The Radiosensitizing Effect of AZD0530 in Glioblastoma and Glioblastoma Stem-like Cells

Abstract

AZD0530, a potent small-molecule inhibitor of Src family kinases, is an anti-cancer drug used in the treatment of various cancers. In the case of Glioblastoma (GBM), where resistance to radiotherapy frequently occurs, Src kinase is known as one of the molecules responsible for imparting radioresistance to GBM. Thus, we evaluated the effect of AZD0530 on the radiosensitivity of human GBM cells and human glioblastoma stem-like cells (GSCs). We show that Src activity of GBM and GSC is increased by radiation and inhibited by AZD0530, and using clonogenic assays, AZD0530 enhances the radiosensitivity of GBM and GSCs. Also, AZD0530 induced a prolongation of radiation-induced γH2AX without specific cell cycle and mitotic index changes suggesting that AZD0530-induced radiosensitization in GBM cells and GSCs results from the inhibition of DNA repair. Additionally, AZD0530 was shown to inhibit the radiation-induced EGFR/PI3K/AKT pathway which in known to promote and regulate radioresistance and survival of GBM cells by radiation. Finally, mice bearing orthotopic xenografts initiated from GBM cells were then used to evaluate the in vivo response to AZD0530 and radiation. The combination of AZD0530 and radiation showed the longest median survival compared to any single modality. Thus, these results show that AZD0530 enhances the radiosensitivity of GBM cells and GSCs and suggests the possibility of AZD0530 as a clinical radiosensitizer for treatment of GBM.

Introduction

Glioblastoma (GBM) is the most aggressive malignant brain tumor of which the median survival is estimated to be 12 to 15 months. This limited survival is mostly due to the resistance and subsequent recurrence of GBM following radiotherapy and other treatments (1). Despite this resistance, surgical resection and radiotherapy remain the standard of care for GBM (2). Although a clear mechanism for the radioresistance of GBM has not yet been established, several studies have found the presence of glioma stem-like cells (GSCs) among the heterogeneous tumor cell population of GBM (3). Like stem-like cells found in other cancers, GSCs can act as potential sources of radioresistance, which has provided motivation for studies elucidating the mechanisms of treatment response and radiation resistance in these cell types (4).

Src is one of nine non-receptor tyrosine kinases that holds a central role in the regulation of multifunctional signaling in GBM (5). Indeed, expression of Src is common in human cancer (6). With respect to GBM, Src is a major component in the processes and pathways that regulate the tumorigenesis of GBM such as proliferation, invasion, migration, and EGF receptor (EGFR), Ras/Raf/MEK, and PI3K/AKT pathways (5, 7, 8). Importantly, Src activation by radiation plays a critical role in increasing radioresistance in GBM, lung, breast cancer, and esophageal squamous cells (ESCC cells) (8–11). In GBM, radiation-induced activation of Src kinase activates the EGFR/PI3K/AKT or PI3K/AKT signaling pathway, contributing significantly to the radiation resistance of GBM (8, 12). Inhibition of Src by small molecule inhibitors in GBM, human head and neck squamous cell carcinoma (HNSCC), and non-small cell lung cancer (NSCLC) suppressed radiation-induced activated Src, resulting in increased radiosensitivity (8, 13, 14). With respect to GBM, the combination the Src inhibitor, Si306, with proton therapy was more effective in treating the U87 glioblastoma cell line compared to proton therapy alone, a finding shown to be a result of Si306 suppression of proton therapy-induced upregulation of PI3K and AKT (15).

