Maintaining Glycogen Synthase Kinase-3 Activity Is Critical for Mtor Kinase Inhibitors to Inhibit Cancer Cell Growth

Published OnlineFirst March 13, 2014; DOI: 10.1158/0008-5472.CAN-13-2946

Cancer Therapeutics, Targets, and Chemical Biology Research

Maintaining Glycogen Synthase Kinase-3 Activity Is Critical for mTOR Kinase Inhibitors to Inhibit Cancer Cell Growth

Junghui Koo1, Ping Yue1, Anthony A. Gal2, Fadlo R. Khuri1, and Shi-Yong Sun1

Abstract mTOR kinase inhibitors that target both mTORC1 and mTORC2 are being evaluated in cancer clinical trials. Here, we report that glycogen synthase kinase-3 (GSK3) is a critical determinant for the therapeutic response to this class of experimental drugs. Pharmacologic inhibition of GSK3 antagonized their suppressive effects on the growth of cancer cells similarly to genetic attenuation of GSK3. Conversely, expression of a constitutively activated form of GSK3b sensitized cancer cells to mTOR inhibition. Consistent with these findings, higher basal levels of GSK3 activity in a panel of human lung cancer cell lines correlated with more efficacious responses. Mechanistic investigations showed that mTOR kinase inhibitors reduced cyclin D1 levels in a GSK3b-dependent manner, independent of their effects on suppressing mTORC1 signaling and cap binding. Notably, selective inhibition of mTORC2 triggered proteasome-mediated cyclin D1 degradation, suggesting that mTORC2 blockade is responsible for GSK3-dependent reduction of cyclin D1. Silencing expression of the ubiquitin E3 ligase FBX4 rescued this reduction, implicating FBX4 in mediating this effect of mTOR inhibition. Together, our findings define a novel mechanism by which mTORC2 promotes cell growth, with potential implications for understanding the clinical action of mTOR kinase inhibitors. Cancer Res; 74(9); 2555–68. 2014 AACR.

Introduction with weak activity against mTORC2. Although some rapalogs The mTOR, a serine–threonine protein kinase related tightly (e.g., everolimus) are approved by the U.S. Food and Drug to the family of the phosphoinositide 3-kinase–related kinases, Administration for the treatment of advanced renal cell cancer exerts different biologic functions primarily through forming and pancreatic neuroendocrine tumors, the single-agent activ- two complexes with the essential partner protein raptor ity of rapalogs in most other tumor types has been modest at (mTOR complex 1; mTORC1) and rictor (mTOR complex 2; best (4). Hence, great efforts have been made to identify novel mTORC2; refs. 1, 2). Compared with the mTORC1, which is mTOR inhibitors that suppress both mTORC1 and mTORC2 involved in regulation of many key cellular processes, including activity. As a result, several ATP-competitive inhibitors of cell growth and metabolism primarily via regulating cap- mTOR kinase, including INK128 and PP242, have been devel- dependent protein translation initiation, relatively little is oped and are being tested in clinical trials (5, 6). These mTOR known about the biologic functions of the mTORC2 other kinase inhibitors (TORKinibs) in general more dramatically than its regulation of cytoskeleton and Akt-mediated cell inhibit protein synthesis, suppress Akt phosphorylation, and survival (2). Nonetheless, mTOR signaling is dysregulated in induce G1 arrest and/or apoptosis in some cancer cells than the – various types of human cancers and, hence, has emerged as an conventional allosteric mTOR inhibitor, rapamycin (7 10). A in vivo attractive cancer therapeutic target (3). robust anticancer activity of these inhibitors against The conventional mTOR inhibitors, rapamycin and its ana- certain types of cancers was also observed (8, 11, 12). Some logs (rapalogs), are specific allosteric inhibitors of mTORC1 TORKinibs have been tested in clinical trials (5, 6). Therefore, these TORKinibs not only represent novel potential cancer therapeutic agents, but also are valuable research tools for understanding the biology of mTORCs. fi 1 Authors' Af liations: Departments of Hematology and Medical Oncology Glycogen synthase kinase-3 (GSK3) is a ubiquitous serine/ and 2Pathology, Emory University School of Medicine and Winship Cancer Institute, Atlanta, Georgia threonine kinase that is present in mammals in two isoforms: a and b (13). GSK3 was initially identified as an enzyme involved Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/). in the regulation of glycogen metabolism. Increasing evidence during the past decades indicates that GSK3 has a key role in Note: F.R. Khuri and S.-Y. Sun are Georgia Research Alliance Distinguished Cancer Scientists. regulating a diverse range of cellular functions, including cell survival and death (13). Thus, GSK3 inhibition has been con- Corresponding Author: Shi-Yong Sun, Emory University School of Med- icine and Winship Cancer Institute, 1365-C Clifton Road NE, C3088, sidered an attractive therapeutic strategy for certain diseases Atlanta, GA 30322. Phone: 404-778-2170; Fax: 404-778-5520; E-mail: such as diabetes, neurodegenerative diseases, and mental dis- [email protected] orders (14, 15). GSK3 has been implicated in the regulation of doi: 10.1158/0008-5472.CAN-13-2946 oncogenesis with complex patterns: It acts paradoxically as a 2014 American Association for Cancer Research. tumor suppressor in some cancer types while potentiating

www.aacrjournals.org 2555

Koo et al.

growth of cancer cells in others (16, 17). One well-known have not been authenticated. These cell lines were cultured important cancer-related function of GSK3 is to positively in RPMI-1640 or Dulbecco's Modified Eagle Medium con- regulate proteasomal degradation of several oncogenic pro- taining 5% fetal calf serum at 37C in a humidified atmo- – teins such as c-Myc, c-Jun, cyclin E, Mcl-1, and cyclin D (18 20). sphere of 5% CO2 and 95% air. Plasmid transfection was For example, GSK3-dependent cyclin D1 phosphorylation is conducted in H1299 or HEK-293T cells largely because of required for cyclin D1 degradation mediated by the E3 ubiqui- their high transfection efficiency. tin ligase FBX4 (20, 21). It has been suggested that GSK3 can inhibit the mTOR Cell growth assay pathway by phosphorylating TSC2 in a manner dependent on Cells were seeded in 96-well cell culture plates and treated AMPK-priming phosphorylation (22). A recent study has the next day with the given agents. Viable cell numbers were shown that GSK3 phosphorylates the turn motif of p70S6K determined using a sulforhodamine B (SRB) assay as described and cooperates with mTOR to control the activity of p70S6K previously (26). Combination index (CI) for drug interaction and cell proliferation (23), thus providing a rationale for (e.g., synergy) was calculated using the CompuSyn software cotargeting mTOR and GSK3 to treat diseases such as cancer. (ComboSyn, Inc.). The IC50, which represents the concentra- In a longstanding effort to identify strategies or agents that can tion required for 50% growth inhibition, was estimated from a potentially enhance the therapeutic efficacy of mTOR inhibi- concentration-dependent growth curve. tors in cancer therapy, we unexpectedly found that the activity of GSK3 is crucial for TORKinibs to exert their inhibitory effects Cell-cycle analysis on the growth of cancer cells. Thus, this work has focused on Cells were harvested after a given treatment and stained demonstrating the impact of GSK3 on the therapeutic activity with propidium iodide for cell-cycle analysis as described of TORKinibs against cancer cells and on understanding the previously (27). underlying mechanisms. Colony formation assay Materials and Methods The effects of the given drugs on colony formation on plates Reagents were measured as previously described (28). PP242, INK128, and AZD8055 were purchased from Active Biochem. Torin 1 was purchased from Tocris. The GSK3 inhib- Western blot analysis itor SB216763, the proteasome inhibitor MG132, and the protein Preparation of whole-cell protein lysates and Western blot synthesis inhibitor cycloheximide (CHX) were purchased from analysis was performed as described previously (28). Sigma Chemical Co. The NEDD8-activating enzyme inhibitor Gene knockdown by siRNA or small hairpin RNA MLN4924 was provided by Millennium Pharmaceuticals, Inc. 0 0 Cyclin D1, p-GSK3a/b (S21/9), p-AKT (S473), AKT, p-S6 (S235/ Rictor #1 (5 -AAGCAGCCTTGAACTGTTTAA-3 ), rictor #2 (50-AAACTTGTGAAGAATCGTATC-30), raptor #1 (50-AAGGC- 236), and S6 antibodies were purchased from Cell Signaling 0 0 Technology, Inc. GSK3a/b antibody was purchased from TAGTCTGTTTCGAAAT-3 ), raptor #2 (5 -AAGGACAACGGC- 0 a 0 Upstate/EMD Millipore. Polyclonal rictor and raptor antibodies CACAAGTAC-3 ), GSK-3 (5 -AAGTGATTGGCAATGGCT- 0 b 0 0 were purchased from Bethyl Laboratories, Inc. Both polyclonal CAT-3 ), and GSK-3 (5 -AAGTAATCCACCTCTGGCTAC-3 ) a b and monoclonal actin antibodies were purchased from Sigma siRNAs were synthesized by Qiagen. GSK3 / siRNA (#6301) Chemical Co. Myc-tagged constitutively active form of were purchased from Cell Signaling Technology, Inc. The GSK3b (GSK3bCA; ref. 24) was provided by Dr. Binhua P. nonsilencing control siRNA duplexes were described previous- Zhou (The University of Kentucky, College of Medicine, ly (29). Transfection of these siRNA duplexes was conducted in Lexington, Kentucky). Flag-cyclin D1 expression plasmid 6-well plates using the Lipofectamine 2000 transfection was provided by Dr. Alan Diehl (Abramson Family Cancer reagent (Invitrogen) following the manufacturer's manual. Research Institute, University of Pennsylvania, Philadelphia, Short hairpin RNA (shRNA) sets in lentiviral pLKO.1 vector PA). Myc-Rictor and HA-raptor expression plasmids were for FBX4 were purchased from Open Biosystems and used as purchased from Addegene. the manufacturer instructed.

