Perillyl Alcohol as a Radio-/Chemosensitizer in Malignant Glioma*

The prognosis for patients with malignant glioma has not significantly changed in two decades, despite advances in surgery, radiation, and chemotherapy, empha-sizing the growing need for novel approaches to glioma therapy. Perillyl alcohol (POH) is a naturally occurring monoterpene that has been shown to possess chemotherapeutic as well as chemopreventive activity in ani-mal tumor models and is currently in Phase I and Phase II clinical trials. In the present study, we have demonstrated that POH is an effective radiosensitizer at clinically relevant doses of radiation using established glioma cell lines. POH caused a transient arrest in the G 2 /M phase of the cell cycle and induced apoptosis in glioma cells. POH treatment sensitized glioma cells to Fas-me-diated apoptosis, which was further augmented in the presence of ionizing radiation and abrogated in the presence of antagonistic antibody. POH-induced radiosensitization was partially inhibited in glioma cells expressing dominant negative Fas-associated death domain and completely inhibited in glioma cells overexpressing the cytokine response modifier A. In addition, POH treatment resulted in a dose-dependent sensitization to cisplatin and doxorubicin induced cytotoxicity in glioma cells, highlighting its usefulness as a potent radio/chemosensitizer in the treatment of malignant glioma. malignant Quantification

Over the past 2 decades, minimal survival gains for malignant glioma have been recorded, and the disease remains almost universally fatal (1)(2)(3). To date, radiotherapy has proven to be the most effective treatment for malignant glioma, extending median survival to 8 -9 months, and hence the development of an effective radiosensitizer should be particularly relevant in the management of malignant glioma, since 90% of these patients will ultimately develop recurrences immediately within the treatment field (4,5).
Monoterpenes are naturally occurring nonnutritive dietary components with significant antitumor activity, and their mechanism of action is not similar to classic cytotoxic chemotherapeutic agents. Perillyl alcohol (POH), 1 also called p-metha, 1,7-diene-6-ol, or 4-isopropenylcyclohexenecarbinol, is a monoterpene, isolated from the essential oils of lavendin, peppermint, spearmint, and several other plants and synthesized by the mevalonate pathway (6). It has chemotherapeutic activity in pancreatic, mammary, and prostatic model tumor systems and also has chemopreventive activity in rodent mammary, skin, liver, lung, and forestomach cancers (6 -8). Crowell et al. (9) measured the effects of POH on human and hamster pancreatic tumor cell lines and determined the IC 50 values of 290 and 480 M for the human and hamster cell lines, respectively.  studied the antitumor effect of POH in a rat mammary cancer model employing chemical carcinogens, dimethylbenz(a)anthracene and nitrosomethylurea. The results showed 84% regression of rat mammary carcinoma induced by dimethylbenz(a)anthracene and 60% regression of rat mammary carcinoma induced by nitrosomethylurea on a 2% POH diet and detected 0.82 mM terpene metabolites in the plasma. Stayrook et al. (14,15) studied thymidine incorporation of POH-treated malignant and nonmalignant hamster pancreatic ductal epithelial cells and concluded that the inhibitory effect of POH was due to stimulation of apoptosis and an increase in the proapoptotic protein Bak. The malignant cells treated with 100, 300, and 500 M POH exhibited a 2.6, 8.8, and 18-fold higher rate of apoptosis than the untreated malignant cells. As little as 2000 ppm of POH in the diet inhibited azoxymethane-induced adenoma and adenocarcinoma development in rat colons (6,16). The biological activities of POH are related to its metabolites perillic acid and dihydroperillic acid (17,18). Results from rat studies showed that tumor regression occurred when plasma levels of the POH metabolites perillic acid and dihydroperillic acid reached 390 -480 and 110 -230 M, respectively.