Because of the involvement of Src kinase in several cancers, small-molecule inhibitors blocking the ATP binding site of Src have been developed (16) and more recently tested in clinical trials for several human cancers (17). A potent and orally available small-molecule Src inhibitor, AZD0530 (saratinib), has anti-tumor activity against several human cancer cell lines and xenograft models (18, 19). To date, AZD0530 has been shown to reduce metastasis in orthotopic colon cancer and prostate cancer models (20), inhibit proliferation of breast cancer cells (21), and suppress the mobility and invasiveness of anaplastic thyroid, lung, and head and neck squamous cancer cells (22–24). Additionally, AZD0530 is known to confer radiosensitivity to lung cancer cells by inhibiting AKT activity by blocking radiation-induced activation of Src (23). Interestingly, recent reports show that treatment of AZD0530 in ESCC cells, a type of esophageal carcinoma, increases accumulation of γH2AX foci in the nucleus compared to untreated groups, weakening repair of DNA damage and finally reducing radioresistance in ESCC cells (11). However, while anti-cancer and clinical effects of AZD0530 on various cancers have been reported, there is no study on the regulation of radiosensitivity of GBM through AZD0530.

In this study we evaluated the in vitro and in vivo effects of AZD0530 on the radiation sensitivity of GBM and GSCs. The present study indicated that AZD0530 enhances the radiosensitivity of GBM and GSCs in vitro and in vivo. Furthermore, the radiosensitization by AZD0530 in GBM and GSCs was associated with the persistence of radiation-induced γH2AX foci, suggesting an inhibition of the repair of radiation-induced DNA double-strand breaks (DSBs) and inhibition of radiation-induced activation of the EGFR/PI3K/AKT signaling pathway.

Materials and Methods

Cell lines and treatment

The adherent glioblastoma cell line U251 and the glioblastoma stem-like cell line (GSCs - NSC11) were used in this study. U251 cells were obtained from the National Cancer Institute (Frederick, MD) and grown in DMEM (Invitrogen) with 10% fetal bovine serum and maintained at 37°, 5% CO₂. The neurosphere-forming NSC11 (25) cells, isolated from human GBM surgical specimens, were kindly provided by Dr. Frederick Lang (MD Anderson Cancer Center) and maintained in stem cell medium (DMEM/F12), supplemented with B27 (Thermo Fisher Scientific), basic fibroblast growth factor (bFGF), and epidermal growth factor (EGF) (50 ng/ml each Sigma-Aldrich) at 37℃, 5% CO₂. Cells were used less that 6 passages from the most recent verification by IDEXX Bioresearch which also verified no mycoplasma by PCR. For in vitro experiments neurospheres of NSC11 cells were disaggregated into single cell suspensions using TryplE Express (Thermo Fisher Scientific) and seeded onto Biocoat™ poly-l-lysine-coated multi-well plates (Corning) or chamber slides (Sigma-Aldrich). AZD0530, provided by the Developmental Therapeutics Program of the National Cancer Institute (National Cancer Institute - Chemotherapeutic Agents Repository), was reconstituted in DMSO (100 mM) and stored at −20°C. Cells were irradiated as monolayer cultures using a XRad 320 X-ray source (Precision XRay Inc.) at a dose rate of 2.5 Gy/min.

Src kinase assay

Cells were grown to 70 – 80% confluency and treated with different concentrations of AZD0530 depending on the cell line (U251 – 3 μmol/L and NSC11 – 10 μmol/L) for 6 h (U251) or 2 h (NSC11) prior to 2 or 6 Gy irradiation. After irradiation the cells were collected at 2 h (NSC11) and 6 h (U251). Src kinase activity was quantified using a Universal Tyrosine Kinase Assay kit (Takara Bio Inc.) according to the manufacturer’s instructions. Src protein from the lysates of tumor cell lines was collected by immunoprecipitation using an anti-Src (IgG Rabbit; Cell Signaling (#2109)) antibody and Protein G Sepharose™ 4 Fast Flow beads (GE Healthcare. Src kinase activity was assessed by measuring absorbance at 450 nm using a Synergy H1 microplate reader (BioTek).