Cell lines and cell culture m7GTP pull-down for analysis of the eIF4F complex Human non–small cell lung cancer (NSCLC) cell lines The eIF4F complex in cell extracts was detected using used in this study and H157-scramble, H157-shRaptor, and affinity chromatography m7GTP-sepharose as described pre- H157-shRictor stable cell lines were described in our previ- viously (30) ous work (25). Wild-type (WT), GSK3a-KO, and GSK3b-KO murine embryonic fibroblasts (MEF) were generously pro- Immunoprecipitation for detection of cyclin D1 vided by Dr. Jim Woodgett (Samuel Lunenfeld Research ubiquitination Institute, Mount Sinai Hospital, Toronto, Canada). HEK- H1299 cells were cotransfected with HA-ubiquitin only or 293T cells were provided by Keqiang Ye (Emory University, with HA-ubiquitin plus Flag-cyclin D1 plasmids using Lipo- Atlanta, GA). Except for H157 and A549 cells, which were fectamine 2000 transfection reagent (Invitrogen) based on the authenticated by Genetica DNA Laboratories, Inc., through manufacturer's instructions. After 24 hours, the cells were analyzing short tandem repeat DNA profile, other cell lines treated with INK128 or INK128 plus MG132 for 4 hours. Cells

2556 Cancer Res; 74(9) May 1, 2014 Cancer Research

GSK3 Affects Cell Response to mTOR Kinase Inhibitors

were collected and lysed for immunoprecipitation using cyclin lines (Fig. 1A). Using 7.5 mmol/L SB216763 as an example, the D1 antibody (Santa Cruz Biotechnology, Inc.) or Flag M2 CIs in every tested cell line were far greater than 1 (Supple- monoclonal antibody (Sigma) as previously described (31), mentary Fig. S1), indicating antagonistic effects. In a long-term followed by detection of ubiquitinated cyclin D1 with Western colony formation assay that allows us to repeat the treatments, blot analysis using anti-HA (Abgent) or antiubiquitin (Cell we also found that INK128 alone had greater inhibitory effect Signaling Technology, Inc.) antibody. than the combination of SB216763 and INK128 on the growth of NSCLC colonies (Fig. 2A), further confirming the antago- Reverse transcription PCR to detect FBX4 mRNA nistic effect. In our tested cell lines, INK128 at the tested expression concentration (e.g., 100 nmol/L) primarily induced G1 arrest; Total cellular RNA was isolated from the given cell line with however, this effect was abrogated in the presence of SB216763 TRIzol (Sigma Chemical Co.). Reverse transcription was per- (Fig. 1B). Furthermore, we tested the effects of SB216763 on formed with the iScript Select cDNA Synthesis Kit (Bio-Rad), the growth-inhibitory effects of other TORKinibs, including followed with PCR using primers as follows: FBX4 (50- AZD8055, Torin 1, and PP242. Consistently, the presence AGCCGGTACAGTGTGATTCC-30, forward) and (50-CCAAA- of SB216763 significantly attenuated the growth-inhibitory GGCTGGATCTGTCAT-30, reverse). Glyceraldehyde-3-phos- effects of all TORKinibs tested in both H460 and H292 cells phate dehydrogenase was used as an internal control. (Fig. 1C and D). Collectively, these results strongly suggest that GSK3 inhibition impairs the ability of TORKinibs to suppress Lung cancer xenografts and treatments cancer cell growth. Lung cancer xenograft experiments were approved by the To validate our above finding in vivo, we conducted two Institutional Animal Care and Use Committee of Emory Uni- human NSCLC xenograft (H460 and A549) experiments in nude versity and conducted as previously described (25). Five- to 6- mice. In both models, INK128 alone at 1 mg/kg (orally) week-old (about 20 g of body weight) female athymic (nu/nu) effectively inhibited the growth of xenografts. SB216763 at 5 mice were ordered from Harlan. Mice (n ¼ 5 or 6/group) mg/kg (i.p.) did not significantly inhibit the growth of xeno- received the following treatments: vehicle control, INK128 grafts. The combination of INK128 and SB216763 treatment fomulated in 5% polyvinylpropyline/15% N-methy-2-pyrroli- was significantly less effective than INK128 alone in inhibiting done/80% water (1 mg/kg/d, oral gavage; ref. 32), SB216763 in the growth of xenografts as measured by both tumor size (Fig. dimethyl sulfoxide (DMSO; 5 mg/kg/d, i.p.; ref. 33), and the 2B) and tumor weight (Fig. 2C). Moreover, the combination combination of INK128 and SB216763. treatment slightly reduced the body weights of mice in comparison with other treatment groups (Supplementary Fig. S2). Immunohistochemistry Hence, it is clear that the presence of SB216763 attenuates the Immunohistochemistry (IHC) on formalin-fixed and paraf- anticancer efficacy of INK128 in vivo. fin-embedded human NSCLC tissues collected by the Emory To ensure that SB216763 indeed antagonizes growth-inhib- University/Winship Cancer Institute human tissue procure- itory effects of TORKinibs through GSK3 inhibition, we used ment services core was performed using Cytomation EnVi- specific genetic inhibition of GSK3, including gene knockdown þ sionþDual Link System-HPR (DAB ; Dako North America, Inc.) and knockout, and then examined the impact of these manip- following the standard manufacturer's protocol. Tissues were ulations on cell responses to TORKinibs. siRNA-mediated incubated overnight at 4C with the primary antibody against knockdown of GSK3a or GSK3b was confirmed with Western p-GSK3a/b (S21/9; Cell Signaling Technology, Inc.) at 1:75 blotting (Fig. 3A, left). In a 3-day SRB assay, knockdown of dilution. p-GSK3a/b staining was scored as negative (<10% either GSK3a or GSK3b significantly reduced cell sensitivity to staining) and positive (10% staining) staining, respectively. INK128 and to another mTORKinib, PP242 (Fig. 3A, right). Similar results were also generated with another siRNA that Statistical analysis knocks down the expression of both GSK3a and GSK3b (Fig. The statistical significance of differences between two 3B). Consistently, GSK3a-KO or GSK3b-KO MEFs were signif- groups was analyzed with two-sided unpaired Student t tests icantly less sensitive to both INK128 and PP242 in comparison when the variances were equal or with the Welch corrected t with WT MEFs (Fig. 3C). Together, these data further support test when the variances were not equal by use of GraphPad the notion that GSK3 inhibition impairs cell responses to InStat 3 software (GraphPad Software). Data were examined as TORKinibs. suggested by the same software to verify that the assumptions Furthermore, we asked whether GSK3 activation could for use of the t tests held. Results were considered to be increase cell sensitivity to INK128. To this end, we transfected statistically significant at P < 0.05. a constitutively active form of GSK3b (GSK3bCA) into H1299 cells and then analyzed its impact on cell response to INK128. Results Compared with vector control-transfected cells, GSK3bCA- GSK3 is required for the growth-inhibitory effects of transfected cells were more sensitive to INK128 treatment INK128 and other TORKinibs in NSCLC cells (Fig. 3D). Thus, it seems that increased GSK3 activation, in INK128 is a representative TORKinib and is currently under contrast with GSK3 inhibition, sensitizes cells to INK128. clinical testing (5, 6). In the presence of the GSK3 inhibitor Taking the above results together, we conclude that GSK3 SB216763, we found that the growth suppressive effects of activity is required for the growth-inhibitory effects of INK128 INK128 were substantially attenuated in several NSCLC cell and other TORKinibs, at least in NSCLC cells.

www.aacrjournals.org Cancer Res; 74(9) May 1, 2014 2557

Koo et al.

