Defects in Translational Regulation Mediated by the {alpha} Subunit of Eukaryotic Initiation Factor 2 Inhibit Antiviral Activity and Facilitate the Malignant Transformation of Human Fibroblasts

Suppression of protein synthesis through phosphorylation ofthe translation initiation factor ${alpha}$ subunit of eukaryotic initiationfactor 2 (eIF2 ${alpha}$ ) is known to occur in response to many formsof cellular stress. To further study this, we have developednovel cell lines that inducibly express FLAG-tagged versionsof either the phosphomimetic eIF2 ${alpha}$ variant, eIF2 ${alpha}$ -S51D, or thephosphorylation-insensitive eIF2 ${alpha}$ -S51A. These variants showedauthentic subcellular localization, were incorporated into endogenousternary complexes, and were able to modulate overall rates ofprotein synthesis as well as influence cell division. However,phosphorylation of eIF2 ${alpha}$ failed to induce cell death or sensitizecells to killing by proapoptotic stimuli, though it was ableto inhibit viral replication, confirming the role of eIF2 ${alpha}$ inhost defense. Further, although the eIF2 ${alpha}$ -S51A variant has beenshown to transform NIH 3T3 cells, it was unable to transformthe murine fibroblast 3T3 L1 cell line. To therefore clarifythis issue, we explored the role of eIF2 ${alpha}$ in growth control anddemonstrated that the eIF2 ${alpha}$ -S51A variant is capable of collaboratingwith hTERT and the simian virus 40 large T antigen in the transformationof primary human kidney cells. Thus, dysregulation of translationinitiation is indeed sufficient to cooperate with defined oncogenicelements and participate in the tumorigenesis of human tissue.

INTRODUCTION

Top

Introduction

The initiation of protein synthesis in eukaryotes is a highlycomplex and conserved process involving at least 13 initiationfactors, many of which are themselves assembled from numeroussubunits (3, 51, 56). The regulation of protein synthesis canbe greatly affected following exposure to various forms of cellstress including nutrient deprivation, contact with biologicpathogens, fluctuations in temperature, or the presence of toxiccompounds (18, 35). A key translation factor that is a frequenttarget of regulation by stress-sensitive kinases is the ${alpha}$ subunitof the eukaryotic translation initiation factor 2 complex (eIF2 ${alpha}$ )(13, 14, 63, 72). eIF2 is a heterotrimer composed of three subunits( ${alpha}$ , ß, and ${gamma}$ ) which functions by associating with GTPand the initiator Met-tRNA_i to form a ternary complex (48, 62).The ternary complex delivers the Met-tRNA_i to the 40S ribosomalsubunit which, along with other translation factors includingeIF3, forms the 43S preinitiation structure (8, 50, 69). Newlyassembled 43S ribosome-eIF complexes associate with an mRNAtranscript near the 5' m⁷G cap and advance along the transcriptin a 3' direction until an AUG start codon is located withinthe context of an appropriate Kozak sequence (25, 44, 53). Oncethe AUG codon has been recognized, GTP bound by eIF2 is hydrolyzedin a reaction catalyzed, in part, by another initiation factor,eIF5 (4). The Met-tRNA_i is subsequently released from the ternarycomplex to initiate nascent peptide chain synthesis, and eIF2dissociates from the 43S initiation complex. The GDP associatedwith the free eIF2 is exchanged for GTP by the activity of theeIF2B complex, which is itself a heteropentamer comprised of ${alpha}$ , ß, ${gamma}$ , ${delta}$ , and ${varepsilon}$ subunits (2, 54, 59). Following GTPexchange, eIF2 is incorporated into a new ternary complex andthe next round of initiation begins (37).

Phosphorylation on serine 51 of eIF2 ${alpha}$ by stress-responsive kinasescauses eIF2 to acquire an increased affinity for, and functionallysequester, the GTP exchange factor eIF2B, which is requiredfor maintaining eIF2 activity (40). Thus, in response to stress,eIF2 ${alpha}$ kinases can depress global translation rates by inhibitingeIF2-GTP recycling and, subsequently, initiation of translation.For example, accumulation of misfolded proteins in the endoplasmicreticulum (ER) leads to activation of an ER-resident eIF2 ${alpha}$ kinasealternately named PKR-like endoplasmic reticular kinase (PERK)and pancreatic eIF2 ${alpha}$ kinase (33, 67). Similarly, the yeast eIF2 ${alpha}$ kinase GCN2 and its mammalian homologues function as cytoplasmicsensors of amino acid levels via two His-tRNA-like domains intheir carboxy termini (70). GCN2 kinase activity is up-regulatedunder starvation conditions during which the levels of chargedtRNAs fall (21). The heme-regulated inhibitor kinase, in contrast,is predominantly expressed in erythroid cells and is negativelyregulated by hemin binding (11, 12). In addition, the interferon-inducible,double-stranded RNA (dsRNA)-regulated kinase PKR is activatedby dsRNA produced during viral infections and functions in hostdefense to prevent translation of viral transcripts (6, 7, 43,52, 71).

A number of recent reports, however, have indicated that elevatedlevels of phosphorylated eIF2 ${alpha}$ may actually serve to specificallyenhance the translation of selected mRNAs encoding proteinsthat require production in response to stress in mammalian cells.The mechanism of transcript-specific translational up-regulationhas been reported to be, in part, dependent upon upstream openreading frames (uORFs) in the 5' untranslated region (UTR) ofthe mRNA (31, 65). Under normal physiologic conditions, theseshort uORFs lower the efficiency of translation, presumablyby impeding the progress of the scanning ribosome. However,under conditions in which the levels of phosphorylated eIF2 ${alpha}$ rise and levels of available ternary complex fall, these uORFsmay favor the association of the transcript with active ribosomes.Examples to date of transcripts that are regulated in this mannerinclude the transcription factors GCN4 in Saccharomyces cerevisiaeand ATF4 in mammalian cells (17, 20, 31). Additionally, somemRNAs with 5' UTRs containing internal ribosome entry site elementsare also translationally up-regulated when eIF2 ${alpha}$ is phosphorylated(26).

To further clarify the role of eIF2 ${alpha}$ in apoptosis, transformation,and gene regulation, we have developed inducible and constitutiveexpression systems for wild-type (WT) and variant forms of eIF2 ${alpha}$ .Here we report that regulation of translation initiation througheIF2 ${alpha}$ is sufficient to inhibit viral replication in the absenceof other eIF2-independent stress-responsive pathways but cannotaccount for the induction of apoptosis or for aspects of thebroad gene regulation observed upon activation of stress-responsivekinases. Additionally, we demonstrate for the first time thata translation factor can transform human cells in collaborationwith defined genetic elements hTERT and simian virus 40 largeT antigen, collectively confirming the importance of translationalregulation in tumorigenesis.