Clonogenic assay

U251 cells were trypsinized and seeded onto 6-well tissue culture plates (50 to 3,200 cells per well depending radiation dose). After allowing cells time to attach (6 h), cultures received AZD0530 (3 μmol/L) for 6 h prior to 1 - 6 Gy irradiation. 24 h after irradiation, drug containing media was then removed, cells rinsed, and fresh, drug-free media was added. For analysis of GSCs, NSC11 cells were disaggregated into single-cell suspensions and seeded into poly-l-lysine-coated six-well plates (200 to 6,400 cells per well depending on radiation dose). After allowing cells time to attach (16 h), cultures received AZD0530 (10 μmol/L) for 2 h prior to 1 - 4 Gy irradiation. 24 h after irradiation, drug containing media was then removed, cells rinsed, and fresh, drug-free media was added. Colonies were stained with 0.1% crystal violet 10 to 12 days after seeding U251 cells, or 14 to 21 days after seeding NSC11 cells. The number of colonies containing at least 50 cells were counted and the surviving fractions calculated. Data presented are the mean ± SE from at least three independent experiments.

Flow cytometric analysis of cell cycle and mitotic index

Evaluation of cell-cycle phase distribution and mitotic index was performed by flow cytometry. The drug treatment protocols were as described for the clonogenic assays. Following irradiation (2 Gy), cells were harvested at various time points, fixed with 70% ethanol, and stained with propidium iodide (Sigma-Aldrich). DNA content for cell-cycle analysis was determined with a LSR Fortessa™ flow cytometer (BD Biosciences). Mitotic index was evaluated using propidium iodide to identify cells with 4N DNA content and then quantifying the expression of the phosphorylated form of histone H3 in this population (anti-phospho H3 [Ser10], clone 3H10 Alexa Fluor 488 conjugate; Millipore (FCMAB104A4)) (26). Fluorescence was measured using an LSR Fortessa™ flow cytometer.

Immunoblot analysis

Whole-cell pellets of U251 and NSC11 cells were collected, 6h (U251) or 2h (NSC11) after irradiation, for protein extraction in RIPA lysis buffer (Thermo Fisher Scientific). Total protein was quantified using a BCA protein assay (Thermo Fisher Scientific). Proteins were separated by SDS-PAGE (Mini-Protean TGX™ Gels [Bio-Rad]), transferred to a nitrocellulose membrane (Bio-Rad), and probed with antibodies targeting Src (1:1000; Cell Signaling Technology (#2109)), phospho-Src (Y416) (1:1000; Cell Signaling (#2101)), EGFR (1:1000; Cell Signaling (#4267)), phospho-EGFR (Y845) (1:1000; Invitrogen (44-784G)), PI3K (1:1000; Cell Signaling (#4257)), phospho-PI3K (Y458) (1:1000; Cell Signaling (#4228)), AKT (1:1000; Cell Signaling (#9272)), phospho-AKT (T308 and S473) (1:1000; Cell Signaling (#13038 and #9271)), β-actin (1:10000; Cell Signaling (#3700)) and actin (1:5000; Sigma-Aldrich (#T9026)). Bands were visualized with IR Dye secondary antibodies (LI-COR) and quantified using an Odyssey CLx Image System (LI-COR).

Immunofluorescent staining for detection of γH2AX foci and mitotic catastrophe

Visualization of γH2AX foci and mitotic catastrophe was performed as described previously (27). U251 cells were grown on 18×18 mm cover glass and treated with AZD0530 (3 μmol/L) before irradiation (2 Gy). NSC11 cells were grown on 18×18 mm cover glass coated by poly-l-lysine solution (Sigma-Aldrich) and treated with AZD0530 (10 μmol/L) before irradiation (2 Gy). Cells were imaged on a Zeiss upright fluorescent microscope and foci counted in 50 cells per experimental group. Nuclear fragmentation was defined as the presence of ≥ 2 lobes within single cells, and for each condition 100 cells were scored.

