Dysregulating IRES-Dependent Translation Contributes...

Chromatin, Gene, and RNA Regulation

Dysregulating IRES-Dependent Translation Contributes toOverexpression of Oncogenic Aurora A Kinase

Tara Dobson1, Juan Chen2, and Les A. Krushel2

AbstractOverexpression of the oncoprotein Aurora A kinase occurs in multiple types of cancer, often early during cell

transformation. To identify themechanism(s) contributing to enhanced Aurora A protein expression, a comparisonbetween normal human lung fibroblast and breast epithelial cells to nontumorigenic breast (MCF10A andMCF12A) and tumorigenic breast (MCF-7) and cervical cell lines (HeLa S3) was performed. A subset of theseimmortalized lines (MCF10A, MCF12A, and HeLa S3) exhibited increased levels of Aurora A protein,independent of tumorigenicity. The increase in Aurora A protein in these immortalized cells was not due toincreased transcription/RNA stability, protein half-life, or cap-dependent translation. Assays utilizing monocis-tronic and dicistronic RNA constructs revealed that the 5'-leader sequence of Aurora A contains an internalribosomal entry site (IRES), which is regulated in a cell cycle–dependent manner, peaking in G2/M phase.Moreover, IRES activity was increased in the immortalized cell lines in which Aurora A protein expression was alsoenhanced. Additional studies indicated that the increased internal initiation is specific to the IRES of Aurora A andmay be an early event during cancer progression. These results identify a novel mechanism contributing to Aurora Akinase overexpression.Implications: The current study indicates that Aurora A kinase contributes to immortalization and tumor-

igenesis. Mol Cancer Res; 11(8); 887–900. �2013 AACR.

IntroductionAurora A is a serine/threonine kinase that plays a crucial

regulatory role during mitotic events including centrosomeduplication, separation, and maturation as well as mitoticspindle stabilization (1–3). Regulation of Aurora A expres-sion is tightly controlled with both the mRNA and proteindetected in late S and early G2 phase, peaking in G2–M, andrapidly degrading before G1-phase (4, 5). Aberrant expres-sion of this kinase is detrimental to the cell. Overexpressioncontributes to centrosome amplification and a failure incytokinesis creating aneuploidy, setting the stage for carci-nogenesis (6, 7). Numerous tumor cell lines and humantumors exhibit elevated levels of the Aurora A kinase,suggesting itmay play a role in tumorigenesis (8–11). Indeed,ectopic Aurora A overexpression leads to cell transformation(12). Alternatively, loss of Aurora A leads to centrosomal

separation defects resulting in a monopolar spindle, which inturn activates theG2–Mcheckpoint and eventually apoptosis(13). For this reason, the Aurora A kinase is considered atarget for the development of anticancer drugs.Enhanced expression of Aurora A protein in tumors is

reportedly due to a concomitant increase in Aurora AmRNAowing to gene amplification and/or increased transcription(7, 8, 14–16). However, there are examples in many cancerswhereby increased Aurora A protein expression is notaccompanied by changes in mRNA levels (9, 10, 17). Theseresults suggest that posttranscriptional processes includingenhanced protein synthesis and/or protein stability are alsocontributing to the increased Aurora A kinase levels.The major regulatory step in protein synthesis occurs at

the initiation of translation (18). Most eukaryotic mRNAsare thought to initiate translation in a cap-dependent man-ner. This mechanism involves the binding of the preinitia-tion complex to the methyl-7-guansine (m7G) cap structureat the 50 end of themRNA and scanning of the 40S ribosometo the first initiator codon in a proper context (ref. 19;reviewed in refs. 20, 21). In a subset of cellular mRNAs, analternative mechanism to initiate translation occurs in whichthe preinitiation complex internally binds the 50 leader oruntranslated region (UTR; ref. 22, 23). The binding site isreferred to as an internal ribosome entry site (IRES). Duringperiods in which cap-dependent translation is decreased,including in response to cellular stress or throughout theG2–Mphase of the cell cycle, IRES-dependent translation isproposed to be maintained or elevated (24–27). Indeed,

Authors' Affiliations: 1Department of Biochemistry andMolecular Genet-ics, University of Colorado Anschutz Medical Center, Aurora, Colorado;and 2Department of Biochemistry and Molecular Biology, University ofTexas MD Anderson Cancer Center, Houston, Texas

Note: Supplementary data for this article are available at Molecular CancerResearch Online (http://mcr.aacrjournals.org/).

Corresponding Author: Les Krushel, Department of Biochemistry andMolecular Biology, University of Texas MD Anderson Cancer Center, 6767Bertner Ave, Houston, TX 77030. Phone: 713-834-6310; Fax: 713-834-6273; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-12-0707

�2013 American Association for Cancer Research.

MolecularCancer

Research

www.aacrjournals.org 887

on April 30, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst May 9, 2013; DOI: 10.1158/1541-7786.MCR-12-0707

http://mcr.aacrjournals.org/

manymRNAs translated duringmitosis contain IRESes. Forexample, IRES-dependent translation of the ornithine decar-boxylase (26) and PITSRLE p58 (28) mRNA occurs exclu-sively during mitosis (25). In addition, most of the smallsubset of eukaryotic IRESes identified to date are located inmRNAs encoding proteins that affect tumorigenesis. Theseproteins include oncogenes (c-myc; ref. 29), growth factors[fibroblast growth factor 2 (FGF2); refs. 30–32], growthfactor receptors (TrkB; ref. 33), pro- and anti-apoptoticfactors [X-linked inhibitor of apoptosis protein (XIAP) andAPAF-1, respectively; refs. 34, 35], and angiogenic factors(VEGF; ref. 32).Deregulating IRES-dependent translation ofthese mRNAs could be a mechanism to promote cell survivaland uncontrolled cellular proliferation during carcinogenesis.In the present report, we chose multiple cell lines that

differentially express Aurora A protein to identify mecha-nism(s) contributing to its overexpression. Transcription,mRNA stability, cap-dependent translation, and proteinstability could not account for the increased Aurora Aprotein expression in a subset of cell lines. However, anIRES was identified in the 50 leader of the Aurora A mRNA.Aurora A IRES activity positively correlated with Aurora Aprotein levels. Moreover, Aurora A expression in these cellswas unaffected when cap-dependent translation was reduced.We propose there is a switch from cap- to IRES-dependenttranslation of the Aurora A mRNA that contributes to over-expression of the protein. In turn, this enhanced IRES activitymay be a key determinant in generating genomic instabilitythat may eventually result in cellular immortalization.

Materials and MethodsConstructsThe Aurora A (Supplementary Fig. S1) and b-globin

(GenBank: V00497.1) 50 leaders were PCR-amplified froma human fetal brain cDNA library (Clontech) and insertedinto the dual luciferase vector-RP (refs. 36, 37; a generousgift from Dr. AnneWillis, University of Leicester, Leicester,United Kingdom) with EcoRI and NcoI endonucleaserestriction sites. The dicistronic construct for in vitro tran-scription was created by digesting the RP vector with EcoRVand BamHI releasing the Renilla and Photinus luciferasegenes and the SV40 30-UTR. The 2 luciferase genes wereinserted into the multiple cloning site of the SKþ Bluescriptvector (Stratagene) downstream of the T7 promoter. Themonocistronic construct for in vitro transcription was cre-ated by digesting the RP vector with EcoRI and BamHI. Thedigest released the 50 leader, the Photinus luciferase gene andthe SV40 30-UTR, which were inserted into the EcoRI andBamHI sites of the SKþ Bluescript vector (Stratagene)downstream of a T7 promoter.In experiments using a hypophosphorylated form of 4E-

binding protein 1 (4E-BP1; containing Thr-37-Ala/Thr-46-Ala/Ser-65-Ala/Thr-70-Ala/Ser-83-Ala mutations), HeLacells were transfected with 4 mg of a plasmid expressinghypophosphorylated 4E-BP1 or a control vector (Paltag)using Fugene 6 transfection reagent (Roche). After 48 hours,expression was analyzed via Western blotting. The 4E-BP1

mutant was generously provided by Dr. Davide Ruggero(University of California, San Francisco, San Francisco, CA).

In vitro transcriptionThe dicistronic andmonocistronic SKþBluescript vectors

were linearizedwithBamHI and used as templates for in vitrotranscription. For the in vitro translation assay, monocis-tronic templates were transcribed using mMessage mMa-chine T7Ultra (Ambion) producing cappedmRNA. For theRNA transfection assays, dicistronic and monocistronictemplates were transcribed using MEGAScript T7(Ambion) producing either A capped (New England Bio-labs) or uncapped RNA. The m7G cap was added usingScript Cap m7G capping system (CELLSCRIPT, Inc.) andtranscripts were poly(A) tailed using poly(A) polymerasetailing kit (Epicentre) as per manufacturer's instructions. AllmRNAwas extractedwith phenol/chloroform and run on anagarose gel to ensure RNA integrity.

In vitro translationAn aliquot of 0.5 mg of in vitro transcribed mRNA, cap

analog (Ambion), and 1.6 nmol/L methionine was added torabbit reticulocyte lysate (Speed Read; Novagen) and incu-bated for 1 hour at 30�C. The sample was subsequentlyassayed for Photinus and Renilla luciferase activity.

siRNA plasmid transfectionsTen mmol/L of nonsense siRNA (Dharmacon; D-

001206-10-20) or human eIF-4E siRNA (Dharmacon;M-003884-03, J-003884-08, or J-003884-10) were incu-bated in 35-mm plate wells with 12 mL of INTERFERintransfection reagent (PolyPlus-Transfection) at 37�C for 10minutes. HeLa cells were plated at 1.0� 106 in wells alreadycontaining siRNA complexes and serum-free growth media.Complete growth media was then added to a final volume of2 mL. After 48 hours, cells were harvested in cell lysis buffer(Promega) with protease (Roche) and phosphatase inhibitors(Pierce) and analyzed via Western blot analysis.

