Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
Prec
RapExp
DavidAna MMien-
Abst
Intro
Magrowtkaryo(4E-BPlimitinnase 1lationmTOR
Authorand Co5MolecUniversand 7CBiology
Note: SCancer
Corresvard, U4926. E
doi: 10
©2010
Mol C2770
D
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980
Molecular
Cancerapeutics
linical Development
amycin Regulates Stearoyl CoA Desaturase 1
Ther
ression in Breast Cancer
Luyimbazi1, Argun Akcakanat1, Priscilla F. McAuliffe1, Li Zhang2, Gopal Singh1,
aria Gonzalez-Angulo3, Huiqin Chen3, Kim-Anh Do4, Yuhuan Zheng1, Chie Hung5,7, Gordon B. Mills6, and Funda Meric-Bernstam1ractMam
growtmTORrates fthe midentiftotal athreeCoA dmTORproteicline istabiliscriptismallbe reg
. Rapamysignali
s' Affiliationmputationaular andity of Texaenter for M, China Me
upplementTherapeuti
ponding Aunit 444, Hou-mail: fmer
.1158/1535-
American A
ancer Ther
ownloade
malian target of rapamycin (mTOR) signaling is a central regulator of protein translation, cellh, and metabolism. Alterations of the mTOR signaling pathway are common in cancer, makinga promising therapeutic target. In clinical trials, rapamycin analogs have shown modest response
or most cancer types, including breast cancer. Therefore, there is an urgent need to better understandechanism of action of rapamycin to improve patient selection and to monitor pathway inhibition. Toy novel pharmacodynamic markers of rapamycin activity, we carried out transcriptional profiling ofnd polysome-associated RNA in three breast cancer cell lines representing different subtypes. In allcell lines, we found that rapamycin significantly decreased polysome-associated mRNA for stearoyl-esaturase 1 (SCD1), the rate-limiting enzyme in monounsaturated fatty acid synthesis. Activators ofincreased SCD1 protein expression, whereas rapamycin, LY294002, and BEZ235 decreased SCD1
n expression. Rapamycin decreased total SCD1 RNA expression without inducing a significant de-n its relative polysomal recruitment (polysome/total ratio). Rapamycin did not alter SCD1 mRNAty. Instead, rapamycin inhibited SCD1 promoter activity and decreased expression of mature tran-on factor sterol regulatory element binding protein 1 (SREBP1). Eukaryotic initiation factor 4E (eIF4E)interfering RNA (siRNA) decreased both SCD1 and SREBP1 expression, suggesting that SCD1 mayulated through the mTOR/eIF4E-binding protein 1 axis. Furthermore, SCD1 siRNA knockdown in-breast cancer cell growth, whereas overexpression increased growth. Taken together these findings
hibitedshow that rapamycin decreases SCD1 expression, establishing an important link between cell signaling andcancer cell fatty acid synthesis and growth. Mol Cancer Ther; 9(10); 2770–84. ©2010 AACR.
12 (1)to thecal mclinicwith bhavethe grof dev
duction
mmalian target of rapamycin (mTOR) regulates cellh and metabolism. Its best studied targets are eu-tic initiation factor 4E (eIF4E)-binding protein 11), which regulates the availability of eIF4E, a rateg factor for cap-dependent translation, and S6 ki-(S6K1), which regulates ribosomal S6 phosphory-
cin is a macrolide fungicide that inhibitsng by binding to FK506-binding protein
Canparedand mgrowtrateswaysCoAacid sfirst dacyl-Crate-dacids,in turunsaton theAn u
s: Departments of 1Surgical Oncology, 2Bioinformaticsl Biology, 3Breast Medical Oncology, 4Biostatistics,Cellular Oncology, and 6Systems Biology, Thes M.D. Anderson Cancer Center, Houston, Texas;olecular Medicine and Graduate Institute of Cancerdical University and Hospital, Taichung, Taiwan
ary material for this article is available at Molecularcs Online (http://mct.aacrjournals.org/).
thor: Funda Meric-Bernstam, 1400 Holcombe Boule-ston, TX 77030. Phone: 713-745-4453; Fax: [email protected]
7163.MCT-09-0980
ssociation for Cancer Research.
; 9(10) October 2010
on March 15, 2020. © 201mct.aacrjournals.org d from
. Althoughmany breast cancer cell lines are sensitivegrowth-inhibitory effects of rapamycin in preclini-odels (2–4), the overall response rates observed inal studies involving mTOR inhibitors in patientsreast cancer have beenmodest (5–7). These findingsprompted further studies of the mechanism behindowth-inhibitory effects of rapamycin with the goaleloping effective rational combinatorial therapies.cer cells have increased metabolic autonomy. Com-with normal tissues, cancer cells take up nutrientsetabolize them at high rates to support rapidh and proliferation. How cancer cells achieve highof fatty acid synthesis, and how oncogenic path-influence this, is not well understood. Stearoyl-desaturase 1 (SCD1) is a critical mediator of fattyynthesis. SCD1 catalyzes the introduction of theouble bond in the cis-Δ9 position of saturated fattyoAs, such as palmitoyl- and stearoyl-CoA. Thisetermining step produces monounsaturated fattynamely palmitoleoyl- and oleoyl-CoA (8), whichn provide substrates for the development of poly-urated fatty acids. SCD1 therefore has a direct effect
ratio of saturated to monounsaturated fatty acids.nbalanced ratio contributes to altered membrane
0 American Association for Cancer Research.
fluiditincludof lysotidic aprogrof fattcell linsaturaflectinthe cegeneting thFurthferencThus,of fattThe
to inteno accent sadipolulartumorfor inTo
sponscombmicrobreastsubtyor tracauseexpreand itelemeinvolvWe alsmTOReIF4Ebreascreaselink bcanceimpoand sufor in
Mate
Cell lBT-
MCF7MDA75-1 cCultuwerethus wlecula
ty oflines,PCRshortand fAuthetabid/supplhumidchasecell li231 ceSCD1were2 mg/ern bl
ReagRap
chasegrowtvinolwas kcal Rerapamshow
OligoOli
The hand tcontroInc. TpGL3of Wiwas oactiveTsichl4E-BPphoryproviCharl
SucroAft
for 24scribepolys
RNABot
by phsampmanuxenog(Qiag
Rapamycin Regulates SCD1 Expression
www.a
D
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980
y and has been implicated in a variety of diseases,ing cancer (9, 10). Further, the activity of a numberphospholipid growth factors such as lysophospha-cid, which is implicated in the development andession of breast cancer, is altered by the saturationy acyl chains (11). Lipid analyses from transformedes have indicated an increase in the conversion ofted stearic acid to monounsaturated oleic acid, re-g a significant decrease in the saturation index ofll (12, 13). SCD1 is overexpressed in mouse modelsically susceptible to hepatocarcinogenesis, suggest-at SCD1 may play a role in carcinogenesis (14, 15).er, SCD1 has recently been identified in a RNA inter-e screen as a potential target for cancer therapy (16).further work is needed to determine the regulatorsy metabolism and specifically SCD1 in cancer cells.role of mTOR in metabolism is shown by its abilitygrate nutrient availability and energy through ami-id biosynthesis and glucose homeostasis (17). Re-tudies have also pointed to a role for mTOR ingenesis (18–20). Clarifying the role of mTOR in cel-fat metabolism is important in understanding howcells support the accelerated metabolism essential
creased cell proliferation.identify novel pharmacodynamic markers of re-e to rapamycin as well as potential targets forinatorial therapy, we carried out a high-throughputarray polysome analysis of three rapamycin-sensitivecancer cell lines representing the key breast cancerpes to identify genes regulated at the transcriptionalnslational level. We show here that rapamycins a dramatic decrease in SCD1 mRNA and proteinssion. Rapamycin inhibits SCD1 promoter activity,decreases expression of mature sterol regulatorynt binding protein 1 (SREBP1), a transcription factored in fatty acid and cholesterol homeostasis (21).o show that SCD1 is regulated by eIF4E, suggestingmay regulate SCD1 through the mTOR/4E-BP1/axis. Furthermore, suppression of SCD1 inhibitst cancer cell growth, whereas overexpression in-s cell growth. These findings support an importantetween the oncogenic cell signaling of mTOR andr cell fat metabolism. This novel observation isrtant in understanding how tumor cells establishpport the accelerated fatty acid synthesis essential
creased cell proliferation.
rials and Methods
ines and cultures20, BT-474, BT-483, BT-549, HCC1143, HCC1937,, MDA-MB-157, MDA-MB-231, MDA-MB-361,-MB-436, MDA-MB-468, SK-Br-3, T47D, and ZR-ells were obtained from the American Type Tissuere Collection (ATCC) in January 2006. Cell linespassaged for <6 months after resuscitation, and
ere not tested for characterization. The ATCC Mo-r Authentication Resource Center provides a varie-izedchips
acrjournals.org
on March 15, 2020. © 201mct.aacrjournals.org ownloaded from
applications to identify and characterize the cellincluding cloning and gene synthesis, real-timeanalyses, site-directed mutagenesis, sequencing,tandem repeat, single nucleotide polymorphism,ingerprint analyses (http://www.atcc.org/Science/nticationandPreservation/AuthenticationTechnology/209/Default.aspx). Cells were cultured in DMEM/F12emented with 10% fetal bovine serum at 37°C andified 5%CO2. Lipoprotein-deficient serumwas pur-
d from Sigma-Aldrich. Stable SCD1-overexpressingnes and controls were generated using MDA-MB-lls transfected with either myc/DDK-tagged humanORF DNA clone or its control vector. Coloniesthen selected with G418 at a final concentration ofmL, and positive colonies were confirmed by West-ot analysis.
