Background
The DNA damage response (DDR) is a complex network of signaling pathways that have evolved to protect cells from DNA damage or interference with DNA synthesis. A series of cell cycle checkpoints at G1/S, intra-S or S, and G2/M protect cells from undergoing aberrant division in the presence of DNA damage thereby allowing DNA repair, regulation of transcription and apoptosis [
1‐
4]. The serine-threonine checkpoint kinases Chk1 and Chk2 are often described as the “central transducers” of the DDR and are activated by the ATM kinase in response to DNA breaks and ATR kinase by single-stranded regions of DNA and form the key link between the sensing kinases ATM/ATR and the cell cycle machinery. Recognition of DNA double strand breaks by the Mre11 complex (Mre11, Rad50 and Nbs1) or replication stress by the Rad9-Hus1-Rad1 complex results in the activation of the ATR and ATM kinases respectively. These kinases, in turn, activate the effector kinases Chk1 and Chk2. Chk1 activation occurs predominantly by three phosphorylation events on S317 and 345 by ATR [
5,
6] and auto-phosphorylation on S296 [
7]. Chk1 and Chk2 negatively regulate the Cdc25 family of phosphatases thereby preventing cell cycle progression as well as directly modulating repair proteins resulting in effective lesion repair. Biochemical and genetic studies have demonstrated Chk1 to be essential and indispensable for the S- and G2/M checkpoints [
1,
8]. In the vast majority of human cancers, p53 (an important effector of the G1/S checkpoint) is mutated or functionally inactivated, rendering cancer cells reliant on Chk1/Chk2 for checkpoint activation, in the presence of endogenous or exogenous DNA damage.
DNA damaging cytotoxic chemotherapeutic agents and ionizing radiation are the mainstay of current cancer treatment regimens. These agents target the DNA in cancer cells and induce DNA damage either directly through DNA adduct formation (for example cisplatin) or indirectly via inhibition of DNA synthesis (for example gemcitabine and 5-fluoruracil) or DNA unwinding (for example etoposide). All of these processes result in DNA strand breaks, activation of the DDR and cell cycle checkpoints, and ultimately cell cycle arrest. Targeting the DDR through Chk1 inhibition, therefore, represents a novel therapeutic strategy to increase DNA-damaging chemotherapeutic drug induced tumor cell death in p53 pathway defective cancers [
9,
10] by abrogating the remaining intact checkpoint. This “synthetic lethality” approach should increase the therapeutic index of a given chemotherapeutic drug as normal cells remain protected by their functional p53 pathway. This approach has started to be tested clinically with multiple small molecule inhibitors of Chk1 in clinical evaluation in Phase I (GDC-0425 and GDC-0575) or Phase II (LY2603618 [
11] and MK-8776 (SCH 900776) [
12]) trials in combination with gemcitabine, pemetrexed and cisplatin [
13].
The advent of molecularly targeted cancer therapeutics has resulted in increased emphasis on identifying pharmacological biomarkers of drug/target interaction to help accelerate the progress of novel agents through clinical trials [
14‐
16]. To date, biomarker and clinical studies of Chk1 inhibitors have predominantly focused on the combination with gemcitabine. However, Chk1 inhibitors have the potential to be combined with a wide range of cytotoxic chemotherapeutics. In this study, we evaluated the potential for a novel, highly selective Chk1/2 inhibitor, V158411, to potentiate the cytotoxicity of a range of agents in p53 mutant colon cancer cells and the corresponding changes in a panel of potential pharmacodynamic biomarkers for predictors of V158411 combinatorial activity.
Methods
Cell lines and cell culture
All cell lines were purchased from the American Type Culture Collection (ATCC), established as a low passage cell bank and then routinely passaged in our laboratory for less than 3 months after resuscitation. HT29, Colo205 and HCT116 cells were routinely cultured in DMEM containing 10% FCS and 1% penicillin/streptomycin at 37°C in a normal humidified atmosphere supplemented with 5% CO2.
