Role of Glycogen Synthase Kinase 3B in Rapamycin-Mediated Cell Cycle

Research Article

Role of Glycogen Synthase Kinase 3 B in Rapamycin-Mediated Cell Cycle Regulation and Chemosensitivity

JinJiang Dong,1 Junying Peng,1 Haixia Zhang,1 Wallace H. Mondesire,1 Weiguo Jian,1 Gordon B. Mills,2 Mien-Chie Hung,1,3 and Funda Meric-Bernstam1

Departments of 1Surgical Oncology, 2Molecular Therapeutics, and 3Molecular and Cellular Oncology, University of Texas M.D. Anderson Cancer Center, Houston, Texas

Abstract Introduction The mammalian target of rapamycin is a serine-threonine Rapamycin and its analogues are being actively investigated in kinase that regulates cell cycle progression. Rapamycin and clinical trials as novel targeted anticancer agents. The mammalian its analogues inhibit the mammalian target of rapamycin target of rapamycin (mTOR) is a serine-threonine kinase that and are being actively investigated in clinical trials as novel regulates cell cycle progression. The two best-studied targets of targeted anticancer agents. Although cyclin D1 is down- mTOR, eukaryotic initiation factor 4E-binding protein 1 and regulated by rapamycin, the role of this down-regulation in ribosomal p70 S6 kinase-1, are thought to modulate translation; rapamycin-mediated growth inhibition and the mechanism thus, rapamycin is thought to alter the translation of mRNA of cyclin D1 down-regulation are not well understood. Here, involved in control of the cell cycle. However, how rapamycin we show that overexpression of cyclin D1 partially over- blocks cell growth and proliferation is not well understood. comes rapamycin-induced cell cycle arrest and inhibition of Prior studies have suggested that cyclin D1 is a key target of anchorage-dependent growth in breast cancer cells. Rapa- mTOR (1–4). Cyclin D1 is overexpressed in a variety of tumor mycin not only decreases endogenous cyclin D1 levels but types and is proposed to contribute to cancer development. also decreases the expression of transfected cyclin D1, Cyclin D1 plays a critical role in G progression by activating suggesting that this is at least in part caused by accelerated 1 cyclin-dependent kinases 4 and 6, leading to phosphorylation of proteolysis. Indeed, rapamycin decreases the half-life of tumor suppressor pRb, with depression of E2F-mediated cyclin D1 protein, and the rapamycin-induced decrease in transcription (5). Independent of cyclin-dependent kinase cyclin D1 levels is partially abrogated by proteasome activity, cyclin D1 also modulates other transcription factors, inhibitor N-acetyl-leucyl-leucyl-norleucinal. Rapamycin treat- such as estrogen and androgen receptors, signal transducers and ment leads to an increase in the kinase activity of glycogen activators of transcription 3, and PPARg (6, 7). Recently, it has synthase kinase 3B (GSK3B), a known regulator of cyclin D1 been reported that transcription factor CCAAT/enhancer-binding proteolysis. Rapamycin-induced down-regulation of cyclin protein h is involved in regulating genes affected by cyclin D1 D1 is inhibited by the GSK3B inhibitors lithium chloride, overexpression (8). Although the exact role of cyclin D1 SB216763, and SB415286. Rapamycin-induced G arrest is 1 overexpression in human tumors is controversial, there are abrogated by nonspecific GSK3B inhibitor lithium chloride several lines of evidence indicating that cyclin D1 plays a crucial but not by selective inhibitor SB216763, suggesting that role in mammary gland carcinogenesis. First, cyclin D1 is GSK3B is not essential for rapamycin-mediated G arrest. 1 overexpressed in ductal carcinoma in situ and invasive ductal However, rapamycin inhibits cell growth significantly more in breast carcinoma and cyclin D1 overexpression is associated GSK3B wild-type cells than in GSK3B-null cells, suggesting with a poorer prognosis (9, 10). Second, mammary gland– that GSK3B enhances rapamycin-mediated growth inhibition. targeted cyclin D1 overexpression in mouse mammary tumor In addition, rapamycin enhances paclitaxel-induced apopto- virus-cyclin D1 transgenic mice leads to mammary hyperplasia sis through the mitochondrial death pathway; this is and development of carcinoma, suggesting that cyclin D1 at inhibited by selective GSK3B inhibitors SB216763 and least is a weak oncogene (11). Third, mice lacking cyclin D1 are SB415286. Furthermore, rapamycin significantly enhances resistant to mammary carcinogenesis by the H-ras and HER-2/ paclitaxel-induced cytotoxicity in GSK3B wild-type but not in neu oncogenes, suggesting that cyclin D1 expression is critical to GSK3B-null cells, suggesting a critical role for GSK3B in H-ras and HER-2/neu-mediated carcinogenesis (12). Finally, rapamycin-mediated paclitaxel-sensitization. Taken together, B deregulation of cyclin D1 is implicated as the central pathway these results show that GSK3 plays an important role in involved in chemical carcinogenesis models of breast cancer rapamycin-mediated cell cycle regulation and chemosensi- (13). Therefore, cyclin D1 may represent an important down- tivity and thus significantly potentiates the antitumor effects stream target of signaling pathways that have a role in of rapamycin. (Cancer Res 2005; 65(5): 1961-72) mammary carcinogenesis, making the effect of rapamycin on cyclin D1 of particular interest. Cyclin D1 expression is regulated at multiple levels. Cyclin D1 transcription is up-regulated by mitogen stimulation through extracellular signal-regulated kinase 1/2 and 5, c-Jun NH -terminal Note: J. Dong and J. Peng contributed equally to this work. 2 Requests for reprints: Funda Meric-Bernstam, Department of Surgical kinase, signal transducers and activators of transcription 5, nuclear Oncology, University of Texas M.D. Anderson Cancer Center, 1515 Holcombe factor-nB, and h-catenin and down-regulated by stress-activated Boulevard, Unit 444, Houston TX 77030. Phone: 713-745-4453; Fax: 713-745-4926; E-mail: [email protected]. kinase p38 (14–16). Cyclin D1 mRNA stability is regulated by a I2005 American Association for Cancer Research. sequence in its 3V untranslated region; its stability is increased by www.aacrjournals.org 1961 Cancer Res 2005; 65: (5). March 1, 2005