Orthotopic in vivo experiments

For the three in vivo studies, U251 cells engineered to express luciferase and GFP (Lentivirus, LVpFUGW-UbC-ffLuc2-eGFP2) were orthotopically implanted into the right striatum of 8-10 week-old athymic female nude mice (nu/nu; NCI Animal Production Program); bioluminescent imaging (BLI) and local irradiation were all performed as previously described (28). At 6 days post-implantation, BLI was detected in all mice, which were then randomized into 4 treatment group. There were 5 mice per group: control (vehicle), radiation (2 Gy for 3 consecutive days), AZD0530 (25 mg/kg once a day × 3 days, oral gavage) and combination (AZD0530 25 mg/kg qd x 3 days, 4h pre-IR + radiation 2 Gy qd × 3 days). Additionally, we extracted the human tumor tissue (U251 cells: Lentivirus, LVpFUGW-UbC-ffLuc2-eGFP2) from the mice brains to perform immunoblot analysis of Src and phospho-Src (Y416). Tumors from mice from each of the 4 treatment groups (2 mice/group): control (vehicle), radiation (4 Gy), AZD0530 (25 mg/kg, oral gavage) and combination (AZD0530 25 mg/kg + IR 4 Gy), was dissected rapidly 18 h after treatment of AZD0530 or 1 h after radiation treatment. Mice were euthanized by CO₂ inhalation, brains rapidly removed, and placed in PBS containing 100 mg/ml cycloheximide. Tumor tissue was then isolated on the basis of GFP fluorescence under a stereoscope and flash frozen in liquid nitrogen. Isolated tumor tissue from mice brains were homogenized in RIPA buffer (Thermo Fisher Scientific) using Biomasher II® Disposable Micro Tissue Homogenizer (BioMasher II). Homogenized tumor tissue samples follow the immunoblot analysis outlined previously. Mice were monitored every day until the onset of neurologic symptoms (morbidity). All experiments were performed as approved by the principles and procedures in the NIH guide for care and use of animals.

Dosimetry

For all irradiation, a Pantak X-ray source was used at a dose rate of 3.59 Gy/min. Dose rate was calibrated based on the procedures described in American Association of Physicist in Medicine (AAPM) Task Group Report 61 (TG-61) with regard to the following conditions: X-ray tube potential was 300 kV, half-value layer (HVL) was 0.8 mm copper, source-to-surface distance (SSD) was 50 cm. Dose rate was measured at 2 cm depth in solid water phantom using a PTW model N23342 ion chamber (Freiburg, Germany) and Inovision® model 35040 electrometer (Chesterfield, MI).

Statistical analysis

GraphPad Prism 6, (GraphPad Software, La Jolla, CA) was used for all analyses. Src kinase activity, γH2AX foci, and mitotic catastrophe were compared according to Student’s t-test. For in vivo survival studies, Kaplan–Meier curves were generated and log-rank values calculated. Sample size was calculated assuming a type 1 error of 5%, 80% power and an effect size of 6 days as 4.8 animals/group.

Results

Radiation increases Src kinase activity in GSCs which is reduced by AZD0530

Src kinase was previously shown to become activated following irradiation in glioblastoma cells (8, 12). To further define the activation of Src kinase by radiation, immunoblotting for phospho-Src (Y416) and a Src kinase assay were performed on GBM (U251) and GSC lines (NSC11) treated with radiation (6 Gy). As shown in Fig. 1A and B, phospho-Src (Y416) protein levels and Src activity were significantly increased in each cell line at distinct times after radiation when compared to unirradiated controls. Following irradiation, Src kinase activity and phospho-Src (Y416) protein levels reached maximum levels at 6 h (U251) and 2 h (NSC11) after which activity and protein levels progressively declined out to 24 h. These data indicate that Src kinase activity in GBM cell lines and GSCs are differentially affected by radiation.

Figure. 1.

