Oncogene (2015), 1–9 © 2015 Macmillan Publishers Limited All rights reserved 0950-9232/15 www.nature.com/onc
ORIGINAL ARTICLE
Drug repurposing: sulfasalazine sensitizes gliomas to gamma knife radiosurgery by blocking cystine uptake through system Xc−, leading to glutathione depletion L Sleire1, BS Skeie2,3, IA Netland1, HE Førde1, E Dodoo4, F Selheim5, L Leiss1,6, JI Heggdal7, P-H Pedersen2,3, J Wang1,8 and PØ Enger1,3,8 Glioblastomas (GBMs) are aggressive brain tumors that always recur after radiotherapy. Cystine, mainly provided by the system X−c antiporter, is a requirement for glioma cell synthesis of glutathione (GSH) which has a critical role in scavenging free radicals, for example, after radiotherapy. Thus, we hypothesized that the X−c -inhibitor sulfasalazine (SAS) could potentiate the efficacy of radiotherapy against gliomas. Here, we show that the catalytic subunit of system X−c , xCT, was uniformly expressed in a panel of 30 human GBM biopsies. SAS treatment significantly reduced cystine uptake and GSH levels, whereas it significantly increased the levels of reactive oxygen species (ROS) in glioma cells in vitro. Furthermore, SAS and radiation synergistically increased DNA doublestrand breaks and increased glioma cell death, whereas adding the antioxidant N-acetyl-L-cysteine (NAC) reversed cell death. Moreover, SAS and gamma knife radiosurgery (GKRS) synergistically prolonged survival in nude rats harboring human GBM xenografts, compared with controls or either treatment alone. In conclusion, SAS effectively blocks cystine uptake in glioma cells in vitro, leading to GSH depletion and increased ROS levels, DNA damage and cell death. Moreover, it potentiates the anti-tumor efficacy of GKRS in rats with human GBM xenografts, providing a survival benefit. Thus, SAS may have a role as a radiosensitizer to enhance the efficacy of current radiotherapies for glioma patients. Oncogene advance online publication, 23 March 2015; doi:10.1038/onc.2015.60
INTRODUCTION Glioblastomas (GBMs) are lethal cancers and inherently resistant to radiotherapy. Established treatments including surgery, radio- and chemotherapy have a limited efficacy, and the median survival is ~ 14.6 months.1 Thus, treatment resistance represents a major challenge in the clinical management of these patients, and new therapies are urgently needed. The effect of radiotherapy is mediated by creation of reactive oxygen species (ROS) that cause DNA damage, subsequently leading to apoptosis when cells pass through check points during the cell cycle. However, ROS are also byproducts of the normal cellular metabolism, and low concentrations are required for various cellular processes in healthy cells.2 Thus, ROS levels are tightly regulated, and antioxidants are crucial in maintaining redox homeostasis.3 Moreover, studies show that cancer cells synthesize antioxidants at increased rates 4 to scavenge free radicals that are produced for example, under hypoxia 5 or following radiotherapy.6 Notably, combined magnetic resonance imaging (MRI) and immuno-spin trapping for detection of free radicals in experimental glioma models, have demonstrated increased levels in vivo, compared with normal mouse brain.7 Glutathione (GSH) is one of the most important antioxidants in mammalian cells, and is abundantly present in human brain tumors.8 Availability of cystine is the rate-limiting step in GSHsynthesis in glioma cells, and is provided through system X−c .9,10 1
In exchange for cystine uptake, this antiporter releases intracellular glutamate which impacts on multiple aspects of glioma biology.11,12 As such, glutamate reportedly promotes glioma cell growth 13 and has been linked to tissue destruction 14 and tumorassociated epilepsy.15 Studies in glioma cells have shown that system X−c is upregulated under oxidative stress,16 and Takeuchi et al.17 reported that increased X−c expression in human GBMs was associated with a shorter survival. Intriguingly, sulfasalazine (SAS), a drug approved for treatment of rheumatoid arthritis 18 and inflammatory bowel diseases,19 has been shown to block the X−c antiport system, thereby inhibiting glutamate release.20 Notably, it also suppresses growth of experimental gliomas,21 an effect that has been attributed to reduced synthesis of GSH. However, whereas these studies imply major roles for both GSH and glutamate in glioma progression, the therapeutic potential of blocking GSH-synthesis concomitantly with treatment-induced oxidative stress has received relatively little attention. As GSH represents a major defense system against oxidative stress, and is the principal antioxidant in glioma cells,11,21,22 its depletion may increase glioma cell vulnerability to increased ROS levels, such as during radiotherapy. We therefore hypothesized that SAS may potentiate the effect of radiotherapy against glioma cells. The current primary radiation treatment for GBMs includes fractionated radiation to 60 Gy, and has been shown to prolong survival compared with the untreated patients.1,23 The patients do,
Department of Biomedicine, Oncomatrix Research Lab, University of Bergen, Bergen, Norway; 2Department of Clinical Medicine, K1, University of Bergen, Bergen, Norway; Department of Neurosurgery, Haukeland University Hospital, Bergen, Norway; 4Department of Neurosurgery, Karolinska University Hospital, Stockholm, Sweden; 5Department of Biomedicine, Proteomics Unit (PROBE), University of Bergen, Bergen, Norway; 6Neuro Clinic, Haukeland University Hospital, Bergen, Norway and 7Department of Oncology and Medical Physics, Haukeland University Hospital, Bergen, Norway. Correspondence: Professor PØ Enger, Department of Biomedicine, Oncomatrix Research Lab, University of Bergen, Jonas Lies vei 91, Bergen 5020, Norway. E-mail: [email protected] 8 These authors contributed equally to this work. Received 11 July 2014; revised 27 November 2014; accepted 16 December 2014 3
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
2 however, relapse with focal tumors and may experience severe cognitive effects because of radiation damage of healthy brain tissue surrounding the tumor.24 Alternatively, gamma knife radiosurgery (GKRS) is a radiation treatment characterized by a sharp dose fall-off outside the tumor target that may limit cognitive deterioration.25 In contrast to the fractionated radiotherapy, GKRS is a single session treatment and is currently offered at several treatment centers to patients with local GBM recurrences.26,27 Previously, we have established an animal model for GKRS in order to investigate different dose regimens and accompanying radiobiological effects.28 In this study, we used this model to study the effects of GKRS alone, or in combination with SAS, thereby validating this compound as a radiosensitizer for gliomas. RESULTS xCT is expressed in human GBMs and glioma cell lines Initially, we performed immunohistochemistry for xCT, the catalytic subunit of system X−c , in 30 GBM patient biopsies (Figure 1a and Supplementary Figure 1a). All samples expressed xCT and displayed a homogeneous predominantly cytoplasmic/ membrane staining confined to the tumor cells, whereas the vasculature showed no visible staining. As the samples expressed xCT at varying levels, three independent observers scored each section (0–3) for xCT expression. A total of 28 of the 30 sections expressed xCT at moderate-to-high levels (Supplementary Figure 1b). Notably, xCT expression in sections from normal brain could hardly be detected (Figure 1a, left panels). Moreover, western blotting confirmed xCT expression in protein lysates from three GBM tissue biopsies (Figure 1b), as well as in a panel of seven glioma cell lines (Figure 1c). SAS inhibits the growth of glioma cells and blocks the uptake of cystine, leading to GSH depletion and increased ROS levels As immunohistochemistry and western blotting confirmed the expression of the system X−c in glioma biopsies and cell lines, we next wanted to investigate the effects of pharmacologically blocking this antiporter. Thus, we established monolayer cultures with the A172, U251 and LN18 glioma cell lines that were treated
with escalating doses of SAS. We performed ECIS measurements to monitor cell growth longitudinally and observed that controls and glioma cell cultures receiving SAS treatment at different doses produced distinct impedance curves (Figure 2a). These growth curves therefore established a dose–response relationship for the glioma cell lines tested. We then performed measurements of carbon-14 labeled cystine uptake in the A172, LN18 and U251 glioma cell lines preincubated with or without SAS (Figure 2b). SAS reduced cystine uptake significantly in all the cell lines in a dose-dependent manner when added at concentrations 250, 500 and 1000 μM. In a similar way, we observed a significant and dose-dependent reduction in GSH levels in the same glioma cell lines with increasing doses of SAS (Figure 2c). Conversely, assessment of ROS levels showed a consistent increase in ROS that accompanied the reduction in cystin uptake and GSH levels (Figure 2d). In the LN18 and U251 glioma cell lines we again observed a significant dosedependent effect of SAS. Although ROS levels also increased with the concentration of SAS for the A172 cell line, only the highest concentration of SAS produced significantly higher ROS levels than the other groups. Furthermore, adding the antioxidant N-Acetyl-L-Cysteine (NAC) abrogated the effects on ROS levels in all cell lines tested (Figure 2d). SAS and radiation synergistically increases DNA damage in glioma cells As the ROS levels consistently increased following SAS treatment, we wanted to see if SAS also caused increased DNA damage, alone or in combination with radiation. We therefore quantified double-strand DNA breaks by measuring phosphorylation of Histone 2AX (Figure 3a). We conducted the experiment using non-irradiated or irradiated glioma cells grown in medium only, or with increasing concentrations of SAS. Whereas SAS at 500 μM increased DNA damage, 8 Gy radiation without SAS only caused a marginal, insignificant increase in DNA damage. In the presence of 1000 μM SAS however, radiation significantly increased the levels of DNA damage, both in the U251 and LN18 glioma cell lines. In order to further elucidate the biological effects of SAS treatment, alone or in combination with radiation, we assessed cell viability for the U251, LN18 and A172 glioma cell lines using the MTS assay (Figure 3b). In these, radiation alone or 250 μM SAS
Figure 1. Expression of xCT in human GBM biopsies and glioma cell lines. (a) immunohistochemistry (IHC) staining for xCT (brown) in normal brain and human GBM biopsies as indicated, at x10 (upper panels) and x40 magnification (lower panels). Nuclear counterstaining: hematoxylin (blue), scale bars = x10, 100 μM; x40, 50 μM. (b) Western blot for xCT in three GBM biopsies and (c) human glioma cell lines as indicated. β-actin was used as loading control. Pt, patient. Oncogene (2015) 1 – 9
© 2015 Macmillan Publishers Limited
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
3
Figure 2. Dose-dependent effects of SAS on human glioma cell lines. (a) ECIS measurements for monolayer cultures of the A172, LN18 and U251 glioma cell lines in medium only (Ctrl) and when treated with 250, 500 and 1000 μM SAS. (b) Uptake of carbon-14 labeled l-cystine in glioma cell lines, as indicated, grown in medium only (Ctrl) and in 250 and 500 μM of SAS. (c) A luminescence-based assay was used to measure intracellular GSH levels when grown in medium only (Ctrl) and in 250, 500 and 1000 μM of SAS. (d) ROS levels measured using a fluorescencebased assay in glioma cell lines as indicated when grown in medium only (Ctrl), and in 250, 500 and 1000 μM of SAS, in the presence (+NAC) or absence ( − NAC) of NAC. Ctrl: control. Error bars represent s.d. of two (b) and three (c, d) independent experiments. *P o0.05, **Po 0.01, ***P o0.001, ****Po0.0001.
