Cancer Cells Exploit eIF4E2-Directed Synthesis of Hypoxia ... · Microenvironment and Immunology Cancer Cells Exploit eIF4E2-Directed Synthesis of Hypoxia Response Proteins to Drive

Microenvironment and Immunology

Cancer Cells Exploit eIF4E2-Directed Synthesis of HypoxiaResponse Proteins to Drive Tumor Progression

James Uniacke, J. Kishan Perera, Gabriel Lachance, Camille B. Francisco, and Stephen Lee

AbstractHuman tumors display considerable diversity in their genetic makeup but share common physiologic

attributes such as a hypoxic microenvironment that contribute to the malignant phenotype. Hypoxic cellsswitch from eukaryotic initiation factor 4E (eIF4E) to eIF4E2 cap-dependent translation to synthesize a portion oftheir proteins. Here, we show that genetically distinct human cancer cells exploit eIF4E2-directed proteinsynthesis to form cellular masses larger than approximately 0.15 mm, the diffusion limit of oxygen. Cancer cellsdepleted of eIF4E2 are indistinguishable from control cells under normoxic conditions, but are unable to surviveand proliferate in low oxygen conditions. Activation of eIF4E2-directed translation is essential for cancer cells toform a hypoxic tumor core in in vitro spheroids and to form detectable tumors in in vivo xenograft assays. Incontrast, the eIF4E-directed protein synthesis pathway alone cannot sustain cellular adaptation to hypoxia invitro or confer tumorigenic potential in xenograft assays. These data demonstrate that the phenotypic expressionof the cancer genome requires translation by the eIF4E2-directed hypoxic protein synthesis machinery. CancerRes; 74(5); 1379–89. �2014 AACR.

IntroductionA salient feature of expanding tumors is their ability to

outgrow their vasculatures, causing regions in which cells areexposed to low oxygen tension (hypoxia; refs. 1, 2). As oxygendiffuses through approximately 10 layers of cells (3), themajority of cancer cells that populate a tumor are thought tobe exposed to hypoxic conditions. Tumor cells must, therefore,adapt to transient fluctuations in the supply of oxygen andnutrients (4). One of the primary cellular responses to lowoxygen availability is the activation of the hypoxia-induciblefactor (HIF) transcription program (5–8). HIFs are degraded inthe presence of oxygen but stabilized by hypoxia to activate anarray of genes involved in cellular adaptation to low oxygenavailability, including VEGF and glucose transporter 1 (9).Outside of embryogenesis, these extreme oxygen-depletedconditions are rarely experienced by the cells of the bodyexcept during ischemic diseases such as cancer. It is believedthat activation of theHIF transcription programplays a centralrole in the ability of cancer cells to thrive in the hypoxic core oftumors.

Once synthesized, mRNAs must undergo translation toproduce their corresponding proteins. Typically, protein syn-thesis is initiated by the binding of the eukaryotic initiationfactor 4F to the 50 cap found on the majority of mRNAs (10).eIF4F is composed of the cap-binding protein eIF4E, the RNAhelicase eIF4A, and the scaffold protein eIF4G (11). Theactivity of this complex is primarily regulated by the mamma-lian target of rapamycin complex 1 (mTORC1; refs. 12, 13).During periods of active translation, mTORC1 phosphorylatesmembers of the 4E-binding proteins (4E-BP) to prevent theirability to interactwith, and inhibit, eIF4E. Incontrast, inhibitionofmTORC1 results in the accumulationof hypophosphorylated4E-BPs that are capable of assembling with eIF4E and preventcap-dependent translation (14–16). Hypoxia is a potent inhib-itor of mTORC1 activity causing 4E-BP–mediated inhibition ofeIF4E-dependent translation and a decrease in global transla-tion rates (17, 18). Themost commonly accepted interpretationof the hypoxic inhibition of eIF4E-mediated translation is aprocess whereby cells attenuate protein synthesis to preserveenergy, whereas anaerobic metabolism is the sole provider ofATP. Indeed, there is a hypoxia-controlled switch in breastcancer from global eIF4E-mediated cap-dependent translationto selective cap-independent translation (19) that enables cellsto synthesize at least a fraction of their proteins (20). Althoughcancer cells may reduce their energy consumption by decreas-ing the rate of global translation, they are still fully capable ofengaging in intense protein synthesis during the adaptation tohypoxia (20–22). This raises a fundamental question in ourunderstanding of tumor biology about howmalignant cells cantranslate their genetic makeup into a hypoxic cancer cellphenotype during periods of oxygen scarcity.

We recently demonstrated that various normal cells andcancer cells use an alternative translation initiation scheme to

Authors' Affiliation: Department of Cellular and Molecular Medicine,Faculty of Medicine, University of Ottawa, Ottawa, Ontario, Canada

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

J. Uniacke and J.K. Perera contributed equally to this work.

Corresponding Author: Stephen Lee, Department of Cellular and Molec-ular Medicine, Faculty of Medicine, University of Ottawa, 451 Smyth Road,Ottawa, Ontario, K1H 8M5, Canada. Phone: 613-562-5800, ext. 8385; Fax:613-562-5434; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-13-2278

�2014 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 1379

on January 16, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst January 9, 2014; DOI: 10.1158/0008-5472.CAN-13-2278

http://cancerres.aacrjournals.org/

synthesize a portion of their proteins during hypoxia (21). Thissystem relies on the participation of several components. First,the assembly of the oxygen-regulated HIF-2a and the RNA-binding protein RBM4 at 30 untranslated region (UTR) RNAhypoxia response elements (rHRE) found in hundreds ofmRNAs that code for proteins with roles in cancer hallmarkssuch as survival, proliferation, invasion, angiogenesis, and theevasion of apoptosis. Second, the recruitment of the cap-binding protein eIF4E2, a homolog of eIF4E, to the HIF-2a/RBM4 complex followed by the capture of the 50cap of rHRE-containing mRNAs. eIF4E2 is an inhibitor of translation duringDrosophila development (23) and in normoxic human-derivedcells (24). Consistent with these observations, silencing ofeIF4E2 has little detectable effect on the global translationcapacity of cells maintained in normal oxygen tension (21). Incontrast, eIF4E2-depleted cells have a reduced capacity tosynthesize proteins under hypoxic conditions. This is likelybecause 4E-BPs have a stronger affinity for eIF4E than eIF4E2,enabling rHRE-containing mRNAs to evade mTORC1-mediat-ed arrest of eIF4E-dependent translation and contribute sig-nificantly to hypoxic protein synthesis (21, 25, 26). Therefore,we hypothesized that tumors require eIF4E2-mediated proteinsynthesis to sustain hypoxic regions and, consequently, grow tosignificant sizes.

In this report, we show that in vitro and in vivo cancer cellsexploit eIF4E2-directed translation to express their tumori-genic potential. Depletion of eIF4E2 protein prevents geneti-cally diverse cancer cells from forming tumors and slows thegrowth of already-established tumors as they are unable toadapt to the hypoxic microenvironment. In addition, theeIF4E-mediated protein synthesis machinery alone is insuffi-cient to confer adaptation to low oxygen availability. Thesedata provide evidence that cancer cells exploit the hypoxictranslation initiation machinery to express the hypoxic cancercell phenotype.

