• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of neoplasiaLink to Publisher's site
Neoplasia. Apr 2012; 14(4): 297–310.
PMCID: PMC3349256

The Ezrin Metastatic Phenotype Is Associated with the Initiation of Protein Translation1


We previously associated the cytoskeleton linker protein, Ezrin, with the metastatic phenotype of pediatric sarcomas, including osteosarcoma and rhabdomyosarcoma. These studies have suggested that Ezrin contributes to the survival of cancer cells after their arrival at secondary metastatic locations. To better understand this role in metastasis, we undertook two noncandidate analyses of Ezrin function including a microarray subtraction of high-and low-Ezrin-expressing cells and a proteomic approach to identify proteins that bound the N-terminus of Ezrin in tumor lysates. Functional analyses of these data led to a novel and unifying hypothesis that Ezrin contributes to the efficiency of metastasis through regulation of protein translation. In support of this hypothesis, we found Ezrin to be part of the ribonucleoprotein complex to facilitate the expression of complex messenger RNA in cells and to bind with poly A binding protein 1 (PABP1; PABPC1). The relevance of these findings was supported by our identification of Ezrin and components of the translational machinery in pseudopodia of highly metastatic cells during the process of cell invasion. Finally, two small molecule inhibitors recently shown to inhibit the Ezrin metastatic phenotype disrupted the Ezrin/PABP1 association. Taken together, these results provide a novel mechanistic basis by which Ezrin may contribute to metastasis.


Osteosarcoma (OS) is the most common primary tumor of bone. Despite effective control of the primary tumor and both neoadjuvant and adjuvant chemotherapy, the development of metastases to the lungs is the most common cause of death in OS patients. Furthermore, the long-term outcome for patients who present with metastatic disease is grave. The development of new and effective treatments based on a more thorough understanding of metastasis biology is needed.

Toward this end, we previously identified Ezrin as a protein associated with the metastatic phenotype of two highly metastatic pediatric cancers: rhabdomyosarcoma and OS [1,2]. Since then, Ezrin expression has been linked to clinical outcome, clinical stage, or histologic grade in a number of cancers including mammary carcinoma, pancreatic carcinoma, cutaneous and uveal melanoma, uterine carcinoma, breast carcinoma, and soft tissue sarcoma [3–9]. Ezrin is a member of the ERM (Ezrin-Radixin-Moesin) protein family. Ezrin functions as a linker protein connecting the actin cytoskeleton (Ezrin C-terminus) to integral plasma membrane proteins (Ezrin N-terminus) [10,11]. Ezrin is proposed to exist in a dormant form in which the C-terminal tail binds to and masks the N-terminal FERM domain [12,13]. Therefore, amino-terminal Ezrin interactions are critical in determining not only the repertoire of proteins Ezrin can interact with but also the corresponding cellular functions that may be positively or negatively affected. This linkage to the cell membrane allows the cell to physically engage and potentially sense the tumor microenvironment. This linker function makes ERM proteins essential for many fundamental cellular processes, including the determination of cell shape, polarity and formation of surface structures, cell adhesion, motility, cytokinesis, phagocytosis, and integration of membrane transport with signaling pathways [14–17]. The linkage also leads to efficient signal transduction through membrane-associated proteins and receptors. In our previous studies of Ezrin and OS metastasis, we found that Ezrin was not constitutively phosphorylated but rather was dynamically regulated during metastatic progression [18]. Ezrin was phosphorylated early after cells arrived in the lung and again as they progressed into the lung microenvironment. Indeed, high Ezrin expression provided an advantage to highly metastatic cells during this time, resulting in the retention of greater cells in the lung compared with cells with low Ezrin expression. We hypothesize that Ezrin protects cells against apoptosis resulting from stresses faced by metastatic cells at critical periods during metastatic progression, namely early after cells arrive at secondary sites (Hong et al., unpublished observations). In addition and in support of this hypothesis, we recently identified two small molecule inhibitors that bind the N-terminus of Ezrin and inhibit the Ezrin-dependent metastatic phenotype [19]. However, the mechanism(s) by which Ezrin contributes to metastasis are still not understood.

To better define the role of Ezrin in metastasis, we undertook two noncandidate analyses of Ezrin protein function. First, we conducted complementary DNA (cDNA) microarray subtraction of high-and low-Ezrin-expressing tumor cells to identify differentially regulated genes, pathways, and processes. Microarray analysis revealed 124 genes with greater expression in high-Ezrin-expressing cells. Functional analysis of this transcriptional signature suggested that the process of protein translation, specifically translation initiation, was significantly associated with the Ezrin metastatic phenotype. Second, we used tandem mass spectrometry (MS-MS) analysis to identify proteins in tumor lysates that bound the N-terminus of Ezrin. Affinity chromatography and curation of MS-MS data identified 138 Ezrin-binding proteins. Functional analysis of these Ezrin-interacting proteins also revealed a link to the protein translation machinery. The surprising convergence of these noncandidate analyses of Ezrin function led to the hypothesis that Ezrin may contribute to the metastatic phenotype by modulating the efficiency of protein translation in tumor cells. Whereas there were no quantitative differences in total protein synthesis in high-and low-Ezrin-expressing cells, there was an Ezrin-dependent effect on the ability to translate a messenger RNA (mRNA) containing a structured 5′ untranslated region (UTR). As an explanation for this role, we found Ezrin interacts with the 3′ UTR binding protein, poly A binding protein 1 (PABP1; PABPC1), an association that has not been previously described. Furthermore, two Ezrin-binding small molecule inhibitors (NSC305787 and NSC668394) that decrease the Ezrin-dependent metastatic phenotype also inhibited Ezrin/PABP1 interaction. The relevance of these findings was supported by our identification of Ezrin and components of the translational machinery in cellular extensions of highly metastatic cells during the process of cell invasion. Taken together, our results linking Ezrin to the process of protein synthesis are novel and provide a new perspective for understanding how Ezrin contributes to the process of metastasis.

