|Home | About | Journals | Submit | Contact Us | Français|
Inhibition of checkpoint kinase 1 (Chk1) has been shown to enhance the cytotoxicity of DNA damaging targeted chemotherapy through cell cycle checkpoint abrogation and impaired DNA damage repair. A novel Chk1/2 inhibitor, AZD7762 was evaluated for potential enhancement of radiosensitivity for human tumor cells in vitro and in vivo xenografts.
Survival of both p53 wild type and mutant human cell lines was evaluated by clonogenic assay. Dose modification factors (DMF) were determined from survival curves (ratio of radiation doses for control versus drug-treated at 10% survival). Flow cytometry, western blot, and radiation-induced tumor regrowth delay assays were conducted.
AZD7762 treatment enhanced radiosensitivity of p53 mutated tumor cell lines (DMFs ranging from 1.6–1.7) to a greater extent than for p53 wild type tumor lines (DMFs ranging from 1.1–1.2). AZD7762 treatment alone exhibited little cytotoxicity to any of the cell lines and did not enhance the radiosensitivity of normal human fibroblasts (SF1522). AZD7762 treatment abrogated radiation-induced G2 delay, inhibited radiation damage repair (assessed by γ-H2AX), and suppressed radiation-induced cyclin B expression. HT29 xenografts exposed to 5 daily radiation fractions and 2 daily AZD7762 doses exhibited significant radiation enhancement compared to radiation alone.
AZD7762 effectively enhanced the radiosensitivity of mutated p53 tumor cell lines and HT29 xenografts and was without untoward toxicity when administered alone or in combination with radiation. The results of this study support combining AZD7762 with radiation in clinical trials.
Actively proliferating cells experience blocks in the cell cycle after exposure to ionizing radiation. Blocks that occur in G1 and G2 that occur following treatment with radiation and DNA damaging drugs have been referred to as checkpoints and are presumed to allow DNA damage repair prior to further cell cycle progression (1). There has been considerable interest in targeting molecular pathways involved with these checkpoints to inhibit repair, particularly in cancer cells (2, 3). Since nearly half of all human tumors have abnormal p53 and thus are unable to arrest in G1 following DNA damage, attention has primarily focused on the G2 checkpoint (4). There are several lines of evidence suggesting that the G2 checkpoint can be exploited to enhance radiosensitivity. The marked radiosensitivity of Ataxia telangiectasia fibroblasts is related to the lack of G2 arrest (5). Caffeine enhances the radiosensitivity of cells primarily through abrogation of the G2 checkpoint (6). 7-hydroxystaurosporine (UCN01) has been shown to radiosensitize human tumor cells by abrogation of the G2 checkpoint (7); however, UCN01 can target multiple pathways and has been difficult to develop due to its poor drug like properties (8).
Both the G1 and G2 checkpoints are controlled by the ATM/ATR signaling pathway (2). Critical downstream molecules in these pathways are the Chk1 and Chk2 threonine kinases, which facilitate both the G1 and G2 checkpoints. Inhibition of these kinases (particularly Chk1) can result in abrogation of cell cycle progression, premature entry into the cell cycle following DNA damage, and insufficient DNA repair (8). Recently, a novel checkpoint kinase inhibitor (AZD7762) was shown to enhance the cytotoxicity of DNA damaging chemotherapy agents by abrogation of the cell cycle arrest (9). The current study shows that AZD7762 is also a potent radiation sensitizer of p53-compromised cells both in vitro and in vivo. The mechanism of AZD7762 radiosensitization involved inhibition of radiation-induced DNA damage repair; however, the abrogation of the G2 checkpoint was not an absolute requirement for AZD7762-mediated radiosensitization. Collectively, the pre-clinical data presented in this study support evaluation of AZD7762 in human trials as a radiation sensitizer.