Materials and MethodsCell culture and cell lines

U87MG glioblastoma, 786-O renal cell carcinoma, HCT116colorectal carcinoma, and renal proximal tubular epithelialcells were used within 6 months of being obtained from theAmerican Type Culture Collection and maintained as sug-gested. These were characterized by short tandem repeat, Y-chromosome paint and Q-band assays. Cells were incubated at37�C in ambient O2 levels and a 5% CO2 environment. Hypoxiawas induced by incubating at 1% O2, for 24 hours unlessotherwise indicated. Cell number assays were performed byplating 105 cells in 6-cm plates and measuring the cell numberafter 48 hours with a hemocytometer.

Polysomal analysisPolysome analysis was performed as previously described

(27).

Protein synthesis by 35S-Met incorporationDe novo protein synthesis was measured as previously

described (21).

In vitro spheroidsPerformed as previously described (27).

Constructs and short hairpin RNAsLuciferase constructs were generated and used as previ-

ously described (21). GIPZ Lentiviral Human EIF4E2 shRNA-mir (ThermoScientific) was used to target the eIF4E2CDS [V2LHS_68041 short hairpin RNA (shRNA)-1 sequenceTGAACAGAATATCAAA] or the 30UTR (V3LHS_405000shRNA-2 sequence CAGCTGAGATCACTTAATAA). A non-targeting shRNA in a pGIPZ vector was used as a control.Clones are identified by their shRNA sequence and clonenumber (i.e., 1.1 represents shRNA sequence 1 clone 1).Rescue clones were generated by transfecting U87MGeIF4E2 knockdown cells (clones 2.1 and 2.3) with an eIF4E2ORF cDNA construct (GeneCopoeia) or a vehicle controlcontaining a neomycin resistance gene.

Western blot analysis and antibodiesMembranes were incubated with primary antibodies anti-

EGFR (Ab-12; LabVision), anti-GAPDH (glyceraldehyde-3-phosphate dehydrogenase; Genetex), anti-HIF-2a (Novus),anti-actin (Sigma), anti-L5 (Abcam), anti-S13 (Abcam), anti-eIF4E (Genetex), and anti-eIF4E2 (Genetex). Secondary anti-bodies were horseradish peroxidase–conjugated anti-mouse(Amersham Biosciences) or anti-rabbit (Jackson ImmunoRe-search Inc.).

ImmunohistochemistryFour-day-old spheroids or biopsied xenografts were frozen

in optimum cutting temperature compound. Of note, 10-mmsections were fixed in cold acetone for 10minutes, and blockedwith 10% goat serum in PBS for 1 hour at room temperaturefollowed by incubation overnight at 4�Cwith primary antibody(1:300) in 5% goat serum. Incubation with secondary antibody(1:200) diluted in 5% goat serum for 1 to 3 hours at roomtemperature followed. After three washes in PBS, slides weremounted onto coverslips. Antibodies: Cell Signaling Technol-ogy (P4E-BP, 4E-BP, and S6-P) and Novus (CAIX). Spheroidhypoxia determined by labeling with 100 mmol/L hypoxyprobefor 2 hours according to the manufacturer's protocol. TUNEL(terminal deoxynucleotidyl transferase–mediated dUTP nickend labeling) was performed using the In Situ Cell DeathDetection Kit (Roche) as per the manufacturer's protocol. Allspecimens counterstained with Hoechst 33242 reagent (Sig-ma). Blood vessels visualized with hematoxylin and eosin(H&E) as previously described (28).

Cell proliferation and activated programmed cell deathassays

For measuring proliferation, cells were incubated for 24hours in normoxia or hypoxia, then incubated in 10 mmol/LBrdUrd (bromodeoxyuridine; Jackson ImmunoResearch Inc.)for 3 hours, and fixed in 70% ethanol in 50 mmol/L glycine (pH2.0) for 30 minutes at room temperature. An anti-BrdUrdantibody (1:40) and anti-mouse Alexa594–conjugated second-ary antibody (1:200; Invitrogen) were used for detection. Formeasuring the activation of apoptosis, cells were incubated for

Uniacke et al.

Cancer Res; 74(5) March 1, 2014 Cancer Research1380




48 hours in normoxia or hypoxia, andfixed in coldmethanol for10 minutes followed by cold acetone for 1 minute. Cells wereblocked overnight in 10% FBS and incubated with antiactivecaspase-3 primary antibody (Abcam; 1:300) followed by incu-bation with an anti-rabbit Alexa594–conjugated secondaryantibody (1:200). For measuring proliferation in a spheroidsection, acetone-fixed slides were blocked with 5% goat serumfor 1 hour and washed for 5 minutes in PBS. Slides wereincubated with anti-mouse Ki-67 antibody (Dako; 1:150) for1 hour followed by three washes of PBS. Slides were incubatedwith anti-mouse Alexa594–conjugated secondary antibody(1:200) for 1 hour at room temperature.

Clonogenic and sub-G1 analysisFor clonogenic analysis, 200 cells were plated and incubated

in normoxia or hypoxia for 48 hours. Plates were then incu-bated in normoxia for 2 weeks to allow colony formation.Colonies were visualized by fixing in 6% glutaraldhehyde andstaining with 0.5% crystal violet. For sub-G1 analysis, 10

7 cellswere fixed in 70% ethanol for 30 minutes on ice. Cells werepelleted and washed in phosphate–citrate buffer. Cells weretreatedwith 50mLRNAse A (100mg/mL) and 450mLpropidiumiodide (50 mg/mL) for 15 minutes and analyzed by flowcytometry (Beckman Coulter Cyan ADP 9 Analyzer).

Nude mice xenograft studiesFemale CD-1 nude mice (Charles River Laboratories) were

injected in theirflankswith 106 or 107 cells in 200mL sterile PBS.

Mice were euthanized 6 to 8 weeks after injection according tofacility protocols (University of Ottawa, Ottawa, Canada).Lentiviral infection efficiency was monitored by GFP expres-sion from the pGIPZ vector. Titers yielded 4� 108 transductionunits per sample. Lentiviral injections for xenografts weredesigned for a multiplicity of infection of at least four per cellinjected. For a xenograft injection of 106 cells, lentiviral injec-tions contained 4� 106 transduction unit diluted in 50 mL PBS.When tumors reached a volume of 50 mm3, injections wereperformed every 48 hours for 8 days with lentivirus containingeither control shRNAor shRNA-targeting eIF4E2. Todeterminetumor volume by external caliper, the greatest longitudinaldiameter (length) and the greatest transverse diameter (width)

were determined: 1=2 ðlength�width2Þ. All experimentswere performed double-blinded.

Statistical analysisP values associated with all comparisons were based on two-

tailed Student t tests. Results are mean (n � 3) � SEM.