Cell culture/luciferase assaysWI-38 cells (CCL-75) were obtained by American Type

Culture Collection (ATCC) and cultured in minimumessential medium (MEM) plus 10% FBS and 5% penicil-lin/streptomycin. MCF-7 (HTB-22), HeLa (CCL-2), andHeLa S3 cells (CCL-2.2) were also obtained by ATCC andcultured in Dulbecco's modified Eagle medium (DMEM)plus 10% FBS and 5% penicillin/streptomycin. Humanmammary epithelial cells (HMEC; A10565) were obtainedfrom Invitrogen and cultured in the recommendedmedium.HMEC-t cells were generously provided by Dr. JamesDeGregori (University of Colorado Anschutz Medical Cen-ter, Aurora, CO) and cultured in HMEC medium.MCF10A, MCF12A, and the 21 T series were generouslyprovided by Dr. Heide Ford (University of ColoradoAnschutz Medical Center). MCF10A and MCF12A werecultured as previously described (38). 21PT, 21NT, and 21MT2 were cultured as previously described (39). Cell lineswere validated by short tandem repeat (STR) DNA

Dobson et al.

Mol Cancer Res; 11(8) August 2013 Molecular Cancer Research888




fingerprinting using the AmpF‘STR Identifiler Kit accord-ing to the manufacturer's instructions (Applied Biosystems,cat 4322288). The STR profiles were compared with knownATCC fingerprints (ATCC.org), with the Cell Line Inte-gratedMolecular Authentication database (CLIMA) version0.1.200808 (http://bioinformatics.istge.it/clima/; NucleicAcids Research 37:D925-D932 PMCID: PMC2686526),and with the MD Anderson fingerprint database. The STRprofiles matched known DNA fingerprints or were unique.Cells were transfected with 2 mg of mRNA using Trans-

Messenger transfect kit (Qiagen; RNA/TransMessengerlipid 1:8) as per manufacturer's instruction. After 4 hours,the cells were lysed with 500 mL of lysis buffer (Promega).Forty microliter of the supernatant were used for the lucif-erase assays using theDual-Luciferase Reporter Assay System(Promega) and analyzed in a Luminoskan luminometer.

Western blot analysisCells were harvested in cell lysis buffer (Promega) with

protease (Roche) and phosphatase inhibitors (Pierce). Thecell lysate was analyzed by Western blot analysis by sepa-rating the proteins on a 12% SDS-PAGE, with subsequenttransfer onto nitrocellulose. The membranes were blockedand probed with an antibody directed against Aurora A(35C1; Calbiochem), eukaryotic elongation factor 2 kinase(eEF2K; Cell Signaling Technology), eIF4E (BD Transduc-tion), glyceraldehyde-3-phosphate dehydrogenase (Gapdh;FL-335; Santa Cruz Biotechnology), 4E-BP1 (53H11; CellSignaling Technology), phospho-4E-BP1 (Thr37/46; CellSignaling Technology), or eIF-4G (C45A4; Cell SignalingTechnology) in 5% nonfat dried milk in a solution of PBScontaining 0.1% Tween 20. The blots were then incubatedwith horseradish peroxidase–conjugated secondary antibo-dies (Promega). Immunoreactive bands were detected usingAmersham enhanced chemiluminescence (ECL) plus (GEHealthcare) and quantified using ImageQuant version 5.2(Molecular Dynamics) or ImageJ software (NIH http://rsbweb.nih.gov/ij/).

Polysome analysisCells were treated with cycloheximide (50 ng/mL) for 30

minutes, then harvested on ice in PBS containing 50 ng/mLof cycloheximide, and finally lysed with 400 mL of 100mmol/L KCL, 50 mmol/L Tris–Cl, 1.5 mmol/L MgCl2, 1mmol/L dithiothreitol, 1.5% NP-40, protease inhibitors(Roche), 100 mg/mL cycloheximide, and 100 U RNasinplus RNase inhibitor (Promega). Then 300 mL of lysate wasloaded on a 20% to 60% sucrose gradient created using aBIOCOMP Gradient Station and centrifuged at 39,000rpm for 2 hours at 4�C using a SW40Ti rotor (BeckmanCoulter). The gradient was fractionated and RNA wasextracted from each fraction using TRIzol Reagent (Sigma)followed by PureLink RNA Mini Kit (Invitrogen) andanalyzed by quantitative real-time PCR (qRT-PCR).

RNA extraction/qRT-PCRTotal RNA was extracted using TRIzol Reagent (Sigma)

followed by PureLink RNA Mini Kit (Invitrogen). cDNA

libraries were synthesized using iScript cDNA Synthesis Kit(Bio-Rad), including a (�)RT control. The primer pairsused for the qRT-PCR were as follows: 50-TCTTCACAG-GAGGCAAATCCA-30 (forward) and 50-AATAAGTTA-CACACTCACTCAGGTACTA-30 (reverse) for Aurora AmRNA, 50-ACAGTCAGCCGCATCTTCTT-30 (forward)and 50-GTTAAAAGCAGCCCTGGTGA-30 (reverse) forGapdh mRNA, and 50-AAAGCTCCCAATCATCCAAA-30 (forward) and 50-GAGATGTGACGAACGTGT-30(reverse) for Photinus luciferase mRNA. qRT-PCR wasconducted using a Roche Lightcycler 480 with either Light-Cycler 480 SYBR Green I Master (Roche) or SsoAdvancedSYBR Green Supermix (Bio-Rad) as per manufacturer'sinstructions.

ResultsEnhanced protein synthesis contributes tooverexpression of Aurora A kinaseIncreased Aurora A kinase expression is proposed to

contribute to cellular immortalization and to the epitheli-al–mesenchymal transition (EMT; refs. 36, 37). To identifymechanisms contributing to this enhancement, Aurora Aprotein expression was examined in a variety of cells, focus-ing on breast epithelial cell lines. Aurora A protein levels werequantified in 2 finite lines, normal human lung fibroblasts(WI-38) and HMEC. Aurora A protein expression in thesecells was chosen to represent basal protein levels and com-pared with the following immortalized cell lines: nontu-morigenic breast epithelial cell lines (MCF10A andMCF12A), and tumorigenic epithelial cell lines (MCF-7and HeLa S3) from breast and cervix, respectively. Lysateswere analyzed via Western blotting for Aurora A and Gapdh(as a loading control). Expression of Aurora A kinase in thecontrol cell lines, WI-38 and HMEC, were similar andcomparable with that observed in MCF-7 cells (Fig. 1A).However, Aurora A kinase expression was 2- to 4.2-foldhigher in MCF10A, MCF12A, and HeLa S3 cells (Fig. 1A).Three of 4 immortalized cell lines showed Aurora A proteinlevels similar to those observed in breast and cervical cancer(36).The Aurora A gene is transcribed and the mRNA is

translated during late S-phase and peaks during G2–Mphaseof the cell cycle (4, 5). Therefore, it is possible that differ-ences in the number of cells in these phases from anasynchronous population could contribute to the observeddifferences in Aurora A protein levels. However, fluores-cence-activated cell sorting (FACS) analysis of the cell linesdid not show a correlation with the percentage of cells inG2–Mand Aurora A protein levels (Supplementary Fig. S2).Therefore, alterations in transcription, translation, and/orprotein stability are potentially contributing to increasedAurora A kinase expression in MCF10A, MCF12A, andHeLa S3 cells.To quantify Aurora A mRNA levels, cDNA libraries were

constructed from total RNA isolated from the individual celllines. Aurora A transcript levels from each cell line weremeasured by qRT-PCR and showed that mRNA levels were

Misregulation of IRES-Dependent Translation of Aurora A mRNA

www.aacrjournals.org Mol Cancer Res; 11(8) August 2013 889




similar between each cell line except for a modest increase inMCF-7 cells (Fig. 1B). The ratio of Aurora A protein tomRNA within each cell line was significantly elevated(ranging from approximately 2–5-fold higher) in MCF10A,MCF12A, and HeLa S3 cells compared with WI-38 andHMEC cells, and even higher when compared MCF-7 cells(Fig. 1C). These results indicate that increased Aurora Aprotein expression in theMCF10A,MCF12A, andHeLa S3cells is not due to enhanced transcription of the Aurora Agene or Aurora AmRNA stability, but the result of increasedprotein synthesis and/or protein stability.To identify if alterations in protein stability contributed to

the differential Aurora A expression, the half-life of theAurora A protein was determined. Cells were harvested at2 to 12 hours after being treated with cycloheximide. TheAurora A protein levels were then analyzed via Westernblotting (Supplementary Fig. S3) with Gapdh as a loadingcontrol as it has a half-life of 90 to 120 hours (40). The half-life of the Aurora A protein was similar between the cell lines,

ranging between 2.2 to 2.9 hours (Fig. 1D). This range isconsistent with a previous study that found the Aurora Aprotein half-life to be approximately 2.5 hours (5). Inaddition, this result indicates that differences in proteinstability do not contribute to the differential expression ofAurora A protein.By order of elimination, the previous results implicated

protein synthesis as one of the remaining mechanismscontributing to the variable Aurora A protein levels. Toconfirm this hypothesis, a polysome gradient analysis wasconducted. Aurora A mRNA levels were quantified innonpolysomal, low-molecular weight (LMW) polysome,and high-molecular weight (HMW) polysome fractionsbetween the low Aurora A protein–expressing MCF-7 cellsand high Aurora A protein–expressing MCF12A cells.Association with HMW fractions suggests increased trans-lation initiation and/or reinitiation as the result of moreefficient loading of the ribosomes onto the mRNA (41).Aurora A mRNA was most concentrated in the HMW