entsamycin, LY294002, and cyclohexamide were pur-d from LC Laboratories, Inc. DMSO, insulin-likeh factor-I (IGF-I), insulin, actinomycin D, and me-in were purchased from Sigma-Aldrich. BEZ235indly provided by Novartis Institutes for BioMedi-search-Novartis Oncology. Chemical structures ofycin (22), LY294002 (23), and BEZ235 (24) are
n in Supplementary Fig. S1.
nucleotides and plasmidsgonucleotides were synthesized by Sigma-Aldrich.uman SCD1 cDNA, the human β-actin cDNA,he myc/DDK-tagged human SCD1 ORF and itsl were all obtained from Origene Technologies,he promoter-reporter gene plasmid hSCD-Lucwas a gift from Dr. James M. Ntambi (Universitysconsin, Madison, WI; ref. 25). The pRL-TK vectorbtained from Promega Corporation. Constitutively(CA) Akt plasmid was provided by Dr. Philip N.is (Fox Chase Cancer Center, Philadelphia, PA).1 wild-type (4E-BP1 WT) and 4E-BP1 five phos-lation sites mutated (4E-BP1 5A) plasmids wereded by Dr. Thurl E. Harris (University of Virginia,ottesville, VA).
se density gradientser cells were treatedwith either rapamycin or controlhours, sucrose density gradients were done as de-d elsewhere (26). The resulting monosomal andomal fractions were pooled and RNAwas extracted.
extraction for microarray analysish monosomal and polysomal RNA were extractedenol and chloroform. Total RNA from all in vitroles was extracted using TRIzol reagent per thefacturer's protocol (Invitrogen). Total RNA fromrafts was extracted using the RNeasy Mini Kiten Inc.). Total and polysomal RNA were hybrid-
to Affymetrix Human Genome U133 Plus 2.0(Affymetrix, Inc.).Mol Cancer Ther; 9(10) October 2010 2771
0 American Association for Cancer Research.
Real-Rea
SequeAssaycludeβ-actinhydroTaqMtems)Detecfold-cendog
WestCel
100 merol, osodiu5 mmleupe1 mmphatearated0.2 μmLaborin TBThe
nologboxylphospS6K1proteiSignasynthBiosciSigmaeither(Li-Cocence
NorthEle
wereThe Sa plasing 5′primereverslarly,cDNAas theCAAAwereLabeltopeumnsusing(GE H
SmalThe
smallDharmThe mGUCAwas Gwas GsingleCUAUGGUGACUCAUGUACAAsingleCAUUGAUCUACUUCAU(Dharfor expoolSCD1Ambi
RNAMD
mycinAll saconce0 houfor eaand a
TransTra
GENEcol (Rwas dSCD1wellTransfor 24prior100 nbinatiwereLucifeBP1 ptransfTK, amids.mediuwith
Luyimbazi et al.
Mol C2772
D
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980
time PCR analysisl-time PCR was done on the ABI PRISM 7900HTnce Detection System (Applied Biosystems). Thes-on-Demand gene expression products used in-d human SCD1 (Hs01682761m1), 18S (4319413E),(4326315E), and glyceraldehyde 3-phosphate de-
genase (GAPDH; 4326317E; Applied Biosystems).an Universal PCR Master Mix (Applied Biosys-was used to amplify the cDNA. The Sequencetion System software automatically determinedhange for SCD1 in each sample relative to theenous control.
ern blot analysisls were washed with cold PBS and lysed in eithermol/L Tris-HCl (pH 6.8), 4% SDS, and 20% glyc-r 1% Triton X-100, 150 mmol/L NaCl, 20 mmol/Lm phosphate buffer (pH 7.4), 1% aprotonin,ol/L phenylmethylsulfonylfluoride, 10 μg/mLptin, 100 mmol/L NaF, 2 mmol/L Na3VO4,ol/L EGTA, and 5 mmol/L sodium pyrophos-. Cell lysates containing 50 μg of protein were sep-by SDS-PAGE. The protein was transferred to aol/L polyvidine difluoride membrane (Bio-Rad
atories). Membranes were blocked in 0.1% caseinS.SCD1 antibodies were from Santa Cruz Biotech-y, Inc. Antibodies against total Acetyl CoA Car-ase (ACC), total Akt, phospho-Akt (Thr 308),ho-Akt (Ser 473), mTOR, total S6K1, phospho-(Thr 389), eIF4E, and phospho-S6 ribosomaln (S6RP; Ser 240/244) were purchased from Cellling Technology, Inc. Antibodies against fatty acidase (FAS) and SREBP1 were purchased from BDences, Inc. Antibodies against β-actin were from-Aldrich. The immunoblots were visualized usingthe Odyssey IR imaging system or softwarer Biosciences), or the enhanced chemilumines-detection kit (ECL) from GE Healthcare Corp.
ern blot analysisctrophoresis of total RNA and Northern analysisdone using the NorthernMax kit (Ambion Inc.).CD1 probe was constructed via PCR reaction frommid containing the human SCD1 cDNA clone us--CCACAGCATATCGCAAGAAA-3′ as the forwardr and 5′-CCCAGCTGTCAAAGAGAAGG-3′ as thee primer. The β-actin probe was constructed simi-from a plasmid containing the human β-actinclone using 5′-GGCATCCTCACCCTGAAGTA-3′forward primer and 5′-GGGGTGTTGAAGGTCT--3′ as the reverse primer. The respective probes
labeled with α P32 dCTP using the Prime-a-Geneing System (Promega). Unincorporated radioiso-was removed using Illustra MicroSpin G-25 col-(GE Healthcare Corp.). All blots were developed
a Molecular Dynamics Storm 860 phosphorimagerealthcare Corp.).hourRepor
ancer Ther; 9(10) October 2010
on March 15, 2020. © 201mct.aacrjournals.org ownloaded from
l interfering RNAsilencing of SCD1, mTOR, S6K1, or eIF4E withinterfering RNA (siRNA) was achieved by usingaFECT 1 transfection reagent (Dharmacon, Inc.).TOR single siRNA sequence was CCCUGCCUUU-UGCCU (27). The eIF4E single siRNA sequence 1GAUGGUAUUGAGCCUAUG and sequence 2CAAACCUGCGGCUGAUCU (28). The S6K1siRNA sequence 1 was CAGUGGAGGAGAA-UU (29), sequence 2 was CUUCUGGCUCGAAA-
GG, and sequence 3 was UGUAUGACAUGCU‐GG. The S6K1 pool sequences were CAUGGAA-
UGUGAGAAAUU, GGAAUGGGCAUAAGUU-UU, GUAAAUGGCUUGUGAUACUUU, andAUUAGCAUGCAAGCUUUU (29). The SCD1siRNA sequencewasCUACGGCUCUUUCUGAU-and sequence 2 was GAGAUAAGUUGGAGAC-
UU (16). The SCD1 pool siRNA sequences wereCAAGAGUGGCUGAGUUUU, CUACGGCU-UCUGAUCAUU, GCACAUCAACUUCACCA-U, and GAACAGUGCUGCCCACCUCUUmacon). A negative control single siRNAwas usedperiments involving single sequence siRNA, andsiRNA sequences were used in the silencing of. All siRNAs were purchased from Dharmacon,on, or Sigma.
stability studyA-MB-468 cells were inoculated with either rapa-or vehicle at 12 time points spanning 24 hours.
mples were incubated with actinomycin D at a finalntration of 5 μg/mL. No treatment was given at ther time points, which were the designated controlsch study. Total RNA was extracted with TRIzolnalyzed via Northern blotting.
ient transfection and dual reporter gene assaynsient cotransfections were done using the Fu-6 transfection reagent per manufacturer's proto-oche Applied Science, Inc.). Each experimentone in triplicate. For rapamycin regulation ofexperiments, cells were transfected with 0.5 μg/
of both hSCD-Luc pGL3 and pRL-TK plasmids.fected cells were incubated in complete mediumhours, followed by serum-free medium 12 hoursto treatment. The cells were then treated in
mol/L rapamycin, 5 μmol/L mevinolin, or a com-on in media with 10% lipid-deficient serum. Cellsharvested 24 hours after treatments using the Dualrase Reporter Assay System (Promega). For 4E-lasmid cotransfection experiments, cells wereected with 1 μg/well of hSCD-Luc pGL3, pRL-nd empty vector, or 4E-BP1 WT or 4E-BP1 5A plas-Transfected cells were incubated in completem for 72 hours, followed by a combination media10% lipid-deficient serum. Cells were harvested 24
s after treatments using the Dual Luciferaseter Assay System (Promega). All readings wereMolecular Cancer Therapeutics
0 American Association for Cancer Research.
doneDetec
In vivAll
AnimgramAssesInternSpragMDAAfterweeklDMSOthe firand R
ReverPro
graftswere240/2was vThe d
SREBNu
FractifactorhumaChem
Cell gCel
treate(SRB)platescordinwas mand ICanaly
StatisAll
meanone-wwhere(GrapThe
usingto estmixtuThe gcoverwereparisofalsepolys
meanray dwere
Resu
Rapain vitWe
threetypesrecep(HERceptocin ha(36, 3tal celally arapamrapamthreea clinitrialssignifmRNtranslThere0.010)5.7-fotreatmdeclin0.026)mRNcells,els upand acells (a falsBT-47ysis wmarildecreTo e
rapamPCRRNA,MDApresenhourshicle.centrimal Rrelativand thwerebut o
Rapamycin Regulates SCD1 Expression
www.a
D
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980
using a Sirius Single Tube Luminometer (Bertholdtion Systems).