The effect of AZD0530 on radiation-induced phosphorylation and activation of Src kinase in GBM and GSCs. A and B), U251 and NSC11 cells were irradiated with 6 Gy and protein was isolated at the indicated time points. A) representative immunoblots for phosphor-Src (Y416) along with the corresponding Src levels from each cell-line. B) Src kinase activity in the U251 and NSC11 cells as a function of time after irradiation. C and D) AZD0530 3 μmol/L (U251) and 10 μmol/L (NSC11) were given for 6 h (U251) or 2 h (NSC11) prior to irradiation. C) immunoblots for phosphor-Src (Y416) at 6 h (U251) and 2 h (NSC11) after irradiation. D) Src kinase activity at 6 h (U251) and 2 h (NSC11) after irradiation. β-actin used as a loading control in A and C. Data are the mean ± SEM for 3 independent experiments and statistical significance was determined by Student’s t-test. *P < 0.05, **P < 0.005 and ***P < 0.0005 (Radiation versus AZD0530 + radiation).

AZD0530 has been previously shown to potently inhibit Src activity (18). To investigate whether AZD0530 attenuates the radiation-induced increase in Src kinase activity, immunoblots for phospho-Src (Y416) (Fig. 1C) and kinase activity assays (Fig. 1D) were measured in U251 and NSC11 cells after 3 μmol/L (U251) or 10 μmol/L (NSC11) of AZD0530 alone, 6 Gy radiation alone, or a combination of AZD0530 with radiation, doses chosen from a dose-finding experiment (supplemental figure 1). Radiation-induced levels of phospho-Src (Y416) were reduced by AZD0530 for both cell lines (Fig 1C). Also, as shown in Fig. 1D, treatment with AZD0530 alone had no effect on baseline Src activity in U251 and NSC11 cells at 6 h (U251) and 2 h (NSC11) when compared with their control. Notably, with radiation alone, the Src kinase activity increased dramatically for both cell lines. However, the addition of AZD0530 reduced this radiation-induced Src activity to baseline. Overall, AZD0530 significantly inhibited post-radiation increases in Src kinase activity.

AZD0530 increases the radiosensitivity of GBM and GSCs

To determine the effect of AZD0530 on the radiosensitivity of these cells, clonogenic cell survival assays were performed. U251 and NSC11 cell cultures were treated with AZD0530 (U251 - 3 μmol/L and NSC11 - 10 μmol/L) for 6 h (U251) or 2 h (NSC11), irradiated (1 - 6 Gy), and fed fresh growth medium at 24 h after irradiation. As shown in Fig. 2, treatment with AZD0530 increased the radiosensitivity of U251 and NSC11 cells with dose-enhancement factors (DEF) at a surviving fraction of 0.1 ranging from 1.35 to 1.47. The plating efficiencies of control (DMSO treated) U251 and NSC11 cells were 0.48 and 0.17, respectively. Treatment with AZD0530 reduced these plating efficiencies to 0.27 and 0.14. The data herein indicates that inhibition of radiation-induced Src kinase activity by AZD0530 increases the radiosensitivity of GBM (U251) and GSCs (NSC11).

Figure. 2.

The effect of AZD0530 on radiosensitivity in GBM and GSCs. Each cell-line was treated with the designated concentrations of AZD0530 for 6 h (U251) or 2 h (NSC11) before radiation. Surviving fraction (Log) curves were generated after normalizing for the cytotoxicity generated by AZD0530 alone. DEF values were calculated at a surviving fraction (Log) of 0.1.

AZD0530 enhances radiation-induced DNA damage without affecting the cell cycle or mitotic index (G2 checkpoint) in GBM and GSCs

To further define the AZD0530-induced radiosensitization in GBM and GSC lines, we evaluated the presence of γH2AX foci, an indicator of DNA damage (29), and their dispersal which correlates with radiation-induced DNA damage repair (30). In this study, cells were exposed to AZD0530 (U251 - 3 μmol/L and NSC11 - 10 μmol/L) for 6 h (U251) or 2 h (NSC11), before radiation (2 Gy), and γH2AX foci determined at 1 and 24 h post-radiation. When U251 and NSC11 cells were pretreated with AZD0530 alone, there was no significant increase in the number of γH2AX foci, indicating the drug alone did not cause DNA damage (Fig. 3). Radiation induced a significant increase in the number of γH2AX foci at 1 h, which decreased by 24 h as DNA damage repair occurred in both cell lines (Fig. 3). At 1 h after irradiation there was no significant difference in the number of γH2AX foci detected between control (DMSO) and AZD0530 treated cells. However, in both cell lines, the addition of AZD0530 with radiation treatment resulted in a significantly higher number of foci at 24 h when compared with cells receiving only radiation consistent with an inhibition of DNA damage repair.