Compounds
Solid stocks were purchased from the indicated suppliers and prepared as concentrated stock solutions in the appropriate solvent: gemcitabine (Apin Chemicals Inc), 20 mM in H2O; camptothecin (LC Laboratories), 5 mM in DMSO; cisplatin (David Bull Laboratories), 3.33 mM in 1% NaCl in H2O; oxaliplatin (Tocris), 5 mM in H2O; etoposide (Selleckchem), 20 mM in DMSO; doxorubicin (Selleckchem), 5 mM in DMSO; 5-fluorouracil (Sigma), 50 mM in DMSO; LY2603618 (Selleckchem), 20 mM in DMSO and MK-8776 (ChemieTek), 20 mM in DMSO.
Potentiation assays
5000 cells per well were seeded in 96-well plates and incubated overnight. Cells were treated with a 10-point titration of cytotoxic chemotherapeutic agent in the presence of a fixed concentration of Chk1 inhibitor for 72 or 168 hours. The effect on cell proliferation was determined using CellTiter 96 AQueous One Solution Cell Proliferation Assay (MTS, Promega) and read on a Victor plate reader (Perkin Elmer).
Anchorage independent growth assays
1500 cells/well in 0.4% low melting point agarose (SeaPlaque, Lonza) in complete media were plated on to 96-well plates coated with 0.8% low melting point agarose in complete media. Wells were subsequently overlaid with complete media containing cytotoxic chemotherapeutic agents and Chk1 inhibitor. Following incubation for 168 hours, cell viability was determined using CellTiter Blue (Promega) and fluorescence determined using a Victor plate reader (Perkin Elmer).
Spheroid growth assays
Multi-cellular tumor spheroid assays were preformed essentially as described previously [
17]. 1000 HT29 cells/well were seeded in 96-well round bottomed ultra-low attachment microplates (Corning Costar), centrifuged at 1000 × g for 3 minutes and spheroids formed for 72 hours. Spheroid cell viability after incubation with chemotherapeutic drug plus V158411 for 168 hours was determined using CellTiter-Glo Luminescent Cell Viability Assay (Promega).
Immunoblotting
Antibodies against Chk1, pChk1 (S317), pChk1 (S345), pChk2 (T68), pChk2 (S516), γH2AX, pCdc2 (Y15), pCdc25c (S216), Cdc25a, phH3 (S10), PARP, cleaved PARP, 53BP1, cyclin A, cyclin B1, cyclin D, cyclinE, pCDK2 (T160) and RPA70 were purchased from Cell Signaling Technologies and pChk1 (S296) from Abcam.
Cells were washed once with PBS and lysed in RIPA buffer containing protease and phosphatase inhibitor cocktails (Roche). Protein concentration was determined using a BCA kit (Pierce). Equal amounts of lysate were separated by SDS-PAGE and western blot analysis conducted using the antibodies indicated above
Statistical analysis
Results were analyzed using a Student’s t-Test tool within the data analysis package provided by Microsoft Excel.
Ethical approval
None of the research in this manuscript involved human subjects, human material, or human data, or used regulated vertebrates or invertebrates.
Discussion
Seven structurally distinct inhibitors of the serine/threonine checkpoint kinase Chk1 have been evaluated or are currently being actively tested in combination clinical trials with a range of cytotoxic chemotherapy drugs such as irinotecan, cisplatin, gemcitabine, pemetrexed and cytarabine. These include XL844, AZD7762 and PF477736 which completed Phase I trials and LY2603618 which completed Phase II, but further development of these agents has subsequently been discontinued. GDC-0425 and GDC-0575 continue to be actively developed in a Phase I setting and MK-8776 (SCH 900776) in Phase II. It is interesting to note that all Chk1 inhibitors so far tested in combination clinical trials (seven to date) have undergone clinical testing in combination with gemcitabine (ClinicalTrials.gov) whilst pemetrexed, cisplatin, irinotecan or cytarabine have been tested with only one Chk1 inhibitor each [
25‐
28].