MATERIALS AND METHODS

Top

Materials and Methods

Plasmid construction. The pTREFLAG vector was derived from the pTRE vector (pUHD15-neo1)(28) by the insertion of the FLAG tag coding sequence precededby an in-frame ATG initiation codon into the SacII/NdeI restrictionsites in pTRE. pTREFLAG-eIF2 ${alpha}$ WT was derived from pTREFLAG bythe insertion of the WT eIF2 ${alpha}$ sequence into the NdeI/BamHI restrictionsites. pTREFLAG-eIF2 ${alpha}$ S51A and pTREFLAG-eIF2 ${alpha}$ S51D were developedfrom pTREFLAG-eIF2 ${alpha}$ WT by using the QuickChange site-directedmutagenesis kit (Stratagene, La Jolla, Calif.) and contain pointmutations rendering a serine-to-alanine or serine-to-aspartic-acidchange, respectively, at position 51. pFB-NeoWTeIF2 ${alpha}$ , pFB-NeoeIF2 ${alpha}$ S51D,and pFB-NeoeIF2 ${alpha}$ S51A were constructed by inserting the WT orvariant eIF2 ${alpha}$ sequences into the EcoRI/BamHI restriction sitesin the pFB-Neo vector (Stratagene). pFB-NeoWTRas and pFB-NeoRasV12vectors were generated by insertion of the WT Ras or RasV12coding sequences into the EcoRI/BamHI restriction sites in pFB-Neo.pVPack-GP and pVPack env expression vectors were purchased fromStratagene. The luciferase reporter constructs were obtainedby cloning the entire 5' leader sequence of the Fas gene intothe PGL3 control vector (Promega, Madison, Wis.), in frame withthe ATG start codon from the luciferase ORF.

Cell lines. 3T3 L1 cells (Clontech, Palo Alto, Calif.) stably transfectedwith a pTETOFF vector (pUHD15-neo1) (28) were subsequently cotransfectedwith the pTREFLAG-eIF2 ${alpha}$ WT, pTREFLAG-eIF2 ${alpha}$ S51A, or pTREFLAG-eIF2 ${alpha}$ S51Dvector along with pTK-HYG (Clontech) by using Lipofectamine(Gibco-BRL, Grand Island, N.Y.), according to the manufacturer'sprotocol. Immediately upon removal of the DNA-liposome complexes,doxycycline (DOX; Sigma Chemical, St. Louis, Mo.) was addedto the medium (5 µg/ml). After 48 h of recovery, cellswere selected with 250 µg of G418 (Gibco-BRL)/ml and 100µg of hygromycin (Clontech)/ml. Resistant colonies wereisolated and expanded. Protein expression and inducibility ofindividual clones were examined by Western blotting with lysatefrom cells passaged for 1 week in the presence or absence ofDOX. NIH 3T3 and 3T3 L1 cell lines were obtained from the AmericanType Culture Collection.

Immunoblot analysis. Protein extracts from cell lines were prepared by disruptingcells in lysis buffer (10 mM Tris-HCl [pH 7.5], 50 mM KCl, 1mM dithiothreitol, 2 mM EDTA, 0.2 mM phenylmethylsulfonyl fluoride,100 U of aprotinin/ml, 1% NP-40) (all reagents obtained fromSigma). Supernatants were added to an equal volume of loadingbuffer (5% sodium dodecyl sulfate [SDS], 5% ß-mercaptoethanol,20% glycerol, and 150 mM Tris-HCl [pH 7.50]) and boiled for2 min prior to being loaded on an SDS-polyacrylamide gel. Afterelectrophoretic resolution, proteins were transferred to nitrocellulosemembranes, incubated for 1 h in blocking solution (phosphate-bufferedsaline [PBS], 0.1% Tween 20, 10% nonfat dry milk) at room temperature,and incubated with specific monoclonal or polyclonal antibodiesovernight at 4°C. Membranes were washed three times in PBS-Tween20 and incubated with horseradish peroxidase-conjugated secondaryanti-mouse or anti-rabbit antibodies (Jackson ImmunoResearchLaboratories, West Grove, Pa.). Following a second washing,proteins were visualized by addition of chemiluminescent substrate(Pierce Chemicals, Rockford, Ill.).

FLAG M5 monoclonal antibody (MAb) and ß-actin MAbswere obtained from Sigma. Antibodies to Fas, Bax, and Bcl-Xand Bcl-2 were purchased from Santa Cruz Biotechnology (SantaCruz, Calif.). Anti-FADD MAb was obtained from Upstate BiotechnologyInc. (Lake Placid, N.Y.). Anti-Ras MAb was purchased from OncogeneInc. (Cambridge, Mass.).

Protein synthesis analysis. Equal numbers of cells passaged in either the presence or theabsence of DOX were split into six-well dishes and allowed tosettle for 48 h. Cells were labeled with 75 µCi of [³⁵S]methionine(Amersham Pharmacia Biotech, Piscataway, N.J.)/ml in methionine-freemedium (Gibco-BRL) supplemented with 10% dialyzed fetal bovineserum (FBS; Gibco-BRL) for 30 min at 37°C. Cells were washedthree times in warm PBS, detached with trypsin-EDTA, counted,and finally disrupted with lysis buffer. Lysate volumes of equivalentcell numbers were added to bovine serum albumin-methionine carrierbuffer (0.002% bovine serum albumin [Pierce], 0.01% methionine[Sigma], 0.1 volume of urea sample buffer). Urea sample bufferconsisted of 50% urea (Sigma), 0.2% NP-40, 0.05% Coomassie blue,and 0.5% ß-mercaptoethanol. Protein bound counts wereprecipitated by the addition of 0.5 volume of ice-cold 50% trichloroaceticacid (TCA; Sigma) and centrifuged at 14,000 x g for 5 min at4°C. Pellets were washed with 10% TCA at 4°C and resuspendedin Soluene 350 (Packard BioScience, Boston, Mass.). Counts wereanalyzed by liquid scintillation.

Ternary complex immunoprecipitations. Coimmunoprecipitations were performed by transfecting subconfluentmonolayers of 293T cells (American Type Culture Collection)in six-well dishes with either a vector expressing FLAG-taggedWT eIF2 ${alpha}$ or a control plasmid expressing a nonspecific FLAG-taggedprotein. Forty-eight hours following transfection, cells werelysed with lysis buffer. Whole-cell lysate was added to low-saltbuffer (20 mM Tris-HCl [pH 7.4], 150 mM NaCl, 5 mM EDTA, 0.2%Triton X-100, 0.1% ß-mercaptoethanol) and preclearedby incubation with 50 µl of protein G agarose (InvitrogenCorp., Carlsbad, Calif.) and 2.5 µl of normal rabbit immunoglobulinG (Santa Cruz Biotechnology) for 1 h at 4°C. Protein G wasremoved by centrifugation, and an additional 50 µl ofprotein G slurry was added along with 5 µl of rabbit polyclonalanti-FLAG antiserum (Sigma). Lysates were incubated for 2 hat 4°C. Protein G was pelleted by centrifugation, and thelysate was removed. Protein G pellet was washed three timeswith 1 ml of low-salt buffer and then heated.

RNA analysis. Cell lines were induced in the absence of DOX for the indicatedperiods, and total RNA was harvested with the RNeasy minikit(Qiagen, Valencia, Calif.). RNA was incubated with an ${alpha}$ -³²P-labeledmAPO-2 or mAPO-3 probe set (Riboquant; PharMingen, San Diego,Calif.) according to the manufacturer's protocol. FollowingRNase treatment, protected probes were resolved using 5% polyacrylamidegels and imaged with autoradiography.