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 2005 American Association for Cancer Research. Cancer Research the phosphatidylinositol 3-kinase (PI3K) pathway and is decreased Colonies of >120 um in diameter (20 cells) were counted using a by prostaglandin A (14, 17). Cyclin D1 mRNA translation is thought microscope. to be controlled by the PI3K/Akt pathway and by mTOR signaling Cell Cycle Analysis. Cells were trypsinized and washed with PBS, and (18). Cyclin D1 protein stability is regulated by glycogen synthase cells used in GFP experiments were fixed with 0.5% paraformaldehyde for 10 minutes. These cells were washed with PBS and then fixed with 75% ethanol kinase 3h (GSK3h) and p38, which phosphorylate the protein, overnight at 4jC. Following this, the cells were then washed twice with PBS triggering ubiquitination (19, 20). Cyclin D1 mRNA translation is and resuspended in hypotonic propidium iodide solution (10 Ag propidium often called the critical mode of cyclin D1 regulation by rapamycin. iodide, 10 Ag RNase A, 0.5% Tween 20 in 1 mL PBS) for 0.5 hour at 37jC and In recent work, Gera et al. have shown that rapamycin leads to a kept in the dark at 4jC before analysis. Cell cycle distribution was 5-fold decrease in cyclin D1 translation in cell lines that have high determined with a FACScan flow cytometer and Cell Quest software levels of Akt activity (3). However, in NIH3T3 cells, rapamycin was (Becton Dickinson, San Jose, CA). also found to decrease cyclin D1 mRNA and protein stability Western Blot Analysis. Cultured cells were washed with cold PBS and through unknown mechanisms (21). Therefore, the mechanism by lysed in radioimmunoprecipitation buffer [20 mmol/L Tris (pH 7.5), 150 which rapamycin down-regulates cyclin D1 expression needs mmol/L NaCl, 5 mmol/L EDTA, 1% NP40, 1 mmol/L Na3VO4, 1 mmol/L further study. phenylmethylsulfonyl fluoride, 50 mmol/L NaF] supplemented with complete protease inhibitors on ice. To extract cytosolic proteins for In our previous work, we found that rapamycin leads to down- detection of cytochrome c release, 4 Â 106 cells were harvested and regulation of cyclin D1 levels in rapamycin-sensitive breast cancer washed twice with ice-cold PBS and resuspended in 300 AL ice-cold buffer cell lines MCF-7 and MDA-MB-468 but not in rapamycin-resistant [20 mmol/L HEPES-KOH (pH 7.0), 10 mmol/L KCl, 1.5 mmol/L MgCl2, cell lines MDA-MB-231, MDA-MB-435, and NCI/ADR-RES, suggesting 1 mmol/L EDTA, 1 mmol/L EGTA, 1 mmol/L DTT, 250 mmol/L sucrose, that rapamycin-mediated cyclin D1 down-regulation may be 1 Ag/mL leupeptin and pepstatin, 2 Ag/mL aprotinin]. After incubation on critical to its growth-inhibitory effect (22). In this study, we found ice for 15 minutes, cells were homogenized with a Dounce homogenizer that cyclin D1 plays an important role in rapamycin-mediated cell (B pestle per 25 strokes) and centrifuged at 1,000 Â g for 10 minutes. The cycle arrest and growth inhibition. We show that rapamycin supernatants were centrifuged at 14,000 Â g for 15 minutes in a mediates cyclin D1 down-regulation in part by accelerated microcentrifuge to pellet membranes, including mitochondria. The proteolysis. Furthermore, we show that rapamycin activates resulting supernatants were used as cytosolic extracts. Cell lysates or cytosolic extracts were separated by SDS-PAGE and transferred to a 0.2 Am GSK3h and that GSK3h activity is critical for rapamycin-mediated h polyvinylidene difluoride membrane (Bio-Rad Laboratories, Hercules, CA). cyclin D1 down-regulation. We show that although GSK3 is not Membranes were blocked with 5% nonfat dry milk in TBST (TBS with essential for rapamycin-mediated growth inhibition, it significantly 0.1% Tween 20) and immunoblotted with antibodies. The immunoblots enhances the growth-inhibitory effect of rapamycin. Further, were visualized by an enhanced chemiluminescence detection system rapamycin-mediated chemosensitization is also attenuated in (Amersham Life Sciences, Arlington Heights, IL). GSK3h-null cells compared with GSK3h wild-type cells. Thus, we Kinase Assay. Cells treated with rapamycin (100 nmol/L) for 0, 3, or 6 show that GSK3h plays a significant role in the antitumor effects of hours or with the PI3K inhibitor LY294004 (10 or 20 Amol/L) for 1 hour rapamycin. were lysed in the NP40 immunoprecipitation lysis buffer (KPL, Gaithersburg, MD). Cell lysates (200 Ag protein) were precleared for 1 hour at 4jC with 50 AL (50%) protein G-agarose beads (KPL). The Materials and Methods precleared protein lysates were incubated with 3 AL rabbit anti-GSK3h Cell Cultures. Human breast cancer cell lines MCF-7 and MDA-MB-468 antibody (Cell Signaling Technology) overnight at 4jC with gentle were obtained from the American Type Culture Collection (Manassas, VA). agitation. The immunoprecipitates were incubated with 50 AL (50%) GSK3h+/+ and GSK3hÀ/À mouse embryo fibroblasts have been described protein G-agarose for 2 hours at 4jC with gentle agitation. The previously (23, 24). Cells were cultured in DMEM/F-12 supplemented with immobilized immunocomplexes were washed once with the NP40 lysis 10% fetal bovine serum, 2 mmol/L glutamine, and 1% penicillin- buffer (500 AL) and twice with kinase buffer (500 AL) [20 mmol/L Tris j streptomycin at 37 C and humidified 5% CO2. (pH 7.5), 5 mmol/L MgCl2, 1 mmol/L DTT]. The immunoprecipitates Materials. Rapamycin and cycloheximide was purchased from A.G. were equally aliquoted into two tubes for the Western blotting and the Scientific, Inc. (San Diego, CA). N-acetyl-leucyl-leucyl-norleucinal (ALLN); kinase assay, respectively. GSK3h inhibitors lithium chloride (LiCl), SB216763, and SB415286; and In vitro kinase assays were carried out by mixing the beads with 30 AL antibodies against h-actin were purchased from Sigma Chemical Co. (St. kinase buffer [20 mmol/L Tris (pH 7.5), 5 mmol/L MgCl2, 1 mmol/L DTT, Louis, MO). Antibodies against cyclin D1, p21, p27, Rb, phospho-Rb (Ser780), ATP (Sigma Chemical) 250 Amol/L, 1.4 ACi [g-32P]ATP (MP Biomedicals, and phospho-GSK3h (Ser9), cytochrome c, caspase-9, caspase-3, caspase-7, Inc., Irvine, CA), 0.1 Ag/AL recombinant H protein (Panvera, Madison, WI)] 112 j poly(ADP-ribose)polymerase (PARP), Bad, Bad (Ser ), Bak, Bax, Bcl-xL, and subsequently incubated at 30 C for 20 minutes. The reaction was Bcl-2 (Ser70), survivin, and X-linked inhibitor of apoptosis were purchased stopped by adding 6 ALof6Â Laemmli sample buffer and denaturing at from Cell Signaling Technology (Beverly, MA). Antibodies against cyclin A, 98jC for 5 minutes. Proteins were fractionated in 10% SDS-PAGE gels, cyclin E, GSK3h, and green fluorescent protein (GFP) were purchased from vacuum dried, and exposed to a phosphoscreen. The phosphorimage was BD Biosciences PharMingen (San Jose, CA). Effectene transfection reagent captured using PhosphorImager (Molecular Dynamics, Sunnyvale, CA) and was purchased from Qiagen, Inc. (Valencia, CA). Phospho-GSK3 Y216 followed by densitometry using Scion Image (Scion, Inc., Frederick, MD). antibody was purchased from Upstate Biotechnology (Lake Placid, NY). The Relative kinase activities were based on densities of the untreated samples. GFP-cyclin D1 plasmid was a kind gift of Dr. Li-Huo Su (previously at The efficiency of GSK3h immunoprecipitation was determined by University of Texas M.D. Anderson Cancer Center, currently at Cancer Cell). immunodetection for GSK3h.A30ALof1Â Laemmli sample buffer was The pcDNA3.HA-tagged cyclin D1 and the pcDNA3.T286A cyclin D1 mixed with the beads and denatured at 98jC for 5 minutes. Proteins were constructs were kind gifts of Dr. Doris Germain (University of Melbourne, fractionated in 10% SDS-PAGE, transferred to Immunoblot polyvinylidene Melbourne, Victoria, Australia). difluoride membrane (Bio-Rad Laboratories), and underwent immunode- Colony Formation Assay. Transfected cells were washed, trypsinized tection using the anti-GSK3h antisera at 1:2,000. and, resuspended in DMEM. Cells were counted and plated in 100 mm Cell Growth Assays. For 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetra- dishes in triplicate, and cultured in medium containing 500 Ag/mL G418 zolium bromide assays, cells were plated into 96-well, flat-bottomed plates with or without 100 nmol/L rapamycin. The medium was changed twice a at 2 Â 103 to 4 Â 103 cells/100 AL/well, and following the overnight week for 3 weeks, and then colonies were stained with the crystal violet. incubation, triplicate wells were treated with varying concentration of

Cancer Res 2005; 65: (5). March 1, 2005 1962 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 2005 American Association for Cancer Research. Role of GSK3bb in Rapamycin-Mediated Antitumor Effects rapamycin for 4 days. Relative percentage of metabolically active cells Trypan Blue Exclusion Assay. A trypan blue exclusion study was done relative to untreated controls were then determined based on the after treating cells with the signal transduction inhibitors. After 48 hours, mitochondrial conversion of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetra- cells were trypsinized and washed in PBS. Cells were mixed thoroughly and zolium bromide to formazine. The results were assessed in a 96-well cell suspension (20 AL) was added to 20 AL of 0.08% trypan blue dye format plate reader by measuring the absorbance at a wavelength of solution. At least 500 cells were microscopically counted in the hemocytometer. Cell death was expressed as the percentage of cells 540 nm (A540 nm). The rates of DNA synthesis were determined by the 3 staining blue. percentage of cells showing [ H]thymidine incorporation into DNA. In Statistical Analysis. Experiments were independently done three or brief, after the cells were treated with rapamycin in the same manner as in more times. Two-tailed Student’s t test or linear regression analysis was the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay, 0.5 used to analyze growth inhibition assays. P < 0.05 was considered A 3 Ci [ H]thymidine was added to each well and the cells were incubated for statistically significant. an additional 16 hours before being harvested. The incorporation of [3H]thymidine was measured by liquid scintillation counting. The rates of DNA synthesis in the treated cells were compared with the rates seen for Results control cells not treated with rapamycin in the same culture plate. Direct cell counts were done by seeding 10,000 cells/well and after 24 hours Cyclin D1 Overexpression Partially Overcomes Rapamycin- culturing the cells in the absence or presence of rapamycin. At the Mediated Cell Cycle Arrest and Growth Inhibition. In previous specified intervals, the cells were trypsinized and the viable cells were work, we found that rapamycin-sensitive cell lines treated with counted with a hemocytometer. 100 nmol/L rapamycin for 4 days showed a decline in cyclin D1 Annexin V Labeling. Apoptosis was determined using the ApoAlert levels, whereas cell lines resistant to the growth-inhibitory effects Annexin V apoptosis kit (Clontech, Palo Alto, CA) according to the of rapamycin did not (22). As rapamycin is known to mediate cell manufacturer’s protocol. In brief, cells were trypsinized and rinsed in cycle arrest as one of its early effects, we evaluated the time course binding buffer. They were then resuspended in 200 AL binding buffer to of rapamycin-mediated cyclin D1 down-regulation in two rapamycin- which 5 AL Annexin V (20 Ag/mL in Tris-NaCl buffer) and 10 AL propidium iodide (1 Ag/mL) were added. Cells were incubated for 10 to sensitive cell lines, MCF-7 and MDA-MB-468. We found that cyclin 15 minutes at ambient temperature in the dark and then analyzed by D1 levels started to decrease by 6 hours of rapamycin treatment flow cytometry. (Fig. 1A). Rapamycin treatment also led to a decrease in Rb