Figure. 3.

The effect of AZD0530 on radiation-induced γH2AX foci in GBM and GSCs. The quantitative assessment of γH2AX foci per cell at 1h and 24 h after radiation is shown. Foci were counted in at least 50 cells per experiment and represent histograph images obtained from control, AZD0530 alone, radiation (2 Gy) alone, and combination treatment of AZD0530 with radiation. Data are the mean ± SEM for 3 independent experiments and statistical significance was determined by Student’s t-test. *P < 0.05 and **P < 0.005 (radiation versus AZD0530 + radiation).

To further elucidate the mechanism of AZD0530-induced radiosensitization in U251 and NSC11 cells, we investigated whether AZD0530 alone or in combination with radiation causes changes in the cell cycle or mitotic index. As shown in Supplementary Fig. S2, there was essentially no difference in the cell cycle distribution for radiation alone compared to the combination of AZD0530 with radiation for each cell line, at every time point. Another potential source of radiosensitization is the abrogation of the G2/M checkpoint (26). As shown in Supplementary Fig. S2B, AZD0530 alone had no effect on the mitotic index nor did it affect the reduction in mitotic index induced by radiation. Together the data presented in Fig. 3 and Supplementary Fig. S2 suggests that treatment with AZD0530 impairs the cellular repair of radiation-induced DNA damage without alteration of cell cycle or abrogation of the G2/M checkpoint.

AZD0530 increases radiation-induced mitotic catastrophe in GBM and GSCs.

Following exposure to radiation, cells with unrepaired DNA-DSBs undergo cell death by mitotic catastrophe. To investigate whether the AZD0530 affects radiation-induced mitotic catastrophe, we performed immunostaining (Fig. 4). The treatment regimens of AZD0530 and radiation were the same as previously used in γH2AX studies (Fig. 3). As shown in Fig. 4, at 1 h the percentage of cells undergoing mitotic catastrophe after treatment with AZD0530 alone was not significantly different compared to control. However, there was a time dependent increase in the number of cells undergoing mitotic catastrophe after the AZD0530 treatment out to 72 h. More importantly, there was a significant increase in mitotic catastrophe with the combination of AZD0530 and radiation at 48 h and 72 h in both U251 and NSC11. Therefore, these data, in combination with previous γH2AX results, suggest that radiosensitization following treatment with AZD0530 is mediated by an attenuation of radiation-induced DNA damage repair, resulting in a greater number of cells undergoing mitotic catastrophe compared with radiation alone.

Figure. 4.

The effect of AZD0530 on radiation-induced mitotic catastrophe in GBM and GSCs. The quantitative assessment of nuclear fraction per cell at 48h and 72 h after radiation is shown. Nuclear fraction (defined as the presence of two or more distinct lobes within a single cell) was evaluated in at least 100 cells per treatment per experiment and represent histograph images obtained from control cells and cell that pretreated AZD0530 alone, radiation (2 Gy) alone and combination treatment of AZD0530 with radiation. Data are the mean ± SEM for 3 to 4 independent experiments. Statistical significance was determined by Student’s t-test. *P < 0.05 (radiation versus AZD0530 + radiation).

