk1 (G-4) and polyclonal anti-c-Myc (A-14) (Santa Cruz Biotechnology), monoclonal anti-Myc (Clontech), polyclonal anti-␤-actin (Cell Signaling), and monoclonal anti-caspase-6 (Medical & Biological Laboratories). Monoclonal antibodies against DDB1 or glyceraldehyde-3-phosphate dehydrogenase were generated as described previously (24). The signals were visualized using SuperSignal West Femto maximum sensitivity substrate (Pierce) or Immobilon western chemiluminescent horseradish peroxidase substrate (Millipore) and a LAS1000 lumino-image analyzer (Fuji Film).

Apoptosis Induces Chk1 Truncation in Chicken DT40 Cells-
We have previously established a conditional DDB1 knockout DT40 clone (DDB1 Ϫ/Ϫ/Ϫ /hDDB1), which lacks all DDB1 alleles and contains human DDB1 transgene under control of the tetracycline-responsive element. The addition of doxycycline leads to complete loss of DDB1 and thereby induces a severe growth defect and subsequently cell death by apoptosis (24). During the analyses of various cellular factors under DDB1-depleted conditions, anti-Chk1 (G-4) monoclonal antibody revealed a time-dependent decrease of full-length Chk1 and concomitant appearance of a faster migrating band around 35 kDa following 96-and 120-h doxycycline treatment (Fig.  1A), when a significant population of DDB1 Ϫ/Ϫ/Ϫ /hDDB1 cells undergoes apoptosis (24). Another anti-Chk1 (FL-476) polyclonal antibody also detected this band (data not shown), indicating an intrinsic signal of Chk1. The ϳ35-kDa band was similarly observed in DDB1 Ϫ/Ϫ/Ϫ /hDDB1 cells treated with UV light (Fig. 1B), topoisomerase I poison camptothecin, and topoisomerase II poison etoposide (Fig. 1C) under doxycyclinefree conditions. These results suggest that the truncated form of Chk1 is generated during apoptosis caused by DDB1 depletion as well as genotoxic treatment.
To ascertain the relationship between Chk1 truncation and apoptosis, we treated wild-type DT40 cells with various concentrations of etoposide and analyzed them with Western blotting and flow cytometry (Fig. 2, A and B). Five M or higher concentrations of etoposide evoked Chk1 truncation and apoptotic sub-G 1 cells, whereas 1 M etoposide induced neither. Kinetic analysis using 50 M etoposide also revealed a similar  time course between the appearance of truncated Chk1 and sub-G 1 cells (Fig. 2, C and D). Moreover, pan-caspase inhibitor Z-VAD-fmk completely abolished Chk1 truncation as well as the increase of sub-G 1 population. Taken together, these results strongly indicate that genotoxic stress-induced Chk1 truncation is associated with apoptosis induction.
Chicken Chk1 Is Cleaved at Asp-299 by Active Caspase during Apoptosis-We next tried to understand the molecular basis of Chk1 truncation during apoptosis. Wild-type DT40 cells were electroporated with a plasmid expressing chicken Chk1 tagged with Myc and His 6 at the C terminus (chChk1-Myc-His) and treated with 50 M etoposide. The ectopic expression of tagged Chk1 was verified by immunoblotting with anti-Chk1 (G-4) and anti-Myc antibodies (Fig. 3A, lane 3). Upon etoposide treatment, anti-Chk1 (G-4) antibody showed a more intense signal of truncated Chk1, although this band is invisible with anti-Myc antibody (Fig. 3A, lane 2 versus lane 4). On the other hand, anti-Myc antibody, but not anti-Chk1 (G-4) antibody, detected new doublet bands around 20 kDa. The upper band seems to be a phosphorylated form of the lower band because -phosphatase treatment led to a single band (data not shown). These results strongly suggest that chicken Chk1 is cleaved at a single site in the process of apoptosis, generating ϳ35-kDa N-terminal and ϳ20-kDa C-terminal fragments. Consistent with this notion, double-tagged Chk1 containing an additional Myc tag at the N terminus (Myc-chChk1-Myc-His) exhibited a slightly larger product (ϳ38 kDa) following etoposide treat-ment, which is also detectable with anti-Myc antibody (Fig. 3A, lane 4  versus lane 6).
To test the possibility that Chk1 is cleaved by caspase(s), we performed an in vitro caspase assay using the lysate prepared from DT40 cells expressing Myc-chChk1-Myc-His. Following incubation with active recombinant caspase-3, -6, -7, or -8, only caspase-7 exhibited truncated Chk1 with the same mobility as the etoposide-induced Chk1 fragment in DT40 cells expressing Myc-chChk1-Myc-His (Fig. 3B, lane 4  versus lane 6). Furthermore, the caspase-7-induced Chk1 cleavage was totally abolished by the addition of Z-VAD-fmk to the reaction (Fig.  4C, lane 2 versus lane 4). We conclude that Chk1 is cleaved by active caspase, most likely caspase-7, during apoptosis.