Figure 1. Role of cyclin D1 in rapamycin-mediated growth inhibition. A, MCF-7 and MDA-MB-468 cells were incted with rapamycin (100 nmol/L) for 6, 12, or 24 hours, and Western blot analysis was done to assess cyclin D1 and actin levels. B, MCF-7 cells were cultured in the absence or presence of 100 nmol/L rapamycin for 24 hours. Western blot analysis was done with phospho-Rb (Ser780) and total Rb antibodies. C, modulation of cell cycle and expression of cell cycle regulatory proteins by continued rapamycin treatment. MCF-7 and MDA-MB-468 cells were incubated with rapamycin (100 nmol/L) for 1, 2, or 4 days; cell cycle distribution was evaluated with flow cytometry; and expression of cyclin D1, p21, p27, cyclin A, cyclin E, and actin was determined by Western blot analysis. D, effect of cyclin D1 overexpression on rapamycin-mediated G1 arrest. MCF-7 cells were transfected with GFP or GFP-cyclin D1 and incubated with or without rapamycin (100 nmol/L) for 2 days. Cells were than stained with propidium iodide, and cell cycle distribution was determined by flow cytometry. Cell cycle distribution of all cells (total cells, right) was determined as an additional control and compared with the cell cycle distribution of the GFP-positive cells (left), representing the cells transfected with either GFP or GFP-cyclin D1 constructs. E, effect of cyclin D1 overexpression on rapamycin-mediated growth inhibition. MCF-7 and MDA-MB-468 cells transfected with GFP and GFP-cyclin D1, incubated with or without rapamycin (100 nmol/L) under G418 selection for 3 weeks, and then stained with crystal violet. Colonies of >120 Amin diameter were counted using a microscope. Columns, mean; bars, SD. **, P < 0.05 GFP versus GFP-cyclin D1–transfected, rapamycin-treated cells. www.aacrjournals.org 1963 Cancer Res 2005; 65: (5). March 1, 2005

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 2005 American Association for Cancer Research. Cancer Research phosphorylation (Fig. 1B). We then evaluated the effects of mRNA stability, respectively. MCF-7 cells were transfected with rapamycin on a panel of cell cycle regulatory molecules. MCF-7 GFP-cyclin D1, GFP alone, or both plasmids together. Cells were and MDA-MB-468 cells were maintained in 100 nmol/L rapamycin, cultured in the absence or presence of 100 nmol/L rapamycin, and and cell cycle analysis was done with flow cytometry after different cyclin D1 levels were evaluated with Western blotting (Fig. 2A). durations of treatment (Fig. 1C). In both cell lines, the population Rapamycin treatment did not affect exogenous GFP levels. In of cells in G1 phase increased, whereas the population of cells in contrast, rapamycin reduced levels of both exogenous and the S and G2 phases decreased after treatment with rapamycin endogenous cyclin D1 protein. Similarly, we observed that (Fig. 1C). The expression of cyclins D1, A, and E and cell cycle rapamycin decreased the expression of hemagglutinin-tagged cyclin inhibitors p21 and p27 were evaluated with Western blot analysis at D1 transfected in MDA-MB-468 cells (data not shown). Quantifica- the same time points. Cyclin D1 levels were found to decrease in tion of the effects in MCF-7 cells (Fig. 2A, right) shows that both cell lines with rapamycin treatment, with variable changes in rapamycin reduced endogenous cyclin D1 levels more than cyclin A levels and no change in cyclin E levels. Rapamycin did not exogenous cyclin D1 levels (70% versus 38%). The decrease in increase levels of cell cycle inhibitors p21 and p27 in MCF-7 and expression of exogenous cyclin D1 protein in this experiment MDA-MB-468 cells; rather, their levels were decreased within 3 days suggests that rapamycin-mediated cyclin D1 down-regulation may of treatment. Based on these data, we conclude that rapamycin be at least in part regulated at the level of protein stability. However, induces G1 cell cycle arrest, and this is associated with a decline in the fact that the expression of endogenous cyclin D1 was decreased cyclin D1 levels. more than that of the exogenous cyclin D1, which lacked sequence To determine the role of cyclin D1 down-regulation in the effects elements regulating transcriptional, translational, and mRNA of rapamycin on breast cancer cells, we tested whether the forced stability, suggests that in addition to protein stability other overexpression of cyclin D1 is able to rescue cells from rapamycin- mechanisms are likely at play in rapamycin-mediated down- mediated cell cycle arrest and growth inhibition. We transfected regulation of the endogenous cyclin D1 gene. MCF-7 cells with either a plasmid encoding for a GFP-tagged cyclin To determine whether rapamycin affects cyclin D1 protein D1 protein (GFP-cyclin D1) or a plasmid encoding for GFP only. stability, we evaluated the effect of rapamycin on the half-life of Cells were incubated in the absence or presence of 100 nmol/L cyclin D1 protein. MCF-7 and MDA-MB-468 cells were pretreated rapamycin; then, cell cycle analysis was done on the GFP-positive with rapamycin and then treated with an inhibitor of protein cells (Fig. 1D, GFP-positive cells, left) and on the entire cell synthesis, 100 Ag/mL cycloheximide, for different time intervals. population as a control (Fig. 1D, total cells, right). Rapamycin The level of cyclin D1 protein in rapamycin-treated MCF-7 cells induced G1 arrest in the cells transfected with GFP protein and in (Fig. 2B) and MDA-MB-468 cells (Fig. 2C) declined faster over time the MCF-7 cell population. In comparison, the effect of rapamycin than it did in rapamycin-treated cells. In the average of three on the cell cycle was less prominent in the cells transfected with independent experiments, the half-life of cyclin D1 protein in GFP-cyclin D1 (Fig. 1D, GFP-positive cells, left). untreated and rapamycin-treated MCF-7 cells was found to be 48 To examine the effect of cyclin D1 on the inhibition of cell and 27 minutes, respectively, demonstrating a 44% decrease in growth and proliferation mediated by rapamycin, we determined cyclin D1 half-life on rapamycin treatment. The half-life of cyclin the effect of cyclin D1 overexpression on colony formation ability in D1 protein in untreated and rapamycin-treated MDA-MB-468 cells the absence or presence of rapamycin. We transfected MCF-7 and was 44 and 36 minutes, respectively, demonstrating a more subtle, MDA-MB-468 cells with either GFP control or GFP-cyclin D1. Cells but reproducible, decrease in the cyclin D1 half-life on rapamycin were incubated with or without rapamycin for 3 weeks under G418 treatment in three independent experiments. These results selection. Results in Fig. 1E show that rapamycin leads to a illustrate that rapamycin decreases the half-life of cyclin D1 significant decrease in colony formation in both cell lines. protein and suggest that the degradation of cyclin D1 protein Rapamycin treatment led to a 13.6-fold reduction in colony induced by rapamycin contributes to the decline of cyclin D1 levels formation in control MCF-7 cells and a 4-fold reduction in cyclin after rapamycin treatment. D1–transfected MCF-7 cells. Rapamycin inhibited colony formation To further show the role of proteolysis in rapamycin-mediated by the control cells 3.4-fold more than by cyclin D1–transfected cyclin D1 down-regulation, we carried out studies with the cells (P < 0.05). Further, rapamycin led to a 3.7-fold reduction in proteasome inhibitor ALLN. We incubated cells without inhibitors, colony formation in control MDA-MB-468 cells and a 1.7-fold with 100 nmol/L rapamycin alone, with ALLN alone, or with ALLN reduction in cyclin D1–transfected MDA-MB-468 cells. Rapamycin and rapamycin for 12 hours. As shown in Fig. 2D and E, ALLN inhibited colony formation by control cells 2.2-fold more than by inhibited the rapamycin-induced decrease of cyclin D1 levels. cyclin D1–transfected cells (P < 0.05). Taken together, these results These results taken together show that rapamycin reduces the show that cyclin D1 overexpression partially overcomes rapamycin- cyclin D1 protein levels at least in part by accelerating proteasome- induced cell cycle arrest and growth inhibition. This suggests that dependent proteolysis. the rapamycin-mediated cyclin D1 down-regulation observed plays Rapamycin Activates GSK3B. Because cyclin D1 is known to a critical role in the growth-inhibitory effects of rapamycin. undergo proteasome-dependent proteolysis in a GSK3h-dependent Rapamycin Accelerates the Proteolysis of Cyclin D1. To better fashion (19), we evaluated the effect of rapamycin treatment on delineate the mechanism of rapamycin-mediated cyclin D1 down- GSK3h kinase activity. The kinase activity of GSK3h was evaluated regulation, we tested the effect of rapamycin treatment on in MCF-7 cells at baseline and after incubation with 100 nmol/L exogenously overexpressed cyclin D1 compared with endogenous rapamycin (Fig. 3A). Rapamycin treatment was found to enhance cyclin D1. The transcription of the exogenous GFP-cyclin D1 GSK3h kinase activity by 110% and 65% at 3 and 6 hours after construct is driven by a cytomegalovirus promoter; thus, it should treatment, respectively. Next, GSK3h kinase activity was deter- not be under cyclin D1–specific transcriptional regulation. Further, mined in MDA-MB-468 cells at baseline and after incubation with the construct lacks the cyclin D1 5Vand 3Vuntranslated regions that 100 nmol/L rapamycin. GSK3h kinase activity increased by 136% are proposed to be critical for cyclin D1 translational regulation and and 107% after 3 and 6 hours of rapamycin treatment, respectively.