Retrovirus production. Individual retrovirus was produced according to the manufacturer'sinstructions included with pVPack Moloney murine leukemia virus-basedretroviral expression vectors (Stratagene). Briefly, 293T cellsat 50% confluency in 10-cm-diameter dishes were transfectedwith 5 µg each of pVPack-GP, pVPack-Eco or pVPack-VSV-G,and pFB-Neo or pFB-Neo containing the appropriate inserted gene,with the use of Lipofectamine. Retrovirus-containing supernatantswere harvested after 48 h and frozen at -80°C until needed.Target cells were subsequently infected by the addition of retrovirus-containingsupernatants to the medium along with DEAE-dextran to a finalconcentration of 10 µg/ml. Target cells were selected24 h following infection by the addition of 400 µg ofneomycin (Gibco-BRL)/ml.

Apoptosis analysis. Terminal deoxynucleotidyltransferase-mediated dUTP-biotin nickend labeling (TUNEL) was done using the fluorescein in situcell death detection kit (Roche, Mannheim Germany) accordingto the manufacturer's instructions. The annexin V-propidiumiodide (PI) binding assay kit was purchased from R & D Systems(Minneapolis, Minn.). Quantitation of cell killing in responseto apoptotic stimuli was done using trypan blue exclusion following12 and 24 h of treatment. Jo-2 antibody was used from 0.1 to1.0 µg/ml. Tumor necrosis factor (TNF) was used from 10to 750 ng/ml. Poly(IC) was used in concentrations from 0.25to 2 µg/ml. Soluble recombinant TRAIL was used in concentrationsfrom 10 to 750 ng/ml.

Oncogenic collaboration. One hundred thousand HEK cells stably transduced with largeT antigen and hTERT (a gift from Robert Weinberg) (29) weretransduced with retrovirus encoding WT eIF2 ${alpha}$ , eIF2 ${alpha}$ -S51A, or RasV12,according to the manufacturer's instructions. Forty-eight hoursfollowing transduction, cells were selected with 0.5 µgof puromycin (Invitrogen)/ml for 10 days. Remaining colonieswere subcloned and photographed for morphology.

Anchorage-independent growth. Cell lines were trypsinized and counted, and 500 or 5,000 cellswere mixed with 1 ml of warm Dulbecco modified Eagle medium(DMEM), 10% FBS, and 0.5% low-melting-temperature agarose (LMTA).Cells were then layered onto 1 ml of DMEM-10% FBS with 0.75%LMTA that had previously been added to individual wells of asix-well plate. Finally an additional layer consisting of DMEMwith 0.75% LMTA was added over the top. Cells were culturedat 37°C and 5% CO₂ for approximately 3 weeks (21 days).Colony growth was scored using light microscopy.

RESULTS

Top

Results

Inducible expression of eIF2 ${alpha}$ variants in 3T3 L1 cells. To analyze the specific effects of eIF2 ${alpha}$ phosphorylation on thecell in the absence of other eIF2-independent stress signalingcascades, we established novel 3T3 L1 cell lines that induciblyexpress a FLAG-tagged phosphomimetic variant of eIF2 ${alpha}$ , referredto as eIF2 ${alpha}$ -S51D, or the phosphorylation-incompetent varianteIF2 ${alpha}$ -S51A (61). Clonal 3T3 L1 cell populations were isolatedwhich express the exogenous forms of eIF2 ${alpha}$ under the controlof a DOX-repressible promoter (28). Removal of DOX from theculture medium induced a dose-dependent accumulation of boththe eIF2 ${alpha}$ -S51A and the eIF2 ${alpha}$ -S51D variants that was detectableafter 3 days by using a MAb specifically recognizing the FLAGepitope tag (42) (Fig. 1A). Maximal protein levels of eIF2 ${alpha}$ -S51Awere observed approximately 3 days postinduction. In contrast,eIF2 ${alpha}$ -S51D accumulated more slowly, reaching apparent maximalexpression after 6 days (Fig. 1A, lanes 4 to 9). These datacollectively demonstrate the successful inducible expressionof eIF2 ${alpha}$ variants in 3T3 L1 cells.

View larger version (60K):
[in this window]
[in a new window]
FIG. 1. (A) Inducible expression of eIF2 ${alpha}$ variants in 3T3 L1 cells. Cells carrying vector alone or tetracycline-inducible plasmids expressing empty vector, FLAG eIF2 ${alpha}$ -S51D, or FLAG eIF2 ${alpha}$ -S51A were induced by the withdrawal of DOX, and whole-cell lysates were harvested at 0, 3, and 6 days postinduction. Cell extracts were analyzed for the expression of FLAG-tagged proteins by immunoblotting with a MAb directed against the FLAG epitope as described in Materials and Methods. Lane 1 is whole-cell lysate harvested from 293T cells transiently transfected for 48 h with a plasmid expressing FLAG-tagged WT eIF2 ${alpha}$ . (B) Morphological changes induced by expression of eIF2 ${alpha}$ variants. The vector-, FLAG eIF2 ${alpha}$ -S51D-, and FLAG eIF2 ${alpha}$ -S51A-expressing cell lines were induced to express the indicated variants for 6 days, and cell morphology was monitored at 0 and 6 days postinduction by phase-contrast microscopy at x4 and x20 magnifications. (C) Levels of total and phosphorylated eIF2 ${alpha}$ in DOX-inducible cell lines. Whole-cell lysates from 3T3 L1 DOX-inducible cells either uninduced or induced to express the indicated eIF2 ${alpha}$ variants for 6 days were harvested, electrophoresed, and probed with MAbs directed against either total eIF2 ${alpha}$ or eIF2 ${alpha}$ specifically phosphorylated on serine 51.

It was found that, following removal of DOX from the culturemedium, a pronounced change in the morphology of the eIF2 ${alpha}$ -S51D-and eIF2 ${alpha}$ -S51A-expressing cell lines was evident, which beganapproximately 48 h postinduction (Fig. 1B). A number of clonesexpressing comparable levels of the two FLAG-tagged constructswere screened, and a similar morphological change was evident,indicating that the observed effect was not clone dependent(data not shown). Similar morphological changes were not observedin vector-containing cell lines (Fig. 1B), and those eIF2 ${alpha}$ -expressingclones demonstrating the most profound alteration in morphologywere selected for further study. Typically cells expressingthe eIF2 ${alpha}$ -S51D variant exhibited a thin and spindle-like morphologywhile, in contrast, cells expressing the eIF2 ${alpha}$ -S51A variant becameoverall smaller and developed a refractile cytoplasm (Fig. 1B)but did not form foci in culture and were not found to exhibitother hallmarks of malignancy (data not shown).

Recent reports have shown that phosphorylation of eIF2 ${alpha}$ inducesa GADD34-dependent phosphatase activity that acts as a negativeregulator of the stress response by specifically dephosphorylatingeIF2 ${alpha}$ (55). To further characterize our inducible cell linesand more specifically to ascertain whether such a stress responsewas being activated by the expression of eIF2 ${alpha}$ -S51D in our celllines, the levels of endogenous phosphorylated eIF2 ${alpha}$ were assayedusing a MAb specifically recognizing phosphorylated eIF2 ${alpha}$ before,and immediately following, induction of our eIF2 ${alpha}$ variants (Fig.1C). We were, however, unable to demonstrate any changes inthe levels of the endogenous pools of phosphorylated eIF2 ${alpha}$ inresponse to expression of either the eIF2 ${alpha}$ -S51D or eIF2 ${alpha}$ -S51Avariant.