A Z D 7 7 6 2 ([3-(carbamoylamino)-5-(3-fluorophenyl)-N-[(3S)-3-piperidyl]thiophene-2 carboxamide) was obtained from AstraZeneca. Stock solutions of AZD7762 (500 μM) in DMSO/PBS were diluted with cell culture medium with a final DMSO concentration of less than 0.1%. For in vivo studies, AZD7762 was dissolved in 11.3% 2-hydroxypropyl-β-cyclodextrin (Sigma-Aldrich) in 0.9% sterile saline at a final concentration of 2.75 mg/ml. Mouse monoclonal anti-phospho histone H2AX (Ser139), clone JBW301, rabbit antiserum to histone H2A (acidic patch) and mouse monoclonal beta-actin were purchased from Millipore (Temecula, CA). Rabbit polyclonal pChk1(Ser296) and pChk1(Ser345) antibodies were purchased from Cell Signaling (Danvers, MA). Mouse monoclonal Chk1 antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA) and mouse monoclonal cyclin B antibody was purchased from BD Transduction Laboratories (San Diego, CA).
The following cell lines were purchased from American Type Culture Collection (Manassas, VA): MCF7 (human breast carcinoma), A549 (human lung adenocarcinoma), H460 (human large cell lung carcinoma), PC3 and DU145 (human prostate carcinoma), MiaPaca2 (human pancreatic carcinoma), and HT29 (human colon carcinoma). Normal human fibroblasts (1522) were purchased from the Coriell Institute for Medical Research (Camden, NJ). PC-Sw (human pancreatic adenocarcinoma) cells were obtained from Dr. William Sindelar (10). The glioblastoma brain tumor cell line (SF-295) was kindly supplied by Dr. Kevin Camphausen. H460-DNp53 cells were constructed by retroviral infection of a dominant-negative p53 construct as previously described (9). All cell lines were grown in RPMI 1640 medium supplemented with 10% fetal calf serum and antibiotics. For cell survival studies, cells were plated (5 × 105 cells/100 mm culture dishes) and incubated for 16 hr at 37°C. AZD7762 was added to the exponentially growing cells 1 hr prior to radiation. A range of radiation doses was delivered to cell samples using an Eldorado 8 cobalt-60 teletherapy unit (Theratronics International Ltd. Kanata, Ontario, Canada) at dose rates of 2.0–2.5 Gy/min. Vehicle control radiation survival curves were conducted in parallel. Twenty-four hours after radiation and drug treatment, cells were trypsinized, counted, plated, and incubated for 10–14 days. Colonies were fixed with methanol/acetic acid (3:1) and stained with crystal violet. Colonies with >50 cells were scored and cell survival determined after correcting for the plating efficiency and for AZD7762 cytotoxicity alone. Survival curve data were fit using a linear-quadratic model according to Albright (11). Survival curves for each cell were repeated 2–3 times. The dose modification factor (DMF) was determined by taking the ratio of radiation doses at the 10% survival level (control radiation dose divided by the AZD7762 treated radiation dose). DMF values > 1 indicate enhancement of radiosensitivity. For plateau phase studies, cells were grown to confluence and maintained without medium change for 3 days after which they were treated with AZD7762/radiation as described above. Flow cytometry analysis confirmed that these cell cultures were enriched in G1 phase.
Abrogation of cell cycle checkpoints was evaluated by flow cytometry. Exponentially growing cells were exposed to a single dose of radiation without or with AZD7762 as described above. Cells were collected as a function of time following radiation. Cells were washed with PBS, trypsinized, fixed in cold 70% ethanol in HBSS, centrifuged at 1000 rpm for 5 min and the supernatant discarded. The pellet was washed in cold PBS and suspended in 1 ml of 20μg/ml of propidium iodide solution containing 0.1% Triton X-100 and 500 ng of DNase free RNase. Cell cycle distribution was immediately analyzed using a BD FACS Calibur (BD Biosciences, San Jose, CA).