A INK128 Alone INK128 + SB 5 μmol/L INK128 + SB 7.5 μmol/L INK128 + SB 10 μmol/L

90 90 90 90 H157 H292 H460 A549 75 75 75 75

60 60 60 60

45 45 45 45

30 30 30 30

15 15 15 15 Growth(%) inhibition

0 0 0 0 0 6.25 12.5 25 50 100 0 6.25 12.5 25 50 100 0 6.25 12.5 25 50 100 0 6.25 12.5 25 50 100 INK128 (nmol/L) B C H157 H292 − SB216763 100 100 100 + SB216763 G = 37.8% G = 42.3% 1 1 75 75 75

** DMSO 50 ** 50 50 ** ** ** ** 25 ** 25 ** 25 ** ** *

G = 64.5% G = 65.5% Growth(%) inhibition 1 1 0 0 0

INK128 μ D AZD8055 (nmol/L) Torin 1 (nmol/L) PP242 ( mol/L)

− SB216763 G = 33.7% G = 42.7% 100 100 100 1 1 + SB216763 75 75 75

SB216763 ** 50 ** ** 50 ** ** 50 ** ** * ** G = 35.6% G = 47.3% 25 ** ** 25 25 * 1 1 ** * * ** * * Growth(%) inhibition 0 0 0 INK + SB INK +

AZD8055 (nmol/L) Torin 1 (nmol/L) PP242 (μmol/L)

Figure 1. The presence of the GSK3 inhibitor SB216763 antagonizes the growth-inhibitory effects of INK128 (A) and other TORKinibs (C and D), and blocks INK128-induced G1 arrest (B). A, the given lung cancer cell lines were plated on 96-well cell culture plates and treated the next day with the indicated doses of INK128 alone, SB216763 (SB) alone, or their combinations for 3 days. B, the indicated lung cancer cell lines were treated with DMSO, 100 nmol/L INK128, 10 mmol/L SB216763, or INK128 plus SB216763 for 24 hours and then harvested for cell-cycle analysis with ﬂow cytometry. C and D, H460 (C) or H292 (D) cells were plated on 96-well cell culture plates and treated the next day with the indicated doses of a TORKinib alone, 7.5 mmol/L SB216763 alone, or their combinations for 3 days. After these treatments in A, C, and D, cell numbers were estimated using the SRB assay. Data, means of four replicate determinations; bars, SDs. , P < 0.001; , P < 0.0001 compared with a TORKinib treatment alone using the Student t test.

Basal levels of GSK3 activity are signiﬁcantly associated INK128 against these cells (r ¼ 0.689, P ¼ 0.0064; Fig. 4C), with cell sensitivity to INK128 implying that GSK3 activity is associated with cell sensitivity to We next determined whether basal levels of GSK3 activity INK128. are associated with cell sensitivity to TORKinibs. Here, we To evaluate the prevalence of GSK3 activity in human detected basal levels of p-GSK3 as an indication of inactivated NSCLC tissues, we further conducted IHC to detect p-GSK3 or low GSK3 activity in a panel of NSCLC cell lines by Western in 50 cases of human NSCLC specimens. We found that there blotting (Fig. 4A). These cell lines possessed different sensi- were 41 cases positive for p-GSK3 staining (82%) and only 9 tivities to INK128 as determined in a 3-day growth-inhibitory cases negative for p-GSK3 staining (18%; Fig. 4D). Hence, these assay (Fig. 4B). Correlation analysis showed that high p-GSK3 data suggest that the majority of NSCLC tissues possess low levels were signiﬁcantly associated with high IC50 values of GSK3 activity.

2558 Cancer Res; 74(9) May 1, 2014 Cancer Research

GSK3 Affects Cell Response to mTOR Kinase Inhibitors

A A549 H460 DMSO INK SB SB + INK DMSO INK SB SB + INK

A549 H460 B 1,000 Vehicle 2,800 Vehicle INK128 INK128 2,400 800 SB216763 SB216763 ) 3 ) INK + SB 3 2,000 INK + SB 600 1,600

400 1,200 *** 800 * Tumor size (mm *** *

Tumor size (mm * 200 * ** *** * * * * 400 * ** * 0 0 2824201612840 181614121086420 Treatment time (d) Treatment time (d) C 0.5 A549 1.4 H460 1.2 0.4 1.0 0.3 P = 0.0229 0.8

0.2 P = 0.0005 0.6 0.4 Tumor weight(g)

0.1 Tumor weight(g) 0.2 0.0 0.0 Vehicle INK INK + SB SB Vehicle INK INK + SB SB

Figure 2. The presence of SB216763 impairs the inhibitory effects of INK128 on the colony formation of cancer cells in dishes (A) and the growth of NSCLC xenografts in nude mice (B and C). A, the indicated cell lines were seeded in 12-well plates at a density of approximately 400 cells per well. On the second day, the cells were treated with DMSO, 25 nmol/L INK128, 7.5 mmol/L SB216763, or INK128 plus SB216763. After 9 days, the plates were stained for the formation of cell colonies with crystal violet dye and pictured using a digital camera. B and C, both A549 and H460 xenografts were treated as described in Materials and Methods. Tumor sizes (A) and mouse body weights (C) were measured once every 2 days. Each measurement is a mean SE (n ¼ 6 or 5). At the end of the experiment, tumor xenografts were removed and weighed (B). , P < 0.05; , P < 0.01; , P < 0.001 compared with the INK128 group alone using the Student t test.

INK128 and other TORKinibs decrease cyclin D1 levels press mTOR signaling and cap-dependent translation. In dif- through a GSK3-mediated mechanism, which is not ferent NSCLC cell lines, we found that INK128 was equally associated with their ability to inhibit mTOR signaling effective in decreasing the levels of p-4EBP1 and p-S6, two well- and eIF4F complex formation known readouts of the mTORC1, both in the absence and To understand the mechanism by which GSK3 activity presence of SB216763 (Fig. 5A), indicating that inhibition of regulates cell response to INK128, we ﬁrst determined whether GSK3 does not interfere with the suppression of mTORC1 GSK3 inhibition interferes with the ability of INK128 to sup- signaling by INK128. Furthermore, we compared the effects of

www.aacrjournals.org Cancer Res; 74(9) May 1, 2014 2559

Koo et al.

A 80 80 70 70 60 60 **** siRNA: 50 **** 50 **** **** 40 α 40 **** **** GSK3α/β 30 siCtrl 30 **** β α 20 **** siGSK3 20 **** β **** 10 **** siGSK3 10 *** Growth(%) inhibition Tubulin Growth(%) inhibition 0 0 1 10 100 1,000 0.01 0.1 1 10 INK128 (nmol/L) PP242 (μmol/L)

48 h 120 h B 75 75

**** siRNA: 50 **** 50 **** **** **** α **** α/β **** GSK3 β 25 **** 25 siCtrl **** **** **** siGSK3α/β Tubulin **** Growth(%) inhibition Growth(%) inhibition 0 0 1 10 100 1,000 0.01 0.1 1 10 INK128 (nmol/L) PP242 (μmol/L)

C 100 GSK3α-WT 100 GSK3β-WT D GSK3α-KO GSK3β-KO 75 75 **** *** **** **** 50 50 GSK3βCA **** **** GSK3β **** 25 **** 25 **** **** **** **** Actin

Cell number (% of control) (% Cell number 0 0 1 10 100 1,000 1 10 100 1,000 INK128 (nmol/L) INK128 (nmol/L) 80 α 70 100 GSK3 -WT 100 GSK3β-WT 60 GSK3α-KO β **** *** GSK3 -KO 50 75 75 * 40 **** *** ** 30 **** *** Vector 50 50 **** **** 20 GSK3βCA **** 10 25 25 **** Growth(%) inhibition **** **** **** **** 0 **** 1 10 100 1,000

Cell number (% of control) (% Cell number 0 0 INK128 (nmol/L) 0.1 1 10 0.1 1 10 PP242 (μmol/L) PP242 (μmol/L)