ResultsTumor cells activate the eIF4E2 translation machinery

Normal cells and cancer cells incubated in the absence ofoxygen switch from eIF4E to eIF4E2-mediated cap-dependenttranslation (Fig. 1A and B) in a HIF-2a–dependent manner(21). It is known that a considerable fraction of cells that form atumor are exposed to a hypoxic microenvironment (3, 29). Wethus hypothesized that cancer cells in the hypoxic core of

1 2 3 4 5 6 7 8 9

Polysomes

eIF4E2

eIF4E

L5

Monolayer 21% O2

1 2 3 4 5 6 7 8 9

Polysomes

Monolayer 1% O2A

Abso

rbance

254 n

m

eIF4E2

eIF4E

S13

L5

1 2 3 4 5 6 7 8 9

Polysomes

Spheroid

1 2 3 4 5 6 7 8 9

Polysomes

XenograftB

Abso

rbance

254 n

m

Merge4E-BPHoechst

F

MergeP4E-BPHoechst

MergeHypoxyprobeTM

Hoechst

C

D

MergeS6-PHoechst

ES13

Figure 1. Tumor cells activate the eIF4E2-directed translation machinery during mTORC1 inhibition. A and B, polysomal distribution of eIF4E and eIF4E2 incultured monolayers of cells exposed to 21% O2 (normoxia) and 1% O2 (hypoxia; A) or an in vitro tumor model (spheroid) and xenograft of U87MGglioblastoma cells (B). Protein integrity verified by blotting for ribosomal proteins L5 and S13. C–F, immunolabeling of 4-day-old spheroid sections of HCT116colorectal cancer cells incubated with hypoxyprobe (C), antiphosphorylated 4E-BP (P4E-BP; D), antiphosphorylated ribosomal protein S6 (S6-P; E), or anti-4E-BP. F, Hoechst was used as a DNA counterstain. Solid line, section border; dotted line, border between edge and core. Scale bar, 100 mm.

Tumorigenesis Requires eIF4E2-Directed Translation

www.aacrjournals.org Cancer Res; 74(5) March 1, 2014 1381




tumors require eIF4E2-directed protein synthesis for tumor-igenesis. We began our investigation by examining the phos-phorylation status of mTORC1 targets in tumors. The kinaseactivity of mTORC1 is reliant on normal oxygen tension (30).Tumors are exposed to an oxygen environment that is moreheterogeneous than in monolayers. Therefore, to determinewhether tumors are appropriate models for studying hypoxictranslation, we measured their oxygenation and mTORC1activity. The oxygenation of a spheroid, an avascular in vitrotumor model, was revealed by incubating with hypoxyprobe.This compound is a pimonidazole hydrochloride that formsreductively activated protein adducts at, or below, 10 mmHg(1.3% O2), which can be recognized by specific antibodies (31).Spheroids of HCT116 colorectal carcinoma and U87MG glio-blastoma cells revealed a thin nonhypoxic edge (�100 mm) anda hypoxic core that accounted for 80% to 90% of the area of thesection (Fig. 1C and Supplementary Fig. S1A). ImmunostainingofmTORC1 targets revealed a biphasic staining pattern where-by phosphorylated inactive 4E-BP or ribosomal protein S6weredetected solely at the oxygenated edge (Fig. 1D and E andSupplementary Fig. S1B). In contrast, total 4E-BP levels wereinduced in themajority of cells that composed the hypoxic coreof a spheroid (Fig. 1F and Supplementary Fig. S1C). Consistentwith these observations, eIF4E protein was observed mostly inthe monosome fractions of spheroids, suggesting that it doesnot participate in active protein synthesis (Fig. 1B). Thisdiffered considerably with eIF4E2, which was observed inmonosomes of oxygenatedmonolayers of cells, but abundantlypresent in polysomes of spheroids (Fig. 1B). An enrichmenttoward the eIF4E2-directed translation machinery was alsoapparent in vivo in which there is an added layer of complexitydue to abnormal vascularization creating pockets of oxygen(32, 33). We confirmed that there are indeed hypoxic pocketswith inactivemTORC1 by probing for phosphorylated 4E-BP inxenografts of glioblastoma cells. Sections immunolabeled forphosphorylated 4E-BP revealedmTORC1 inactivity in the core,as well as in a region between the core and the surface(Supplementary Fig. S1D). A similar region was stained withH&E to highlight the presence of blood vessels (SupplementaryFig. S1E). Furthermore, eIF4E2 was observed in polysomefractions of xenografts (Fig. 1B). These data suggest that eIF4E2actively participates in the translationmachinery of themajor-ity of cancer cells that constitute a tumor.

Tumor cells require eIF4E2-directed protein synthesis toproliferate and survive under hypoxia

Activation of eIF4E2-mediated translation occurs in hypoxicbut not normoxic conditions (21). We reasoned that inhibitingeIF4E2 could selectively inhibit the survival of tumor cellsexposed to hypoxia. To address this, we silenced eIF4E2 inthree genetically distinct cancer cell lines: U87MG glioblasto-ma (PTEN-null), 786-O renal cell carcinoma (VHL-null), andHCT116 colorectal carcinoma (KRAS mutation). Partial orcomplete reduction of eIF4E2 protein was achieved using twoindependent shRNAs (Fig. 2A and Supplementary Fig. S2A).The eIF4E2 knockdown cell lines were viable, displayed unal-tered morphology compared with control cells, and prolifer-ated normally under normoxic conditions. In contrast, we

failed to generate stable cell lines with detectable reductionof eIF4E. This was likely due to the arrest of normoxic proteinsynthesis followed by cell death during the selection process.Cell lines stably expressing shRNA-targeting eIF4E2 failed toinduce the expression of the EGF receptor (EGFR) underhypoxic conditions (Fig. 2B and Supplementary Fig. S2B). Thehypoxic induction of EGFR has been previously shown to bedependent on HIF-2a (27, 28) and eIF4E2 (21). Renal cellcarcinoma cells did not require hypoxia to induce EGFRbecause they constitutively express HIF-2a (SupplementaryFig. S2B). Stably silencing eIF4E2 considerably reduced the rateof global hypoxic translation highlighting its role as an acti-vator of global hypoxic protein synthesis (Fig. 2C). Further-more, eIF4E2 was required to activate the hypoxic expressionof a luciferase reporter containing a 30UTR rHRE (Fig. 2D).Silencing eIF4E2 had no effect on proliferation and numberof glioblastoma, renal, and colorectal cancer cells maintainedin normoxic conditions (Fig. 2E and Supplementary Fig. S2Cand S2D). However, a significant decrease in proliferation andnumber of the same cells relative to control cells was observedin hypoxic conditions (Fig. 2E and Supplementary Fig. S2Cand S2D). These same cells failed to adapt to low oxygentension and subsequently entered apoptosis (SupplementaryFig. S3A–S3C) or failed to form colonies (Supplementary Fig.S3D). Importantly, normal human renal proximal tubularepithelial cells transiently expressing shRNA-targeting eIF4E2displayed decreased survival and proliferation only underhypoxia (Supplementary Fig. S4). These data suggest thateIF4E2 dependence is not a cancer-dependent event, but aphysiologic mechanism that is exploited by cancer cellsattempting to survive low oxygen conditions.