Materials and Methods

Cell Culture and Tissue Harvest

Characterization and maintenance of high-Ezrin/high-metastatic K7M2-WT, K7M2-neo (vector clone), intermediate-Ezrin/intermediate-metastasis AS13, and low-Ezrin/low-metastatic K12-WT, AS2.13, AS2.15, AS1.52, AS1.46 cells have been previously described [1,20]. Human OS (HOS-MNNG, U2 and 143B) and Ewing sarcoma (TC32) cell lines were maintained in Dulbecco modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum, l-glutamine (2 mM), streptomycin (100 U/mL), and penicillin (100 U/mL; Invitrogen, Carlsbad, CA). Ezrin antisense clones were cultured in media containing G418 (Invitrogen). For affinity chromatography studies, K7M2 cells were grown to 80% confluence and harvested by trypsinization before injection to 4-to 6-week-old BALB/c mice by tail vein injection, as previously described [1]. The resultant pulmonary metastases in mice were dissected from surrounding lung and frozen for later use.

Plasmids and Transient Transfection Assays

The Ezrin N-ERMAD (Ezrin-Radixin-Moesin association domain) construct used for recombinant protein production has been previously described [21]. The Stem-Loop-luciferase reporter construct (pcDNA-SL-LUC) was kindly provided by Nancy Colburn [22]. For luciferase reporter assays, 5 x 105 cells were seeded on six-well plates in complete medium and transfected with 2.5 µg of pcDNA-SL-LUC using 7.5 µl of TransIT-LT1 reagent (Mirus, Madison, WI). Protein was harvested 24 or 48 hours after transfection. Firefly luciferase expression from pcDNA-SL-LUC was measured using the Luciferase Assay System (Promega, Madison, WI) in a Victor3 luminometer (PerkinElmer, Boston, MA).

Gene Silencing with Small Interfering RNA

The Ezrin small interfering RNA (siRNA) sequence was generously provided by Natasha Caplen from the Gene Silencing Section of the National Cancer Institute and manufactured along with a nonsilencing negative control by Qiagen Corporation (Valencia, CA). The Ezrin target sequence was 5′-CAGGACTGATTGAATTACGGA-3′ beginning at nucleotide 2149 of human Ezrin mRNA. The siRNA (50 nM final concentration) was added to cultured cells using siLentFect reagent (Bio-Rad, Hercules, CA). Ezrin suppression was verified using Western blot analysis.

Western Blot and Immunoprecipitation

Cells were lysed in either RIPA buffer (Upstate Biotechnology, Waltham, NY) plus protease and phosphatase inhibitor cocktail (Roche, Basel, Switzerland), 1 x sodium dodecyl sulfate (SDS) or cell lysis buffer (Cell Signaling Technology, Danvers, MA). About 30 to 50 µg of protein was separated on Tris-glycine gels (Invitrogen) and transferred to nitrocellulose or polyvinylidene fluoride. The following primary antibodies were used: eIF4E, ribosomal S6, ERM, phospho-ERM, PABP1, and YB-1 (Cell Signaling); β-actin and Ezrin (Sigma-Aldrich, St Louis, MO); RACK1, actinin, IQGAP1, annexin II, and RAN (BD Biosciences, Sparks, MD); CLIC4 and CLIC1 antibodies kindly provided by Dr Mark Berryman, Ohio University College of Osteopathic Medicine; and AHNAK and Rab14 provided by Dr Jacques Baudier, INSERM, Grenoble, France. Horseradish peroxidase-conjugated secondary antibodies and SuperSignal West Pico chemiluminescence substrate were used for detection (Pierce Biotechnology, Rockford, IL). For immunoprecipitation analysis, 24 hours after plating, cells were treated with 10 µM NSC305787 or NSC668394 for 5 hours in serum-free DMEM. Cells were lysed with buffer containing 50 mM HEPES, pH 7.9, 100 mM NaCl, 4 mM NaPP, 10 mM EDTA, 10 mM NaF, 1% Triton X-100, 2 mM sodium vanadate, 1 mM PMSF, 2 µg/ml aprotonin, 2 µg/ml leupeptin, and 1 µM calyculin A. Lysates were precleared with a 50% slurry of protein G agarose resin and incubated overnight with 2 µl of Ezrin antibody (no. E8897; Sigma-Aldrich). Antigen-antibody complexes were selected with protein G agarose beads, washed three times, and analyzed by SDS-PAGE.

cDNA Microarray Procedure

Total RNA was extracted from eight murine OS cell lines (K7M2-WT, K7M2-neo, AS13, AS2.13, AS2.15, AS1.52, AS1.46, and K12) for cDNA microarray experiments; RNA preparation and microarray experiments were done as previously described [23]. The RNA obtained from K7M2-neo (high-Ezrin transfection control) was used as a reference (Cy3 labeled) in all two-color cDNA microarrray experiments. Data normalization and principal component analysis were performed as previously described [23,24]. The normalized expression ratios of experiments (seven cell lines) versus control (high-Ezrin control K7M2-neo) for this filtered list were obtained for further analyses. Two criteria were used to identify genes of interest. First, we selected the cDNAs with a mean expression ratio of 0.7 to 1.3 in the high-Ezrin group. From this list of cDNAs, only those with a mean expression ratio of greater than 2 or less than 0.5 in the low-Ezrin group were then selected. These filtering steps yielded 181 cDNAs of interest, which were then analyzed by hierarchical clustering using Cluster software version 3.0 (Tokyo University, Tokyo, Japan). The heat map was generated using Java TreeView1.0.8 (http://jtreeview.sourceforge.net). Expression analysis systematic explorer (EASE) analysis was performed as previously described [25]. In addition, BLAST2GO (http://www.blast2go.org) was used to determine overrepresented GO terms attributed to Ezrin-interacting proteins identified by affinity chromatography and MS-MS identification [26]. Quantitative polymerase chain reaction analysis of select differentially expressed genes was performed using total cellular RNA. Primers were custom designed and synthesized by Invitrogen, and sequences are available on request.