Although the exact mechanism of POH-induced tumor regression is unknown, POH has been reported to modulate cellular processes that control cell growth and differentiation including 1) G 1 cell cycle arrest and induction of apoptosis (13,19,20), 2) inhibition of isoprenylation of small GTP-binding proteins involved in signal transduction (14,15,(21)(22)(23), and 3) differential gene regulation by overexpression of the mannose-6-phosphate/insulin-like growth factor II and transforming growth factor-␤ type II receptor genes (7,24). POH is currently being tested in Phase I and Phase II clinical trials in patients with refractory solid malignancies. Pharmacokinetic studies in humans show that POH plasma levels between 390 and 480 M can be achieved in some patients when nontoxic doses of POH are given orally (8,(25)(26)(27)(28). The present paper demonstrates the usefulness of POH as an effective, radio-/chemosensitizer using malignant glioma cell lines (primarily T98G) via activation of the Fas/Fas-ligand (Fas-L) signaling cascade. The Fas receptor (CD95/APO-1) and Fas-ligand (CD95 L) are an interacting receptor ligand pair that elicit apoptosis in many cell types (29). Faulty regulation of the Fas system has been described in a variety of human tumors with different histogenetic origin (30 -33). For as yet unknown reasons, glioma cells co-express Fas and Fas-L in vitro without undergoing suicide or fratricide (34). Several laboratories have demonstrated the apoptotic potential of human glioma cell lines in response to cross-linking of Fas with agonistic anti-Fas antibodies (35)(36)(37); cytotoxic drugs like doxorubicin, vincristin, taxol cis-diamminedichloroplatinum, VP-16, and campothecin (36,37); and ionizing radiation (38 -41). Recently, viral vector transduction of the Fas-L gene has been suggested as a potential tool for gene therapy of glioma (42)(43)(44). The ability of POH to modulate the expression of Fas-L on the surface of glioma cells and enhance their susceptibility to Fas-mediated apoptosis by chemotherapeutic agents and ionizing radiation is a novel treatment approach that can be extended to other malignancies harboring abnormalities in the Fas signaling cascade (colon, pancreas, and lung) and could form the basis for clinical studies evaluating POH as a radiosensitizer in newly diagnosed malignant glioma patients.

EXPERIMENTAL PROCEDURES
Materials-Perillyl alcohol, propidium iodide, RNase H, bromodeoxyuridine, pepsin, puromycin, cisplatin, and doxorubicin were purchased from Sigma. An annexin staining kit was purchased from Clontech. Antibodies to Fas-L, Fas receptor, Fas-associated death domain (FADD), and bromodeoxyuridine (BrdUrd); secondary antibodies streptavidin PE and streptavidin APC; and the isotype control antibodies were purchased from BD Pharmingen (San Diego, CA). The antibodies CH11 and ZB4 were purchased from Kamiya Biochemical company (Seattle, WA), and the antibody to FLAG epitope was purchased from Upstate Laboratories (Waltham, MA). Cycloheximide was purchased from Calbiochem. The Ready to Go PCR kit was purchased from Amersham Biosciences. The primers for the quantitative PCR were purchased from Invitrogen. The SYBR® Green PCR Master Mix was purchased from Applied Biosystems (Foster City, CA).
Cell Lines-The T98G cell line (45) (49) and U251 (50) cell lines were obtained from Dorothy Dougherty (Department of Neurological Surgery, University of California, San Francisco). All of the cell lines were maintained in a humidified incubator at 37°C and grown in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum, 1% penicillin, and streptomycin.
Cell Treatments and Clonogenic Survival Assays-The clonogenic survival assay was performed to determine the effect of different doses of POH or radiation and a combination of both in a panel of glioma cell lines. A stock solution of POH was made in medium. U251, T98G, MO59K, and C6 cells were treated with varying doses of POH (0.1-1 mM) for 72 h. Control dishes were treated with different doses of radiation, and the plating efficiency was determined by treating cells with medium alone. All cells were irradiated at a dose of ϳ7.5 Gy/min using a 137 Cs irradiator. The dose of irradiation varied from 1 to 8.5 Gy. Following irradiation, both POH-treated and untreated cells were harvested, washed with PBS, plated at the desired cell number, and incubated for 2 weeks. The colonies were stained with crystal violet and subsequently counted. Survival was determined as the ratio of plating efficiencies for each irradiated group to that of the unirradiated control. The same protocol was used to determine the surviving fraction in T98G cells expressing dominant negative FADD and cytokine response modifier A (Crm A) proteins. For chemosensitization experiments, T98G cells were pretreated with 0.1-0.5 mM POH for 36 h and subsequently treated with increasing doses of cisplatin (1-5.5 M) or doxorubicin (0.01-0.1 M) for 36 h. Control cells were treated with the above mentioned doses of POH, cisplatin, doxorubicin, or medium alone. At the end of the incubation, the cells were harvested, and the surviving fraction was determined by clonogenic survival assays as described above.
Apoptosis Assays-The percentages of cells undergoing apoptosis and the different phases of the cell cycle were quantified by the method described by Darzynkiewicz et al. (51,52). Briefly, the cells were treated with 0.1-1 mM POH for 72 h. The cells were harvested, washed with PBS, fixed in ice-cold 95% ethanol, and stained using a propidium iodide staining solution containing 20 g/ml PI, 200 g/ml RNase H, and 0.1% Nonidet P-40 solution. The cells were filtered through a 40-micron pore nylon mesh (Tetkop, Inc.) and analyzed using a Beckton-Dickinson FACStar Plus flow cytometer with excitation of 488 nm. The percentage of cells in sub-G 0 /G 1 (apoptotic) peak and the distribution of cells in G 0 /G 1 , S, and G 2 /M phases were determined using the ModfitLT version 2.0 (Verity Software, Topham, ME) and Cell Quest software (Becton Dickinson, San Jose, CA).