Exponentially growing cells were exposed to a single dose of radiation without or with AZD7762 as described above. As a function of time after treatment, cell samples were rinsed with PBS, lysed with RIPA lysis buffer (Santa Cruz Biotechnology) in the presence of sodium orthovanadate and protease inhibitors (Sigma-Aldrich), incubated for 30 min on ice and centrifuged at 14,000 × g, supernatant removed, protein concentration determined (DC Protein Assay (BioRad)), aliquoted and stored at −70°C. For xenograft protein analysis, studies, tumors were snap frozen in liquid nitrogen and stored at −70°C. Tumor pieces were homogenized in ice cold RIPA buffer with protease inhibitors, incubated on ice for 30 min, centrifuged at 10,000 × g for 10 min at 4°C and supernatant was removed and re-centrifuged at 10,000 × g for 30 min. Supernatant was removed, aliquoted, and stored at −70°C. Protein samples of equal amounts were subjected to PAGE on 4–20% Tris-glycine acrylamide gels (Novex-Invitrogen). Following transfer to nitrocellulose samples were probed with primary antibodies (1:200~2000), followed by the appropriate secondary antibody diluted to 1:2000 and visualized by chemiluminescence (PerkinElmer). To confirm equal protein loading and transfer, membranes were stripped by ReBlot Plus (Chemicon) and reprobed using anti-actin antibody (or other control protein antibody). Densitometric analysis was accomplished with image analyzer software coupled with the Fluorchem FC800 system (Alpha Innotech, San Leandro, CA). Density values for each protein were normalized to actin or other control protein values.
Mitotic catastrophe (MC) was assessed using a modified procedure (12). Briefly, H460 and 460DNp53 cells were seeded (60,000 cells/chamber) in 4-well chamber slides and incubated overnight at 37°C. Cells were exposed to AZD7762 (100 nM) for 1 hr and then exposed to 2 Gy. After 24 hours the cell monolayer was rinsed and fresh media added. At 24, 48 and 72 hr, medium was removed from the slides and the cells were fixed with cold methanol for 15 min at −20°C. After a PBS wash, slides were blocked with 1% BSA/5% goat serum/PBS for 1 hr at room temperature followed by incubation with anti-α-tubulin antibody (1:1000, Sigma) in 1% BSA/PBS overnight at 4°C. Texas Red-conjugated secondary antibody (Jackson ImmunoResearch Laboratories, Inc.) was added at a concentration of 1:200 in 1% BSA/PBS and incubated at room temperature for 1 hr followed by PBS washes. Chambers were removed from the slides and 8 μL of DAPI mounting medium (Vector Laboraories) was added. Nuclear fragmentation was defined as the presence of more than two distinct nuclear lobes within a single cell. Two separate experiments were performed, each with 300 cells per sample scored on a Zeiss AxioImager.A1 upright flourescent microscope using Axiovision 4.7.2 software.
Female athymic nude mice, 5–6 weeks of age, bred in the National Cancer Institute Animal Production Area (Frederick, MD), were used for this study. All experiments were carried out under a protocol approved by the National Cancer Institute Animal Care and Use Committee and were in compliancewith the Guide for the Care and Use Of Laboratory Animal Resource,(1996) National Research Council. For radiation re-growth delay studies, 1.0 × 106 HT-29 cells were injected into the subcutaneous space of the right hind leg. Mice were ear-tagged to monitor tumor volume measurements in individual mice. Tumor growth was followed until the diameter of tumor reached 0.6–0.8 mm as measured by caliper. At this point animals were randomized into 4 groups (8 mice/group): control; fractionated radiation; AZD7762 control; and AZD7762 + fractionated radiation. Fractionated radiation treatment consisted of 5 daily 2 Gy fractions (Monday-Friday, total radiation dose 10 Gy). AZD7762 (25 mg/kg) was administered by i.p. injection immediately after each radiation fraction in one study and immediately after each radiation fraction and again 8 hr later in a second study. Tumor volume as a function of time is plotted for the various treatments and represents an average tumor volume for each group. Documenting the tumor volume of each individual mouse enabled the determination of the time required (days) for a tumor to reach 3 times the starting tumor volume. All tumor growth data were fit using an exponential growth equation, the tumor growth time (days) for control animals was calculated, and then subtracted from all treated groups. Standard deviations (SD) of the derived values (treated and control) were obtained using the propagation of error formula (13) and then the SDs were used to calculated the Students t-test and p-values for the differences between the various groups (14).