Since available data indicate that alterations in cellular morphologycan be associated with changes in the organization of the underlyingcytoskeleton, we therefore further characterized the alteredmorphology caused by the S51A and S51D variants, by examiningthe structure of polymerized F-actin stress fibers, before andafter induction by immunoblotting and immunofluorescence (Fig.2A). These data indicated that expression of either eIF2 ${alpha}$ -S51Dor eIF2 ${alpha}$ -S51A did indeed produce changes in the organizationof the polymerized actin filaments, although the levels of actinprotein did not necessarily appear to fluctuate as determinedby Western blotting (Fig. 1A).

View larger version (36K):
[in this window]
[in a new window]
FIG. 2. (A) Polymerized F-actin filament organization in FLAG eIF2 ${alpha}$ variant-expressing cells. Vector-, FLAG S51A-, and FLAG S51D-expressing cell lines grown on glass coverslips were permeabilized and stained with fluorescein isothiocyanate-conjugated phalloidin proteins that specifically bind to polymerized F-actin filaments on day 0 or day 6 postinduction. Actin filaments were visualized by immunofluorescence microscopy. (B) Subcellular localization of endogenous and FLAG-tagged eIF2 ${alpha}$ . Staining of 3T3 L1 cells with nonspecific normal mouse immunoglobulin G was included as a negative control (i). Expression and localization of endogenous eIF2 ${alpha}$ in normal 3T3 L1 cells were determined using immunofluorescence with a MAb directed against native eIF2 ${alpha}$ (ii). Localization of FLAG-tagged eIF2 ${alpha}$ in normal 3T3 L1 cells was determined by immunofluorescence with a MAb directed against the FLAG epitope (clone M5; Sigma) following transient transfection for 48 h with a vector expressing FLAG-tagged WT eIF2 ${alpha}$ (iii). Localization of inducible FLAG-tagged eIF2 ${alpha}$ variants in empty vector-, FLAG eIF2 ${alpha}$ -S51A-, or FLAG eIF2 ${alpha}$ -S51D-expressing 3T3 L1 cell lines was determined by immunofluorescence with an antibody to the FLAG epitope following 6 days of induction in the absence of DOX (iv to vi). IgG, immunoglobulin G.

As the inducible eIF2 ${alpha}$ variants carry an 8-amino-acid N-terminalFLAG epitope tag, we next attempted to confirm that these epitope-taggedversions displayed authentic subcellular localization. The cellularlocalization of endogenous eIF2 ${alpha}$ was first determined using immunofluorescencestaining in normal 3T3 L1 cells with monoclonal and polyclonalantibodies directed against eIF2 ${alpha}$ (Fig. 2B). Unexpectedly, endogenouseIF2 ${alpha}$ was almost exclusively localized to the nucleus (Fig. 2B,panel ii). 3T3 L1 cells transiently transfected with WT FLAG-taggedeIF2 ${alpha}$ and stained with a MAb directed against the FLAG epitopealso indicated that the FLAG-tagged versions appeared exclusivelynuclear (Fig. 2B, panel iii). Immunofluorescence staining ofour vector-, eIF2 ${alpha}$ -S51A-, or eIF2 ${alpha}$ -S51D-expressing cell line withthe anti-FLAG MAb confirmed nuclear localization of our FLAG-taggedeIF2 ${alpha}$ variants (Fig. 2B, panels iv to vi), strongly suggestingthat the recombinant proteins were being authentically compartmentalizedwithin the cell. These data are in agreement with other recentreports indicating that a large fraction of the recombinanteIF2 ${alpha}$ localizes to the nucleus (16, 41). As a control to ensurethat the FLAG tag was not sterically interfering with the incorporationof eIF2 ${alpha}$ into eIF2 along with endogenous eIF2ß andeIF2 ${gamma}$ , we also transiently overexpressed either our WT FLAG eIF2 ${alpha}$ or a nonspecific FLAG-tagged protein and subsequently immunoprecipitatedwhole-cell lysates by using a polyclonal antiserum to the FLAGepitope. Our results showed that immunoprecipitated eIF2 ${alpha}$ specificallyassociated with endogenous eIF2ß, indicating thatFLAG-eIF2 ${alpha}$ variants are correctly combining with their naturalbinding partners in vivo (Fig. 3A). As additional confirmation,we then coimmunoprecipitated FLAG eIF2 ${alpha}$ and eIF2ß fromthe inducible 3T3 L1 cell lines, establishing that recombinanteIF2 ${alpha}$ is indeed associated with at least one of the other membersof the ternary complex and likely is fully functional (Fig.3B). To ascertain the relative levels of overexpression of FLAG-taggedeIF2 ${alpha}$ compared to the endogenous protein, we separated the twoforms on the basis of size by SDS-14% polyacrylamide gel electrophoresis(PAGE) and probed with a pan-eIF2 ${alpha}$ MAb (Fig. 3C). Densitometricanalysis indicated that the FLAG-tagged proteins were expressedat levels two- to threefold greater than was the endogenouseIF2 ${alpha}$ . Collectively our data would indicate that the FLAG-taggedeIF2 ${alpha}$ is authentically recognized in the cell.

View larger version (38K):
[in this window]
[in a new window]
FIG. 3. (A) Association of FLAG-tagged eIF2 ${alpha}$ with native ternary complexes in 293T cells. 293T cells were mock transfected or transiently transfected with plasmids encoding WT FLAG eIF2 ${alpha}$ or a nonspecific FLAG-tagged protein (NS FLAG). Forty-eight hours later, whole-cell lysates were harvested and split equally. One half was used in Western blotting to quantitate expression of transfected proteins. The blot was subsequently probed with antibodies to eIF2ß and actin. The remaining lysate was used in an immunoprecipitation with a polyclonal antiserum recognizing the FLAG epitope. Immunoprecipitated proteins were resolved by SDS-PAGE and probed with antibodies recognizing eIF2 ${alpha}$ or eIF2ß. (B) Association of FLAG-tagged eIF2 ${alpha}$ with native ternary complexes in inducible 3T3 L1 cells. Whole-cell lysates from empty vector-, eIF2 ${alpha}$ -S51A-, or eIF2 ${alpha}$ -S51D-expressing inducible 3T3 L1 cell lines were harvested following 6 days of induction and used in an immunoprecipitation with apolyclonal antiserum directed against the FLAG tag. (C) Quantitation of expression levels of endogenous and FLAG-tagged eIF2 ${alpha}$ in inducible 3T3 L1 cell lines. Whole-cell lysates harvested from empty vector-, eIF2 ${alpha}$ -S51A-, or eIF2 ${alpha}$ -S51D-expressing cell lines were harvested after 6 days of induction. Lysates were resolved by SDS-14% PAGE and probed with a MAb directed against eIF2 ${alpha}$ . (D) Regulation of translation rates by eIF2 ${alpha}$ variants. Triplicate samples of vector-, S51D-, or S51A-expressing inducible cell lines were passaged in the presence or absence of DOX for 6 days. Cells were transiently labeled with [³⁵S]methionine for 15 min. Total protein from whole-cell lysates was precipitated with TCA and resuspended. Protein bound radioactive counts were measured via scintillation. Standard deviations are shown. (E) Regulation of growth rates by eIF2 ${alpha}$ variants. Equal numbers of cells from vector- and eIF2 ${alpha}$ variant-expressing cell lines were seeded in triplicate samples in 12-well dishes and passaged in the presence or absence of DOX for 6 days. Cells were trypsinized, and cell counts were determined daily.