o studiesanimal studieswere approved by theM.D.Andersonal Care and Use Committee. The animal care pro-is fully accredited by the Association for thesment and Accreditation of Laboratory Animal Care,ational. Eight-week-old female nu/nu mice (Harlanue Dawley Inc.) were inoculated with 1.5 × 107
-MB-468 or MCF7 cells in their mammary fat pads.75 to 150 mm3 tumors formed, mice were giveny i.p. injections of either rapamycin (15 mg/kg) orfor 3 weeks. Mice were euthanized 24 hours after
st or fourthweekly injection. Tumorswere harvested,NAwas extracted using RNAlater (Ambion).
se phase proteomic arraytein lysates were prepared from frozen mice xeno-and printed on nitrocellulose-coated slides. Slidesprobed with phospho-S6 ribosomal protein (Ser44) antibody (Cell Signaling Technology), whichalidated for reverse phase proteomic array (RPPA).etailed procedure is described elsewhere (30).
P1 transcription factor assayclear protein was extracted using a Nuclear/Cytosolonation Kit (BioVision Inc.). Specific transcriptionDNA binding activity was then assayed using an SREBP1 Transcription Factor Assay Kit (Caymanical Co.).
rowth, dose-effect, and cell cycle analysisl growth was determined by the protein content ofd and untreated cells through the sulforhodamine Bassay (31). The experiment was done in 96-wellin triplicate. Plated cell numbers were adjusted ac-g to the growth rate of the cell lines. Absorbanceeasured at a wavelength of 570 nm. Dose-effect50 analysis are described elsewhere (32). Cell cycle
sis was done as previously described (33).
tical analysisresults formatted as bar graphs are presented ass ± SE. Data were analyzed using Student's t-test,ay ANOVA, and Spearman rank correlation testsappropriate by GraphPad Prism v5.01 software
hPad Software).probe intensities on microarrays were processedthe Position-Dependent Nearest Neighbor modelimate gene expression values (34). The β Uniformre model (35) was used to estimate false discovery.ene list was considered significant at a false dis-y rate of 20%. For all other analyses, differencesconsidered significant at P < 0.05. Two group com-ns were done by using Student's t-test using the
discovery rate cutoff of 20% for both total andomal RNA groups, and the data are presented asSCD1MCF7
acrjournals.org
on March 15, 2020. © 201mct.aacrjournals.org ownloaded from
± SE. Calculations and presentations of microar-ata were done in log 2. Calculations of RPPA datadone in log 2.
lts
mycin treatment suppresses SCD1 expressionrosought to identify genes regulated by rapamycin incell lines representing different breast cancer sub-: MDA-MB-468, estrogen receptor/progesteronetor/human epidermal growth factor receptor 22) negative, i.e., triple negative; MCF7, estrogen re-r positive; and BT-474, HER2 positive. As rapamy-s been shown to affect transcription and translation7), we compared the gene expression profiles of to-lular mRNA and polysome-associated, translation-ctive mRNA in the presence and absence ofycin (Fig. 1A). Cells were treated with 100 nmol/Lycin, a dose that significantly inhibits growth in allbreast cancer cell lines and that has been shown to becally achievable peak concentration in temsirolimus(38, 39). In all three cell lines, we found a statisticallyicant drop in the levels of polysome-associatedA for SCD1, suggesting that there is a decline inationally active SCD1 mRNA in all three cell lines.was a 2.8-fold decline in MDA-MB-468 cells (P =, a 2.1-fold decline in MCF7 cells (P = 0.001), and ald decline in BT-474 cells (P = 0.002) after rapamycinent. SCD1 total RNA levels in MDA-MB-468 cellsed by 2.3-fold after rapamycin treatment (P =, raising the possibility that the decline in polysomalA could be due to a decrease in total RNA. In MCF7there was a minimal drop in total SCD1 mRNA lev-on rapamycin treatment by microarray analysis,lthough there was a decline in total RNA in BT-474P = 0.011), this was not statistically significant usinge discovery rate cutoff of 20%. Thus, in MCF7 and4 cells we could not determine by microarray anal-hether the drop in polysomal SCD1 levels was pri-y attributable to a decrease in total RNA or aase in polysomal recruitment of SCD1 mRNA.valuate the mechanism of SCD1 downregulation byycin, we used quantitative reverse transcriptase-(Q-PCR) to assess SCD1 mRNA levels in totaland monosomal and polysomal RNA fractions in-MB-468 and MCF7 cells cultured in the absence orce of rapamycin. Total RNA was isolated after 24of exposure to either 100 nmol/L rapamycin or ve-Cell lysates were separated with sucrose gradientfugation to obtain pooled monosomal and polyso-NA batches. Real-time PCR was done to quantitatee RNA levels for SCD1 and actin in both total RNAe monosomal and polysomal fractions. The resultssimilar to those observed by microarray analysis,n Q-PCR a statistically significant decline in total
mRNA was shown in both MDA-MB-468 andcells. Q-PCR revealed a significant decrease inMol Cancer Ther; 9(10) October 2010 2773
0 American Association for Cancer Research.
Figure24 houGenomsamplethat meand pocontrolC, Nort24 andfor 1 da(top lefexpressthree tu
Luyimbazi et al.
Mol C2774
D
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980
1. Rapamycin decreases SCD1 expression. A, MDA-MB-468, MCF7, and BT-474 cells were treated with 100 nmol/L rapamycin or 0.01% DMSO forrs. Polysomal RNA was separated by sucrose gradient centrifugation. Total polysomal RNA was extracted and hybridized to Affymetrix Humane U133 Plus 2.0 chips. The RNA expression in the rapamycin-treated samples was compared with that of untreated total and polysomal RNAs using Student's t-test. The gene of interest was considered significant in each cell line if it met a false discovery rate of 20%. All comparisonst this cutoff are demarcated by an asterisk (*). Data are means ± SE. B, Q-PCR analysis was done to quantitatively assess total RNA, and monosomallysomal fractions in MDA-MB-468 and MCF7 cells treated with rapamycin versus vehicle for 24 hours. Actin was used as the endogenous. RNA expression in rapamycin-treated and untreated samples was compared by using Student's t-test. Data are means ± range (min–max).hern blot analysis for SCD1 and actin was done on total RNA isolated from MDA-MB-468 and MCF7 cells grown in either rapamycin or vehicle for96 hours. D, to study the effect of rapamycin on SCD1 in vivo, MDA-MB-468 or MCF7 xenografts were treated with either rapamycin or vehicley or 3 weeks. Tumor volumes at day 22 are shown as means ± SE. Vehicle versus rapamycin groups were compared using Student's t-testt). Protein lysates prepared from three xenografts were printed on RPPA slides and probed with P-S6RP (Ser 240/244) antibody. Relative P-S6RP
ion in rapamycin-treated and untreated groups were compared by using Student's t-test (top right). Data are means ± SE. Total RNA frommor samples from each group was evaluated using Q-PCR to assess SCD1 and actin expression (bottom). Data are means ± SE.ancer Ther; 9(10) October 2010 Molecular Cancer Therapeutics
on March 15, 2020. © 2010 American Association for Cancer Research. mct.aacrjournals.org ownloaded from
SCD1RNAhicle (was nnot shSCD1
tionalfractistead,in bo
Figurewith 10done (t(bottommediumThe expRelativeAfter ovof IGF-and actaddedMDA-MSCD1 aculturedblotting DA-MBEZ235 P-Akt
Rapamycin Regulates SCD1 Expression
www.a
D
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980
mRNA in both the monosomal and the polysomalfractions exposed to rapamycin compared with ve-Fig. 1B). Similar results were obtained when SCD1ormalized to GAPDH or 18S ribosomal RNA (data
was done using P-Akt (Thr 308), Akt, SCD1, and actin antibody. Right, M. Lysates were collected at 6 hours and Western blotting was done using
own). These results argue against the regulation ofby rapamycin at the translational level, as transla-
that thily at
acrjournals.org
on March 15, 2020. © 201mct.aacrjournals.org ownloaded from
regulation would have led to a drop in polysomalons and an increase in monosomal fractions. In-the decline in total SCD1 mRNA as well as mRNAth polysomal and monosomal fractions suggests
B-468 cells were cultured with DMSO or 1, 10, or 100 nmol/L(Ser 473), Akt, and actin antibody.
