Differential Roles of Checkpoint Kinase 1, Checkpoint Kinase 2, and Mitogen-Activated Protein Kinase–Activated Protein Kinase

1935

Differential roles of checkpoint kinase 1, checkpoint kinase 2, and mitogen-activated protein kinase–activated protein kinase 2 in mediating DNA damage–induced cell cycle arrest: implications for cancer therapy

Zhan Xiao, John Xue, Thomas J. Sowin, MK2-deficient cells, Cdc25A protein, which is critically and Haiying Zhang required for the mitotic progression following checkpoint abrogation, becomes greatly depleted. In summary, our Cancer Research, Abbott Laboratories, Abbott Park, Illinois findings show that Chk1 is the only relevant checkpoint kinase as a cancer drug target and inhibition of other Abstract checkpoint kinases in addition to Chk1 would be nonpro- ductive. [Mol Cancer Ther 2006;5(8):1935–43] Mammalian cells initiate cell cycle arrest at different phases of the cell cycle in response to various forms of genotoxic stress to allow time for DNA repair, and thus Introduction preserving their genomic integrity. The protein kinases DNA-targeted agents are among the most effective in checkpoint kinase 1 (Chk1), checkpoint kinase 2 (Chk2), clinical cancer therapy and constitute the cornerstones of and mitogen-activated protein kinase–activated protein modern cancer treatment. These agents can be divided into kinase 2 (MK2) have all been shown to be involved in cell four main classes: alkylating agents, antimetabolites, top- cycle checkpoint control. Recently, cell cycle checkpoint oisomerase inhibitors, and radiomimetics. Their exact abrogation has been proposed as one way to sensitize mechanisms of action vary greatly; however, a common cancer cells to DNA-damaging agents due to the expected theme is that they all confer DNA damage, directly or induction of mitotic catastrophe. Due to their overlapping indirectly, and induce cell cycle checkpoints. Despite their substrate spectra and redundant functions, it is still not effectiveness at elevating the overall survival rates of cancer clear which kinase is mainly responsible for the cell cycle patients, they have serious intrinsic limitations such as arrests conferred by clinically relevant chemotherapeutics. widespread tumor-resistance and severe toxicity in normal Thus, the issue remains about which kinase is the most tissues, resulting in a narrowtherapeutic window(1). therapeutically relevant target and, more importantly, Mammalian cells have established highly elaborate sur- whether multiple kinases might need to be targeted to veillance systems to detect DNA damages and other forms of achieve the best efficacy in light of recent studies showing genotoxic stress, which is essential to maintain the genomic superior efficacy for pan-receptor tyrosine kinase inhib- integrity and, hence, cellular viability. When damage is itors. To clarify this issue, we investigated the roles of the detected, cells activate sophisticated pathways, called cell three kinases in response to different genotoxic stresses cycle checkpoints, to arrest cells in different phases of the cell through small interfering RNA–mediated specific target cycle to allowsufficient time for DNA repair. In normal cells, knockdowns. Our result showed that only the down- checkpoint responses are a critical safeguard to prevent regulation of Chk1, but not of Chk2 or MK2, abrogated tumorigenesis promoted by genetic instability; however, camptothecin- or 5-fluorouracil–induced S-phase arrest or in tumor cells, checkpoints constitute a major mechanism doxorubicin-induced G2-phase arrest. This was followed of resistance to chemotherapeutic drugs that damage DNA by mitotic catastrophe and apoptosis. Moreover, double because they reduce the effects of these drugs (2, 3). inhibition of Chk1 and Chk2 failed to achieve better It has previously been postulated that targeting the efficacy than Chk1 inhibition alone; surprisingly, inhibition cellular checkpoint pathway could be an attractive ap- of MK2, in addition to Chk1 suppression, partially reversed proach to circumvent the cancer resistance as discussed the checkpoint abrogation and negated mitotic catastro- above. The rationale is that in normal cells, DNA damage phe. We further showed that this is due to the fact that in would arrest the cells mostly in G1 in a p53-dependent manner, whereas p53-deficient tumors, accounting for over half of all tumor-types, have to rely on the checkpoint Received 2/9/06; revised 4/11/06; accepted 5/31/06. mediators to arrest cells at S or G2-M checkpoint. Therefore, The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked checkpoint inhibition would only abrogate the cell cycle advertisement in accordance with 18 U.S.C. Section 1734 solelyto blocks in tumor cells to induce mitotic catastrophe and indicate this fact. apoptosis, but mostly spare the normal cells. This may offer Requests for reprints: Haiying Zhang, Cancer Research, Abbott a potential feasible therapeutic window (4, 5). Laboratories, 100 Abbott Park Road, Abbott Park, IL 60064-6101. Phone: 847-938-4857; Fax: 847-935-7551. Two structurally unrelated but functionally similar E-mail: [email protected] protein serine/threonine kinases, checkpoint kinase 1 Copyright C 2006 American Association for Cancer Research. (Chk1) and checkpoint kinase 2 (Chk2), have emerged as doi:10.1158/1535-7163.MCT-06-0077 the major mediators of cell cycle checkpoints in response