POH-induced cell death was confirmed by a dual staining technique of glioma cells with annexin V and PI, via flow cytometry (53). Cell lines were treated with POH, harvested by cell dissociation solution, washed, and subsequently stained with antibody to annexin V conjugated to FITC and with PI (10 g/ml) using the Apoalert kit (Clontech) according to the manufacturer's instructions. Viable (annexin V Ϫ /PI Ϫ ], preapoptotic (annexin V ϩ /PI Ϫ ), apoptotic (annexin V ϩ /PI ϩ ), and residual damaged cells (annexin V Ϫ /PI ϩ ) were quantified using Cell Quest software (Becton Dickinson).
Labeling Cells with BrdUrd-Subconfluent cells of T98G and MO59K cell lines were treated with 0.3-1 mM POH for 48 h. Both POH-treated and untreated cells were pulsed with 10 mM BrdUrd solution 30 min prior to harvesting of cells. The cells were harvested by trypsinization and fixed in ethanol. The cells were washed with PBS, resuspended in 0.5 ml of pepsin HCl solution (0.4 mg/ml pepsin in 0.1 N HCl), and incubated for 30 min in the dark to digest the plasma membrane. The nuclei were pelleted by centrifuging at 2000 rpm and resuspended in solution containing 2 N HCl and 0.1 M sodium tetraborate. The nuclear pellet was centrifuged and resuspended in PBS-TB (i.e. PBS containing 0.1% bovine serum albumin) and stained with anti-BrdUrd antibody (BD Pharmingen, San Diego, CA) for 90 min at room temperature, after which the pellet was washed again and incubated with anti-mouse secondary antibody conjugated to FITC (BD Pharmingen, San Diego, CA). The pellet was washed and resuspended in propidium iodide (50 g/ml) staining solution overnight. The distribution of cells in G 0 /G 1 , S, and G 2 /M phases were determined using Cell Quest software (Becton Dickinson).
Dual Staining of Fas-L and Receptor-Glioma cell lines were treated with 0.1-0.5 mM POH in the presence and absence of 5.5-Gy radiation for 72 h. At the end of the incubation, the cells were harvested using cell dissociation solution. The resulting cell pellet was incubated with biotinylated antibody specific to the human Fas-L or biotinylated IgG1 k isotype control for 45 min at 4°C. The cells were washed and incubated with APC-conjugated antibody specific to the human Fas receptor (Pharmingen) and streptavidin-conjugated phycoerythrin. The cells incubated with biotinylated IgG1 k isotype were incubated with streptavidin-conjugated phycoerythrin as well as streptavidin conjugated to APC to determine the nonspecific fluorescence of each treated sample.
The cells were washed, and before analysis, propidium iodide was added to the cells at a concentration of 20 g/ml. This was done to determine the live cells, and the staining in this population was determined. The cells were analyzed on a FACScan flow cytometer (Becton Dickinson). The mean fluorescence intensity of Fas-L was determined on a linear scale, whereas the Fas staining was determined on a log scale using Cell Quest software (Becton Dickinson).
Assessment of Anti-Fas Antibody-mediated Cytotoxicity and Apoptosis-Glioma cell lines were treated with 1 mM POH in the presence and absence of 5.5 Gy radiation. To determine Fas-mediated cell death, anti-Fas monoclonal antibody (CH11) was added to the cultures at a concentration of 100 ng/ml in the presence or absence of cycloheximide (10 g/ml). For inhibition experiments, the antagonistic antibody ZB4 was added at a concentration of 250 ng/ml to the cells. Following a 48-h  The cells were harvested using cell dissociation solution, and the resulting pellet was washed and incubated with biotinylated anti-human antibody to Fas-L (BD Pharmingen) or with an isotype-matched anti-human antibody, which served as a control to eliminate nonspecific binding. Following the primary antibody incubation, the cells were washed and subsequently stained with secondary antibody conjugated to phycoerythrin (BD Pharmingen) and antibody to CD95 directly conjugated to allophycocyanin. The stained cells were analyzed using a Becton-Dickinson FACStar Plus, and the membrane-bound Fas-L and Fas staining was recorded as mean fluorescence intensity for each sample. Each point in the graph represents the average mean fluorescence intensity value Ϯ S.E. of triplicate dishes after subtracting the mean fluorescence intensity from the corresponding isotype controls. The cells were harvested, and a clonogenic survival assay was performed as previously described.