Figure 1. Enhanced translationcontributes to overexpression ofAurora A protein in a subset ofimmortalized cell lines. A,endogenous Aurora A proteinlevels were analyzed via Westernblotting and normalized to Gapdh.The Aurora A/Gapdh ratio fromeach cell line was compared withthe ratio from WI-38 cells, whichwas set to 1. n ¼ 6 � SD. B, totalRNA was isolated and analyzed viaqRT-PCR with Gapdh as areference target. Results werenormalized to the Aurora A mRNAlevel in WI-38 cells. n ¼ 3 � SD. C,the ratio of the normalized Aurora Aprotein levels to normalized AuroraA mRNA levels are shown. D,quantitation from Western blottingof Aurora A protein expressionlevels in cells treated for 0 to 12.5hours with cycloheximide.Expression level of Aurora Aprotein in the untreated cells (0hour) was normalized to 100. n ¼ 3� SD. E, UV detection readout ofsucrose gradient fractionation ofMCF12A and MCF-7 cells (top).Quantitation of Aurora A mRNAlevels associated with the differentgradient fractions. Levels weremeasured using qRT-PCR afterisolation of total RNAfrom each fraction (bottom).�, P < 0.05; ��, P < 0.005; Studentt test.

Dobson et al.





fractions from MCF12A cells. In contrast, the majority ofAurora A mRNA levels from MCF-7 cells resided with theLMW fractions (Fig. 1E). These results indicated transla-tional upregulation as a contributing mechanism toenhanced protein expression in MCF12A cells comparedwith MCF-7 cells.

Cap-dependent translation initiation is increased in theimmortalized cell linesThe major mechanism by which translation is regulated is

at the step of initiation. And the primary mode by whichtranslation is initiated is through recognition of them7G capstructure (42, 43). Misregulating cap-dependent translationcan lead to elevated levels of protein synthesis and tumor-igenesis. For example, increased expression of the rate-limiting canonical factor eIF-4E (which binds the capstructure) is found in many tumors (44–46). Furthermore,ectopically induced overexpression of eIF-4E has transform-ing capabilities leading to malignant phenotypes (47–49).To determine if elevated eIF-4E levels were contributing tothe overexpression of Aurora A, Western blot analyses oflysates were analyzed. eIF-4E levels were similar in all 6 celllines (Supplementary Fig. S4A). This observation suggeststhat a global increase in cap-dependent translation may notbe occurring. On the other hand, the eIF-4E that is presentmay be differentially restricted in its ability to bind the capstructure.Steady-state expression of 4E-BP1, a negative regulator of

eIF-4E, is often decreased in cancer cells (50). Analysis of 4E-BP1 protein levels viaWestern blotting showed there was nodifference between 4E-BP1 levels in WI-38, MCF12A, orMCF-7 cells (Supplementary Fig. S4B). However, there wasactually a 1.5- to 2.2-fold increase in the expression of 4E-BP1 in HMEC, MCF10A, and HeLa S3 cells. The phos-phorylation state of 4E-BP1 is integral to its activity with thehypophosphorylated 4E-BP1 able to bind eIF-4E and inhib-it cap-dependent translation. Interestingly, there was a 1.5-to 3-fold increase of hypophosphorylated 4E-BP1 in all thecell lines compared with WI-38 cells (Supplementary Fig.S4B, right). Taken together, these results do not support thehypothesis that cap structure accessibility is altered in thedifferent cell lines.The scaffolding protein eIF-4G is the key structural

component in the preinitiation complex. Its expression levelis proposed to contribute to the rate of translation (51).However, eIF-4G levels did not correlate with Aurora Aprotein levels as shown by the increased expression of eIF-4Gin MCF-7 cells compared with WI-38 and HMEC cells(Supplementary Fig. S4C). In addition, eIF-4G proteinexpression was low in MCF-12A cells.The expression patterns of the critical cap-dependent

regulatory proteins eIF-4E, 4EBP, and eIF-4G did not yieldany candidates that may be contributing to the overexpres-sion of Aurora A kinase. To definitively compare cap-dependent translation initiation between these cell lines, weused a reporter assay. In vitro transcribed mRNA frommonocistronic DNA containing the Photinus luciferase openreading frame (ORF), the b-globin 50 leader, and the SV40

poly-adenylation (A) site was created. The b-globin 50 leaderwas chosen because it is short (53 nt), unstructured, withoutan upstream ORF (uORF), and the b-globin mRNA isexclusively translated in a cap-dependent manner (43). Thetranscripts were capped with a m7G cap, poly(A) tailed, andtransfected into each cell line. After 4 hours, the cells wereharvested and assayed for Photinus luciferase activity. Theluciferase activity from each cell line was compared withtranscript levels quantitated using qRT-PCR to normalizefor transfection efficiency as well as mRNA stability. Trans-lation of the reporter was enhanced by 1.9- to 2.6-fold in allof the immortalized cell lines compared with the normal cells(Fig. 2A). Interestingly, MCF-7 cells, which do not over-express the Aurora A protein, elicited the second largestincrease of cap-dependent translation compared withWI-38and HMEC cells (Fig. 2A). This result indicates that othermechanisms aside from eIF-4E, 4E-BP, and eIF-4G expres-sion are contributing to enhanced cap-dependent transla-tion. However, increased cap-dependent translation did notcorrelate with increased expression of Aurora A protein,thereby suggesting that an alternate mechanism may beregulating translation of the Aurora A mRNA.

Aurora A protein expression level is unaffected byinhibiting cap-dependent translation initiationTo further determine the role of cap-dependent transla-

tion initiation in the synthesis of the Aurora A protein, cap-dependent translation was inhibited using 2 differentapproaches. Initially, eIF-4E expression was knocked down.HeLa cells were transfected with siRNA (Dharmacon)targeting eIF-4E mRNA or nonsense siRNA for 48 hours.Quantification of Western blot analyses showed that eIF-4Eprotein expression was reduced by nearly 70% (Fig. 2B).eEF2K expression in the eIF-4E siRNA-treated cellsdecreased by 57%, similar to previous observations (52).In contrast, the level of Aurora A protein was equivalent inthe eIF-4E and nonsense siRNA conditions (Fig. 2B).As a second approach to modulate cap-dependent trans-

lation, HeLa cells were transfected with a DNA plasmidencoding a hypophosphorylatedmutant of 4E-BP1 in whichthe main 5 phosphorylation sites are mutated rendering theprotein nonphosphorylated (generously provided by Dr.Davide Ruggero, University of California, San Francisco)or a control vector (Paltag). The constitutively expressedhypophosphorylated 4E-BP1 should sequester eIF-4E,thereby disrupting eIF-4F formation and inhibit cap-depen-dent translation (53). Forty-eight hours after transfection,cells were harvested and protein levels were analyzed byWestern blotting. eEF2K expression decreased by 75% inthe presence of hypophosphorylated 4E-BP1 (Fig. 2C). eIF-4E expression decreased by 53%, similar to previous results(52), indicating that the eIF-4E mRNA is translated via acap-dependent mechanism and a decrease in cap-dependenttranslation has occurred. On the other hand, the Aurora Aprotein level decreased by only 16% (Fig. 2C). Takentogether, these results show that synthesis of Aurora A kinaseprotein is relatively unaffected when cap-dependent trans-lation is inhibited inHeLa cells indicating that an alternative






mechanismmay be contributing to the translation initiationof the Aurora A mRNA.To determine if cap-dependent translation regulates

expression of the Aurora A protein in other cells, 2 siRNAs

targeting eIF-4E were individually transfected into WI-38,HMEC, MCF10A, MCF12A, and MCF-7 cells. eIF-4Ewas significantly knocked down by each siRNA in all celllines (Fig. 2D and E). Aurora A protein expression was

Figure 2. Inhibiting cap-dependenttranslation initiation differentiallyaffects Aurora A proteinexpression. A, translation of aPhotinus luciferase reporter mRNAcontaining the b-globin 50 leaderwas transfected into the 6 cell lines.Luciferase activity was measuredand themRNA quantitated by qRT-PCR after 7 hours. Shown is theratio of luciferase activity toluciferase RNA. n¼ 3 in triplicate�SD. B, Western blot analyses oflysates fromHeLa cells transfectedwith nonsense siRNA or siRNAtargeting eIF-4E for 48 hours. n¼ 3� SD. C, Western blot analyses oflysates fromHeLa cells transfectedwith a control vector or a plasmidencoding for mutanthypophosphorylated 4E-BP1 for48 hours. n ¼ 3 � SD. D, Westernblot analyses of lysates from WI-38,HMEC,MCF10A,MCF12A, andMCF-7 cells transfected withnonsense siRNA, siRNA1 targetingeIF-4E, or siRNA 2 targeting eIF-4Efor 48 hours. E, quantification ofeIF-4E knockdowns in WI-38,HMEC, MCF10A, MCF12A, andMCF-7 cells. n ¼ 3 � SD;��, P < 0.005; ��, P < 0.0001;Student t test.