alone did not cause a significant reduction in viability. Strikingly however, 250 μM SAS in combination with radiation significantly reduced viability in all the cell lines. Similarly, 500 μM SAS potentiated the effect of radiation of the U251 and LN18 glioma cell lines, whereas 500 μM SAS alone reduced viability to almost zero in the A172 cell line. We then performed siRNA knockdown of the xCT subunit, to confirm that the effects of SAS were mediated by the system X−c . (Figure 3c). Successful siRNA knockdown of xCT in A172 cells was confirmed by western blot (Figure 3c, left panel). A172 cells treated with siRNA were then subjected to ionizing radiation (Figure 3d). Similar to the SAS experiment, we observed that also siRNA-mediated knockdown of xCT potentiated the effects of radiation. SAS and radiation synergistically increases glioma cell death The viability changes we observed after SAS treatment prompted us to investigate how SAS specifically impacted on cell death, alone or in combination with radiation. First we performed live/ dead staining of U251 glioma cells 1 week after treatment with SAS and 8 Gy administered separately or in combination (Figure 4a). Although both radiation and SAS treatment alone caused a modest reduction in live cells (green), 250 and 500 μM SAS significantly potentiated the effect of radiation. © 2015 Macmillan Publishers Limited
Next, we assessed cell death by measuring the release of cytosolic glucose 6-phosphate dehydrogenase after SAS treatment alone or in combination in the U251 and U87 glioma cell lines (Figure 4b). Whereas radiation alone did not significantly increase cell death, 1000 μM SAS significantly increased cell death in both the cell lines compared with untreated controls. However, the combination of 1000 μM SAS and radiation significantly increased cell death in both the cell lines, compared with 1000 μM SAS alone or radiation alone. Furthermore, we subsequently conducted propidium iodide staining for flow cytometric assessment of cell death for the A172 and LN18 glioma cell lines (Figure 4c). Again, we observed a significant increase in cell death after treatment with 1000 μM SAS (Figure 4c, middle panels). Upon adding the antioxidant NAC however, the effect of SAS was almost fully reversed in the A172 cell line, and significantly reduced in the LN18 glioma cell line (Figure 4c, lower panels). SAS potentiates the effect of GKRS on GBM xenografts Finally, we implanted GBM xenograft spheroids intracerebrally in 31 nude rats to compare the survival outcome for animals receiving SAS or GKRS alone or in combination, compared with untreated controls (Figure 5). Upon confirming tumor engraftment Oncogene (2015) 1 – 9
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
4
Figure 3. Effects of SAS and radiation on DNA damage and viability in glioma cells. (a) Quantification of DNA damage by measuring H2AXSer139 phosphorylation in the U251 and LN18 glioma cell lines, grown in medium (Ctrl) or in 250, 500 and 1000 μM of SAS, without radiation (no rad) or after 8 Gy radiation (n = 2). (b) Cell viability in the U251, LN18 and A172 glioma cell lines when grown in medium (Ctrl) or in 250, 500 and 1000 μM of SAS, without radiation (no rad) or after 8 Gy radiation (n = 3). (c) Western blot (left panel) and densitometry (right panel) for xCT in the A172 glioma cell line with xCT siRNA at 5 and 10 nM concentrations, β-actin was used as loading control. (d) MTS assay with absorbance measurements upon 5 and 10 nM xCT siRNA-mediated knockdown with (8 Gy) and without radiation (no rad). *Po 0.05, **P o0.01, ***P o0.001, ****P o0.0001.
with MRI (Figure 5a, upper panel), two groups received SAS for 3 days, before one of the groups receiving SAS also underwent GKRS. The presence of SAS in two brain tumor samples was confirmed using high-performance liquid chromatography–mass spectrometry (HPLC–MS; Supplementary Figure 2). We then performed a new MRI, 14 days after GKRS to monitor tumor growth (Figure 5a, lower panels). Notably, SAS and GKRS synergistically prolonged survival compared with either treatment alone (Figure 5b). The rats treated with GKRS combined with SAS (mean volume 4.4 ± 05 mm3) had a median survival of 72 days from implantation, compared with 57 days for the GKRS treatment group, 48 days for the SAS group and 45 days for the untreated controls (P o 0.0001). Furthermore, 21 animals underwent longitudinal monitoring of tumor growth using MRI. Tumor growth differed significantly between the groups, with the smallest volumetric increase (Figure 5c) in the group receiving SAS and GKRS (P o 0.0001). DISCUSSION In the present work, we have validated inhibition of system X−c as a strategy to potentiate the effect of radiotherapy for gliomas. We found that 28 of 30 GBM biopsies stained moderately to strongly positive for xCT. This is similar to the data published by Takeuchi et al.,33 who reported that 31 of 40 GBMs were moderately or Oncogene (2015) 1 – 9
strongly positive for xCT. Importantly, we found that sections from normal brain tissue displayed only weak immunopositivity. Our findings therefore suggest that xCT expression is common to most GBMs, which together with its low expression in normal brain tissue could provide a therapeutic window. Accordingly, xCT is a potential target in GBMs. Furthermore, as xCT expression was preserved in glioma cell lines, this also allowed us to investigate its importance for GSHsynthesis and protection against oxidative stress. Whereas SAS treatment consistently slowed the growth of different glioma cell lines in a dose-dependent manner, the inhibitory effect was strongest for the A172 glioma cell line. The stronger inhibition of this cell line may be owing to a higher constitutive expression of xCT as shown on the western blots (Figure 1c). Furthermore, the growth-inhibitory effect of SAS was accompanied by a dosedependent reduction of cystine uptake and intracellular GSH levels. Similar to the growth experiments, the inhibitory effect at the lowest SAS concentration was strongest for the A172 glioma cell line, again suggesting the importance of endogenous expression levels of xCT. SAS was initially identified as an inhibitor of the system X−c antiporter in lymphoid cells,20 whereas Chung et al.21 later reported the inhibition of cystine uptake in glioma cell lines at 250 μM SAS. They also found that the effect upon cystine uptake in astrocytes was more limited, which also suggest that SAS may selectively target the glioma cells. In this context, our © 2015 Macmillan Publishers Limited
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
5
Figure 4. Glioma cell death after SAS and radiation treatment. (a) Live (green)/dead (red) staining of U251 glioma cells under treatment conditions as indicated. Magnification: x20, scale bar = 100 μm. (b) Cytotoxicity as measured by the release of cytosolic G6PD for the U251 and U87 glioma cell lines when grown in medium (Ctrl) or in 250, 500 and 1000 μM SAS. Error bars represent s.d. of three independent experiments. (c) Flow cytometric detection of cell death using PI staining in the A172 and LN18 glioma cell lines when grown in medium (Ctrl), with 1000 μM SAS or with 1000 μM SAS+NAC. SSC, side scatter; FSC, forward scatter; Ctrl, Control; PI, propidium iodide. **Po 0.01.
data demonstrate a dose-dependent effect on cystine uptake outside the dose range previously reported, at concentrations higher than 250 μM SAS. Furthermore, Chung et al.21 reported a dose-dependent depletion of GSH levels from 50–500 μM SAS. Although we used additional cell lines and a different dose range, including 1000 μM SAS, we again observed essentially the same dose-dependent effect on intracellular GSH. Importantly, © 2015 Macmillan Publishers Limited
depletion of GSH was accompanied by a significant increase in the ROS levels, clearly demonstrating the crucial role of GSH as a scavenger of free radicals in glioma cells. In agreement with these findings, adding the antioxidant NAC significantly reduced ROS levels in the presence of SAS. In relation to radiosensitivity, the significance of increased ROS levels was demonstrated by assessments of double-strand breaks. Radiation and 1000 μM SAS Oncogene (2015) 1 – 9
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
6
Figure 5. Validation of SAS as a radiosensitizer in nude rats harboring GBM xenografts. (a) MRI T2-weighted brain scans from the treatment groups, as indicated, the day before GKRS and 14 days after GKRS. (b) Survival of nude rats with human GBM xenografts after treatments, as indicated (n = 31). (c) Volumetric assessments of tumors based on contrast-enhanced MRI at the time of GKRS and 14 days post GKRS (n = 22).
treatment synergistically increased DNA damage, that is, more than the additive effects of either treatment alone. Notably, the increase in DNA damage was accompanied by reduced viability as measured with the MTS assay, and increased cell death as measured by propidium iodide staining and release of glucose 6-phosphate dehydrogenase. Importantly, whereas 1000 μM SAS alone increased cell death, adding the antioxidant NAC effectively reversed this effect. This finding unambiguously establishes GSH depletion as the driver of glioma cell death following SAS treatment. This is also in compliance with previous findings showing increased cell death upon decreased glioma cell influx of cystine with subsequent GSH depletion.29 Previously, studies have demonstrated increased efficacy of chemo- and radiotherapy upon GSH depletion.8,30 Also in other solid tumors such as breast cancer, SAS has been shown to increase the effect of the chemotherapeutic drug doxorubicin.31 In order to validate SAS as a radiosensitizer in a clinically relevant model for brain tumors, we established GBM xenografts in nude rats that subsequently underwent GKRS radiation. This treatment is currently offered to patients with GBM recurrences at several treatment centers. Although radiotherapy in the primary setting is usually administered in 30 fractions, GKRS is a single session treatment and was therefore more feasible within our experimental design. Moreover, we reasoned that the risk of side effects from SAS in this setting would be reduced compared with that of fractionated radiotherapy which would have required long-term usage of SAS. Importantly, SAS and GKRS synergistically prolonged survival compared with either treatment alone, suggesting that X−c inhibition is a viable strategy to overcome radioresistance in gliomas. Two clinical studies have so far validated SAS in glioma treatment.32,33 On the basis of the studies on NF-κB’s role in glioma progression,34 as well as other pre-clinical data with SAS, Robe et al.32 initiated a phase I/II trial with patients with recurrent gliomas. The trial reported no clinical response, as well as a high frequency of side effects. It should be emphasized however, Oncogene (2015) 1 – 9
that the authors did not administer SAS in combination with radiotherapy. As such, this trial did not validate SAS as a radiosensitizer. Moreover, the study included patients with a generally low Karnofsky performance score, a factor significantly associated with poor prognosis. The drug was also given daily over an extended period, 35 days in average, which was associated with severe toxicities leading to termination of the trial. Recently, Takeuchi et al.33 enrolled 12 patients in a clinical trial with SAS in the primary treatment setting. Although the patients in general had a better Karnofsky performance score compared with Robe et al.,32 they observed no clinical benefit of SAS. Again however, SAS was scheduled for continuous administration over several weeks. In addition, patients simultaneously received temozolomide. On this regimen, SAS was discontinued in the large majority owing to hematological side effects. Thus, both these clinical studies indicate that caution should be exercised with prolonged SAS administration to glioma patients. Our current data nevertheless suggest that SAS sensitizes glioblastoma to radiation treatment. Thus, SAS administered over a short time period to potentiate the effect of GKRS may therefore be a more favorable approach. MATERIALS AND METHODS Cell culture and patient material In this study the A172, GaMg, LN18, LN229, T98G, U251 and U87 human glioma cell lines were used. The GaMg cell line was previously established from a human GBM biopsy at Haukeland University Hospital (Bergen, Norway).35 All other cell lines were commercially available and purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). Cell lines were all confirmed mycoplasma free, as well as genetically fingerprinted to confirm their origin. Cells were cultured in DMEM (D5671, Sigma-Aldrich, St Louis, MO, USA) supplemented with 10% fetal bovine serum, 3.2% non-essential amino acids, 100 U/ml Penicillin/ Streptomycin, 400 mol/l L-glutamine (all purchased from Lonza, Cologne, Germany) and 0.005 mg/ml Plasmocin (InvivoGen, San Diego, CA, USA) at 37 °C and 5% CO2. © 2015 Macmillan Publishers Limited
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
7 GBM biopsies were obtained at the Department of Neurosurgery at Haukeland University Hospital. The biopsy spheroids were prepared as described previously 36 and passaged in immunodeficient rats.37 All patient material was used with informed consent of the patients and with approval by the regional ethical committee.