Ezrin N-ERMAD Affinity Chromatography

The generation and purification of recombinant Ezrin N-ERMAD protein and purification have been described previously [27]. Metastatic lung nodules were grossly dissected into pieces no larger than 2 x 2 mm, weighed and suspended in cold extraction buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 1 mM EDTA, 0.1% Triton X-100 protease inhibitors) at 5 ml/g of tissue. The tissue was homogenized, sonicated, and centrifuged at 50,000g for 30 minutes, followed by ultracentrifugation at 200,000g for 30 minutes. Ten milliliters of tumor protein extract (5 mg/mL) was incubated with 100 µlof Ezrin N-ERMAD or BSA-coated sepharose beads. Soluble Ezrin N-ERMAD protein was added to selected reactions for competition. Samples were incubated overnight at 4°C, washed eight times with cold TBS/0.1% Triton X-100, and bound proteins eluted by boiling in 2x SDS sample buffer.

Tandem Mass Spectrometry

After affinity chromatography as described, protein bands were excised and subsequently subjected to in-gel tryptic digestion to extract peptides [28]. Each peptide sample was desalted, lyophilized, and resuspended in 0.1% formic acid for liquid chromatography-MS-MS analysis using an Agilent 1100 capillary high-performance liquid chromatography system (Agilent Technologies, Santa Clara, CA) with a 10-cm integrated micro reverse phase liquid chromatography-electrospray ionization emitter column, coupled online with a LCQ Deca XP mass spectrometer (Thermo Fisher Scientific, Fair Lawn, NJ). Peptides were eluted using a linear gradient of 2% mobile phase B (acetonitrile with 0.1% formic acid) to 42% mobile phase B. The ion trap-mass spectrometer was operated in a data-dependent MS-MS mode in which the three most intense peptide molecular ions in the MS scan were sequentially and dynamically selected for subsequent collision-induced dissociation using a normalized collision energy of 35%. The mass spectra were acquired at the mass range of m/z 475 to 2000. The ion-source capillary voltage and temperature were set at 1.7 kV and 180°C, respectively. The MS-MS data were used to search the EBI UniProt Homo sapiens database (http://www.ebi.ac.uk/integr8). Up to two missed cleavage sites were allowed during the database search. The cutoff for legitimate identifications were charge state-dependent cross-correlation (Xcorr) ≥ 2.0 for [M + H]1+, ≥2.5 for [M + 2H]2+, and ≥3.0 for [M + 3H]3+ with delta correlation (ΔCn) ≥ 0.10. Additional protein identification was carried out on a 4800 matrix-assisted laser desorption/ionization-tandem time of flight (MALDI-TOF-TOF) analyzer (Applied Biosystems, Carlsbad, CA) in reflector-positive mode and then validated in MS-MS mode. Select MS-MS-identified Ezrin-interacting proteins were identified by subtracting nonspecific interactors by subtracting peptides found to interact with BSA and then for proteins not represented by two or more peptides in the MS-MS data. Detected peptides were then prioritized for assessment based on actual versus predicted molecular weight (MW) and peptide coverage. Proteins with a MW observed/actual between 0.7 and 1.5 were selected.

m7-GTP Cap Affinity Binding Assay

Cells were grown to ~50% confluence and lysed, and the supernatant was precleared with 50 µl of sepharose CL-4B beads (50/50 slurry in wash buffer 1). Samples were centrifuged and supernatant transferred to a new tube where 50 µl of m7-GTP sepharose beads (GE Lifesciences, Piscataway, NJ) were added. Negative control reactions received sepharose CL-4B beads without m7-GTP coating. Samples were washed twice with wash buffer 1 (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, complete protease, and phosphatase inhibitors) and once with wash buffer 2 (20 mM Tris-HCl, pH 7.5, 300 mM NaCl, complete protease, and phosphatase inhibitors). Bound protein was eluted with 2x SDS sample buffer with heating.

Polysome Analysis by Sucrose Density Gradient Fractionation

Cells were grown in complete DMEM and harvested at ~50% confluence. Immediately before harvesting, cells were incubated with cycloheximide (200 µg/ml) for 10 minutes at 37°C. Samples were then placed on ice and washed twice with ice-cold PBS. Cells were then lysed in polysome lysis buffer (250 mM KCl, 10 mM Tris-HCl, pH 7.4, 25 mM MgCl2, 0.5% NP-40, 0.5% sodium deoxycholate, 1 µg/µl RNasin, 200 µg/ml cycloheximide, EDTA-free protease inhibitors [Roche, Indianapolis, IN], phosphatase inhibitors, and nuclease-free water) and incubated on ice for 15 minutes followed by centrifugation at 4°C for 10 minutes at 12,000g. EDTA (25 mM final concentration) was added to the lysis buffer in place of Mg2+ for control samples to disrupt ribosomes. Supernatant was layered on 15% to 45% (wt/vol) sucrose gradients and centrifuged at 180,000g for 2.5 hours at 4°C. An ISCO Model 640 fractionator with a UV-6 absorbance monitor, 10-mm path-length flow cell, was used to measure the optical density of the gradients at 254 nm in real time. A Dataq model DU-158 analog-to-digital converter was used for data acquisition using Windaq software (Dataq Instruments, Akron, OH).

Enzyme-Linked Immunosorbent Assay

ELISA experiments were performed by coating a 96-well high-protein-binding ELISA plate (M0661; Sigma) with 200 ng of recombinant Ezrin protein. Nonspecific binding sites were blocked with 4% bovine serum albumin for 2 hours at room temperature. Five hundred micrograms of total protein from K12-WT cell lysate was added and incubated for 2 hours at room temperature. PABP1 bound to Ezrin was detected using a PABP1 antibody (no. ab6125; Abcam, Cambridge, MA) and a horseradish peroxidase-conjugated antimouse antibody (Sigma-Aldrich). Finally, 2,2′-azino-di-3-ethyl-benzthiazline sulfonate (Sigma-Aldrich) substrate was added, and the optical density at 405 nm was measured after 15 minutes.

Protein Isolation from Pseudopodia and Cell Bodies

Protein was isolated from migrating pseudopodia (“feet”) and cell bodies using the Pseudopodia Purification Kit (Chemicon International, Temecula, CA) according to manufacturer's instructions.