Cancer Res 2005; 65: (5). March 1, 2005 1964 www.aacrjournals.org

Figure 2. Effects of rapamycin on endogenous and exogenous cyclin D1. A, MCF-7 cells were transfected with GFP alone, GFP-cyclin D1, or both GFP and GFP-cyclin D1 and incubated with or without rapamycin (100 nmol/L) for 2 days. Cell lysates were separated by 12% SDS-PAGE, and immunoblot analysis was done using antibodies against cyclin D1, GFP, and actin. Relative protein levels of endogenous cyclin D1, GFP-cyclin D1, and GFP in the presence and absence of rapamycin, as shown in the Western blots, were quantified by densitometry and normalized according to the level of actin control. Protein expression in rapamycin-untreated cells was depicted as 100%. B, effect of rapamycin on cyclin D1 protein half-life. MCF-7 cells were cultured in the absence or presence of 100 nmol/L rapamycin for 2 hours. Cells were then treated with 100 Ag/mL cycloheximide (CHX) and harvested at indicated time intervals. Western blot analysis was done with anti-cyclin D1 and anti-actin antibodies. Bottom, quantitation of cyclin D1 expression, normalized to the level of actin control, as an average of three independent experiments. Cyclin D1 expression at the 0-hour time point of both untreated and treated cells was set as 100%. C, MDA-MB-468 cells were cultured in the absence or presence of 100 nmol/L rapamycin for 4 hours and then treated with 100 Ag/mL cycloheximide. Western blot analysis and quantitation was done as in B. D, effect of proteasome inhibitor ALLN on rapamycin-mediated cyclin D1 down-regulation. MCF-7 cells were incubated with or without 100 Amol/L ALLN in the absence or presence of 100 nmol/L rapamycin. Western blot analysis was done using an anti-cyclin D1 antibody, with longer exposure and shorter exposure of the same membrane shown. Western blotting with anti-actin antibody used as a loading control. E, MDA-MB-468 cells were incubated with or without 100 Amol/L ALLN in the absence or presence of 100 nmol/L rapamycin for 12 hours, and Western blot analysis was done using an anti-cyclin D1 and anti-actin antibody.

Alternatively, as GSK3h is inhibited by Akt (25), the cells were It is known that Ser9 phosphorylation of GSK3h decreases incubated with PI3K inhibitor LY294002 (Fig. 3B). Treatment with GSK3h activity, whereas Tyr216 phosphorylation is required for 10 or 20 Amol/L LY294002 increased GSK3h kinase activity by 73% GSK3h activity (26, 27). Interestingly, although activation of GSK3h and 163%, respectively. In three independent experiments, by LY294002 was associated with Ser9 dephosphorylation as rapamycin induced a statistically significant increase in GSK3h expected, rapamycin-mediated activation was not (Fig. 3B). kinase activity in MCF-7 and MDA-MB-468 cells (P < 0.001; Fig. 3C). Further, no significant alteration in phosphorylation of GSK3h Of note, treatment with the DMSO vehicle alone was not Tyr216 was observed in either cell line with rapamycin treatment associated with GSK3h activation or cyclin D1 down-regulation (Fig. 3D). Our results suggest that rapamycin activates GSK3h (data not shown). through an as yet unknown mechanism. www.aacrjournals.org 1965 Cancer Res 2005; 65: (5). March 1, 2005

Figure 3. Regulation of GSK3h kinase activity by rapamycin. MCF-7 cells (A) and MDA-MB-468 cells (B) were treated with rapamycin or LY294002 for the durations indicated. Cells were lysed and immunoprecipitated with anti-GSK3h antibody. In vitro kinase activity was measured by the GSK3h immunoprecipitate–catalyzed [g-32P]ATP incorporation into recombinant H protein. Band densities were quantified. Cell lysates from the same experiments were separated by SDS-PAGE, and Western blot analysis was done using antibodies to phospho-GSK3h (Ser9), total GSK3h antibody, and actin antibody. C, effect of rapamycin on GSK3h kinase activity in MCF-7 and MDA-MB-468 cells in three independent experiments. Columns, mean (untreated cells normalized to 100% kinase activity); bars, SE. **, P < 0.001, difference between rapamycin-treated and untreated cells. D, effect of rapamycin on GSK3h Ser9 and Y216 phosphorylation. MDA-MB-468 and MCF-7 cells were treated with 100 nmol/L rapamycin for 12 or 24 hours. Phospho-GSK3h (Y216), phospho-GSK3h (Ser9), total GSK3h, and actin protein levels were determined by Western blot analysis.

GSK3B Inhibitors LiCl, SB216763, and SB415286 Inhibit GSK3B Wild-type Cells Are More Sensitive to Rapamycin- Rapamycin-Mediated Down-regulation of Cyclin D1. To inves- Mediated Growth Inhibition Than GSK3B-Null Cells. As GSK3h tigate whether GSK3h is involved in the rapamycin-induced contributes to rapamycin-mediated cyclin D1 down-regulation, degradation of cyclin D1, we examined the effect of LiCl, a known we hypothesized that GSK3h plays a significant role in inhibitor of GSK3h, on rapamycin-induced cyclin D1 down- rapamycin-mediated cell cycle arrest and growth inhibition. We regulation (28). We incubated cells with or without 100 nmol/L first tested the effect of GSK3h inhibitors LiCl and SB216763 on rapamycin in the presence or absence of 20 mmol/L LiCl. As rapamycin-mediated cell cycle arrest in MCF-7 cells. In three shown in Fig. 4A, LiCl almost completely inhibited the decrease of independent experiments, we found that coincubation with 20 cyclin D1 levels induced by rapamycin. We then tested the effect mmol/L LiCl inhibited the G1 arrest induced after 24 hours of of two specific inhibitors of GSK3h, SB216763 and SB415286, at 30 rapamycin treatment (Fig. 5A). SB216763 has been shown and 10 Amol/L, respectively (29). As shown in Fig. 4B, SB216763 previously to inhibit GSK3h activity by 96% at 10 Amol/L (29); and SB415286 both inhibited the rapamycin-induced decline in however, SB216763 was not able to inhibit rapamycin-mediated cyclin D1 levels. These results suggest that rapamycin treatment cell cycle arrest at 30 Amol/L (Fig. 5B) or at 10, 50, and activates GSK3h, which induces proteolysis of the cyclin D1 100 Amol/L (data not shown). These data suggest that although protein. GSK3h activity plays a role in rapamycin-mediated cyclin D1 Cyclin D1 T286A Mutant Is Resistant to Rapamycin-Induced down-regulation, GSK3h is not essential for rapamycin-mediated Down-regulation. Because cyclin D1 proteolysis is thought to be cell cycle arrest. These data also suggest that LiCl may be mediated by phosphorylation of its Thr286 residue by GSK3h (21), inhibiting targets in addition to GSK3h to abrogate the effect of we evaluated whether a T286A cyclin D1 mutant that is resistant to rapamycin on the cell cycle. GSK3h-induced phosphorylation and degradation is also resistant We then tested the role of GSK3h in rapamycin-mediated cell to rapamycin. MDA-MB-468 cells were transiently transfected with growth by comparing the growth-inhibitory effects of rapamycin in pcDNA3 vector control or the same vector encoding for the T286A GSK3h wild-type (GSK3h+/+) and GSK3h-null (GSK3hÀ/À) cyclin D1 mutant. Twenty-four hours later, the transfected cells fibroblasts (Fig. 5C; ref. 24). We first evaluated the effect of were incubated in the absence or presence of 100 nmol/L rapamycin on cell proliferation by direct cell counts (Fig. 5D). rapamycin for another 48 hours. Western blot analysis showed a GSK3h wild-type and GSK3h-null cells were cultured in the significant reduction in cyclin D1 levels in the control vector- absence or presence of 10 or 100 nmol/L rapamycin on triplicate transfected cells on rapamycin treatment, with a much less plates. Three days after rapamycin treatment, compared with dramatic reduction in cyclin D1 levels in the T286A cyclin D1– untreated GSK3h wild-type cells, a significant decline was seen in transfected cells (Fig. 4C). These results suggest that GSK3h- the number of GSK3h wild-type cells treated with 10 nmol/L induced T286 phosphorylation is critical in rapamycin-mediated rapamycin (10.7 Â 104 versus 2.4 Â 104; P = 0.0009) and 100 nmol/L cyclin D1 down-regulation. rapamycin (10.7 Â 104 versus 1.8 Â 104; P = 0.0006). However, there