Figure 3. Knockdown (A and B) or knockout of GSK3 (C) and enforced expression of a constitutively activated form of GSK3b (D) alters cell responses to mTORKinibs. A and B, H292 cells were transfected with control (Ctrl), GSK3a, GSK3b, or GSK3a/b siRNA for 24 hours and then reseeded in 96-well plates. After 48 hours, the cells were exposed to different concentrations of INK128 or PP242 as indicated for an additional 3 days. Cell numbers were estimated with the SRB assay. GSK3 knockdown effects at 48 hours or the indicated times after transfection were detected with Western blotting. C, the indicated MEFs were seeded in 96-well plates and the next day treated with different concentrations of INK128 or PP242 as indicated. After 3 days, cell numbers were estimated with the SRB assay. D, H1299 cells were transfected with empty vector or expression plasmid carrying GSK3bCA and after 24 hours, were reseeded in 96-well plates. After 48 hours, the cells were exposed to different concentrations of INK128 as indicated for an additional 3 days. Cell numbers were estimated with the SRB assay. GSK3bCA expression was conﬁrmed with Western blotting (top). Data, means of four replicate determinations; bars, SDs. , P < 0.05; , P < 0.01; , P < 0.001; , P < 0.0001 compared with its matched control using the Student t test.

2560 Cancer Res; 74(9) May 1, 2014 Cancer Research

GSK3 Affects Cell Response to mTOR Kinase Inhibitors

A D H23 EKVX H1944 H1299 H1792 Calu-1 H596 H2228 H460 A549 H322 H157 H1651 H292 p-GSK3α/β (S21/9)

GSK3α/β

GAPDH B 2.5 SQCC 2.0 1.5 1.0 0.5

p-GSK3/GSK3 0 60 50 40

30 (82%) 41/50 staining: Positive (nmol/L) 20 50 10 IC 0 ADC

C 60 r = 0.689 (P = 0.0064) 50 40 ; nmol/L) 50 30 20 10 INK128 (IC ADC (18%) 9/50 Negative: 0 0 0.5 1.0 1.5 2.0 2.5 p-GSK3/GSK3

Figure 4. Detection of basal levels of p-GSK3 in human NSCLC cells (A) and tissues (D) and its correlation with cell sensitivity to INK128 (B and C). Whole-cell protein lysates were prepared from the listed cell lines and subjected to Western blotting for detection of the indicated proteins (A). The intensities of these proteins were quantiﬁed with NIH ImageJ software. The IC50 values of INK128 were determined from growth curves of the indicated cell lines exposed to different concentrations of INK128 for 3 days. The correlation between p-GSK3/GSK3 and IC50 values was calculated with GraphPad InStat software (B and C). p-GSK3 in human NSCLC tissues (D) was detected with IHC. SQCC, squamous cell carcinoma; ADC, adenocarcinoma.

INK128 on cap-binding of the eIF4F complex in the absence (Fig. 5C). In agreement, knockdown of GSK3 rescued cyclin D1 and presence of SB216763. Again, INK128 effectively reduced reduction induced by INK128 in both A549 and H460 cell lines the amounts of eIF4G bound to eIF4E with increased amounts (Fig. 5D). These data clearly indicate that GSK3 is required for of 4EBPI bound to eIF4E regardless of the presence or absence INK128 to decrease cyclin D1 levels. of SB216763 (Fig. 5B), suggesting that inhibition of GSK3 INK128 under the tested conditions (100 nmol/L for 10 also does not impair the suppression of cap-dependent trans- hours) reduced p-Akt (S473) levels in the tested three cell lines. lation initiation by INK128. Interestingly, INK128 effectively When combined with SB216763, reduction of p-Akt levels was decreased the levels of cyclin D1, an oncogenic protein known not altered (Fig. 5A). Thus, GSK3 inhibition does not interfere to be regulated by cap-dependent translation. Cotreatment of with the suppression of Akt phosphorylation (S473) by INK128. the cells with SB216763 and INK128 prevented cyclin D1 Interestingly, we noted that INK128 did not suppress GSK3a/b reduction by INK128 in all of the tested cell lines (Fig. 5A). phosphorylation (S21/9) in either of the tested cell lines, INK128 also reduced cyclin D1 levels in lung cancer xenografts; despite its potent inhibition of Akt S473 phosphorylation (Fig. this effect could be prevented by cotreatment with SB216763 5A). Hence, INK128 has no effect on activation of GSK3.

www.aacrjournals.org Cancer Res; 74(9) May 1, 2014 2561

Koo et al.

A H157 A549 H460 B H460 SB216763: −−++ −−++−−++ INK128: −+−+ −+−+−+−+ SB216763: − − + + INK128: − + − + p-Akt (S473) eIF4G Akt GTP

4EBP1 7

p-4EBP1 (T37/46) m eIF4E p-4EBP1 (S65)

eIF4G 4EBP1

p-S6 (S235/S236) 4EBP1 Lysate

S6 eIF4E

Cyclin D1 C Actin Vehicle INK128 SB216763 INK/SB

p-GSK3α/β (S21/9) Cyclin D1

Tubulin GSK3α/β

Tubulin E D siRNA: Ctrl GSK3α/β SB216763: − − + + SB216763: − − + + SB216763: − − + + INK128: − + − + AZD8805: − + − + Torin 1: − + − + PP242: − + − + α/β Cyclin D1 Cyclin D1 Cyclin D1 GSK3

Tubulin Tubulin Tubulin Cyclin D1 A549 Tubulin p-Akt (S473) p-Akt (S473) p-Akt (S473)

Akt Akt Akt GSK3α/β p-GSK3α/β p-GSK3α/β p-GSK3α/β (S21/9) (S21/9) Cyclin D1

(S21/9) H460 GSK3α/β GSK3α/β GSK3α/β Tubulin

Tubulin Tubulin Tubulin

Figure 5. Inhibition of GSK3 with SB216763 (A, C, and E) or siRNA (D) rescues cyclin D1 reduction induced by INK128 (A, C, and D) or other TORKinibs (E) without blocking INK128-mediated suppression of mTOR signaling (A and E) and cap binding (B). A and B, the indicated cell lines were treated with DMSO, 100 nmol/L INK128, 10 mmol/L SB216763 (SB), or INK128 plus SB216763 for 10 hours and then harvested for preparation of whole-cell protein lysates and subsequent Western blotting (A). Moreover, lysates from H460 cells were also used for the m7GTP pull-down and subsequent detection of the given proteins by Western blot analysis (B). C, whole-cell protein lysates were prepared from H460 xenografts described in Fig. 2 and subjected to Western blotting for detection of the indicated proteins. D, the indicated cell lines were transfected with the given siRNAs, and after 48 hours, were exposed to 100 nmol/L INK128 for an additional 4 hours. The cells were harvested for preparation of whole-cell protein lysates and subsequent Western blotting. E, A549 cells were treated with 100 nmol/L AZD8055, 100 nmol/l Torin 1, or 1 mmol/L PP242 in the absence and presence of 10 mmol/L SB216763 for 10 hours and then harvested for preparation of whole-cell protein lysates and subsequent Western blotting. Ctrl, control.

INK128 and other TORKinibs decrease cyclin D1 levels INK128 on cyclin D1 expression. INK128-induced cyclin D1 through promoting its degradation reduction could be detected at 3 hours and was sustained up to Because INK128 effectively decreased cyclin D1 levels, which 12 hours. At concentration ranges between 25 and 200 nmol/L, could be prevented by GSK3 inhibition as demonstrated above, INK128 decreased the levels of cyclin D1 (Supplementary Fig. we conducted experiments to further explore the effect of S3A and S3B). Thus, it is clear that INK128 rapidly and potently