Immunohistochemistry Paraffin-embedded sections of GBM biopsies were deparaffinized using xylene 100% and 96% ethanol, and antigen retrieval was done with incubation in 10 mM citrate buffer, pH 6.0 at 98 °C for 20 min. Sections were treated with peroxidase and protein block (both Dako, Glostrup, Denmark) for 5 and 30 min, respectively. The primary antibody rabbit anti-xCT (Thermo Fisher Scientific, Waltham, MA, USA) was diluted to 1:200 in TrisBSA buffer, and added to the slides for 1 h at room temperature. Following washing with TBS-Tween, sections were incubated with anti-rabbit secondary antibody (Dako) for 45 min at room temperature. Slides were developed with DAB chromogen (Dako) and counterstained with hematoxylin. Double staining with xCT and von Willebrand factor (vWF) was done with Dako EnVision G|2 Doublestain System (Dako) according to the manufacturer’s protocol. Antigen retrieval for vWF was done by 10 min incubation with Proteinase K (Dako) in Tris-HCl, pH 7.5, and slides were incubated with the rabbit anti-vWF antibody (Dako, 1:500 in Tris-BSA) for 40 min at room temperature. Double-stained slides were counterstained with hematoxylin and rinsed in distilled water before mounting with Shandon Immu-mount (Thermo Fisher Scientific).
Western blot Cell lysates were prepared by scraping and resuspending cells in kinexus buffer, containing 20 mM MOPS, 5 mM EDTA, 2 mM EGTA, as well as protease and phosphatase inhibitor tablets (Roche, Basel, Switzerland), followed by water bath sonication for 2 × 10 seconds. Protein concentrations were determined using a conventional Bradford assay (BioRad, Hercules, CA, USA). A 20 μg sample was mixed with LDS sample-loading buffer and the reducing agent dithiothreitol and incubated at 70 °C for 10 min. Samples were run on a pre-cast SDS-gel (NuPage, Invitrogen) at 200 V for 50 min. Transfer to a nitrocellulose membrane was done at 30 V for 80 min. Following blocking for 1 h at room temperature, the membrane was incubated with rabbit anti-xCT at 1:1000 dilution (Thermo Fisher Scientific) and rabbit anti-β-actin at 1:1000 dilution (Abcam, Cambridge, UK) overnight at 4 °C. After washing, the membrane was incubated with antirabbit HRP secondary antibody (Thermo Fisher Scientific) for 1.5 h. Detection was done using Supersignal West Femto Maximum Sensitivity Substrate (Pierce Biotechnology, Rockford, IL, USA), and the membranes were developed using a Fuji LAS 3000 Imager (Fuji Photo Film, Tokyo, Japan). Western blots of GBM biopsies and glioma cell lines were conducted twice.
Cell growth Effect on cell growth upon drug treatment was investigated using an electric cell substrate impedance sensor (ECIS, Applied Biophysics, Troy, NY, USA), a validated method for studying changes in cell confluence by measuring changes in impedance to currents going through the monolayers from electrodes underlying the cell-culture wells.38 A total of 20 000 cells were seeded into 8-well chambers, with electrodes attached to the bottom of each well. Upon treatment with SAS (Sigma-Aldrich) at 250, 500 and 1000 μM for up to 90 h, cell growth was displayed as an increase in the impedance signal, thus more confluent cell layers increased resistance to the electrode. ECIS measurments of glioma cell growth were conducted three times.
Measurements of carbon-14 labeled cystine uptake A total of 50 000 cells per well were pre-incubated with or without SAS for 2 h before addition of L-[3,3`-14C]- cystine, (0.05 μCi/ml, PerkinElmer, Boston, MA, USA). The cells were incubated for 20 h, washed twice with PBS before addition of RIPA buffer and scintillation cocktail. Radioactive cystine uptake was counted with a Tri Carb 3100 TR liquid scintillation analyzer (Packard Instrument Co., Inc., Downers Grove, IL, USA).
GSH assay Intracellular levels of GSH were measured using the GSH-Glo GSH assay (Promega, Madison, WI, USA) according to the manufacturer’s instructions. 5000 cells per well were seeded in a 96-well luciferase plate (Nunc, © 2015 Macmillan Publishers Limited
Penfield, NY, USA) and treated with 250 μM, 500 μM and 1000 μM SAS. After 48 h, cells were washed with PBS and GSH-Glo reagent was added to the plate. Following incubation, luciferin detection reagent was added, and luminescence was measured using a luminometer (Anthos Labtec Instruments GmbH, Salzburg, Austria). All measurments were normalized to cell numbers by cell counting. Total GSH concentration was determined from a standard curve with known GSH concentrations. All experiments were conducted three times.
ROS and antioxidant assay A total of 25 000 cells per well were seeded in a black 96-well plate the day before analysis. ROS levels were detected using the DCFDA cellular ROS kit (Abcam) according to the manufacturer’s instructions. Cells were marked with DCFDA for 45 min before treatment with SAS at increasing concentrations (250, 500 and 1000 μM) with or without 10 mM of the antioxidant NAC (Sigma) for 6 h. The plate was analyzed using a fluorescent plate reader with excitation at 485 nm and emission at 535 nm. ROS levels were determined as relative to untreated control. The assay was performed three times.
Ionizing radiation Glioma cells were seeded in cell-culture plates at a density of 1000 cells per well, and treated with 250, 500 and 1000 μM SAS. After 24 h, 6 MV x-ray radiation was given as an 8 Gray (Gy) single dose at the Department of Oncology and Medical Physics, Haukeland University Hospital.
MTS viability assay A total of 1000 cells per well were seeded in 96-well plates. After 24 h of adding SAS, the cells were radiated as described above, and were subsequently incubated for 7 days. Medium was removed from the cells before the MTS reagent (Promega) diluted in standard DMEM medium was added. Cells were incubated for 4 h in the dark at 37 °C, and absorption was measured at 490 nm using a plate reader (Asys UVM340, Biochrom, Cambridge, UK). Viability was determined relative to untreated controls. The assay was performed three times.
siRNA knockdown Cells were seeded in medium without antibiotics the day before transfection, giving a confluence of 30–50% at transfection. siRNA against xCT (sense: 5′-GGAGUUAUGCAGCUAAUUAtt-3′, antisense: 5′-UAAUUA GCUGCAUAACUCCag-3′, Ambion, Life Technologies, Carlsbad, CA, USA) was diluted to final concentrations of 5 and 10 nM in Opti-MEM I Reduced Serum Medium (Invitrogen) and mixed with diluted oligofectamine (Invitrogen) according to the manufacturer’s instructions. Control siRNA was also prepared at the same concentration as the highest xCT siRNA (10 nM). Growth medium was removed from the cells and replaced by medium without serum and antibiotics. Following 15–20 min incubation, the siRNA mix was added to the cells. After 4 h, medium containing 30% serum was added. Knockdown was assessed by western blot performed 5 days after treatment. Knockdown was also combined with radiation, in which 8 Gy doses were given after siRNA addition to the cells. Viability was assessed after 1 week using the MTS assay.
DNA damage/γH2AX assay A total of 5000 cells per well were seeded in 96-well plates. The following day SAS was added at increasing concentrations (250, 500 and 1000 μM) and cells were incubated for 24 h. Ionizing radiation was administered as described above. After 15 min incubation detection of double-stranded breaks was done using the EpiQuik In Situ DNA Damage Assay kit (EpiGentec, Farmingdale, NY, USA) according to the manufacturer’s instructions. Absorbance signal was normalized to cell number in each sample, and samples were calculated as relative to untreated, nonirradiated control. The assay was performed two times.
Live/dead assay Equal numbers of cells were seeded in culture plates and treated with 250, 500 and 1000 μM SAS for 24 h before administering a single dose of 8 Gy radiation. After 1 week of incubation, live/dead reagents were added according to the manufacturer’s instructions (Invitrogen, Paisley, UK), staining live cells green and dead cells red. Fluorescence was examined Oncogene (2015) 1 – 9
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
8 using a Nikon TE2000 fluorescent microscope (Nikon Instruments, Melville, NY, USA). The experiment was performed three times.
and on follow-up MRI were measured in Gamma Plan (Elekta Instrument AB, Stockholm, Sweden) for 22 animals.
Cytotoxicity assay
Gamma-knife treatment
Cells were seeded in black 96-well plates at 5000 cells per well. The following day, SAS was added at increasing concentrations (250, 500 and 1000 μM) and cells were incubated overnight. Ionizing radiation was subsequently administered as described above, and cells were incubated for 4 days. Cytotoxicity was analyzed using the Vybrant cytotoxicity assay kit (Life Technologies) that detects the release of cytosolic glucose 6-phosphate dehydrogenase. In short, medium was removed from the cells and 50 μl of reaction mixture was added to each well. After 20 min fluorescence signal was read using a fluorescent plate reader (Victor3 V1420 Multilabel Counter, PerkinElmer). Cytotoxicity levels were normalized to cell number for each sample and determined as relative to untreated and non-irradiated control. The assay was performed three times.
Animals were irradiated with the Leksell Gamma Knife Perfextion (Elekta Instrument AB) during anesthesia (subcutaneous Hypnorm-Dormicum; 0.4 ml/kg). The rats were immobilized in a Regis-Valliccioni stereotactic frame 5 (Neuropace, Neuilly, France).40 Stereotactic CT scanning was coregistered with MRI images in Leksell Gamma Plan 10.1. The rats were treated with a single 4 mm shot with a margin dose of 8 Gy to the 90% isodose line with 99–100% tumor coverage.
Propidium iodine staining for cell death analysis Glioma cells were seeded in T25 culture flasks a day before treatment with a density of 300 000 cells for the control and 400 000 cells for treatment with 1000 μM SAS and 1000 μM SAS+10 mM NAC. After 24 h, the medium was removed and the cells were washed with PBS. Following trypsination cells were spun down at 1700g for 6 min, and the pellet was resuspended in 100 μl of 50 μg/ml PI solution (Sigma) prior to flowcytometry using a a BD Accuri C6 flow cytometer (BD Biosciences, Erembodegem, Belgium). Samples were analyzed using FlowJo 7.6 (TreeStar, Ashland, OR, USA). Cells were gated according to SSC/FSC and doublets were excluded by FSC-H/ FSC-A. The analysis was performed twice.
Animal experiment The animal experiment was performed with 31 athymic, homozygous nude rats (rnu/rnu, Rowett) between 200 and 250 g. They were kept on a standard pellet diet in a pathogen-free environment at a constant temperature and humidity, and on a standard 12/12 h light and dark cycle with unlimited access to water. The experiment was approved by the Norwegian Animal Research Authority (Oslo, Norway) and is in accordance with Guide for the Care and Use of Laboratory Animals (Institute for Laboratory Animal Research, National Research Council. Washington DC: National Academy Press, 1996). Prior to implantation, all animals were anesthetized with isoflurane gas (1.5% mixed with 50% air and 50% O2) and Marcain subcutaneously, and tumor implantation was conducted as previously described.39 Animals with comparable tumor volumes derived from a patient GBM xenograft, Pt3, were randomly assigned to four different groups: (i) Ctrl (7 rats), (ii) SAS (9 rats), (iii) GKRS (8 rats) and (iv) SAS+GKRS (7 rats). SAS was administered by an 8 mg intraperitonal injection during gas anesthesia twice daily for 3 days, starting the day before GKRS. Animals were inspected daily and killed with CO2 at the onset of symptoms such as passiveness, neurological deficits or other signs of illness.