Activation of pChk1 by radiation was rapid and persisted for several hours post-radiation as shown in Supplementary Fig. S1 for DU145 and HT29 cells. Consistent with this activation profile, pilot studies showed that AZD7762 treatment post-radiation was more effective than pre-treatment protocols and that an AZD7762 concentration of 100 nM yielded maximal radiation enhancement with minimal cytotoxicity alone (data not shown). Subsequently for all in vitro studies AZD7762 (100 nM final concentration) was added to cells 1 hr prior to radiation (to ensure pChk1 inhibition at the time of radiation) and left on for 24 hr after radiation followed by clonogenic survival assessment. AZD7762 enhanced the radiosensitivity of multiple cancer cell lines (Fig. 1, Supplementary Fig. S2, and Table 1). AZD7762 cytotoxicity alone was minimal for all cell lines studied. Radiation DMFs for AZD7762 were substantially greater for cell lines with p53 mutations (HT29, DU145, MiaPaca2). Normal p53 WT human fibroblasts (Fig. 1B) showed no radiosensitization with AZD7762. To further test the dependency of AZD7762-mediated radiation sensitization on p53 status, two H460 cell lines were compared that differed only in their p53 status. As shown in Fig. 1C and D, AZD7762 radiosensitized H460 DN p53 cells to a greater extent than H460 WT cells (DMF = 1.58 versus 1.11, respectively). The radiosensitivity of two human pancreatic and one glioblastoma cell lines was also enhanced by AZD7762 (Table 1). All of the studies described above used exponentially growing cell cultures. When confluent (plateau phase) cultures of HT29 cells were used no radiosensitization by AZD7762 was observed (see Table 1, Supplementary Fig. S3). Compared to exponentially growing HT29 cells, the plateau phase HT29 cells were enriched in the G1 cell cycle phase (45% versus 86%, respectively). Thus, active movement through the cell cycle is necessary for maximal AZD7762 radiation sensitization.
Chk1 inhibition has been shown to result in an abrogation of the G2 checkpoint following treatment with DNA-damaging cytotoxic drugs (9). To determine if AZD7762 might similarly abrogate radiation-induced G2 arrest, flow cytometry studies were conducted for irradiated cells treated with or without AZD7762. A series of flow profiles were generated for several cell lines as a function of time after treatment (Supplementary Fig. S4–6) and the effects of AZD7762 treatment on the radiation-induced G2 arrest are summarized in Fig. 2 and Supplementary Fig. S6A, B. Regardless of the p53 status, all cell lines evaluated exhibited a G2 arrest following radiation treatment. Likewise, AZD7762 abrogated the radiation-induced G2 arrest for all cell lines. Thus, there was no relationship between abrogation of the G2 arrest and AZD7762-mediated radiation sensitization.
To determine the influence of AZD7762 on radiation-induced DNA damage repair and direct DNA damage, phosphorylated γH2AX induction (15) and mitotic catastrophe were evaluated respectively. Fig. 3A and B and Supplementary Fig. S7A and B show the effects of AZD7762 on radiation-induced γH2AX induction for four cell lines. In response to radiation alone, phosphorylated γH2AX levels rapidly increased following radiation (0.5 hr), but with time returned to near control values by 24 hr indicating the repair of DNA double strand breaks. For HT29, DU145, and A549 cells AZD7762 inhibited repair at 8 and 24 hr post-radiation with the most inhibition noted in HT29 and DU145 cells, a small amount repair inhibition in A549 cells and very little inhibition observed for 1522 cells. AZD7762 treatment alone resulted in elevated levels of phosphorylated γ-H2AX at 24 hr in HT29, DU145, and A549 cells (Fig. 3A and Supplementary Fig. S7A and B), suggesting that AZD7762 mediated a DNA damage/repair response. This was not observed in 1522 cells (Fig. 3B). To determine if DNA damage as manifested by nuclear fragmentation was increased by combination of radiation and AZD7762 treatment, mitotic catastrophe (MC) studies were conducted as shown in Fig. 3C, D and Supplementary Fig. S8. For H460 DN p53 cells radiation treatment alone increased MC at 24 hr, but returned to near control values at 48 and 72 hr. The combination of AZD7762 and radiation significantly elevated MC at all time points for H460 DN p53 (Fig. 3C) and HT29 cells (Supplementary Fig. S8). In contrast, there were no significant differences in MC among the various groups for H460 WT p53 cells across the time course. These data clearly demonstrate a correlation between elevated MC and radiosensitization (see Table 1).