Dobson et al.





significantly decreased in MCF-7 cells, indicating cap-dependent initiation is the primary mechanism for transla-tion of Aurora A mRNA in these cells. However, Aurora Aprotein levels were unaffected when eIF-4E levels werereduced in WI-38, HMEC, MCF10A, and MCF12A cells(Fig. 2D and E). Taken together, these results suggest analternate mechanism may be used to initiate translation ofthe Aurora A mRNA.

The Aurora A 50 leader contains an IRESCap-dependent translation did not seem to be a significant

mechanism for the synthesis of Aurora A protein in theHMEC cells and in the cell lines in which Aurora A kinasewas overexpressed. This result indicated it may be translatedin a cap-independent manner. To determine if the Aurora A50 leader has an IRES, it was inserted into the intercistronicregion of a dicistronic luciferase construct coding for Renillaand Photinus luciferase in the first and second cistrons,respectively (29). The b-globin 50 leader, which does notcontain an IRES, was used as a negative control. Two viralIRESes, the encephalomyocarditis virus (EMCV) IRES andthe IRES in the intergenic region of the cricket paralysis virus

(CrPV) were chosen as positive controls (54, 55) Theconstructs were in vitro transcribed to eliminate the possi-bility of cryptic promoter activity or alternative splicing (56,57). The resulting dicistronic transcripts were transfectedinto HeLa cells. After 7 hours, the cells were harvested andassayed for Renilla and Photinus luciferase activity. ThePhotinus:Renilla (P:R) luciferase ratio obtained from theb-globin dicistronic construct was normalized to 1. Conse-quently, a ratio above 1 would indicate IRES activity. TheP:R ratio obtained from the dicistronic luciferase mRNAcontaining the Aurora A 50 leader was 5-fold higher than thatobtained from the dicistronic mRNA containing the b-glo-bin 50 leader (Fig. 3A, left). This ratio was similar to the oneobtained from the mRNA containing the CrPV IRES, butless than the 20-fold increase in the P:R ratio from themRNA containing the EMCV IRES. These results indicatedthat the Aurora A 50 leader contains an IRES, which exhibitsactivity similar to the CrPV IRES in HeLa cells.To further validate the presence of an IRES in the Aurora

A 50 leader, in vitro transcribed monocistronic mRNAcontaining the Photinus luciferase ORF and the b-globinor Aurora A 50 leader was translated in rabbit reticulocyte

Figure 3. The 50 leader of the Aurora A mRNA contains an IRES. A, dicistronic luciferase mRNA (left) containing the b-globin, Aurora A, CrPV, and EMCV 50

leaders or IRES elements were transfected individually into HeLa cells. Luciferase activity is shown as the ratio of P:R and is normalized to that obtained fromthe dicistronic mRNA containing the b-globin 50 leader. n ¼ 3 in triplicate � SD. Monocistronic Photinus luciferase mRNA (right) containing the b-globin orAurora A 50 leader were translated in rabbit reticulocyte lysate in the presence of increasing concentrations of cap analog. The initial level of Photinusluciferase activity from each monocistronic mRNA was normalized to 100. n¼ 3� SD. B, cell-cycle progression following the release of a double thymidineblock inHeLa cells was confirmed by flowcytometry analysis (left). HeLa cells transfectedwith dicistronic luciferasemRNAcontaining the b-globin andAuroraA 50 leaders were synchronized in G1–S with a double thymidine block and released. Luciferase activity is shown as the ratio of P:R and is normalized to thatobtained from the dicistronic mRNA containing the b-globin 50 leader at the 0 time point (right). n ¼ 3 in triplicate � SD; ��, P < 0.0001; Student t test.






lysate. Cap-dependent translation was inhibited by addingincreasing amounts of cap analog to compete with the m7Gcap. Luciferase activity measured from the b-globin con-struct decreased by 70% in the presence of 100 mmol/L ofcap analog, whereas luciferase activity from the Aurora Aconstruct was relatively unchanged (Fig. 3A, right). Thisresult showed that the Aurora A 50 leader initiates translationindependently of the cap structure and supports the hypoth-esis that it contains an IRES.Expression levels of both Aurora A mRNA and Aurora A

protein peak during the G2–M phase of the cell cycle (4, 5).This temporal expression coincides with reports indicatingthat IRES activity is regulated through the cell cycle andpeaks at G2–M (58, 59). To determine if the Aurora A IRESis regulated by the cell cycle, HeLa cells were synchronizedby a double thymidine block. During the last thymidineblock the cells were transfected with in vitro transcribeddicistronic mRNA containing either the b-globin or AuroraA 50 leader in the intercistronic region. Cells were harvestedbetween 0 and 8 hours following release from the doublethymidine block into normal growth medium with the totaltransfection time being 11 hours for all cells. FACS analysisshowed that the majority of cells were synchronized in G1–Sand proceeded through the cell cycle, entering G2–M atapproximately 6 to 8 hours after release (Fig. 3B, left). The P:R ratio obtained from the b-globin construct at the time ofthymidine release was set to 1 and did not change during thephases of the cell cycle. On the other hand, immediately afterthymidine release, the Aurora A P:R ratio was 4.2-fold higherthan that obtained with b-globin. Themaximum P:R ratio, a6.4-fold increase occurred 8 hours later (Fig. 3B, right). Thisresult indicates that the Aurora A IRES is active throughoutthe cell cycle but peaks during the same portion of the cellcycle in which translation of the endogenous Aurora AmRNA occurs—G2–M.

Aurora A IRES activity is increased in the cell lines thatoverexpress Aurora A proteinTo determine if Aurora A IRES activity parallels Aurora A

protein expression, dicistronic mRNA containing the Auro-ra A or b-globin 50leaders was transfected into all 6 cell lines.After 7 hours, the cells were harvested and assayed for Renillaand Photinus luciferase activity. The P:R ratio for the mRNAcontaining the Aurora A 50 leader was at least 2-fold higherthan the control in the 6 cell lines indicating the Aurora AIRES is operative in all cells (Fig. 4A). Moreover, the AuroraA P:R ratio was highest in the cell lines that also over-expressed the Aurora A protein (Fig. 4A). These results showa positive correlation between Aurora A protein expressionand Aurora A IRES activity. This correlation supports thehypothesis that IRES-dependent translation is contributingto the overexpression of the Aurora A protein observed in thesubset of cell lines.All eukaryotic transcripts contain a m7G cap. According-

ly, translation of the Aurora A mRNA may be initiatedthrough the cap structure and/or the IRES.To quantitate thecontribution of the Aurora A IRES in a monocistronicmRNA, the b-globin or Aurora A 50 leader were placed

upstream of the Photinus luciferase ORF and in vitro tran-scribed. The transcripts were capped with an m7G orApppG cap (A cap) and poly(A) tailed (Fig. 4B). The Acap is not recognized by eIF-4E preventing cap-dependenttranslation initiation of the mRNA. In addition, the poly(A)–specific 30 exonuclease PARN does not recognize the Acap, thereby inhibiting deadenylation and preventing both50 and 30 exonucleolytic mRNA decay (60, 61). The mRNAwas cotransfected into each cell line with a m7G-capped andpoly(A) tailed mRNA containing the humanized Renillaluciferase ORF to normalize for transfection efficiency. Foreach cell line, the P:R ratio obtained from the m7G capPhotinus constructs was set to 100. The P:R ratio obtainedfrom the A cap Photinus luciferase mRNA was comparedwith theP:R ratio obtained from the correspondingm7GcapPhotinus luciferase constructs. The A cap reporter mRNAcontaining the b-globin leader was poorly translated (5% to16% of the m7G cap P:R ratio; Fig. 4C, left). This resultshowed that the translation of the mRNA containing theb-globin 50 leader was dependent on the presence of them7G cap structure. In the WI-38 and HMEC cells, trans-lation of the A cap mRNA containing the Aurora A 50 leaderwas higher than the mRNA with the b-globin leader. Thisresult indicates that the Aurora A 50 leader was initiating cap-independent translation, although it was only approximately25% of that obtained by a m7G cap mRNA containing theAurora A 50 leader (Fig. 4C, right). On the other hand, the Acap P:R ratio from the Aurora A construct was considerablyhigher in the cell lines that overexpress the Aurora A protein(MCF10A, MCF12A, and HeLa S3). Remarkably, in theMCF12A cells, Photinus luciferase activity from the A capmRNAwas 95% of that obtained from the m7G cap AuroraA mRNA (Fig. 4C). These results suggest cap-dependenttranslation is most likely a contributing mechanism forAurora A protein synthesis in the low-expressing lines, butIRES-dependent translation is the predominant mechanismin the Aurora A overexpressing lines.To determine if IRES activity in general was upregulated

in the Aurora A overexpressing cell lines, the experiment wasrepeated using IRESes from the PITSLRE kinase, EMCV,and FMR1 mRNAs. Although all 3 IRESes are used inproliferating cells, PITSLRE and EMCV IRES activity isupregulated during G2–M phase of the cell cycle, whereasthe FMR1 IRES is predominantly used in postmitoticneurons (52, 62). Internal initiation mediated by the FMR1IRES varied between the cell lines but the activity did notcorrelate with that observed from the Aurora A IRES(Supplementary Fig. S5A). FMR1 IRES activity was similarin HMEC, MCF12A and MCF-7 cell lines, significantlyelevated in MCF10A, slightly decreased in WI-38 cells, andsignificantly decreased in HeLa S3 cells. Interestingly, activ-ity from the EMCV and PITSLRE IRESes also did notcorrelate with that observed from the Aurora A IRES(Supplementary Fig. S5B and S5C, respectively). EMCVIRES activity was significantly lower inMCF10A andMCF-7 cells compared with HMEC cells. There was no differencein activity betweenWI-38,HMEC,MCF12A, andHeLa S3cells. On the other hand, activity from the PITSLRE IRES