HPLC–MS/MS An HPLC–MS/MS method for the determination of SAS in rat brain tissue was developed at Analytical Services, using a 5 μm, 2.1 × 100 mm HPLC column (Venusil, Agela Technologies, Wilmington, DE, USA) and Micromass Quattro Micro API mass spectrometer (Waters, Milford, MA, USA). For the analysis, dimenhydrinate was used as internal standard. Protein precipitation was used for sample preparation of SAS from diluted rat brain tumor tissue extracts. The rat brain tissue samples were homogenized, weighed and spiked with working solutions and diluent in 1.5 ml Eppendorf caps together with a small magnetic stirring bar, vortexed for 2 min, stored in an ultrasonic bath for 15 min and vortexed again for 2 min. For injection, supernatants of the centrifuged samples (5 min) were filled in autosampler vials. The HPLC system was equilibrated with the eluent for at least 1 h before analyzing the samples.
MRI All animals were imaged after implantation of tumor spheroids to confirm tumor take, using a Bruker Pharmascan 7Tesla small animal MRI (Bruker Biospin MRI GmnH, Ettingen, Germany). Axial T1- and T2-weighted images were acquired as previously described.39 The tumor volumes at treatment Oncogene (2015) 1 – 9
Statistical analysis For cystine uptake, GSH assay, ROS assay, glucose 6-phosphate dehydrogenase Cytotoxicity assay, MTS assay and γH2AX assay, we performed two-way analysis of variance with Tukey’s multiple comparions’ test with a P-valueo0.05 considered significant. For statistical analysis of the animal experiment we constructed Kaplan– Meier survival curves. Median survival times for the treatment groups were compared using the log-rank (Mantel–Cox) test using the GraphPad Prism 6 statistical software (GraphPad Software Inc., La Jolla, CA, USA). Tumor volumes on MRI images at the day of treatment and at 14 days follow-up were measured in the Gamma plan and compared using two-way analysis of variance. A probability value ⩽ 0.05 was regarded as significant.
CONFLICT OF INTEREST The authors declare no conflict of interest.
REFERENCES 1 Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJ et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005; 352: 987–996. 2 Cairns RA, Harris IS, Mak TW. Regulation of cancer cell metabolism. Nat Rev Cancer 2011; 11: 85–95. 3 Fruehauf JP, Meyskens FL Jr. Reactive oxygen species: a breath of life or death? Clin Cancer Res 2007; 13: 789–794. 4 Hayes JD, McMahon M. NRF2 and KEAP1 mutations: permanent activation of an adaptive response in cancer. Trends Biochem Sci 2009; 34: 176–188. 5 Lluis JM, Buricchi F, Chiarugi P, Morales A, Fernandez-Checa JC. Dual role of mitochondrial reactive oxygen species in hypoxia signaling: activation of nuclear factor-{kappa}B via c-SRC and oxidant-dependent cell death. Cancer Res 2007; 67: 7368–7377. 6 McDonald JT, Kim K, Norris AJ, Vlashi E, Phillips TM, Lagadec C et al. Ionizing radiation activates the Nrf2 antioxidant response. Cancer Res 2010; 70: 8886–8895. 7 Towner RA, Smith N, Saunders D, De Souza PC, Henry L, Lupu F et al. Combined molecular MRI and immuno-spin-trapping for in vivo detection of free radicals in orthotopic mouse GL261 gliomas. Biochim Biophys Acta 2013; 1832: 2153–2161. 8 Allalunis-Turner MJ, Day RS 3rd, McKean JD, Petruk KC, Allen PB, Aronyk KE et al. Glutathione levels and chemosensitizing effects of buthionine sulfoximine in human malignant glioma cells. J Neurooncol 1991; 11: 157–164. 9 Bannai S, Kitamura E. Transport interaction of L-cystine and L-glutamate in human diploid fibroblasts in culture. J Biol Chem 1980; 255: 2372–2376. 10 Sato H, Tamba M, Ishii T, Bannai S. Cloning and expression of a plasma membrane cystine/glutamate exchange transporter composed of two distinct proteins. J Biol Chem 1999; 274: 11455–11458. 11 Ye ZC, Rothstein JD, Sontheimer H. Compromised glutamate transport in human glioma cells: reduction-mislocalization of sodium-dependent glutamate transporters and enhanced activity of cystine-glutamate exchange. J Neurosci 1999; 19: 10767–10777. 12 Robert SM, Sontheimer H. Glutamate transporters in the biology of malignant gliomas. Cell Mol Life Sci 2014; 71: 1839–1854. 13 Lyons SA, Chung WJ, Weaver AK, Ogunrinu T, Sontheimer H. Autocrine glutamate signaling promotes glioma cell invasion. Cancer Res 2007; 67: 9463–9471. 14 Rothstein JD, Brem H. Excitotoxic destruction facilitates brain tumor growth. Nat Med 2001; 7: 994–995. 15 Buckingham SC, Campbell SL, Haas BR, Montana V, Robel S, Ogunrinu T et al. Glutamate release by primary brain tumors induces epileptic activity. Nat Med 2011; 17: 1269–1274. 16 Kim JY, Kanai Y, Chairoungdua A, Cha SH, Matsuo H, Kim DK et al. Human cystine/ glutamate transporter: cDNA cloning and upregulation by oxidative stress in glioma cells. Biochim Biophys Acta 2001; 1512: 335–344.
© 2015 Macmillan Publishers Limited
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
9 17 Takeuchi S, Wada K, Toyooka T, Shinomiya N, Shimazaki H, Nakanishi K et al. Increased xCT expression correlates with tumor invasion and outcome in patients with glioblastomas. Neurosurgery 2013; 72: 33–41 discussion 41. 18 Neumann VC, Grindulis KA. Sulphasalazine in rheumatoid arthritis: an old drug revived. J R Soc Med 1984; 77: 169–172. 19 Dick AP, Grayson MJ, Carpenter RG, Petrie A et al. Controlled trial of sulphasalazine in the treatment of ulcerative colitis. Gut 1964; 5: 437–442. 20 Gout PW, Buckley AR, Simms CR, Bruchovsky N. Sulfasalazine, a potent suppressor of lymphoma growth by inhibition of the x(c)- cystine transporter: a new action for an old drug. Leukemia 2001; 15: 1633–1640. 21 Chung WJ, Lyons SA, Nelson GM, Hamza H, Gladson CL, Gillespie GY et al. Inhibition of cystine uptake disrupts the growth of primary brain tumors. J Neurosci 2005; 25: 7101–7110. 22 Ogunrinu TA, Sontheimer H. Hypoxia increases the dependence of glioma cells on glutathione. J Biol Chem 2010; 285: 37716–37724. 23 Salcman M. Survival in glioblastoma: historical perspective. Neurosurgery 1980; 7: 435–439. 24 Archibald YM, Lunn D, Ruttan LA, Macdonald DR, Del Maestro RF, Barr HW et al. Cognitive functioning in long-term survivors of high-grade glioma. J Neurosurg 1994; 80: 247–253. 25 Soffietti R, Kocher M, Abacioglu UM, Villa S, Fauchon F, Baumert BG et al. A European organisation for research and treatment of cancer phase III trial of adjuvant whole-brain radiotherapy versus observation in patients with one to three brain metastases from solid tumors after surgical resection or radiosurgery: quality-of-life results. J Clin Oncol 2013; 31: 65–72. 26 Skeie BS, Enger PØ, Brøgger J, Ganz JC, Thorsen F, Heggdal JI et al. gamma knife surgery versus reoperation for recurrent glioblastoma multiforme. World Neurosurg 2012; 78: 658–669. 27 Hsieh PC, Chandler JP, Bhangoo S, Panagiotopoulos K, Kalapurakal JA, Marymont MH et al. Adjuvant gamma knife stereotactic radiosurgery at the time of tumor progression potentially improves survival for patients with glioblastoma multiforme. Neurosurgery 2005; 57: 684–692 discussion 684-92. 28 Skeie BS, Wang J, Dodoo E, Heggdal JI, Grønli J, Sleire L et al. Gamma knife surgery as monotherapy with clinically relevant doses prolongs survival in a human GBM xenograft model. Biomed Res Int 2013; 2013: 139674.
29 Kato S, Negishi K, Mawatari K, Kuo CH. A mechanism for glutamate toxicity in the C6 glioma cells involving inhibition of cystine uptake leading to glutathione depletion. Neuroscience 1992; 48: 903–914. 30 Bump EA, Yu NY, Brown JM. Radiosensitization of hypoxic tumor cells by depletion of intracellular glutathione. Science 1982; 217: 544–545. 31 Narang VS, Pauletti GM, Gout PW, Buckley DJ, Buckley AR. Sulfasalazine-induced reduction of glutathione levels in breast cancer cells: enhancement of growthinhibitory activity of Doxorubicin. Chemotherapy 2007; 53: 210–217. 32 Robe PA, Martin DH, Nguyen-Khac MT, Artesi M, Deprez M, Albert A et al. Early termination of ISRCTN45828668, a phase 1/2 prospective, randomized study of sulfasalazine for the treatment of progressing malignant gliomas in adults. BMC Cancer 2009; 9: 372. 33 Takeuchi S, Wada K, Nagatani K, Otani N, Osada H, Nawashiro H et al. Sulfasalazine and temozolomide with radiation therapy for newly diagnosed glioblastoma. Neurol India 2014; 62: 42–47. 34 Robe PA, Bentires-Alj M, Bonif M, Rogister B, Deprez M, Haddada H et al. In vitro and in vivo activity of the nuclear factor-kappaB inhibitor sulfasalazine in human glioblastomas. Clin Cancer Res 2004; 10: 5595–5603. 35 Akslen LA, Andersen KJ, Bjerkvig R. Characteristics of human and rat glioma cells grown in a defined medium. Anticancer Res 1988; 8: 797–803. 36 Bjerkvig R, Tønnesen A, Laerum OD, Backlund EO. Multicellular tumor spheroids from human gliomas maintained in organ culture. J Neurosurg 1990; 72: 463–475. 37 Sakariassen PO, Prestegarden L, Wang J, Skaftnesmo KO, Mahesparan R, Molthoff C et al. Angiogenesis-independent tumor growth mediated by stem-like cancer cells. Proc Natl Acad Sci USA 2006; 103: 16466–16471. 38 Wegener J, Keese CR, Giaever I. Electric cell-substrate impedance sensing (ECIS) as a noninvasive means to monitor the kinetics of cell spreading to artificial surfaces. Exp Cell Res 2000; 259: 158–166. 39 Wang J, Miletic H, Sakariassen PØ, Huszthy PC, Jacobsen H, Brekkå N et al. A reproducible brain tumour model established from human glioblastoma biopsies. BMC Cancer 2009; 9: 465. 40 Rey M, Valliccioni PA, Vial M, Porcheron D, Regis J, Kerlerian-le Goff L et al. Experimental radiosurgery in rats using gamma a ‘gamma knife’. Description of a stereotactic device for small laboratory animals. Neurochirurgie 1996; 42: 289–293.