We further tried to determine the cleavage site of Chk1 using site-directed mutagenesis (Fig. 4A). Based on the size of the cleavage product, we mutated three Asp sites (Asp-299, Asp-329, or Asp-336) to Ala individually in Myc-chChk1-Myc-His and tested for the etoposide-induced cleavage. As shown in Fig. 4B, the substitution of Asp-299 to Ala (D299A) resulted in no Chk1 cleavage, whereas other two mutants (D329A and D336A) showed a comparable cleavage with wild-type Chk1. Consistently, the lysate from DT40 expressing Myc-chChk1(D299A)-Myc-His showed no Chk1 cleavage after incubation with active recombinant caspase-7 (Fig. 4C, lane 2 versus lane 3). Furthermore, the ectopic expression of truncated Chk1, Myc-chChk1-(1-299), exhibited identical mobility with the etoposide-induced cleavage fragment of Myc-chChk1-Myc-His (data not shown). These results demonstrate that apoptosis-dependent Chk1 cleavage takes place at Asp-299 in chicken DT40 cells.
Human Chk1 Is Cleaved at Asp-299 and Asp-351 upon Apoptotic Stress-We asked whether Chk1 cleavage also occurs in human cells undergoing apoptosis. Human B-lymphocyte line Jurkat cells were treated with 1 M staurosporine, a nonselective protein kinase inhibitor, and apoptosis induction was verified by the cleavage of caspase-6 ( Fig. 5A) as well as the increase of the sub-G 1 population (Fig. 5B). Under these conditions, anti-Chk1 (G-4) antibody detected two truncated products (Fig. 5A, lanes 2 and 3), which were not observed in the presence of Z-VAD-fmk (lanes 4 and 5), indicating that Chk1 is also cleaved in apoptotic human cells. We similarly constructed plasmids expressing double-tagged human Chk1 (Myc-hChk1-Myc-His), wild-type, or D299A mutant and electroporated them into Jurkat cells. Following staurosporine treatment, wild-type Chk1 exhibited two cleavage fragments (Fig. 5C, lane  (Fig. 5A), whereas D299A mutant Chk1 showed only one product with lower mobility (Fig. 5C, lane 5), suggesting that a smaller band is a cleavage product at Asp-299. To identify the second cleavage site in human cells, we also constructed a plasmid expressing D351A mutant Chk1, based on the estimated size as well as the fact that Asp-351 is not conserved in chicken (Fig. 4A). As expected, Jurkat cells transiently expressing D351A mutant Chk1 showed only one product with higher mobility following staurosporine treatment (Fig. 5C, lane 6). Taken together, these results clearly indicate that Chk1 is cleaved at Asp-299 and/or Asp-351 in human cells undergoing apoptosis.

4) as seen in endogenous Chk1
The N-terminal Cleavage Fragment of Chk1 Exhibits Elevated Kinase Activity and Induces Abnormal Nuclear Morphology as Well as H2AX Phosphorylation-It has been reported previously that Xenopus Chk1 contains the AIR at the C terminus, which suppresses its N-terminal kinase activity (17). Interestingly, the common cleavage site Asp-299 is mapped between the N-terminal kinase domain and the C-terminal AIR (Fig. 4A), prompting us to compare the kinase activities between full-length Chk1 and truncated Chk1 mimicking the N-terminal cleavage fragment (residues 1-299). HEK293T cells were transfected with a plasmid expressing N-terminally Myc-tagged human Chk1: wild-type Myc-hChk1, kinasedead mutant Myc-hChk1(D130A), or truncated mutant Myc-hChk1-(1-299). The equal amounts of Chk1 were immunoprecipitated with anti-Myc antibody, and their kinase activities were measured in vitro using a Chk1 substrate (CHKtide) and 32 P-labeled ATP (Fig. 6A). As a negative control, the immunoprecipitant from HEK293T cells transfected with a vector alone exhibited only marginal radioactive incorporation into the CHKtide. In addition, the immunoprecipitant from wild-type Chk1 transfectant showed negligible radioactivity in the absence of CHKtide (data not shown). Under these conditions, wild-type Chk1, but not kinase-dead mutant Chk1(D130A), moderately phosphorylated the CHKtide substrate. Interestingly, truncated Chk1-(1-299) exhibited markedly higher kinase activity (ϳ8-fold), consistent with the observation by Katsuragi and Sagata (17).