Regulation of protein synthesis and growth rates by eIF2 ${alpha}$ variants. To assess the capacity of our FLAG-tagged versions of the eIF2 ${alpha}$ variants to influence translation, we measured the rates ofprotein synthesis in each cell line in the presence of DOX (uninduced)and following 6 days of induction by using [³⁵S]methionine labeling(Fig. 3D). As expected, vector-expressing control cells didnot display significant perturbation in protein synthesis rates.However, eIF2 ${alpha}$ -S51D-expressing cells demonstrated a significant40% decrease in synthesis rates as measured by label incorporation,while the eIF2 ${alpha}$ -S51A-expressing cell line evidenced an approximately18% increase in protein synthesis. These data are in accordwith reported values from an eIF2 ${alpha}$ -S51A knock-in mouse modelthat reported a 17 to 30% increase in translation rates andconfirm the functional capacity of the FLAG-tagged variants(65).

As translation has been reported to influence cell growth anddivision, the growth characteristics of the inducible 3T3 L1cell lines were analyzed following induction of both eIF2 ${alpha}$ variants.Cell numbers were measured in triplicate experiments for 6 daysfollowing the removal of DOX from the medium for 24 h. All celllines remained viable and continued to divide. However, we observeda marked reduction in the growth rates of the eIF2 ${alpha}$ -S51D-expressingcells, with the reduction in cell number after 6 days approaching35% compared to the identical cell line passaged in the presenceof DOX (Fig. 3E). A slight but reproducible increase in cellnumber in the induced eIF2 ${alpha}$ -S51A-expressing cell lines was observedat the later time points, while no significant change was observedin the growth rates of the vector-expressing cell lines in thepresence or absence of DOX (Fig. 3E). These data confirm thatvariant forms of eIF2 ${alpha}$ can strongly influence rates of proteinsynthesis and govern cell growth.

Phosphorylation of eIF2 ${alpha}$ does not induce apoptosis. Previous investigations into eIF2 ${alpha}$ have shown that transienttransfection of Cos cells with the eIF2 ${alpha}$ -S51D variant is capableof inducing apoptosis (68). In contrast, the eIF2 ${alpha}$ -S51A varianthas been shown to be protective against apoptosis in some systems,implying a role for this translation factor in the regulationof cell death (27). To assess the capacity of eIF2 ${alpha}$ to regulateapoptosis in the absence of eIF2 ${alpha}$ -independent stress-responsivepathways, we assayed for the ability of eIF2 ${alpha}$ -S51D expressionalone to induce apoptosis in our 3T3 L1 cells. However, TUNELof induced and uninduced vector-, eIF2 ${alpha}$ -S51A-, and eIF2 ${alpha}$ -S51D-expressingcell lines did not demonstrate any significant induction ofapoptosis attributable to overexpression of the eIF2 ${alpha}$ -S51D construct(Fig. 4A). Agonistic anti-Fas antibody was used as a positivecontrol for the induction of apoptosis (Fig. 4A). As the TUNELassay lacks sensitivity against cells that are in later stagesof apoptosis and have detached from the culture dish, we subsequentlyassayed for cell killing by utilizing annexin V binding andPI staining. After induction for 4 days, the tissue culturemedium was changed and the cells were induced for an additional3 days. At this point cells from the tissue culture supernatantand those remaining attached were stained and analyzed by flowcytometry. Using this methodology, we observed only a slightdecrease in viability in the induced eIF2 ${alpha}$ -S51D-expressing cells(Fig. 4B). Curiously, these dead cells accumulated as a PI singlystaining population, indicating that they were in the very latestages of apoptosis or had undergone necrosis. However, eIF2 ${alpha}$ -S51D-dependentcell death represented only a 5% decrease in viability comparedto uninduced cells, implying that phosphorylation of eIF2 ${alpha}$ doesnot potently induce spontaneous apoptosis (Fig. 4B).

View larger version (28K):
[in this window]
[in a new window]
FIG. 4. (A) Expression of eIF2 ${alpha}$ -S51D does not induce apoptosis in 3T3 L1 cells. Equal numbers of cells from vector-, S51A-, and S51D-expressing inducible cell lines were seeded onto glass coverslips in 12-well dishes and passaged in the presence or absence of DOX for 6 days. Cells were TUNEL stained and imaged with fluorescence microscopy to detect fragmented nuclear DNA. One microgram of agonistic anti-Fas (Jo-2) antibody per milliliter was added to vector cells for 24 h as a positive control. (B) Annexin V-PI staining of 3T3 L1 cell lines. Vector- and eIF2 ${alpha}$ -S51D-expressing cell lines were left uninduced or induced for 6 days. All cells remaining attached to the substrate and those that had detached into the medium were harvested and pooled. Cells were subsequently stained with annexin V-PI and analyzed by flow cytometry to detect apoptotic death. (C) eIF2 ${alpha}$ variant expression does not sensitize cells to apoptotic death. Duplicate samples of vector-, S51A-, or S51D-expressing cell lines in the induced or uninduced states were treated with the indicated inducers of apoptosis for 24 h. Cells were subsequently harvested, and cell viability was determined by trypan blue exclusion. (D) General inhibition of protein synthesis does not initiate apoptosis. Normal 3T3 L1 cells were treated with the indicated concentrations of CHX or transfected with poly(IC). Six hours later protein synthesis rates were determined by metabolic labeling with [³⁵S]methionine for 15 min. Protein bound counts were precipitated with TCA, resuspended, and quantitated by scintillation. Following 24 h of treatment cells were harvested, and cell viability was determined by trypan blue exclusion.

The lack of obvious apoptosis induced by the eIF2 ${alpha}$ -S51D variantled us to examine whether elevated levels of phosphorylatedeIF2 ${alpha}$ might potentiate or protect eukaryotic cells from deathinduced by common proapoptotic stimuli. To investigate thispossibility, we subjected our cell lines to treatment with avariety of prodeath agents including agonistic anti-mouse Fasantibody (Jo-2), murine TNF alpha (TNF- ${alpha}$ ), poly(I · C),soluble recombinant TRAIL, and the DNA-damaging drug etoposide.However, we did not observe a differential killing in any ofour cell lines with or without induction in response to anyof the agents that were used (Fig. 4C). Multiple concentrationsof each agent yielded similar results (Fig. 4C and data notshown). The unexpected absence of cell death caused by expressionof eIF2 ${alpha}$ -S51D led us to ask whether inhibition of translationin itself is sufficient to induce apoptosis or whether additionalsignaling events are required. To test this hypothesis, we treated3T3 L1 cells with the general protein synthesis inhibitor cycloheximide(CHX) and with poly(IC), a potent activator of the eIF2 ${alpha}$ kinasePKR, in doses sufficient to reduce protein synthesis to ratesroughly 50% of physiologic levels. Following 24 h of each treatment,we assayed for cell viability. Although CHX and poly(IC) reducedprotein synthesis to an equivalent degree, the dsRNA-treatedcells demonstrated a markedly decreased cell viability comparedto CHX-treated populations (Fig. 4D). This result argues that,at least in the 3T3 L1 background, apoptosis is not initiatedby a reduction in protein synthesis rates per se.