Supplementary Information accompanies this paper on the Oncogene website (http://www.nature.com/onc)
© 2015 Macmillan Publishers Limited
Oncogene (2015) 1 – 9
ORIGINAL ARTICLE
Drug repurposing: sulfasalazine sensitizes gliomas to gamma knife radiosurgery by blocking cystine uptake through system Xc−, leading to glutathione depletion L Sleire1, BS Skeie2,3, IA Netland1, HE Førde1, E Dodoo4, F Selheim5, L Leiss1,6, JI Heggdal7, P-H Pedersen2,3, J Wang1,8 and PØ Enger1,3,8 Glioblastomas (GBMs) are aggressive brain tumors that always recur after radiotherapy. Cystine, mainly provided by the system X−c antiporter, is a requirement for glioma cell synthesis of glutathione (GSH) which has a critical role in scavenging free radicals, for example, after radiotherapy. Thus, we hypothesized that the X−c -inhibitor sulfasalazine (SAS) could potentiate the efficacy of radiotherapy against gliomas. Here, we show that the catalytic subunit of system X−c , xCT, was uniformly expressed in a panel of 30 human GBM biopsies. SAS treatment significantly reduced cystine uptake and GSH levels, whereas it significantly increased the levels of reactive oxygen species (ROS) in glioma cells in vitro. Furthermore, SAS and radiation synergistically increased DNA doublestrand breaks and increased glioma cell death, whereas adding the antioxidant N-acetyl-L-cysteine (NAC) reversed cell death. Moreover, SAS and gamma knife radiosurgery (GKRS) synergistically prolonged survival in nude rats harboring human GBM xenografts, compared with controls or either treatment alone. In conclusion, SAS effectively blocks cystine uptake in glioma cells in vitro, leading to GSH depletion and increased ROS levels, DNA damage and cell death. Moreover, it potentiates the anti-tumor efficacy of GKRS in rats with human GBM xenografts, providing a survival benefit. Thus, SAS may have a role as a radiosensitizer to enhance the efficacy of current radiotherapies for glioma patients. Oncogene advance online publication, 23 March 2015; doi:10.1038/onc.2015.60
INTRODUCTION Glioblastomas (GBMs) are lethal cancers and inherently resistant to radiotherapy. Established treatments including surgery, radio- and chemotherapy have a limited efficacy, and the median survival is ~ 14.6 months.1 Thus, treatment resistance represents a major challenge in the clinical management of these patients, and new therapies are urgently needed. The effect of radiotherapy is mediated by creation of reactive oxygen species (ROS) that cause DNA damage, subsequently leading to apoptosis when cells pass through check points during the cell cycle. However, ROS are also byproducts of the normal cellular metabolism, and low concentrations are required for various cellular processes in healthy cells.2 Thus, ROS levels are tightly regulated, and antioxidants are crucial in maintaining redox homeostasis.3 Moreover, studies show that cancer cells synthesize antioxidants at increased rates 4 to scavenge free radicals that are produced for example, under hypoxia 5 or following radiotherapy.6 Notably, combined magnetic resonance imaging (MRI) and immuno-spin trapping for detection of free radicals in experimental glioma models, have demonstrated increased levels in vivo, compared with normal mouse brain.7 Glutathione (GSH) is one of the most important antioxidants in mammalian cells, and is abundantly present in human brain tumors.8 Availability of cystine is the rate-limiting step in GSHsynthesis in glioma cells, and is provided through system X−c .9,10 1
In exchange for cystine uptake, this antiporter releases intracellular glutamate which impacts on multiple aspects of glioma biology.11,12 As such, glutamate reportedly promotes glioma cell growth 13 and has been linked to tissue destruction 14 and tumorassociated epilepsy.15 Studies in glioma cells have shown that system X−c is upregulated under oxidative stress,16 and Takeuchi et al.17 reported that increased X−c expression in human GBMs was associated with a shorter survival. Intriguingly, sulfasalazine (SAS), a drug approved for treatment of rheumatoid arthritis 18 and inflammatory bowel diseases,19 has been shown to block the X−c antiport system, thereby inhibiting glutamate release.20 Notably, it also suppresses growth of experimental gliomas,21 an effect that has been attributed to reduced synthesis of GSH. However, whereas these studies imply major roles for both GSH and glutamate in glioma progression, the therapeutic potential of blocking GSH-synthesis concomitantly with treatment-induced oxidative stress has received relatively little attention. As GSH represents a major defense system against oxidative stress, and is the principal antioxidant in glioma cells,11,21,22 its depletion may increase glioma cell vulnerability to increased ROS levels, such as during radiotherapy. We therefore hypothesized that SAS may potentiate the effect of radiotherapy against glioma cells. The current primary radiation treatment for GBMs includes fractionated radiation to 60 Gy, and has been shown to prolong survival compared with the untreated patients.1,23 The patients do,
Department of Biomedicine, Oncomatrix Research Lab, University of Bergen, Bergen, Norway; 2Department of Clinical Medicine, K1, University of Bergen, Bergen, Norway; Department of Neurosurgery, Haukeland University Hospital, Bergen, Norway; 4Department of Neurosurgery, Karolinska University Hospital, Stockholm, Sweden; 5Department of Biomedicine, Proteomics Unit (PROBE), University of Bergen, Bergen, Norway; 6Neuro Clinic, Haukeland University Hospital, Bergen, Norway and 7Department of Oncology and Medical Physics, Haukeland University Hospital, Bergen, Norway. Correspondence: Professor PØ Enger, Department of Biomedicine, Oncomatrix Research Lab, University of Bergen, Jonas Lies vei 91, Bergen 5020, Norway. E-mail: [email protected] 8 These authors contributed equally to this work. Received 11 July 2014; revised 27 November 2014; accepted 16 December 2014 3
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
2 however, relapse with focal tumors and may experience severe cognitive effects because of radiation damage of healthy brain tissue surrounding the tumor.24 Alternatively, gamma knife radiosurgery (GKRS) is a radiation treatment characterized by a sharp dose fall-off outside the tumor target that may limit cognitive deterioration.25 In contrast to the fractionated radiotherapy, GKRS is a single session treatment and is currently offered at several treatment centers to patients with local GBM recurrences.26,27 Previously, we have established an animal model for GKRS in order to investigate different dose regimens and accompanying radiobiological effects.28 In this study, we used this model to study the effects of GKRS alone, or in combination with SAS, thereby validating this compound as a radiosensitizer for gliomas. RESULTS xCT is expressed in human GBMs and glioma cell lines Initially, we performed immunohistochemistry for xCT, the catalytic subunit of system X−c , in 30 GBM patient biopsies (Figure 1a and Supplementary Figure 1a). All samples expressed xCT and displayed a homogeneous predominantly cytoplasmic/ membrane staining confined to the tumor cells, whereas the vasculature showed no visible staining. As the samples expressed xCT at varying levels, three independent observers scored each section (0–3) for xCT expression. A total of 28 of the 30 sections expressed xCT at moderate-to-high levels (Supplementary Figure 1b). Notably, xCT expression in sections from normal brain could hardly be detected (Figure 1a, left panels). Moreover, western blotting confirmed xCT expression in protein lysates from three GBM tissue biopsies (Figure 1b), as well as in a panel of seven glioma cell lines (Figure 1c). SAS inhibits the growth of glioma cells and blocks the uptake of cystine, leading to GSH depletion and increased ROS levels As immunohistochemistry and western blotting confirmed the expression of the system X−c in glioma biopsies and cell lines, we next wanted to investigate the effects of pharmacologically blocking this antiporter. Thus, we established monolayer cultures with the A172, U251 and LN18 glioma cell lines that were treated
with escalating doses of SAS. We performed ECIS measurements to monitor cell growth longitudinally and observed that controls and glioma cell cultures receiving SAS treatment at different doses produced distinct impedance curves (Figure 2a). These growth curves therefore established a dose–response relationship for the glioma cell lines tested. We then performed measurements of carbon-14 labeled cystine uptake in the A172, LN18 and U251 glioma cell lines preincubated with or without SAS (Figure 2b). SAS reduced cystine uptake significantly in all the cell lines in a dose-dependent manner when added at concentrations 250, 500 and 1000 μM. In a similar way, we observed a significant and dose-dependent reduction in GSH levels in the same glioma cell lines with increasing doses of SAS (Figure 2c). Conversely, assessment of ROS levels showed a consistent increase in ROS that accompanied the reduction in cystin uptake and GSH levels (Figure 2d). In the LN18 and U251 glioma cell lines we again observed a significant dosedependent effect of SAS. Although ROS levels also increased with the concentration of SAS for the A172 cell line, only the highest concentration of SAS produced significantly higher ROS levels than the other groups. Furthermore, adding the antioxidant N-Acetyl-L-Cysteine (NAC) abrogated the effects on ROS levels in all cell lines tested (Figure 2d). SAS and radiation synergistically increases DNA damage in glioma cells As the ROS levels consistently increased following SAS treatment, we wanted to see if SAS also caused increased DNA damage, alone or in combination with radiation. We therefore quantified double-strand DNA breaks by measuring phosphorylation of Histone 2AX (Figure 3a). We conducted the experiment using non-irradiated or irradiated glioma cells grown in medium only, or with increasing concentrations of SAS. Whereas SAS at 500 μM increased DNA damage, 8 Gy radiation without SAS only caused a marginal, insignificant increase in DNA damage. In the presence of 1000 μM SAS however, radiation significantly increased the levels of DNA damage, both in the U251 and LN18 glioma cell lines. In order to further elucidate the biological effects of SAS treatment, alone or in combination with radiation, we assessed cell viability for the U251, LN18 and A172 glioma cell lines using the MTS assay (Figure 3b). In these, radiation alone or 250 μM SAS
Figure 1. Expression of xCT in human GBM biopsies and glioma cell lines. (a) immunohistochemistry (IHC) staining for xCT (brown) in normal brain and human GBM biopsies as indicated, at x10 (upper panels) and x40 magnification (lower panels). Nuclear counterstaining: hematoxylin (blue), scale bars = x10, 100 μM; x40, 50 μM. (b) Western blot for xCT in three GBM biopsies and (c) human glioma cell lines as indicated. β-actin was used as loading control. Pt, patient. Oncogene (2015) 1 – 9
© 2015 Macmillan Publishers Limited
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
3
Figure 2. Dose-dependent effects of SAS on human glioma cell lines. (a) ECIS measurements for monolayer cultures of the A172, LN18 and U251 glioma cell lines in medium only (Ctrl) and when treated with 250, 500 and 1000 μM SAS. (b) Uptake of carbon-14 labeled l-cystine in glioma cell lines, as indicated, grown in medium only (Ctrl) and in 250 and 500 μM of SAS. (c) A luminescence-based assay was used to measure intracellular GSH levels when grown in medium only (Ctrl) and in 250, 500 and 1000 μM of SAS. (d) ROS levels measured using a fluorescencebased assay in glioma cell lines as indicated when grown in medium only (Ctrl), and in 250, 500 and 1000 μM of SAS, in the presence (+NAC) or absence ( − NAC) of NAC. Ctrl: control. Error bars represent s.d. of two (b) and three (c, d) independent experiments. *P o0.05, **Po 0.01, ***P o0.001, ****Po0.0001.