Mol Cancer Ther 2006;5(8). August 2006

to genotoxic stress. Homozygous Chk1 knockout is lethal in Mitogen-activated protein kinase–activated protein ki- mouse embryonic stem cells. The mouse embryonic stem nase-2 (MK2) was discovered as a protein kinase activated cells that conditionally lack the Chk1 gene cannot prevent by extracellular signal–regulated kinase 1/2 from rabbit mitotic entry in response to ionizing radiation, showing skeletal muscle. Extracellular signal–regulated kinase 1/2 that Chk1 is required for the G2-M checkpoint (6). In adult and p38 were first reported to activate MK2 and MK3 somatic cells, Chk1 deficiency does not result in lethality in vitro, but it was later found that extracellular signal– and cells are viable and display normal cell cycle profiles regulated kinase 1 and 2 are not physiologic kinases for (7, 8). Complete deficiency of Chk1 in avian DT-40 MK2. Rather, MK2 activity is potently stimulated by h lymphoma cells abolished DNA damage–induced G2 various activators of p38 and p38 (19, 20). The targeted arrest and undermined S-phase checkpoint in response to deletion of the MK2 gene in mice provided the unexpected replication stress (8). Small interfering RNA (siRNA)– result that although p38 mediates the activation of many mediated knockdown of Chk1 in various human cancer cell similar kinases, MK2 seems to be the key kinase responsible lines also revealed an essential role of this kinase in both S for p38-dependent biological processes involving cytokine and G2-M DNA damage checkpoints (9, 10). Unlike Chk1, synthesis (21). Loss of MK2 leads to a defect in lipopoly- Chk2 knockout mice are viable and fertile (11). Although saccharide-induced synthesis of cytokines such as TNF-a structurally distinct from Chk1, Chk2 shares overlapping and g-IFN. Consistent with a role for MK2 in inflammatory substrate specificity with Chk1 and can phosphorylate responses, MK2-deficient mice showincreased susceptibil- critical Chk1 substrates, such as Cdc25A and Cdc25C, ity to infection (22). A recent finding has added new both in vitro and in vivo. Additionally, Chk2 is rapidly complexities to the above paradigm. Manke et al. showed phosphorylated and activated following exposure to that MK2 recognizes the same phosphorylation sites on various DNA-damaging agents such as ionizing radiation Cdc25B/C as Chk1/Chk2. More importantly, MK2 is or topoisomerase inhibitors. This indicated that Chk2 also directly responsible for Cdc25B/C phosphorylation and plays a role in cell cycle checkpoints (12). Consistently, their subsequent 14-3-3 binding in response to UV-induced studies with dominant-negative Chk2, siRNA-mediated DNA damage in mammalian cells. Down-regulation of Chk2 ablation, or intrinsic cellular Chk2 deficiency have all MK2 eliminates DNA damage–induced G2-M and intra-S confirmed a role of Chk2 in the S and G2 checkpoints in phase checkpoints. Therefore, it was proposed that MK2 response to double-strand breaks in various immortalized is a newmember of the DNA damage checkpoint kinase human cell types (13–15). In contrast to these reports family that functions in parallel with Chk1 and Chk2 to showing that Chk2 is required for checkpoint induction in integrate DNA damage signaling responses and cell cycle human cell lines, murine fibroblasts with Chk2 deletion arrest in mammalian cells (23, 24). showed no significant defects in the S-phase checkpoints, Judging from these results, it is still far from clear which indicating that it is not a major checkpoint mediator in checkpoint kinase is the major mediator of cell cycle arrest mice (11, 16, 17). This discrepancy was attributed to the in response to various DNA-damaging agents, especially differential requirements of Chk2 between human and the clinically relevant ones. It is possible that different murine systems or to the slowproliferation rate of murine forms of DNA damage require different checkpoint fibroblasts versus the fast growth rate of immortalized kinases to enforce the arrest. This issue will be critical human cell lines (12). for the development of small-molecule inhibitors targeting In the current paradigm of S or G2 checkpoint pathway, the checkpoint kinase(s). Additionally, it remains un- genotoxic stress activates ataxia telangiectasia mutated known whether inhibition of multiple checkpoint kinases or ataxia telangiectasia mutated and Rad3-related, which may confer better efficacy in potentiation of chemo- phosphorylate and activate both Chk1 and Chk2 (5, 12). therapeutics. In comparison, it has been well established Key substrates for these two checkpoint kinases are the that, targeting multiple receptor tyrosine kinases, a pan- Cdc25A, B, and C tyrosine phosphatases, which regulate receptor tyrosine kinase inhibitor confers better efficacy the timely activation of cyclin-dependent kinases at the in both antiangiogenesis and overall tumor regression G1-S and G2-M transitions. For agents that induce S-phase (25–28). Here we attempted to clarify these pressing arrest, Chk1/Chk2 target Cdc25A to proteolysis through issues with three frequently used chemotherapeutics the ubiquitin pathway, resulting in the suppression of including the topoisomerase-I inhibitor camptothecin, the cyclin-dependent kinase-2 activity and, hence, S-phase topoisomerase-II inhibitor doxorubicin, or the antimetab- arrest. Agents that confer G2-M arrest also activate Chk1/ olite 5-fluorouracil (5-FU). Using siRNA technology, we Chk2 to target Cdc25A to degradation. Additionally, they selectively down-regulated each of the checkpoint kinases induce the cytoplasmic sequestration of Cdc25B and singly or in various combinations and examined the Cdc25C through the 14-3-3 protein, eventually leading effects on checkpoint abrogation, mitotic progression, and to G2 arrest due to the failure to activate the mitotic cell survival. Our result showthat only Chk1 inhibition is cyclin-dependent kinase 1/Cdc2 kinase (5). Therefore, it is required for significant efficacy whereas knockdown of reasonable to conclude that simultaneous inhibition of other kinases in addition to Chk1 not only fails to deliver both Chk1 and Chk2 should be necessary to fully abrogate better efficacy but also, in some cases, attenuates or the checkpoint and, hence, confer maximal efficacy of eliminates the potentiation effect seen with Chk1 inhibi- potentiation (12, 18). tion alone.