Cancer Res 2005; 65: (5). March 1, 2005 1966 www.aacrjournals.org

caused a greater inhibition in GSK3h wild-type cells compared with GSK3h-null cells (P < 0.0001 for all). In addition, the data show that the effect of GSK3h on rapamycin-mediated growth inhibition increases as the rapamycin dose level increases (P < 0.0001). We also determined the effect of rapamycin on GSK3h wild-type and GSK3h-null cell proliferation by culturing cells in the absence or presence of rapamycin for 4 days and then measuring DNA synthesis by [3H]thymidine incorporation (Fig. 5F). Rapamycin treatment decreased [3H]thymidine incorporation in both GSK3h wild-type and GSK3h-null cells (P < 0.05). However, treatment with 10 or 100 nmol/L rapamycin led to a significantly greater decline in DNA synthesis in GSK3h wild-type cells compared with GSK3h-null cells (P = 0.0011 and 0.0035, respectively). These results were reproducible in three independent experiments. We further tested the effect of GSK3h on rapamycin on anchorage-dependent growth in the absence and presence of rapamycin long-term. GSK3h wild-type and GSK3h-null cells were trypsinized and plated in 100 mm Petri dishes (1 Â 104 cells per dish), and after cell adherence (24 hours), cells were cultured in the absence or presence of 10 nmol/L rapamycin. Three weeks later, the cells were stained with crystal violet. On crystal violet assay, rapamycin completely inhibited anchorage-dependent growth in GSK3h wild-type cells but not in GSK3h-null cells (Fig. 5G). These results were reproducible in two independent experiments. These data taken together show that GSK3h activity is not essential for rapamycin-mediated cell cycle arrest; however, GSK3h-mediated mechanisms significantly enhance rapamycin-mediated growth inhibition. Rapamycin Enhances Paclitaxel-Mediated Apoptosis in a GSK3B-Dependent Fashion. As GSK3h has been shown previously to be critical in regulation of cell survival (30), we next investigated Figure 4. GSK3h as a mediator of rapamycin-induced cyclin D1 whether GSK3h plays a role in rapamycin-mediated chemo- down-regulation. A, inhibition of rapamycin-induced cyclin D1 own-regulation by sensitivity. In our previous work, we observed additive interactions LiCl. Cells were incubated in the absence or presence of 100 nmol/L rapamycin for 12 hours with or without LiCl (20 mmol/L). Immunoblotting was done using when rapamycin was given synchronously with doxorubicin or cyclin D1 and actin antibodies. Quantitation of cyclin D1 levels with the untreated gemcitabine, and synergistic interactions when rapamycin was cells normalized to 100% for each group.B,inhibition of rapamycin-induced given synchronously with carboplatin, vinorelbine, or paclitaxel cyclin D1 down-regulation by GSK3h inhibitors SB216763 and SB415286. Cells were incubated in the absence or presence of 100 nmol/L rapamycin for 12 hours (31). As we observed the greatest synergy between rapamycin and with or without SB216763 (30 Amol/L) or SB415286 (10 Amol/L). Immunoblotting paclitaxel, we explored the molecular mechanism behind this was done using cyclin D1 and actin antibodies. Quantitation of cyclin D1 levels finding. MDA-MB-468 cells were treated with 0.01 Ag/mL paclitaxel, with the untreated cells normalized to 100% for each group. C, effect of rapamycin on T286A cyclin D1 mutant. MDA-MB-468 cells were transfected with 100 nmol/L rapamycin, or with the combination, and cells were a pcDNA3 vector control or pcDNA3-T286A cyclin D1. After 24 hours, the cells harvested after 48 hours to determine apoptosis as determined by were incubated in the absence or presence of 100 nmol/L rapamycin for another 48 hours. Immunoblotting was done using cyclin D1 and actin antibodies. Annexin V labeling. The level of Annexin V labeling was much higher after treatment with the combination of rapamycin and paclitaxel than after treatment with each agent alone (Fig. 6A). was no statistically significant decline in the number GSK3h-null After 24 hours, cells treated with the combination of rapamycin cells treated with rapamycin 3 days after treatment. By 6 days of and paclitaxel, but not with either agent alone, showed cytoplasmic rapamycin treatment, there was a significant decrease in the cytochrome c release and cleavage of caspase-9, caspase-3, number GSK3h-null cells treated with rapamycin compared with caspase-7, and PARP (Fig. 6B). This suggests that the cytotoxicity untreated GSK3h-null cells, but there was a significantly greater of the combination of rapamycin and paclitaxel is mediated inhibition in direct cell counts in GSK3h wild-type cells compared through the mitochondrial death pathway. As rapamycin has been with GSK3h-null cells on treatment with both 10 nmol/L reported to decrease the level of the antiapoptotic protein survivin rapamycin (P = 0.0014) and 100 nmol/L rapamycin (P = 0.0010). (32, 33), we hypothesized that rapamycin-induced changes in These results were reproducible in three independent experiments. expression of apoptotic and proapoptotic molecules may play a Next, GSK3h wild-type and GSK3h-null cells were treated with role in rapamycin-mediated chemosensitivity. We therefore deter- increasing doses of rapamycin (1-1,000 nmol/L), and the metabol- mined the expression of a panel of proapoptotic and antiapoptotic ically active cells were determined based on the mitochondrial molecules after 24 hours of treatment with paclitaxel, rapamycin, conversion of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium or the combination (Fig. 6C). We found no dramatic changes in the 112 bromide to formazine, assessed 4 days after treatment (Fig. 5E). As levels of Bad, phospho-Bad (Ser ), Bak, Bax, Bcl-xL, Bcl-2, Bcl-2 assessed by linear regression analysis, rapamycin led to a dose- (Ser70), survivin, or X-linked inhibitor of apoptosis with the dependent growth inhibition in both cell lines (P < 0.0001). combination therapy. Thus, our data show that rapamycin However, at all four rapamycin concentrations tested, rapamycin enhances paclitaxel-induced cytotoxicity at least in part through www.aacrjournals.org 1967 Cancer Res 2005; 65: (5). March 1, 2005

Figure 5. GSK3h as a mediator of rapamycin-induced growth inhibition. A, MCF-7 cells were cultured in the presence or absence of 100 nmol/L rapamycin with or without 20 mmol/L LiCl. After 24 hours, FACS analysis was done to determine the percentage of total cells in G1 and S phases in each treatment group. Columns, mean of three independent experiments; bars, SD. B, MCF-7 cells were cultured in the presence or absence of 100 nmol/L rapamycin with or without 30 Amol/L SB216763 (SB21). After 24 hours, FACS analysis was done. Columns, mean percentage of cells in G1 and S phases; bars, SD. C, Western blot analysis of GSK3h wild-type (GSK3h+/+) and GSK3h-null (GSK3hÀ/À) cells with GSK3h and actin antibodies. D, direct cell counts were done by seeding GSK3h wild-type (GSK3h+/+) and GSK3h-null (GSK3hÀ/À) cells (1 Â 104 per well) in triplicate and after 24 hours culturing the cells in the absence or presence of 10 or 100 nmol/L rapamycin. Cells were trypsinized at 1, 3, and 6 days after rapamycin treatment, and the viable cells were counted with a hemocytometer. Points, mean cell counts results; bars, SD. **, P < 0.05, difference in rapamycin-treated and untreated cells. Representative of three independent experiments. E, GSK3h wild-type and GSK3h-null cells were grown in triplicate in the presence and absence of rapamycin (1, 10, 100, and 1,000 nmol/L). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay was done at 96 hours. Columns, mean metabolically active cells expressed in arbitrary units; bars, SD. Representative of three independent experiments. **, P < 0.0001, difference in rapamycin-induced growth inhibition in GSK3h wild-type and GSK3h-null cells. F, GSK3h wild-type and GSK3h-null cells were cultured cells in the absence or presence of rapamycin (1, 10, and 100 nmol/L) for 4 days, and DNA synthesis was measured by [3H]thymidine incorporation. Representative of three independent experiments. Columns, mean; bars, SD. **, P < 0.0005, difference in rapamycin-induced inhibition of [3H]thymidine incorporation in GSK3h wild-type and GSK3h-null cells. G, GSK3h wild-type and GSK3h-null cells were trypsinized and plated in 100 mm Petri dishes (104 cells per dish) and after cell adherence cultured in the absence or presence of 10 nmol/L rapamycin. Three weeks later, the cells were stained with crystal violet.

Cancer Res 2005; 65: (5). March 1, 2005 1968 www.aacrjournals.org

Figure 6. GSK3h as a mediator of rapamycin-induced paclitaxel sensitization. A, MDA-MB-468 cells were treated with 0.01 Ag/mL paclitaxel alone (P), 100 nmol/L rapamycin alone (R), and combination (P/R). After 24 hours, Annexin V–positive cells were determined by FACS analysis. B, MDA-MB-468 cells were treated for 24 hours. Fifty micrograms of cytoplasmic extract (top)or50Ag of total protein from each treatment group or control cells (No Rx) were separated by SDS-PAGE. Antibodies against cytoplasmic cytochrome c (top), caspase-9, caspase-3, caspase-7, and actin were used for Western blot analysis. C, expression of proapoptotic and antiapoptotic molecules. MDA-MB-468 cells were treated with as above for 24 hours. Western blot analysis was done with antibodies against Bad, Bad (Ser112), Bak, 70 Bax, Bcl-xL, Bcl-2 (Ser ), survivin, X-linked inhibitor of apoptosis, and actin. D, MDA-MB-468 cells were treated with 100 nmol/L rapamycin alone, 0.01 Ag/mL paclitaxel alone, and combination. LiCl (10 mmol/L), SB216763 (20 Amol/L), or SB415286 (10 Amol/L) was used in combination with rapamycin and paclitaxel. After 24 hours, intact and cleaved PARP products and actin were detected by Western blot analysis. E, following the same treatment schedule, Annexin V–positive cells were determined by FACS analysis. F, trypan blue dye exclusion assay was done after treating cells as above or after treating cells with 10 mmol/L LiCl, 20 Amol/L SB216763 (SB21), or 10 Amol/L SB415286 (SB41) alone for 24 hours. Columns, mean of three independent experiments. G, GSK3h wild-type (GSK3h+/+) and GSK3h-null (GSK3hÀ/À) cells were treated for 24 hours with 100 nmol/L rapamycin alone, 0.01 Ag/mL paclitaxel alone, and combination. Annexin V–positive cells were determined by FACS analysis. **, P < 0.05, difference between P and P/R cells. H, GSK3h wild-type and GSK3h-null cells were treated for 24 hours as above. Western blot analysis was done with antibodies against cytochrome c (on cytoplasmic extract) and caspase-9, PARP, and actin (on total protein lysate). www.aacrjournals.org 1969 Cancer Res 2005; 65: (5). March 1, 2005