alone did not cause a significant reduction in viability. Strikingly however, 250 μM SAS in combination with radiation significantly reduced viability in all the cell lines. Similarly, 500 μM SAS potentiated the effect of radiation of the U251 and LN18 glioma cell lines, whereas 500 μM SAS alone reduced viability to almost zero in the A172 cell line. We then performed siRNA knockdown of the xCT subunit, to confirm that the effects of SAS were mediated by the system X−c . (Figure 3c). Successful siRNA knockdown of xCT in A172 cells was confirmed by western blot (Figure 3c, left panel). A172 cells treated with siRNA were then subjected to ionizing radiation (Figure 3d). Similar to the SAS experiment, we observed that also siRNA-mediated knockdown of xCT potentiated the effects of radiation. SAS and radiation synergistically increases glioma cell death The viability changes we observed after SAS treatment prompted us to investigate how SAS specifically impacted on cell death, alone or in combination with radiation. First we performed live/ dead staining of U251 glioma cells 1 week after treatment with SAS and 8 Gy administered separately or in combination (Figure 4a). Although both radiation and SAS treatment alone caused a modest reduction in live cells (green), 250 and 500 μM SAS significantly potentiated the effect of radiation. © 2015 Macmillan Publishers Limited
Next, we assessed cell death by measuring the release of cytosolic glucose 6-phosphate dehydrogenase after SAS treatment alone or in combination in the U251 and U87 glioma cell lines (Figure 4b). Whereas radiation alone did not significantly increase cell death, 1000 μM SAS significantly increased cell death in both the cell lines compared with untreated controls. However, the combination of 1000 μM SAS and radiation significantly increased cell death in both the cell lines, compared with 1000 μM SAS alone or radiation alone. Furthermore, we subsequently conducted propidium iodide staining for flow cytometric assessment of cell death for the A172 and LN18 glioma cell lines (Figure 4c). Again, we observed a significant increase in cell death after treatment with 1000 μM SAS (Figure 4c, middle panels). Upon adding the antioxidant NAC however, the effect of SAS was almost fully reversed in the A172 cell line, and significantly reduced in the LN18 glioma cell line (Figure 4c, lower panels). SAS potentiates the effect of GKRS on GBM xenografts Finally, we implanted GBM xenograft spheroids intracerebrally in 31 nude rats to compare the survival outcome for animals receiving SAS or GKRS alone or in combination, compared with untreated controls (Figure 5). Upon confirming tumor engraftment Oncogene (2015) 1 – 9
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
4
Figure 3. Effects of SAS and radiation on DNA damage and viability in glioma cells. (a) Quantification of DNA damage by measuring H2AXSer139 phosphorylation in the U251 and LN18 glioma cell lines, grown in medium (Ctrl) or in 250, 500 and 1000 μM of SAS, without radiation (no rad) or after 8 Gy radiation (n = 2). (b) Cell viability in the U251, LN18 and A172 glioma cell lines when grown in medium (Ctrl) or in 250, 500 and 1000 μM of SAS, without radiation (no rad) or after 8 Gy radiation (n = 3). (c) Western blot (left panel) and densitometry (right panel) for xCT in the A172 glioma cell line with xCT siRNA at 5 and 10 nM concentrations, β-actin was used as loading control. (d) MTS assay with absorbance measurements upon 5 and 10 nM xCT siRNA-mediated knockdown with (8 Gy) and without radiation (no rad). *Po 0.05, **P o0.01, ***P o0.001, ****P o0.0001.
with MRI (Figure 5a, upper panel), two groups received SAS for 3 days, before one of the groups receiving SAS also underwent GKRS. The presence of SAS in two brain tumor samples was confirmed using high-performance liquid chromatography–mass spectrometry (HPLC–MS; Supplementary Figure 2). We then performed a new MRI, 14 days after GKRS to monitor tumor growth (Figure 5a, lower panels). Notably, SAS and GKRS synergistically prolonged survival compared with either treatment alone (Figure 5b). The rats treated with GKRS combined with SAS (mean volume 4.4 ± 05 mm3) had a median survival of 72 days from implantation, compared with 57 days for the GKRS treatment group, 48 days for the SAS group and 45 days for the untreated controls (P o 0.0001). Furthermore, 21 animals underwent longitudinal monitoring of tumor growth using MRI. Tumor growth differed significantly between the groups, with the smallest volumetric increase (Figure 5c) in the group receiving SAS and GKRS (P o 0.0001). DISCUSSION In the present work, we have validated inhibition of system X−c as a strategy to potentiate the effect of radiotherapy for gliomas. We found that 28 of 30 GBM biopsies stained moderately to strongly positive for xCT. This is similar to the data published by Takeuchi et al.,33 who reported that 31 of 40 GBMs were moderately or Oncogene (2015) 1 – 9
strongly positive for xCT. Importantly, we found that sections from normal brain tissue displayed only weak immunopositivity. Our findings therefore suggest that xCT expression is common to most GBMs, which together with its low expression in normal brain tissue could provide a therapeutic window. Accordingly, xCT is a potential target in GBMs. Furthermore, as xCT expression was preserved in glioma cell lines, this also allowed us to investigate its importance for GSHsynthesis and protection against oxidative stress. Whereas SAS treatment consistently slowed the growth of different glioma cell lines in a dose-dependent manner, the inhibitory effect was strongest for the A172 glioma cell line. The stronger inhibition of this cell line may be owing to a higher constitutive expression of xCT as shown on the western blots (Figure 1c). Furthermore, the growth-inhibitory effect of SAS was accompanied by a dosedependent reduction of cystine uptake and intracellular GSH levels. Similar to the growth experiments, the inhibitory effect at the lowest SAS concentration was strongest for the A172 glioma cell line, again suggesting the importance of endogenous expression levels of xCT. SAS was initially identified as an inhibitor of the system X−c antiporter in lymphoid cells,20 whereas Chung et al.21 later reported the inhibition of cystine uptake in glioma cell lines at 250 μM SAS. They also found that the effect upon cystine uptake in astrocytes was more limited, which also suggest that SAS may selectively target the glioma cells. In this context, our © 2015 Macmillan Publishers Limited
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
5
Figure 4. Glioma cell death after SAS and radiation treatment. (a) Live (green)/dead (red) staining of U251 glioma cells under treatment conditions as indicated. Magnification: x20, scale bar = 100 μm. (b) Cytotoxicity as measured by the release of cytosolic G6PD for the U251 and U87 glioma cell lines when grown in medium (Ctrl) or in 250, 500 and 1000 μM SAS. Error bars represent s.d. of three independent experiments. (c) Flow cytometric detection of cell death using PI staining in the A172 and LN18 glioma cell lines when grown in medium (Ctrl), with 1000 μM SAS or with 1000 μM SAS+NAC. SSC, side scatter; FSC, forward scatter; Ctrl, Control; PI, propidium iodide. **Po 0.01.
data demonstrate a dose-dependent effect on cystine uptake outside the dose range previously reported, at concentrations higher than 250 μM SAS. Furthermore, Chung et al.21 reported a dose-dependent depletion of GSH levels from 50–500 μM SAS. Although we used additional cell lines and a different dose range, including 1000 μM SAS, we again observed essentially the same dose-dependent effect on intracellular GSH. Importantly, © 2015 Macmillan Publishers Limited
depletion of GSH was accompanied by a significant increase in the ROS levels, clearly demonstrating the crucial role of GSH as a scavenger of free radicals in glioma cells. In agreement with these findings, adding the antioxidant NAC significantly reduced ROS levels in the presence of SAS. In relation to radiosensitivity, the significance of increased ROS levels was demonstrated by assessments of double-strand breaks. Radiation and 1000 μM SAS Oncogene (2015) 1 – 9
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
6
Figure 5. Validation of SAS as a radiosensitizer in nude rats harboring GBM xenografts. (a) MRI T2-weighted brain scans from the treatment groups, as indicated, the day before GKRS and 14 days after GKRS. (b) Survival of nude rats with human GBM xenografts after treatments, as indicated (n = 31). (c) Volumetric assessments of tumors based on contrast-enhanced MRI at the time of GKRS and 14 days post GKRS (n = 22).
treatment synergistically increased DNA damage, that is, more than the additive effects of either treatment alone. Notably, the increase in DNA damage was accompanied by reduced viability as measured with the MTS assay, and increased cell death as measured by propidium iodide staining and release of glucose 6-phosphate dehydrogenase. Importantly, whereas 1000 μM SAS alone increased cell death, adding the antioxidant NAC effectively reversed this effect. This finding unambiguously establishes GSH depletion as the driver of glioma cell death following SAS treatment. This is also in compliance with previous findings showing increased cell death upon decreased glioma cell influx of cystine with subsequent GSH depletion.29 Previously, studies have demonstrated increased efficacy of chemo- and radiotherapy upon GSH depletion.8,30 Also in other solid tumors such as breast cancer, SAS has been shown to increase the effect of the chemotherapeutic drug doxorubicin.31 In order to validate SAS as a radiosensitizer in a clinically relevant model for brain tumors, we established GBM xenografts in nude rats that subsequently underwent GKRS radiation. This treatment is currently offered to patients with GBM recurrences at several treatment centers. Although radiotherapy in the primary setting is usually administered in 30 fractions, GKRS is a single session treatment and was therefore more feasible within our experimental design. Moreover, we reasoned that the risk of side effects from SAS in this setting would be reduced compared with that of fractionated radiotherapy which would have required long-term usage of SAS. Importantly, SAS and GKRS synergistically prolonged survival compared with either treatment alone, suggesting that X−c inhibition is a viable strategy to overcome radioresistance in gliomas. Two clinical studies have so far validated SAS in glioma treatment.32,33 On the basis of the studies on NF-κB’s role in glioma progression,34 as well as other pre-clinical data with SAS, Robe et al.32 initiated a phase I/II trial with patients with recurrent gliomas. The trial reported no clinical response, as well as a high frequency of side effects. It should be emphasized however, Oncogene (2015) 1 – 9
that the authors did not administer SAS in combination with radiotherapy. As such, this trial did not validate SAS as a radiosensitizer. Moreover, the study included patients with a generally low Karnofsky performance score, a factor significantly associated with poor prognosis. The drug was also given daily over an extended period, 35 days in average, which was associated with severe toxicities leading to termination of the trial. Recently, Takeuchi et al.33 enrolled 12 patients in a clinical trial with SAS in the primary treatment setting. Although the patients in general had a better Karnofsky performance score compared with Robe et al.,32 they observed no clinical benefit of SAS. Again however, SAS was scheduled for continuous administration over several weeks. In addition, patients simultaneously received temozolomide. On this regimen, SAS was discontinued in the large majority owing to hematological side effects. Thus, both these clinical studies indicate that caution should be exercised with prolonged SAS administration to glioma patients. Our current data nevertheless suggest that SAS sensitizes glioblastoma to radiation treatment. Thus, SAS administered over a short time period to potentiate the effect of GKRS may therefore be a more favorable approach. MATERIALS AND METHODS Cell culture and patient material In this study the A172, GaMg, LN18, LN229, T98G, U251 and U87 human glioma cell lines were used. The GaMg cell line was previously established from a human GBM biopsy at Haukeland University Hospital (Bergen, Norway).35 All other cell lines were commercially available and purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). Cell lines were all confirmed mycoplasma free, as well as genetically fingerprinted to confirm their origin. Cells were cultured in DMEM (D5671, Sigma-Aldrich, St Louis, MO, USA) supplemented with 10% fetal bovine serum, 3.2% non-essential amino acids, 100 U/ml Penicillin/ Streptomycin, 400 mol/l L-glutamine (all purchased from Lonza, Cologne, Germany) and 0.005 mg/ml Plasmocin (InvivoGen, San Diego, CA, USA) at 37 °C and 5% CO2. © 2015 Macmillan Publishers Limited
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
7 GBM biopsies were obtained at the Department of Neurosurgery at Haukeland University Hospital. The biopsy spheroids were prepared as described previously 36 and passaged in immunodeficient rats.37 All patient material was used with informed consent of the patients and with approval by the regional ethical committee.