Dobson et al.





was significantly elevated inMCF12A and HeLa S3 cells yetsignificantly decreased in MCF-7 cells compared with thenormal cells, similar to Aurora A IRES activity. However,unlike the Aurora A IRES, the PITSLRE IRESwas less activein MCF10A cells. In addition, PITSLRE IRES activity washigher in theWI-38 andHMEC cells comparedwith AuroraA IRES activity. These results indicate that not all IRESes areupregulated in the Aurora A overexpressing cell lines. Fur-thermore, they suggest possible variations in the regulationof IRESes that are upregulated under similar circumstances,such as the G2–M phase of the cell cycle.To determine if the correlation between Aurora A IRES

activity and protein expression extends to other cell lines, weconducted the same assay with 3 lines from the 21T cell lineseries (21PT, 21NT, and 21MT2). These cell lines werederived from a single patient diagnosed with infiltratingand intraductal mammary carcinoma and are often used asan in vitro model for cancer progression. The nontumori-genic 21PT and the tumorigenic 21NT cell lines werederived from a primary tumor. The 21MT2 line was derivedfrom a metastatic tumor (63–65). All 3 cell lines overexpress

Aurora A protein (Fig. 5A), yet mRNA levels and proteinhalf-life (2.8–3.1 hours) were equivalent to the normalcell lines (Fig. 5B and D). The elevated Aurora A pro-tein/mRNA ratio indicated an increase in Aurora A trans-lation (Fig. 5C). And the A/m7G cap monocistronic trans-fection experiment showed a high level of Aurora A IRESusage. The percentage of A cap–mediated translation to thatobtained from the m7G cap with the Aurora A 50 leaderranged from 60% (21PT) to 83%–85% (21MT2 and21NTm respectively; Fig. 5E, bottom). These results aresimilar to what was observed with the other Aurora-Aoverexpressing cell lines.To resolve whether increased proliferative activity con-

tributed to alterations in Aurora A IRES activity, this assaywas conducted in aHMECcell line that ectopically expressestelomerase (HMEC-t); a cellular process that prevents rep-licative cellular aging. Overexpressing telomerase did notalter Aurora A protein levels, mRNA levels, or proteinstability (Fig. 5A–D). In addition, the A cap P:R ratioexhibited by these cells was similar to that obtained fromWI-38 and HMEC cells (Fig. 5E).

Figure 4. Aurora A IRES activity is increased in the subset of immortalized cell lines that exhibit enhanced Aurora A protein expression. A, dicistronic mRNAcontaining theb-globin andAurora 50 leaderswere transfected into each cell line. TheP:R ratio obtained from the b-globin construct fromeach cell linewas setto 1. The P:R ratio from the Aurora A construct was normalized to b-globin. n¼ 3 in triplicate� SD. B, schematic representation of cotransfections of an A- orm7G-capped monocistronic Photinus luciferase mRNA containing the Aurora A or b-globin 50 leaders with an m7G-capped Renilla luciferase mRNA. C,normalized luciferase activity obtained from theA-cappedPhotinus luciferase constructs are represented as a percentage of the normalized luciferase activityobtained from them7G-cappedPhotinus luciferase construct for the b-globin 50 leader (left) or the Aurora A 50 leader (right). n¼ 3 in triplicate�SD; �,P < 0.05;��, P < 0.0001; Student t test.






In summary, these results show that IRES-dependenttranslation initiation mediated by the Aurora A 50leader isused to a greater extent in cells overexpressing the Aurora Aprotein. Simply extending proliferative activity does not alterAurora A IRES activity or protein levels. These resultssupport the hypothesis that enhanced IRES-dependenttranslation of the Aurora A mRNA contributes to theincrease of Aurora A kinase expression, which predisposesthe cell to immortalization and tumorigenesis.

DiscussionOverexpression of the Aurora A kinase is observed in a

broad range of malignancies including breast, brain, andpancreatic tumors (8–10, 66–68). Misregulation of itsexpression is proposed to be both an early event in theimmortalization of the cell as well as a contributor to theEMT (37, 69). In the present report, the goal was to identifymechanism(s) contributing to Aurora A protein overexpres-

sion. To this end, we chose 6 cell lines, which includedprimary, immortalized, and tumorigenic cell types. Theywere categorized as either high or low Aurora A protein–expressing cells. Transcription/mRNA stability, protein sta-bility, nor cap-dependent translation could account forenhanced levels of Aurora A protein in the high-expressinglines. On the other hand, we identified an IRES in theAurora A 50 leader and IRES activity strongly correlated withAurora A protein expression.Using RNA dicistronic luciferase constructs, Aurora A

IRES activity was shown to correlate with Aurora A proteinexpression. However, this assay does not measure the con-tribution of the IRES in an m7G-capped monocistronicmRNA, the state in which the Aurora AmRNA is present. Ithas been suggested that eukaryotic IRESes initiate transla-tion at a considerably reduced rate compared with m7G cap-dependent initiation (21, 70). Thus, even if IRES activity isenhanced in a subset of cell lines, the overall increasemay not

Figure5. Additional breast epithelialcell lines show a correlationbetweenAurora A IRESactivity andAurora A protein expression. A,endogenous Aurora A proteinlevels were analyzed via Westernblotting and normalized to Gapdh.The Aurora A/Gapdh protein ratiofrom each cell line was comparedwith the ratio from WI-38, whichwas set to 1. n ¼ 3 � SD. B, totalRNA was isolated from the 4 celllines and analyzed via qRT-PCRwith Gapdh as a reference target.Results were normalized to theAurora AmRNA level inWI-38 cells.n ¼ 3 � SD. C, the ratio of thenormalized Aurora A protein levelsto normalized Aurora A mRNAlevels are shown. D, quantitationfrom Western blotting of Aurora Aprotein expression levels from the4cell lines treated for 0 to 12.5 hourswith cycloheximide. Expressionlevel of Aurora A protein in theuntreated cells (0 hour) wasnormalized to 100. n ¼ 3 � SD. E,normalized luciferase activity fromthe A-capped Photinus luciferaseconstructs are represented as apercentage of the normalizedluciferase activity obtained fromthe m7G-capped Photinusluciferase construct for theb-globin 50 leader (top) or theAurora A 50 leader (bottom). n¼ 3 intriplicate � SD; ��, P < 0.005;��, P < 0.0001; Student t test.

Dobson et al.





be physiologically significant. To address this issue, wecreated monocistronic mRNA containing the Aurora AIRES with an A or m7G cap. These RNAs would permitcomparison between translation initiated in a cap-indepen-dent manner to translation mediated by an m7G cap and anIRES (60). The resulting A cap/m7G cap ratio we termed"IRES usage." In WI-38 and HMEC cells, A-capped trans-lation of the RNA containing the Aurora 50 leader was lessthan 25% of that obtained from an m7G-capped mRNA(Fig. 4C). This result indicated that cap-dependent trans-lation is likely the predominant mechanism. However,transfection into the other cell lines yielded dramaticallydifferent results. InMCF12A cells, translation of themRNAwas similar irrespective of the cap structure (Fig. 4C). This isone of the first examples whereby a eukaryotic IRES initiatestranslation at a rate similar to anm7G-capped transcript.Weinterpret this result as showing that IRES-dependent trans-lation is elevated and is the principal mechanism used toinitiate translation of the m7G-capped mRNAs. Alterna-tively, reduced cap-dependent translation without any con-comitant alteration in IRES activity could yield a similarresult. To differentiate between these 2 explanations, weexamined cap-dependent translation of a monocistronicmRNA and found that it was relatively similar between the4 immortalized lines and actually higher than the 2 primarycell lines (Fig. 2A). Accordingly, these results indicate anapparent switch from low level cap-dependent translation ofthe Aurora A mRNA in normal, finite lifespan cells toincreased IRES-dependent translation in immortalized cellsbefore malignant transformation. Identifying the mecha-nism(s) that play a role in this switch could be crucial tounderstanding oncogenesis mediated by Aurora A kinaseoverexpression and possibly of other oncogenes, which aretranslated in an IRES-dependent manner. For example, asubset of breast tumors exhibiting overexpression of HER2and EGFR has been implicated in affecting Aurora A kinaseexpression (69).Examination of additional cell lines reinforced the positive

correlation between IRES usage and Aurora A proteinexpression. The 21T series exhibited both elevated AuroraA protein levels and IRES activity; whereas Aurora mRNAand protein half-life was similar to thefinite cell lines (see Fig.5). This group of cells included immortalized, but nontu-morigenic (21PT), tumorigenic (21NT), and finally meta-static (21MT2). These results are consistent with previousreports suggesting enhanced Aurora A expression contri-butes to immortalization and continues through the EMT(37, 69).Internal initiation of translation has been invoked as a

mechanism involved in carcinogenesis (reviewed in ref. 71).Multiple mRNAs encoding proteins contributing to cellproliferation [FGF2 and platelet-derived growth factor 2(PDGF2)], cell cycle (PITSLRE p58), cell death/survival(Bcl-X and Apaf1), and tumor development (p53, c-myc,and c-jun) contain IRESes (28, 29, 32, 72–75). Misregulat-ing IRES-dependent translation contributes to cancer pro-gression throughmultiplemeans. For example, in the diseaseX-linked dyskeratosis congenita, mutations in the dyskerin

