Abstract
Purpose: Checkpoint kinase 1 (Chk1) plays a critical role in the activation of mitotic spindle checkpoint and DNA damage checkpoint. We examined the preclinical use of the Chk1 inhibitor PF-00477736 as a docetaxel-sensitizing agent. Specifically, we investigated the correlation between PF-00477736–mediated modulation of biomarkers and the sensitization of docetaxel efficacy.
Experimental Design:In vitro and in vivo studies using COLO205 and other cell lines were done to assess PF-00477736–induced enhancement of docetaxel efficacy and effects on associated biomarkers.
Results: PF-00477736 significantly enhanced the docetaxel-induced efficacy in tumor cells and xenografts. Docetaxel induced dose- and time-dependent increase in the levels of phosphorylated Chk1 (Ser345), phosphorylated histone H3 (Ser10), and γH2AX foci and promoted the cytoplasmic localization of phosphorylated Cdc25C (Ser216). PF-00477736 cotreatment suppressed docetaxel-induced changes in phosphorylated histone H3 and cytoplasmic phosphorylated Cdc25C (Ser216) levels and concurrently sensitized the docetaxel-induced apoptosis. Docetaxel alone or in combination with PF-00477736 induced significant antiproliferative activity in xenografts, shown via [18F]FLT-PET imaging. However, changes in [18F]FLT uptake did not reflect the potentiation of docetaxel efficacy. In contrast, bioluminescence imaging showed that PF-00477736 sensitized docetaxel-induced suppression of tumor survival.
Conclusions: Docetaxel triggers mitotic spindle checkpoint activation at low concentrations and activates both the DNA damage checkpoint and the spindle checkpoint at high concentrations. In combination with docetaxel, PF-00477736 abrogates the mitotic checkpoint, as well as the DNA damage checkpoint, and results in sensitization to docetaxel. Chk1 inhibitor PF-00477736 offers a therapeutic potential for the enhancement of taxane therapy.
Multiple checkpoint kinase 1 (Chk1) inhibitors have been actively pursued in clinical trials for their ability to sensitize the efficacy of DNA-damaging agents or radiation. The critical involvement of Chk1 in the spindle checkpoint also suggests that Chk1 inhibitors may be used to potentiate taxane therapy. In this report, PF-00477736, a selective Chk1 inhibitor, showed significant potentiation of docetaxel-induced antitumor activity in multiple xenograft models without introducing noticeable adverse effects. Several assays, including immunohistochemical and imaging approaches, were used to reveal the underlying mechanism of PF-00477736, as well as to correlate biomarker modulation with the potentiation effect, from a translational perspective. Docetaxel treatment leads to the dose- and time-dependent activation of the DNA damage checkpoint, as well as the mitotic spindle checkpoint. Cotreatment of PF-00477736 and docetaxel showed target-associated efficacy by disrupting both checkpoints and enhancing cell death. Docetaxel is a highly effective “standard-of-care” spindle poison in clinical use, and PF-00477736 could potentially increase the therapeutic window of docetaxel to benefit a broader spectrum of the patient population.
Eukaryotic cell cycle transitions are closely regulated by checkpoint signaling pathways. There are at least three DNA damage checkpoints, at the G1-S, S, and G2-M transitions, as well as one mitotic spindle checkpoint (1). Checkpoint kinase 1 (Chk1) is a key regulator of S, G2-M, and mitotic spindle checkpoints (2, 3). In normal cells, DNA damage activates the p53-dependent G1 checkpoint and triggers DNA repair. In p53-deficient tumors, which account for over half of human cancers, the Chk1-mediated cell cycle arrest becomes the dominant defense mechanism. In response to DNA damage, Chk1 is phosphorylated by ATR at Ser346 and Ser317 sites and become activated (4). Chk1 then inactivates the Cdc25C protein phosphatase by phospohorylating Cdc25C at Ser216, which promotes its sequestration into the cytoplasm and in turn prevents activation of the cyclin B/Cdc2 mitotic kinase complex, finally resulting in G2 cell cycle arrest (5) and suppression of mitotic entry (6). Chk1 also triggers the mitotic spindle checkpoint by activating Aurora B, which signals through the Bub and Mad proteins to inhibit APC/Cdc20 to result in mitotic arrest (3). Thus, antagonizing the Chk1-mediated cell cycle checkpoints have emerged as an attractive target for anticancer therapy (7). If Chk1 activity is blocked, DNA-damaged (8) or spindle-disrupted cells (9) would exit cell cycle arrest before full repair and subsequently undergo mitotic catastrophe or cell death. Chk1 inhibitors consequently increase the therapeutic index of DNA-damaging or antimitotic agents.