2562 Cancer Res; 74(9) May 1, 2014 Cancer Research

GSK3 Affects Cell Response to mTOR Kinase Inhibitors

decreases cyclin D1 levels. Several other TORKinibs, includ- Indeed, INK128 failed to decrease cyclin D1 levels in the ing AZD8055, Torin 1, and PP242, effectively reduced the presence of MLN4924 (Supplementary Fig. S5). Moreover, we levels of cyclin D1 as well (Supplementary Fig. S3C); these used shRNA to stably knockdown FBX4 and then examined its effects were also GSK3-dependent because inhibition of impact on INK128-induced cyclin D1 reduction. Because of the GSK3 with SB216763 abrogated the ability of these TORKi- lack of suitable antibodies to detect endogenous FBX4, we used nibs to decrease cyclin D1 (Fig. 5E). Hence, GSK3-dependent reverse transcription PCR (RT-PCR) to detect FBX4 expression reduction of cyclin D1 is a general phenotype caused by at the mRNA level to demonstrate its knockdown efficiency TORKinibs. Here, we again observed that these TORKinibs (Fig. 6E). We found that INK128 reduced cyclin D1 levels decreased p-Akt (S473) levels without suppressing GSK3a/b effectively in pLKO.1 cells, but not in two cell lines expressing phosphorylation (S21/9). Moreover, SB216763 did not affect FBX4 shRNA (shFBX4#1 and shFBX4#5). Similar results were the ability of the TORKinibs to inhibit Akt phosphorylation generated when the cell lines were treated with AZD8055 (Fig. (Fig. 5E). We noted that SB216763 reduced p-GSK3b (S9) 6F). These results indicate that FBX4 is involved in mediating levels in the tested cell lines (Fig. 5A and E) although we do TORKinib-induced cyclin D1 degradation. not know the underlying mechanism. By examining the effects of INK128 or AZD8055 on cyclin mTORC2, but not mTORC1, is responsible for positive D1 reduction in different NSCLC cell lines, we found that regulation of cyclin D1 stability both INK128 and AZD8055 decreased cyclin D1 levels more To understand which mTORC is responsible for the regu- effectively in three sensitive cell lines (H460, A549, and lation of cyclin D1 stability, we inhibited mTORC1 and H1648) than in three relatively less sensitive cell lines mTORC2 by knocking down raptor or rictor, respectively, and (Calu-1, H23, and EKVX). Interestingly, the basal levels of then looked at their impact on cyclin D1 levels. We found that cyclin D1 were higher in the less sensitive than in the siRNA-mediated transient knockdown of both raptor and sensitive cell lines. However, both agents effectively sup- rictor decreased cyclin D1 levels in H157 cells. In A549 cells, pressed the phosphorylation of S6 and Akt in all of the tested transient knockdown of rictor, but not raptor, resulted in cyclin cell lines regardless of the level of cell sensitivity (Supple- D1 reduction (Fig. 7A). Similarly, we also detected reduced mentary Fig. S4). These data indicate that the inhibitory cyclin D1 levels in H157 cells in which raptor or rictor expres- effects of TORKinibs on cyclin D1 are not associated with sion was stably knocked down with shRNA (Fig. 7B). These their abilities to inhibit either mTORC1 (e.g., p-S6) or results indicate that inhibition of both mTORC1 and mTORC2 mTORC2 (e.g., p-Akt) signaling. reduces cyclin D1 levels. Given that GSK3 inhibition blocks the reduction of cyclin To further demonstrate the role of mTORCs in the reg- D1 by INK128 without inhibiting mTORC1 signaling and cap- ulation of cyclin D1 levels, we then compared the effects of dependent translation initiation, we then examined the rictor and raptor knockdown on cyclin D1 stability. As effects of INK128 on cyclin D1 stability. Compared with presented in Fig. 7C, we detected a shorter half-life of cyclin DMSO control, INK128 apparently shortened the half-life of D1 in H157-shRictor cells (about 15 minutes) than that in cyclin D1 (Fig. 6A), indicating that INK128 decreases the H157-scramble cells (about 30 minutes), meaning that cyclin stability of cyclin D1 protein. In the presence of the protea- D1 is degraded faster in H157-shRictor cells than in H157- some inhibitor MG132, INK128-induced cyclin D1 reduction scramble cells. Thus, it is clear that knockdown of rictor was prevented in all of the tested cell lines, including H460, decreases cyclin D1 stability. In contrast, H157-scramble and H157, and A549 (Fig. 6B), suggesting that INK128 facilitates H157-shRaptor cell lines had comparable cyclin D1 degra- cyclin D1 degradation. In agreement, INK128 also increased dation rates, indicating that raptor knockdown does not cyclin D1 ubiquitination (Fig. 6C), a required step for protea- alter cyclin D1 stability. some-dependent protein degradation. Collectively, we con- In our previous study, we have shown that enforced expres- clude that INK128 decreases cyclin D1 levels through pro- sion of ectopic rictor interacts with endogenous mTOR and moting its degradation. Consistently, inhibition of the pro- increases Akt phosphorylation (38). To provide additional teasome with MG132 rescued cyclin D1 reduction induced by evidence in support of the role of the mTORC2 in regulation other TORKinibs (Fig. 6D). Therefore, induction of cyclin D1 of cyclin D1 stability, we asked whether increasing mTORC2 degradation is not specific to INK128, and is rather a general activity by enforcing expression of ectopic rictor enhances phenomenon caused by TORKinibs. cyclin D1 stability. Hence, we coexpressed myc-rictor with Flag-cyclin D1 or vector with Flag-cyclin D1 and then com- The ubiquitin E3 ligase FBX4 is involved in mediating pared cyclin D1 stabilities. As shown in Fig. 7D, enforced TORKinib-induced cyclin D1 degradation expression of rictor substantially slowed down the cyclin D1 The F-Box ubiquitin E3 ligase FBX4 has been suggested to degradation rate in comparison with the vector control, indi- mediate GSK3-dependent cyclin D1 degradation (21, 34–36). cating that rictor expression stabilizes cyclin D1 protein. In Because cullin neddylation is essential for activating the SCF contrast, enforced expression of raptor did not affect cyclin D1 E3 ligase complex (consisting of Skp1, cullins, F-box proteins, degradation rate. These results, thus, provide complementary and RING box protein/regulator of Cullins ring finger pro- evidence supporting the notion that mTORC2 indeed positive- teins), through which F-box E3 ligases function (37), we then ly regulates cyclin D1 stability. tested whether inhibition of neddylation by the NEDD8 inhib- Moreover, we found that cyclin D1 reduction caused by itor MLN4924 prevented cyclin D1 reduction by INK128. rictor knockdown could be rescued by siRNA-mediated

www.aacrjournals.org Cancer Res; 74(9) May 1, 2014 2563

Koo et al.

100 A DMSO INK128 DMSO Time after CHX (h): 0 15 30 45 60 75 0 15 30 45 60 75 75 INK128

Cyclin D1 50

25 (% of 0 time) 0 of (%

Tubulin levels D1 Cyclin 0 0 153045607590 BC Time after CHX (min) DMSO MG132 DMSO MG132 INK128: − + − + INK128: − + − + Cyclin D1 D MG132: − − + +

Tubulin H460 AZD8055: − + − + WB: HA (Ub) Cyclin D1

Cyclin D1 D1 Ub-cyclin Tubulin Tubulin H157 WB: Flag MG132: − − + + IP: Anti-Flag (cyclin IP: Anti-Flag (cyclin D1) Cyclin D1 (Cyclin D1) Torin 1: − + − +

A549 DMSO MG132 Tubulin Cyclin D1 INK128: − + − + Tubulin E A549 − − + + shFBX4 MG132: PP242: − + − + pLKO.1 1 2 3 4 5 WB: Ub Cyclin D1 Cyclin D1 Ub-cyclin D1 Ub-cyclin

Tubulin WB Tubulin IP: Anti-cyclin D1 IP: Anti-cyclin WB: cyclin D1 FBX4

A549 A549

GAPDH RT-PCR F pLKO.1 shFBX4#1 shFBX4#5 pLKO.1 shFBX4#1 shFBX4#5 INK128: − + − + − + AZD8055: − + − + − +

Cyclin D1 Cyclin D1

Tubulin Tubulin

Figure 6. INK128 (A–C) and other TORKinibs (D) decrease cyclin D1 levels through facilitating its degradation involving FBX4 (E and F). A, A549 cells were treated with DMSO or 100 nmol/L of INK128 for 6 hours. The cells were then washed with PBS three times and re-fed with fresh medium containing 10 mg/mL CHX. At the indicated times, the cells were harvested for preparation of whole-cell protein lysates and subsequent Western blot analysis. Protein levels were quantiﬁed with NIH ImageJ Software and were normalized to tubulin. B, the indicated cell lines were pretreated with 10 mmol/L MG132 for 30 minutes and then cotreated with 100 nmol/L INK128 for 6 hours. C, H1299 cells were cotransfected with HA-ubiquitin and Flag-cyclin D1 plasmids (top) or transfected with HA-ubiquitin plasmid alone (bottom) using Lipofectamine 2000 reagent. After 24 hours, the cells were then pretreated with 20 mmol/L MG132 for 30 minutes and then cotreated with 100 nmol/L INK128 for another 4 hours. Whole-cell protein lysates were then prepared for immunoprecipitation (IP) using anti-Flag or cyclin D1 antibody followed by Western blotting (WB) using the indicated antibodies. D, A549 cells were pretreated with 10 mmol/L MG132 for 60 minutes and then cotreated with 100 nmol/L AZD8055, 100 nmol/L Torin 1, or 1 mmol/L PP242 for an additional 4 hours. E, whole-cell protein lysates and total cellular total RNA were prepared from the indicated stable cell lines and used for detection of cyclin D1 and FBX4 expression by Western blot analysis and RT-PCR, respectively. F, the indicated cell lines were treated with 100 nmol/L INK128 for 8 hours. After the aforementioned treatments in B, D, and F, the cells were harvested for preparation of whole-cell protein lysates and subsequent Western blotting to detect the given proteins.

knockdown or inhibition of GSK3 (Fig. 7E) and by shRNA- FBX4-dependent. These data provide strong support for the mediated knockdown of FBX4 (Fig. 7F), demonstrating that notion that mTORC2 negatively regulates GSK3-dependent rictor knockdown–mediated cyclin D1 is also GSK3- and and FBX4-mediated cyclin D1 degradation.

2564 Cancer Res; 74(9) May 1, 2014 Cancer Research

GSK3 Affects Cell Response to mTOR Kinase Inhibitors

A H157 C 100 Scramble Scramble shRictor 75 shRictor Time (min): 0 15 30 45 60 75 0 15 30 45 60 75 50 Cyclin D1 Rictor 25 (% of 0 time) 0 of (%

Tubulin levels D1 Cyclin 0 Raptor 0 1530456075 Time after CHX (min) Cyclin D1 Scramble shRaptor 100 Scramble GAPDH shRaptor Time (min): 0 15 30 45 60 75 0 15 30 45 60 75 75

A549 Cyclin D1 50 25

Tubulin time) 0 of (%

Cyclin D1 levels D1 Cyclin 0 0 1530456075 Time after CHX (min) Rictor D Raptor Vector + cyclin D1 Rictor + cyclin D1 100 75 Cyclin D1 Time (min): 0 15 30 45 60 0 15 30 45 60 50 Tubulin Cyclin D1 Vector 25 Rictor Rictor time) 0 of (%

Cyclin D1 levels levels D1 Cyclin 0 H157 0 15304560 B Tubulin Time after CHX (min)

Vector + Cyclin D1 Raptor + Cyclin D1 100 Vector Rictor Raptor Time (min): 0 15 30 45 60 0 15 30 45 60 75 Raptor 50 Cyclin D1 Cyclin D1 25 (% of 0 time) 0 of (%

Raptor levels D1 Cyclin 0 Tubulin 015304560 Tubulin Time after CHX (min) EFA549 siGSK3α/β: −− ++ A549 siRictor: −+ −+ pLKO.1 shFBX4#1 shFBX4#5 GSK3α siCtrl: +− +− +− siRictor: −+ −+ −+ GSK3β Rictor Rictor

Cyclin D1 Cyclin D1

Tubulin Tubulin

Figure 7. Genetic manipulation of rictor expression alters cyclin D1 stability (A–D) through a GSK3- and FBX4-dependent mechanism (E and F). A and B, whole-cell protein lysates were prepared from the indicated cell lines 48 hours after transfection with control (Ctrl), rictor or raptor siRNAs (A) or from the stable transfectants carrying scrambled, rictor or raptor shRNA. Western blot analysis was used to detect the indicated proteins. C, the indicated H157 stable transfectants were exposed to 10 mg/mL CHX. At the indicated times, the cells were harvested for preparation of whole-cell protein lysates and subsequent Western blot analysis. Protein levels were quantiﬁed with NIH ImageJ Software and were normalized to tubulin. D, HEK293T cells were cotransfected with cyclin D1 and rictor expression plasmids. After 48 hours, the cells were exposed to CHX as in C for the cyclin D1 stability assay. E and F, the indicated cell lines were transfected with the indicated siRNAs alone or their combination, and after 48 hours, were harvested for preparation of whole-cell protein lysates and subsequent Western blotting.

www.aacrjournals.org Cancer Res; 74(9) May 1, 2014 2565

Koo et al.

Discussion interfering with the ability of INK128 to inhibit mTORC1 This study provides the first preclinical evidence that GSK3 signaling (e.g., suppression of p-S6 and p-4EBP1) and eIF4F activity affects cell response or sensitivity to mTORKinibs based complex formation (Fig. 5B); and (ii) both INK128 and on the following findings: (i) the presence of the GSK3 inhibitor, AZD8055 preferentially and potently decreased cyclin D1 levels SB216763, antagonizes the suppression of NSCLC cell growth by in TORKinib-sensitive NSCLC cell lines (e.g., H460, A549, and INK128 and other TORKinibs both in vitro and in vivo (Figs. 1 H1648), but had comparable effects on inhibiting mTORC1 and 2); (ii) SB216763 abrogates the ability of INK128 to induce signaling (e.g., S6 phosphorylation), irrespective of cell sensi- a G1 arrest (Fig. 1B); (iii) knockdown or knockout of GSK3 , tivities to TORKinibs (Supplementary Fig. S4). GSK3b, or both attenuates the ability of INK128 or PP242 to Protein degradation is another important mechanism that inhibit cell growth (Fig. 3A–C); (iv) expression of constitutively regulates cyclin D1 levels. GSK3 has been suggested to be active GSK3bCA sensitizes cancer cells to INK128 (Fig. 3D); and involved in this process (20, 44). In this study, we have (v) low basal GSK3 activity (i.e., high basal p-GSK3 levels) is demonstrated that INK128 and other TORKinibs reduce significantly associated with low sensitivity of NSCLC cells to cyclin D levels through facilitating its proteasomal degra- INK128 (Fig. 4). Thus, our findings warrant further study of the dation. This conclusion is strongly supported by our findings potential impact of GSK3 activity on the therapeutic outcomes that INK128 and other TORkinibs enhanced cyclin D1 ubi- of TORKinibs both preclinically and clinically. quitination, induced cyclin D1 proteasomal degradation, and It is well known that GSK3 is inactivated through phos- decreased cyclin D1 stability as presented in Fig. 6. In phorylation by Akt, which is usually hyperactivated in many addition, we have further demonstrated that the SCF E3 types of cancer cells (39). Previous studies have shown that ligase FBX4 is involved in mediating cyclin D1 degradation Akt phosphorylation (or activation) is positive in 50% to 80% induced by TORKinibs primarily because knockdown of of NSCLC tumors (40–42). In our preliminary or pilot study, FBX4 prevented cyclin D1 reduction by INK128 or AZD8055 GSK3 phosphorylation was positive in approximately 80% of (Fig.6).Collectively,weconcludethatTORKinibssuchas NSCLC tumors (Fig. 4). Because low basal GSK3 activity (i.e., INK128 induce GSK3-dependent, FBX4-mediated proteaso- high p-GSK3 levels) was significantly associated with low mal degradation of cyclin D1, resulting in decreased cyclin sensitivity of NSCLC cells to INK128, our prediction is that D1 levels. Accordingly, we suggest a novel function that the subset of NSCLCs (approximately 20% based on our pilot mTOR signaling positively regulates cyclin D1 levels through study shown in Fig. 4) negative for p-GSK3 staining (i.e., preventing its degradation. having activated GSK3) may represent a population who Following this important finding, we have further shown may respond well to TORKinibs. Whether this implies that that mTORC2 is in fact responsible for the positive regulation TORKinibs may not be very active in cancer types with of cyclin D1 stability through preventing GSK3-dependent hyperactivated Akt or whether GSK3 activity can be used and FBX4-mediated cyclin D1 degradation. In our study, as a predictive biomarker for TORKinib-based cancer ther- although knockdown of both raptor and rictor reduced cyclin apy is worthy of further investigation. D1 levels, knockdown of rictor, but not raptor, enhanced Cyclin D1 is an oncogenic protein that promotes cell-cycle cyclin D1 degradation or decreased its stability (Fig. 7). progression through the G1 phase by forming active holoen- Moreover, enforced expression of ectopic rictor, but not zymes with cyclin-dependent kinase 4 and 6 (20). In our study, raptor, substantially stabilized cyclin D1 and rictor-induced INK128 at the tested concentration ranges (e.g., 100 nmol/L) cyclin D1 reduction could be prevented by knockdown of primarily induced G1 arrest with limited effect on induction of GSK3 or FBX4 (Fig. 7). Cyclin D1 reduction induced by raptor apoptosis. Moreover, INK128 and other TORKinibs effectively knockdown is likely to be the consequence of suppression of decreased cyclin D1 levels particularly in NSCLC cell lines cap-dependent translation of cyclin D1 due to inhibition of sensitive to TORKinibs (Supplementary Figs. S3 and S4). Inhi- the mTORC1. To the best of our knowledge, this is the first bition of GSK3 with SB216763 or siRNA-mediated knockdown study revealing that the mTORC2, like the mTORC1, also prevented cyclin D1 reduction induced by INK128 (Fig. 5) and positively regulates cyclin D1 levels, but through a different abrogated the ability of INK128 to induce G1 arrest (Fig. 1B). mechanism (i.e., stabilizing protein). Given that the mTORC2 Consistently, other TORKinibs also induced GSK3-dependent has been implicated in tumorigenesis (45, 46), our findings cyclin D1 reduction (Fig. 5E). Hence, cyclin D1 reduction is warrant further study of the role of cyclin D1 in mTORC2- likely to be a key event that mediates the growth-inhibitory mediated oncogenesis. effects of INK128 or TORKinibs. Accordingly, the blockage of Rictor has been shown to be involved in the regulation of cyclin D1 reduction may be critical in mediating the antago- protein degradation (e.g, SGK1 and cyclin E1) independent of nistic effect of GSK3 inhibition on the therapeutic efficacy of mTOR (47, 48). Our findings differ from these in clearly showing TORKinibs in cancer cells. an mTOR-dependent regulation of cyclin D1 degradation. One Cyclin D1 is known to be regulated by mTORC1 signaling significant implication of our findings is to suggest a new role through cap-dependent translation (3, 43). An intriguing and for mTORC2 in the regulation of protein (e.g., cyclin D1) stability novel finding in this study is that TORKinibs reduce cyclin D1 or degradation. Further study in this direction will shed light on levels through a GSK3-dependent mechanism independent of the biology of the mTORC2. Although Akt may positively inhibition of mTORC1 signaling and cap-dependent transla- regulate cyclin D1 stability through inactivation of GSK3 (34), tion based on the following results: (i) inhibition of GSK3 with cyclin D1 reduction induced by TORKinibs is unlikely to be a SB216763 prevented cyclin D1 reduction by INK128 without consequence of Akt inhibition because our data clearly show

2566 Cancer Res; 74(9) May 1, 2014 Cancer Research

GSK3 Affects Cell Response to mTOR Kinase Inhibitors

that inhibition of GSK3 with SB216763 rescued TORKinib- Disclosure of Potential Conflicts of Interest induced cyclin D1 reduction without impairing the ability of No potential conflicts of interest were disclosed. TORKinibs to suppress Akt phosphorylation in any of the tested cell lines (Fig. 5A and E). In addition, both INK128 and AZD8055 Authors' Contributions Conception and design: J. Koo, F.R. Khuri, S.-Y. Sun effectively inhibited Akt phosphorylation regardless of cell Development of methodology: J. Koo, F.R. Khuri sensitivities to TORKinibs, but potently reduced cyclin D1 levels Acquisition of data (provided animals, acquired and managed patients, preferentially in TORKinib-sensitive cell lines (Supplementary provided facilities, etc.): J. Koo, A.A. Gal Analysis and interpretation of data (e.g., statistical analysis, biostatistics, Fig. S4). Moreover, none of the tested TORKinibs could reduce computational analysis): J. Koo, A.A. Gal, F.R. Khuri, S.-Y. Sun p-GSK3a/b (S21/9) levels despite their potent suppression of Writing, review, and/or revision of the manuscript: J. Koo, A.A. Gal, F.R. Khuri, S.-Y. Sun Akt S473 phosphorylation (Fig. 5); this result is in line with a Administrative, technical, or material support (i.e., reporting or orga- previous report showing that SIN1 knockout induced suppres- nizing data, constructing databases): J. Koo, P. Yue sion of mTORC2, which abolished Akt S473 phosphorylation Study supervision: S.-Y. Sun without affecting GSK3 phosphorylation (49). Our ongoing work is attempting to understand the mechanism by which mTORC2 Acknowledgments The authors thank Drs. Binhua P. Zhou, Alan Diehl, and Jim Woodgett for positively regulates cyclin D1 stability. providing us with plasmids or cell lines used in this work. The authors also thank In summary, this study has demonstrated the critical role of Dr. A. Hammond in our department for editing the article. GSK3 activation in determining the response of cancer cells to TORKinibs, and thus has high translational significance and Grant Support This work was supported by NIH R01 CA118450 (S.-Y. Sun), R01 CA160522 impact in regard to cancer treatment with TORKinibs. Our (S.-Y. Sun), and P01 CA116676 (project 1; F.R. Khuri and S.-Y. Sun). finding of the regulation of GSK3-dependent and FBX4-medi- The costs of publication of this article were defrayed in part by the payment ated cyclin D1 degradation by mTORC2 suggests a novel of page charges. This article must therefore be hereby marked advertisement biologic function of the mTORC2 and increases our under- in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. standing of the involvement of the mTORC2 in the regulation Received October 11, 2013; revised February 20, 2014; accepted February 24, of cancer development. 2014; published OnlineFirst March 13, 2014.

References 1. Guertin DA, Sabatini DM. Defining the role of mTOR in cancer. Cancer 14. Jope RS, Yuskaitis CJ, Beurel E. Glycogen synthase kinase-3 (GSK3): Cell 2007;12:9–22. inflammation, diseases, and therapeutics. Neurochem Res 2007;32: 2. Oh WJ, Jacinto E. mTOR complex 2 signaling and functions. Cell Cycle 577–95. 2011;10:2305–16. 15. Cohen P, Goedert M. GSK3 inhibitors: development and therapeutic 3. Wang X, Sun SY. Enhancing mTOR-targeted cancer therapy. Expert potential. Nat Rev Drug Discov 2004;3:479–87. Opin Ther Targets 2009;13:1193–203. 16. Mills CN, Nowsheen S, Bonner JA, Yang ES. Emerging roles of 4. Abraham RT, Gibbons JJ. The mammalian target of rapamycin sig- glycogen synthase kinase 3 in the treatment of brain tumors. Front naling pathway: twists and turns in the road to cancer therapy. Clin Mol Neurosci 2011;4:47. Cancer Res 2007;13:3109–14. 17. Mishra R. Glycogen synthase kinase 3 beta: can it be a target for oral 5. Sun SY. mTOR kinase inhibitors as potential cancer therapeutic drugs. cancer. Mol Cancer 2010;9:144. Cancer Lett 2013;340:1–8. 18. Welcker M, Clurman BE. FBW7 ubiquitin ligase: a tumour suppressor 6. Zhang YJ, Duan Y, Zheng XF. Targeting the mTOR kinase domain: the at the crossroads of cell division, growth and differentiation. Nat Rev second generation of mTOR inhibitors. Drug Discov Today 2011;16: Cancer 2008;8:83–93. 325–31. 19. Inuzuka H, Fukushima H, Shaik S, Liu P, Lau AW, Wei W. Mcl-1 7. Feldman ME, Apsel B, Uotila A, Loewith R, Knight ZA, Ruggero D, et al. ubiquitination and destruction. Oncotarget 2011;2:239–44. Active-site inhibitors of mTOR target rapamycin-resistant outputs of 20. Takahashi-Yanaga F, Sasaguri T. GSK-3beta regulates cyclin D1 mTORC1 and mTORC2. PLoS Biol 2009;7:e38. expression: a new target for chemotherapy. Cell Signal 2008;20:581–9. 8. Yu K, Toral-Barza L, Shi C, Zhang WG, Lucas J, Shor B, et al. 21. Jia L, Sun Y. F-box proteins FBXO31 and FBX4 in regulation of cyclin Biochemical, cellular, and in vivo activity of novel ATP-competitive D1 degradation upon DNA damage. Pigment Cell Melanoma Res and selective inhibitors of the mammalian target of rapamycin. Cancer 2009;22:518–9. Res 2009;69:6232–40. 22. Inoki K, Ouyang H, Zhu T, Lindvall C, Wang Y, Zhang X, et al. TSC2 9. Thoreen CC, Kang SA, Chang JW, Liu Q, Zhang J, Gao Y, et al. An integrates Wnt and energy signals via a coordinated phosphorylation ATP-competitive mammalian target of rapamycin inhibitor reveals by AMPK and GSK3 to regulate cell growth. Cell 2006;126:955–68. rapamycin-resistant functions of mTORC1. J Biol Chem 2009;284: 23. Shin S, Wolgamott L, Yu Y, Blenis J, YoonSO. Glycogen synthase kinase 8023–32. (GSK)-3 promotes p70 ribosomal protein S6 kinase (p70S6K) activity and 10. Garcia-Martinez JM, Moran J, Clarke RG, Gray A, Cosulich SC, cell proliferation. Proc Natl Acad Sci U S A 2011;108:E1204–13. Chresta CM, et al. Ku-0063794 is a specific inhibitor of the mammalian 24. Zhou BP, Deng J, Xia W, Xu J, Li YM, Gunduz M, et al. Dual regulation of target of rapamycin (mTOR). Biochem J 2009;421:29–42. Snail by GSK-3beta-mediated phosphorylation in control of epithelial– 11. Hoang B, Frost P, Shi Y, Belanger E, Benavides A, Pezeshkpour G, mesenchymal transition. Nat Cell Biol 2004;6:931–40. et al. Targeting TORC2 in multiple myeloma with a new mTOR kinase 25. Wang X, Yue P, Kim YA, Fu H, Khuri FR, Sun SY. Enhancing mam- inhibitor. Blood 2010;116:4560–8. malian target of rapamycin (mTOR)-targeted cancer therapy by pre- 12. Chresta CM, Davies BR, Hickson I, Harding T, Cosulich S, Critchlow venting mTOR/raptor inhibition-initiated, mTOR/rictor-independent SE, et al. AZD8055 is a potent, selective, and orally bioavailable ATP- Akt activation. Cancer Res 2008;68:7409–18. competitive mammalian target of rapamycin kinase inhibitor with in 26. Sun SY, Yue P, Dawson MI, Shroot B, Michel S, Lamph WW, et al. vitro and in vivo antitumor activity. Cancer Res 2010;70:288–98. Differential effects of synthetic nuclear retinoid receptor-selective 13. Frame S, Cohen P. GSK3 takes centre stage more than 20 years after retinoids on the growth of human non–small cell lung carcinoma cells. its discovery. Biochem J 2001;359:1–16. Cancer Res 1997;57:4931–9.