eIF4E2 depletion prevents tumorigenesis of geneticallydiverse human cancers

The ability of cancer cells to adapt to a hypoxic tumormicroenvironment is believed to be a key determinant in theevolution of malignancy. In principle, cancer cells would berequired to engage in eIF4E2-directed hypoxic protein synthe-sis during the initial steps of tumorigenesis. To test this, nudemice were injected subcutaneously in their flanks with cancercells stably expressing either control shRNA or shRNA-target-ing eIF4E2. Four to 5 weeks after injection, glioblastoma cellsstably expressing control shRNA produced large tumors morethan 200 to 300 mm3 in volume relative to cells stably expres-sing shRNA-targeting eIF4E2, which formed initial minimalmasses that were eventually no longer visible (Fig. 3A andSupplementary Fig. S5A). Analogous results using eIF4E2-depleted renal (Fig. 3B) and colorectal carcinomas (Fig. 3C)were obtained. To emphasize eIF4E2 as an attractive antitu-mor target, established tumors (50 mm3) in nude mice wereinjected every 2 days for 8 days with lentiviruses harboringshRNA-targeting eIF4E2 or nontargeting control shRNA. Glio-blastoma and colorectal carcinoma xenografts injected withshRNA-targeting eIF4E2 decreased in size after the secondinjection and had an average final volume that was 5-fold and2-fold smaller relative to the control, respectively (Fig. 3D andE). Immunohistochemistry (IHC) revealed that the hypoxicregions of these xenografts had similar levels of proliferation,

Uniacke et al.





GAPDH

eIF4E2

Colorectal

Control

C1.1 C1.2 2.2 2.32.1

3´UTRshRNA

Colorectal

1

Normoxia Hypoxia

C1.1 2.1 2.2 C1.1 2.1 2.2

***

**

Re

lative

Brd

Uin

co

rpo

ratio

n

Colorectal

1

2

3

4

5

C1.1 2.1 2.2 C1.1 2.1 2.2

*** **Normoxia

Hypoxia

Re

lative

active

ca

sp

ase

-3

Glioblastoma

1

Normoxia Hypoxia

C1.1 1.1 2.1 C1.1 1.1 2.1

*

**

Re

lative

Brd

Uin

co

rpo

ratio

n

Re

lative

active

ca

sp

ase

-3

Glioblastoma

1

2

3

4

5

C1.1 1.1 2.1 C1.1 1.1 2.1

**

**

Normoxia

Hypoxia

21 1 21 1 21 1 21 1

C1.1 2.1 2.2 2.3

Colorectal

GAPDH

HIF-2α

EGF R

eIF4E2 mRNA

shRNA-1

shRNA-2

3´UTR

CDS

A

Renal

1

Normoxia Hypoxia

C1.1 1.1 2.1 C1.1 1.1 2.1

*

**

Re

lative

Brd

Uin

co

rpo

ratio

n

1

Renal

Normoxia

Hypoxia

C1.1 1.1 2.1

2

3

4

5

C1.1 1.1 2.1

**

Re

lative

active

ca

sp

ase

-3

E

F

B

21 1 21 1 21 1 21 1 21 1 21 1

C1.1 1.1 C1.1 2.1

ColorectalGlioblastoma

% O 2

GLUT mRNA

Total RNA

% S inc.35

100 70 103 13 100 69 99 14 100 72 60 8 1

Renal

Normoxia Hypoxia

C1.1

2

3

4

5

1.1R

ela

tive

fo

ld in

cre

ase

Glioblastoma

6

7

C1.1 1.1 C1.1 2.1Colorectal

******

***

Renal

C1.1 1.1

C D

% O2

Figure 2. Cancer cells require eIF4E2-directed protein synthesis to survive and proliferate during hypoxia. A, two shRNAs were used independently.shRNA-1 targeted the coding sequence (CDS) and shRNA-2 the 30UTR of eIF4E2 mRNA. Three clones were selected from three geneticallydiverse cancer cell lines (U87MG glioblastoma, 786-O renal cell carcinoma, and HCT116 colorectal carcinoma) stably expressing shRNA targeting theeIF4E2 CDS (1.1, 1.2, and 1.3) or eIF4E2 30UTR (2.1, 2.2, and 2.3). Clones stably expressing a nontargeting control shRNA (C1.1, C1.2, and C1.3) werealso selected. See also Supplementary Fig. S2A. B, the ability of each clone to induce EGFR protein when exposed to normoxia or hypoxia wasdetermined. GAPDH was used as a loading control. C, global translation rates in normoxia or hypoxia of cells stably expressing shRNA-targetingeIF4E2 (1.1 and 2.1) or control shRNA (C1.1). GLUT mRNA used as a marker for hypoxia. D, luciferase expression was measured in eIF4E2-depletedcells (1.1 and 2.1) relative to control cells (C1.1) expressing a reporter construct containing a 30UTR RNA hypoxia response element. E, the ability of cellsstably expressing control shRNA (C1.1) or shRNA-targeting eIF4E2 (1.1, 2.1 and 2.2) to proliferate was measured following incubation in normoxia orhypoxia for 24 hours. F, active caspase-3–positive cells stably expressing control shRNA (C1.1) or shRNA-targeting eIF4E2 (1.1, 2.1, and 2.2)incubated in normoxia or hypoxia for 48 hours. E and F, data reported relative to cells stably expressing nontargeting shRNA (C1.1). Columns, mean(n ¼ 3); error bars, SEM. Significance measured by the Student t test; �, P < 0.05; ��, P < 0.01; ��, P < 0.001.






but eIF4E2-depleted xenografts had smaller regions of hypoxiathat were adjacent to large areas of cell death (Fig. 3F). Thexenografts were confirmed to be infected with lentiviruses bydetecting the GFP expressed by the pGIPZ backbone (Supple-mentary Fig. S5B and S5C). Although not every cell in the tumorwas infected, a corresponding decrease in eIF4E2 protein levelswas observed (Supplementary Fig. S5B and S5C). We wereunable to excise and study tumors from cells stably expressingshRNA-targeting eIF4E2 because of their absence.

Formation of a hypoxic tumor core necessitates eIF4E2-directed protein synthesis

Regions of hypoxia occur early in tumor formation as oxygendiffuses through approximately 10 cellular layers (3). There-fore, we used 4-day-old spheroids, an in vitro tumor model, to

observe the effects of eIF4E2 depletion at the early stages oftumor formation. Typically, cancer cells are able to producehypoxic spheroids that enlarge over time. However, eIF4E2-depleted spheroids formed fragile, loosely packed aggregates(Fig. 4A and Supplementary Fig. S6A) that failed to increase tosizes larger than 500 mm even after several days in culturerelative to control spheroids (Fig. 4B and Supplementary Fig.S6B). Interestingly, eIF4E2-depleted spheroids had no detect-able expression of the hypoxia markers HIF-1a or HIF-2a,compared with spheroids expressing control shRNA (Fig. 4Cand Supplementary Fig. S6C). Furthermore, staining of spher-oids with hypoxyprobe revealed that they were unable toproduce a hypoxic tumor microenvironment (Fig. 4D), unlikespheroids generated from control cells (Fig. 1C and Supple-mentary Fig. S1A), even though their radii exceeded the oxygen