Tremendous progress has been made in the past two decades in the research and development of Chk1 inhibitors, in combination with radiation therapy and treatment with DNA-damaging agents. Chk1 inhibitors, such as UCN-1 (10), EXEL-9844 (11), CHIR-124 (12), AZD7762 (13), and PF-00477736 (14), have been shown to abrogate cell cycle arrest and sensitize cells to DNA-damaging therapy in preclinical settings. The current clinical development of Chk1 inhibitors is primarily focused on their combination with radiation and DNA-damaging agents (15).
Recent advances with Chk1 have shown its critical role in the mitotic spindle checkpoint (3), suggesting that Chk1 inhibitors may also potentiate the efficacy of antimitotics. Taxanes, including paclitaxel and docetaxel, are antimitotic agents that are frequently used as standard of care treatments for advanced metastatic cancer patients (16). Docetaxel has received significant interest owing to its high potency compared with other taxanes. Docetaxel displays similar cytotoxic mechanism to paclitaxel, which was found to be dependent on drug concentration and on the molecular profile of the tumor. At lower concentrations, taxanes suppress spindle microtubule dynamics, thus conferring mitotic arrest through spindle checkpoint activation and inducing tumor cell death (17); higher doses of taxane quickly induce cytotoxicity through cell cycle–independent signaling pathways (18). Zachos and coworkers (3) showed that Chk1-deficient cells fail to arrest in paclitaxel-induced mitosis and clearly showed that Chk1 activity is required during spindle checkpoint-regulated mitotic arrest. Additionally, Chk1 down-regulation enhanced paclitaxel-induced inhibition of tumor cell proliferation, mitotic catastrophe, and apoptosis (19). These results offered further rationale for the therapeutic combination of Chk1 inhibitors and taxanes.
PF-00477736, a selective small molecule inhibitor of Chk1, is currently in phase 1 clinical trials in combination with gemcitabine (14). Previous work has shown that PF-00477736 abrogates the S and G2-M checkpoints and prevents cell cycle arrest for subsequent repair in DNA-damaged tumor cells, resulting in sensitized apoptosis. In tumor cell and xenograft models, PF-00477736 showed chemopotentiation in combination with DNA-damaging agents.
In this report, we evaluated PF-00477736–induced sensitization of docetaxel. The pharmacodynamic end points associated with the Chk1 signaling pathway were assessed with immunohistochemical and imaging-based assays. Noninvasive bioluminescent imaging (BLI) and [18F]FLT (3′-fluoro-3′ deoxythymidine) PET imaging were used to assess treatment-induced cell activities. A common PET imaging tracer, [18F]FLT, is taken up by cells and phosphorylated by thymidine kinase 1, at which point it becomes trapped in the cells (20). Because thymidine kinase 1 is overexpressed in tumor cells, [18F]FLT uptake is a reliable tumor proliferation marker (21). [18F]FLT-PET imaging is considered to be more translatable than the conventional biopsy approach due to its noninvasive nature and lower risk to the patient (22). [18F]FLT-PET imaging has been increasingly used in the clinic as an early indication of therapeutic response before a reduction in tumor burden becomes apparent (23). Where feasible, [18F]FLT-PET imaging can significantly accelerate the development of novel cancer therapies.
BLI, although only used in preclinical settings, is a cost-effective optical imaging modality (24, 25). Because BLI output relies on ATP-dependent oxidation of d-luciferin by the endogenous luciferase of the cells, in vivo BLI is a reliable indicator of cell metabolic activity in real-time (26). By combining [18F]FLT-PET and BLI imaging modalities, the proliferating and metabolic activity of tumor cells in living animals can be monitored simultaneously and longitudinally (27).
Here, we show that PF-00477736 enhances docetaxel activity in tumor cells and xenografts by abrogating the mitotic spindle checkpoint, as well as the DNA damage checkpoint. We used multiple assay approaches to correlate Chk1 target modulation, apoptosis, reduction of tumor cell proliferation, and PF-00477736–induced efficacy in tumor cells and xenografts. These findings offer insights for the future development of Chk1 inhibitors in combination with taxanes.