2. SCD1 expression is inhibited by PI3K/mTOR inhibitors and increased by insulin signaling. A, MDA-MB-468, MCF7, and BT-474 cells were treated0 nmol/L rapamycin or DMSO for 24, 48, 72, and 96 hours, and 10% SDS-PAGE and Western blotting using SCD1 and actin antibody wereop). MCF7 cells were treated with various concentrations of rapamycin for 96 hours. Western blotting was done using SCD1 and actin antibody). B, MCF7 cells were incubated overnight in serum-free media. Cells were then cultured for 8 hours in one of the following conditions: no treatment,containing 10 μg/mL of insulin, or 100 ng/mL of IGF-I. SDS-PAGE (10%) and Western blotting using SCD1 antibody and actin were done (left).eriments were replicated in triplicate and quantified according to a relative expression of SCD1/β-actin. Top band was used for quantification.SCD1 expression in the treatment groups was compared with that of the no treatment group using Student's t-test (middle). Data are means ± SE.ernight serum starvation, cells were cultured for 8 hours in one of the following conditions: no treatment or medium containing 25 or 100 ng/mLI in the absence or presence of pretreatment with 100 nmol/L rapamycin (right). SDS-PAGE (10%) and Western blotting using SCD1 antibodyin were done. C, MCF7 cells were transfected with control and constitutively active Akt (CA-Akt) plasmids. Sixty hours later, serum-free media wereand cells were incubated for an additional 36 hours. Western blotting was done for SCD1, Akt, P-S6RP (Ser 240/244), and actin (right). D, left,B-468 cells were cultured for 24 hours with no treatment, DMSO, 100 nmol/L rapamycin, and 50 μmol/L LY294002. Western blotting was done usingnd actin antibody. S6K1 and phospho-S6K1 (Thr 389) were used to confirm inhibition of the mTOR pathway. Middle, MDA-MB-468 cells werewith DMSO, or 1, 10, or 100 nmol/L BEZ235. Lysates were collected at 6 hours (for P-Akt and Akt) or 24 hours (for SCD1 and actin). Western
e regulation of SCD1 by rapamycin occurs primar-the level of total RNA.
Mol Cancer Ther; 9(10) October 2010 2775
0 American Association for Cancer Research.
Tolines wmycinThe dknowcin trSCD1represfirm t
RapasupprTo d
and SMDAforme1 day
statistrapamxenogin thetreategraftsand 1(Fig.mTORtreatetumo24 hotreatm(Ser 2grafts
Luyimbazi et al.
Mol C2776
D
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980
confirm this observation, total RNA from both cellas isolated after 24 or 96 hours of exposure to rapa-or vehicle, andNorthern analysis was done (Fig. 1C).ata reveal two transcripts (3.0 and 5.2 kb) that aren to encode the same size polypeptide (40). Rapamy-eatment was associated with a >50% reduction inmRNA levels. Both transcripts were equally
sed in the presence of rapamycin. These findings con-he regulation of total SCD1 RNA by rapamycin.
mycin treatment decreases growth andesses SCD1 expression in vivoetermine the effect of rapamycin on tumor growthCD1 expression in vivo, we inoculated mice with-MB-468 or MCF7 cells and once tumors had
d treated the mice with rapamycin or vehicle foror 3 weeks. Xenografts of both cell lines showed91% (of rap
ancer Ther; 9(10) October 2010
on March 15, 2020. © 201mct.aacrjournals.org ownloaded from
ically significant inhibition of tumor growth withycin treatment. Tumor volumes of MDA-MB-468rafts on day 22 were 355 ± 80 mm3 (mean ± SE)DMSO group and 140 ± 13 mm3 in the rapamycin-d group (P = 0.03). Tumor volumes of MCF7 xeno-on day 22 were 388 ± 54 mm3 in the DMSO group87 ± 29 mm3 in the rapamycin group (P = 0.004)1D). Next, we assessed the effect of rapamycin on
downstream signaling by RPPA in xenograftsd with rapamycin for 1 day, or weekly for 3 weeks;rs were in both cases harvested approximatelyurs after rapamycin treatment. After 3 weeks ofent, rapamycin significantly reduced P-S6RP40/244) phosphorylation in MDA-MB-468 xeno-by 90% (P < 0.001) and in MCF7 xenografts by
P < 0.001; Fig. 1D). Then we established the effectamycin on SCD1 RNA in vivo. SCD1 RNA levelsureD1 pSCDA10
iclehouubatnaltreaourdesdy.ea
theNoow.F7 cctstheuctioplemum.licatiferar coSC
CD1L) plicatressrapamycin.
FigSCnotA, Mwithveh24inca fiNo0 hthestufromNorthebelMCeffeonindsupserreplucaftewith(hS(pRtriprepby
Molecular
0 American Association for C
3. Rapamycin regulatesromoter activity, butD1 mRNA stability.-MB-468 cells were treated0 nmol/L rapamycin orat 12 time points spanningrs. All samples wereed with actinomycin D atconcentration of 5 μg/mL.tment was used at thetime points, which wereignated controls for eachTotal RNA was extractedch sample and analyzed viarn blotting. Quantitation ofrthern blot was shownB, Western blot analyses inells were used to study theof rapamycin (100 nmol/L)mevinolin-based (5 μmol/L)n of SCD1 in mediumented with lipid-deficientThese results wereed in triplicate. C, dualse assays were donetransfecting MCF7 cellsD1 promoter reporter-Luc pGL3) and controllasmids. Results frome experiments show aion of the SCD1 promoter
Cancer Therapeutics
ancer Research.
showechang(P = 0declin(P = 0resultexpre
SCD1PI3KWe
SCD1pamydownlinesmentbandfragm
We inof rapdecre10 nmTo
(PI3Kstudieknowhoursto eitwas cblot awithindepsulin(P = 0
Figure24 andsmallertreatedreplicated in triplicate and quantified in comparison with vehicle using Student's t-test. Data are means ± SE. D, MDA-MB-468 cells were treated with100 nm , FAS,
Rapamycin Regulates SCD1 Expression
www.a
D
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980
d some decline in MDA-MB-468 tumors with thee at 3 weeks approaching statistical significance.051; Fig. 1D). MCF7 tumors showed a significante in SCD1 RNA levels at both day 1 and week 3.006 and P = 0.025, respectively; Fig. 1D). Theses show that rapamycin regulates SCD1 total RNAssion both in vivo and in vitro.
expression is increased with activation of/Akt/mTOR signalingsought to determine whether rapamycin regulatesprotein expression. After exposing the cells to ra-cin for 24 to 96 hours, we observed a robustregulation of SCD1 protein levels in all three cell(Fig. 2A). A dominant band and a smaller frag-were noted in most Western blots; the smaller
ol/L rapamycin for 96 hours. SDS-PAGE and Western blotting using ACC
showed equal modulation by rapamycin. Thisent may represent a SCD1 cleavage product (41).
Rapamoverc
acrjournals.org
on March 15, 2020. © 201mct.aacrjournals.org ownloaded from
cubated MCF7 cells with increasing concentrationsamycin for 24 hours and found a dose-dependentase in SCD1 expression, which became evident atol/L (Fig. 2A).investigate the role of phosphoinositide 3-kinase)/Akt/mTOR signaling in SCD1 expression, wed the effects of insulin and IGF-I, two mitogensn to induce mTOR activity (42, 43). After 12of serum starvation, MCF7 cells were exposed
her insulin or IGF-I, and their SCD1 expressionompared with cells grown in no serum. Westernnalyses showed an increase in SCD1 expressionboth stimuli (Fig. 2B, left). The results from threeendent experiments showed that, indeed, both in-and IGF-I significantly increased SCD1 expression.0002 and P = 0.04, respectively; Fig. 2B, middle).
SCD1, SREBP1, and actin antibodies were done.
4. Rapamycin regulates expression of mature SREBP1. A, MCF7 and MDA-MB-468 cells were treated with 100 nmol/L rapamycin or vehicle for96 hours. SDS-PAGE (8%) and Western blotting using SREBP1 and actin antibodies were done. Each SREBP1 blot had a precursor band (P) and amature band (M). These results were confirmed in triplicate experiments. Nuclear protein extracts from MCF7 (B) and MDA-MB-468 (C) cellswith 100 nmol/L rapamycin or vehicle for 1 and 4 days were assayed for specific transcription factor-DNA binding activity. All experiments were
ycin pretreatment was at least in part able toome IGF-induced SCD1 overexpression (Fig. 2B,
Mol Cancer Ther; 9(10) October 2010 2777
0 American Association for Cancer Research.
right)and tconsttrans
starvephory
FigureformTOB,MDAactin wAkt, P-A72 hourfirst tracontrolin triplic(hSCD1as cont
Luyimbazi et al.
Mol C2778
D
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980
. We conducted a gain-of-function experimentransfected the MCF7 cell line with control and
itutively active Akt plasmids. Sixty hours after targe-Luc pGL3) and control (pRL-TK) plasmids were cotransfected, and 96 hours later,rol. Analysis was done by using one-way ANOVA and Tukey post hoc test (right).
ancer Ther; 9(10) October 2010
on March 15, 2020. © 201mct.aacrjournals.org ownloaded from
d for 36 hours. There was an increase in phos-lation of S6 ribosomal protein, a downstream
t of mTOR, and accompanying increase in ex-fection, serum was removed and cells were pression of SCD1 (Fig. 2C).