Previous work in our laboratory has demonstrated pronouncedregulation of key apoptotic genes in 3T3 L1 cells, such as Fas,in response to overexpression of WT or mutant PKR (5). To assesswhether PKR-dependent regulation of apoptotic proteins is aresult of differential regulation of translation initiation,we examined the RNA and protein levels of these same genes inour inducible cell system including Fas, Bax, and Bcl-2. However,no difference in gene expression at the transcriptional or translationallevel was observed for the genes assayed (Fig. 5A to C). TheFas death receptor in particular has been identified previouslyas a dsRNA-responsive gene and has been shown to be regulatedby PKR activity (5, 22). Although we did not observe an overtchange in Fas expression levels in our inducible cell system,we confirmed this by obtaining the 5' UTR of the Fas gene, whichcontains a uORF, and fused it to the ORF of firefly luciferase.We then assayed for possible subtle contributions of the UTRto translational regulation which may be dependent upon levelsof phosphorylated eIF2 ${alpha}$ (17). In addition, a second constructwas generated in which the putative start codon in the 5' UTRwas mutated to eliminate the uORF (Fig. 5D, panel i). Theseconstructs, along with a control vector, were stably transfectedinto 3T3 L1 cells, and the pooled populations were treated withpoly(IC) and the glycosylation inhibitor tunicamycin. Both ofthese agents have been previously shown to be powerful stimulatorsof eIF2 ${alpha}$ phosphorylation (5, 31). Phosphorylation of eIF2 ${alpha}$ wasconfirmed by Western blotting with whole-cell lysates harvestedfrom 3T3 L1 cells probed with antibodies directed against phosphorylatedeIF2 ${alpha}$ (Fig. 5D, panel iii). However, we did not observe any preferentialtranslation of the constructs which contained the intact uORFunder conditions in which eIF2 ${alpha}$ was phosphorylated (Fig. 5D,panel ii). These observations indicate that the PKR-inducedtranslational regulation of Fas previously reported likely functionsthrough pathways other than through the phosphorylation of eIF2 ${alpha}$ .

View larger version (57K):
[in this window]
[in a new window]
FIG. 5. (A) Phosphorylation of eIF2 ${alpha}$ does not regulate the transcription of key apoptotic genes. Total RNA was harvested from vector-, S51A-, and S51D-expressing cell lines after 0, 3, or 6 days of induction and quantified by an RNase protection assay with radiolabeled probes derived from the indicated genes. (B) Total RNA was harvested from cells as described for panel A and quantitated by an RNase protection assay with radiolabeled probes derived from the indicated cytoplasmic apoptotic signaling genes. (C) eIF2 ${alpha}$ phosphorylation does not affect the level ofapoptotic proteins. Total-cell lysates harvested from vector-, S51A-, or S51D-expressing inducible cell lines on 0, 3, or 6 days postinduction were resolved by SDS-PAGE and probed with antibodies to the indicated proteins. (D) Translational regulation of Fas requires elements in addition to the 5' UTR. Pooled populations of normal 3T3 L1 cells stably expressing luciferase, luciferase fused in frame with the WT Fas 5' UTR, or luciferase fused to a mutant Fas 5' UTR (i) were transfected with either 10 µg of poly(IC) or 10 µg of tunicamycin/ml to induce phosphorylation of eIF2 ${alpha}$ . Relative luciferase counts were determined after 3 h in the presence or absence of treatment (ii). Additionally, whole-cell lysates derived from normal 3T3 L1 cells or cells treated with tunicamycin or poly(IC) were resolved by SDS-PAGE and probed with antibodies recognizing total or phosphorylated eIF2 ${alpha}$ (iii). GAPDH, glyceraldehyde-3-phosphate dehydrogenase; P.I., postinduction; SV40, simian virus 40.

Inhibition of viral replication by eIF2 ${alpha}$ . Earlier experiments performed in our laboratory have shown thenecessity for functional PKR in the innate immune response tovesicular stomatitis virus (VSV) (6). However, the downstreamsubstrates of PKR that potentiate the observed antiviral immunityin mammalian cells have never been definitively shown, and whileeIF2 ${alpha}$ remains the most widely studied substrate for this kinase,other proteins have been shown to be in vitro and in vivo substratesfor PKR (45, 64). To rigorously demonstrate the sufficiencyof phosphorylated eIF2 ${alpha}$ in the inhibition of viral replicationin mammalian tissue, we have utilized a version of the VSV thathas been engineered to express the green fluorescent protein(VSV-GFP) (24). Uninduced and 6-day-induced 3T3 L1 cell lineswere infected with VSV-GFP at a multiplicity of infection (MOI)of 0.1 and were monitored during the course of viral replicationover the following 24 h. By observing GFP fluorescence, we wereable to examine the relative rates of viral protein synthesis.These experiments indicated that eIF2 ${alpha}$ -S51D-expressing cellsshowed a profoundly diminished production of viral proteinsover the 24-h time course (Fig. 6). Similar results were observedat a higher MOI (data not shown). In contrast, neither the vector-expressingcontrol cells nor the eIF2 ${alpha}$ -S51A-expressing cells showed a DOX-dependentalteration of viral replication. To definitively assay viralyield and to control for possible GFP-specific effects of eIF2 ${alpha}$ -S51Dexpression, we determined VSV titers at the 12- and 24-h timepoints after infection. These analyses indicated that expressionof the eIF2 ${alpha}$ -S51D variant resulted in a dramatic 100-fold reductionin VSV-GFP titer at all times measured. In agreement with ourGFP fluorescence observations, no reduction of viral titer wasobserved in the vector- or eIF2 ${alpha}$ -S51A-expressing cells (Fig.6). The failure of the S51A construct to enhance viral replicationmay be due to the fact that the S51A is not a true dominant-negativemutation. While eIF2 ${alpha}$ -S51A is phosphorylation incompetent, eIF2 ${alpha}$ bearing the S51A substitution nonetheless requires eIF2B-mediatedGDP-GTP exchange in order to function in protein synthesis.In the context of a viral infection of cells expressing theeIF2 ${alpha}$ -S51A mutation, activated PKR may still phosphorylate enoughof the endogenous eIF2 ${alpha}$ to sequester eIF2B and inhibit translationof viral transcripts.

View larger version (40K):
[in this window]
[in a new window]
FIG. 6. Phosphorylation of eIF2 ${alpha}$ is sufficient to mediate antiviral immunity. Triplicate samples of uninduced or induced vector-, S51A-, or S51D-expressing cell lines were infected with VSV-GFP at an MOI of 0.1. Viral GFP intensity was measured at 0-, 12-, and 24-h time points by fluorescence microscopy. Additionally, cell culture supernatants were harvested at 0, 12, and 24 h, and viral replication was determined by plaque titration with BHK cells.