Materials and Methods
PF-00477736 was synthesized by Pfizer chemists and prepared as previously described (14). Docetaxel (Taxotere, Henry Schein) was prepared according to the manufacturer's instructions. Luciferase-expressing COLO205 cells were stably transfected to constitutively express luciferase (28). Antibodies recognizing phosphorylated Chk1 (Ser345) 133D3 (2348, dilution 1:25), caspase-3 (9661, dilution 1:200), phosphorylated Cdc25C (Ser216) 63F9 (4901S, dilution 1:50), phosphorylated histone H3 (Ser10; 9701, dilution 1:200), and phosphorylated histone H2A.X (Ser139; 2577, dilution 1:200) were ordered from Cell Signaling. Anti-Chk1 antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, G-4).
In vitro assays. Cell cycle analysis was done with a FACSCalibur system (Becton Dickinson) using the CycleTEST Plus kit (Becton Dickinson) and Annexin V/PE apoptosis detection kit (Becton Dickinson). Western blot analysis for phosphorylated Chk1 (Ser345) was done as described previously (14). Immunofluorescently stained slides were scored based on the percentage of positively stained cells versus 4′,6-diamidino-2-phenylindole (DAPI)–stained cell counts (total).
In vivo studies and drug administration. All animal experimental procedures complied with the Guide for the Care and Use of Laboratory Animals (Institute for Laboratory Animal Research, 1996) and were approved by the Pfizer Global Research and Development Institutional Animal Care and Use Committee. Two million cells were implanted in the dorsal region of athymic NCr-nu/nu mice (Charles River Breeding Laboratories). When tumors reach 150 to 200 mm3, mice were randomly assigned to groups such that the mean value of tumor size was identical for all groups. Mice were then i.p. administered with (a) vehicle, (b) PF-00477736 at 7.5 or 15 mg/kg twice daily (0 and 6 h), (c) docetaxel once daily at 15 or 30 mg/kg (0 h), and (d) docetaxel and PF-00477736 at 7.5 or 15 mg/kg. Mice were dosed on days 1, 8, and 15 for BLI, [18F]FLT-PET imaging, and caliper measurements. For experiments using immunohistochemical assays, tumor-bearing mice were treated on day 1 and tumors were collected at 24 or 48 h after the initial dosing.
Immunohistochemistry analysis. Six tumor-bearing mice were used in each group for immunohistochemical studies. All immunostained sections were counterstained with hematoxylin. Quantitative analysis of section staining was done using the Chromovision automated cell imaging system.
Bioluminescence imaging.In vivo BLI was conducted as previously described (29) using the IVIS100 system with Living Image Acquisition and Analysis Software (Caliper Life Sciences). Before imaging, mice were anesthetized with 2.5% isofluorane and i.p. administered with 75 mg/kg d-luciferin firefly potassium salt (Caliper Life Sciences). BLI of the tumor burden was done at 10 min after injection of luciferin.
[18F]FLT-PET imaging. [18F]FLT-PET and computed tomography (CT) imaging was done with MicroPET Focus F220 Scanner (Siemens Medical Solutions) and GE eXplore MicroCT Scanner (GE Healthcare), respectively. COLO205 tumor-bearing mice were anesthetized and given 250 μCi of [18F]FLT tracer i.v. At 60 min after dosing, PET imaging was done, followed by a subsequent CT scan (5 min) for anatomic reference. Reconstructed PET and CT data were coregistered, and volumes of interest (VOI) were hand-drawn to fit the primary tumor according to the CT and PET data sets. Volumes of interest were overlaid on the PET data for objective PET quantification. Standardized uptake value was calculated using the following formula:
where CPET is the measured activity/cm3 in the volume of interest, ID is the injected dose (μCi), and W is the mouse body weight. In this report, [18F]FLT uptake in tumors was normalized and reported as the ratio of tumor-to-liver SUV.
Data analysis. Tumors were measured two to three times weekly with calipers, and the tumor volume was calculated as [0.5 × (length × width2)]. Statistical analyses of caliper or bioluminescence imaging readouts were conducted using the Prism GraphPad software (GraphPad) for one-way ANOVA analysis followed by the Dunnet's t test.
Results
PF-00477736 enhanced the efficacy of docetaxel in COLO205 and MDA-MB-231 xenograft models. The effect of PF-00477736 on docetaxel treatment was assessed in COLO205 tumor model using 30 mg/kg (Fig. 1A) and 15 mg/kg docetaxel (Fig. 1B), as well as in a MDA-MB-231 model using 15 mg/kg docetaxel (Fig. 1C). A combination study in the MDA-MB-231 model using the maximum tolerated dose (MTD; 30 mg/kg) of docetaxel could not be done, because docetaxel treatment alone caused complete regression. Response to treatment was assessed by tumor growth inhibition and tumor growth delay, and toxicity was assessed by body weight changes in the mouse (COLO205 study). The results of these studies are summarized in Supplementary Table S1.