Mol Cancer Ther 2006;5(8). August 2006

Materials and Methods Colony Formation Assay Cell Culture H1299 cells in six-well plates were transfected with the Human cervical cancer cell line HeLa and human non– indicated siRNA (or combinations of siRNAs). Twenty- small-cell lung cancer cell line H1299 were obtained from four hours posttransfection, the cells were trypsinized and American Type Culture Collection (Manassas, VA). H1299 split into new six-well plates at 500 per well. After cells were grown in RPMI 1640 supplemented with 10% fetal attachment overnight, the cells were treated with either bovine serum, 1 mmol/L sodium pyruvate, 1% penicillin- DMSO (control) or lowdoses of camptothecin for 9 days. j The formed colonies were then fixed and stained with streptomycin, and 0.45% glucose at 37 C in a 5% CO2 incubator. HeLa cells were grown in DMEM supplemented methylene blue. Total colony formation (considering both with 10% fetal bovine serum, 1 mmol/L sodium pyruvate, surface area and staining intensity) was scanned as pixel 1% penicillin-streptomycin, and 0.45% glucose at 37jCina counts and calculated with Image-Pro program. Caspase Assay 5% CO2 incubator. Before transfection, the cells were switched to medium without any penicillin-streptomycin. Fluorometric assay for caspase-3 was done essentially Antibodies according to the recommendations of the manufacturer Antibodies against Chk1, Chk2, and Cdc2 Y15P were (Roche Applied Sciences, Nutley, NJ) in a 96-well plate. purchased from Santa Cruz Biotechnology (Santa Cruz, HeLa cells transfected with the indicated siRNAs were CA). Antibody against MK2 was purchased from Cell treated or not with 150 nmol/L of camptothecin for 32 hours Signaling Technology (Beverly, MA). Antibodies for and assayed for caspase activity. Excitation was at 400 nm phospho-histone H3 (P-H3) and phospho-histone H2AX and fluorescence emission was detected at 505 nm. Caspase- (P-H2AX) were obtained from Upstate Technology 3 activity was calculated as the differential emission at (Waltham, MA). 505 nm between treated sample and blank buffer only. Tr a n s f e c t i o n Cell Cycle Analysis Human Chk1 siRNA and control luciferase siRNA were Cells undergoing the indicated treatments were washed as previously described and obtained from Dharmacon once in PBS and fixed in 70% ethanol. The fixed cells were washed again twice with PBS and treated with RNase A at Technology (Lafayette, CO; ref. 29). siRNA transfection j protocols were also as previously described (29). HeLa cells 37 C for 30 minutes. Finally, the cells were stained with were transfected with Oligofectamine whereas H1299 cells propidium iodide and incubated in the dark for 60 minutes were transfected with Lipofectamine-2000 (Invitrogen, or overnight before analysis. The samples were analyzed by Carlsbad, CA). The final concentration of each siRNA was flowcytometry witha fluorescence-activated cell sorter 50 nmol/L. manufactured by BD Bioscience (San Jose, CA) using the Western Blot Analysis CellQuest program. Adherent cells in the well were rinsed with PBS and directly lysed in Laemmli sample buffer. Floating cells or Results cell fragments in the same well were collected, lysed, and To clarify the unresolved issue of which checkpoint kinase combined with the above lysates. Samples were heated at represents the best cancer target for sensitization of tumor j 95 C for 5 minutes and resolved on the Novex mini-gel cells to various clinically relevant DNA-damaging agents, system (Invitrogen) under denaturing conditions and and additionally, whether inhibition of multiple kinases blotted to polyvinylidene difluoride membrane using a confers superior efficacy than inhibition of a single kinase, semidry transfer device (Amersham Biosciences, Piscat- we transfected HeLa cells with siRNAs that specifically away, NJ). The membrane was blocked with 5% nonfat dry targeted Chk1, Chk2 or MK2, either singly or in various milk and probed with various antibodies. Enhanced combinations. We first investigated the morphologic chemiluminescent detection was done with enhanced profiles of HeLa cells undergoing the various knockdowns, chemiluminescence reagents according to the protocols of treated with the topoisomerase-I inhibitor camptothecin at the vendor (Santa Cruz Biotechnology). 150 nmol/L or with the topoisomerase II inhibitor Cell Proliferation Assay [3-(4,5-Dimethyl-Thiazol- doxorubicin at 150 nmol/L for 1 day (Fig. 1). The doses 2yl)-5-(3-Carboxymethoxyphenyl)-2-(4-Sulfophenyl)- f correspond to 25% of the cytotoxic EC50 values of the 2H-Tetrazolium Assay] drugs in cell proliferation assay so that they would not HeLa cells were seeded in 96-well plates and transfected confer extensive cell death by themselves, making it easier with the indicated siRNA. Eight hours after transfection, the to detect the potentiation effect of the siRNAs. cells were treated with the indicated DNA-damaging agents IntheabsenceoftheDNA-damagingagents,the for 48 hours. After treatment, 3-(4,5-dimethyl-thiazol-2yl)- knockdown of the various checkpoint kinases, either by 5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazo- themselves or in different combinations, did not confer lium (MTS) reagent, which measures the amount of live obvious signs of cell death or other abnormalities. In cells (Promega, Madison, WI), was added to the cells and contrast, in the presence of camptothecin or doxorubicin, allowed to develop for 20 minutes to 2 hours. Colorimetric we observed dramatically different outcomes among the measurement was taken at 490 nm on Spectra MAX 190 different knockdowns. Because we used a relatively low from Molecular Devices (Sunnyvale, CA). dose of the DNA-damaging agents, control cells transfected