HT29 tumor xenografts were next evaluated to determine if AZD7762 would enhance the radiation response in vivo. As shown in Fig. 4A, compared to vehicle control, five daily injections of AZD7762 had no effect on tumor growth. Fractionated radiation (2 Gy × 5) delayed tumor growth and the combination of AZD7762 and fractionated radiation further enhanced the tumor delay; however, the enhancement was not significant (p = 0.37, compared to fractionated radiation). Since only 1 dose of AZD7762 was given after each daily radiation dose, it was questioned whether adequate drug levels were present to inhibit activated Chk1 between the radiation fractions. An in vivo study was conducted to determine the duration of pChk1 activation following a single dose of radiation in HT29 xenografts (see Fig. 4B). Radiation treatment activated pChk1 beginning at 3 hr and persisting to 30 hr compared to unirradiated controls. Based on the first xenograft study and the reported half life of AZD7762 in mice of 1–2 hr (16) another study was conducted where following each daily radiation fraction two injections of AZD7762 were given, immediately after radiation treatment and 8 hr later as shown in Fig. 4C. As was seen in Fig. 4A, AZD7762 treatment alone had little effect on tumor growth while fractionated radiation delayed tumor growth similar that observed in Fig. 4A. The time for tumors to reach 3 times the initially measured tumor volume relative to the control for AZD7762 alone, fractionated radiation, and AZD7762 plus fractionated radiation was 2.3 (p < 0.53), 7.4 (p < 0.07), and 18.7 (p < 0.00014) days, respectively. Relative to fractionated radiation alone, the combination of AZD7762 and fractionated radiation was also highly significant (p < 0.0038). Thus, the combination of AZD7762 and fractionated radiation showed a greater tumor growth delay than the sum of the individual treatments alone. From this study it was concluded that two injections of AZD7762 given immediately after each radiation treatment and again 8 hours later provided longer systemic drug levels for pChk1 inhibition. Lastly, AZD7762 treatment alone or in combination with radiation resulted in no untoward toxicity. Animal weights throughout the study for both groups were similar to the untreated controls.
To identify specific proteins that might be useful in guiding future clinical trials combining radiation with AZD7762, a HT29 xenograft study was conducted. Three proteins were evaluated: γ-H2AX, pChk1, and cyclin B as shown in Fig. 5. As was seen for in vitro studies (Fig. 3A-B; Supplementary Fig. S7A-B), radiation treatment induced γ-H2AX in a time dependent manner returning to near control levels by 24 hr. AZD7762 plus radiation inhibited the return of γ-H2AX levels at 24 hr consistent with radiation repair inhibition. Interestingly, AZD7762 alone induced γ-H2AX at all time points evaluated. Both radiation and AZD7762 activated pChk1. In response to radiation treatment, cyclin B was up-regulated and AZD7762 when combined with radiation clearly decreased this induction across all time points.
Successful cancer treatment with radiation depends heavily on whether a therapeutic gain can be achieved. Sophisticated radiation delivery instrumentation can minimize the normal tissue included in the radiation field; however, invariably normal tissues are included necessitating a need to identify agents that might differentially radiosensitize tumor as opposed to normal tissues. Cytotoxic chemotherapy combined with radiation is currently used to enhance local tumor control at the expense of increasing normal tissue toxicity (17). Ideally what is needed are approaches that result in selective tumor radiosensitization.
The current findings suggest that AZD7762-mediated Chk1/2 inhibition may offer considerable selective tumor radiosensitization. AZD7762 did not exert appreciable cytotoxicity alone both in vitro and in vivo. In addition, the normal human fibroblast cell line 1522 was not radiosensitized by AZD7762, suggesting that other normal tissues would not be radiosensitized by AZD7762. In general there was a relationship between AZD7762-mediated radiation sensitization and the p53 status of the cell line. Cell lines that carried p53 mutations were enhanced to a greater extent than p53 WT lines. This was particularly apparent in the H460 cell line pair, where the only difference between the cell lines was the p53 status. Consistent with the in vitro data for HT29 cells, when AZD7762 and fractionated radiation treatment were evaluated in a HT29 xenograft tumor model, significant enhancement in radiation-induced tumor regrowth delay was observed. It should be noted that AZD7762-mediated enhancement of tumor regrowth delay required two daily doses of AZD7762 separated by 8 hr after each radiation fraction consistent with the prolonged radiation-induced activation of pChk1 (Fig. 4B).