Within 2 weeks after initial dosing, treatment of PF-00477736 at 7.5 or 15 mg/kg resulted in significant (P < 0.05) enhancement in docetaxel-induced tumor growth inhibition (Supplementary Table S1) in all three settings (Fig. 1A-C). Tumor growth delay was measured using time (days) to reach a fixed tumor size. PF-00477736 at 15 mg/kg significantly (P < 0.05) extended growth delays compared with docetaxel treatment alone in both COLO205 and MDA-MB-231 models. In the COLO205 study, cotreatment of 15 mg/kg PF-00477736 with the MTD (30 mg/kg) of docetaxel caused complete remission in 3 of 12 mice over the whole course of disease (∼94 days), whereas all tumors eventually relapsed in mice treated with docetaxel alone. This result further suggests that PF-00477736 has a strong sensitization effect for docetaxel.
When combined with the MTD of docetaxel, 15 mg/kg PF-00477736 caused an additional 5% body weight loss when toxicity reached a nadir (day 21; Fig. 1D). The PF-00477736–induced sensitization of this side effect seemed to be reversible, as the body weight loss recovered by day 36. When combined with a lower dose of docetaxel (15 mg/kg), both 7.5 and 15 mg/kg PF-00477736 enhanced the antitumor activity without a concomitant increase in toxicity (Supplementary Table S1), showing a PF-00477736–induced therapeutic index increase of docetaxel.
PF-00477736 suppresses docetaxel-induced phosphorylation of histone H3 (Ser10) and potentiates apoptosis. PF-00477736 and docetaxel-mediated changes in the Chk1 signaling pathway were measured by immunofluorescence microscopy in COLO205 cells. Docetaxel activates the mitotic spindle checkpoint, and Chk1 has been shown to mediate spindle checkpoint activation by phosphorylating Aurora B kinase (3), which subsequently phosphorylates histone H3 at Ser10, resulting in cell arrest at M phase. Phosphorylation of histone H2AX at Ser139 (γH2AX), an indicator of DNA damage (6) during apoptosis, was assayed to determine the mechanism of cell death. For the in vitro assay, PF-00477736 was used at 0.36 μmol/L, a biological concentration to suppress Chk1 activity with high selectivity.
As expected, docetaxel treatment (1 nmol/L) alone significantly induced the increase of phosphorylated histone H3 expression at 24 hours (Fig. 2A and B). An increase in γH2AX foci staining was also observed indicative of apoptosis. Simultaneous treatment with PF-00477736 reversed the increase of phosphorylated histone H3 staining and increased the number of γH2AX foci, indicating potentiation of apoptosis. Fluorescence-activated cell sorting analysis further confirmed that PF-00477736 abolished the docetaxel-induced M-phase arrest and induced an increase in apoptosis (Fig. 2C). Thus, PF-00477736 was able to abrogate the mitotic spindle checkpoint and sensitize docetaxel-induced apoptosis. Similar results were observed in HeLa cells (data not shown).
Docetaxel induces phosphorylation of Chk1 (Ser345); PF-00477736 suppresses the cytoplasmic localization of phosphorylated Cdc25C (Ser216) and enhances docetaxel-induced apoptosis. Western blots showed that when COLO205 cells were treated with higher concentrations (5-10 nmol/L) of docetaxel for 8 hours, we observed an increase in the phosphorylation of Chk1 at Ser345 whereas the level of total Chk1 protein remained unchanged (Fig. 3A). The phosphorylation of Chk1 (Ser345), as an indication of up-regulated Chk1 activity during DNA damage checkpoint activation (30, 31), was not observed previously during the taxane-regulated spindle checkpoint activation (3, 19). In the presence of DNA damage, activated Chk1 phosphorylates Cdc25C (Ser216), promotes its sequestration in the cytoplasm, and initiates G2 cell cycle arrest. Immunofluorescence and cell fractionation confirmed these events.
Figures 3B and C depict representative images of immunostaining for each biomarker in cells treated with vehicle alone or 10 nmol/L docetaxel and quantitative assessments for all tested doses of docetaxel, respectively. After 8 hours of treatment with docetaxel, no significant increase in the level of phosphorylated histone H3 was observed for any of the concentrations tested. A dose-dependent increase in γH2AX foci was observed indicative of double-DNA strand breaks. Concurrently, a significant number of cells exhibited cytoplasmic localization of phosphorylated Cdc25C (Ser216) compared with vehicle-treated cells, in agreement with reports that double-stranded DNA breaks induce the activation of the ATM/Chk1 signaling pathway, which leads to the phosphorylation of Cdc25C (Ser216) and its subsequent sequestration in the cytoplasm.