C1.2 1.2Tum

or

volu

me (

mm

)3

Glioblastoma

1 2 3 4

Weeks after injection

C1.1 X 1.1 * 2.1

C1.2 1.2 2.2

A

400

300

200

100

0 X X X X* * * *

*

Tu

mo

r vo

lum

e (

mm

)3

Colorectal

1 3 4 5


C1.1 * 2.1

C1.2 2.2

C

400

300

200

100

02

*

*

*

* * * * *C1.1 2.2

C1.1 1.1Tum

or

vo

lum

e (

mm

)3

Renal

2 4 5 6


C1.1 X 1.1 * 2.1

C1.2 1.2 2.2

B

200

100

031

X X X X X X* * * * * **

**

*

*

**

*

Re

lative

tu

mo

r vo

lum

e

2 4 6 8

Days after first injection

D

4

3

0

control shRNAeIF4E2 shRNA-1

1 *** **

Glioblastoma

2

2 4 6 8Days after first injection

E

6

0

control shRNAeIF4E2 shRNA-2

Colorectal

5

Re

lative

tu

mo

r vo

lum

e

**

*4

3

2

1

TUNELKi-67CAIX

TUNELKi-67CAIX

F

Co

ntr

ol sh

RN

Ae

IF4

E2

sh

RN

A-1

Figure 3. Silencing of eIF4E2 prevents tumorigenesis in genetically diverse human cancers. A–C,mice were injected with 107 U87MG glioblastoma (A), 786-Orenal cell carcinoma (B), and HCT116 colorectal carcinoma (C) cells stably expressing nontargeting shRNA (C1.1 and C1.2) or shRNA-targeting eIF4E2(1.1, 1.2, 2.1, and 2.2). D and E, established tumors (�50 mm3) of glioblastoma (D) or colorectal carcinoma (E) were injected every 2 days for 8 days withlentiviruses harboring shRNA-targeting eIF4E2 (shRNA-1 or shRNA-2) or control shRNA. Change in tumor volume reported relative to volume beforefirst lentiviral injection. Data, mean � SEM of at least three independent experiments. Significance measured by the Student t test; �, P < 0.05; ��, P < 0.01;��,P<0.001. F, IHCof serial sections fromDandE, staining for hypoxia (CAIX), proliferation (Ki-67, brownnuclei), andcell death (TUNEL). Dotted line, hypoxia.Scale bar, 100 mm.

Uniacke et al.





diffusion limit (3). Consistent with the presence of ampleoxygen in the eIF4E2-depleted in vitro tumors, mTORC1 activ-ity was observed throughout the spheroids. IHC revealedphosphorylated 4E-BP to be evenly distributed throughout thespheroid (Fig. 4E), as opposed to being localized to the outer,more oxygenated edge in the controls (Fig. 1D and Supple-mentary Fig. S1B). Immunostaining for antiactive caspase-3and TUNEL assays revealed less cell death in controls (Fig. 5A)compared with eIF4E2-depleted spheroids (Fig. 5B). Finally,the distribution of proliferative cells differed between controland eIF4E2-depleted spheroid sections. In a control spheroid, a

higher concentration of proliferative cells was observed in theoxygenated edge than in the core (Fig. 5C). This compartmen-talization is consistent with the spheroid oxygen gradientsshownwith hypoxyprobe (Fig. 1B and Supplementary Fig. S1A).Conversely, an eIF4E2-depleted spheroid did not have a dis-tinct border between the edge and core, but a uniform distri-bution of proliferative cells (Fig. 5D). These data demonstratethat eIF4E2-directed translation confers the ability to cancercells to form large tumors with hypoxic cores in vitro. Theseresults also suggest that eIF4E2-depletion leads to more rapidcell death in the cores of early tumors.

2.1

C

HIF-1α

eIF4E2

GAPDH

C1.1 1.1Glioblastoma

C1.1 2.1

Colorectal

HIF-1α

eIF4E2

GAPDH

Days

Rela

tive

sphero

id a

rea

1

GlioblastomaB

1 2 3 4

C1.1 X 1.1

* 2.1

*

**

**

**

***

X

X

*X

X

Days

Rela

tive

sphero

id a

rea

1

Renal

1 2 3 4

C1.1 X 1.1

* 2.1

X

X

XX

** * *

*****

D

HypoxyprobeTM

Hoechst

eIF4E2 shRNA 1.1

eIF4E2 shRNA 1.1E

P4E-BPHoechst

Merge

Merge

Days

Rela

tive

sphero

id a

rea

1 2 3 4

1 XX

X X

* ** *

C1.1 X 2.1

* 2.2**

*

Colorectal

A 1.1C1.1 C1.1Glioblastoma Colorectal

Figure 4. Formation of a hypoxic tumor core necessitates eIF4E2-directed translation. A, light micrographs of U87MG glioblastoma and HCT116colorectal control spheroids (C1.1) compared with eIF4E2-depleted spheroids (1.1 and 2.1). B, spheroid growth was monitored in cells stably expressingcontrol shRNA (C1.1) or shRNA-targeting eIF4E2 (1.1, 2.1, and 2.2). Data, mean � SEM of three independent experiments. Significance measured bythe Student t test; �,P < 0.05; ��,P < 0.01; ��,P < 0.001. C,Western blot analysis of 4-day-old spheroids stably expressing control shRNA or shRNA-targetingeIF4E2.HIF-1awasusedasamarker for hypoxia.GAPDHwasusedasa loading control. D andE, immunolabeling of a 4-day-old glioblastomaspheroid stablyexpressing shRNA-targeting eIF4E2 incubated with hypoxyprobe (D) or anti-P4E-BP (E), revealing active mTORC1 throughout. Hoechst was used as a DNAcounterstain. Solid line, spheroid border. Scale bar, 100 mm.






Reintroduction of exogenous eIF4E2 restores tumor cellcharacteristics

To substantiate that the observed phenotypes were specificto eIF4E2 knockdown, eIF4E2was stably reintroduced into twoglioblastoma cell lines stably expressing shRNA targeting theeIF4E2 30UTR (clones 2.1 and 2.3). Because the exogenouslyexpressed eIF4E2 harbors the vehicle-derived 30UTR, only theendogenous eIF4E2 is targeted by the stably expressed shRNA.Western blot analysis revealed the reappearance of eIF4E2 inknockdown cells stably expressing exogenous eIF4E2 (Y andZ),but not in vehicle control cells (v1 and v2; Fig. 6A). Further-more, these rescued cells regained the ability to induce EGFR inhypoxia, whereas vehicle control cells did not (Fig. 6B andSupplementary Fig. S7A). Knockdown cells stably expressingexogenous eIF4E2 also recovered the ability to divide inhypoxia. The cell number significantly increased for rescuedcells in hypoxia relative to eIF4E2-depleted vehicle control cells(Supplementary Fig. S7B). Conversely, cell numbers did notchange during normoxic exposure regardless of eIF4E2 proteinlevels (Supplementary Fig. S7B). Cellular proliferation was alsorescued by eIF4E2 reintroduction in cells cultured in hypoxia,but there was no difference in normoxia (Fig. 6C and Supple-mentary Fig. S7C). To address whether the reintroduction ofeIF4E2 reduced the initiation of apoptosis under low oxygentension, the percentage of cells with active caspase-3 wascompared between hypoxia and normoxia. Indeed, significant-ly less active caspase-3–positive cells were observed in hypoxicrescued cells relative to vehicle control cells, but there was nochange in normoxic conditions (Fig. 6D and SupplementaryFig. S7D). Importantly, larger (Fig. 6E and Supplementary Fig.