www.aacrjournals.org Cancer Res; 74(9) May 1, 2014 2567

Koo et al.

27. Sun SY, Zhou Z, Wang R, Fu H, Khuri FR. The farnesyltransferase 38. Zhao L, Yue P, Khuri FR, Sun SY. mTOR complex 2 is involved in inhibitor lonafarnib induces growth arrest or apoptosis of human lung regulation of Cbl-dependent c-FLIP degradation and sensitivity of cancer cells without downregulation of Akt. Cancer Biol Ther 2004;3: TRAIL-induced apoptosis. Cancer Res 2013;73:1946–57. 1092–8. 39. Manning BD, Cantley LC. AKT/PKB signaling: navigating downstream. 28. Sun SY, Rosenberg LM, Wang X, Zhou Z, Yue P, Fu H, et al. Cell 2007;129:1261–74. Activation of Akt and eIF4E survival pathways by rapamycin-medi- 40. Cappuzzo F, Magrini E, Ceresoli GL, Bartolini S, Rossi E, Ludovini V, ated mammalian target of rapamycin inhibition. Cancer Res 2005; et al. Akt phosphorylation and gefitinib efficacy in patients with 65:7052–8. advanced non–small cell lung cancer. J Natl Cancer Inst 2004;96: 29. Liu X, Yue P, Zhou Z, Khuri FR, Sun SY. Death receptor regulation and 1133–41. celecoxib-induced apoptosis in human lung cancer cells. J Natl Can- 41. Tsurutani J, Fukuoka J, Tsurutani H, Shih JH, Hewitt SM, Travis WD, cer Inst 2004;96:1769–80. et al. Evaluation of two phosphorylation sites improves the prognostic 30. Fan S, Li Y, Yue P, Khuri FR, Sun SY. The eIF4E/eIF4G interaction significance of Akt activation in non-small-cell lung cancer tumors. inhibitor 4EGI-1 augments TRAIL-mediated apoptosis through c-FLIP J Clin Oncol 2006;24:306–14. downregulation and DR5 induction independent of inhibition of cap- 42. Yoshizawa A, Fukuoka J, Shimizu S, Shilo K, Franks TJ, Hewitt SM, dependent protein translation. Neoplasia 2010;12:346–56. et al. Overexpression of phospho-eIF4E is associated with survival 31. Liu X, Yue P, Schonthal AH, Khuri FR, Sun SY. Cellular FLICE-inhibitory through AKT pathway in non–small cell lung cancer. Clin Cancer Res protein downregulation contributes to celecoxib-induced apoptosis in 2010;16:240–8. human lung cancer cells. Cancer Res 2006;66:11115–9. 43. Bjornsti MA, Houghton PJ. Lost in translation: dysregulation of cap- 32. Hsieh AC, Liu Y, Edlind MP, Ingolia NT, Janes MR, Sher A, et al. The dependent translation and cancer. Cancer Cell 2004;5:519–23. translational landscape of mTOR signalling steers cancer initiation and 44. Alao JP. The regulation of cyclin D1 degradation: roles in cancer metastasis. Nature 2012;485:55–61. development and the potential for therapeutic invention. Mol Cancer 33. Mai W, Kawakami K, Shakoori A, Kyo S, Miyashita K, Yokoi K, et al. 2007;6:24. Deregulated GSK3{beta} sustains gastrointestinal cancer cells survival 45. Guertin DA, Stevens DM, Saitoh M, Kinkel S, Crosby K, Sheen JH, et al. by modulating human telomerase reverse transcriptase and telome- mTOR complex 2 is required for the development of prostate cancer rase. Clin Cancer Res 2009;15:6810–9. induced by Pten loss in mice. Cancer Cell 2009;15:148–59. 34. Diehl JA, Cheng M, Roussel MF, Sherr CJ. Glycogen synthase kinase- 46. Lee K, Nam KT, Cho SH, Gudapati P, Hwang Y, Park DS, et al. Vital 3beta regulates cyclin D1 proteolysis and subcellular localization. roles of mTOR complex 2 in Notch-driven thymocyte differentiation Genes Dev 1998;12:3499–511. and leukemia. J Exp Med 2012;209:713–28. 35. Lin DI, Barbash O, Kumar KG, Weber JD, Harper JW, Klein-Szanto AJ, 47. Gao D, Wan L, Inuzuka H, Berg AH, Tseng A, Zhai B, et al. Rictor forms a et al. Phosphorylation-dependent ubiquitination of cyclin D1 by the complex with Cullin-1 to promote SGK1 ubiquitination and destruc- SCF(FBX4-alphaB crystallin) complex. Mol Cell 2006;24:355–66. tion. Mol Cell 2010;39:797–808. 36. Barbash O, Zamfirova P, Lin DI, Chen X, Yang K, Nakagawa H, et al. 48. Guo Z, Zhou Y, Evers BM, Wang Q. Rictor regulates FBXW7-depen- Mutations in Fbx4 inhibit dimerization of the SCF(Fbx4) ligase and dent c-Myc and cyclin E degradation in colorectal cancer cells. contribute to cyclin D1 overexpression in human cancer. Cancer Cell Biochem Biophys Res Commun 2012;418:426–32. 2008;14:68–78. 49. Jacinto E, Facchinetti V, Liu D, Soto N, Wei S, Jung SY, et al. SIN1/ 37. Jia L, Sun Y. SCF E3 ubiquitin ligases as anticancer targets. Curr MIP1 maintains rictor-mTOR complex integrity and regulates Akt Cancer Drug Targets 2011;11:347–56. phosphorylation and substrate specificity. Cell 2006;127:125–37.

2568 Cancer Res; 74(9) May 1, 2014 Cancer Research

Maintaining Glycogen Synthase Kinase-3 Activity Is Critical for mTOR Kinase Inhibitors to Inhibit Cancer Cell Growth

Junghui Koo, Ping Yue, Anthony A. Gal, et al.

Cancer Res 2014;74:2555-2568. Published OnlineFirst March 13, 2014.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-13-2946

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2014/03/13/0008-5472.CAN-13-2946.DC1

Cited articles This article cites 49 articles, 17 of which you can access for free at: http://cancerres.aacrjournals.org/content/74/9/2555.full#ref-list-1

Citing articles This article has been cited by 7 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/74/9/2555.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 Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/74/9/2555. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.