Immunohistochemistry Paraffin-embedded sections of GBM biopsies were deparaffinized using xylene 100% and 96% ethanol, and antigen retrieval was done with incubation in 10 mM citrate buffer, pH 6.0 at 98 °C for 20 min. Sections were treated with peroxidase and protein block (both Dako, Glostrup, Denmark) for 5 and 30 min, respectively. The primary antibody rabbit anti-xCT (Thermo Fisher Scientific, Waltham, MA, USA) was diluted to 1:200 in TrisBSA buffer, and added to the slides for 1 h at room temperature. Following washing with TBS-Tween, sections were incubated with anti-rabbit secondary antibody (Dako) for 45 min at room temperature. Slides were developed with DAB chromogen (Dako) and counterstained with hematoxylin. Double staining with xCT and von Willebrand factor (vWF) was done with Dako EnVision G|2 Doublestain System (Dako) according to the manufacturer’s protocol. Antigen retrieval for vWF was done by 10 min incubation with Proteinase K (Dako) in Tris-HCl, pH 7.5, and slides were incubated with the rabbit anti-vWF antibody (Dako, 1:500 in Tris-BSA) for 40 min at room temperature. Double-stained slides were counterstained with hematoxylin and rinsed in distilled water before mounting with Shandon Immu-mount (Thermo Fisher Scientific).
Western blot Cell lysates were prepared by scraping and resuspending cells in kinexus buffer, containing 20 mM MOPS, 5 mM EDTA, 2 mM EGTA, as well as protease and phosphatase inhibitor tablets (Roche, Basel, Switzerland), followed by water bath sonication for 2 × 10 seconds. Protein concentrations were determined using a conventional Bradford assay (BioRad, Hercules, CA, USA). A 20 μg sample was mixed with LDS sample-loading buffer and the reducing agent dithiothreitol and incubated at 70 °C for 10 min. Samples were run on a pre-cast SDS-gel (NuPage, Invitrogen) at 200 V for 50 min. Transfer to a nitrocellulose membrane was done at 30 V for 80 min. Following blocking for 1 h at room temperature, the membrane was incubated with rabbit anti-xCT at 1:1000 dilution (Thermo Fisher Scientific) and rabbit anti-β-actin at 1:1000 dilution (Abcam, Cambridge, UK) overnight at 4 °C. After washing, the membrane was incubated with antirabbit HRP secondary antibody (Thermo Fisher Scientific) for 1.5 h. Detection was done using Supersignal West Femto Maximum Sensitivity Substrate (Pierce Biotechnology, Rockford, IL, USA), and the membranes were developed using a Fuji LAS 3000 Imager (Fuji Photo Film, Tokyo, Japan). Western blots of GBM biopsies and glioma cell lines were conducted twice.
Cell growth Effect on cell growth upon drug treatment was investigated using an electric cell substrate impedance sensor (ECIS, Applied Biophysics, Troy, NY, USA), a validated method for studying changes in cell confluence by measuring changes in impedance to currents going through the monolayers from electrodes underlying the cell-culture wells.38 A total of 20 000 cells were seeded into 8-well chambers, with electrodes attached to the bottom of each well. Upon treatment with SAS (Sigma-Aldrich) at 250, 500 and 1000 μM for up to 90 h, cell growth was displayed as an increase in the impedance signal, thus more confluent cell layers increased resistance to the electrode. ECIS measurments of glioma cell growth were conducted three times.
Measurements of carbon-14 labeled cystine uptake A total of 50 000 cells per well were pre-incubated with or without SAS for 2 h before addition of L-[3,3`-14C]- cystine, (0.05 μCi/ml, PerkinElmer, Boston, MA, USA). The cells were incubated for 20 h, washed twice with PBS before addition of RIPA buffer and scintillation cocktail. Radioactive cystine uptake was counted with a Tri Carb 3100 TR liquid scintillation analyzer (Packard Instrument Co., Inc., Downers Grove, IL, USA).
GSH assay Intracellular levels of GSH were measured using the GSH-Glo GSH assay (Promega, Madison, WI, USA) according to the manufacturer’s instructions. 5000 cells per well were seeded in a 96-well luciferase plate (Nunc, © 2015 Macmillan Publishers Limited
Penfield, NY, USA) and treated with 250 μM, 500 μM and 1000 μM SAS. After 48 h, cells were washed with PBS and GSH-Glo reagent was added to the plate. Following incubation, luciferin detection reagent was added, and luminescence was measured using a luminometer (Anthos Labtec Instruments GmbH, Salzburg, Austria). All measurments were normalized to cell numbers by cell counting. Total GSH concentration was determined from a standard curve with known GSH concentrations. All experiments were conducted three times.
ROS and antioxidant assay A total of 25 000 cells per well were seeded in a black 96-well plate the day before analysis. ROS levels were detected using the DCFDA cellular ROS kit (Abcam) according to the manufacturer’s instructions. Cells were marked with DCFDA for 45 min before treatment with SAS at increasing concentrations (250, 500 and 1000 μM) with or without 10 mM of the antioxidant NAC (Sigma) for 6 h. The plate was analyzed using a fluorescent plate reader with excitation at 485 nm and emission at 535 nm. ROS levels were determined as relative to untreated control. The assay was performed three times.
Ionizing radiation Glioma cells were seeded in cell-culture plates at a density of 1000 cells per well, and treated with 250, 500 and 1000 μM SAS. After 24 h, 6 MV x-ray radiation was given as an 8 Gray (Gy) single dose at the Department of Oncology and Medical Physics, Haukeland University Hospital.
MTS viability assay A total of 1000 cells per well were seeded in 96-well plates. After 24 h of adding SAS, the cells were radiated as described above, and were subsequently incubated for 7 days. Medium was removed from the cells before the MTS reagent (Promega) diluted in standard DMEM medium was added. Cells were incubated for 4 h in the dark at 37 °C, and absorption was measured at 490 nm using a plate reader (Asys UVM340, Biochrom, Cambridge, UK). Viability was determined relative to untreated controls. The assay was performed three times.
siRNA knockdown Cells were seeded in medium without antibiotics the day before transfection, giving a confluence of 30–50% at transfection. siRNA against xCT (sense: 5′-GGAGUUAUGCAGCUAAUUAtt-3′, antisense: 5′-UAAUUA GCUGCAUAACUCCag-3′, Ambion, Life Technologies, Carlsbad, CA, USA) was diluted to final concentrations of 5 and 10 nM in Opti-MEM I Reduced Serum Medium (Invitrogen) and mixed with diluted oligofectamine (Invitrogen) according to the manufacturer’s instructions. Control siRNA was also prepared at the same concentration as the highest xCT siRNA (10 nM). Growth medium was removed from the cells and replaced by medium without serum and antibiotics. Following 15–20 min incubation, the siRNA mix was added to the cells. After 4 h, medium containing 30% serum was added. Knockdown was assessed by western blot performed 5 days after treatment. Knockdown was also combined with radiation, in which 8 Gy doses were given after siRNA addition to the cells. Viability was assessed after 1 week using the MTS assay.
DNA damage/γH2AX assay A total of 5000 cells per well were seeded in 96-well plates. The following day SAS was added at increasing concentrations (250, 500 and 1000 μM) and cells were incubated for 24 h. Ionizing radiation was administered as described above. After 15 min incubation detection of double-stranded breaks was done using the EpiQuik In Situ DNA Damage Assay kit (EpiGentec, Farmingdale, NY, USA) according to the manufacturer’s instructions. Absorbance signal was normalized to cell number in each sample, and samples were calculated as relative to untreated, nonirradiated control. The assay was performed two times.
Live/dead assay Equal numbers of cells were seeded in culture plates and treated with 250, 500 and 1000 μM SAS for 24 h before administering a single dose of 8 Gy radiation. After 1 week of incubation, live/dead reagents were added according to the manufacturer’s instructions (Invitrogen, Paisley, UK), staining live cells green and dead cells red. Fluorescence was examined Oncogene (2015) 1 – 9
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
8 using a Nikon TE2000 fluorescent microscope (Nikon Instruments, Melville, NY, USA). The experiment was performed three times.
and on follow-up MRI were measured in Gamma Plan (Elekta Instrument AB, Stockholm, Sweden) for 22 animals.
Cytotoxicity assay
Gamma-knife treatment
Cells were seeded in black 96-well plates at 5000 cells per well. The following day, SAS was added at increasing concentrations (250, 500 and 1000 μM) and cells were incubated overnight. Ionizing radiation was subsequently administered as described above, and cells were incubated for 4 days. Cytotoxicity was analyzed using the Vybrant cytotoxicity assay kit (Life Technologies) that detects the release of cytosolic glucose 6-phosphate dehydrogenase. In short, medium was removed from the cells and 50 μl of reaction mixture was added to each well. After 20 min fluorescence signal was read using a fluorescent plate reader (Victor3 V1420 Multilabel Counter, PerkinElmer). Cytotoxicity levels were normalized to cell number for each sample and determined as relative to untreated and non-irradiated control. The assay was performed three times.
Animals were irradiated with the Leksell Gamma Knife Perfextion (Elekta Instrument AB) during anesthesia (subcutaneous Hypnorm-Dormicum; 0.4 ml/kg). The rats were immobilized in a Regis-Valliccioni stereotactic frame 5 (Neuropace, Neuilly, France).40 Stereotactic CT scanning was coregistered with MRI images in Leksell Gamma Plan 10.1. The rats were treated with a single 4 mm shot with a margin dose of 8 Gy to the 90% isodose line with 99–100% tumor coverage.
Propidium iodine staining for cell death analysis Glioma cells were seeded in T25 culture flasks a day before treatment with a density of 300 000 cells for the control and 400 000 cells for treatment with 1000 μM SAS and 1000 μM SAS+10 mM NAC. After 24 h, the medium was removed and the cells were washed with PBS. Following trypsination cells were spun down at 1700g for 6 min, and the pellet was resuspended in 100 μl of 50 μg/ml PI solution (Sigma) prior to flowcytometry using a a BD Accuri C6 flow cytometer (BD Biosciences, Erembodegem, Belgium). Samples were analyzed using FlowJo 7.6 (TreeStar, Ashland, OR, USA). Cells were gated according to SSC/FSC and doublets were excluded by FSC-H/ FSC-A. The analysis was performed twice.
Animal experiment The animal experiment was performed with 31 athymic, homozygous nude rats (rnu/rnu, Rowett) between 200 and 250 g. They were kept on a standard pellet diet in a pathogen-free environment at a constant temperature and humidity, and on a standard 12/12 h light and dark cycle with unlimited access to water. The experiment was approved by the Norwegian Animal Research Authority (Oslo, Norway) and is in accordance with Guide for the Care and Use of Laboratory Animals (Institute for Laboratory Animal Research, National Research Council. Washington DC: National Academy Press, 1996). Prior to implantation, all animals were anesthetized with isoflurane gas (1.5% mixed with 50% air and 50% O2) and Marcain subcutaneously, and tumor implantation was conducted as previously described.39 Animals with comparable tumor volumes derived from a patient GBM xenograft, Pt3, were randomly assigned to four different groups: (i) Ctrl (7 rats), (ii) SAS (9 rats), (iii) GKRS (8 rats) and (iv) SAS+GKRS (7 rats). SAS was administered by an 8 mg intraperitonal injection during gas anesthesia twice daily for 3 days, starting the day before GKRS. Animals were inspected daily and killed with CO2 at the onset of symptoms such as passiveness, neurological deficits or other signs of illness.