S7E), more hypoxic (Fig. 6F and Supplementary Fig. S7F)spheroidswere observedwhen eIF4E2was stably reintroduced.Finally, the ability of eIF4E2 reintroduction to restore tumorcell growth in an in vivo setting wasmeasured. Nudemice wereinjected with vehicle control eIF4E2 knockdown cells on oneflank and rescues on the other flank. Significantly largertumors were observed as early as 2 weeks after injection ofrescued cells relative to vehicle control cells (Fig. 6G andSupplementary Fig. S7G). These results establish a critical rolefor eIF4E2 and hypoxic protein synthesis in the early stages oftumor formation and highlight the central role of this alter-native translation initiation pathway regardless of the tissue oforigin or the mutational profile of the cancer cells.

DiscussionProkaryotic infections such as pneumonia and tuberculosis

were the leading cause of death before the advent of antibiotics.Many antibiotics are protein synthesis inhibitors that selec-tively inhibit the bacterial machinery, but not that of the host,providing a window for effective treatment. It is generallybelieved that cancer cells exploit the same protein synthesismachinery as normal cells. This has hampered efforts inidentifying compounds that would selectively prevent theprotein synthesis of cancer cells with limited toxicity to normalcells. The data shown here demonstrate that tumor cellsexploit a parallel alternative cap-dependent protein synthesismachinery that functions in hypoxia and relies on eIF4E2. Thisoccurs when cancer cells rapidly encounter hypoxia en route toforming a tumor requiring eIF4E2-directed translation forsurvival. This dependence on eIF4E2 provides a uniquewindow

C1.1

Hoechst

C

Hoechst

eIF4E2 shRNA 1.1D

Merge

MergeKi-67

Ki-67

C1.1

Hoechst

A

Hoechst

eIF4E2 shRNA 1.1B

Merge

Mergeactive caspase-3

active caspase-3

Edge Core

50

% o

f K

i-67/H

oech

st

25

0

Edge Core

50

% o

f K

i-67/H

oech

st

25

0

*

TUNEL

TUNEL

Figure 5. Early spheroids requireeIF4E2-mediated protein synthesisto proliferate and avoid cell death.A and B, immunolabeling withantiactive caspase-3 and TUNELof 4-day-old spheroids fromU87MG glioblastoma cells stablyexpressing nontargeting shRNA (A)or shRNA-targeting eIF4E2 (B).C and D, Ki-67–stained 4-day-oldspheroid from glioblastoma cellsstably expressing nontargetingshRNA (C) or shRNA-targetingeIF4E2 (D). Hoechst was used as aDNA counterstain. Bar graphsrepresent percentage of Ki-67–positive cells at the outer edge andthe core of the spheroid sections.Solid line, section border; dottedline, border between edge andcore. Data, mean � SEM of threeindependent experiments.Significance measured by theStudent t test;�, P < 0.05.

Uniacke et al.





to develop anticancer agents that prevent the protein synthesisof tumor cells without affecting normoxic somatic cells, anal-ogous to the function of antibiotics.Our data demonstrate that as tumors grow, and their

cores become increasingly hypoxic, eIF4E2 depletion resultsin less proliferation (Fig. 2E and Supplementary Fig. S2C)and an increase in cell death (Fig. 2F and Supplementary Fig.S3). Tumors lacking eIF4E2 are too small to notice beneaththe skin of a nude mouse, whereas a parallel control reachesthe ethical endpoint in the same time frame (Fig. 3A–C).Furthermore, in vitro eIF4E2-depleted spheroids could notexceed a diameter of 500 mm and were loose, fragile, and fullyoxygenated (Fig. 4). The data also indicate that severalgenetically diverse cancer cells in a physiologic tumor set-ting do not solely exploit the eIF4E-dependent translationsystem, even though it is sometimes hyperactive or func-tionally deregulated compared with normal cells. Indeed,

eIF4E is not as strongly associated with polysomes as eIF4E2in spheroids and xenografts (Fig. 1B). In addition, eIF4E-mediated translation cannot sustain the formation of hyp-oxic spheroids or detectable tumors in xenograft assays inthe absence of eIF4E2. Thus, cancer cells must activate theeIF4E2-directed hypoxic translation system to adapt tohypoxia and to form tumors independently of the functionalattributes or expression profiles of eIF4E.

The link between eIF4E2 and cancer is not unprecedented.One study identified candidate polymorphisms that influenceoverall survival in patients with advanced non–small cell lungcancer (34). In this genome-wide association study, the stron-gest predictor of shortened overall survival was a single-nucleotide polymorphism (SNP) in eIF4E2. Interestingly, thefrequency of this SNP differed across ethnic backgrounds(refSNP rs1656402). Another study described a molecularsignature in primary solid tumors that is a predictor of

Tu

mor

volu

me (

mm

)3

1 32 4Weeks after injection

G

300

200

100

0

2.1v1 X 2.1Y2.3Y2.3v1 *

X

X

X X

** *

**

*

2.1Y 2.1v1

GAPDH

eIF4E2

v1 v2 ZY v1 v2 ZY

2.1 2.3CloneA

GAPDH

HIF-2α

EGFR

(%)O2 21 1 21 1 21 1 21 1

2.1v1 2.1v2 2.1Y 2.1ZB

Days

E

Re

lative

sp

he

roid

are

a

21 3 4

*2.1v1 X 2.1Y

2.3Y2.3v1

1 X X

X

X

* **

****

***

1

Rela

tive B

rdU

incorp

ora

tion

2.1Y

C

2.1v12.3Y

2.3v12.1Y

2.1v12.3Y

2.3v1

**

**

Normoxia

Hypoxia

D

Rela

tive

active c

aspase-3

Normoxia

Hypoxia

2.1Y2.1v1

2.3Y2.3v1

2.1Y2.1v1

2.3Y2.3v1

1

2

3

4

5

******

F

2.1v1 2.1Y

HIF-1α

eIF4E2

GAPDH

Figure 6. Reintroduction ofexogenous eIF4E2 restores tumorcell characteristics. A, two U87MGglioblastoma clones stablyexpressing shRNA targeting theeIF4E2 30UTR (2.1 and 2.3) wererescued with exogenous eIF4E2.Two rescued clones from eachparent (Y and Z) are shown. eIF4E2-depleted cells stably expressingvehicle-derived nontargetingshRNA were generated as controls(v1 and v2). B, the ability of eachrescued cell line (2.1Y and 2.1Z) toinduce EGFR when exposed tohypoxia (1% O2) was determined.GAPDH was used as a loadingcontrol. C, the ability of controls(2.1v1 and 2.3v1) to proliferate inculture relative to eIF4E2 rescuedcells (2.1Yand2.3Y)wasmeasured.D, active caspase-3–positivecultured control cells (2.1v1 and2.3v1) incubated in normoxia orhypoxia was reported relative toeIF4E2 rescued cells (2.1Y and2.3Y). E, spheroid growth wasmonitored incontrolcells (2.1v1and2.3v1) and eIF4E2 rescued cells(2.1Y and 2.3Y). Data, mean� SEMof three independent experiments.F, Western blot analysis of a 4-day-old spheroid from control cells(2.1v1) and eIF4E2 rescues (2.1Y).HIF-1a used as a marker forhypoxia.G, tumormeasurementsofnude mouse xenografts wereperformed with control cells (2.1v1and2.3v1)andeIF4E2 rescuedcells(2.1Y and 2.3Y). Data, mean� SEMof at least three independentexperiments. C and D, columns,mean (n ¼ 3); error bars, SEM.Significance measured by theStudent t test; �, P < 0.05;��, P < 0.01; ��, P < 0.001.