eIF2 ${alpha}$ as a regulator of malignant transformation. A number of proteins such as eIF4G and eIF4F and eEF1 have beenshown to be capable of influencing the transformation of mammaliancells (1, 47, 49). Overexpression of the eIF2 ${alpha}$ -S51A variant inparticular has been shown to transform NIH 3T3 cells (23). However,although we observed distinct morphological changes in our 3T3L1 cells upon induction of the eIF2 ${alpha}$ -S51A variant in all clonestested, we did not observe evidence of cellular transformation.To complement these studies, we subsequently retrovirally transducedthe eIF2 ${alpha}$ -S51A into normal 3T3 L1 cells to produce cell linesconstitutively expressing this variant. Following this approach,we were able to obtain a large number of colonies after selectionwith neomycin. Most colonies expressing recombinant eIF2 ${alpha}$ -S51Aexhibited a reduced cell size, but none became transformed,fully consistent with results from the inducible cell system(Fig. 7A). As 3T3 L1 cells retain expression of the INK4 tumorsuppressor locus, unlike NIH 3T3 cells, this product may befunctioning to inhibit cellular transformation (66). Thus, theeIF2 ${alpha}$ -S51A variant was transduced by retrovirus into the NIH3T3 background in an attempt to recapitulate previously reportedphenotypes (23). Expression of the transduced constructs inNIH 3T3 and 3T3 L1 cell lines was confirmed by Western blotting(Fig. 7B). Interestingly, colonies transfected with eIF2 ${alpha}$ -S51Awere readily obtained after 7 days of selection in neomycinthat displayed all of the well-known hallmarks of transformation.Such NIH 3T3 cells expressing S51A had a marked reduction incell size and a dramatic reduction in doubling time (Fig. 7Cand data not shown). In addition, these cell lines exhibitedreduced adherence to tissue culture dishes and were able tosustain anchorage-independent growth (Fig. 7C). Thus, 3T3 L1cells do not undergo cellular transformation in response tooverexpression of eIF2 ${alpha}$ -S51A, while NIH 3T3 cells are rapidlytransformed.

View larger version (93K):
[in this window]
[in a new window]
FIG. 7. (A) eIF2 ${alpha}$ -S51A can transform NIH 3T3 but not 3T3 L1 cell lines. Normal 3T3 L1 cells were transduced with retrovirus carrying the indicated genes. Stable colonies were selected following selection in G418. Morphology was assessed by phase-contrast microscopy. To measure transformation, two colonies per construct were selected and plated in 0.5% soft agar for 21 days. (B) Stable expression of eIF2 ${alpha}$ variants in 3T3 L1 cell lines was determined by Western blotting with a MAb directed against the FLAG tag. (C) NIH 3T3 cells were retrovirally transduced and assayed for transformation as described for panel A. (D) Stable expression of RasV12 or indicated eIF2 ${alpha}$ variants was confirmed by Western blotting with appropriate MAbs. Cloning efficiency for each cell line was determined following growth in soft agar for 21 days, by dividing the colony number by the total number of cells plated.

As there was a discrepancy in our observations concerning cellulartransformation induced by eIF2 ${alpha}$ -S51A, we performed oncogeniccollaboration experiments in a more defined, human genetic backgroundwith previously described HEK cells that have been stably transfectedwith the large T antigen and hTERT (29). Such HEK cells areimmortal and require a single additional genetic element tobecome fully transformed. Retroviral transduction of WT eIF2 ${alpha}$ into these cells produced no obvious morphological changes (Fig.8A). While transduction of eIF2 ${alpha}$ -S51D produced some coloniesfollowing selection, we were not able to expand these into celllines presumably due to aberrancies in protein synthesis affectingcell growth (data not shown). In contrast, transduction of eIF2 ${alpha}$ -S51Aor RasV12 produced colonies with a reduced cell size and a refractilecytoplasm (Fig. 8A). Expression of RasV12, a mutation that hasbeen previously shown to transform human or mouse tissue, andthe FLAG-eIF2 ${alpha}$ constructs was confirmed by Western blotting (Fig.8B). Transformation of these cells was confirmed by their abilityto sustain anchorage-independent growth, while WT eIF2 ${alpha}$ -expressingcells were unable to grow in soft agar (Fig. 8C). Thus, dysregulationof eIF2 ${alpha}$ activity is able to collaborate in the transformationof primary human cells. The results of the colony-forming assayare quantitated in Fig. 8D.

View larger version (90K):
[in this window]
[in a new window]
FIG. 8. (A) eIF2 ${alpha}$ -S51A can collaborate in the transformation of primary human tissue. HEK cells stably transduced with simian virus 40 large T antigen and hTERT (29) were subsequently transduced with either RasV12 or eIF2 ${alpha}$ -S51A. Following selection in puromycin, two colonies per construct were examined by phase-contrast microscopy to determine morphology. (B) Expression of the indicated constructs in individual HEK clones was confirmed by Western blotting. (C) eIF2 ${alpha}$ -S51A and RasV12 confer anchorage-independent growth on HEK cells. Transformation of HEK clones expressing RasV12 or eIF2 ${alpha}$ -S51A was confirmed by plating in 0.5% soft agar for 21 days. (D) Soft agar cloning efficiency for each HEK cell line was determined by dividing the number of colonies observed by the total number of cells plated.

DISCUSSION

Top

Discussion

Considerable evidence exists to indicate that in mammalian cellshormonal, mitogenic, and stress response signaling cascadescan regulate the process of protein synthesis through posttranslationalmodification of eukaryotic translation initiation factors (eIFs)(34, 39). The eIF2 complex, perhaps the most comprehensivelystudied of the translation initiation factors, is a substratefor a number of highly specific stress-responsive kinases suchas PERK, PKR, heme-regulated inhibitor kinase, and GCN2 (63).Nutrient deprivation, inadequate reticular protein processing,arsenical exposure, oxidative stress, and dsRNA all lead toactivation of one or more of these kinases and to the subsequentinhibition of translation via phosphorylation of eIF2 ${alpha}$ on serine51 (57). In addition to limiting protein synthesis, signalingcascades triggered in response to stress often result in complexcellular phenotypes that may include cell cycle arrest, regulationof gene expression at the transcriptional and translationallevels, and/or induction of apoptosis (10, 13, 32). The preciserole, if any, that translational control through eIF2 ${alpha}$ playsin these responses is unclear. Phosphorylation of eIF2 ${alpha}$ by stress-responsivekinases such as GCN2 has, however, been shown in yeast to besufficient to account for the pronounced inhibition of translationthat accompanies amino acid deprivation (17, 21, 38). Additionally,eIF2 ${alpha}$ has also been demonstrated to be a transcript-specificregulator of protein expression for the transcription factorsGCN4 in yeast and ATF4 in mammals (19, 31).

To examine those processes that are subject to eIF2 ${alpha}$ regulationand to further explore the relationship between eIF2 ${alpha}$ regulationand the stress-responsive kinases, we developed clonal, inducible,3T3 L1 cell lines expressing FLAG-tagged versions of the phosphomimeticeIF2 ${alpha}$ variant (eIF2 ${alpha}$ -S51D) and the phosphorylation-insensitiveeIF2 ${alpha}$ -S51A variant. Each of these variants was successfully overexpressedtwo- to threefold over endogenous levels of native eIF2 ${alpha}$ andwas incorporated into eIF2; though it remains formally possiblethat the FLAG tag is affecting eIF2 ${alpha}$ activity in subtle ways,we have seen no evidence for this. Through this approach, weobserved a reproducible and distinct change in the cellularmorphology of both the eIF2 ${alpha}$ -S51D- and the eIF2 ${alpha}$ -S51A-expressingcell lines. The pathways regulated downstream of eIF2 ${alpha}$ that areresponsible for this cytoskeletal reorganization remain to beelucidated. Nevertheless, the distinct morphologies producedby the S51D and the S51A variants suggest that they may be deliveringqualitatively different signals. For example, regulation ofcellular morphology by components of the translation initiationmachinery has been reported previously (15). In particular,antisense RNA against the mRNA cap binding protein eIF4E inHeLa cells has been shown to result in altered cell shape andgrowth patterns (15). Additionally, it has been demonstratedthat eIF4E is capable of modulating signaling by the ras oncogene,which is itself a potent regulator of the activity of the Rhofamily of small GTPases (58). Signaling through such GTPasesis an extensively characterized mechanism for regulation ofmorphology in response to extracellular signals such as platelet-derivedgrowth factor and insulin (58). Whether eIF2 ${alpha}$ also specificallyinfluences these GTP-dependent pathways regulating morphologyremains to be determined.