Microarray Analysis Defines an Ezrin-Dependent Gene Expression Signature

To define the Ezrin-associated metastatic phenotype at the level of transcription, cDNA microarray subtraction of high-and low-Ezrin-expressing OS cell lines was carried out. Three cell line clones with high Ezrin expression, three with low Ezrin expression and one with intermediate Ezrin expression were hybridized with the high-Ezrin transfection control, K7M2-neo, in two-color cDNA microarray experiments. One hundred eighty-one (181) genes were differentially expressed (vs the high-Ezrin control) with 57 downregulated and 124 upregulated (Figure 1). As predicted, AS13, the cell line with intermediate Ezrin levels showed intermediate expression (between high-and low-Ezrin cell lines) of the 181 genes of interest. Characterization of differentially expressed genes, based on non-mutually exclusive gene ontology (GO) function assignment, was conducted using EASE analysis. It revealed several specific gene functions significantly overrepresented in the 181 genes of interest compared with the total gene set of non-differentially expressed transcripts. Functions such as development, microtubule cytoskeleton, cytoskeleton, and microtubule-associated complex were expected based on existing knowledge of Ezrin biology (Table 1). However, other functions were unexpectedly found to correlate with Ezrin expression. For example, an area of biologic interest not previously associated with Ezrin activity but highly overrepresented in our analysis was protein translation. Translation, translation initiation, and translation initiation factor activity were overrepresented GO functions that were associated with the following four genes: eIF4e, Ormdl2, eIF3S4, and eIF1AY. The increase in eIF4e expression in high-Ezrin cell lines, K7M2-wt and K7M2-neo, has been confirmed by Western blot (C. Khanna, unpublished observations, personal communication). This functional approach to the analysis of our microarray data suggested the hypothesis that Ezrin expression was associated with altered regulation of the translation machinery in these OS cells.

Figure 1
Hierarchical clustering of 181 differentially expressed cDNAs define the Ezrin phenotype. A 21k cDNA microarray subtraction of high (K7M2 WT, and AS 2.13), intermediate (AS13), and low (AS 1.52, AS 1.46, and K12 WT) compared with a high-Ezrin control ...
Table 1
Functions That Define the Ezrin Microarray Transcription Phenotype Ranked by EASE Score.

Ezrin Affinity Chromatography Independently Identifies Proteins Involved in Translation

In parallel to the microarray studies described, a noncandidate proteomic-based evaluation of Ezrin included the identification of Ezrin-binding proteins, using affinity chromatography followed by MS-MS. The amino (N) terminus of Ezrin has previously been shown to be involved in numerous protein-protein interactions while the carboxy (C) terminus primarily binds actin. In addition, the C-terminus of activated Ezrin primarily binds f-actin, whereas the diversity of Ezrin N-terminal interactions is far greater. Therefore, we chose to use the N-terminus for affinity chromatography experiments to “capture” Ezrin-interacting proteins from OS tumor lysates. The proteins identified using MS-MS were prioritized to eliminate those proteins that nonspecifically bound to BSA and were not identified by multiple unique peptides in the MS-MS data (Figure W1). This filtering algorithm yielded 138 proteins that bound Ezrin with strong coverage in OS tumor lysates (Table 2). Western blot analysis of selected proteins eluted from N-Ezrin sepharose beads validated the MS-MS identification of putative Ezrin-interacting proteins (Figure W2). Disruption of the actin cytoskeleton in the harvested protein lysates using latrunculin B demonstrated the requirement of actin for the observed interactions between most candidate proteins and Ezrin (suggestive of indirect protein interactions with Ezrin). RACK1, the protein translation-associated scaffolding protein, and, to a lesser extent, Rab-14, the large GTPase trafficking protein, demonstrated persistent strong elution from N-Ezrin beads after actin disruption. A direct interaction between RACK1 and Ezrin could not be demonstrated by coimmunoprecipitation with available antibodies. Functional classification of the 138 direct and indirect Ezrin-binding proteins included GO terms for membrane trafficking and cytoskeleton that were expected based on known functions of Ezrin (Table 3). In addition, proteins associated with translation and translation initiation were also enriched, including YB-1 (Y-box binding protein 1), RACK1 (receptor for activated C-protein kinase), and several eukaryotic translation initiation and elongation factors. To the best of our knowledge, Ezrin has not previously been shown to interact directly or indirectly with these proteins or to be associated with the process of protein synthesis. Taken together, the results from both noncandidate genomic and proteomic studies suggested a novel link between Ezrin and the process of protein translation.

Table 2
List of Ezrin N-ERMAD Interacting Proteins Identified by Affinity Chromatography and Tandem Mass Spectrometry.
Table 3
Functional Assessment of Ezrin N-terminal Binding Proteins Identified by Affinity Chromatography and MS-MS.