metastatic potential (35). Eight genes made up the strongestpredictors of metastasis, including eIF4E2 referred to by itsalias eIF4EL3.

eIF4E is overexpressed in a variety of cancers and is a targetof modern therapies (36–41). Furthermore, eIF4E contributesto tumorsphere growth in hypoxic breast cancer cells (42).Indeed, eIF4E remains an attractive target to inhibit the growthof cancer cells. According to our model, the more oxygenatededge of tumors would be more susceptible to eIF4E inhibitorsbecause it has more translation competent eIF4E due to activemTORC1 (Fig. 1C–F and Supplementary Fig. S1A–S1D). Inter-estingly, studies are emerging that directly target eIF4E causingsmaller tumors in mouse models. One study uses antisenseoligonucleotides (ASO) targeting eIF4E and is the first eIF4E-specific therapeutic to advance to clinical trials (36). Theseauthors report similar effects of silencing eIF4E as we do witheIF4E2, such as a reduction in tumor growth in mice, areduction in cellular proliferation and an increase in apoptosis.However, the observed eIF4E silencing is incomplete and thephenotype not rescued by reintroduction of eIF4E. The celllines produced in our study display a strong reduction ofeIF4E2 and, more importantly, the phenotype was rescued bythe expression of exogenous eIF4E2 that evaded shRNA rec-ognition. We attempted to create cancer cell lines stablyexpressing shRNA-targeting eIF4E, but, as silencing eIF4Esignificantly hampers normoxic protein synthesis, clones werenot viable in multiple attempts. The clones that we did obtainretained levels of eIF4E protein andnormoxic protein synthesisthat we found too high and, thus, unsatisfactory to furtherinvestigate in tumor assays. Therefore, it seems that there is alimitedwindow inwhich eIF4E can be reduced enough to affectthe ability of cancer cells to form tumors, but have minimaltoxic effects to the organism.Many questions remain for such a

clinical trial, however, as the toxicity of targeting a ubiqui-tous and important protein synthesis component, and thedosage of ASOs required to achieve the desired effect inhumans. In conclusion, we propose that targeting eIF4E2-directed hypoxic translation would be a less toxic and moredirect alternative to selectively inhibit the protein synthesismachinery of tumor cells regardless of their tissue of originor mutation profile.

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

Authors' ContributionsConception and design: J. Uniacke, J.K. Perera, G. Lachance, S. LeeDevelopment of methodology: J. Uniacke, J.K. Perera, G. Lachance, S. LeeAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): J.K. Perera, C.B. FranciscoAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): J. Uniacke, J.K. Perera, G. Lachance, S. LeeWriting, review, and/or revision of the manuscript: J. Uniacke, J.K. Perera,G. Lachance, S. LeeAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructingdatabases): J.K. Perera, G. Lachance, C.B. Francisco,S. LeeStudy supervision: S. Lee

AcknowledgmentsThe authors thank Martin Holcik for reagents and technical advice and

Josianne Payette for technical assistance.

Grant SupportThis work was funded by the Canadian Institutes of Health Research (grant

#220392; S. Lee).The costs of publication of this article were defrayed in part by the payment of

page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received August 8, 2013; revised December 6, 2013; accepted December 20,2013; published OnlineFirst January 9, 2014.

References1. Hockel M, Schlenger K, Knoop C, Vaupel P. Oxygenation of carcino-

mas of the uterine cervix: evaluation by computerized O2 tensionmeasurements. Cancer Res 1991;51:6098–102.

2. Vaupel P, Schlenger K, Knoop C, Hockel M. Oxygenation of humantumors: evaluation of tissue oxygen distribution in breast cancersby computerized O2 tension measurements. Cancer Res 1991;51:3316–22.

3. Franko AJ, Sutherland RM. Oxygen diffusion distance and devel-opment of necrosis in multicell spheroids. Radiat Res 1979;79:439–53.

4. Harris AL. Hypoxia–a key regulatory factor in tumour growth. Nat RevCancer 2002;2:38–47.

5. KondoK,Klco J,Nakamura E, LechpammerM,KaelinWGJr. Inhibitionof HIF is necessary for tumor suppression by the von Hippel–Lindauprotein. Cancer Cell 2002;1:237–46.

6. Maranchie JK, Vasselli JR, Riss J, Bonifacino JS, Linehan WM, Klaus-ner RD. The contribution of VHL substrate binding and HIF1-alpha tothe phenotype of VHL loss in renal cell carcinoma. Cancer Cell2002;1:247–55.

7. Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC, Cock-man ME, et al. The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 1999;399:271–5.

8. Wang GL, Jiang BH, Rue EA, Semenza GL. Hypoxia-inducible factor 1is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2tension. Proc Natl Acad Sci U S A 1995;92:5510–4.

9. Denko NC, Fontana LA, Hudson KM, Sutphin PD, Raychaudhuri S,Altman R, et al. Investigating hypoxic tumor physiology through geneexpression patterns. Oncogene 2003;22:5907–14.

10. Sonenberg N, Hinnebusch AG. Regulation of translation initiation ineukaryotes: mechanisms and biological targets. Cell 2009;136:731–45.

11. Gebauer F, Hentze MW. Molecular mechanisms of translational con-trol. Nat Rev Mol Cell Biol 2004;5:827–35.

12. DowlingRJ, Topisirovic I, Alain T,BidinostiM, FonsecaBD,PetroulakisE, et al. mTORC1-mediated cell proliferation, but not cell growth,controlled by the 4E-BPs. Science 2010;328:1172–6.

13. Wullschleger S, Loewith R, Hall MN. TOR signaling in growth andmetabolism. Cell 2006;124:471–84.

14. Brunn GJ, Hudson CC, Sekulic A,Williams JM, Hosoi H, Houghton PJ,et al. Phosphorylation of the translational repressor PHAS-I by themammalian target of rapamycin. Science 1997;277:99–101.

15. Lin TA, Kong X, Haystead TA, Pause A, BelshamG, Sonenberg N, et al.PHAS-I as a link between mitogen-activated protein kinase and trans-lation initiation. Science 1994;266:653–6.

16. Richter JD, Sonenberg N. Regulation of cap-dependent translation byeIF4E inhibitory proteins. Nature 2005;433:477–80.