5. eIF4E knockdown decreases SCD1 and SREBP1 expression and SCD1 promoter activity. A, MDA-MB-468 cells were transfected with siRNAR and eIF4E. After 72 hours,Western blottingwith SCD1,mTOR, and eIF4E and actinwas done. These resultswere confirmed in triplicate experiments.-MB-468 cells were transfected with control siRNA or two separate sequences of eIF4E siRNA. After 72 hours, Western blotting with SCD1, eIF4E, andas done. C, MDA-MB-468 cells were transfected with single or pool siRNA for S6K1, and 72 hours later, Western blotting was done using S6K1,kt (Thr 308), P-Akt (Ser 473), SCD1, SREBP1, and actin antibody (left). MDA-MB-468 cells were transfected with siRNA for eIF4E and S6K1. Afters, Western blotting with SCD1, SREBP1, eIF4E, and S6K1 was done (right). These results were confirmed in triplicate experiments. D, MCF7 cells werensfected with siRNA for eIF4E. Dual luciferase assays were then done after cotransfecting with SCD1 promoter reporter (hSCD1-Luc pGL3) and(pRL) plasmids. Treatment with and without rapamycin served as control. These results reflect an average of three independent experiments doneate (left). MDA-MB-468, MCF7, and BT-474 cells were transfected with vector, 4E-BP1 WT, or 4E-BP1 5A plasmids. SCD1 promoter reporter
dual luciferase assays were carried out. Vector transfection servedThis experiment was repeated three times in triplicate. Bars, SE.
Molecular Cancer Therapeutics
0 American Association for Cancer Research.
Toin SCa duamycinSCD1new(45). WcreasemTOR
RapaTo d
SCD1rapammycintrendsthat r(Fig. 3SCD1We cupresemoterrapampressimoterSCD1controactivirapamand Pport tscript
RapaWe
pressSREBposedand Scells ea declern blsor pOncespecifSCD1bent ation fdoublmobilfromrapamin spethe pbothMDAthis fmedia
causemultistudiescriptrapamACCof SCthesis
eIF4Eand SAct
its effS6K1phoryeIF4Eexpredownthat mour excurs teIF4EdeclinmTORfirmeddeclinlencinducedWe
SREBfor eIcreaseproteieIF4Edid afectedLuc pwe fosupprlevelsiRNA(P =togethulatioTo
SCD,hSCDemptymid u(Fig.promfectiovector
SCD1cell g
Rapamycin Regulates SCD1 Expression
www.a
D
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980
further confirm the role of PI3K/mTOR signalingD1 expression, we studied the effects of LY294002,l PI3K and mTOR inhibitor (44). Similar to rapa-, treatment with LY294002 led to the repression of(Fig. 2D). In addition, we tested the effect of a
generation dual PI3K/mTOR inhibitor, BEZ235e found that BEZ235 treatment also led to a de-in SCD1 levels, consistent with a role for PI3K/signaling in SCD1 expression (Fig. 2D).
mycin regulates SCD1 promoter activityetermine whether rapamycin affects the stability ofmRNA, MDA-MB-468 cells were cultured in eitherycin or vehicle in the presence of 5 μg/mL actino-D. The lack of a difference between the 24-hourof SCD1 mRNA levels in the two groups suggestsapamycin does not affect SCD1 mRNA stabilityA). We thus hypothesized that rapamycin regulatestranscription by inhibiting SCD1 promoter activity.ltured MCF7 cells with rapamycin in the absence ornce of mevinolin, a known inducer of SCD1 pro-activity (21, 46). Western blot analysis showed thatycin repressed the induction of SCD1 protein ex-on by mevinolin (Fig. 3B). We next did a dual pro-assay (25). After cotransfecting MCF7 cells withpromoter reporter plasmid hSCD1-Luc pGL3 andl plasmid pRL-TK in triplicate, SCD1 promoter
ty was significantly suppressed in the presence ofycin, both with and without mevinolin (P = 0.013= 0.009, respectively; Fig. 3C). These results sup-he notion that rapamycin modulates SCD1 tran-ion by inhibiting SCD1 promoter activity.
mycin regulates SREBP1next sought to determine how rapamycin sup-es SCD1 promoter activity. We focused on theP1 transcription factor based on a recently pro-association between the PI3K/mTOR pathwayREBP1 (47, 48). Both MDA-MB-468 and MCF7xposed to rapamycin for 24 and 96 hours showedine in the mature active form of SREBP1 on West-ot (Fig. 4A). In addition, levels of SREBP1 precur-rotein decreased in MCF7 at both time points.in the nucleus, the mature form of SREBP1 bindsically to the SREBP response element (SRE) on thepromoter (25). The enzyme-linked immunosor-ssay (ELISA) was used to detect specific transcrip-actor DNA binding activity. The assay kit has ae-stranded DNA sequence containing the SRE im-ized onto wells of a 96-well plate. Nuclear extractsboth MCF7 and MDA-MB-468 cells treated withycin for 24 and 96 hours showed a clear decreasecific transcription factor DNA binding activity inresence of rapamycin. The tabulated P values forMCF7 (P = 3.1 × 10−10, P = 0.00014; Fig. 4B) and-MB-468 (P = 0.0014, P = 0.00054; Fig. 4C) show
inding to be supportive of a role for rapamycin-ted decrease in SREBP1 DNA binding activity. Be-Asing, w
acrjournals.org
on March 15, 2020. © 201mct.aacrjournals.org ownloaded from
SREBP1 is known to regulate transcription ofple genes in the fatty acid synthesis pathway, wed the effects of rapamycin on two additional tran-ion targets (21). MDA-MB-468 cells treated withycin for 96 hours also showed a suppression ofand FAS protein expression in addition to thatD1, suggesting a broader effect on fatty acid syn-(Fig. 4D).
plays a role in the regulation of both SCD1REBP1ivation of mTOR results in the phosphorylation ofectors, the best studied of which are 4E-BP and(49). Because rapamycin decreases 4E-BP1 phos-lation (50), thereby decreasing the availability of, we assessed the effect of eIF4E siRNA on SCD1ssion in MDA-MB-468 cells. We used the knock-of mTOR by siRNA as a positive control. We foundTOR knockdown decreased SCD1, consistent withpectation that the effect of rapamycin on SCD1 oc-hrough decreased mTOR signaling. We found thatknockdown, like mTOR knockdown, also led to ae in SCD1, suggesting a potential role for the/4E-BP1 axis in SCD1 regulation (Fig. 5A). We con-that two separate eIF4E siRNA sequences led to a
e in SCD1 levels (Fig. 5B). In MDA-MB-468 cells, si-g S6K1, an mTOR downstream target, neither re-SCD1 levels nor increased P-Akt levels (Fig. 5C).then further studied the role of mTOR effectors onP1. MDA-MB-468 cells were transfected with siRNAF4E and S6K1. Western blot analysis showed a de-in both precursor and mature forms of SREBP1
n expression, as well as SCD1 expression for thesiRNA–transfected cells only (Fig. 5C). We nextdual promoter assay using MCF7 cells cotrans-with SCD1 promoter reporter plasmid hSCD1-GL3 and control plasmid pRL-TK in triplicate;und SCD1 promoter activity to be significantlyessed with knockdown of eIF4E by siRNA. Theof relative promoter activity decline with eIF4Ewas similar to that observed with rapamycin
0.005 and P = 0.019, respectively; Fig. 5D). Takener, these results show a role for eIF4E in the reg-n of SCD1.support the role of 4E-BP1/eIF4E on expression ofwe cotransfected SCD1 promoter reporter plasmid1-Luc pGL3 and control plasmid pRL-TK withvector or 4E-BP1 WT plasmid or 4E-BP1 5A plas-sing MDA-MB-468, BT-474, and MCF7 cell lines5D). Using dual promoter assay, we found SCD1oter activity to be significantly reduced with trans-n of 4E-BP1 5A plasmid compared with emptyand wild-type in all cell lines (P < 0.05).
suppression and overexpression modulatesrowth
SCD1 expression is regulated by mTOR signal-e hypothesized that baseline SCD1 expressionMol Cancer Ther; 9(10) October 2010 2779
0 American Association for Cancer Research.
itselftest thsion ocell li
cell litions
Luyimbazi et al.