Functional Analysis of the Translation Machinery in OS Cells

Next, to determine the extent to which Ezrin expression correlated with changes in global translation, we used sucrose density gradient centrifugation to analyze and quantify polysome profiles from high-and low-Ezrin-expressing cells. This approach allowed for quantification of total protein and mRNA associated with free messenger ribonucleoprotein complexes (mRNPs), individual ribosomal subunits (40S and 60S), monosomes (80S), and actively translating polysomes (mRNA containing more than two 80S particles). A global increase in translation initiation would be expected to result in an increase in the A254 signal intensity for the heavy polysome fractions (areas of active protein translation). Quantification of the area under the peak for fractions 8 to 20 (containing more than two ribosomes) revealed no difference between the high-and low-Ezrin-expressing cell lines, suggesting similar rates of global translation (Figure 2A). Cells treated with thapsigargin, a well-characterized inhibitor of translation initiation, demonstrated an expected shift in polysome profiles in both cell lines. Given the similarities in global translational rates, we hypothesized that differences may exist in the efficiency of translation of specific mRNA, namely those containing complex 5′UTR (Figure 2B). mRNA with complex 5′UTR have been shown to be “weakly” translated compared with the majority of other mRNA species. Numerous proto-oncogenes and antiapoptotic genes possess 5′UTRs that may form complex secondary structures. The rate-limiting step in synthesizing peptides from these messages seems to be during assembly of the preinitiation complex or initiation [29–32]. To model the expression of such complex 5′UTRs, we expressed a luciferase reporter with a hairpin/loop structure (δG = -44.8 kcal/mol) placed upstream of a luciferase protein coding sequence in cells with high and low Ezrin (Figure 2B). This stem-loop sequence was previously shown to dramatically inhibit protein synthesis of the downstream cistron [22]. Cells with more efficient translational machinery would be expected to better translate the stem-loop reporter mRNA resulting in increased luciferase activity. Indeed, cells with high Ezrin were able to express this complex synthetic (weakly translated) reporter more efficiently than cells in which Ezrin was suppressed. A similar correlation in stem-loop luciferase activity was seen after siRNA-mediated knockdown of Ezrin expression in human OS (HOS-MNNG, U2OS, 143B), rhabdomyosarcoma (Rh30), and Ewing sarcoma (TC32) cell lines (data not shown). Taken together, these results suggest that Ezrin expression specifically correlates with an increase in the ability to express an mRNA with a complex 5′UTR, without changes to overall protein synthesis.

Figure 2
Ezrin expression does not influence global protein translation but contributes to the expression of complex mRNAs. (A) Polysome analysis of high-(K7M2) and low-(AS 1.46) Ezrin-expressing cells was undertaken to assess active protein translation in cells. ...

To provide an explanation for how Ezrin contributes to the initiation of protein translation, we analyzed OS cell lysates, again by sucrose density gradient centrifugation, to determine the extent to which Ezrin protein was present in fractions containing complete, intact ribosomes (80S monomers) as well as actively translating polysomes (Figure 3A). Western blot analysis of Ezrin and previously identified proteins linked to translation was conducted on protein isolated from each fraction. As shown in Figure 3A, PABP1 and RACK1 were found in fractions containing ribosomal subunits, monosomes, or polysomes and undetectable in the free mRNP fractions. This is consistent with previous reports of RACK1 as a core ribosome-binding protein and PABP1's role in binding the poly A tail of mRNA as part of the protein translation complex [33,34]. eIF4E was detected mainly in fractions containing free mRNPs and 40S–80S complexes consistent with its role in translation before initiation. YB-1 was enriched in polysome fractions consistent with its role as an mRNA binding protein involved in regulating translation [35]. Ezrin was detected primarily in free mRNP fractions as well as fractions corresponding to 40S, 60S and 80S particles (Figure 3A). However, Ezrin was not detected in high-MW polysomes. Using an antibody that detects the ERM (Ezrin, Radixin, Moesin) family members, including Ezrin, a similar pattern of protein detection was seen (data not shown). These results suggest that several functional pools of Ezrin may be present and that Ezrin detected in fractions containing eIF4E, RACK1, PABP1, and/or YB-1 may represent a functionally important multiprotein complex associated with translation.

Figure 3Figure 3
Ezrin cosediments with messenger ribonucleoprotein (mRNP) complexes, ribosomal subunits and also interacts with PABP1. (A) Ezrin is part of the ribonucleoprotein complex. Sucrose density gradient centrifugation was used to isolate and resolve messenger ...

To demonstrate that Ezrin cosedimentation with translation components was not the result of interaction with nonrelated complexes of a similar density, we treated lysates with EDTA before loading on to sucrose density gradients. EDTA treatment chelates magnesium, a necessary cofactor for ribosomal subunit integrity. As expected, EDTA treatment caused a dramatic decrease in polysomes while promoting a concomitant increase in the levels of dissociated ribosomal subunits (Figure 3B). In addition, Ezrin protein levels, along with RACK1, decreased in 40S fractions. This suggested that cosedimentation of Ezrin in 40S fractions of non-EDTA-treated samples (Figure 3A) was dependent on intact ribosomes, indicating that Ezrin may be a part of the translation preinitiation complex. To investigate this possibility, we used an approach to enrich for functional, cap-bound translation initiation factors in K7M2 OS cell lysates. Alterations in protein translation are often mediated by the expression and availability of eukaryotic translation initiation factors, such as eIF4E, or as a result of alterations in the regulation of signaling pathways. eIF4E binds to the 7-methylguanosine “cap” present at the 5′ end of all eukaryotic mRNAs and is central to formation of the translation preinitiation complex. As shown in Figure 3C, eIF4E, was efficiently captured from tumor cell lysates as were eIF4G and 4E-BP1. However, we were unable to detect the binding of Ezrin, or other ERM family members using this approach. These results show that, although Ezrin cosediments with factors involved in translation initiation, there is no direct interaction of Ezrin and cap-bound proteins. We next asked if Ezrin might interact with proteins that may stabilize the translational loop structure at the 3′ UTR of complex proteins. Indeed, we found a direct interaction between Ezrin and the poly A binding protein (PABP1; Figure 3, D and E). In support of the interaction between PABP1 and Ezrin, the use of small molecules that directly bind to Ezrin and inhibit metastasis in vivo also blocked the interaction between PABP1 and Ezrin (Figure 3F) [19]. Collectively, these findings provide new insights into the association of Ezrin with the ribonucleoprotein complex and may suggest a role for Ezrin in the stabilization of the translational machinery during its interaction with mRNA early during the initiation or regulation of translation. Our data do not support a role for Ezrin in active and ongoing mRNA translation.