Quantification of Fas-Ligand and Receptor Transcripts by Real Time
Quantitative PCR Assay-T98G control cells and two stable transfectants expressing FADD-DN (T98G-FADD-DN6) and overexpressing Crm A (T98GCrmA4) proteins were treated with 0.1, 0.3, and 0.5 mM POH in the presence and absence of 5.5-Gy radiation. The cells were harvested at 2, 6, 24, and 72 h, respectively. Total RNA was extracted by using the Qiagen RNA extraction kit according to the manufacturer's instructions. The quality of the RNA samples was determined by electrophoresis through agarose gels and staining with ethidium bromide, and the 18 and 28 S RNA bands were visualized under UV light. For cDNA synthesis, the reverse transcription polymerase reaction was performed using Ready to Go reverse transcriptase-PCR beads (Amersham Biosciences) using 1 g of purified RNA according to the manufacturer's instructions.
Primers-Primers and probes for the Fas-ligand and Fas receptor genes and 18 S genes were chosen with the assistance of the Primer Express (PerkinElmer Life Sciences) computer programs. We conducted searches on GenBank TM and EMBL data base sequences to confirm the total gene specificity of the nucleotide sequences chosen for the primers. Primers were purchased from Invitrogen. The sequences of the primers used for the real time quantitative PCR were 5Ј-CCCTGAGCCACAAG-GTCTACA-3Ј (up) and 5Ј-GCCCACATCTGCCCAGTAGT-3Ј (down) for The analysis of each sample was normalized on the basis of its 18 S content, which served as an endogenous control to account for variability in the initial concentration quality of the total RNA and the conversion efficiency of the reverse transcription reaction. N-fold differences in Fas-L and Fas receptor gene expression relative to the 18 S gene were calculated using the formula N(Fas-L or Fas) ϭ 2 (⌬Ct(Fas-L or Fas) Ϫ (⌬Ct(18 S)) (56).

Perillyl Alcohol Induced Radiosensitization in T98G Cells-
Clonogenic survival assays were performed on T98G cells pretreated with varying doses of POH (0.1-1 mM) and radiation (1 to 5.5 Gy). As seen in Fig. 1, pretreatment of T98G with 0.1, 0.3, and 0.5 mM POH for 72 h did not drastically affect cell survival. Treatment of cells with higher concentrations between 0.7 and 1 mM for 72 h resulted in the reduction of the surviving fraction of these cells. Hence, radiosensitization experiments were designed in the dose range between 0.1 and 0.5 mM of POH. T98G (Fig. 1) showed a dose-dependent radiosensitization by POH pretreatment to radiation-induced cell death. Since pretreatment with 0.5 mM POH revealed the greatest radiosensitization, a time course radiosensitization experiment was conducted with 0.5 mM POH and 5.5-Gy radiation. T98G cells were pretreated with 0.5 mM POH for various time intervals between 2 and 24 h. At the end of each time point, the cells were irradiated with 5.5 Gy radiation, immediately replated at an optimum cell number, and incubated for 2 weeks. The colonies were stained with crystal violet, and the surviving fraction was   11. Western blot of T98G transfectants expressing FADD-DN and Crm A proteins. The Western blotting for detecting the Fas-L, FADD, or FLAG epitope-tagged FADD-DN and Crm A proteins was carried out according to the method described by Herr et al. (59). Briefly, were harvested by scraping and lysed for 5 min on ice using 30 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 0.5% Triton X-100, 0.5% sodium deoxycholate, 1 mM dithiothreitol, 1 mM Na 3 VO 4 , 1 mM phenylmethylsulfonyl fluoride, 2 g/ml pepstatin, 2 g/ml aprotinin, 2 g/ml leupeptin. After centrifugation, 40 g of protein/lane were separated on 12% SDS-PAGE and probed with the relevant antibodies (anti-FLAG or -FADD or ␣-tubulin) and subsequently with anti-rabbit or anti-mouse secondary antibody conjugated to horseradish peroxidase, and finally the bands were detected by ECL reagents (Amersham Biosciences) according to the manufacturer's protocol.
determined as the ratio of plating efficiencies for each irradiated group to that of the unirradiated control. At each time point, there was a parallel set of cells treated with either 0.5 mM POH or 5.5-Gy radiation or medium alone. The results show that the earliest onset of radiosensitization was observed after 8 h of POH treatment (Fig. 2).
Perillyl Alcohol Induced Apoptosis and Induced Changes in Cell Cycle Progression-T98G cells treated with POH (0.1-1 mM) for a period of 24, 48, and 72 h demonstrated a dose and time-dependent increase in the sub-G 0 /G 1 population (Fig. 3A). POH-induced cell death was confirmed by staining for the presence of annexin V on the cell membrane via dual colored (propidium iodide and annexin V-FITC) flow cytometry. Cells treated with 1 mM POH showed an increase the early apoptotic (annexin V ϩ /PIϪ) and apoptotic (annexin V ϩ /PI ϩ ) population of cells (Fig. 3B).
Cell cycle analysis of POH-treated cells in the presence and absence of radiation revealed an increase in the G 2 /M phase of the cell cycle and a decrease in the G 0 /G 1 phase of the cell cycle (Fig. 4). The increase in the G 2 /M phase of the cell cycle was obvious at POH concentrations higher than 0.3 mM, and it appeared to precede the onset of apoptosis. In order to accurately determine the changes in the cell cycle profile following POH treatment, T98G cells were treated with varying concentrations of POH for 48 h and subsequently stained with anti-BrdUrd antibody conjugated to FITC and PI, and the distribution of cells in G o /G 1 , S, and G 2 /M phases was determined. The results of this experiment (Fig. 5)  mediated Apoptosis-Radiation-induced apoptosis can be mediated through pathways initiated by either DNA damage or ceramide-induced Fas signaling (57)(58)(59). The Fas receptor/ligand system has been shown to synergize with several cytotoxic drugs like doxorubicin, vincristin, teniposide taxol cisdiaminedichloroplatinum, VP-16, and campo thecin (36,37,60,61) and ionizing radiation (40) and in response to cross-linking with agonistic anti-Fas antibodies (62,63). Hence, we determined the role of the Fas receptor/ligand system in POHmediated radiosensitization in T98G cells. T98G cells treated with varying concentrations of POH for 72 h in the presence and absence of radiation were stained for a membrane-bound form of the Fas-L and Fas receptor by dual color flow cytometry. The results in Fig. 6 demonstrate the dose-dependent up-regulation of the membrane-bound Fas-L in T98G, treated with 0.1-0.5 mM POH in the presence and absence of radiation (5.5 Gy), whereas the levels of Fas receptor remained unchanged.
To further evaluate the involvement of the Fas-mediated death in POH-mediated radiosensitization, T98G cells were treated with 1 mM POH in the presence and absence of agonistic anti-Fas antibody CH11 (which recognizes a functional epitope of the Fas antigen), 5.5-Gy radiation, cycloheximide (CHX), and the antagonistic antibody ZB4 (that inhibits Fasmediated apoptosis). As shown in Fig. 7, T98G cells appeared to be sensitive to cell death mediated by the anti-Fas antibody CH11 in the presence of CHX, confirming the intactness of the Fas signaling cascade. The augmentation of Fas-mediated cell death by CHX treatment suggested that short lived cytoprotective proteins synthesized by glioma cells could confer resistance to Fas-mediated cell death. POH pretreatment sensitized all the cell lines to Fas-mediated apoptosis in the absence of CHX. The highest amount of cell death was induced by a combination of POH, radiation, and anti-Fas antibody CH11. Pretreatment with the antagonistic antibody ZB4 abrogated the activation of the Fas cascade by POH and CH11 (Fig. 8). The above findings confirm the involvement of the Fas cascade in POH-mediated cell death.
Perillyl Alcohol Pretreatment Enhanced the Cell Death Induced by Cisplatin and Doxorubicin-DNA-damaging agents, like cisplatin (64), bleomycin (65), doxorubicin (66 -68), and ionizing radiation (40,41,69) enhance Fas-mediated apoptosis. To assess the ability of POH to synergize with potent inducers of the Fas cascade, T98G cells were pretreated with POH and subsequently treated with increasing doses of cisplatin (1-5.5 M) or doxorubicin (0.01-0.1 M), and surviving fraction was determined by clonogenic survival assays. The results shown in Fig. 9, A and B, revealed a dose-dependent sensitization to cell death induced by cisplatin (Fig. 9A) and doxorubicin (Fig. 9B) by POH pretreatment. Thus, POH could synergize with other potential inducers of the Fas pathway. A parallel set of cultures were harvested and fixed to estimate the percentage of cells undergoing apoptosis and determine the effects on cell cycle via flow cytometric analysis of PI-stained cells. The results (Fig.  10, A and B) revealed a slight increase in the sub-G 0 /G 1 population of the cells at higher concentrations of the drug. The combination of POH with cisplatin or doxorubicin caused a substantial increase in the G 2 /M phase of the cell cycle and a decrease in the G 0 /G 1 phase of the cell cycle. Thus, accumulation of cells in the G 2 /M phase of the cell cycle appeared to play a crucial role in POH-mediated radio-/chemosensitization.

Perillyl Alcohol-mediated Apoptosis and Radiosensitization Was Partially Abrogated in Cell Lines Expressing FADD-DN
Transformants-Aggregation of Fas with its ligand or with a monoclonal antibody transmits a signal into the cytoplasm that induces sequential recruitment of the protein FADD and the proenzyme form of caspase-8 (70,71). Once recruited to FADD, procaspase-8 is autocatalytically cleaved to its active form and can initiate further activating cleavage of downstream caspases, including caspase-3, the central effector caspases. To identify the mechanistic link between up-regulation of Fas-L and POH-induced p53-independent enhancement in Fas-mediated apoptosis, stable transfectants of glioma cell line T98G expressing dominant negative mutants (FADD-DN) were generated, and the clones were characterized. Since Fas-induced apoptosis involved the activation and the participation of caspase 8, we generated stable transfectants of T98G cells overexpressing Crm A. Crm A is a product of the cowpox virus and acts as the endogenous inhibitor of exogenous Fas-L-induced apoptosis. The clones were selected in the presence of puromycin and screened for the presence of the transfected protein by immunoblotting with anti-FLAG antibody (Fig. 11). Clonogenic survival assays were performed with selected clones treated with 1-5.5-Gy doses of radiation. The results in Fig. 12A revealed that the presence of FADD-DN or Crm A proteins did not alter the radiation-induced cell death. POHinduced radiosensitization was significantly abrogated in T98G cells expressing FADD-DN (FADD-DN6) (Fig. 12B) and (FADD-DN3) (Fig. 12C) and the Crm A protein (T98GCrmA4) (Fig. 12D). T98G overexpressing Crm A demonstrated complete inhibition of POH-induced radiosensitization.
Thus, the presence of FADD-DN/Crm A proteins specifically protected cells from POH-induced radiosensitization. The lack of complete inhibition to POH-induced radiosensitization in cells expressing FADD-DN could be attributed to the presence of endogenous FADD, which would compete with the dominant negative FADD in these cells. The above findings confirm the involvement of the Fas cascade and caspase 8 in POH mediated cell death and radiosensitization. Clones overexpressing FADD-DN and Crm A were relatively more resistant to POHinduced apoptosis and did not exhibit a block in the G 2 /M phase of the cell cycle (Fig. 13).
Real time PCR was performed to determine the differences in the levels of Fas-L and Fas receptor mRNA in control and T98G cells expressing dominant negative FADD-DN and Crm A4 proteins. The induction of the Fas receptor expression did not change significantly following POH treatment. This could be explained by the presence of mutated p53 in this cell line. Radiation treatment induced a slight increase in the -fold induction (Ͻ10%) of the Fas receptor (Fig. 14 A-D). The results depicted in Fig. 15 A-D, revealed the induction of the Fas-L in POH-treated T98G control cells in the presence and absence of radiation between 6 and 24 h of treatment. The time kinetics of Fas-L induction support the data generated by other investigators . T98G cells expressing dominant negative FADD-DN and overexpressing Crm A revealed a decreased expression of the Fas-L and receptor. POH treatment marginally altered the expression of the ligand as well as the receptor. Thus, the lack of Fas-L up-regulation and the presence of FADD-DN/Crm A protected these cells from POH-induced radiosensitization.
POH-induced Apoptosis, Radiosensitization, and Up-regulation of the Fas-L in U251, C6, and U87 Glioma Cell Lines-POH-mediated radiosensitization via up-regulation of the Fas-L was confirmed on three other glioma cell lines. As depicted in Table I, POH induced a dose-dependent radiosensitization in U251 and C6 cell lines. Thus, the ability of POH to induce radiosensitization was not restricted to T98G cells. POH pretreatment induced apoptosis caused an increase in the G 2 /M phase of the cell cycle (Fig. 16) and increased the expression of the membrane-bound form of the Fas-L (Fig. 17A) and Fas (Fig.  17B) in U87, U251, and C6 cell lines and sensitized them to Fas mediated apoptosis (Fig. 18). POH pretreatment caused a significant up-regulation of the Fas receptor in U87 and a marginal up-regulation in C6 cells harboring wild type p53. DISCUSSION Although the process of targeting radiation therapy specifically to tumors continues to be redefined, achieving additional advances with radiotherapy may require a combinatorial ap-proach. Knowledge about the regulatory molecules involved in the growth and malignancy of glioma is essential for the rational design of agents to manage the treatment of malignant glioma. Several reports have demonstrated impairment of proteins encoding for apoptotic regulators (e.g. Fas-L and Fas) in malignant glioma cells. Paradoxically, malignant glioma cells co-express Fas and Fas-L without undergoing spontaneous apoptosis (suicide) or fratricide. An increase in Fas expression but not of the Fas-L has been demonstrated during the progression from low grade to high grade glioma (73,74), and the ratio of Fas-L/Fas mRNA has been correlated with a decrease in disease free survival in breast cancer patients (75). Since the critical threshold of the Fas-L expression on the cell surface determines the outcome of Fas-mediated cell death, glioma cells may acquire resistance to Fas-mediated apoptosis by down-regulating the expression of the Fas-L, by cleaving into the soluble form, or by rendering it nonfunctional. The ability of glioma cells to undergo apoptosis may be a critical determinant of the outcome of treatment, since most of the glioma cells are resistant to radio-/chemotherapy.
The present study evaluates the usefulness of the monoterpene POH as an effective radio-/chemosensitizer in glioma cells at clinically relevant doses of POH and radiation. POH has been previously shown to induce apoptosis and cause a G 0 /G 1 arrest in breast cancer, pancreatic cancer, and leukemic cell lines. In this report, we have demonstrated the ability of POH treatment to induce a transient G 2 /M arrest either alone or in combination with radiation/cisplatin/doxorubicin. POH treatment of T98G clones expressing FADD-DN or Crm A did not result in a G 2 /M arrest. Since the G 2 /M phase of the cell cycle is the point at which maximum radiosensitivity is observed, FIG. 14. Real Time Quantitative PCR for Fas expression. T98G-FADD-DN6, T98G-Crm A4, and control T98G cells were treated with 0.1, 0.3, and 0.5 mM POH in the presence and absence of 5.5-Gy radiation. The cells were harvested at 2, 4, 24, and 72 h. Total RNA was isolated using the RNeasy RNA isolation kit (Qiagen), and 1 g was used to prepare cDNA using Ready to Go reverse transcriptase-PCR beads (Amersham Biosciences). Quantitative real time PCR was performed by monitoring in real time the increase in fluorescence of the SYBR Green dye as described using the iCycler detection system (Bio-Rad) using specific primers for the Fas and the 18 S gene. The 18 S ribosomal RNA served as an endogenous RNA control, and each sample was normalized on the basis of its 18 S content. A-D reveal N-fold differences in Fas expression in T98G control and T98G expressing FADD-DN6 and Crm A following POH and radiation treatment at 2, 6, 24, and 72 h, respectively. Each point is an average of duplicate samples.
accumulation of cells at this position in the cell cycle could be a potential mechanism for the radiosensitizing properties of POH.
POH induced up-regulation of Fas-L and sensitization to Fas-mediated apoptosis in T98G cells. The Fas-L was functional, since it induced killing of the Fas-sensitive Jurkat cell line in a dose-dependent manner in a cell-mediated cytotoxicity experiment (data not shown). T98G cells expressing the FADD-DN and Crm A exhibited inhibition of the radiosensitizing effect caused by POH treatment and lack of up-regulation of the Fas-L protein (data not shown), suggesting a role of the Fas signaling cascade in POH-mediated radiosensitization. The presence of FADD-DN has been shown to protect radiation-induced cell death in some glioma cell lines (62). FADD-DN-induced radiation resistance was observed at higher doses of radiation (Ͼ6 Gy). There have also been reports indicating the lack of involvement of the Fas and FADD signaling cascade in apoptosis induced by ionizing radiation (55). In the present FIG. 15. Real time quantitative PCR for Fas-L expression. T98G-FADD-DN6, T98G-Crm A4, and control T98G cells were treated with 0.1, 0.3, and 0.5 mM POH in the presence and absence of 5.5-Gy radiation. The cells were harvested at 2, 4, 24, and 72 h. Total RNA was isolated using the RNeasy RNA isolation kit (Qiagen), and quantitative real time PCR was performed by monitoring in real time the increase in fluorescence of the SYBR Green dye as described using the iCycler detection system (Bio-Rad) using specific primers for the Fas-L and the 18 S gene. The 18 S ribosomal RNA served as an endogenous RNA control, and each sample was normalized on the basis of its 18 S content. A-D reveal N-fold differences in Fas-L in T98G control cells and T98G expressing FADD-DN6 and Crm A after 2, 6, 24, and 72 h. Each point is an average of duplicate samples.  study, there was no significant difference in the radiation survival between FADD-DN3-, FADD DN6-, and Crm A4-expressing clones compared with the control T98G cells at a clinically relevant dose range of radiation (1-5.5 Gy).