Mol C2780
D
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980
may be a predictor of rapamycin sensitivity. Tois hypothesis, we quantified the baseline expres-
f SCD1 protein in a panel of 15 breast cancernes by Western blotting (Fig. 6A). This panel ofinhibitween
ancer Ther; 9(10) October 2010
on March 15, 2020. © 201mct.aacrjournals.org ownloaded from
nes was also treated with increasing concentra-of rapamycin and 5 days after exposure, growth
tion was measured by SRB assay. Correlation be-SCD1 expression and IC50 was not significantMolecular Cancer Therapeutics
0 American Association for Cancer Research.
(SpeabaselitermininhibiAs
soughexpreMDAand p(Fig.transfmentin boplateysis (moreand MknockWestetisticawithquencknockshownage oNA vwithregulMDAthesesignifWe
can inline wstablespectiMB-2an ascant iof SCconfir(Fig. 6
Discu
Rapextenantituthe hiis anmarkerole oits abmeostsignathe linactionmor gmT
pamydephowhich4E-BPwhichcap-dsomaltion otract (al regprogrfactoras a mcin trprimaels duthan dment.by inswas sing mtivatelipogeinclud
Figureusing SSCD1/a5 dayswere trgrowthdesignaMCF7 awere trdone wStudentransfec96 houcontrolreflecttheir re
Rapamycin Regulates SCD1 Expression
www.a
D
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980
rman r = −0.3424, P = 0.2116; Fig. 6A). Thus,ne level of SCD1 protein expression did not de-e sensitivity of breast cancer cells to the growth-tory action of rapamycin.mTOR is known to regulate cell growth, wet to determine whether the suppression of SCD1ssion has any effect on cell growth. MCF7 and-MB-468 cells were transfected with both singleool SCD1 siRNA and their respective controls6B). One day after transfection, the cells wereerred to 96-well plates in quadruplicates. Treat-with rapamycin, a known suppressor of growthth cell lines, was used as control (2). A parallelwas maintained for subsequent Western blot anal-Fig. 6B). Cell proliferation was assessed after 3days of culture by SRB assay. In both MCF7 cellsDA-MB-468 cells, SCD1 pool and single siRNAdown led to a significant decrease in growth;rn blot analysis confirmed SCD1 knockdown. Sta-lly significant growth inhibition was obtainedSCD1 knockdown using two separate siRNA se-es in MCF7 and BT474 cells (Fig. 6B). SCD1down in MCF7 cells induced cell cycle arrest, asby a statistically significant increase in percent-
f cells in G0-G1 phase (50% G0-G1 with control siR-ersus 70% with SCD1 siRNA, P = 0.002, and 61%SCD1 siRNA 2, P = 0.011; Fig. 6C). SCD1 down-ation did not significantly affect cell cycle in-MB-468 and BT474 cells. Taken together, however,results show that suppression of SCD1 yields aicant decline in cell growth.next determined whether overexpression of SCD1crease cell growth. We used MDA-MB-231, a cellith relative low expression of SCD1, to develop aSCD1-overexpressing cell line along with its re-
ve control. Both overexpresser and control MDA-31 were grown in 96-well plates. After 96 hours,say for cell number was done, yielding a signifi-ncrease in proliferation with the overexpressionD1 (P = 1.8 × 10−4; Fig. 6D). Western blot analysis
med the presence of exogenous SCD1 expressionD).of S6Khowe
experiments done in triplicate (left). Western blotting with SCD1 and actin was dspective control to confirm expression of the transfected myc-tagged SCD1 prot
acrjournals.org
on March 15, 2020. © 201mct.aacrjournals.org ownloaded from
ssion
amycin and its analogs are currently undergoingsive study to evaluate the determinants of theirmor efficacy. Identifying how rapamycin affectsgh rate of metabolism required for tumor growthimportant first step that could lead to useful bio-rs or effective rational combinatorial therapy. Thef mTOR in cellular metabolism is emphasized byility to control amino acid biosynthesis, glucose ho-asis, and adipogenesis (17). We show that mTORling regulates SCD1 expression. We propose thatk between mTOR and SCD1 is one of the key inter-s that may play a role in the effect of mTOR on tu-rowth and metabolism.OR is a critical regulator of protein translation. Ra-cin inhibits mTOR complex 1 signaling, resulting insphorylation of its effectors, the best studied ofare 4E-BP1 and S6K1. Hyperphosphorylation of
1 leads to inhibition of binding of 4E-BP to eIF4E,decreases the amount of eIF4E available to initiateependent translation. S6K1 phosphorylates ribo-protein S6, in some models enhancing the transla-f mRNAs possessing a 5′ terminal oligopyrimidine51). S6K1 phosphorylates several other translation-ulators, including eukaryotic initiation factor 4B,ammed cell death 4, and eukaryotic elongation-2 kinase (52). Although we initially identified SCD1RNA that is decreased in polysomes upon rapamy-eatment in all three cell lines, we found that this isrily a reflection of a drop in total SCD1 mRNA lev-e to inhibition of SCD1 promoter activity, ratherue to a significant inhibition of polysomal recruit-Induction of SCD1 expression or promoter activityulin and abolishment of this activity by rapamycinhown in liver cancer (47), but the players connect-TOR and SCD1 are not very well defined. S6K1 ac-s liver X receptor α (LXRα), and LXRα promotesnesis by regulating expression of lipogenic genes,ing SREBP1 (53). We could not show involvement
1 in regulation of SCD1 expression; interestingly,ver, we found that silencing eIF4E also resulted in
6. SCD1 regulates breast cancer cell growth. A, left, baseline SCD1 protein expression in 15 breast cancer cell lines. Western blotting was doneCD1 and actin antibody. Right, SCD1 and actin bands were quantified; the average of three independent experiments is shown as relativectin expression. The same chart shows IC50 values of breast cancer cell lines. Cells were treated with increasing concentrations of rapamycin forand IC50 was determined on the basis of the dose-response curves using SRB assay. B, MCF7 cells (top left) and MDA-MB-468 cells (top right)ansfected with single or pool control siRNA and SCD1 siRNA. Cells were grown in medium supplemented with lipid-deficient serum and cellanalysis was then done with readings recorded after 96 hours of treatment by SRB assay. Cells treated with and without rapamycin were theted controls. These results reflect experiments done in triplicate. Western blotting with SCD1 and actin were done on the treated cells of bothnd MDA-MB-468 cells to confirm adequate suppression of SCD1 by siRNA after 96 hours. MCF7 cells (bottom left) and BT474 cells (bottom right)ansfected with single control siRNA, SCD1 siRNA, or a SCD1 siRNA 2. Cells were grown in regular medium and cell growth analysis was thenith readings recorded after 96 hours of treatment by SRB assay. SCD1 siRNA–treated cells were compared with control siRNA–treated cells usingt's t-test. These results are representative of two independent experiments. C, MCF7 (left), MDA-MB-468 (middle), and BT474 cells (right) wereted with single control siRNA, SCD1 siRNA, or a SCD1 siRNA 2. Cells were grown in regular medium and analyzed for cell cycle profile afterrs. Percentage of cells in G0-G1 was compared using Student's t-test. D, MDA-MB-231 SCD1 overexpressing stable cells were grown alongsidein media supplemented with lipid-deficient serum. Cell growth analysis was then done with readings recorded after 96 hours of growth. These results
one on MDA-MB-231–overexpressing stable cells alongsideein (right).
Mol Cancer Ther; 9(10) October 2010 2781
0 American Association for Cancer Research.
the inexpreeIF4Ethe efthe reour reof SREsion. Ator ofshowmay ptranscpresenSREBregulasis duregulaObe
develadipoeffectsmammin risSCD1acidsolismassocidownwhenshowtabolicurrentherapgest tapy tbreastof bobreastof SCeratioliferat
is a nther scanceIn s
pathwbreastpreciacer cerole osess tmTORmarkeare necer mthe roin com
Discl
F. Mhonorasuppor
Ackn
We tDr. JamhSCD1activemutant
Grant
NIH(F. MerAward1UL1R(F. Merfor MoMills) a(CA 16
Theof pagein acco
Refe1. Hid
tio668
2. Nocin101
3. Mogetcyt703
4. Yucanels
5. Me
Luyimbazi et al.
Mol C2782
D
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980
hibition of SCD1 promoter activity and subsequentssion, suggesting that SCD1 is downstream of. Further work is needed to determine whetherfect of rapamycin on SCD1 is mediated throughgulation of SREBP1 expression by eIF4E. Indeed,sults show that both precursor and mature formsBP1 are affected in the presence of eIF4E suppres-dditionally, although eIF4E is a principal modula-translation initiation (54), recent studies have
n that it maintains a presence in the nucleus andlay a role in both selective mRNA export and post-riptional modulation (55, 56). This suggests thece of a signaling module from mTOR to eIF4E toP1 culminating in SCD1. The module is likely tote a suite of enzymes involved in fatty acid synthe-e to the activity of SREBP1 in coordinate promotertion of this group of enzymes.sity in postmenopausal women is a risk factor foroping breast cancer (57). Hormones (estrogen) andkines such as leptin and IGF-I all show mitogenicin MCF7 cells and stimulate cell proliferation inary glands of obese rats, suggesting an increase
k for breast cancer (58). The role that humanplays in the production of monounsaturated fattymakes it an important factor in cellular fat metab-and a novel target in the prevention of obesity-ated breast cancer. Studies in mice point to a clearregulation of the genes involved in lipogenesisSCD1 expression is decreased (59). These studiesthat SCD1 plays a central role in cellular fat me-sm, further highlighting the significance of ourt findings. SCD1 is already being pursued as aeutic target for obesity (60, 61). Recent data sug-hat SCD1 may also hold promise as a cancer ther-arget (16). To better define the role of SCD1 incancer biology, we sought to determine the effectsth knockdown and overexpression of SCD1 oncancer cell growth in vitro. Transient knockdown
D1 yields a significant drop in cell growth/prolif-
n, whereas overexpression yields elevated cell pro-ion. These findings support the notion that SCD1Recepublish
cer: the effect of CCI-779, an mTOR inhibitor, in preclinical mod-of breast cancer. Endocr Relat Cancer 2001;8:249–58.ric-Bernstam F, Esteva FJ. Potential role of mammalian target of
rap20
6. Chlimtie20
7. EllacomeOn
8. Ensoen19
ancer Ther; 9(10) October 2010
on March 15, 2020. © 201mct.aacrjournals.org ownloaded from
ovel mediator of breast cancer cell growth and fur-upports the case that SCD1 is indeed a promisingr therapy target.ummary, we have established the mTOR signalingay as a major regulator of SCD1 expression incancer cells. This highlights a previously underap-ted link between oncogenic cell signaling and can-ll fat metabolism. We have also begun to define thef SCD1 in breast cancer. Studies are ongoing to as-he potential of SCD1 as a marker of activity ofsignaling, as well as potential pharmacodynamicr of response to mTOR inhibitors. Further studieseded to evaluate the role of SCD1 in the breast can-alignant phenotype, and ultimately to determinele of SCD1 as a therapeutic target either alone orbination with rapamycin.