The Cellular Relevance of Enhanced Protein Translation in Metastatic Cells

The active, phosphorylated form of Ezrin is enriched at the cell membrane, and it regulates signaling dynamics associated with actin reorganization during cell motility [18]. To determine whether components of the translation machinery colocalized with active Ezrin/ERM family proteins in this context, we purified pseudopodia from migrating OS cells. Cells were plated on collagen/laminin-coated membranes containing 0.2-µm pores. This selective pore size allows invasion of pseudopod extensions through the pores in response to a chemotactic gradient (e.g., high serum) while excluding cell bodies. Western blot analysis was then undertaken for Ezrin and other proteins involved in protein translation from invading pseudopodia (extracted from the lower part of the membrane) or cell bodies (extracted from the upper part of the membrane) of metastatic human and murine OS cells (Figure 4). Samples were normalized so that equal amounts of protein were loaded for each condition. As expected, phosphorylated ERM family proteins were enriched in migrating pseudopodia. The relative levels of total ERM (unphosphorylated) protein were similar to the cell bodies. Interestingly ribosomal S6, the 40S core protein, RACK1, and eIF4E (proteins needed for the initiation of translation) were all found in these invading pseudopodia. Levels of β-actin were similar in both fractions. These data support the hypothesis that, during metastatic progression, Ezrin may provide efficiency to metastatic cells by allowing the translation of needed proteins at distinct points in time and in distinct subcellular localizations (i.e., invadapodia). A corollary of this hypothesis is that the inability of nonmetastatic cells to deliver these needed proteins prevents them from successfully metastasizing.

Figure 4
Active Ezrin and elements of the translational machinery are present in invading pseudopodia. Migrating highly metastatic human and murine OS cells (143B and K7M2, respectively) were plated on collagen/laminin-coated membranes containing 0.2-µ ...


We previously identified the cytoskeleton linker protein, Ezrin, as part of our effort to expand our understanding of the biology of metastasis in pediatric sarcomas [1,20]. Since then, several groups have identified associations between Ezrin expression and clinical outcome in a variety of human cancers [9,36–40]. We previously hypothesized that Ezrin enhances the ability of metastatic cells to endure specific stresses related to metastatic progression [1]. The inefficiency of metastasis is believed to be primarily influenced by an inability of the majority of cancer cells to manage stresses that are faced after cells arrive at distant secondary sites. High-Ezrin-expressing cells are better able to resist this inefficiency early during progression at secondary sites [41]. From these data, we hypothesized that Ezrin expression allowed cells to rapidly adapt to the foreign lung microenvironment and manage cellular stresses. However, it has been unclear how Ezrin provided this advantage during the process of metastasis. To address this question, we used two noncandidate approaches to identify biologic processes affected by Ezrin in OS cells. First, a genomic comparison of high-and low-Ezrin-expressing OS cells resulted in a transcriptional signature of 181 genes that correlated with high Ezrin expression. Functional analysis of these cDNAs revealed unexpected enrichment of gene functions related to protein biosynthesis. In parallel, we used affinity chromatography in which the N-terminus of Ezrin was used to capture putative Ezrin-interacting proteins present in tumor extracts. A total of 138 bound proteins were identified by tandem mass spectrometry and surprisingly, we again found enrichment of proteins associated with protein translation and the translational machinery. These included known RNA-binding proteins, translation initiation factors, the core 40S ribosomal subunit protein, RACK1, and the 3′ mRNA binding protein, PABP1. Previous studies have demonstrated that cytoskeletal dynamics and translation may be coregulated [42–46]. Furthermore, to our knowledge, our study is the first to associate Ezrin with protein translation and with elevated expression of components of the translation machinery such as eIF4E.

Because eIF4E binding to the 5′ cap of cellular mRNA is considered a major rate-limiting step in protein synthesis, increased expression in the presence of Ezrin may provide a means for enhanced anabolic processes such as tumor cell growth. Increased expression of eIF4E has been shown to enhance translation of mRNA with a complex, highly structured 5′ untranslated region [29]. Consistent with this idea, expression of a weakly translated stem-loop reporter was increased in high-Ezrin-expressing cells. We considered the possibility that enhanced translation may be a general phenomenon affecting global rates of protein synthesis. However, quantitative polysome analysis of steady-state translation levels failed to reveal differences between high-and low-Ezrin-expressing cells. In addition, polysome profiles from physiologically stressed cells were also similar regardless of Ezrin levels. Taken together, these data suggested that Ezrin might have an effect on the translation of specific mRNAs with complex 5′ untranslated regions rather than on global protein synthesis. This may not be surprising because both high-and low-Ezrin-expressing OS cells are viable, proliferate in vitro and yield similar primary tumor growth features in vivo. The distinction among these cells is the inability of cells with low Ezrin to metastasize [1]. A model of selective protein translation describes proteins as either “strongly translated” or “weakly translated” [47]. The strongly translated proteins undergo very limited regulation of translation, whereas the “weakly translated proteins” are largely maintained as stable mRNAs and await cues that activate the translational machinery to translate these complex proteins [48]. Common among such “weakly translated proteins” is complexity in the 5′ and 3′ untranslated regions. Because it is energetically impossible to express all proteins at all times, this proposed model suggests that cells in dynamic environments activate the translational machinery in times of need; as such, these “weakly translated proteins” undergo translation only when they are most needed. Examples of “weakly translated proteins” include proteins involved in acute phase responses and several oncogenes [49].

Extending this hypothesis, we next asked whether Ezrin and components of the translational machinery were present in subcellular locations relevant to metastasis such as invading pseudopodia. Indeed, we demonstrated the presence of phosphorylated forms of Ezrin, and critical parts of the translational machinery (i.e., RACK1, eIF4E) are present in these invading pseudopodia. These studies support the notion that enhanced efficiency of metastasis may not be limited to a kinetic advantage alone but may also involve the opportunity to deliver proteins in the areas of greatest need during metastasis. It is attractive to consider the dynamic needs of highly metastatic cells and an efficiency in metastasis being conferred on cells that are able to meet these needs through the delivery of necessary proteins at specific times and locations within the cell [29]. Previous studies have shown that actin dynamics can alter translation by directly affecting both transport of specific mRNA within the cell as well as signaling to and from the protein synthesis machinery [45]. For example, during development, axons become polarized, migrating in response to chemotactic gradients. During this migration, transmembrane receptors in the growth cone become activated stimulating formation of actin-rich filopodia and lamellipodia. In this context, the cytoskeleton serves as a scaffold for RNA binding proteins, translation factors, and signaling intermediates, enabling the spatiotemporal regulation of protein synthesis [50]. In light of our current data and given the classic role of Ezrin as an actin-binding protein, one can envision that a similar scenario may occur in migrating and invading tumor cells during metastatic progression.