An overnight treatment of PF-00477736 is required to suppress the enzymatic activity of Chk1. Therefore, to test the effect of PF-00477736 on Chk1-mediated DNA damage checkpoint, treatment of docetaxel (5 nmol/L) was stopped by removing the media at 8 hours and cells were subsequently treated with vehicle or PF-00477736 (360 nmol/L) for additional 16 hours (Fig. 3D and E). Docetaxel treatment maintained the cytoplasmic localization of phosphorylated Cdc25C and high levels of γH2AX foci formation but also exhibited elevated level of phosphorylated histone H3, suggesting the activation of DNA damage checkpoint and spindle checkpoint. PF-00477736 cotreatment clearly suppressed the cytoplasmic localization of phosphorylated Cdc25C (Ser216) and enhanced apoptosis as measured by an increase in γH2AX foci, indicating that the disruption of the DNA damage checkpoint contributed to the sensitization of cell death. However, the docetaxel-induced increase in phosphorylated histone H3 was not affected by PF-00477736 treatment.
Pharmacodynamic assessment of Chk1 signaling pathway mediated by PF-00477736 in combination with docetaxel. The mechanism for PF-00477736–induced in vivo efficacy or pharmacodynamic endpoint assessments were done by immunohistochemical analysis. COLO205 tumors were harvested at 24 and 48 hours after the mice were i.p. injected with docetaxel (30 mg/kg) alone or docetaxel in combination with PF-00477736 (15 mg/kg). Selected immunohistochemical images of tumor at 24 hours after treatment and quantitative analyses are shown in Fig. 4.
Docetaxel treatment resulted in a time-dependent increase in γH2AX foci, phosphorylated Chk1 (Ser345) and phosphorylated histone H3 levels (Fig. 4A and B), in agreement with in vitro assay results, at high concentrations of docetaxel. These data further confirmed the involvement of Chk1 in docetaxel-induced activation of DNA damage checkpoint and spindle checkpoint. Levels of γH2AX, phosphorylated Chk1 (Ser345) and phosphorylated histone H3 were diminished at 48 hours compared with 24 hours but remained higher than in untreated controls. Total Chk1 protein remained constant under these conditions (data not shown).
When mice were cotreated with PF-00477736 and docetaxel for 24 hours, the docetaxel-mediated increase in phosphorylated histone H3 was partially reversed. Concurrently, PF-00477736 cotreatment suppressed the cytoplasmic staining of phosphorylated Cdc25C (Ser216) compared with docetaxel treatment alone (Fig. 4C and D). Consistent with previous reports (32), high levels of phosphorylated Cdc25C (Ser216) were observed in vehicle-treated mice, indicating the constitutive expression of Cdc25C in normal cells. The increase in cell number with positive γH2AX foci and caspase-3 staining observed in PF-00477736/docetaxel–cotreated tumors indicates enhancement of apoptosis, providing evidence that abrogation of the Chk1-mediated DNA damage checkpoint and mitotic spindle checkpoint contribute to PF-00477736–induced sensitization to docetaxel therapy.
Bioluminescence imaging showing PF-00477736 enhanced the suppression of tumor cell survival. BLI was used to accurately measure tumor cell survival and treatment response in real time. COLO205 tumor-bearing mice were treated on days 1, 8, and 15. Images of three representative mice from each group on days 2 and 16 are shown in Fig. 5A. On day 2, caliper measurements showed marginal tumor growth inhibition in mice treated with docetaxel (30 mg/kg) alone or in combination with PF-00477736, whereas BLI showed an improved treatment effect in all treated groups except PF-00477736 alone (Fig. 5B, top). Using BLI readout, PF-00477736 (15 mg/kg)–induced enhancement of docetaxel activity against tumor cell survival was detected as early as 24 hours after the first dose (P < 0.05); the difference in caliper measurements between these two groups was marginal. On day 16 (Fig. 5B, bottom), both BLI and caliper measurements showed that PF-00477736 at 7.5 and 15 mg/kg significantly (P < 0.05) enhanced docetaxel-mediated tumor growth inhibition. BLI allowed an early assessment of PF-00477736 treatment response and also offered a robust measure of PF-00477736–induced potentiation of docetaxel efficacy.
[18F]FLT-PET imaging. To assess treatment-induced antiproliferation, COLO205 tumor-bearing mice were scanned using CT and [18F]FLT-PET imaging at 24 hours after each treatment on days 1, 8, and 15. PET and CT images of one representative mouse from each group are displayed in Fig. 6A. Figure 6B shows the tumor-to-liver standardized uptake value ratio of [18F]FLT uptake on selected days. Mice treated with docetaxel alone or in combination with PF-00477736 displayed a significant reduction in [18F]FLT uptake compared with vehicle-treated or PF-00477736–treated groups on days 2 and 16, suggesting that proliferation was suppressed in both groups. However, no difference in [18F]FLT uptake was observed between the groups treated with docetaxel alone and in combination with PF-00477736.
Discussion
Here, we have shown that PF-00477736 displayed potentiation of the docetaxel-induced efficacy in tumor cells and xenografts. In vitro and in vivo mechanistic assessment showed that the sensitization by PF-00477736 was associated with the modulation of the Chk1 signaling pathway, suggesting that Chk1 inhibitors could enhance the therapeutic window of docetaxel. From a translational research perspective, we have used [18F]FLT-PET imaging and immunohistochemical approaches to link Chk1-mediated target or mechanism biomarkers to PF-00477736–induced sensitization. This preclinical work suggests a novel therapeutic opportunity for the clinical development of PF-00477736 or Chk1 inhibitors.
Potentiation induced by PF-00477736 was observed in multiple xenograft models with different molecular backgrounds, including MDA-MB-231and COLO205. Pharmacokinetic analysis indicated that there was no increased plasma exposure to PF-00477736 or docetaxel in the combination setting compared with either single agent (data not shown), indicating that potentiation was induced by PF-00477736–mediated biological effects. In the COLO205 tumor model, PF-00477736 at 7.5 and 15 mg/kg displayed significant (P < 0.05) potentiation, measured by tumor growth inhibition when combined with docetaxel both at MTD (30 mg/kg) and below MTD (15 mg/kg). BLI, which offers a more accurate measurement of viable tumor cells in real time, further confirmed significant sensitization (P < 0.05) by PF-00477736 at both doses. In the tested models, PF-00477736 significantly increased the efficacy of sub-MTD (15 mg/kg) docetaxel without noticeably increasing adverse effects, suggesting that this Chk1 inhibitor primarily sensitized tumor cells and did not significantly affect normal tissues. As docetaxel is one of the most highly effective “standard-of-care” spindle poisons, increasing the therapeutic window of this agent through combination with Chk1 inhibitors could potentially benefit a broad spectrum of patient population.
Mechanistic studies were done in COLO205 cells. At 24 hours after treatment, docetaxel at or below nanomolar concentrations (0.5-1 nmol/L) induced an increase in phosphorylated histone H3 levels and M-phase arrest (Fig. 2), suggesting that the mitotic spindle checkpoint was activated (33, 34). When cells were exposed to higher concentrations (5-10 nmol/L) of docetaxel, we observed concurrent increases in double-stranded DNA breaks (γH2AX foci), cytoplasmic staining of phosphorylated Cdc25C (Ser216), and phosphorylation of Chk1 (Ser345), which was not observed at low concentrations, within 8 hours of treatment. These effects were maintained for an additional of 16 hours (24 hours in total) after drug removal.
Additionally, cells exhibited a significantly increased level of phosphorylated histone H3 at 24 hours compared with vehicle-treated cells. These results were further confirmed in the COLO205 tumor model. As the plasma concentrations of docetaxel at 30 mg/kg were maintained at higher than 10 nmol/L for a short period (data not shown), docetaxel treatment resulted in increased levels of γH2AX foci, phosphorylated Chk1 (Ser345), phosphorylated histone H3, and cytoplasmic sequestration of phosphorylated Cdc25C (Ser216) at 24 and 48 hours. The phosphorylation of Chk1 at Ser345, indicative of Chk1 activation through the double-strand DNA break-regulated ATM signaling pathway, occurs in G2 phase (30, 31, 35). Upon activation, Chk1 phosphorylates Cdc25C (Ser216), causes its sequestration in the cytoplasm, and activates the G2-M checkpoint (5). No increase in phosphorylated Chk1 (Ser345) has been observed for taxane-regulated spindle checkpoint activation (3, 19).
Collectively, these data suggest that, at high concentrations, docetaxel not only serves as an antimicrotubule agent to activate the mitotic spindle checkpoint, as measured by the increased level of phosphorylated histone H3 (Figs. 3E and 4), but also elicits the activation of DNA damage checkpoint in a cell cycle–independent manner. The data are consistent with previous reports that a high concentration of docetaxel induces rapid and abundant onset of apoptosis through multiple apoptotic signal transduction pathways (18, 36) and results in DNA damage checkpoint activation. However, we cannot exclude the possibility that the phosphorylation of Chk1 at Ser345 can directly contribute to the activation of the mitotic checkpoint or through cross-talk between the DNA damage and spindle checkpoints (37). The pleiotropy of the cytotoxic mechanism (17) allows docetaxel to trigger the activation of multiple Chk1 signaling pathways, presumably leading to increased vulnerability of S-G2-M cells to Chk1 inhibitors. Therefore, combination with a Chk1 inhibitor seems to be a rational approach to enhance the therapeutic window of docetaxel.
PF-00477736 was profiled for its ability to disrupt the spindle checkpoint. Chk1 was recently reported to be critically involved in taxane-mediated spindle checkpoint activation, in which Aurora B is activated as a downstream effect (3). Docetaxel (1 nmol/L) treatment leads to an increase in phosphorylated histone H3 level, indicating that Aurora B kinase has been activated (38, 39). In COLO205 cells and xenografts, PF-00477736 cotreatment with docetaxel reversed this increase in phosphorylated histone H3 and corresponded to an increase in apoptosis, suggesting that cell death followed the M-phase arrest. Because PF-00477736 alone did not exhibit any phenotypic modulation of the Aurora kinase pathways (14) nor down-regulation of Chk1 protein (data not shown), we conclude that the diminished level of phosphorylated histone H3 indicates that PF-00477736 abrogated the docetaxel-activated spindle checkpoint to enhance apoptosis.
To assess the effect of PF-00477736 on docetaxel-mediated DNA damage checkpoint activation, COLO205 cells were treated with PF-00477736 for 16 hours (Fig. 3D and E) after a short-term (8 hours) exposure to high doses of docetaxel. The results showed that cells maintained high levels of phosphorylated Cdc25C cytoplasmic staining in the absence of PF-00477736, whereas PF-00477736 treatment suppressed the cytoplasmic staining of phosphorylated Cdc25C with a concurrent increase in γH2AX foci level. This indicated that the PF-00477736–induced enhancement of apoptosis was due to the abrogation of the DNA damage checkpoint. A similar outcome was shown in COLO205 xenografts, as PF-00477736 cotreatment suppressed the cytoplasmic staining of phosphorylated Cdc25C (Ser216) and enhanced docetaxel-induced apoptosis, as shown by the increase in the number of γH2AX foci and level of caspase-3 (Fig. 4). PF-00477736 showed an ability to abrogate the DNA damage checkpoint and enhance docetaxel-induced apoptosis, as was reported for PF-00477736 in combination with other DNA damage agents (14).
PF-00477736 also abolished the docetaxel-induced increase in phosphorylated histone H3 in a time- and dose-dependent manner. When COLO205 cells were treated with high concentrations of docetaxel for 8 hours, followed by a 16-hour treatment of PF-00477736 (Fig. 3E), PF-00477736 failed to suppress the docetaxel-induced increase in phosphorylated histone H3. The same observation holds true in the immunohistochemical examination of COLO205 tumors harvested at 48 hours after cotreatment (data not shown), in which suppression was seen at 24 hours (Fig. 4C and D). Under these conditions, cells or tumors treated with docetaxel alone or in combination with PF-00477736 treatment continued to express higher levels of phosphorylated histone H3 than those treated with vehicle or PF-00477736 alone. In these situations, PF-00477736–induced sensitization of apoptosis was observed, possibly due to its dual function in abrogating the DNA damage checkpoint and spindle checkpoint. This explanation is supported by the observation that PF-00477736 affects both the S and G2 checkpoints by blocking DNA damage agent-induced Cdc25C phosphorylation and causing premature mitotic entry (14). However, PF-00477736 also showed an ability to abrogate the spindle checkpoint and induce a reduction in mitosis. Therefore, the net change of phosphorylated histone H3 in samples treated with docetaxel in the presence and absence of PF-00477736 would depend on how much the individual checkpoint is involved at a given time point or dose of docetaxel. In COLO205 tumor-bearing mice, the plasma level of docetaxel fluctuates over a wide range due to the nature of i.p. administration, resulting in the time- and dose-dependent suppression of phosphorylated histone H3 by PF-00477736. The discrepancy in the timing of PF-00477736–induced phosphorylated histone H3 suppression between in vitro and in vivo settings may stem from differences in the timing and duration of drug exposure.