The eIF2 ${alpha}$ -S51D-expressing cells also demonstrated a marked reductionin growth rate, suggesting a connection between cell cycle regulationand translation initiation. Previous studies examining the activationof the ER-resident eIF2 ${alpha}$ kinase PERK have revealed that ER stressinduces a transient G₁ cell cycle arrest with kinetics similarto those of eIF2 ${alpha}$ phosphorylation (9, 10). The kinase activityof PERK was shown to be essential for this arrest, stronglyimplying that eIF2 ${alpha}$ phosphorylation plays a role in this process.Elevated levels of phosphorylated eIF2 ${alpha}$ were subsequently shownto inhibit translation of cyclin D1 (36, 60, 66). Inducibleexpression of the eIF2 ${alpha}$ -S51D variant in our system may be exertinga similar effect, albeit at low levels since to date we havebeen unable to demonstrate a significant perturbation of thecell cycle profile in exponentially growing populations of theinducible eIF2 ${alpha}$ -S51D-expressing cell lines (data not shown).In contrast, expression of the eIF2 ${alpha}$ -S51A variant in our 3T3L1 system did result in a very mild increase in growth ratesequivalent to approximately 15%. This is in agreement with dataderived from constitutive overexpression of S51A in the immortalizedmurine fibroblast cell line NIH 3T3, where a 20% increase ingrowth rates and an approximately threefold increase in translationrates were observed (23). Similarly, fibroblasts isolated froma knock-in mouse model carrying two copies of the S51A alleleexhibited translation rates which were elevated a modest 18to 35% (65). Such knock-in mice die within hours after birthdue to hypoglycemia caused by deterioration of pancreatic betaislet cells. Thus, translation regulation by eIF2 ${alpha}$ may directlyor indirectly govern the abundance of a selected set of proteinsinvolved in cell growth and division.

The failure of the eIF2 ${alpha}$ -S51A mutant to cause transformationin our inducible 3T3 L1 system is unlikely to be the resultof insufficient expression levels given that we obtained significantoverexpression of our mutants in our inducible cell lines. Additionally,retroviral transduction of eIF2 ${alpha}$ -S51A into 3T3 L1 cells alsofailed to produce transformed colonies. In contrast, NIH 3T3cells were readily transformed by eIF2 ${alpha}$ -S51A. Presumably, 3T3L1 cells are more refractory to transformation than the NIH3T3 cells are, given that the 3T3 L1 cells retain expressionof the p16 tumor suppressor and exhibit a longer doubling time(9). The capacity of the eIF2 ${alpha}$ -S51A variant to collaborate intransformation with large T antigen and hTERT in HEK cells suggeststhat stimulation of translation initiation may be deliveringmitogenic signals similar to those transmitted by activatedRas (46). To our knowledge, this is the first time that a translationinitiation factor has been shown in a defined genetic systemto collaborate in the transformation of human tissue.

Stress which produces activation of eIF2 ${alpha}$ kinases, such as liposome-mediatedtransfection of synthetic dsRNA [poly(IC)], is known to initiatean apoptotic cascade (5, 30, 32). Transient expression of eIF2 ${alpha}$ -S51Ahas been reported to partially ameliorate dsRNA- and TNF- ${alpha}$ -inducedapoptosis, while, in contrast, expression of the eIF2 ${alpha}$ -S51D varianthas been reported to exert apoptotic activity (27, 68). We were,however, unable to demonstrate a marked induction of apoptosisin response to DOX-dependent induction of eIF2 ${alpha}$ -S51D in our 3T3L1 system (Fig. 3). We did observe the transition of approximately5% of the S51D-expressing population to a late-apoptotic-necroticpopulation. While this is a small fraction of the total cellnumber, it does represent an approximately 50-fold increaseover the same population in the uninduced cells. Thus, it appearsthat in our inducible system the eIF2 ${alpha}$ -S51D mutant causes a modestelevation in background cell death but does not induce globalcell death. Surprisingly, apoptosis in response to well-characterizedstimuli was unaltered by expression of either of the eIF2 ${alpha}$ variants.This is in apparent contradiction to an earlier report in whicheIF2 ${alpha}$ -S51A was shown to partially inhibit TNF- ${alpha}$ - and PKR-mediatedtoxicity in NIH 3T3 cells (27, 68). This discrepancy may bedue to the fact that the NIH 3T3 cells are malignantly transformedby eIF2 ${alpha}$ -S51A and, as a result, may have sustained additionalgenetic lesions that inactivate components of the apoptoticsignaling cascades.

Previous work in our laboratory using 3T3 L1 cell lines induciblyoverexpressing human PKR has also shown that activation of overexpressedPKR by dsRNA transfection results in an apparent 10-fold translationalinduction in the levels of the Fas death receptor protein (5).It is, therefore, plausible that translational regulation ofkey apoptotic genes may occur in response to dsRNA. Such PKR-dependentinduction might function through a mechanism similar to ATF4regulation via untranslated ORFs in the 5' UTR of the ATF4 transcript.However, we were unable to explain the potent regulation ofFas in response to activation of PKR by dsRNA through phosphorylationof eIF2 ${alpha}$ . Additionally, we could not demonstrate a role for the5' uORF found in the human Fas transcript. Finally, no experimentalevidence was found to support the translational up-regulationof any of the other key proapoptotic genes examined. While itis possible that our system may not accurately reflect the levelsof phosphorylated eIF2 ${alpha}$ that occur under normal conditions oftransient physiologic stress, our present data indicate a greaterlikelihood that the dsRNA-dependent regulation of Fas and apoptosisoccurs independently of eIF2 ${alpha}$ through other dsRNA signaling pathways.

Finally, while popular dogma concerning the role of PKR in innateimmunity suggests that it functions primarily by inhibitingtranslation through the phosphorylation of eIF2 ${alpha}$ in the presenceof replicating virus, this has not been shown definitively.In this report, we find that, while expression of the phosphomimeticeIF2 ${alpha}$ -S51D is singularly sufficient to suppress viral replication,phosphorylated eIF2 ${alpha}$ does not play a significant role in inducingprogrammed cell death. Our data collectively indicate that regulationof translation through eIF2 ${alpha}$ is an important component of innateimmunity to viral infection, the dysregulation of which cancontribute to tumorigenesis in human cells.

ACKNOWLEDGMENTS

We gratefully acknowledge Robert Weinberg for the contributionof the immortalized HEK cell lines. We also thank Yoshi Nakanishifor the Fas 5' UTR constructs and Scott Kimball for antibodiesto eIF2 ${alpha}$ and eIF2ß.

FOOTNOTES

* Corresponding author. Mailing address: Room 511, Papanicolaou Bldg., 1550 NW 10th Ave. (M710), University of Miami School of Medicine, Miami, FL 33136. Phone: (305) 243-5914. Fax: (305) 243-5885. E-mail: gbarber{at}med.miami.edu.

REFERENCES

Top