gene reduce pseudouridylation of rRNA, which inhibitsIRES-dependent translation of tumor suppressors (p53 andp27) and apoptotic factors (Bcl-X and XIAP) predisposingindividuals to cancer (76–78). Alternatively, the low oxygenenvironment in the center of solid tumors increases IRES-dependent translation of the VEGF mRNA, which in turnpromotes angiogenesis and tumor growth (79). In thepresent study, we found that the Aurora A IRES is regulatedthrough the cell cycle and by cell type. As a first step towardidentifying the mechanism, other IRESes that are upregu-lated in G2–M were examined. However, there was nocorrelation of IRES usage between the PITSLRE, EMCV,and the Aurora A IRES. It would be of interest to determineif there is a subset of IRESes, perhaps those encoding otheroncogenes, which exhibit IRES usage patterns similar toAurora A.The mechanism regulating the Aurora A IRES is

unknown. Identifying cis-elements that control IRES activ-ity would be informative. Viral IRESes generally rely onextensive secondary structure, a feature that is often con-served phylogenetically, although the sequences can vary(80–82). The presence of uORFs is observed as well (23, 83).Alternatively, eukaryotic IRESes are quite variable; sequenceand/or secondary structure does not imply the presence of anIRES. Previous studies have reported eukaryotic IRESmotifsas short as 9 to 50 nt (84, 85), but they can extend to over1,000 nt (86).Viral IRESes exhibit a variable requirement for trans-

factors. Some IRESes recruit the ribosome directly, whereasothers use both canonical factors (including eIF-4E incertain situations) and IRES-transacting factors (ITAF) suchas polypyrimidine-tract—binding (PTB) protein or La pro-tein (54, 87–89). The rate-limiting step for eukaryoticIRESes is proposed to be the expression level of ITAFs(reviewed in ref. 90). These are RNA-binding proteins thataid in the recruitment of the translational machinery and/oralter the RNA secondary structure, which in turn promotesbinding of the translation complex. ITAFmis-expression canalter translation of cancer-related mRNAs. For example,murine double minute (MDM2) is an oncoprotein thatbinds the IRES in the mRNA encoding the XIAP. IncreasedXIAP expression in the MDM2 overexpressing cells leads toa resistance to radiation-induced apoptosis (91). Alterna-tively, a loss of 2 other ITAFs, TCP80 and RHA diminishesp53 IRES activity and protein expression; promoting cellsurvival in response toDNAdamage (92). Presumably, thereare ITAFs whose increased expression is responsible for theenhanced Aurora A IRES activity. Because there is no in vivoassay to quantify Aurora A IRES activity in normal/immor-talized cells, the level of these regulatory proteins would be auseful biomarker and provide a novel drug target to mod-ulate Aurora A IRES activity in cancer cells. Two ITAFsknown to be expressed inG2–Mare PTB and upstream of n-ras (Unr). They both bind the PISTRLE p58 IRES andregulate the G2–M–specific expression of the PISTLRE p58protein (59, 93). They can also regulate IRESes during otherphases of the cell cycle, such as the Apaf-1 IRES during G1(35). As noted earlier, there was no correlation between






PITSLRE andAurora A IRES usage. Furthermore, knockingdown Unr did not affect Aurora A IRES activity (Unpub-lished Data) and there does not seem to be any potentialPTB-binding sites in the Aurora A 50 leader. Consequently,there are likely to be additional ITAFs that regulate theAurora A and possibly other mitotic IRESes. Indeed, itwould be of interest to determine whether the Aurora AIRES represents another subset ofG2–Mfunctionally relatedIRESes.Current anticancer drugs target Aurora A kinase activity as

its inhibition leads to formation of a monopolar spindle andcell death (94). However, it has been difficult to designreagents that exhibit specificity for the ATP pocket of theAurora A kinase (95, 96). Moreover, overexpressing a kinasedead mutant of Aurora A can still contribute to cellularimmortalization (97).Targeting the Aurora A mRNA could be an effective

alternative approach to regulate expression of the kinase.HMEC (and WI-38 cells) yielded low Aurora A IRESactivity, but surprisingly Aurora A protein expression wasrelatively unaffected when eIF-4E levels were reduced. Thisresult indicates that endogenous Aurora AmRNA in normalcells (e.g., HMEC) can be translated in both a cap- andIRES-dependent manner. We postulate that the HMECcells use both translation initiation mechanisms to ensureexpression of the Aurora A kinase. Inhibiting cap-dependenttranslation in these cells is compensated by the normalupregulation of IRES activity during G2–M. On the otherhand, Aurora A expression was sensitive to the reduction ofeIF-4E inMCF-7 cells. Aurora A IRES activity (and proteinexpression) remains low and cannot compensate for thereduction in cap-dependent translation. This result suggeststhat reducing cap-dependent translation would decreaseendogenous Aurora A protein in tumors that cannot usethe Aurora A IRES. Finally, IRES-dependent translation wasincreased in the Aurora A protein overexpressing cells and isthe principal translation initiationmechanism. Consequent-

ly, designing reagents targeting the IRES would inhibitAurora A protein expression primarily in cancer cells thatoverexpress the kinase.In summary, we have identified an IRES situated in the 50

leader of the human Aurora A mRNA. In an examination ofmultiple cell lines, IRES activity was the only mechanismthat correlated with Aurora A protein expression. Moreover,this mechanism seems to be upregulated early during cancerdevelopment and remains elevated as cell transformationadvances indicating that targeting this mechanism may bebeneficial for multiple stages of cancer progression.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: T. Dobson, L.A. KrushelDevelopment of methodology: T. Dobson, L.A. KrushelAcquisition of data (provided animals, acquired and managed patients, providedfacilities, etc.): T. Dobson, J. ChenAnalysis and interpretation of data (e.g., statistical analysis, biostatistics, compu-tational analysis): T. Dobson, J. Chen, L.A. KrushelWriting, review, and/or revision of the manuscript: T. Dobson, L.A. KrushelAdministrative, technical, or material support (i.e., reporting or organizing data,constructing databases): T. DobsonStudy supervision: L.A. Krushel

AcknowledgmentsThe authors thank Drs. Jessica Tyler for helpful comments on the article, Ford and

DeGregori for cell lines, and Ruggero for the 4E-BP1 mutant construct. The authorsalso thank Shihuang (Sam) Su and the other members of the Krushel laboratory fortheir crucial assistance. STRDNA fingerprinting was conducted by the Cancer CenterSupport grant funded Characterized Cell Line core, NCI # CA16672.

Grant SupportThis work was supported by the NIH (AG02815G; to L.A. Krushel).The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be herebymarked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

Received December 26, 2012; revised March 18, 2013; accepted April 11, 2013;published OnlineFirst May 9, 2013.

References1. Adams RR, Carmena M, Earnshaw WC. Chromosomal passengers

and the (aurora) ABCs of mitosis. Trends Cell Biol 2001;11:49–54.2. Nigg EA. Mitotic kinases as regulators of cell division and its check-

points. Nat Rev Mol Cell Biol 2001;2:21–32.3. Fu J, BianM, Jiang Q, Zhang C. Roles of Aurora kinases in mitosis and

tumorigenesis. Mol Cancer Res 2007;5:1–10.4. Tanaka M, Ueda A, Kanamori H, Ideguchi H, Yang J, Kitajima S, et al.

Cell-cycle–dependent regulation of human aurora A transcription ismediated by periodic repression of E4TF1. J Biol Chem 2002;277:10719–26.

5. Honda K, Mihara H, Kato Y, Yamaguchi A, Tanaka H, Yasuda H, et al.Degradation of human Aurora2 protein kinase by the anaphase-pro-moting complex-ubiquitin–proteasome pathway. Oncogene 2000;19:2812–9.

6. Littlepage LE, Wu H, Andresson T, Deanehan JK, Amundadottir LT,Ruderman JV. Identification of phosphorylated residues that affect theactivity of the mitotic kinase Aurora-A. Proc Natl Acad Sci U S A2002;99:15440–5.

7. Zhou H, Kuang J, Zhong L, Kuo WL, Gray JW, Sahin A, et al. Tumouramplified kinase STK15/BTAK induces centrosome amplification,aneuploidy and transformation. Nat Genet 1998;20:189–93.

8. Bischoff JR, Anderson L, Zhu Y, Mossie K, Ng L, Souza B, et al. Ahomologue of Drosophila aurora kinase is oncogenic and amplified inhuman colorectal cancers. EMBO J 1998;17:3052–65.

9. Gritsko TM, Coppola D, Paciga JE, Yang L, Sun M, Shelley SA,et al. Activation and overexpression of centrosome kinase BTAK/Aurora-A in human ovarian cancer. Clin Cancer Res 2003;9:1420–6.

10. Jeng YM, Peng SY, Lin CY, HsuHC. Overexpression and amplificationof Aurora-A in hepatocellular carcinoma. Clin Cancer Res 2004;10:2065–71.

11. Sakakura C, Hagiwara A, Yasuoka R, Fujita Y, Nakanishi M,Masuda K,et al. Tumour-amplified kinaseBTAK is amplified and overexpressed ingastric cancers with possible involvement in aneuploid formation. Br JCancer 2001;84:824–31.

12. Goepfert TM, Adigun YE, Zhong L, Gay J, Medina D, Brinkley WR.Centrosome amplification and overexpression of aurora A are earlyevents in rat mammary carcinogenesis. Cancer Res 2002;62:4115–22.