AZD0530 inhibits radiation-induced activation of the EGFR/PI3K/AKT signaling pathway

Radiation-induced activation of Src has been shown to activate the EGFR/PI3K/AKT pathway and increases radioresistance in glioblastoma cells (8). Thus, to investigate whether AZD0530 can inhibit this activation of the EGFR/PI3K/AKT signaling pathway, we performed western blotting of cells treated with AZD0530 (U251 - 3 μmol/L and NSC11 - 10 μmol/L) and irradiation (6 Gy) (Fig. 5). As shown in Fig. 5, radiation induced the activation of Src (phospho-Y416), EGFR (phospho-Y845), PI3K (phospho-Y458) and AKT (phospho-T308 and S473) in U251 and NSC11 cells. Treatment with AZD0530 blocked the radiation-induced activation of Src, EGFR, PI3K, and AKT (Fig. 5) in each cell line.

Figure. 5.

The effect of AZD0530 radiation-induced activation of EGFR/PI3K/AKT signaling pathway in GBM and GSCs. Immunoblot to detect Src, p-Src (Y416), EGFR, p-EGFR (Y845), PI3K, p-PI3K (Y458), AKT and p-AKT (T308 and S437) protein from whole cell lysates were performed in U251 and NSC11. Each cell-lines were treated with AZD0530 (U251 - 3 μmol/L and NSC11 – 10 μmol/L) for 6 h (U251) or 2 h (NSC11), and irradiated radiation (6 Gy), and collected at 6 h (U251) and 2 h (NSC11) for analysis. β-actin used as a loading control for immunoblot.

AZD0530 enhancement of radiation in vivo

Finally, to determine whether there was a radiosensitizing effect of AZD0530 under in vivo conditions, we used an orthotopic xenograft mouse model using U251 cells that expressed luciferase and GFP (Lentivirus, LVpFUGW-UbC-ffLuc2-eGFP2) as described previously (28). Mice implanted with U251 cells were randomized according to BLI signal into four groups: control (vehicle), AZD0530 (25 mg/kg × 3), radiation (2 Gy × 3) and combination (AZD0530 (25 mg/kg × 3) + radiation (2 Gy × 3)). Vehicle or AZD0530 (25 mg/kg) were delivered once a day for three consecutive days, and the tumor locally irradiated (2 Gy) 4 h after each drug treatment. Mice were euthanized at the onset of morbidity and Kaplan-Meier survival curves were generated (Fig. 6A). AZD0530 and radiation alone had no effect on mouse survival as compared to control. However, the overall survival of the mice that received the combination of AZD0530 and radiation was increased as compared to control, AZD0530, or radiation alone, demonstrating that AZD0530 enhances the radioresponse of U251 orthotopic xenografts (p=0.032). In addition, to evaluate whether the increased survival of the combination (AZD0530 and radiation) group was caused by AZD0530, we performed immunoblot assays of Src and phospho-Src (Y416) using implanted tumor tissue from a second set of mice (Fig. 6B). Immunoblot analysis of Src and phospho-Src revealed that increased phospho-Src (Y416) protein level in the implanted tumors by radiation was effectively inhibited by AZD0530. These data suggest that AZD0530 crosses the blood-brain-barrier, inhibits irradiation-induced Src activity, and leads to an enhancement of radiosensitivity in vivo.

Figure. 6.

The effect of AZD0530 on the radioresponse of orthotopic xenografts initiated from GBM. A and B) at day 6 post orthotopic implant, mice were BLI and randomized. A) mice were treated with vehicle or AZD0530 (25 mg/kg) and delivered once a day 4 h prior to radiation. Tumor was locally irradiated (2 Gy) after each drug treatment, and this process was carried out for 3 consecutive days. Kaplan-Meier survival curves of different treatment group Control (vehicle or PBS), AZD0530, radiation and Combo (AZD0530 + radiation). B) mice were treated with vehicle or AZD0530 (25 mg/kg) 4h prior to radiation. Tumor was locally irradiated (4 Gy) and all tumors were collected 1h after irradiation for immunoblot analysis of Src and phospho-Src (Y416) .