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 2005 American Association for Cancer Research. Cancer Research the activation of the mitochondrial death pathway. However, this showed that rapamycin mediates cyclin D1 down-regulation in does not seem to be due to a rapamycin-mediated alteration in the part by accelerated proteolysis. We showed that rapamycin levels of the proapoptotic and antiapoptotic molecules examined. activates GSK3h and down-regulates cyclin D1 levels in a We next evaluated the role of GSK3h in rapamycin-enhanced GSK3h-dependent fashion. Further, we showed that both cytotoxicity. We first evaluated the effect of GSK3h inhibitors on rapamycin-dependent growth inhibition and rapamycin-mediated the cytotoxicity of rapamycin in combination with paclitaxel. After chemosensitization are attenuated in GSK3h-null cells compared MDA-MB-468 cells were treated with 100 nmol/L rapamycin and with GSK3h wild-type cells. Taken together, these data show that 0.01 Ag/mL paclitaxel for 24 hours, we evaluated apoptosis as GSK3h plays a significant role in rapamycin-mediated anticancer shown by PARP cleavage. Coincubation of cells with rapamycin and effects, including cell cycle regulation, growth inhibition, and paclitaxel along with GSK3h inhibitors SB216763 (20 Amol/L) or chemosensitization. SB415286 (10 Amol/L) completely inhibited PARP cleavage (Fig. 6D). We have reported recently that rapamycin leads to a reduction in Coincubation with 10 mmol/L LiCl, however, did not inhibit cyclin D1 levels in rapamycin-sensitive MCF-7 and MDA-MB-468 PARP cleavage. When we evaluated apoptosis at 48 hours with cells but not in rapamycin-resistant MDA-MB-231, MDA-MB-435, Annexin V labeling, we showed that treatment with rapamycin in and NCI/ADR-RES cell lines (22). Our findings are consistent with addition to paclitaxel significantly enhanced apoptosis compared the results of Gera et al., who found that cyclin D1 is down- with treatment with paclitaxel alone (7% versus 37%; P = 0.0003; regulated in two rapamycin-sensitive cell lines with activated Akt Fig. 6E). Coincubation with GSK3h inhibitors SB216763 or but not in their isogenic counterparts that have less Akt activity SB415286 decreased the amount of apoptosis seen with rapamycin and are rapamycin resistant (3). In this study, we show that cyclin and paclitaxel (P = 0.0036 and 0.0012, respectively). Interestingly, in D1 overexpression partially overcomes rapamycin-mediated cell contrast, coincubation with LiCl did not decrease apoptosis but cycle arrest and growth inhibition. Similar results obtained in rather was associated with an increase in apoptosis. Following the glioblastoma cells were reported recently by Gera et al. (3). same treatment schedule, when the percentage of nonviable cells Rapamycin was also reported to inhibit the expression of cyclin D1 was determined with a trypan blue dye exclusion assay, we found in hepatocytes, and transfection with cyclin D1 was found to that coincubation with GSK3h inhibitors SB216763 or SB415286, overcome rapamycin-mediated cell cycle arrest in rat hepatocytes but not with LiCl, decreased the percentage of nonviable cells (2). These results taken together suggest that cyclin D1 is a key observed with rapamycin and paclitaxel treatment (Fig. 6F). Of mediator of cell growth and proliferation downstream of mTOR note, the treatments with SB216763 or SB415286 and especially LiCl and that down-regulation of cyclin D1 is critical to rapamycin- alone led to an increase in cell death. Thus, inhibition of GSK3h induced growth inhibition. itself may lead to cell death; this observation is consistent with the However, cyclin D1 is not down-regulated by rapamycin in all hypothesis that maintenance of baseline GSK3h activity is actually experimental systems. For example, cyclin D1 levels were not required for cell survival (30). In our experiments, inhibition of found to be altered by rapamycin analogue CCI-779 in multiple GSK3h decreases the rapamycin-enhanced paclitaxel cytotoxicity myeloma cells, which showed a growth-inhibitory response (34). when GSK3h inhibitors SB216763 or SB415286 were used, Further, rapamycin treatment of the Eker rat model of tuberous h suggesting that GSK3 activity may play a role in rapamycin- sclerosis complex renal tumors led to a significant tumor mediated chemosensitization. Interestingly, LiCl did not decrease response, without altering cyclin D1 levels (35). These results the cell death induced by the combination of rapamycin and show that rapamycin does not uniformly lead to a decrease in paclitaxel; this may be because treatment with LiCl alone induces cyclin D1 levels in all cell types. The effect of rapamycin on cyclin more cell death or may be due to the differences in the specificity D1 may vary with the activity of different signaling pathways in of LiCl compared with SB216763 or SB415286. each cell. For example, in the Eker rat kidney tumor model where To further investigate the role of GSK3h in rapamycin-mediated Akt was found not to be phosphorylated (35), GSK3h may already chemosensitization, we evaluated the effect of rapamycin on be in an activated state due to lack of negative regulation by the paclitaxel-induced apoptosis in GSK3h wild-type and GSK3h-null PI3K/Akt pathway; thus, rapamycin may not lead to significant cells. GSK3h wild-type and GSK3h-null cells were treated for cyclin D1 proteolysis. Further, our results show that over- 48 hours with 100 nmol/L rapamycin alone, 0.01 Ag/mL paclitaxel expression of cyclin D1 only partially overcomes the effect of alone, and combination. Annexin V–positive cells were determined rapamycin on cell cycle and growth. This may be not only by fluorescence-activated cell sorting (FACS) analysis (Fig. 6G). because the exogenous cyclin D1 is down-regulated by rapamycin Rapamycin significantly enhanced paclitaxel-induced apoptosis in but also because cyclin D1 is not the only mediator of the GSK3h wild-type cells (paclitaxel alone 6% versus paclitaxel and downstream effects of rapamycin. rapamycin 41%; P = 0.0128). In contrast, rapamycin did not Rapamycin seems to down-regulate cyclin D1 expression at significantly enhance paclitaxel-induced apoptosis in GSK3h-null many levels. Rapamycin has been found to decrease cyclin D1 cells. Furthermore, the combination of rapamycin and paclitaxel mRNA levels in U87 glioma cells and LAPC prostate cancer cells lead to cytoplasmic cytochrome c release, caspase-9, and PARP transfected with constitutively active Akt (3). Rapamycin treatment cleavage in GSK3h wild-type cells but not in GSK3h-null cells is associated with dephosphorylation of 4E-binding protein 1 (Fig. 6H). These results show that the GSK3h activity is critical to (22, 31); the resulting decrease in eukaryotic initiation factor 4E rapamycin-mediated paclitaxel sensitization. availability may alter cyclin D1 mRNA export, as eukaryotic initiation factor 4E has been shown to facilitate the nuclear- Discussion cytoplasmic export of cyclin D1 mRNA (36, 37). Recently, Gera et al. Rapamycin inhibits cell growth and proliferation and is showed that rapamycin decreases cyclin D1 translation 5-fold currently in clinical trials as an anticancer agent. In the present in U87MG glioblastoma cells and LAPC prostate cancer cells study, we showed that rapamycin causes G1 cell cycle arrest in transfected with constitutively active Akt (3). In this study, we breast cancer cells, with a decrease in cyclin D1 levels. We showed that rapamycin also down-regulates cyclin D1 by enhancing