Mol Cancer Ther 2006;5(8). August 2006

death as shown by rounding up of cells and fragmentation of cell bodies. This was additionally confirmed with 4¶,6- diamidino-2-phenylindole nuclear staining showing con- densed nuclei, suggesting apoptosis induction (data not shown). This showed that Chk1 is the major mediator of cell cycle checkpoints induced by the used topoisomerase inhibitors; hence, only its knockdown potentiated the toxicity of the drugs. Strikingly, when we combined Chk1 with MK2 siRNA (row 6), we failed to observe signs of cell death, indicating that MK2 knockdown reversed the potentiation effect of Chk1 siRNA. This intriguing finding will be investigated later. Western blot confirmed efficient and specific knockdowns of the various kinase targets by only their respective siRNAs but not the other siRNAs. Additionally, double siRNA transfections achieved comparable extents of target knockdown as single siRNA (Fig. 1B). We further probed the underlying mechanisms of the above finding by molecular marker analysis. HeLa cells were transfected with the various siRNAs and treated with 150 nmol/L camptothecin. Twenty-four hours later, the cells were harvested for immunoblot analysis (Fig. 2A). We first investigated the profile of P-H3, the major M-phase marker. Camptothecin treatment in control cells (transfected with Luciferase siRNA) suppressed the P-H3 signal, in line with the expected S-G2 phase arrest and a corresponding depletion of mitotic cells (lanes 1 and 2). Chk1 siRNA rescued the P-H3 signal, indicating a successful abrogation of cell cycle block and renewed mitotic progression (lanes 3 and 4). In contrast, Chk2 or MK2 siRNA failed to rescue P-H3 (lanes 5–8), suggesting a failure to abrogate the camptothecin-induced checkpoint. Chk1 and Chk2 combination knockdown showed similar efficacy as Chk1 knockdown alone (lanes 9 and 10) whereas Chk2 and MK2 siRNA combination did not lead to a recovery of the P-H3 signal (lanes 13 and 14). In contrast, the combination of Chk1 and MK2 siRNAs failed to abrogate the checkpoint and induce mitotic progression (lanes 11 and 12), suggesting that MK2 knockdown not only failed to enhance the efficacy of Chk1 siRNA but actually blocked its ability to abrogate cell cycle checkpoint. This is consistent with the surprising finding in Fig. 1 showing a failure of the double Chk1/MK2 knockout to potentiate the Figure 1. Morphologic studyof HeLa cells transfected with Chk1, cytotoxicity of camptothecin or doxorubicin. Chk2, and MK2 siRNAs, either singlyor in combination, and treated with To corroborate the P-H3 result, we investigated another topoisomerase inhibitors. HeLa cells were transfected in six-well plates with siRNA(s) targeting either control luciferase, Chk1, Chk2, or MK2, downstream effector of checkpoint abrogation, P-H2AX, a either singlyor in combination as indicated. Sixteen hours posttransfec- marker for DNA double-strand breaks. The rationale is that tion, cells were treated or not with 150 nmol/L camptothecin (CPT)or in control cells receiving DNA damage, only a lowlevel of 150 nmol/L doxorubicin (Dox) for 24 h. Cells were examined under phase- contrast microscope to analyze the sensitization effect of the differential P-H2AX will be observed due to the cell cycle arrest and, siRNA-treatments (A). Cells under each condition were also harvested for thus, cells have sufficient time to repair the double-strand Western blot analysis to confirm the specific knockdown of the target of breaks. However, the checkpoint abrogation will force the each siRNA (B). cells into premature cell cycle progression in the presence of DNA damage, which will exacerbate the double-strand with luciferase siRNA showed no overt signs of cell death breaks and induce an elevated P-H2AX signal. Consistent (Fig. 1A, row 1); the same is true for cells transfected with with this expectation, camptothecin induced a moderate Chk2 and MK2 siRNA, either alone or in combination (rows increase of P-H2AX in control cells (Fig. 2A, lanes 1 and 2) 3, 4, and 7). On the other hand, Chk1 siRNA (row 2)or whereas Chk1 down-regulation significantly increased this Chk1 plus Chk2 siRNA (row 5) conferred massive cell signal (3.6-fold increase when comparing lane 4 versus

Mol Cancer Ther 2006;5(8). August 2006

lane 2, Fig. 2A), suggesting that abrogation of the arrest and MK2 failed to elevate the P-H2AX signal (lanes 11 exacerbated the DNA damage, leading to potentiation of and 12), in total agreement with the P-H3 result and the toxicity of camptothecin. In contrast, down-regulation additionally confirming that MK2 down-regulation of either Chk2 or MK2 did not enhance the extent of reverses the efficacy of Chk1 siRNA. DNA damage (lanes 5–8), again showing that Chk1 is the Because we and others have previously shown that major mediator of the checkpoint. In line with the P-H3 besides DNA-damaging agents such as topoisomerase result, Chk1 and Chk2 siRNA combination (lanes 9 and 10) inhibitors, Chk1 inhibition could also greatly potentiate conferred a comparable increase in P-H2AX level as Chk1 the efficacy of antimetabolites and other replication siRNA alone, whereas combination of Chk2 and MK2 inhibitors such as 5-FU, hydroxyurea, and gemcitabine siRNA (lanes 13 and 14) was just as ineffective as Chk2 or (29–32), we tested whether we could extend the above MK2 siRNA alone. Yet again, double knockdown of Chk1 observations to 5-FU. HeLa cells were similarly transfected with various siRNAs and subjected to treatment with 5-FU at 50 Amol/L for 24 hours. This dose corresponds to 20% of the EC50 value of 5-FU in suppressing HeLa cell growth in a cell proliferation assay. We used the same two markers, P-H3 and P-H2AX, to examine the effect of each siRNA on the checkpoint abrogation and potentiation of DNA damages (Fig. 2B). As expected, in control cells, 5-FU successfully depleted the mitotic cell population as shown by the elimination of the P-H3 signal, consistent with the expected G1-S arrest. Chk1 siRNA, but not Chk2 or MK2 siRNA, recovered the P-H3 signal, indicating a successful checkpoint abrogation followed by mitotic progression, which led to enhanced double-strand breaks as denoted by increased P-H2AX signal. Again, additional MK2 knockdown blocked the efficacy of Chk1 siRNA. However, distinct from Fig. 2A, in which Chk1 and Chk2 double knockdown induced similar effect as Chk1 siRNA alone in potentiation of camptothecin, Chk1 and Chk2 double knockdown failed to achieve the full effect of Chk1 inhibition alone in potentiation of 5-FU both in terms of checkpoint abrogation (P-H3 signal recovery) and potentiation of DNA damage (P-H2AX signal elevation), indicating that to sensitize tumor cells to antimetabolites, only Chk1 inhibition is necessary and additional Chk2 or MK2 inhibition may be counterproductive. To correlate the observed checkpoint abrogation with induction of apoptosis, we further tested the caspase activation of the differentially transfected HeLa cells with or without camptothecin treatment for 32 hours (Fig. 2C). Due to the relative short duration of the treatment and the lowdose of camptothecin used, control cells did not showa dramatic induction of caspase activity in response to camptothecin treatment. Although not activating caspase by itself, Chk1 siRNA greatly potentiated the camptothecin-induced caspase activity (>5-fold increase). Again, Chk2 or MK2 siRNAs were not effective in this assay. Combination knockdowns additionally confirmed that Chk1/Chk2 knockdown was modestly lower than Chk1 Figure 2. Cell cycle marker analysis of HeLa cells transfected with siRNAs targeting various checkpoint kinases and treated with camptothe- knockdown alone, and MK2 knockdown significantly cin (A) or 5-FU (B). HeLa cells were treated similarlyas in Fig. 1. Cells were diminished the induction of caspase activity conferred by harvested for protein analysis for P-H3 and P-H2AX to ascertain mitotic Chk1 siRNA. progression and double-strand break induction. Chk1, Chk2, and MK2 immunoblots were done to confirm target knockdowns bythe To ascertain whether the above conclusions can be corresponding siRNAs. Actin was probed as loading control. Representa- applied to other cancer cell lines besides HeLa cells, we tive of three separate tests. C, HeLa cells transfected with the indicated carried out similar siRNA transfections in H1299 cells, siRNAs were treated or not with 150 nmol/L camptothecin for 32 h and a non–small-cell lung cancer cell line (Fig. 3). In addition assayed for caspase activity according to the suggestions of the manufacturer. Excitation was at 400 nm and fluorescence emission was to P-H3, we also characterized the profile of another cell detected at 505 nm. cycle marker, phospho-Cdc2 Y15P. Cdc2 kinases undergo