DISCUSSION
Chk1 has critical roles in G 2 /M-and S-phase checkpoints induced by genotoxic or replication stress and activated via the phosphorylation at Ser-317 and Ser-345 by ATR kinase in concert with several mediator proteins. On the recovery from cell cycle checkpoints after removal of the stress, the activated ATR-Chk1 pathway is down-regulated by proteasomal degradation of its components such as Chk1 and Claspin (18 -20). The ATR-Chk1 signaling pathway is also attenuated when the cells undergo apoptosis. Rad9 and Claspin, essential for Chk1 phosphorylation/activation, are cleaved by caspases following apoptosis induction (21)(22)(23). Those cleavages appear to function in switching the cellular response from cell cycle arrest to apoptosis. Similarly, other proteins involved in checkpoint or DNA repair such as ATM, DNA-PKcs, Bub1, poly(ADP-ribose) polymerase, and Rad51 are known to be substrates for active caspases during apoptosis (25,26). On the other hand, Rad9 cleavage has been suggested to have an additional and more positive role in the apoptosis pathway. The N-terminal cleavage fragment of Rad9 is localized in the cytoplasm, instead of nuclear localization of full-length Rad9, and binds to Bcl-XL, thereby promoting apoptosis (21).
In this study, we found that Chk1 is cleaved during apoptosis in chicken as well as human cells. The Chk1 cleavage was observed in apoptotic cells induced by DDB1 depletion or genotoxic (UV light, camptothecin, and etoposide) and nongenotoxic (staurosporine) treatments (Figs. 1, 2, and 5), suggesting a general response during apoptosis. While we were conducting this study, another group observed a similar truncated product of Chk1 in HeLa cells treated with etoposide or cisplatin (27). We further demonstrated that the apoptosis-dependent Chk1 cleavage is mediated by active caspase, most likely caspase-7 (Fig. 3), and Asp-299 is the common cleavage site in chicken (SDTD-F) and human (SNLD-F) cells, although Asp-351 (TCPD-H) is also cleaved in human cells (Figs. 4 and 5). Asp-299 is highly conserved among different species including human, mouse, rat, and chicken, whereas Asp-351 is conserved only in human and rat (Fig. 4A), suggesting that Asp-299 is a more general target for active caspase.
Intriguingly, Asp-299 and Asp-351 residues are located between the highly conserved N-terminal kinase domain (residues 1-265) and the C-terminal domain containing the AIR and NLS (15)(16)(17). Therefore, the cleavage of Chk1 at Asp-299 or Asp-351 divides full-length Chk1 into two fragments, one containing the kinase domain and another containing the AIR and NLS (Fig. 4A). We showed that truncated Chk1 (residues 1-299) mimicking the N-terminal cleavage fragment exhibited ϳ8-fold higher activity compared with full-length Chk1 (Fig.  6A). Chen et al. (16) reported that C-terminally truncated Chk1 (residues 1-265 or 1-289) is Ͼ20-fold more active than fulllength Chk1. These results suggest that the N-terminal cleavage fragment of Chk1 may have a function in phosphorylating some known or unknown substrates during apoptosis. Identification of the substrates for truncated Chk1 would be helpful for understanding the role of Chk1 cleavage in apoptosis.
Because both cleavage sites, Asp-299 and Asp-351, are near and within the SQ/TQ-rich domain (Fig. 4A), respectively, it is reasonable to ask whether the phosphorylation status affects the caspase-mediated Chk1 cleavage. A human Chk1 phosphorylation mutant (S317A/S345A) transiently expressed in Jurkat cells exhibited a similar level of cleavage at both Asp-299 and Asp-351 sites following staurosporine treatment (data not shown), indicating that the phosphorylation of at least Ser-317 and Ser-345 has neither stimulatory nor inhibitory effects on the caspase-dependent Chk1 cleavage. Consistently, caffeine, a phosphoinositide 3-kinase-related protein kinase inhibitor, showed no significant effects on Chk1 cleavage in DT40 cells treated with etoposide (data not shown).
Apoptotic DNA fragmentation also induces H2AX phosphorylation (31). Our data suggest that the N-terminal Chk1 fragment may induce DNA damage, probably double-strand breaks, enhancing apoptotic signals, or promote some apoptotic reactions thereby causing DNA fragmentation. Further study is required to clarify the mechanism of H2AX phosphorylation caused by the N-terminal Chk1 fragment and uncover a possible function of Chk1 cleavage in apoptosis induction. In summary, this study provides a new finding that Chk1 is activated by caspase-dependent cleavage during apoptosis and raises the possibility that the resultant N-terminal fragment may play a role in promoting apoptotic reactions through phosphorylating some substrates by its highly elevated kinase activity.