17. Koumenis C, Naczki C, Koritzinsky M, Rastani S, Diehl A, Sonen-berg N, et al. Regulation of protein synthesis by hypoxia viaactivation of the endoplasmic reticulum kinase PERK and phos-phorylation of the translation initiation factor eIF2alpha. Mol CellBiol 2002;22:7405–16.

Uniacke et al.





18. Tinton SA, Buc-Calderon PM. Hypoxia increases the association of4E-binding protein 1 with the initiation factor 4E in isolated rat hepa-tocytes. FEBS Lett 1999;446:55–9.

19. 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.

20. Young RM, Wang SJ, Gordan JD, Ji X, Liebhaber SA, Simon MC.Hypoxia-mediated selectivemRNA translation by an internal ribosomeentry site-independent mechanism. J Biol Chem 2008;283:16309–19.

21. Uniacke J, Holterman CE, Lachance G, Franovic A, Jacob MD, FabianMR, et al. An oxygen-regulated switch in the protein synthesismachin-ery. Nature 2012;486:126–9.

22. WoutersBG, van denBeucken T,MagagninMG,KoritzinskyM, Fels D,Koumenis C. Control of the hypoxic response through regulation ofmRNA translation. Semin Cell Dev Biol 2005;16:487–501.

23. Cho PF, Poulin F, Cho-Park YA, Cho-Park IB, Chicoine JD, Lasko P,et al. A newparadigm for translational control: inhibition via 50-30 mRNAtethering by Bicoid and the eIF4E cognate 4EHP. Cell 2005;121:411–23.

24. Morita M, Ler LW, Fabian MR, Siddiqui N, Mullin M, Henderson VC,et al. A novel 4EHP-GIGYF2 translational repressor complex is essen-tial for mammalian development. Mol Cell Biol 2012;32:3585–93.

25. Rom E, Kim HC, Gingras AC, Marcotrigiano J, Favre D, Olsen H, et al.Cloning and characterization of 4EHP, a novel mammalian eIF4E-related cap-binding protein. J Biol Chem 1998;273:13104–9.

26. Tee AR, Tee JA, Blenis J. Characterizing the interaction of the mam-malian eIF4E-related protein 4EHP with 4E-BP1. FEBS Lett 2004;564:58–62.

27. Franovic A, Gunaratnam L, Smith K, Robert I, Patten D, Lee S.Translational up-regulation of the EGFR by tumor hypoxia providesa nonmutational explanation for its overexpression in human cancer.Proc Natl Acad Sci U S A 2007;104:13092–7.

28. Franovic A, Holterman CE, Payette J, Lee S. Human cancers convergeat the HIF-2alpha oncogenic axis. Proc Natl Acad Sci U S A 2009;106:21306–11.

29. Whiteside TL. The tumor microenvironment and its role in promotingtumor growth. Oncogene 2008;27:5904–12.

30. Liu L, Cash TP, Jones RG, Keith B, Thompson CB, Simon MC.Hypoxia-induced energy stress regulates mRNA translation and cellgrowth. Mol Cell 2006;21:521–31.

31. Raleigh JA, Miller GG, Franko AJ, Koch CJ, Fuciarelli AF, Kelly DA.Fluorescence immunohistochemical detection of hypoxic cells inspheroids and tumours. Br J Cancer 1987;56:395–400.

32. Brown JM, Giaccia AJ. The unique physiology of solid tumors: oppor-tunities (and problems) for cancer therapy. Cancer Res 1998;58:1408–16.

33. Krishna MC, Matsumoto S, Yasui H, Saito K, Devasahayam N, Sub-ramanian S, et al. Electron paramagnetic resonance imaging of tumorpO(2). Radiat Res 2012;177:376–86.

34. Sato Y, Yamamoto N, Kunitoh H, Ohe Y, Minami H, Laird NM, et al.Genome-wide association study on overall survival of advanced non-small cell lungcancer patients treatedwith carboplatin andpaclitaxel. JThorac Oncol 2011;6:132–8.

35. Ramaswamy S, Ross KN, Lander ES, Golub TR. Amolecular signatureof metastasis in primary solid tumors. Nat Genet 2003;33:49–54.

36. Graff JR, Konicek BW, Vincent TM, Lynch RL, Monteith D, Weir SN,et al. Therapeutic suppression of translation initiation factor eIF4Eexpression reduces tumor growth without toxicity. J Clin Invest2007;117:2638–48.

37. Kerekatte V, Smiley K, Hu B, Smith A, Gelder F, De Benedetti A. Theproto-oncogene/translation factor eIF4E: a survey of its expression inbreast carcinomas. Int J Cancer 1995;64:27–31.

38. Nathan CO, Carter P, Liu L, Li BD, Abreo F, Tudor A, et al. Elevatedexpression of eIF4E and FGF-2 isoforms during vascularization ofbreast carcinomas. Oncogene 1997;15:1087–94.

39. Rosenwald IB, Chen JJ, Wang S, Savas L, London IM, Pullman J.Upregulation of protein synthesis initiation factor eIF-4E is an earlyevent during colon carcinogenesis. Oncogene 1999;18:2507–17.

40. Topisirovic I, Guzman ML, McConnell MJ, Licht JD, Culjkovic B,Neering SJ, et al. Aberrant eukaryotic translation initiation factor 4E-dependent mRNA transport impedes hematopoietic differentiationand contributes to leukemogenesis. Mol Cell Biol 2003;23:8992–9002.

41. Wang S, Rosenwald IB, Hutzler MJ, Pihan GA, Savas L, Chen JJ, et al.Expressionof the eukaryotic translation initiation factors 4Eand2alphain non-Hodgkin's lymphomas. Am J Pathol 1999;155:247–55.

42. Yi T, Papadopoulos E, Hagner PR, Wagner G. Hypoxia-induciblefactor-1aalpha (HIF-1alpha) promotes cap-dependent translation ofselective mRNAs through upregulating Initiation factor eIF4E1 inbreast cancer cells under hypoxia conditions. J Biol Chem 2013;288:18732–42.






2014;74:1379-1389. Published OnlineFirst January 9, 2014.Cancer Res James Uniacke, J. Kishan Perera, Gabriel Lachance, et al. Response Proteins to Drive Tumor ProgressionCancer Cells Exploit eIF4E2-Directed Synthesis of Hypoxia

Updated version

10.1158/0008-5472.CAN-13-2278doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2014/01/09/0008-5472.CAN-13-2278.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/74/5/1379.full#ref-list-1

This article cites 42 articles, 15 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/74/5/1379.full#related-urls

This article has been cited by 5 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

[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://cancerres.aacrjournals.org/content/74/5/1379To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-13-2278

http://cancerres.aacrjournals.org/content/suppl/2014/01/09/0008-5472.CAN-13-2278.DC1

http://cancerres.aacrjournals.org/content/74/5/1379.full#ref-list-1

http://cancerres.aacrjournals.org/content/74/5/1379.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/74/5/1379


Documents

Cancer Cells Exploit eIF4E2-Directed Synthesis of Hypoxia ... · Microenvironment and Immunology Cancer Cells Exploit eIF4E2-Directed Synthesis of Hypoxia Response Proteins to Drive