In this study, we evaluated the ability of the novel Chk1 inhibitor V158411 to potentiate the in vitro cytotoxicity of seven clinically used cytotoxic chemotherapy drugs with different mechanisms-of-action in two p53 mutant and one p53 wild-type colorectal carcinoma cell lines growing either anchorage dependently, anchorage independently or as multi-cellular tumor spheroids. Analysis of in vitro protein biomarker responses was subsequently undertaken in an attempt to identify biomarkers potentially predictive of combinatorial activity.
V158411 induced moderate to good potentiation of all seven cytotoxic agents tested in the two p53 mutant cell lines but not the p53 wild-type cell line. The only exception was with 5-fluorouracil in the HT29 cell line. The lack of potentiation of 5-fluorouracil activity in this cell line was most likely due to the high intrinsic resistance of HT29 cells to 5-fluorouracil (GI50 > 100 μM for single agent). Greater potentiation was observed for cells growing anchorage dependently than cells growing either anchorage independently or as multi-cellular tumor spheroids and may be a reflection of the increased rate of proliferation and/or the fraction of cells undergoing active DNA replication in the anchorage dependent culture conditions. In short term combination studies (3 day co-incubation), gemcitabine was the only agent for which V158411 not only potentiated the anti-proliferative activity of the cytotoxic agent but also increased the fraction of cells killed at the higher concentrations of drug. This increased cell killing was only observed for the short 3 day incubation and was subsequently lost at longer incubations possibly due to the increased cytotoxicity of gemcitabine as a single agent.
Previously published studies have observed the greatest potentiation of cytotoxicity by Chk1 inhibitors with the anti-metabolite class of drugs, including gemcitabine [
23]. Chk1 activity has been demonstrated to be critical for not only the DNA damage response checkpoint but also for replication fork stabilization, replication origin firing and homologous recombination. These later roles have been suggested to be critical for the increased effectiveness of Chk1 inhibitors in combination with gemcitabine [
29] compared to other cytotoxic chemotherapy drugs such as Topoisomerase inhibitors. Gemcitabine inhibits DNA synthesis, DNA replication and cell proliferation through two distinct but linked mechanisms. Gemcitabine diphosphate binds to and irreversibly inhibits ribonucleotide reductase thereby depleting the pool of deoxyribonucleotides available for
de novo DNA synthesis. The triphosphate analogue of gemcitabine can also be incorporated into DNA (in substitution for cytidine) where it acts as a chain terminator thereby inhibiting further DNA synthesis. Inhibition of ribonucleotide reductase by gemcitabine (or similarly by hydroxyurea) induces replication fork stalling. Chk1 activity is required to maintain replication fork stability and inhibition of Chk1 leads to replication fork collapse and the generation of “new” DNA strand breaks. In p53 mutant cancer cells, the checkpoint is functionally inactivated by Chk1 inhibition, therefore these cells progress through S-phase and enter into a premature, lethal mitosis [
30]. Replication fork collapse and checkpoint abrogation by Chk1 inhibitors induces potentially lethal DNA damage killing gemcitabine treated p53-mutant cancer cells by a “double hit” mechanism. It should however be noted that the potentiation of gemcitabine observed in pre-clinical xenograft studies is not nearly as dramatic as that observed in the
in vitro potentiation studies. The pre-clinical combination studies of gemcitabine in combination with Chk1 inhibitors are generally conducted at gemcitabine concentrations below the gemcitabine maximum tolerated dose and using a schedule (once every 3 days) that is not reflective of the clinical schedule (once weekly). Our studies suggest that other cytotoxic drugs such as cisplatin or oxaliplatin, in addition to gemcitabine, are worthy of further evaluation.