Mol Cancer Ther 2006;5(8). August 2006

phosphorylation at Y15 position in response to DNA We additionally examined the effect of double-siRNA damage, resulting in inhibition of the kinase activity and, knockdowns in H1299 cells (Fig. 3B). Consistent with the hence, S or G2 phase arrest. Therefore, this marker is HeLa cell result, double Chk1 and Chk2 down-regulation commonly used as an indicator of cell cycle arrest. H1299 produced indistinguishable results from Chk1 knockdown cells with various siRNA transfections were treated with alone in terms of Y15P and P-H3 profiles (compare lanes 3 camptothecin or doxorubicin and cell lysates were probed and 4 versus lanes 5 and 6). MK2 knockdown, in addition for Cdc2 Y15P (Y15P) and P-H3. As expected, doxorubicin to Chk1 knockdown, again reversed the checkpoint and camptothecin conferred G2 or S phase arrest in control abrogation seen with Chk1 siRNA alone by increasing cells by inducing the Y15P signal, and correspondingly, the the Y15P level and abolishing the P-H3 signal (lanes 7 depletion of the mitotic marker, P-H3 (lanes 1–3, Fig. 3A). and 8). In summary, cell cycle marker analysis indicated Chk1 knockout completely eliminated the increases in Y15P that only Chk1 inhibition conferred effective checkpoint level and rescued the P-H3 signal (lanes 4–6), indicating abrogation and mitotic progression, which has been shown an efficient checkpoint abrogation followed by mitotic to be necessary for the induction of mitotic catastrophe progression. Interestingly, Chk2 knockdown also signifi- and the potentiation of the toxicity of chemotherapy (5). cantly diminished the elevation of the Y15P signal, albeit to Our results clearly indicated that neither Chk2 nor MK2 a lesser extent than Chk1 siRNA. This suggests that knockdown achieved any effects. More importantly, Chk2 knockdown may have partially activated Cdc2. inhibition of Chk2 or MK2, in addition to Chk1 suppres- However, the P-H3 profile showed no recovery, indicating sion, failed to enhance or even attenuated the efficacy of that partially activated Cdc2 was insufficient for complete Chk1 inhibition alone. checkpoint abrogation or mitotic progression. MK2 siRNA Fluorescence-activated cell sorting analysis was carried did not abolish the doxorubicin-induced increase in Y15P out to corroborate the above molecular marker analysis level but abrogated the camptothecin-induced increase, (Fig. 4). As expected, control cells displayed prominent suggesting that MK2 knockdown may partially relieve G2-M arrest in response to doxorubicin and a hybrid S-G2 camptothecin-induced checkpoint (lanes 10–12). However, arrest in response to camptothecin. Under our experimen- the P-H3 profile again showed that this partial effect failed tal condition (150 nmol/L doxorubicin or camptothecin to translate into productive mitotic progression. for 1 day), no significant cell death population (sub-G1 peak) was observed. Consistent with previous reports (7, 29), Chk1 down-regulation abrogated the doxorubicin- induced G2-phase arrest by lowering the G2 peak, resulting in apoptosis (G2-population decreased from 60% to 33% whereas cells with a sub-G1 DNA content increased from 1.5% to 13.3%; Fig. 4B). Similarly, Chk1 siRNA also abrogated the camptothecin-induced hybrid S-G2 peak and drove the cells into sub-G1 phase (total S-G2 population decreased from 72% to 57% whereas cells with a sub-G1 DNA content increased from 3.6% to 15.7%). Chk2 siRNA only showed a modest abrogation effect compared with Chk1 siRNA (decreasing doxorubicin-induced G2-population from 60% to 52% and camptothecin-induced S-G2 population from 72% to 64% but conferring no significant change to the cells with a sub-G1 DNA content). This is fully consistent with the Chk2 siRNA immunoblot result in Fig. 3A showing a modest abrogation of the Cdc2 Y15P signal but no rescue of the mitotic marker P-H3. Similarly, MK2 siRNA also only showed a very limited abrogation profile. Chk1 and Chk2 double knockdown decreased doxorubicin-induced G2- phase arrest from 60% to 39% and decreased camptothecin-induced S-G2 arrest from 72% to 58%. Concomitantly, it increased the cells with a sub-G1 DNA content from 1.5% to 12.4% for doxorubicin and from 3.6% to 12.8% for camptothecin. These figures are virtually identical to Figure 3. Cell cycle marker analysis of H1299 cells transfected with those achieved by Chk1 siRNA alone, again indicating no siRNAs targeting Chk1, Chk2, and MK2 and treated with doxorubicin or additional benefits from Chk2 knockdown. In contrast, camptothecin. H1299 cells were transfected in six-well plates with the double Chk1 and MK2 knockdown only decreased indicated siRNAs either singly( A) or in various combinations (B) and then doxorubicin-induced G -phase arrest from 60% to 52% treated with 200 nmol/L doxorubicin or 200 nmol/L camptothecin for 24 h. 2 Protein lysates were harvested for immunoblot analysis for the checkpoint (compared with 33% achieved by Chk1 siRNA alone) and marker Cdc2 Y15P and the mitotic marker P-H3. only increased apoptotic cells in sub-G1 phase from 1.5%