Cancer Res 2005; 65: (5). March 1, 2005 1970 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 2005 American Association for Cancer Research. Role of GSK3bb in Rapamycin-Mediated Antitumor Effects its proteolysis. Our results are consistent with those of Hashemol- GSK3h is critical to rapamycin-mediated chemosensitization. It is hosseini et al. who had reported previously that rapamycin decreases unknown at this time which of the many putative substrates of cyclin D1 protein stability in NIH3T3 cells (21). Our finding that the GSK3h (48) mediates or potentiates paclitaxel-induced apoptosis. expression of endogenous cyclin D1 was decreased 70% by Enhanced GSK3h activity has been reported to mediate hypoxia- rapamycin treatment, whereas the expression of exogenous cyclin induced apoptosis via activation of the mitochondrial death D1 that lacked transcriptional, translational, and mRNA stability pathway and has been attributed in part to alterations in glucose elements decreased by 38% suggests that alterations in protein transport and metabolism (49). GSK3h has also been found to play a stability account for only a component of the effect of rapamycin on role in regulating survival downstream of PI3K/Akt signaling (50). In cyclin D1 levels in the cell lines we tested. Interestingly, inhibition of this context, GSK3h has been shown to control cell survival GSK3h dramatically inhibits rapamycin-mediated cyclin D1 down- upstream of mitochondrial cytochrome c release as a result of regulation. Therefore, GSK3h may play a role not only in the phosphorylation of translation eukaryotic initiation factor 2B and proteolysis of cyclin D1 protein but also in other mechanisms of global regulation of protein synthesis (50). Thus, eukaryotic rapamycin-induced cyclin D1 down-regulation. initiation factor 2B may also be involved in rapamycin-mediated The role of mTOR signaling in regulation of glycogen synthase chemosensitization, and if so, this mechanism is likely to sensitize to and GSK3 activity has been controversial to date. We show here a broad range of cytotoxic stimuli. Alternatively, the chemosensitizing that rapamycin treatment is associated with activation of GSK3h. effect of GSK3h may be mediated through its other phosphorylation In contrast, several reports have found no effect of rapamycin on targets, such as microtubule-associated proteins Tau and 1B (48); GSK3 in other cell types, including Chinese hamster ovary cells, if so, the role of GSK3h may be more important for microtubule- human myoblasts, rat hepatocytes, and rat epididymal fat cells targeted agents, such as paclitaxel, and thus more specific. Further (25, 38–42). In L6 muscle cells, rapamycin failed to block GSK3 studies to determine the mechanism of rapamycin-mediated inactivation in response to insulin but inhibited GSK3 inactivation chemosensitization are ongoing to help identify patients who will in response to amino acids, suggesting that rapamycin may benefit the most from combination therapy with mTOR inhibitors. regulate GSK3 in a stimulus-dependent manner (43). The differ- Alterations in GSK3h has been associated with a variety of ences observed between the studies may also be related to pathologic conditions, including diabetes, Alzheimer’s disease, and differences between the cell lines and lineages studied, the bipolar disorder (28, 51, 52). Strikingly, in the phase I clinical trial of activation status of the signaling pathways, and the baseline mTOR inhibitor CCI-779 (Wyeth, Collegeville, PA), higher doses GSK3h activity of cells or to the time points studied. Further, most induced euphoria followed by melancholy, mimicking bipolar of these studies have evaluated the ability of rapamycin to block disorder in three patients with no previous history of psychiatric the inhibition of GSK3 by stimuli, such as insulin, rather than disorders (53). Our finding that rapamycin activates GSK3h may be evaluating the effect of rapamycin treatment alone in the presence able to shed light on the etiology of this unexpected toxicity. of serum and nutrients as was done in our study. Our results show We also show that LiCl interferes with rapamycin-mediated cell that rapamycin activates GSK3h and suggest that GSK3h is a cycle arrest, and SB216763 and SB415286 interfere with rapamycin- previously unrecognized critical downstream effector of the mediated paclitaxel sensitization. Although LiCl is an ATP h antitumor effects of rapamycin. noncompetitive inhibitor of GSK3 activity that has been widely The mechanism of GSK3h activation by rapamycin observed in used to evaluate the effects of GSK3h, it is known to inhibit other breast cancer cells is not known at this time. One possibility is that targets, such as casein kinase-2, p38 kinase, mitogen-activated mTOR directly regulates one of the kinases that phosphorylates protein kinase-2, polyphosphate 1-phosphatase, and inositol GSK3h. The target of mTOR, S6 kinase-1, has been shown monophosphatase (29). SB216763 and SB415286 are ATP compet- previously to phosphorylate GSK3h in vitro (26); however, we did itive compounds that are considered selective inhibitors of GSK3h not observe a significant dephosphorylation of the in vitro GSK3h (29); however, they may potentially have other targets. Thus, we phosphorylation site Ser9 of S6 kinase-1 at the time points in which cannot definitively determine whether our findings with these we observed kinase activation. Although rapamycin also did not small molecule inhibitors are directly attributable to the inhibition affect the GSK3h Ser9 phosphorylation in multiple myeloma cells of GSK3h. However, currently, LiCl is widely used for the treatment and baby hamster kidney cells (44, 45), it was reported to inhibit of bipolar disease, and GSK3h is also being considered as a GSK3h Ser9 phosphorylation induced by opioids in a rat cardiac therapeutic target for diabetes, stroke, and Alzheimer’s disease (54). ischemia/reperfusion model (46). Thus, we cannot exclude the Our results emphasize that these GSK3h inhibitors may have possibility that rapamycin affects GSK3h Ser9 phosphorylation in previously unrecognized drug interactions that may decrease the different models. Alternatively, GSK3h may be regulated by another antitumor activity of mTOR inhibitors. Further study is needed to mTOR-regulated kinase that has not been identified. Another determine the clinical relevance of these interactions. mechanism through which rapamycin may regulate GSK3h is by activating a phosphatase that dephosphorylates GSK3h, such as protein phosphatase 2A (47). A third possible mechanism of GSK3h Acknowledgments regulation by rapamycin may be indirect. In this scenario, a third Received 7/13/2004; revised 11/16/2004; accepted 12/23/2004. molecule involved in GSK3h regulation may be regulated by Grant Support: Goodwin Foundation, NIH grant 1K08-CA-91895-01, 1 R01 CA112199-01, University of Texas M.D. Anderson Cancer Center Physician-Scientist inhibition of the mTOR signaling pathway at the translational or Program (F. Meric-Bernstam), NIH grant 5T32CA09599 (Department of Surgical transcriptional level. Determination of the mechanism of GSK3h Oncology), and NIH Cancer Support grant P30-CA-16672 (University of Texas M.D. Anderson Cancer Center). regulation by rapamycin may lead to identification of novel The costs of publication of this article were defrayed in part by the payment of page pathways or additional regulatory elements for GSK3h. Therefore, charges. This article must therefore be hereby marked advertisement in accordance these possibilities deserve further study. with 18 U.S.C. Section 1734 solely to indicate this fact. We thank Ying Yang, Mark F. Munsel, and Kristine Broglia (Department of In our work, rapamycin enhances paclitaxel-induced cytotoxicity Biostatistics) for assistance with data analysis, Anne Packer for editorial assistance, in GSK3h wild-type cells but not in GSK3h-null cells, suggesting that and Marlen Banda for assistance with article preparation. www.aacrjournals.org 1971 Cancer Res 2005; 65: (5). March 1, 2005