Discussion

Glioblastoma failure after radiochemotherapy typically occurs in the first 7 months post-treatment and within the highest dose radiation field indicating the need to develop targeted agents to enhance the effect of radiotherapy. Radiation-induced activation of Src kinase has been associated with such pathways as DNA repair, cell-cycle checkpoint blockade, and the EGFR/PI3K/AKT signaling axis (8, 31, 32). Though typical hierarchical pathways have EGFR above SRC more recent data has shown that SRC can phosphorylate EGFR as well as it’s traditional downstream targets of PI3K and AKT (33–35). Inhibition of Src kinase activity through Src kinase inhibitors such as Dasatinib, PP2, and Bosutinib have shown anti-cancer effects in multiple tumor lines and enhanced the effects of radiotherapy in non-GBM models (14, 36). AZD0530 (Saracatinib) is a potent, orally administered small molecule that inhibits Src by blocking its ATP binding site and has been tested in multiple anti-cancer clinical trials without benefit (37). However, it has gained FDA approval as an inhibitor of Idiopathic Pulmonary Fibrosis (38). Previous data has shown that AZD0530 can increase radiosensitivity in lung cancer (23) and esophageal cancer (11), however there are no studies investigating AZD0530 as a radiosensitizer in GBM. Therefore, we evaluated in vitro and in vivo whether AZD0530 could serve as a radiosensitizer in human GBM and GSCs. Our study found that the activation and phosphorylation (p-Src (Y416)) of Src kinase by radiation was inhibited by AZD0530 in human GBM cells and GSCs. This inhibition translated to an increase in sensitivity to radiation for each cell line tested accompanied by prolongation of γH2AX foci, increased mitotic catastrophe and inhibition of the radiation-activated Src-EGFR-PI3K-AKT signaling axis. In this study, we showed that AZD0530, a small molecule inhibitor of Src kinase, enhances the radiosensitivity of human GBM cells and GSCs.

Whether our findings of the radiosensitizing properties of AZD0530 are clinically applicable will require additional studies. The first is the general activity of AZD0530 within the CNS. Initial preclinical studies of AZD0530 as a potential treatment for Alzheimer’s disease (AD) reported pharmacokinetics indicative of BBB penetration and reduced Fyn protein in the brains of an AD mouse model (39). Herein, we showed that AZD0530 effectively inhibited radiation-induced phospho-Src (Y416), using immunoblot analysis of tumors from mice bearing orthotopic GBM xenografts, indicating that AZD0530 was able to cross the BBB. These two findings are suggestive of AZD0530 crossing the BBB but a human study proving this is needed. The second additional study is the impact of SRC inhibition on temozolomide efficacy as temozolomide is part of the standard of care therapy in GBM. One preclinical study showed an additive effect of SRC inhibition and temozolomide treatment against GBM cell lines in vitro, however, no clinical data exists (40). The third additional study is the development of EGFR/Src/PI3K/AKT inhibition as radiosensitizing agents for GBM. Two studies of EGFR inhibitors (gefitinib and vandetanib) failed to show a benefit when used in combination with irradiation in patients with GBM (41, 42). However, because SRC inhibition can lead to the inhibition of the radiation-induced phosphorylation of EGFR as well as PI3K and AKT it might prove more efficacious that strict EGFR inhibition. The fourth set of data needed to plan a clinical trial with AZD0530 is the side effect profile of the drug. In a recently completed clinical trial of AZD0530 in patients with metastatic sarcoma to the lung the most common grade 3 side effect of daily treatment with saracatinib was hypophosphatemia, which was reversible with supplementation (37). In the EGFR inhibitor study using gefitinib the side effects were primarily dermatologic and GI (41). Taken together it would be expected that daily AZD0530 might cause GI toxicity and possibly hypophosphatemia, side effects that do not overlap with the most common side effects of radiochemotherapy for patients with GBM. Taken together, it appears AZD0530 could be used in combination with RT/Temodar without overlapping side effects and has the potential to be a potent radiation sensitizer. Therefore, these preclinical data indicate that inhibition of Src kinase activity by AZD0530 is a viable strategy to improve the radiosensitivity of human GBM.