Mol Cancer Ther 2006;5(8). August 2006

S phase arrest, conferring apoptosis. In summary, these results confirmed the above molecular marker analysis and further established that Chk1 is the dominant checkpoint kinase. The same conclusions were reached in a cell proliferation assay (MTS assay). We carried out this assay with HeLa cells transfected with the relevant siRNAs (either single or in combination). Transfected cells were then treated with low doses of camptothecin (150 nmol/L) or 5-FU (50 Amol/L) for 2 days and cell growth under each condition was assessed by MTS assay (Fig. 5A). As expected, in control luciferase siRNA–transfected cells, camptothecin and 5-FU only displayed very moderate antiproliferation effects (f20% inhibition of growth). Chk1 siRNA conferred significant potentiation of both agents by increasing the growth inhibition by 60% to 70%. In comparison, neither Chk2 nor MK2 siRNA conferred any enhanced growth suppression over control. In double knockdowns, Chk1 and Chk2 siRNAs showed similar potentiation effect as Chk1 siRNA alone in enhancing camptothecin toxicity; however, they were moderately lower in potentiation of 5-FU than Chk1 siRNA alone

Figure 4. Cell cycle analysis of H1299 cells transfected with siRNAs targeting various checkpoint kinases and treated with doxorubicin or camptothecin. A, H1299 cells were treated similarlyas in Fig. 2 and subjected to fluorescence-activated cell sorting analysis. All profiles were uniformlyscaled. B, cells in different phases were quantitated with the Figure 5. Knockdown of Chk1, but not of Chk2 or MK2, potentiates CellQuest program and the percentages of cells in sub-G1,G1, S, and G2-M phases were plotted to displaythe extent of checkpoint abrogation and chemotherapeutics in both short-term cell proliferation assayand long- induction of apoptosis. Representative of three independent trials. term colonyformation assay. A, HeLa cells were transfected with the indicated siRNAs (single or combination) and treated or not with 150 nmol/L camptothecin or 50 Amol/L 5-FU for 2 d. Cell proliferation assay(MTS assay)was done to assess cell growth under each condition. Luciferase siRNA – transfected cells with no drug treatment was used as to 6.4% (compared with 13% achieved by Chk1 siRNA 0% growth inhibition. B, H1299 cells transfected with the indicated alone). For camptothecin, the MK2/Chk1 double siRNA siRNAs were trypsinized and replated into six-well plates at 500 per well. increased the cells with a sub-G1 DNA content from Cells were treated with vehicle or 0.3 nmol/L camptothecin for 9 d to allow 3.6% to 5.5%, much less than the 15% accomplished with colonyformation. Colonies were then stained with methyleneblue and scanned. Total colonyformation was calculated with Image-Pro program. Chk1 siRNA alone. Therefore, MK2 siRNA significantly Luciferase siRNA was used as control (100% colonyformation). abolished the ability of Chk1 siRNA to abrogate the G2 or Representative of three separate trials.

Mol Cancer Ther 2006;5(8). August 2006

(47% inhibition for Chk1 + Chk2 versus 71% inhibition for H1299 cells was very low, we used H1299 cells expressing Chk1 siRNA only). This correlates well with Fig. 2B, which lowlevels of ectopic Cdc25A for this study. The cells showed diminished efficacy in inductions of P-H3 and were transfected with siRNAs targeting either luciferase, P-H2AX with Chk1 and Chk2 double knockdown. The MK2, or Chk1 and then were harvested 24 hours later for effect of Chk2 and MK2 double knockdown was indistin- immunoblot analysis of Cdc25A expression profile (Fig. 6). guishable from that of control siRNA, and the same is true In comparison with luciferase siRNA transfection, Chk1 for single knockdown of either gene. Again, MK2 knock- siRNA increased the Cdc25A protein level, consistent with down, in addition to Chk1 down-regulation, abolished the previous studies showing that Chk1 constitutively targets potentiation effect seen with Chk1 siRNA alone with both Cdc25A to degradation even in the absence of DNA camptothecin and 5-FU. damage (10, 13). In contrast, MK2 down-regulation We further ascertained the long-term effect of the down- abolished the Cdc25A signal, indicating that loss of MK2 regulation of the various checkpoint kinases through destabilized the Cdc25A protein. On the other hand, clonogenic assay (Fig. 5B). Because H1299 cells form Cdc25C, another member of the Cdc25 phosphatase family, distinctive colonies that are easily quantifiable, and Chk1 did not showsignificant variation among the different siRNA confers sustained knockdown of Chk1 in this cell knockdowns. Hence, we speculated that MK2 siRNA may line for up to 10 days (data not shown), it was chosen for have prevented Chk1 siRNA from abrogating the cell cycle the clonogenic assay. H1299 cells transfected with the checkpoint due to its ability to destabilize the Cdc25A indicated siRNAs were replated at 500 per well in six-well protein, which is required for the checkpoint abrogation plates. Colony formation was monitored in the absence or and cell cycle progression. We are in the process of further presence of lowdoses of camptothecin for 9 days (due to characterizing this interesting discovery. We also carried the long incubation time and lowcell density, cells became out Chk2 siRNA knockdown in a similar assay, which much more sensitive to cytotoxic agents as compared with showed that Chk2 depletion does not affect Cdc25A the short-term MTS assay). At 0.3 nmol/L camptothecin, stability (data not shown). colony formation was barely disrupted in the control luciferase siRNA–transfected cells. Chk1 siRNA alone Discussions conferred f40% decrease in colony formation, indicating that although Chk1 inhibition has no significant toxicity in To address the issue of whether a pan-inhibitor of short-term growth assays (Fig. 4), it induces significant checkpoint kinases would be necessary to achieve maxi- growth suppression in long-term assays. Despite this mum efficacy in checkpoint abrogation, we carried out baseline shift, Chk1 siRNA still effectively potentiated the selective knockdown studies in different cancer cell lines. toxicity of camptothecin because it decreased the colony Molecular marker, cell cycle, cell proliferation, and formation to <10% of the control. In contrast, neither Chk2 clonogenic studies all showed that Chk1 inhibition alone nor MK2 siRNA displayed any sensitization effect, and is sufficient to sensitize cancer cells to a variety of their combination is virtually indistinguishable from chemotherapeutics including topoisomerase inhibitors control luciferase siRNA. Chk2 knockdown, in addition and antimetabolites, two cornerstones of cancer therapy. to Chk1 knockdown, conferred similar effects as Chk1 down-regulation only and did not increase the efficacy of the potentiation. Again, in line with the results from the cell proliferation assay, MK2 knockout partially blocked the potentiation effect of Chk1 siRNA. Therefore, using molecular marker analysis, fluorescence-activated cell sorting, and cell proliferation and clonogenic assays, we have consistently shown that Chk1 inhibition alone constitutes the best approach to abrogate checkpoint and sensitize tumor cells to chemotherapy. More intriguingly, we also showed the surprising ability of MK2 siRNA to block Chk1 siRNA from abrogating the checkpoint, initiating mitotic progression and potentiating cell death. To identify the underlying mechanisms of this interesting but puzzling finding, we focused on Cdc25A, the critical downstream effector molecule of Chk1 in the checkpoint pathway. It has been shown that Chk1 mediates cell cycle checkpoint by targeting Cdc25A, which is required for both S and G2-M transitions, to proteasome-mediated degrada- Figure 6. MK2 knockdown destabilizes Cdc25A. H1299 cells were tion. Therefore, Chk1 down-regulation rescues Cdc25A transfected with verylow level of Cdc25A expression vector (0.03 Ag/well from proteolysis, which then drives cells into premature in a 12-well plate) and then additionallytransfected with either luciferase, MK2, or Chk1 siRNA at 50 nmol/L. Protein lysates were analyzed 24 h later mitosis despite unrepaired DNA damage or unfinished for Cdc25A or Cdc25C expression. Chk1 and MK2 were also probed to replication (5, 9). Because the endogenous Cdc25A level in confirm the siRNA efficacy.

Mol Cancer Ther 2006;5(8). August 2006

Down-regulation of additional targets, such as Chk2 or 8. Zachos G, RaineyMD, Gillespie DA. Chk1-deficient tumour cells are viable but exhibit multiple checkpoint and survival defects. EMBO J 2003; MK2, not only fails to improve efficacy but actually 22:713 – 23. attenuates or abrogates the efficacy achieved with Chk1 9. Zhao H, Watkins JL, Piwnica-Worms H. Disruption of the checkpoint inhibition alone. kinase 1/cell division cycle 25A pathway abrogates ionizing radiation- induced S and G2 checkpoints. Proc Natl Acad Sci U S A 2002;99: Our cell cycle marker analysis reveals that partial Cdc2/ 14795 – 800. cyclin-dependent kinase activation is not sufficient to 10. Xiao Z, Chen Z, Gunasekera AH, et al. Chk1 mediates S and G2 mount an effective checkpoint abrogation. In H1299 cells arrests through Cdc25A degradation in response to DNA-damaging (Fig. 3A), DNA-damaging agents (doxorubicin or campto- agents. J Biol Chem 2003;278:21767 – 73. thecin) confer cell cycle arrest by increasing the inhibitory 11. Hirao A, Cheung A, Duncan G, et al. Chk2 is a tumor suppressor that regulates apoptosis in both an ATM-dependent and ATM-independent Cdc2 Y15P (G2-phase arrest indicator) level and suppress- manner. Mol Cell Biol 2002;22:6521 – 32. ing the P-H3 signal (mitotic indicator). Chk1 siRNA 12. Bartek J, Lukas J. Chk1 and Chk2 kinases in checkpoint control and efficiently abrogates the increase in Y15P and rescues the cancer. Cancer Cell 2003;3:421 – 9. P-H3 signal, indicating successful mitotic progression 13. Sorensen CS, Syljuasen RG, Falck J, et al. Chk1 regulates the S phase following checkpoint abrogation. Chk2 knockdown also checkpoint bycoupling the physiological turnover and ionizing radiation- induced accelerated proteolysis of Cdc25A. Cancer Cell 2003;3:247 – 58. depressed the Y15P level, albeit to a lesser extent. However, 14. Falck J, Mailand N, Syljuasen RG, Bartek J, Lukas J. The ATM-Chk2 – unlike Chk1 siRNA, the critical difference is that this 25A checkpoint pathwayguards against radioresistant DNA synthesis. abrogation does not result in any recovery of the P-H3 Nature 2001;410:842 – 7. signal, indicating that partial increase of the activity 15. Falck J, Petrini JH, Williams BR, Lukas J, Bartek J. The DNA damage- dependent intra-S phase checkpoint is regulated byparallel pathways.Nat Cdc2 did not result in productive mitotic progression and Genet 2002;30:290 – 4. checkpoint abrogation. In comparison, MK2 knockdown 16. Takai H, Naka K, Okada Y, et al. Chk2-deficient mice exhibit eliminated the camptothecin-induced increase of Cdc2 radioresistance and defective p53-mediated transcription. EMBO J 2002; Y15P, but not the doxorubicin-induced increase, and failed 21:5195 – 205. to rescue the P-H3 signal in both cases. Therefore, it seems 17. Jack MT, Woo RA, Hirao A, Cheung A, Mak TW, Lee PW. Chk2 is dispensable for p53-mediated G1 arrest but is required for a latent p53- that whereas all three siRNAs were able to induce different mediated apoptotic response. Proc Natl Acad Sci U S A 2002;99:9825 – 9. degrees of cyclin-dependent kinase activation, only Chk1 18. Collins I, Garrett MD. Targeting the cell division cycle in cancer: CDK and siRNA induced sufficient extent of cyclin-dependent kinase cell cycle checkpoint kinase inhibitors. Curr Opin Pharmacol 2005;5:366– 73. activation, which is required for the checkpoint abrogation 19. GuayJH, Lambert G, Gingras-Breton JN, Lavoie J, LandryJ. and mitotic transition. Regulation of actin filament dynamics by p38 MAP kinase-mediated phosphorylation of heat shock protein 27. J Cell Sci 1997;110:357 – 68. In summary, our study clarified the differential roles of 20. Rouse J, Cohen P, Trigon S, et al. A novel kinase cascade triggered by the three checkpoint kinases in cell cycle control by stress and heat shock that stimulates MAPKAP kinase-2 and phosphor- convincingly showing that inhibition of Chk1 is both ylation of the small heat shock proteins. Cell 1994;78:1027 – 37. necessary and sufficient for complete checkpoint abroga- 21. Kotlyarov A, Neininger A, Schubert C, et al. MAPKAP kinase 2 is tion, and furthermore, suppression of Chk2 or MK2 not essential for LPS-induced TNF-a biosynthesis. Nat Cell Biol 1999;1:94 – 7. 