In further support of the link between Ezrin and protein translation, sucrose density gradient fractionation of whole cell lysates revealed that a portion of total Ezrin protein was present in fractions containing the RNA binding proteins YB-1 and PABP1 as well as the cap binding protein eIF4E and RACK1. These results provided a functional context in which the majority of Ezrin was present in free mRNP fractions, whereas a smaller portion was present in “lighter” ribosomal units. The absence of Ezrin in more heavily weighted polysomes suggested a role for Ezrin in the early phases of protein synthesis such as mRNA transport, formation of the preinitiation complex, or initiation rather than elongation or termination. These possibilities were further narrowed down after we failed to detect the presence of Ezrin using 7-methyl guanosine cap affinity chromatography to functionally enrich for components of the cap-binding complex. In addition, eIF4A, eIF4G, and eIF4E were all absent from Ezrin immunoprecipitates (data not shown). However, one protein, identified by MS-MS, that did coimmunoprecipitate with Ezrin was poly-A binding protein (PABP1). Similar results have been seen for binding of paxillin to PABP1 [50]. In addition, paxillin/PABP1 complexes demonstrate nucleocytoplasmic shuttling in a CRM1-dependent manner [51]. There have been some reports of Ezrin localizing to the nucleus, and it is interesting to speculate that PABP1 may play a role in nucleocytoplasmic shuttling of Ezrin as well [52,53]. Although we cannot conclude from our current studies whether Ezrin/PABP1 interaction is responsible for Ezrin's role in translation and metastasis, we have identified small molecule inhibitors, which block their interaction and, importantly, also inhibit metastatic progression in vivo [19]. Having established this phenotypic relevance, studies are under way to characterize the nature of Ezrin/PABP1 interaction including the dynamics and subcellular localization. In addition, future studies will be aimed at identifying those endogenous mRNA, which may coassociate with Ezrin/PABP1, specifically those with complex 5′UTR as candidates for enhanced translation and which contribute to metastasis.

Collectively, these data provide support for a broader complex of proteins that may collectively contribute to efficient protein translation. In our current report, we found that Ezrin interacts with RACK1, by Ezrin affinity chromatography and cofractionation with a portion of endogenous Ezrin in sucrose density gradient analysis. A wealth of electron microscopy and crystal structure data have established the location of RACK1 binding to the 40S ribosomal subunit and functional analysis has revealed that RACK1 serves as a docking site and scaffold for multiple proteins, including PABP1 and protein kinase C (PKC), which regulate translation [33,54–56]. Interestingly, we have previously shown that PKC isoforms coimmunoprecipitate with Ezrin and mediate C-terminal Ezrin phosphorylation, believed to regulate Ezrin's activation in OS cells [18]. We hypothesize that a complex involving Ezrin, PABP1, and PKC with close interaction with RACK1 would be well positioned to efficiently regulate the translational machinery in response to the stresses of metastasis.

Based on a confluence of functional data from genomic and proteomic assessment of Ezrin function, we generated a hypothesis that Ezrin contributes to the efficiency of metastasis by enhancing the process of protein translation. In support of this hypothesis, we demonstrated that Ezrin is part of the ribonucleoprotein complex, interacts with PABP1, and may enhance the translation of specific proteins at specific subcellular locations, critical to the metastatic phenotype. A potential clinical impact of these findings is a biologic rationale for the use of inhibitors of translation initiation, such as rapamycin and related synthetic analogs, in the context of metastatic progression rather than solely for established disease. Collectively, these data support new functions for Ezrin in the metastatic phenotype of cancer and serve as a foundation for studies that will elucidate the mechanisms by which this occurs.

Supplementary Material

Supplementary Figures and Tables:


4.1 Ezrin-Radixin-Moesin
tandem mass spectrometry
amino-terminal Ezrin, Radixin, Moesin association domain
untranslated region


1This article refers to supplementary materials, which are designated by Figures W1 and W2 and are available online at www.neoplasia.com.