13. Glover DM, Leibowitz MH, McLean DA, Parry H. Mutations in auroraprevent centrosome separation leading to the formation of monopolarspindles. Cell 1995;81:95–105.

Dobson et al.





14. Tanner MM, Grenman S, Koul A, Johannsson O, Meltzer P, Pejovic T,et al. Frequent amplification of chromosomal region 20q12-q13 inovarian cancer. Clin Cancer Res 2000;6:1833–9.

15. Tanner MM, Tirkkonen M, Kallioniemi A, Collins C, Stokke T, Karhu R,et al. Increased copy number at 20q13 in breast cancer: defining thecritical region and exclusion of candidate genes. Cancer Res1994;54:4257–60.

16. Tanner MM, Tirkkonen M, Kallioniemi A, Holli K, Collins C, Kowbel D,et al. Amplification of chromosomal region 20q13 in invasive breastcancer: prognostic implications. Clin Cancer Res 1995;1:1455–61.

17. Lai CH, Tseng JT, LeeYC, ChenYJ, Lee JC, Lin BW, et al. Translationalup-regulation of Aurora-A in EGFR-overexpressed cancer. J Cell MolMed 2010;14:1520–31.

18. Palmiter RD. Regulation of protein synthesis in chick oviduct. I.Independent regulation of ovalbumin, conalbumin, ovomucoid, andlysozyme induction. J Biol Chem 1972;247:6450–61.

19. Kozak M. An analysis of 50-noncoding sequences from 699 vertebratemessenger RNAs. Nucleic Acids Res 1987;15:8125–48.

20. GingrasAC,RaughtB,SonenbergN. eIF4 initiation factors: effectors ofmRNA recruitment to ribosomes and regulators of translation. AnnuRev Biochem 1999;68:913–63.

21. Merrick WC. Cap-dependent and cap-independent translation ineukaryotic systems. Gene 2004;332:1–11.

22. Jackson RJ, Hunt SL, Reynolds JE, Kaminski A. Cap-dependent andcap-independent translation: operational distinctions andmechanisticinterpretations. Curr Top Microbiol Immunol 1995;203:1–29.

23. Jackson RJ, Kaminski A. Internal initiation of translation in eukaryotes:the picornavirus paradigm and beyond. RNA 1995;1:985–1000.

24. Stein I, Itin A, Einat P, Skaliter R, Grossman Z, Keshet E. Translation ofvascular endothelial growth factor mRNA by internal ribosome entry:implications for translation under hypoxia. Mol Cell Biol 1998;18:3112–9.

25. Holcik M, Lefebvre C, Yeh C, Chow T, Korneluk RG. A new internal-ribosome-entry-site motif potentiates XIAP-mediated cytoprotection.Nat Cell Biol 1999;1:190–2.

26. Pyronnet S, Pradayrol L, Sonenberg N. A cell cycle-dependent internalribosome entry site. Mol Cell 2000;5:607–16.

27. Stoneley M, Willis AE. Cellular internal ribosome entry segments:structures, trans-acting factors and regulation of gene expression.Oncogene 2004;23:3200–7.

28. Cornelis S, Bruynooghe Y, DeneckerG, VanHuffel S, Tinton S, BeyaertR. Identification and characterization of a novel cell cycle-regulatedinternal ribosome entry site. Mol Cell 2000;5:597–605.

29. StoneleyM, Paulin FE, LeQuesne JP, Chappell SA,Willis AE. C-Myc 50

untranslated region contains an internal ribosome entry segment.Oncogene 1998;16:423–8.

30. Creancier L, Morello D, Mercier P, Prats AC. Fibroblast growth factor 2internal ribosome entry site (IRES) activity ex vivo and in transgenicmice reveals a stringent tissue-specific regulation. J Cell Biol2000;150:275–81.

31. Martineau Y, Le Bec C, Monbrun L, Allo V, Chiu IM, Danos O, et al.Internal ribosome entry site structural motifs conserved among mam-malian fibroblast growth factor 1 alternatively splicedmRNAs.Mol CellBiol 2004;24:7622–35.

32. van der Velden AW, Thomas AA. The role of the 50 untranslated regionof an mRNA in translation regulation during development. Int J Bio-chem Cell Biol 1999;31:87–106.

33. Dobson T, Minic A, Nielsen K, Amiott E, Krushel L. Internal initiation oftranslation of the TrkBmRNA ismediated bymultiple regionswithin the50 leader. Nucleic Acids Res 2005;33:2929–41.

34. Holcik M, Korneluk RG. Functional characterization of the X-linkedinhibitor of apoptosis (XIAP) internal ribosome entry site element: roleof La autoantigen in XIAP translation. Mol Cell Biol 2000;20:4648–57.

35. Mitchell SA, SpriggsKA,ColdwellMJ, JacksonRJ,Willis AE. TheApaf-1 internal ribosome entry segment attains the correct structural con-formation for function via interactions with PTB and unr. Mol Cell2003;11:757–71.

36. Kollareddy M, Dzubak P, Zheleva D, Hajduch M. Aurora kinases:structure, functions and their association with cancer. Biomed PapMed Fac Univ Palacky Olomouc Czech Repub 2008;152:27–33.

37. WanXB, LongZJ, YanM, Xu J, Xia LP, Liu L, et al. Inhibition of Aurora-Asuppresses epithelial–mesenchymal transition and invasion by down-regulating MAPK in nasopharyngeal carcinoma cells. Carcinogenesis2008;29:1930–7.

38. Schedin PJ, Eckel-Mahan KL,McDaniel SM, Prescott JD, Brodsky KS,Tentler JJ, Gutierrez-Hartmann A. ESX induces transformation andfunctional epithelial to mesenchymal transition in MCF-12A mammaryepithelial cells. Oncogene 2004;23:1766–79.

39. Ford HL. Homeobox genes: A link between development, cell cycle,and cancer? Cell Biol Int 1998;22:397–400.

40. SukhanovS, Higashi Y, Shai SY, ItabeH,OnoK, Parthasarathy S, et al.Novel effect of oxidized low-density lipoprotein: cellular ATP depletionvia downregulation of glyceraldehyde-3-phosphate dehydrogenase.Circ Res 2006;99:191–200

41. Thomas JD, Johannes GJ. Identification of mRNAs that continue toassociate with polysomes during hypoxia. RNA 2007;13:1116–31.

42. Kozak M. Evaluation of the "scanning model" for initiation of proteinsynthesis in eucaryotes. Cell 1980;22:7–8.

43. LockardRE, LaneC.Requirement for 7-methylguanosine in translationof globin mRNA in vivo. Nucleic Acids Res 1978;5:3237–47.

44. De Benedetti A, Joshi-Barve S, Rinker-Schaeffer C, Rhoads RE.Expression of antisense RNA against initiation factor eIF-4E mRNAin HeLa cells results in lengthened cell division times, diminishedtranslation rates, and reduced levels of both eIF-4E and the p220component of eIF-4F. Mol Cell Biol 1991;11:5435–45.

45. Duncan R, Milburn SC, Hershey JW. Regulated phosphorylation andlow abundance of HeLa cell initiation factor eIF-4F suggest a role intranslational control. Heat shock effects on eIF-4F. J Biol Chem1987;262:380–8.

46. Hiremath LS, Webb NR, Rhoads RE. Immunological detection of themessenger RNA cap-binding protein. J Biol Chem 1985;260:7843–9.

47. Avdulov S, Li S, Michalek V, Burrichter D, Peterson M, Perlman DM,et al. Activation of translation complexeIF4F is essential for thegenesisand maintenance of the malignant phenotype in human mammaryepithelial cells. Cancer Cell 2004;5:553–63.

48. De Benedetti A, Graff JR. eIF-4E expression and its role in malignan-cies and metastases. Oncogene 2004;23:3189–99.

49. Lazaris-Karatzas A, SmithMR, Frederickson RM, JaramilloML, Liu YL,Kung HF, et al. Ras mediates translation initiation factor 4E-inducedmalignant transformation. Genes Dev 1992;6:1631–42.

50. Ramirez-Valle F, Braunstein S, Zavadil J, Formenti SC, Schneider RJ.eIF4GI links nutrient sensing by mTOR to cell proliferation and inhibi-tion of autophagy. J Cell Biol 2008;181:293–307.

51. Haghighat A, Sonenberg N. eIF4G dramatically enhances the bindingof eIF4E to the mRNA 50-cap structure. J Biol Chem 1997;272:21677–80.

52. Dobson T, Kube E, Timmerman S, Krushel LA. Identifying intrinsic andextrinsic determinants that regulate internal initiation of translationmediated by the FMR1 50 leader. BMC Mol Biol 2008;9:89.

53. Gingras AC, Gygi SP, Raught B, Polakiewicz RD, Abraham RT, Hoek-straMF, et al. Regulation of 4E-BP1 phosphorylation: a novel two-stepmechanism. Genes Dev 1999;13:1422–37.

54. Jan E, Sarnow P. Factorless ribosome assembly on the internal ribo-some entry site of cricket paralysis virus. J Mol Biol 2002;324:889–902.

55. Jang SK, Davies MV, Kaufman RJ, Wimmer E. Initiation of proteinsynthesis by internal entry of ribosomes into the 50 nontranslatedregion of encephalomyocarditis virus RNA in vivo. J Virol 1989;63:1651–60.