22. Rou PP, Blenis J. ERK and p38 MAPK-activated protein kinases: a only fails to enhance the efficacy of Chk1 inhibition but familyof protein kinases with diverse biological functions. Microbiol Mol may actually antagonize it. Therefore, Chk1 is the only Biol Rev 2004;68:320 – 44. relevant cancer target among the checkpoint kinases, and 23. Manke IA, Nguyen A, Lim D, Stewart MQ, Elia AE, Yaffe MB. efforts should be made during the compound screening MAPKAP kinase-2 is a cell cycle checkpoint kinase that regulates the G2-M transition and S phase progression in response to UV irradiation. Mol Cell process to ensure that the obtained Chk1 inhibitor does 2005;17:37 – 48. not target Chk2 or MK2. 24. Abraham RT. MAPKAP kinase-2: three’s companyat the G(2) checkpoint. Mol Cell 2005;17:163 – 4. Acknowledgments 25. Glade-Bender J, Kandel JJ, Yamashiro DJ. VEGF blocking therapyin We thank Dr. Saul Rosenberg for his critical reading of the manuscript and the treatment of cancer. Expert Opin Biol Ther 2003;3:263 – 76. valuable inputs. 26. Dewji MR. Earlyphase I data on an irreversible pan-erb inhibitor: CI-1033. What did we learn? J Chemother 2004;16 Suppl 4:44 – 8. References 27. Ahmed SI, Thomas AL, Steward WP. Vascular endothelial growth factor (VEGF) inhibition bysmall molecules. J Chemother 2004;16 Suppl 4:59– 63. 1. HurleyLH. DNA and its associated processes as targets for cancer therapy. Nat Rev Cancer 2002;2:188 – 200. 28. Smith JK, Mamoon NM, Duhe RJ. Emerging roles of targeted small molecule protein-tyrosine kinase inhibitors in cancer therapy. Oncol Res 2. Hartwell LH, Kastan MB. Cell cycle control and cancer. Science 1994; 2004;14:175 – 225. 266:1821 – 8. 29. Xiao Z, Xue J, Sowin TJ, Rosenberg SH, Zhang H. A novel 3. Zhou BB, Elledge SJ. The DNA damage response: putting checkpoints mechanism of checkpoint abrogation conferred byChk1 down-regulation. in perspective. Nature 2000;408:433 – 9. Oncogene 2005;24:1403 – 11. 4. Dixon H, NorburyCJ. Therapeutic exploitation of checkpoint defects in 30. Parsels LA, Parsels JD, Tai DC, Coughlin DJ, Maybaum J. 5-fluoro-2¶- cancer cells lacking p53 function. Cell Cycle 2002;1:362 – 8. deoxyuridine-induced cdc25A accumulation correlates with premature 5. Zhou BB, Bartek J. Targeting the checkpoint kinases: chemosensitiza- mitotic entryand clonogenic death in human colon cancer cells. Cancer tion versus chemoprotection. Nat Rev Cancer 2004;4:216 – 25. Res 2004;64:6588 – 94. 6. Li Q, Guntuku S, Cui XS, et al. Chk1 is an essential kinase that is 31. Arlander SJ, Eapen AK, Vroman BT, McDonald RJ, Toft DO, Karnitz regulated byAtr and required for the G(2)/M DNA damage checkpoint. LM. Hsp90 inhibition depletes Chk1 and sensitizes tumor cells to Genes Dev 2001;14:1448 – 59. replication stress. J Biol Chem 2003;278:52572 – 7. 7. Chen Z, Xiao Z, Xhen J, et al. Human Chk1 expression is dispensable 32. Mesa RA, Loegering D, Powell HL, et al. Heat shock protein 90 for somatic cell death and critical for sustaining G2 DNA damage inhibition sensitizes acute myelogenous leukemia cells to cytarabine. checkpoint. Mol Cancer Ther 2003;2:543 – 8. Blood 2005;106:318 – 27.

Mol Cancer Ther 2006;5(8). August 2006

Downloaded from mct.aacrjournals.org on September 24, 2021. © 2006 American Association for Cancer Research. Differential roles of checkpoint kinase 1, checkpoint kinase 2, and mitogen-activated protein kinase−activated protein kinase 2 in mediating DNA damage−induced cell cycle arrest: implications for cancer therapy

Zhan Xiao, John Xue, Thomas J. Sowin, et al.

Mol Cancer Ther 2006;5:1935-1943.

Updated version Access the most recent version of this article at: http://mct.aacrjournals.org/content/5/8/1935

Cited articles This article cites 32 articles, 13 of which you can access for free at: http://mct.aacrjournals.org/content/5/8/1935.full#ref-list-1

Citing articles This article has been cited by 12 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/5/8/1935.full#related-urls

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

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

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/5/8/1935. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.