References 20. Casanovas O, Miro F, Estanyol JM, et al. Osmotic 39. Lavoie L, Band CJ, Kong M, Bergeron JJ, Posner BI. stress regulates the stability of cyclin D1 in a p38SAPK2- Regulation of glycogen synthase in rat hepatocytes. 1. GreweM,GansaugeF,SchmidRM,AdlerG, dependent manner. J Biol Chem 2000;275:35091–7. Evidence for multiple signaling pathways. J Biol Chem Seufferlein T. Regulation of cell growth and cyclin 21. Hashemolhosseini S, Nagamine Y, Morley SJ, et al. 1999;274:28279–85. D1 expression by the constitutively active FRAP- Rapamycin inhibition of the G1 to S transition is 40. Hurel SJ, Rochford JJ, Borthwick AC, et al. Insulin p70s6K pathway in human pancreatic cancer cells. mediated by effects on cyclin D1 mRNA and protein action in cultured human myoblasts: contribution of Cancer Res 1999;59:3581–7. stability. J Biol Chem 1998;273:14424–9. different signalling pathways to regulation of glycogen 2. Nelsen CJ, Rickheim DG, Tucker MM, Hansen LK, 22. Noh WC, Mondesire WH, Peng J, et al. Determinants synthesis. Biochem J 1996;320:871–7. Albrecht JH. Evidence that cyclin D1 mediates both of rapamycin sensitivity in breast cancer cells. Clin 41. Welsh GI, Foulstone EJ, Young SW, Tavare JM, growth and proliferation downstream of TOR in Cancer Res 2004;10:1013–23. Proud CG. Wortmannin inhibits the effects of insulin hepatocytes. J Biol Chem 2003;278:3656–63. 23. Hoeflich KP, Luo J, Rubie EA, et al. Requirement for and serum on the activities of glycogen synthase kinase-3 3. Gera JF, Mellinghoff IK, Shi Y, et al. AKT activity glycogen synthase kinase-3h in cell survival and NF-nB and mitogen-activated protein kinase. Biochem J determines sensitivity to mammalian target of rapa- activation. Nature 2000;406:86–90. 1994;303:15–20. mycin (mTOR) inhibitors by regulating cyclin D1 and 24. Liu S, Yu S, Hasegawa Y, et al. Glycogen synthase 42. Cross DA, Alessi DR, Vandenheede JR, et al. The c-myc expression. J Biol Chem 2004;279:2737–46. kinase 3h is a negative regulator of growth factor- inhibition of glycogen synthase kinase-3 by insulin or 4. Aguirre D, Boya P, Bellet D, et al. Bcl-2 and CCND1/ induced activation of the c-Jun N-terminal kinase. J Biol insulin-like growth factor 1 in the rat skeletal muscle CDK4 expression levels predict the cellular effects of Chem 2004;279:51075–81. cell line L6 is blocked by wortmannin, but not by mTOR inhibitors in human ovarian carcinoma. Apo- 25. Cross DA, Alessi DR, Cohen P, Andjelkovich M, rapamycin: evidence that wortmannin blocks activa- ptosis 2004;9:797–805. Hemmings BA. Inhibition of glycogen synthase kinase-3 tion of the mitogen-activated protein kinase pathway 5. Sherr CJ, Roberts JM. CDK inhibitors: positive and by insulin mediated by protein kinase B. Nature in L6 cells between Ras and Raf. Biochem J negative regulators of G -phase progression. Genes Dev 1 1995;378:785–9. 1994;303:21–6. 1999;13:1501–12. 43. Peyrollier K, Hajduch E, Blair AS, Hyde R, and 6. Coqueret O. Linking cyclins to transcriptional control. 26. Cohen P, Frame S. The renaissance of GSK3. Nat Rev Hundal HS. L-Leucine availability regulates phosphati- Gene 2002;299:35–55. Mol Cell Biol 2001;2:769–76. dylinositol 3-kinase, p70 S6 kinase and glycogen 7. Wang C, Pattabiraman N, Zhou JN, et al. Cyclin D1 27. Hughes K, Nikolakaki E, Plyte SE, Totty NF, Woodgett JR. synthase kinase-3 activity in L6 muscle cells: evidence repression of peroxisome proliferator-activated recep- Modulation of the glycogen synthase kinase-3 family by for the involvement of the mammalian target of tor g expression and transactivation. Mol Cell Biol tyrosine phosphorylation. EMBO J 1993;12:803–8. rapamycin (mTOR) pathway in the L-leucine-induced 2003;23:6159–73. 28. Klein PS, Melton DA. A molecular mechanism for up-regulation of system A amino acid transport. 8. Lamb J, Ramaswamy S, Ford HL, et al. A mechanism of the effect of lithium on development. Proc Natl Acad Biochem J 2000;350:361–8. cyclin D1 action encoded in the patterns of gene Sci U S A 1996;93:8455–9. expression in human cancer. Cell 2003;114:323–34. 29. Coghlan MP, Culbert AA, Cross DA, et al. Selective 44. Pene F, Claessens YE, Muller O, et al. Role of the 9. Weinstat-Saslow D, Merino MJ, Manrow RE, et al. small molecule inhibitors of glycogen synthase kinase-3 phosphatidylinositol 3-kinase/Akt and mTOR/P70S6- Overexpression of cyclin D mRNA distinguishes invasive modulate glycogen metabolism and gene transcription. kinase pathways in the proliferation and apoptosis in and in situ breast carcinomas from non-malignant Chem Biol 2000;7:793–803. multiple myeloma. Oncogene 2002;21:6587–97. lesions. Nat Med 1995;1:1257–60. 30. Jope RS, Johnson GV. The glamour and gloom of 45. Versteeg HH, Sorensen BB, Slofstra SH, et al. VIIa/ 10. Kenny FS, Hui R, Musgrove EA, et al. Overexpression glycogen synthase kinase-3. Trends Biochem Sci tissue factor interaction results in a tissue factor of cyclin D1 messenger RNA predicts for poor prognosis 2004;29:95–102. cytoplasmic domain-independent activation of protein in estrogen receptor-positive breast cancer. Clin Cancer 31. Mondesire WH, Jian W, Zhang H, et al. Targeting synthesis, p70, and p90 S6 kinase phosphorylation. J Biol Res 1999;5:2069–76. mammalian target of rapamycin synergistically enhan- Chem 2002;277:27065–72. 11. Wang TC, Cardiff RD, Zukerberg L, et al. Mammary ces chemotherapy-induced cytotoxicity in breast cancer 46. Gross ER, Hsu AK, Gross GJ. Opioid-induced hyperplasia and carcinoma in MMTV-cyclin D1 trans- cells. Clin Cancer Res 2004;10:7031–42. cardioprotection occurs via glycogen synthase kinase h genic mice. Nature 1994;369:669–71. 32. PolunovskyVA,RosenwaldIB,TanAT,etal. inhibition during reperfusion in intact rat hearts. Circ 12. Yu Q, Geng Y, Sicinski P. Specific protection against Translational control of programmed cell death: Res 2004;94:960–6. breast cancers by cyclin D1 ablation. Nature eukaryotic translation initiation factor 4E blocks 47. Peterson RT, Desai BN, Hardwick JS, Schreiber SL. 2001;411:1017–21. apoptosis in growth-factor-restricted fibroblasts with Protein phosphatase 2A interacts with the 70-kDa S6 13. Qiu C, Shan L, Yu M, Snyderwine EG. Deregulation of physiologically expressed or deregulated Myc. Mol Cell kinase and is activated by inhibition of FKBP12- the cyclin D1/Cdk4 retinoblastoma pathway in rat Biol 1996;16:6573–81. rapamycinassociated protein. Proc Natl Acad Sci U S A mammary gland carcinomas induced by the food- 33. Li S, Sonenberg N, Gingras AC, et al. Translational 1999;96:4438–42. derived carcinogen 2-amino-1-methyl-6-phenylimi- control of cell fate: availability of phosphorylation sites 48. Frame S, Cohen P. GSK3 takes centre stage more dazo[4,5-b]pyridine. Cancer Res 2003;63:5674–8. on translational repressor 4E-BP1 governs its proapop- than 20 years after its discovery. Biochem J 2001;359: 14. Lin S, Wang W, Wilson GM, et al. Down-regulation of totic potency. Mol Cell Biol 2002;22:2853–61. 1–16. cyclin D1 expression by prostaglandin A(2) is mediated 34. Shi Y, Gera J, Hu L, et al. Enhanced sensitivity of 49. Loberg RD, Vesely E, Brosius FC III. Enhanced by enhanced cyclin D1 mRNA turnover. Mol Cell Biol multiple myeloma cells containing PTEN mutations to glycogen synthase kinase-3h activity mediates hypoxia- 2000;20:7903–13. CCI-779. Cancer Res 2002;62:5027–34. induced apoptosis of vascular smooth muscle cells and 15. Mulloy R, Salinas S, Philips A, Hipskind RA. 35. Kenerson HL, Aicher LD, True LD, Yeung RS. is prevented by glucose transport and metabolism. J Biol Activation of cyclin D1 expression by the ERK5 cascade. Activated mammalian target of rapamycin pathway in Chem 2002;277:41667–73. Oncogene 2003;22:5387–98. the pathogenesis of tuberous sclerosis complex renal 50. Pap M, Cooper GM. Role of glycogen synthase kinase- 16. Lin SY, Xia W, Wang JC, et al. h-catenin, a novel tumors. Cancer Res 2002;62:5645–50. 3 in the phosphatidylinositol 3-kinase/Akt cell survival prognostic marker for breast cancer: its roles in cyclin 36. Rosenwald IB, Kaspar R, Rousseau D, et al. pathway. J Biol Chem 1998;273:19929–32. D1 expression and cancer progression. Proc Natl Acad Eukaryotic translation initiation factor 4E regulates 51. Alonso M, Martinez A. GSK-3 inhibitors: discov- Sci U S A 2000;97:4262–6. expression of cyclin D1 at transcriptional and post- eries and developments. Curr Med Chem 2004;11: 17. Dufourny B, van Teeffelen HA, Hamelers IH, transcriptional levels. J Biol Chem 1995;270:21176–80. 755–63. Sussenbach JS, Steenbergh PH. Stabilization of cyclin 37. Rousseau D, Kaspar R, Rosenwald I, Gehrke L, 52. Gould TD, Einat H, Bhat R, Manji HK. AR-A014418, a D1 mRNA via the phosphatidylinositol 3-kinase Sonenberg N. Translation initiation of ornithine decar- selective GSK-3 inhibitor, produces antidepressant-like pathway in MCF-7 human breast cancer cells. boxylase and nucleocytoplasmic transport of cyclin D1 effects in the forced swim test. Int J Neuropsychophar- J Endocrinol 2000;166:329–38. mRNA are increased in cells overexpressing eukaryotic macology 2004;1–4. Q4Q4 18. Muise-Helmericks RC, Grimes HL, Bellacosa A, et al. initiation factor 4E. Proc Natl Acad Sci U S A 1996;93: 53. Raymond E, Alexandre J, Faivre S, et al. Safety and Cyclin D expression is controlled post-transcriptionally 1065–70. pharmacokinetics of escalated doses of weekly via a phosphatidylinositol 3-kinase/Akt-dependent 38. Moule SK, Welsh GI, Edgell NJ, et al. Regulation of intravenous infusion of CCI-779, a novel mTOR pathway. J Biol Chem 1998;273:29864–72. protein kinase B and glycogen synthase kinase-3 by inhibitor, in patients with cancer. J Clin Oncol 2004; 19. Diehl JA, Cheng M, Roussel MF, and Sherr CJ. insulin and h-adrenergic agonists in rat epididymal fat 22:2336–47. Glycogen synthase kinase-3h regulates cyclin D1 cells. Activation of protein kinase B by wortmannin- 54. Meijer L, Flajolet M, Greengard P. Pharmacological proteolysis and subcellular localization. Genes Dev sensitive and -insensitive mechanisms. J Biol Chem inhibitors of glycogen synthase kinase 3. Trends 1998;12:3499–511. 1997;272:7713–9. Biochem Sci 2004;25:471–80.

Cancer Res 2005; 65: (5). March 1, 2005 1972 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 2005 American Association for Cancer Research. Role of Glycogen Synthase Kinase 3β in Rapamycin-Mediated Cell Cycle Regulation and Chemosensitivity

JinJiang Dong, Junying Peng, Haixia Zhang, et al.

Cancer Res 2005;65:1961-1972.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/65/5/1961

Cited articles This article cites 49 articles, 32 of which you can access for free at: http://cancerres.aacrjournals.org/content/65/5/1961.full#ref-list-1

Citing articles This article has been cited by 11 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/65/5/1961.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/65/5/1961. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.