56. Van Eden ME, Byrd MP, Sherrill KW, Lloyd RE. Demonstrating internalribosome entry sites in eukaryotic mRNAs using stringent RNA testprocedures. RNA 2004;10:720–30.

57. Thompson SR. So you want to know if your message has an IRES?Wiley Interdiscip Rev: RNA 2012;3:697–705.

58. Lewis SM, Holcik M. For IRES trans-acting factors, it is all aboutlocation. Oncogene 2008;27:1033–5.

59. Tinton SA, Schepens B, Bruynooghe Y, Beyaert R, Cornelis S. Reg-ulation of the cell-cycle–dependent internal ribosome entry site of thePITSLRE protein kinase: roles of Unr (upstream of N-ras) protein andphosphorylated translation initiation factor eIF-2alpha. BiochemJ 2005;385:155–63.






60. Bergamini G, Preiss T, Hentze MW. Picornavirus IRESes and the poly(A) tail jointly promote cap-independent translation in a mammaliancell-free system. RNA 2000;6:1781–90.

61. Dehlin E, Wormington M, Korner CG, Wahle E. Cap-dependent dead-enylation of mRNA. EMBO J 2000;19:1079–86.

62. ChiangPW,Carpenter LE,HagermanPJ. The 50-untranslated region ofthe FMR1 message facilitates translation by internal ribosome entry.J Biol Chem 2001;276:37916–21.

63. Band V, editor. Tumor progression in breast cancer. Totowa, NJ:Humana Press; 1991.

64. Band V, Sager R. Distinctive traits of normal and tumor-derived humanmammary epithelial cells expressed in a medium that supports long-term growth of both cell types. Proc Natl Acad Sci U S A 1989;86:1249–53.

65. Band V, Zajchowski D, Swisshelm K, Trask D, Kulesa V, CohenC, et al.Tumor progression in four mammary epithelial cell lines derived fromthe same patient. Cancer Res 1990;50:7351–7.

66. Hata T, Furukawa T, Sunamura M, Egawa S, Motoi F, Ohmura N, et al.RNA interference targeting aurora kinase a suppresses tumor growthand enhances the taxane chemosensitivity in human pancreatic can-cer cells. Cancer Res 2005;65:2899–905.

67. Neben K, Korshunov A, Benner A, Wrobel G, Hahn M, Kokocinski F,et al. Microarray-based screening for molecular markers in medullo-blastoma revealed STK15 as independent predictor for survival. Can-cer Res 2004;64:3103–11.

68. Sen S, Zhou H, White RA. A putative serine/threonine kinase encodinggene BTAK on chromosome 20q13 is amplified and overexpressed inhuman breast cancer cell lines. Oncogene 1997;14:2195–200.

69. Tseng YS, Lee JC, Huang CY, Liu HS. Aurora-A overexpressionenhances cell-aggregation of Ha-ras transformants through theMEK/ERK signaling pathway. BMC Cancer 2009;9:435.

70. Kozak M. A second look at cellular mRNA sequences said tofunction as internal ribosome entry sites. Nucleic Acids Res2005;33:6593–602.

71. HolcikM. Targeting translation for treatment of cancer—anovel role forIRES? Curr Cancer Drug Targets 2004;4:299–311.

72. Coldwell MJ, Mitchell SA, Stoneley M, MacFarlane M, Willis AE.Initiation of Apaf-1 translation by internal ribosome entry. Oncogene2000;19:899–905.

73. Sehgal A,Briggs J,Rinehart-KimJ,BassoJ,BosTJ. Thechicken c-Jun50 untranslated region directs translation by internal initiation. Onco-gene 2000;19:2836–45.

74. Sherrill KW, Byrd MP, Van Eden ME, Lloyd RE. BCL-2 translation ismediated via internal ribosome entry during cell stress. J Biol Chem2004;279:29066–74.

75. Yang DQ, Halaby MJ, Zhang Y. The identification of an internalribosomal entry site in the 50-untranslated region of p53 mRNA pro-vides a novel mechanism for the regulation of its translation followingDNA damage. Oncogene 2006;25:4613–9.

76. Bellodi C, Kopmar N, Ruggero D. Deregulation of oncogene-inducedsenescence and p53 translational control in X-linked dyskeratosiscongenita. EMBO J 2010;29:1865–76.

77. Montanaro L, Calienni M, Bertoni S, Rocchi L, Sansone P, Storci G,et al. Novel dyskerin-mediatedmechanism of p53 inactivation throughdefective mRNA translation. Cancer Res 2010;70:4767–77.

78. Yoon A, Peng G, Brandenburger Y, Zollo O, Xu W, Rego E, et al.Impaired control of IRES-mediated translation in X-linked dyskeratosiscongenita. Science 2006;312:902–6.

79. Braunstein S, Karpisheva K, Pola C, Goldberg J, Hochman T, Yee H,et al. A hypoxia-controlled cap-dependent to cap-independent trans-lation switch in breast cancer. Mol Cell 2007;28:501–12.

80. Filbin ME, Kieft JS. Toward a structural understanding of IRES RNAfunction. Curr Opin Struct Biol 2009;19:267–76.

81. Belsham GJ, Sonenberg N. Picornavirus RNA translation: roles forcellular proteins. Trends Microbiol 2000;8:330–5.

82. PilipenkoEV,BlinovVM,RomanovaLI, SinyakovAN,MaslovaSV,AgolVI. Conserved structural domains in the 50-untranslated region ofpicornaviral genomes: an analysis of the segment controlling transla-tion and neurovirulence. Virology 1989;168:201–9.

83. Vagner S, Galy B, Pyronnet S. Irresistible IRES. Attracting the trans-lation machinery to internal ribosome entry sites. EMBO Rep 2001;2:893–8.

84. Beaudoin ME, Poirel VJ, Krushel LA. Regulating amyloid precursorprotein synthesis through an internal ribosomal entry site. NucleicAcids Res 2008;36:6835–47.

85. Chappell SA, Edelman GM, Mauro VP. A 9-nt segment of a cellularmRNA can function as an internal ribosome entry site (IRES) and whenpresent in linked multiple copies greatly enhances IRES activity. ProcNatl Acad Sci U S A 2000;97:1536–41.

86. Jang GM, Leong LE, Hoang LT, Wang PH, Gutman GA, Semler BL.Structurally distinct elements mediate internal ribosome entry withinthe 50-noncoding region of a voltage-gated potassium channelmRNA.J Biol Chem 2004;279:47419–30.

87. Ali N, Siddiqui A. The La antigen binds 50 noncoding region of thehepatitis C virus RNA in the context of the initiator AUG codon andstimulates internal ribosome entry site-mediated translation. Proc NatlAcad Sci U S A 1997;94:2249–54.

88. Borovjagin A, Pestova T, Shatsky I. Pyrimidine tract binding proteinstrongly stimulates in vitroencephalomyocarditis virusRNA translationat the level of preinitiation complex formation. FEBS Lett 1994;351:299–302.

89. Pfingsten JS, Costantino DA, Kieft JS. Structural basis for ribosomerecruitment and manipulation by a viral IRES RNA. Science 2006;314:1450–4.

90. HellenCU,SarnowP. Internal ribosomeentry sites in eukaryoticmRNAmolecules. Genes Dev 2001;15:1593–612.

91. Gu L, ZhuN, Zhang H, Durden DL, Feng Y, ZhouM. Regulation of XIAPtranslation and induction by MDM2 following irradiation. Cancer Cell2009;15:363–75.

92. Halaby MJ, Hibma JC, He J, Yang DQ. ATM protein kinase med-iates full activation of Akt and regulates glucose transporter 4translocation by insulin in muscle cells. Cell Signal 2008;20:1555–63.

93. Ohno S, Shibayama M, Sato M, Tokunaga A, Yoshida N. Polypyrimi-dine tract-bindingprotein regulates the cell cycle through IRES-depen-dent translation of CDK11(p58) in mouse embryonic stem cells. CellCycle 2011;10:3706–13.

94. Lee EC, Frolov A, Li R, Ayala G, Greenberg NM. Targeting Aurorakinases for the treatment of prostate cancer. Cancer Res 2006;66:4996–5002.

95. Kitzen JJ, de Jonge MJ, Verweij J. Aurora kinase inhibitors. Crit RevOncol Hematol 2010;73:99–110.

96. Mountzios G, Terpos E, Dimopoulos MA. Aurora kinases as targets forcancer therapy. Cancer Treat Rev 2008;34:175–82.

97. Dutertre S, Descamps S, Prigent C. On the role of aurora-A in cen-trosome function. Oncogene 2002;21:6175–83.

Dobson et al.





2013;11:887-900. Published OnlineFirst May 9, 2013.Mol Cancer Res Tara Dobson, Juan Chen and Les A. Krushel Overexpression of Oncogenic Aurora A KinaseDysregulating IRES-Dependent Translation Contributes to

Updated version

10.1158/1541-7786.MCR-12-0707doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2013/05/09/1541-7786.MCR-12-0707.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/11/8/887.full#ref-list-1

This article cites 96 articles, 43 of which you can access for free at:

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

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mcr.aacrjournals.org/content/11/8/887To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-12-0707

http://mcr.aacrjournals.org/content/suppl/2013/05/09/1541-7786.MCR-12-0707.DC1

http://mcr.aacrjournals.org/content/11/8/887.full#ref-list-1

http://mcr.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mcr.aacrjournals.org/content/11/8/887


Dysregulating IRES-Dependent Translation Contributes...

Documents

Transcript of Dysregulating IRES-Dependent Translation Contributes...