1. Khanna C, Wan X, Bose S, Cassaday R, Olomu O, Mendoza A, Yeung C, Gorlick R, Hewitt SM, Helman LJ. The membrane-cytoskeleton linker ezrin is necessary for osteosarcoma metastasis. Nat Med. 2004;10:182–186. [PubMed]
2. Yu Y, Khan J, Khanna C, Helman L, Meltzer PS, Merlino G. Expression profiling identifies the cytoskeletal organizer ezrin and the developmental homeoprotein Six-1 as key metastatic regulators. Nat Med. 2004;10:175–181. [PubMed]
3. Ilmonen S, Vaheri A, Asko-Seljavaara S, Carpen O. Ezrin in primary cutaneous melanoma. Mod Pathol. 2005;18:503–510. [PubMed]
4. Ohtani K, Sakamoto H, Rutherford T, Chen Z, Satoh K, Naftolin F. Ezrin, a membrane-cytoskeletal linking protein, is involved in the process of invasion of endometrial cancer cells. Cancer Lett. 1999;147:31–38. [PubMed]
5. Pang ST, Fang X, Valdman A, Norstedt G, Pousette A, Egevad L, Ekman P. Expression of ezrin in prostatic intraepithelial neoplasia. Urology. 2004;63:609–612. [PubMed]
6. Tokunou M, Niki T, Saitoh Y, Imamura H, Sakamoto M, Hirohashi S. Altered expression of the ERM proteins in lung adenocarcinoma. Lab Invest. 2000;80:1643–1650. [PubMed]
7. Tynninen O, Carpen O, Jaaskelainen J, Paavonen T, Paetau A. Ezrin expression in tissue microarray of primary and recurrent gliomas. Neuropathol Appl Neurobiol. 2004;30:472–477. [PubMed]
8. Weng WH, Ahlen J, Astrom K, Lui WO, Larsson C. Prognostic impact of immunohistochemical expression of ezrin in highly malignant soft tissue sarcomas. Clin Cancer Res. 2005;11:6198–6204. [PubMed]
9. Bruce B, Khanna G, Ren L, Landberg G, Jirstrom K, Powell C, Borczuk A, Keller ET, Wojno KJ, Meltzer P, et al. Expression of the cytoskeleton linker protein ezrin in human cancers. Clin Exp Metastasis. 2007;24:69–78. [PubMed]
10. Reczek D, Berryman M, Bretscher A. Identification of EBP50: a PDZ-containing phosphoprotein that associates with members of the ezrin-radixinmoesin family. J Cell Biol. 1997;139:169–179. [PMC free article] [PubMed]
10. Tsukita S, Oishi K, Sato N, Sagara J, Kawai A. ERM family members as molecular linkers between the cell surface glycoprotein CD44 and actinbased cytoskeletons. J Cell Biol. 1994;126:391–401. [PMC free article] [PubMed]
12. Berryman M, Franck Z, Bretscher A. Ezrin is concentrated in the apical microvilli of a wide variety of epithelial cells whereas moesin is found primarily in endothelial cells. J Cell Sci. 1993;105(pt 4):1025–1043. [PubMed]
13. Bretscher A, Chambers D, Nguyen R, Reczek D. ERM-Merlin and EBP50 protein families in plasma membrane organization and function. Annu Rev Cell Dev Biol. 2000;16:113–143. [PubMed]
14. Wu KL, Khan S, Lakhe-Reddy S, Jarad G, Mukherjee A, Obejero-Paz CA, Konieczkowski M, Sedor JR, Schelling JR. The NHE1 Na+/H+ exchanger recruits ezrin/radixin/moesin proteins to regulate Akt-dependent cell survival. J Biol Chem. 2004;279:26280–26286. [PubMed]
15. Bretscher A, Edwards K, Fehon RG. ERM proteins and merlin: integrators at the cell cortex. Nat Rev Mol Cell Biol. 2002;3:586–599. [PubMed]
16. Ng T, Parsons M, Hughes WE, Monypenny J, Zicha D, Gautreau A, Arpin M, Gschmeissner S, Verveer PJ, Bastiaens PI, et al. Ezrin is a downstream effector of trafficking PKC-integrin complexes involved in the control of cell motility. EMBO J. 2001;20:2723–2741. [PMC free article] [PubMed]
17. Serrador JM, Nieto M, Sanchez-Madrid F. Cytoskeletal rearrangement during migration and activation of T lymphocytes. Trends Cell Biol. 1999;9:228–233. [PubMed]
18. Ren L, Hong SH, Cassavaugh J, Osborne T, Chou AJ, Kim SY, Gorlick R, Hewitt SM, Khanna C. The actin-cytoskeleton linker protein ezrin is regulated during osteosarcoma metastasis by PKC. Oncogene. 2009;28:792–802. [PubMed]
19. Bulut G, Hong SH, Chen K, Beauchamp EM, Rahim S, Kosturko GW, Glasgow E, Dakshanamurthy S, Lee HS, Daar I, et al. Small molecule inhibitors of ezrin inhibit the invasive phenotype of osteosarcoma cells. Oncogene. 2012;31:269–281. [PMC free article] [PubMed]
20. Khanna C, Khan J, Nguyen P, Prehn J, Caylor J, Yeung C, Trepel J, Meltzer P, Helman L. Metastasis-associated differences in gene expression in a murine model of osteosarcoma. Cancer Res. 2001;61:3750–3759. [PubMed]
21. Bretscher A, Reczek D, Berryman M. Ezrin: a protein requiring conformational activation to link microfilaments to the plasma membrane in the assembly of cell surface structures. J Cell Sci. 1997;110(pt 24):3011–3018. [PubMed]
22. Yang HS, Cho MH, Zakowicz H, Hegamyer G, Sonenberg N, Colburn NH. A novel function of the MA-3 domains in transformation and translation suppressor Pdcd4 is essential for its binding to eukaryotic translation initiation factor 4A. Mol Cell Biol. 2004;24:3894–3906. [PMC free article] [PubMed]
23. Wei JS, Khan J. Purification of Total RNA from Mammalian Cells and Tissue. New York, NY: Cold Spring Harbor Laboratory Press; 2002.
24. Son CG, Bilke S, Davis S, Greer BT, Wei JS, Whiteford CC, Chen QR, Cenacchi N, Khan J. Database of mRNA gene expression profiles of multiple human organs. Genome Res. 2005;15:443–450. [PMC free article] [PubMed]
25. Riss J, Khanna C, Koo S, Chandramouli GV, Yang HH, Hu Y, Kleiner DE, Rosenwald A, Schaefer CF, Ben-Sasson SA, et al. Cancers as wounds that do not heal: differences and similarities between renal regeneration/repair and renal cell carcinoma. Cancer Res. 2006;66:7216–7224. [PubMed]
26. Gotz S, Garcia-Gomez JM, Terol J, Williams TD, Nagaraj SH, Nueda MJ, Robles M, Talon M, Dopazo J, Conesa A. High-throughput functional annotation and data mining with the Blast2GO suite. Nucleic Acids Res. 2008;36:3420–3435. [PMC free article] [PubMed]
27. Reczek D, Bretscher A. The carboxyl-terminal region of EBP50 binds to a site in the amino-terminal domain of ezrin that is masked in the dormant molecule. J Biol Chem. 1998;273:18452–18458. [PubMed]
28. Wilm M, Shevchenko A, Houthaeve T, Breit S, Schweigerer L, Fotsis T, Mann M. Femtomole sequencing of proteins from polyacrylamide gels by nano-electrospray mass spectrometry. Nature. 1996;379:466–469. [PubMed]
29. Graff JR, Zimmer SG. Translational control and metastatic progression: enhanced activity of the mRNA cap-binding protein eIF-4E selectively enhances translation of metastasis-related mRNAs. Clin Exp Metastasis. 2003;20:265–273. [PubMed]
30. Parsyan A, Svitkin Y, Shahbazian D, Gkogkas C, Lasko P, Merrick WC, Sonenberg N. mRNA helicases: the tacticians of translational control. