mTOR is a promising therapeutic target both in cisplatin-sensitive and cisplatin-resistant clear cell carcinoma of the ovary.

Ovarian carcinoma is the fourth most common cause of cancer death among women in the United States, with more than 21,000 new cases each year and an estimated 15,520deaths in 2008 ( 1 ). Cytoreductive surgery followed by platinum-based chemotherapy usually combined with paclitaxel is the standard initial treatment and has improved survival in patients with epithelial ovarian cancer ( 2 ). However, there still exists many clinical problems in the treatment of epithelial ovarian cancer. One of the most important problems that needs to be resolved is the management of clear cell carcinoma (CCC) of the ovary, which was first recognized by the World Health Organization as a distinct histological subtype in 1973 ( 3 ). The precise incidence of CCC is unknown, but it is reported to be 3.7~12.1% of all histological subtypes among epithelial ovarian cancer ( 4 ). There have been two major clinical problems in the clinical management of CCC. First is its poor sensitivity to first-line platinum-based chemotherapy and the association with a worse prognosis than the more common serous adenocarcinomas (SAC). In the setting of front-line chemotherapy, the response rate to conventional platinum-based chemotherapy, platinum agent alone or in combination with cyclophosphamide and adriamycin, was reported to be only 11% in CCC. In contrast, patients with SAC had a response rate of 72% ( 5 ). The response to carboplatin-paclitaxel, a current standard regimen, was also reported to be relatively low, ranging from 22% to 56% ( 6 - 8 ). When analyzed by clinical stage, worse clinical outcome in patients with CCC has been more evident in advanced (stage III-IV) than in early stage disease (stage I-II). In a retrospective analysis ( 5 ), a statistically significant difference in overall survival between CCC and SAC was observed in patients with stage III disease (12.7 months vs 26.8 months, respectively; p=0.0015). However, the difference was not significant in stage I-II disease (31.8 months vs 42.3 months, respectively; p=0.2761). Similar results were reported by several groups of investigators ( 9 - 11 ). A more recent retrospective review of six randomized phase III clinical trials also demonstrated that patients with stage III CCC treated with carboplatin-paclitaxel had a shorter survival compared to those with other histological subtypes of epithelial ovarian cancer ( 12 ). The second important clinical problem in the management of CCC is the lack of effective chemotherapy for recurrent CCCs after front-line treatment with platinum-based chemotherapy. A recent report demonstrated that the response rate for various regimens in the setting of second-line chemotherapy for recurrent CCC was only 1% ( 13 ). Therefore, to improve survival of patients with CCC, a better understanding of the mechanism of platinum-resistance and the identification of effective treatment strategies especially for both advanced and recurrent disease are needed. The sensitivity of cancer cells to chemotherapeutic drug-induced apoptosis depends on the balance between pro-apoptotic and anti-apoptotic signals. Therefore, inhibition of anti-apoptotic signals, such as those mediated by the AKT pathway, has been proposed as a promising strategy to enhance the efficacy of conventional chemotherapeutic agents ( 14 ). Among the numerous AKT substrates, mTOR is thought to be one of the major targets of relevance to cancer therapy ( 15 , 16 ). mTOR phosphorylates p70 S6 kinase (p70S6K) and the 4E-BP1 translational repressor, leading to translation of proteins required for cell proliferation ( 17 ). It has been reported that AKT-mTOR signaling is frequently activated in epithelial ovarian cancer ( 18 ). Recently, an orally bioavailable derivative of rapamycin, everolimus (RAD001), has been shown to inhibit the proliferation of ovarian cancer cells and enhance sensitivity to cisplatin in vitro and in vivo ( 19 - 22 ). However, no reports have addressed the impact of mTOR inhibitors on ovarian cancer cells that have acquired resistance after the exposure to platinum agents. Moreover, since most tumor specimens and tumor-derived cell lines used in these investigations have been ovarian SACs ( 19 - 21 ), the role of mTOR in CCC remains largely unknown. It has been reported that loss of PTEN expression is common in CCC of the ovary ( 23 ). It also has been reported that ovarian endometriosis, from which CCC is thought to arise, is characterized by hyperactivation of the AKT-mTOR pathway ( 24 ). Since it is well known that loss of PTEN expression and consequent activation of AKT signaling result in hypersensitivity to mTOR inhibition ( 20 , 25 , 26 ), CCC may be a good candidate for therapy with a mTOR inhibitor. In the current investigation, we examined the activation status of mTOR both in early stage and advanced stage CCC, and we determined whether RAD001 has anti-neoplastic efficacy in both in vitro and in vivo models of CCC. Moreover, we investigated the role of AKT/mTOR signaling in the acquired resistance to cisplatin in CCC cells.

Immunohistochemical analysis of ovarian cancer tissue microarrays for phospho-mTOR expression was performed using 52 CCCs of the ovary and 46 ovarian SACs as described in “Material and Methods.” Representative photographs of CCC and SAC are shown in Fig. 1A . Phospho-mTOR immunoreactivity was scored semiquantitatively ( Fig. 1B ), and stronger immunoreactivity for phospho-mTOR was observed in CCCs than in SACs. Among the 46 SAC, 23 (50%) showed negative staining, 7 (15.2%) were scored as +0.5, 15 (32.6%) were +1, and 1 (2.1%) was scored as +2. In contrast, among the 52 CCCs, 7 tumors (13.4%) had negative staining. Five (9.6%) were scored as +0.5, 23 (44.2%) were +1, and 15 (28.8%) were +2, and 2 (3.8%) were scored as +3. The frequency of strong phospho-mTOR immunoreactivity was significantly higher, and frequency of tumors with no immunoreactivity was significantly lower in CCCs than in SACs ( Fig. 1C ). These results indicate that CCCs may be more strongly dependent on mTOR for tumor progression than SACs. When analyzed by clinical stage, phospho-mTOR expression was observed in 76% of advanced stage CCCs and in 96% of early stage CCCs ( Fig. 1D ). Thus, most patients with CCC may be candidates for therapy with a mTOR inhibitor. In contrast, in SACs, phospho-mTOR expression was uncommon in early stage tumors, although it was significantly increased in advanced stage tumors. Therefore, in SACs, mTOR inhibition may be a therapeutic option only in advanced stage disease. Collectively, these results indicate that pharmacologic inhibition of mTOR may be a promising therapeutic strategy in the management of CCCs, both in early stage and in advanced stage disease. Given the frequent mTOR activation found in human CCC tumor specimens ( Fig. 1 ), we evaluated the expression of phospho-mTOR in four human CCC cell lines by western blotting. As shown in Fig. 2A , under serum-starvation conditions, mTOR was phosphorylated in all CCC cell lines tested, which is consistent with immunohistochemical results observed with tumor samples. We next examined the efficacy of mTOR pathway inhibition by RAD001 on the proliferation of CCC cells in vitro . For this purpose, we performed a MTS assay using two of these CCC cell lines with activated AKT/mTOR signaling. As shown in Fig. 2B , RAD001 inhibited the proliferation of RMG1 and KOC7C cells in vitro , with ~25% inhibition at the highest drug concentration tested ( Fig. 2B ). To determine if the anti-proliferative effects of RAD001 result from inhibition of mTOR signaling, we examined the effect of RAD001 on the phosphorylation of downstream p70S6K in RMG1 and KOC7C cells. As shown in Fig. 2C , AKT, mTOR and p70S6K were phosphorylated in both cell lines, indicative of the hyperactivation of the AKT/mTOR pathway. As expected, phosphorylation of the downstream effector p70S6K was significantly decreased in both cell lines by treatment with RAD001, indicating that RAD001 effectively inhibits mTOR signaling in CCC cells. Although previous studies have shown that mTOR inhibition is associated with a feedback activation of AKT which may result in resistance to mTOR inhibition ( 32 - 35 ), no significant increase in the phosphorylation of AKT was observed in response to RAD001 in these CCC cell lines. Rapamycin and its derivatives are generally regarded as having cytostatic effects; however, in some tumor cells, these agents have also been reported to induce apoptosis ( 36 ). To determine the mechanism by which RAD001 inhibits cell proliferation, we first examined the effect of RAD001 on cell cycle progression by flow cytometry. As shown in Fig. 2D (i), the percentage of cells in G1 phase was significantly increased in both RMG1 and KOC7C cells after 2-day treatment with 10 nM RAD001. In both cell lines, the percentage of apoptotic cells in the sub-G1 peak did not change after treatment with RAD001. Moreover, as shown in Fig. 4B , treatment with 10 nM RAD001 did not induce cleavage of PARP in these cells. We also examined whether treatment with RAD001 induces autophagic cell death in CCC cells. It has been reported that LC3B-I is converted to LC3B-II during autophagy ( 37 ). However, as shown in Fig. 2D (ii), the conversion of LC3B-I to the lower migrating form LC3B-II was not induced in response to treatment with RAD001 in RMG1 or KOC7C cells. Furthermore, as shown in Fig. 2D (iii), treatment with 10 nM of RAD001 did not induce punctate staining for LC3B, an indicator of authophagy associated with the concentration of LC3 in autophagosomalvacuoles ( 37 ). Collectively, these results suggest that RAD001 most likely affects CCC cells by inducing cell cycle arrest ( 17 ). To further examine the in vivo growth-inhibitory effect of RAD001, we employed a subcutaneous (s.c.) xenograft model in which athymic mice were inoculated s.c. with RMG1 or KOC7C cells. When tumors reached ~50 mm 3 , the mice were randomized into two treatment groups receiving placebo or RAD001, as described in “Material and Methods.” Drug treatment was well tolerated, with no apparent toxicity throughout the study. Tumor volume was measured weekly after the start of treatments ( Fig. 3B and 3D ). The appearance of tumors four weeks from the first day of treatment is also shown in Fig. 3A and 3C . Histologically, these subcutaneous tumors were CCCs (data not shown). Mean RMG1-derived tumor burden in mice treated with RAD001 was 332.5 mm 3 compared to 652.5 mm 3 in placebo-treated mice, and mean KOC7C-derived tumor burden in animals treated with RAD001 was 276 mm 3 compared to 605.5 mm 3 in placebo-treated mice. Overall, treatment with RAD001 decreased RMG1 derived- and KOC7C derived-tumor burden by 49% and 55%, respectively, compared to placebo. These results indicate that RAD001 hassignificant anti-tumor effects as a single agent in CCC. Cisplatin-resistance is regarded as a major clinical problem in the management of CCC of the ovary ( 11 , 38 ). It has been previously reported that AKT is involved in the resistance of ovarian SAC cells to cisplatin ( 39 , 40 ). To examine whether AKT/mTOR signaling is involved in cisplatin-resistance in CCC, we established cisplatin-resistant sublines from RMG1 and KOC7C cells, as described in “Material and Methods.” To examine whether these sublines had acquired resistance to cisplatin, we first evaluated the sensitivity of these cell lines to cisplatin by MTS assay. As shown in Fig. 4A , clear differential sensitivity to cisplatin was observed between cisplatin-sensitive parental and respective cisplatin-resistant sublines. We next examined cisplatin-induced apoptosis in these cell lines. Treatment with cisplatin induced cleavage of PARP in parental cells, but not in cisplatin-resistant sublines ( Fig. 4B ). Using these cell lines, we have investigated the activity of AKT/mTOR in both cisplatin-resistant sublines and parental chemosensitive cells by western blotting. As shown in Fig. 4C , higher phospho-AKT and phospho-mTOR expression was observed in both chemoresistant cell lines compared with their respective parental cell lines. Increased activation of AKT/mTOR signaling was also observed in another cisplatin-resistant subline, HAC2-CR, which was established from parental HAC2 cells (data not shown). The increased phosphorylation of AKT and mTOR was inhibited by treatment with a PI3K inhibitor,LY294002 (data not shown). Since it is well known that loss of PTEN expression and consequent activation of AKT result in hypersensitivity to mTOR inhibition ( 20 , 25 ), we considered chemoresistant sublines to be good candidates for treatment with RAD001. Thus, we next examined the inhibitory effect of RAD001 on chemoresistant and parental chemosensitive CCC cell lines by MTS assay ( Fig. 4D ). A clear differential effect was demonstrated depending on the cell sensitivity to cisplatin. Cisplatin-resistant RMG1-CR and KOC7C-CR cells are significantly more sensitive to RAD001 than their respective parental cell lines RMG1 and KOC7C. We also confirmed that treatment with RAD001 effectively inhibited the phosphorylation of p70S6K in vitro , without inducing negative feedback activation of AKT (data not shown). Moreover, using RMG1-CR and KOC7C-CR cells, we next determined whether the treatment with RAD001 enhances the efficacy of cisplatin. As shown in Fig. 4E , in the presence of 10 nM of RAD001, the ability of cisplatin (0-10μM) to inhibit cell proliferation was not enhanced in these cisplatin-resistant cell lines. These results suggest that RAD001 may have efficacy as a single agent for cisplatin-resistant CCCs. To further examine the in vivo effect of RAD001 on cisplatin-resistant sublines, athymic mice were inoculated s.c. with RMG1-CR or KOC7C-CR cells, and were randomized into two treatment groups receiving placebo or RAD001, as described in “Material and Methods. The appearance of the tumors four weeks from the first day of treatment is shown in Fig. 5A , C. Moreover, corresponding graphs depicting diminished tumor volumes for RAD001-treated mice relative to placebo-treated mice are presented in Fig. 5B , D. Mean RMG1-CR-derived tumor burden in mice treated with RAD001 was 163 mm 3 compared to 553 mm 3 in placebo-treated mice, and mean KOC7C-CR-derived tumor burden in animals treated with RAD001 was 218.5 mm 3 compared to 710 mm 3 in placebo-treated mice. Treatment with RAD001 decreased RMG1-CR-derived tumors burden by 72% compared to only 49% reduction in RMG1-derived tumors ( Fig. 3C, D ). Similar results were obtained in mice inoculated with KOC7C-CR cells. Treatment with RAD001 decreased KOC7C-CR derived tumor burden by 69% compared to a 55% reduction in RAD001-treated KOC7C-derived tumors ( Fig. 3A, B ). Collectively, these in vitro and in vivo data suggest that the anti-tumor effect of RAD001 is greater in cisplatin-resistant CCC than in cisplatin-sensitive CCC.

Despite recent developments in platinum-based combination chemotherapy, patients with CCC of the ovary, especially in advanced stage or recurrent disease, have a worse progression free survival and overall survival when compared with patients with a serous histology ( 5 - 11 ). Therefore, to improve survival, new strategies are necessary to more effectively treat CCC. In the present study, we observed activation of mTOR in 86.6% of CCC of the ovary ( Fig. 1 ). Importantly, the frequency of strong phospho-mTOR immunoreactivity in CCCs was significantly higher than that found in SACs, indicating that CCCs are more strongly dependent on mTOR signaling for tumor progression than are SACs. In addition, mTOR was frequently activated in both stage III-IV CCCs (76%) and stage I-II CCCs (96%). Therefore, mTOR appears to be a promising target for the treatment of patients with both early and advanced stage CCC. In contrast, phospho-mTOR expression was uncommon in early stage SACs but was significantly increased in advanced stage SACs. The very high frequency of mTOR activation observed in early stage CCCs suggests that hyperactivation of mTOR kinase is an early event in the development of CCCs. This is noteworthy in light of the fact that activated AKT/mTOR signaling has been reported in ovarian endometriosis, from which CCC is thought to arise ( 24 ). We have recently demonstrated that the mTOR inhibitor RAD001 markedly inhibited tumor onset and progression in a transgenic mouse model of ovarian cancer that develops ovarian SACs with activated AKT/mTOR signaling. Thus, mTOR might be a reasonable target for the chemoprevention of CCC in patients with ovarian endometriosis. Our data demonstrate that treatment with RAD001 effectively attenuates the phosphorylation of p70S6K in vitro and markedly inhibits the proliferation of ovarian CCC cells. There exists a concern in inhibiting mTOR, in that mTOR inhibition may trigger a feedback mechanism that activates AKT to potentially promote tumor growth and may consequently reduce the anti-tumor effect of mTOR inhibitors ( 32 - 35 ). Although such a feedback has been observed in several cancer cell types including breast cancer ( 32 ), rhabdomyosarcoma ( 33 ), non-small cell lung cancer ( 34 ), and multiple myeloma ( 35 ), in the current study treatment with RAD001 did not induce activation of AKT in ovarian CCC cells ( Fig. 2D ). We also evaluated the efficacy of RAD001 in vivo , employing s.c. xenograft models ( Fig 3 ). In mice inoculated s.c. with RMG1 or KOC7C cells, treatment with RAD001 significantly inhibited tumor growth. Moreover, orally administrated RAD001 in our treatment schedule was well tolerated. Taken together, these findings indicate that RAD001 could have significant anti-tumor effects as a single agent for CCC in a setting of front-line therapy. An additional important finding in our study is the anti-tumor activity of RAD001 in cisplatin-resistant CCC. In general, patients with platinum-resistant recurrent epithelial ovarian cancer have been treated with anti-neoplastic agents that do not exhibit cross-resistance with platinum-agents. However, these patients have dismal prognosis, with overall response rate ranging from 9% to 33% ( 41 ). Unfortunately, the prognosis of patients with cisplatin-resistant CCCs is even worse. For example, in one study, the response rate for salvage chemotherapy for cisplatin-resistant CCC was only 1% ( 13 ), indicative of the urgent need of new treatment strategies for recurrent CCC of the ovary. In this study, we found that cisplatin-resistant CCC cell lines exhibit enhanced phospho-mTOR expression compared to the corresponding cisplatin-sensitive parental cell lines ( Fig. 4B ). The increased phospho-mTOR expression was associated with increased activation of AKT. The involvement of AKT in the resistance to cisplatin has been reported previously ( 39 , 42 ). Although we and others have previously reported that inhibition of AKT activity sensitizes human ovarian cancer cells to conventional anticancer agents such as cisplatin ( 39 ) and paclitaxel ( 43 ), there are concerns associated with inhibiting AKT, because AKT also mediates certain biologically important cell processes such as glucose metabolism ( 44 ). Thus, a safer approach may be to target downstream therapeutic effectors such as mTOR. Interestingly, our cisplatin-resistant CCC cells showed significantly higher sensitivity to RAD001 in vitro , compared with the respective cisplatin-sensitive parental cell lines. Furthermore, the in vivo anti-tumor effect of RAD001 was also greater in cisplatin resistant cell-derived tumors than in cisplatin sensitive cell-derived tumors ( Fig. 5 ). It has been previously reported that AKT activation may be a biomarker to predict the sensitivity to mTOR inhibitors ( 20 , 25 , 26 ). Although AKT activation is not the sole determinant of sensitivity to mTOR inhibition ( 26 ), our results indicate that enhanced sensitivity to mTOR inhibitors in cisplatin-resistant CCC cells is associated with, at least in part, the activation of AKT/mTOR signaling. Since the RMG1-CR and KOC7C-CR cells used in this study mimic the clinical situation of resistance development in cisplatin-treated patients, our results may suggest that a mTOR inhibitor might have efficacy for the clinical management of cisplatin-resistant CCCs. We should note, however, that a potential limitation of our experimental design is the use of a subcutaneous xenograft model. Peritoneal dissemination is the main process involved in the progression in human ovarian cancer. Thus, intra-peritoneal injection of cancer cells would more accurately model advanced disease. Therefore, further investigation using an intra-peritoneal model or a genetically engineered mouse model of ovarian cancer would be helpful. Our results indicate that RAD001 is a promising agent for the treatment of CCC of the ovary both as a front-line treatment and as a salvage treatment for recurrence after platinum-based chemotherapy. A recent phase III study demonstrated that RAD001 had significant activity in some patients with advanced renal cell carcinoma ( 45 ). For patients with recurrent ovarian cancer, the Southwest Oncology Group (SWOG) will soon initiate a randomized phase II trial of carboplatin and paclitaxel with or without everolimus in patients with ovarian cancer in first relapse. We believe that our data support the scientific justification or this and future clinical trials with RAD001 in patients with CCC of the ovary, a chemoresistant histological subtype characterized by frequent hyperactivation of mTOR pathway.

RAD001 was obtained from Novartis Pharma AG (Basel, Switzerland). ECL Western blotting detection reagents were from Perkin Elmer (Boston, MA). Antibodies recognizing p70S6K, phospho-p70S6K (Thr389), mTOR, phospho-mTOR (Ser2448), AKT, phospho-AKT (Ser473), PARP, LC3B and β-actin were obtained from Cell Signaling Technology (Beverly, MA). The Cell Titer 96-well proliferation assay kit was obtained from Promega (Madison, WI). Cisplatin was purchased from Sigma (St. Louis, MO). RAD001 was formulated at 2% (w/v) in a microemulsion vehicle (Novartis Pharma AG). RAD001 was prepared according to the manufacturer's protocols. Thus, for animal studies, RAD001 was diluted to the appropriate concentration in double-distilled water just before administration by gavage. For in vitro analyses, RAD001 was prepared in DMSO before addition to cell cultures. All surgical specimens were collected and archived according to protocols approved by the institutional review boards (IRBs) of the parent institutions. Appropriate informed consent was obtained from each patient. The tumors included 46 SACs and 52 CCCs. Based on criteria of the International Federation of Gynecology and Obstetrics (FIGO) criteria, 22 SACs were stage I-II tumors and 24 were stage III-IV tumors. Among CCCs, 27 were stage I-II tumors and 25 were stage III-IV tumors. Tumor samples were fixed in 10% neutral buffered formalin (10% formaldehyde, phosphate-buffered) overnight and then embedded in paraffin. In all patients, the diagnosis was based on a light microscopy examination using conventional hematoxylin and eosin (H&E) stain. Ovarian cancer tissue microarrays consisting of two cores from each tumor sample were prepared by the Tumor Bank Facility at Fox Chase Cancer Center, as described previously ( 18 , 19 ). Tissue sections were cut at 4 μm, mounted on slides, and processed for either H&E or immunohistochemical staining. For immunohistochemical studies, sections were incubated with the primary antibody, followed by the appropriate peroxidase-conjugated secondary antibody, as reported previously ( 19 ). The primary antibody used was anti-phospho-mTOR (Ser 2448) at 1:50 dilution. Negative controls were incubated with primary antibody preabsorbed with blocking peptide (Cell Signaling Technology). Surrounding non-neoplastic stroma served as an internal negative control for each slide. The slides were scored semiquantitatively by a pathologist who was blinded to the clinical outcome. A score of 0 indicated no staining, +0.5 was weak focal staining (less than 10% of the cells were stained), +1 was indicative of focal staining (10-50% of the cells were stained), +2 indicated clearly positive staining (more than 50% of the cells were stained), and a score of +3 was intensely positive, as described in detail elsewhere ( 18 ). The slides were examined under a bright field microscope. Tumors with staining of +2 or +3 were grouped as strong staining-group, whereas tumors with staining of +0.5 or +1 were grouped as a weak staining-group. When the two cores from the same tumor sample showed different positivity results, then the lower score was considered valid. Human ovarian CCC cell lines RMG1, RMG2, KOC7C, and HAC2 were kindly provided by Dr. H. Itamochi (Tottori University, Tottori, Japan). These cells were cultured in phenol red free Dulbecco's Modified Eagles Medium (DMEM Ham's F-12, Gibco Ltd, Paisley, Strathclyde, UK) with 10% FBS, as reported previously ( 27 - 29 ). Cisplatin-resistant sublines from RMG1 and KOC7C were developed in our laboratory by continuous exposure to cisplatin, as described previously ( 30 ). Briefly, cells of both lines were exposed to stepwise increases in cisplatin concentrations. Initial cisplatin exposure was at a concentration of 10nM. After the cells had regained their exponential growth rate, the cisplatin concentration was doubled and then the procedure was repeated until selection at 10μM was attained. The resulting cisplatin-resistant sublines, called RMG1-CR and KOC7C-CR were subcultured weekly and treated monthly with 10 μM cisplatin to maintain a high level of chemoresistance. An MTS assay was used to analyze the effect of RAD001 on cell viability as described ( 31 ). Cells were cultured overnight in 96-well plates (1 × 10 4 cells/well). Cell viability was assessed after addition of RAD001 and/or cisplatin at the indicated concentrations for 48h. The number of surviving cells was assessed by determination of the A 490 nm of the dissolved formazan product after addition of MTS for 1 h as described by the manufacturer (Promega, Madison, WI). Cell viability is expressed as follows: A exp group /A control × 100. Cells were treated with either DMSO (vehicle) or 10 nM RAD001 for 6h. Cells were washed twice with ice-cold PBS and lysed in lysis buffer (20 mM Tris-HCl, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM Na 3 VO 4 , 1 mM β-glycerophosphate, 2.5 mM sodium pyrophosphate, 1 mM 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride, 10 μg/ml aprotinin, 1 μg/ml leupeptin, and 1% Triton X-100) for 10 min at 4°C. Lysates were centrifuged at 12,000 × g at 4°C for 15 min, and protein concentrations of the supernatants were determined using Bio-Rad protein assay reagent. Equal amounts of proteins were separated by SDS-PAGE and transferred to nitrocellulose membranes. Blocking was done in 5% nonfat milk in 1X Tris-buffered saline. Western blot analyses were performed with various specific primary antibodies. Immunoblots were visualized with horseradish peroxidase-coupled goat anti-rabbit or anti-mouse immunoglobulin by using the enhanced chemiluminescence Western blotting system (Perkin Elmer, Boston MA). Cells were incubated with or without 20 nM RAD001 for 2 days. After the cells were washed with PBS, they were fixed with 75% ethanol overnight at 4°C. The cells were then washed twice with PBS and stained with propidium iodide (50 μg/ml) in the presence of RNase A (100 μg/ml; Roth, Karlsruhe, Germany) for 20 min at 4°C. Cell cycle distribution was determined by analyzing 10,000 cells using a FACScan flow cytometer and Cell Quest software (Becton Dickinson, San Jose, CA) Cells were incubated with or without 20 nM RAD001 for 2 days. Cells were washed with ice-cold phosphate-buffered saline (PBS), fixed in 4% paraformaldehyde in PBS for 10 min, and then blocked and incubated with anti-LC3B antibody overnight at 4C. After washing with PBS, the coverslips were incubated with FITC-conjugated secondary antibody (Alexab Fluor 488 anti-rabbit IgG, Molevular Probes, Eugene, OR) for 1 h, followed by 10 min of incubation with 4',6-diamidino-2-phenylindole (DAPI). Slides were washed with PBS, mounted with Vectashield hardest mounting medium (Vector Laboratories). Images were acquired with a fluorescence microscope and processed using Photoshop software. All procedures involving animals and their care were approved by the Institutional Animal Care and Usage Committee of Osaka University, in accordance with institutional and NIH guidelines. 5-7-week-old nude mice (n=40) were inoculated s.c. into the right flank either with 5×10 6 RMG1, RMG1-CR, KOC7C, or KOC7C-CR cells in 200 μl of PBS, with 10 mice in each group. When tumors reached about 50 mm 3 , mice were assigned into two treatment groups, with 10 mice in each group. The first group was treated with placebo twice a week. The second group was treated with RAD001 (2.5 mg/kg) twice a week. RAD001 was administered intragastrically using an animal-feeding needle. Body weight was measured weekly. Caliper measurements of the longest perpendicular tumor diameters were performed every week to estimate tumor volume using the following formula: V = L × W × D × π/6, where V is the volume, L is the length, W is the width, and D is the depth. Cell proliferation was analyzed by Wilcoxon exact test. Tumor volume of RAD001-treated mice was compared with that of placebo-treated mice and analyzed by Wilcoxon exact test. Immunoreactivity was analysed using Fisher's exact test. A p-value of <0.05 was considered significant.