Tumors resistant to Fas-mediated apoptosis require exposure to a sensitizing agent (interferon-␥ or inhibitors of protein synthesis) and a response-enhancing agent such as ionizing radiation to be sensitized to Fas-mediated cell death (39). Conflicting reports exist for brain tumors, where responsiveness to Fas-mediated apoptosis of medulloblastoma and glioblastoma cells has been reported to be increased by ␥-irradiation (21), whereas other studies have indicated that irradiated glioma cells do not become more susceptible to Fas-mediated cell death (76). In the present study, POH could be acting as a sensitizing agent, whereas radiation and exposure to chemotherapeutic drugs could serve as a response-enhancing agent that could trigger activation of the Fas signaling cascade and cause cell death. Thus, glioma cells could acquire resistance to chemo-/radiotherapy by developing resistance to cell death mediated by the Fas pathway. Treatment of these resistant glioma cells with a combination of monoterpenes and a particular chemotherapeutic agent or radiation could potentially overcome this resistance and restore the apoptotic program in resistant glioma cells.
POH-induced chemo-/radiosensitization appeared to be independent of the p53 status. Several reports have demonstrated a direct correlation between wild-type p53 activity and Fas up-regulation after exposure to ionizing radiation or DNAdamaging agents, strongly suggesting that postirradiation Fas up-regulation is dependent on wild-type p53 activity (39 -41, 65, 77-81). Glioma cell lines expressing wild type p53 (U87 and C6) respond to POH treatment by up-regulation of Fas. Conversely, up-regulation of Fas-L has been demonstrated to occur in a p53-independent fashion. Cell lines with wild type p53 (U87 and C6) as well as mutated p53; U25I and U373 (CGT(Arg) 3 CAT(His)); and T98G (mutated CGT(Arg) 3 CAT(His)) cells reveal up-regulation of the Fas-L. The kinetics of the up-regulation of Fas-L in POH-treated T98G cells using the real time quantitative reverse transcription polymerase chain reaction demonstrated the induction of Fas-L between 6 and 24 h of POH treatment. Fas receptor expression was not altered following POH treatment. The kinetics of the Fas-L up-regulation using the real time quantitative reverse transcription polymerase chain reaction correlated with the onset of radiosensitization of T98G cells as illustrated in Fig. 2. The p53-independent up-regulation of Fas-L induced on the surface of glioma cell lines by POH could potentially trigger cell death in an autocrine or paracrine manner via cross-linking of Fas, or alternately, POH in combination with ionizing radiation or chemotherapeutic agent could act synergistically to activate the Fas pathways, resulting in enhanced sensitivity to these agents.
Further experiments are needed to elucidate the mechanism of activation of the Fas-Fas-L cascade by POH and ionizing radiation in glioma cells. Ionizing radiation has been shown to induce the production of ceramide, which in turn can activate the Fas signaling pathway, leading to apoptosis (59). The radio resistance of glioma cells could be attributed to the inability of these tumor cells to hydrolyze sphingomyelin and to produce ceramide in response to radiation (82,83). POH treatment FIG. 17. Dual staining of Fas-L and receptor on perillyl alcohol-treated U373, U251, U87MG, and C6 cells. Subconfluent cultures of U373, U251, U87MG, and C6 cells were treated with 0.1-1 mM perillyl alcohol for 72 h and stained for the membrane-bound Fas-L (A) and Fas (B) by flow cyotmetry as described before. Each point in the graph is an average Ϯ S.E. values from two experimental sets after subtracting the MFI from the corresponding isotype controls.
FIG. 18. Perillyl alcohol-induced sensitization to Fas-mediated apoptosis in glioma cell lines. Glioma cell lines were treated with 1 mM POH in the presence and absence of radiation. Anti-Fas monoclonal antibody (Kamiya Biochemical Co.) was added to cultures as indicated at a concentration of 100 ng/ml in combination with 10 g/ml cycloheximide. Following incubation, the cells were harvested, and quantitative analysis of apoptosis was performed by flow cytometric analysis of fixed cells stained with propidium iodide to identify cells with a subdiploid DNA content as described before. Each point in the graph average values of duplicate dishes. could potentially induce the production of ceramides, which in turn would sensitize the cells to Fas-mediated apoptosis. These present findings can be extended to other malignancies with known abnormalities in the Fas signaling cascade (prostate, colon, pancreas, lung, hepatomas) and eventually be the basis for clinical studies evaluating monoterpenes as radio-/chemosensitizers in newly diagnosed malignant glioma patients.