Inhibition of Chk1/2 by AZD7762 has been shown to enhance the cytotoxicity of DNA damaging chemotherapy drugs in part by abrogation of the G2 checkpoint. The enhancement was greater in cell lines with compromised p53 status (9). In the current study, AZD7762 treatment resulted in abrogation of the radiation-induced G2 delay for every cell line tested (Fig. 2, Supplementary Fig. 6A-B), yet normal 1522 cells were not radiosensitized by AZD7762. Thus, abrogation of the radiation-induced G2 checkpoint by AZD7762 was insufficient to explain the mechanism of radiosensitization. Like AZD7762, the mechanism for caffeine-mediated radiosensitization has been largely attributed to abrogation of the G2 checkpoint (18). However, there are reports, which show no relationship between radiation-induced G2 abrogation with caffeine and radiosensitization (19). Other mechanisms identified in the current study that may be more pertinent include the effects of AZD7762 on radiation-induced repair. It has been proposed that Chk1 is required of homologous recombination repair (HRR) (20), which normally occurs in the S and G2 (21). Likewise, another major repair pathway is the non-homologous end joining (NHEJ), which predominantly occurs in G1 (22). Since p53 mutated cells lack a G1 checkpoint, they may be more dependent on HRR as opposed to NHEJ. Wild type p53 cells, expressing both a G1 and G2 checkpoint following radiation treatment, should be capable of utilizing both types of repair. Thus, it would be anticipated that Chk1/2 inhibition would predominantly impact HRR in p53 mutated cells (20). Consistent with this was our findings that AZD7762 inhibited the repair of radiation-induced damage (γ-H2AX) and enhanced mitotic catastrophe which led to greater radiosensitization in p53 mutated cells. Further support for inhibition of HRR by Chk1/2 inhibition comes from plateau phase HT29 cells, which were not radiosensitized by AZD7762 (Supplementary Fig. S3). Plateau phase HT29 cells were predominantly in G1 (86%) during the radiation and AZD7762 treatment. It is interesting to speculate that repair of radiation damage in plateau phase cells would be through and not affected by Chk1/2 inhibition. Studies are ongoing to test this hypothesis.
Several protein biomarkers from xenograft studies were identified as potential surrogates to guide clinical trials with AZD7762 and radiation. As was seen for in vitro studies, AZD7762 alone and radiation activated γ-H2AX levels. AZD7762 combined with radiation inhibited the return of γ-H2AX to normal levels. The reason for the AZD7762 induction of γ-H2AX is not clear; however, it may be activated as a result of replication stress (23, 24). Interestingly, pChk1 was activated by both radiation and AZD7762 alone. The latter might be indicative of a DNA damage response associated with γ-H2AX activation. Lastly, radiation was shown to induce cyclin B and AZD7762 markedly inhibited its induction, consistent with the radiation-induced G2 abrogation observed in in vitro studies. Collectively, various combinations of these markers may give indication that AZD7762 is targeting necessary pathways to illicit tumor radiosensitization in clinical trials.
This research was supported by the Intramural Research Program of the Center for Cancer Research, National Cancer Institute, NIH.
This study reports the evaluation of a novel checkpoint kinase inhibitor (AZD7762) as a radiation sensitizer. AZD7762 inhibits downstream checkpoint kinases 1 and 2 (Chk1/2) of the ATM/ATR DNA damage response pathway. AZD7762 is currently being evaluated in clinical trials to enhance the cytotoxicity of DNA damaging chemotherapy agents. In the current study tumor cell lines and xenografts with p53 mutations were shown to be significantly radiosensitized with AZD7762 treatment; whereas, normal fibroblasts were not. These data suggest that AZD7762 could also be evaluated in clinical trials as a radiation sensitizer in patients with tumors harboring p53 mutations.