HPLC–MS/MS An HPLC–MS/MS method for the determination of SAS in rat brain tissue was developed at Analytical Services, using a 5 μm, 2.1 × 100 mm HPLC column (Venusil, Agela Technologies, Wilmington, DE, USA) and Micromass Quattro Micro API mass spectrometer (Waters, Milford, MA, USA). For the analysis, dimenhydrinate was used as internal standard. Protein precipitation was used for sample preparation of SAS from diluted rat brain tumor tissue extracts. The rat brain tissue samples were homogenized, weighed and spiked with working solutions and diluent in 1.5 ml Eppendorf caps together with a small magnetic stirring bar, vortexed for 2 min, stored in an ultrasonic bath for 15 min and vortexed again for 2 min. For injection, supernatants of the centrifuged samples (5 min) were filled in autosampler vials. The HPLC system was equilibrated with the eluent for at least 1 h before analyzing the samples.
MRI All animals were imaged after implantation of tumor spheroids to confirm tumor take, using a Bruker Pharmascan 7Tesla small animal MRI (Bruker Biospin MRI GmnH, Ettingen, Germany). Axial T1- and T2-weighted images were acquired as previously described.39 The tumor volumes at treatment Oncogene (2015) 1 – 9
Statistical analysis For cystine uptake, GSH assay, ROS assay, glucose 6-phosphate dehydrogenase Cytotoxicity assay, MTS assay and γH2AX assay, we performed two-way analysis of variance with Tukey’s multiple comparions’ test with a P-valueo0.05 considered significant. For statistical analysis of the animal experiment we constructed Kaplan– Meier survival curves. Median survival times for the treatment groups were compared using the log-rank (Mantel–Cox) test using the GraphPad Prism 6 statistical software (GraphPad Software Inc., La Jolla, CA, USA). Tumor volumes on MRI images at the day of treatment and at 14 days follow-up were measured in the Gamma plan and compared using two-way analysis of variance. A probability value ⩽ 0.05 was regarded as significant.
CONFLICT OF INTEREST The authors declare no conflict of interest.
REFERENCES 1 Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJ et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005; 352: 987–996. 2 Cairns RA, Harris IS, Mak TW. Regulation of cancer cell metabolism. Nat Rev Cancer 2011; 11: 85–95. 3 Fruehauf JP, Meyskens FL Jr. Reactive oxygen species: a breath of life or death? Clin Cancer Res 2007; 13: 789–794. 4 Hayes JD, McMahon M. NRF2 and KEAP1 mutations: permanent activation of an adaptive response in cancer. Trends Biochem Sci 2009; 34: 176–188. 5 Lluis JM, Buricchi F, Chiarugi P, Morales A, Fernandez-Checa JC. Dual role of mitochondrial reactive oxygen species in hypoxia signaling: activation of nuclear factor-{kappa}B via c-SRC and oxidant-dependent cell death. Cancer Res 2007; 67: 7368–7377. 6 McDonald JT, Kim K, Norris AJ, Vlashi E, Phillips TM, Lagadec C et al. Ionizing radiation activates the Nrf2 antioxidant response. Cancer Res 2010; 70: 8886–8895. 7 Towner RA, Smith N, Saunders D, De Souza PC, Henry L, Lupu F et al. Combined molecular MRI and immuno-spin-trapping for in vivo detection of free radicals in orthotopic mouse GL261 gliomas. Biochim Biophys Acta 2013; 1832: 2153–2161. 8 Allalunis-Turner MJ, Day RS 3rd, McKean JD, Petruk KC, Allen PB, Aronyk KE et al. Glutathione levels and chemosensitizing effects of buthionine sulfoximine in human malignant glioma cells. J Neurooncol 1991; 11: 157–164. 9 Bannai S, Kitamura E. Transport interaction of L-cystine and L-glutamate in human diploid fibroblasts in culture. J Biol Chem 1980; 255: 2372–2376. 10 Sato H, Tamba M, Ishii T, Bannai S. Cloning and expression of a plasma membrane cystine/glutamate exchange transporter composed of two distinct proteins. J Biol Chem 1999; 274: 11455–11458. 11 Ye ZC, Rothstein JD, Sontheimer H. Compromised glutamate transport in human glioma cells: reduction-mislocalization of sodium-dependent glutamate transporters and enhanced activity of cystine-glutamate exchange. J Neurosci 1999; 19: 10767–10777. 12 Robert SM, Sontheimer H. Glutamate transporters in the biology of malignant gliomas. Cell Mol Life Sci 2014; 71: 1839–1854. 13 Lyons SA, Chung WJ, Weaver AK, Ogunrinu T, Sontheimer H. Autocrine glutamate signaling promotes glioma cell invasion. Cancer Res 2007; 67: 9463–9471. 14 Rothstein JD, Brem H. Excitotoxic destruction facilitates brain tumor growth. Nat Med 2001; 7: 994–995. 15 Buckingham SC, Campbell SL, Haas BR, Montana V, Robel S, Ogunrinu T et al. Glutamate release by primary brain tumors induces epileptic activity. Nat Med 2011; 17: 1269–1274. 16 Kim JY, Kanai Y, Chairoungdua A, Cha SH, Matsuo H, Kim DK et al. Human cystine/ glutamate transporter: cDNA cloning and upregulation by oxidative stress in glioma cells. Biochim Biophys Acta 2001; 1512: 335–344.
© 2015 Macmillan Publishers Limited
SAS sensitises gliomas to GKRS through GSH depletion L Sleire et al
9 17 Takeuchi S, Wada K, Toyooka T, Shinomiya N, Shimazaki H, Nakanishi K et al. Increased xCT expression correlates with tumor invasion and outcome in patients with glioblastomas. Neurosurgery 2013; 72: 33–41 discussion 41. 18 Neumann VC, Grindulis KA. Sulphasalazine in rheumatoid arthritis: an old drug revived. J R Soc Med 1984; 77: 169–172. 19 Dick AP, Grayson MJ, Carpenter RG, Petrie A et al. Controlled trial of sulphasalazine in the treatment of ulcerative colitis. Gut 1964; 5: 437–442. 20 Gout PW, Buckley AR, Simms CR, Bruchovsky N. Sulfasalazine, a potent suppressor of lymphoma growth by inhibition of the x(c)- cystine transporter: a new action for an old drug. Leukemia 2001; 15: 1633–1640. 21 Chung WJ, Lyons SA, Nelson GM, Hamza H, Gladson CL, Gillespie GY et al. Inhibition of cystine uptake disrupts the growth of primary brain tumors. J Neurosci 2005; 25: 7101–7110. 22 Ogunrinu TA, Sontheimer H. Hypoxia increases the dependence of glioma cells on glutathione. J Biol Chem 2010; 285: 37716–37724. 23 Salcman M. Survival in glioblastoma: historical perspective. Neurosurgery 1980; 7: 435–439. 24 Archibald YM, Lunn D, Ruttan LA, Macdonald DR, Del Maestro RF, Barr HW et al. Cognitive functioning in long-term survivors of high-grade glioma. J Neurosurg 1994; 80: 247–253. 25 Soffietti R, Kocher M, Abacioglu UM, Villa S, Fauchon F, Baumert BG et al. A European organisation for research and treatment of cancer phase III trial of adjuvant whole-brain radiotherapy versus observation in patients with one to three brain metastases from solid tumors after surgical resection or radiosurgery: quality-of-life results. J Clin Oncol 2013; 31: 65–72. 26 Skeie BS, Enger PØ, Brøgger J, Ganz JC, Thorsen F, Heggdal JI et al. gamma knife surgery versus reoperation for recurrent glioblastoma multiforme. World Neurosurg 2012; 78: 658–669. 27 Hsieh PC, Chandler JP, Bhangoo S, Panagiotopoulos K, Kalapurakal JA, Marymont MH et al. Adjuvant gamma knife stereotactic radiosurgery at the time of tumor progression potentially improves survival for patients with glioblastoma multiforme. Neurosurgery 2005; 57: 684–692 discussion 684-92. 28 Skeie BS, Wang J, Dodoo E, Heggdal JI, Grønli J, Sleire L et al. Gamma knife surgery as monotherapy with clinically relevant doses prolongs survival in a human GBM xenograft model. Biomed Res Int 2013; 2013: 139674.
29 Kato S, Negishi K, Mawatari K, Kuo CH. A mechanism for glutamate toxicity in the C6 glioma cells involving inhibition of cystine uptake leading to glutathione depletion. Neuroscience 1992; 48: 903–914. 30 Bump EA, Yu NY, Brown JM. Radiosensitization of hypoxic tumor cells by depletion of intracellular glutathione. Science 1982; 217: 544–545. 31 Narang VS, Pauletti GM, Gout PW, Buckley DJ, Buckley AR. Sulfasalazine-induced reduction of glutathione levels in breast cancer cells: enhancement of growthinhibitory activity of Doxorubicin. Chemotherapy 2007; 53: 210–217. 32 Robe PA, Martin DH, Nguyen-Khac MT, Artesi M, Deprez M, Albert A et al. Early termination of ISRCTN45828668, a phase 1/2 prospective, randomized study of sulfasalazine for the treatment of progressing malignant gliomas in adults. BMC Cancer 2009; 9: 372. 33 Takeuchi S, Wada K, Nagatani K, Otani N, Osada H, Nawashiro H et al. Sulfasalazine and temozolomide with radiation therapy for newly diagnosed glioblastoma. Neurol India 2014; 62: 42–47. 34 Robe PA, Bentires-Alj M, Bonif M, Rogister B, Deprez M, Haddada H et al. In vitro and in vivo activity of the nuclear factor-kappaB inhibitor sulfasalazine in human glioblastomas. Clin Cancer Res 2004; 10: 5595–5603. 35 Akslen LA, Andersen KJ, Bjerkvig R. Characteristics of human and rat glioma cells grown in a defined medium. Anticancer Res 1988; 8: 797–803. 36 Bjerkvig R, Tønnesen A, Laerum OD, Backlund EO. Multicellular tumor spheroids from human gliomas maintained in organ culture. J Neurosurg 1990; 72: 463–475. 37 Sakariassen PO, Prestegarden L, Wang J, Skaftnesmo KO, Mahesparan R, Molthoff C et al. Angiogenesis-independent tumor growth mediated by stem-like cancer cells. Proc Natl Acad Sci USA 2006; 103: 16466–16471. 38 Wegener J, Keese CR, Giaever I. Electric cell-substrate impedance sensing (ECIS) as a noninvasive means to monitor the kinetics of cell spreading to artificial surfaces. Exp Cell Res 2000; 259: 158–166. 39 Wang J, Miletic H, Sakariassen PØ, Huszthy PC, Jacobsen H, Brekkå N et al. A reproducible brain tumour model established from human glioblastoma biopsies. BMC Cancer 2009; 9: 465. 40 Rey M, Valliccioni PA, Vial M, Porcheron D, Regis J, Kerlerian-le Goff L et al. Experimental radiosurgery in rats using gamma a ‘gamma knife’. Description of a stereotactic device for small laboratory animals. Neurochirurgie 1996; 42: 289–293.
Supplementary Information accompanies this paper on the Oncogene website (http://www.nature.com/onc)
© 2015 Macmillan Publishers Limited
Oncogene (2015) 1 – 9