osure of Potential Conflicts of Interest
eric-Bernstam: clinical trial support Novartis Pharma and Abraxis;rium Novartis Pharma. A.M. Gonzalez-Angulo: clinical trialt Novartis Pharma and Abraxis.
owledgments
hank Judy Roehm for assistance with manuscript preparation andes M. Ntambi for the gift of the SCD1 promoter reporter plasmid-Luc pGL3, Dr. Philip N. Tsichlis for the gift of the constitutivelyAkt plasmid, and Dr. Thurl E. Harris for the gift of the 4E-BP1plasmids.
Support
1 R01 CA112199 (F. Meric-Bernstam), Elsa Pardee Foundationic-Bernstam), Society of Surgical Oncology Clinical Investigator(F. Meric-Bernstam), National Center for Research Resources grantR024148 (F. Meric-Bernstam and L. Zhang), NIH 5 T32 CA09599ic-Bernstam, D. Luyimbazi, and P.F. McAuliffe), Kleberg Centerlecular Markers (F. Meric-Bernstam, A.M. Gonzalez-Angulo, G.B.nd the U. T. M. D. Anderson Cancer Center Support Core Grant672).costs of publication of this article were defrayed in part by the paymentcharges. This article must therefore be herebymarked advertisementrdance with 18 U.S.C. Section 1734 solely to indicate this fact.
ived 10/30/2009; revised 08/11/2010; accepted 08/11/2010;ed OnlineFirst 09/28/2010.
rencesalgo M, Rowinsky EK. The rapamycin-sensitive signal transduc-n pathway as a target for cancer therapy. Oncogene 2000;19:0–6.h WC, Mondesire WH, Peng J, et al. Determinants of rapamy-sensitivity in breast cancer cells. Clin Cancer Res 2004;10:3–23.ndesire WH, Jian W, Zhang H, et al. Targeting mammalian tar-of rapamycin synergistically enhances chemotherapy-inducedotoxicity in breast cancer cells. Clin Cancer Res 2004;10:1–42.K, Toral-Barza L, Discafani C, et al. mTOR, a novel target in breast
amycin inhibitors in breast cancer therapy. Clin Breast Cancer05;6:357–60.an S, Scheulen ME, Johnston S, et al. Phase II study of temsiro-us (CCI-779), a novel inhibitor of mTOR, in heavily pretreated pa-nts with locally advanced or metastatic breast cancer. J Clin Oncol05;23:5314–22.rd SL, Clemons M, Gelmon KA, et al. Randomized phase II studymparing two schedules of everolimus in patients with recurrent/tastatic breast cancer: NCIC Clinical Trials Group IND.163. J Clincol 2009;27:4536–41.och HG, Catala A, Strittmatter P. Mechanism of rat liver micro-
mal stearyl-CoA desaturase. Studies of the substrate specificity,zyme-substrate interactions, and the function of lipid. J Biol Chem76;251:5095–103.Molecular Cancer Therapeutics
0 American Association for Cancer Research.
9. Ntarat
10. NtaLip
11. Liulysinv
12. Li Jchachasue
13. Hagen
14. Fal1 (Stion23:
15. Pado17:
16. Morelproget
17. Wume
18. El-adkin191
19. Yukincel
20. Zhthr
21. Broterfac
22. ChThnas
23. Vlaph1-b
24. Maizasitoin v
25. BehuvatpoCo
26. Sobrecer
27. Kimforchi
28. Lynim(eIFtic
29. KimlatiGro
30. Hecartas
31. MoF. A
co25
32. Akrapno
33. Do3btivi
34. Zhsho
35. Poantiti19
36. Gesitiula27
37. RaOntioCe
38. Raicsno23
39. HidcodaCa
40. ZhCoge34
41. HeturMo
42. BrulatSc
43. voThinitrapBio
44. BruAbmator
45. SemTca80
46. TaseregCh
47. MaRoCoCo
48. PoedMe
49. Mena
50. GinBP
Rapamycin Regulates SCD1 Expression
www.a
D
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980
mbi JM. Regulation of stearoyl-CoA desaturase by polyunsatu-ed fatty acids and cholesterol. J Lipid Res 1999;40:1549–58.mbi JM. The regulation of stearoyl-CoA desaturase (SCD). Progid Res 1995;34:139–50.S, Umezu-Goto M, Murph M, et al. Expression of autotaxin andophosphatidic acid receptors increases mammary tumorigenesis,asion, and metastases. Cancer Cell 2009;15:539–50., Ding SF, Habib NA, Fermor BF, Wood CB, Gilmour RS. Partialracterization of a cDNA for human stearoyl-CoA desaturase andnges in its mRNA expression in some normal and malignant tis-s. Int J Cancer 1994;57:348–52.bib NA, Wood CB, Apostolov K, et al. Stearic acid and carcino-esis. Br J Cancer 1987;56:455–8.vella FS, Pascale RM, Gariboldi M, et al. Stearoyl-CoA desaturasecd1) gene overexpression is associated with genetic predisposi-to hepatocarcinogenesis in mice and rats. Carcinogenesis 2002;
1933–6.scale RM, Simile MM, DeMiglio MR, et al. The BN rat strain carriesminant hepatocarcinogen resistance loci. Carcinogenesis 1996;1765–8.rgan-Lappe SE, Tucker LA, Huang X, et al. Identification of Ras-ated nuclear protein, targeting protein for xenopus kinesin-liketein 2, and stearoyl-CoA desaturase 1 as promising cancer tar-s from an RNAi-based screen. Cancer Res 2007;67:4390–8.llschleger S, Loewith R, Hall MN. TOR signaling in growth andtabolism. Cell 2006;124:471–84.Chaar D, Gagnon A, Sorisky A. Inhibition of insulin signaling andipogenesis by rapamycin: effect on phosphorylation of p70 S6ase vs eIF4E-BP1. Int J Obes Relat Metab Disord 2004;28:–8.W, Chen Z, Zhang J, et al. Critical role of phosphoinositide 3-ase cascade in adipogenesis of human mesenchymal stemls. Mol Cell Biochem 2008;310:11–8.ang HH, Huang J, Duvel K, et al. Insulin stimulates adipogenesisough the Akt-TSC2–1 pathway. PLoS One 2009;4:e6189.wn MS, Goldstein JL. The SREBP pathway: regulation of choles-ol metabolism by proteolysis of a membrane-bound transcriptiontor. Cell 1997;89:331–40.oi JH, Bertram PG, Drenan R, Carvalho J, Zhou HH, Zheng XF.e FKBP12-rapamycin-associated protein (FRAP) is a CLIP-170 ki-e. EMBO Rep 2002;3:988–94.hos CJ, Matter WF, Hui KY, Brown RF. A specific inhibitor ofosphatidylinositol 3-kinase, 2-(4-morpholinyl)-8-phenyl-4H-enzopyran-4-one (LY294002). J Biol Chem 1994;269:5241–8.ira SM, Stauffer F, Brueggen J, et al. Identification and character-tion of NVP-BEZ235, a new orally available dual phosphatidylino-l 3-kinase/mammalian target of rapamycin inhibitor with potentivo antitumor activity. Mol Cancer Ther 2008;7:1851–63.ne H, Lasky D, Ntambi JM. Cloning and characterization of theman stearoyl-CoA desaturase gene promoter: transcriptional acti-ion by sterol regulatory element binding protein and repression bylyunsaturated fatty acids and cholesterol. Biochem Biophys Resmmun 2001;284:1194–8.ni A, Akcakanat A, Singh G, et al. eIF4E knockdown decreasesast cancer cell growth without activating Akt signaling. Mol Can-Ther 2008;7:1782–8.DH, Sarbassov DD, Ali SM, et al. mTOR interacts with raptor to
m a nutrient-sensitive complex that signals to the cell growth ma-nery. Cell 2002;110:163–75.ch M, Chen L, Ravitz MJ, et al. hnRNP K binds a core polypyr-idine element in the eukaryotic translation initiation factor 4E4E) promoter, and its regulation of eIF4E contributes to neoplas-transformation. Mol Cell Biol 2005;25:6436–53.D, Akcakanat A, Singh G, Sharma C, Meric-Bernstam F. Regu-
on and localization of ribosomal protein S6 kinase 1 isoforms.wth Factors 2009;27:12–21.nnessy BT, Gilcrease MZ, Kim E, Gonzalez-Angulo AM. Breastcinoma with neuroendocrine differentiation and myocardial me-
tases. Clin Breast Cancer 2007;7:892–4.reno A, Akcakanat A, Munsell MF, Soni A, Yao JC,Meric-Bernstamntitumor activity of rapamycin and octreotide as single agents or intiv50
51. Je
acrjournals.org
on March 15, 2020. © 201mct.aacrjournals.org ownloaded from