Bioluminescence imaging indicated that cotreatment of PF-00477736 and docetaxel caused a greater reduction in viable tumor cells than docetaxel treatment alone in the COLO205 tumor model (Fig. 5). As BLI output is highly correlated with metabolically active cells (26), it provides a more accurate readout of real-time tumor viability compared with caliper assessment alone. The sensitization induced by PF-00477736 (15 mg/kg) was observed via BLI as early as 24 hours after the first treatment, and significant potentiation was observed on day 16 in a dose-dependent manner. These data further confirmed the PF-00477736 enhancement of tumor cell death when combined with docetaxel.
However, PF-00477736–induced potentiation of docetaxel was not reflected in the [18F]FLT-PET imaging study. The uptake of FLT correlates well with the activity of thymidine kinase 1, which is highly expressed in cancer cells during S phase (21, 40). Therefore, FLT uptake predominately reflects the fraction of proliferating cells. As has been reported frequently, docetaxel alone induced significant inhibition of [18F]FLT uptake in tumor cells (41, 42). Our study results (Fig. 6) showed that docetaxel alone and in combination with PF-00477736 substantially suppressed [18F]FLT uptake (tumor-to-liver ratio ∼1.3) compared with the vehicle and PF-00477736 alone (tumor-to-liver ratio ∼2.0). However, [18F]FLT uptake values did not differ significantly between animals treated with docetaxel alone and in combination with PF-00477736. Even at a late stage of the study (day 16), when PF-00477736 showed substantial sensitization to suppression of tumor cell survival as measured by BLI, we observed no additional suppression of [18F] FLT uptake. Although it enhanced docetaxel-induced cellular lethality, PF-00477736 showed no effect on the fraction of cells in S-phase by fluorescence-activated cell sorting analysis (Fig. 2C). These data suggest that the prime causes of sensitization are the abrogation of the G2-M and spindle checkpoints, therefore corresponding cell death would occur during G2-M transition, M phase, or M-phase exit. Consequently, although PF-00477736 enhanced docetaxel-induced tumor cell death as shown by the activation of caspase-3, increase in the number of γH2AX foci and BLI of deceased viable cells, the S-phase cell population, was minimally affected by combination treatment, resulting in no further reduction in [18F]FLT uptake by PF-00477736.
Validation of target-associated translational biomarkers for PF-00477736 or other Chk1 inhibitors (15) with high specificity and sensitivity presents a unique challenge. No direct substrate for Chk1 during spindle checkpoint activation has been identified (37). PF-00477736–induced suppression of phosphorylated histone H3 in combination with docetaxel may not be an ideal biomarker due to its high dose and time dependency. The modulation of phosphorylated Cdc25C (Ser216) was frequently reported as a more direct indicator of Chk1 inhibition during DNA damage checkpoint activation (1, 15). However, it has yet to be validated in a clinical setting, due to its nonspecific constitutive phosphorylation by other kinases in undamaged cells (32). As the overall potentiation of docetaxel is generated by the combined effects of Chk1-mediated spindle and DNA damage checkpoints, γ-H2AX increases as a shared downstream effect may serve as a more robust pharmacodynamic indicator for the effects of PF-00477736. Moreover, a γ-H2AX assay for clinical translation has already been established in human tumor tissues (43).
In summary, our work has showed that the Chk1 inhibitor PF-00477736 sensitizes the efficacy of docetaxel. Mechanistic studies revealed that this overall sensitization results from the combined disruption of the Chk1-mediated mitotic spindle checkpoint and the DNA damage checkpoint. We also used multiple approaches, including immunohistochemical and [18F]FLT-PET imaging, to offer some translational perspective for the development of Chk1-mediated biomarkers. Future studies are warranted to optimize the dosing schedule of PF-00477736 for maximizing its therapeutic potential without adding additional dose-limiting toxicities that are commonly associated with taxanes and will provide additional insights for the clinical utility of Chk1 inhibitors in combination with taxanes.
Disclosure of Potential Conflicts of Interest
C. Zhang, Z. Yan, C.L. Painter, Q. Zhang, E. Chen, K. Kuszpit, K. Zasadny, M. Hallin, J. Hallin, A. Wong, D. Buckman, M. Qiu, and J.G. Christensen were employed by Pfizer, Inc., at the time of the study and are currently employees of Pfizer, Inc. M.E. Arango, G. Sun, and K. Anderes were employed by Pfizer, Inc., at the time of the study.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/).