One of the challenges, and goals, of molecularly targeted cancer therapeutic development is the identification of biomarkers (whether they be genetic, protein or macromolecule based) that allow the translation of the understanding and knowledge gained at the molecular and cellular level into a therapy effective for patients [
14‐
16]. These biomarkers can be identified and developed for one of three specific aims: 1, to stratify a patient population into potential responders and non-responders, 2, to ensure adequate target engagement or inhibition at a given dose or 3, to assess for pathway modulation and a potentially positive or beneficial therapeutic outcome. The overall aim of biomarker development and utilization is to accelerate the clinical development and adoption of new anti-cancer therapies.
Previous studies have demonstrated that a deficiency in p53 improves chemopotentiation by Chk1 inhibitors [
23]. However, mutation of p53 has been found to be important for overall response but is not sufficient to predict a synergistic outcome between a Chk1 inhibitor and cytotoxic chemotherapy. BRCA, XRCC3, DNA-PK [
31] and CYCLIN B1 [
32] levels have all been postulated to be important in modulating the effectiveness of a Chk1 inhibitor in combination with a DNA damaging agent. In our study, we attempted to correlate potentiation of DNA damaging agent cytotoxicity by the Chk1 inhibitor V158411 with protein biomarker changes induced by the combination to identify biomarker changes predictive of a robust, combinatorial effect.
Phosphorylation of Chk1 on serine 345, an activation phosphorylation site on Chk1 phosphorylated in response to DNA damage by ATR, correlated closely with response to the combination of gemcitabine plus AZD7762 in pancreatic tumor xenografts [
33]. In a separate study, cleaved (activated) caspase-2 levels increased in response to DNA damage when Chk1 was inhibited due to checkpoint inactivation and forced mitotic entry [
34]. However, other studies have demonstrated that death induced by the combination of a DNA damaging drug and Chk1 inhibitor was not always dependent on caspase-2 or the PIDDosome [
35]. In our study, moderate to high potentiation was observed with all of the DNA damaging agents tested in the p53 mutant but not the p53 wild-type colon cancer cell lines. No unifying biomarker was identified that would appear predictive of effective combinatorial activity. Instead, as might be predicted, biomarker responses appeared DNA damaging agent specific. Additionally, there was dependence on post-treatment time to observe optimal biomarker changes but these responses were less dependent on the schedule of addition of DNA damaging agent and V158411.
In all combinations, V158411 efficiently reduced the levels of auto-phosphorylated (pSer296) Chk1 suggesting that Chk1 was effectively inhibited by the concentration of V158411 utilized. pChk1 (S296) would therefore make a powerful biomarker for ensuring effective target engagement and Chk1 inhibition in clinical samples. In combination with all DNA damaging drugs, V158411 induced a time dependent degradation of Chk1. This may, in part, reflect the normal homeostatic process of cellular checkpoint resetting. A reduction in total Chk1 S317 and S345 phosphorylation occurred most consistently with all cytotoxics in combination with V158411 but did not predict cell line sensitivity as similar biomarker changes were observed in the non-responsive, p53 wild-type HCT116 cell line. Induction of γH2AX expression was chemotherapeutic dependent and correlated closely with potentiation for gemcitabine and camptothecin in p53 mutant but not wild-type colon cancer cells. These protein biomarker changes appeared to not depend on the chemical structure of the CHk1 inhibitor as a similar pattern of changes was observed with a range of Chk1 inhibitors with diverse chemotypes. Assays to measure γH2AX are reasonably well developed and are currently being tested clinically with different cancer therapeutics and may therefor prove a relatively straightforward marker to include in clinical studies [
36‐
38].
Competing interests
All authors are either present or past employees of Vernalis (R&D) Ltd. and this work was undertaken as part of their employment. AJM is a stock option holder of Vernalis (R&D) Ltd. The authors declare no other competing financial interests. The authors will adhere to the policies outlined on sharing materials, methods and data.
Authors’ contributions
RR and AJM designed the studies. RR performed the majority of the experiments with help from AJM. RR and AJM wrote the manuscript. All authors read and approved the final manuscript.