mbination in neuroendocrine tumors. Endocr Relat Cancer 2008;15:7–66.cakanat A, Zhang L, Tsavachidis S, Meric-Bernstam F. Theamycin-regulated gene expression signature determines prog-sis for breast cancer. Mol Cancer 2009;8:75.ng J, Peng J, Zhang H, et al. Role of glycogen synthase kinaseeta in rapamycin-mediated cell cycle regulation and chemosensi-ty. Cancer Res 2005;65:1961–72.ang L, Miles MF, Aldape KD. A model of molecular interactions onrt oligonucleotide microarrays. Nat Biotechnol 2003;21:818–21.unds S, Morris SW. Estimating the occurrence of false positivesd false negatives in microarray studies by approximating and par-oning the empirical distribution of P-values. Bioinformatics 2003;:1236–42.ra JF, Mellinghoff IK, Shi Y, et al. AKT activity determines sen-vity to mammalian target of rapamycin (mTOR) inhibitors by reg-ting cyclin D1 and c-myc expression. J Biol Chem 2004;279:37–46.jasekhar VK, Viale A, Socci ND, Wiedmann M, Hu X, Holland EC.cogenic Ras and Akt signaling contribute to glioblastoma forma-n by differential recruitment of existing mRNAs to polysomes. Molll 2003;12:889–901.ymond E, Alexandre J, Faivre S, et al. Safety and pharmacokinet-of escalated doses of weekly intravenous infusion of CCI-779, avel mTOR inhibitor, in patients with cancer. J Clin Oncol 2004;22:36–47.algo M, Buckner JC, Erlichman C, et al. A phase I and pharma-kinetic study of temsirolimus (CCI-779) administered intravenouslyily for 5 days every 2 weeks to patients with advanced cancer. Clinncer Res 2006;12:5755–63.ang L, Ge L, Parimoo S, Stenn K, Prouty SM. Human stearoyl-A desaturase: alternative transcripts generated from a singlene by usage of tandem polyadenylation sites. Biochem J 1999;0:255–64.inemann FS, Ozols J. Degradation of stearoyl-coenzyme A desa-ase: endoproteolytic cleavage by an integral membrane protease.l Biol Cell 1998;9:3445–53.nn GJ, Hudson CC, Sekulic A, et al. Phosphorylation of the trans-ional repressor PHAS-I by the mammalian target of rapamycin.ience 1997;277:99–101.n Manteuffel SR, Dennis PB, Pullen N, Gingras AC, Sonenberg N,omas G. The insulin-induced signalling pathway leading to S6 andiation factor 4E binding protein 1 phosphorylation bifurcates at aamycin-sensitive point immediately upstream of p70s6k. Mol Celll 1997;17:5426–36.nn GJ, Williams J, Sabers C, Wiederrecht G, Lawrence JC, Jr.,raham RT. Direct inhibition of the signaling functions of the mam-lian target of rapamycin by the phosphoinositide 3-kinase inhibi-s, wortmannin and LY294002. EMBO J 1996;15:5256–67.rra V, Markman B, Scaltriti M, et al. NVP-BEZ235, a dual PI3K/OR inhibitor, prevents PI3K signaling and inhibits the growth ofncer cells with activating PI3K mutations. Cancer Res 2008;68:22–30.bor DE, Kim JB, Spiegelman BM, Edwards PA. Identification of con-rved cis-elements and transcription factors required for sterol-ulated transcription of stearoyl-CoA desaturase 1 and 2. J Biolem 1999;274:20603–10.uvoisin D, Rocque G, Arfa O, Radenne A, Boissier P, Mounier C.le of the PI3-kinase/mTor pathway in the regulation of the stearoylA desaturase (SCD1) gene expression by insulin in liver. J Cellmmun Signal 2007;1:113–25.rstmann T, Santos CR, Griffiths B, et al. SREBP activity is regulat-by mTORC1 and contributes to Akt-dependent cell growth. Celltab 2008;8:224–36.ric-Bernstam F, Gonzalez-Angulo AM. Targeting the mTOR sig-ling network for cancer therapy. J Clin Oncol 2009;27:2278–87.gras AC, Kennedy SG, O'Leary MA, Sonenberg N, Hay N. 4E-1, a repressor of mRNA translation, is phosphorylated and inac-
ated by the Akt(PKB) signaling pathway. Genes Dev 1998;12:2–13.fferies HB, Fumagalli S, Dennis PB, Reinhard C, Pearson RB,Mol Cancer Ther; 9(10) October 2010 2783
0 American Association for Cancer Research.
Thinh
52. JasRBtheGro
53. Hwmopageleth
54. Methe
55. AnLumR
56. StreIF
57. Clethe
58. Laedan48
59. NtdeAc
60. Doge
Luyimbazi et al.
Mol C2784
D
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980
omas G. Rapamycin suppresses 5′TOP mRNA translation throughibition of p70s6k. EMBO J 1997;16:3693–704.trzebski K, Hannan KM, Tchoubrieva EB, Hannan RD, Pearson. Coordinate regulation of ribosome biogenesis and function byribosomal protein S6 kinase, a key mediator of mTOR function.wth Factors 2007;25:209–26.ahng SH, Ki SH, Bae EJ, Kim HE, Kim SG. Role of adenosinenophosphate-activated protein kinase-p70 ribosomal S6 kinase-1thway in repression of liver X receptor-alpha-dependent lipogenicne induction and hepatic steatosis by a novel class of dithio-iones. Hepatology 2009;49:1913–25.ric F, Hunt KK. Translation initiation in cancer: a novel target forrapy. Mol Cancer Ther 2002;1:971–9.
drei MA, Ingelfinger D, Heintzmann R, Achsel T, Rivera-Pomar R,hrmann R. A role for eIF4E and eIF4E-transporter in targetingNPs to mammalian processing bodies. RNA 2005;11:717–27.61. Liuora20
ancer Ther; 9(10) October 2010
on March 15, 2020. © 201mct.aacrjournals.org ownloaded from
udwick S, Borden KL. The emerging roles of translation factor4E in the nucleus. Differentiation 2002;70:10–22.ary MP, Grossmann ME. Minireview: obesity and breast cancer:estrogen connection. Endocrinology 2009;150:2537–42.
utenbach A, Budde A, Wrann CD, et al. Obesity and the associat-mediators leptin, estrogen and IGF-I enhance the cell proliferationd early tumorigenesis of breast cancer cells. Nutr Cancer 2009;61:4–91.ambi JM, Miyazaki M, Stoehr JP, et al. Loss of stearoyl-CoAsaturase-1 function protects mice against adiposity. Proc Natlad Sci U S A 2002;99:11482–6.brzyn A, Ntambi JM. Stearoyl-CoA desaturase as a new drug tar-t for obesity treatment. Obes Rev 2005;6:169–74.G, Lynch JK, Freeman J, et al. Discovery of potent, selective,
lly bioavailable stearoyl-CoA desaturase 1 inhibitors. J Med Chem07;50:3086–100.Molecular Cancer Therapeutics
0 American Association for Cancer Research.
2010;9:2770-2784. Published OnlineFirst September 28, 2010.Mol Cancer Ther David Luyimbazi, Argun Akcakanat, Priscilla F. McAuliffe, et al. in Breast CancerRapamycin Regulates Stearoyl CoA Desaturase 1 Expression
Updated version
10.1158/1535-7163.MCT-09-0980doi:
Access the most recent version of this article at:
Material
Supplementary
http://mct.aacrjournals.org/content/suppl/2010/09/24/1535-7163.MCT-09-0980.DC1
Access the most recent supplemental material at:
Cited articles
http://mct.aacrjournals.org/content/9/10/2770.full#ref-list-1
This article cites 61 articles, 30 of which you can access for free at:
Citing articles
http://mct.aacrjournals.org/content/9/10/2770.full#related-urls
This article has been cited by 10 HighWire-hosted articles. Access the articles at:
E-mail alerts related to this article or journal.Sign up to receive free email-alerts
Subscriptions
Reprints and
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Permissions
Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)
.http://mct.aacrjournals.org/content/9/10/2770To request permission to re-use all or part of this article, use this link
on March 15, 2020. © 2010 American Association for Cancer Research. mct.aacrjournals.org Downloaded from
Published OnlineFirst September 28, 2010; DOI: 10.1158/1535-7163.MCT-09-0980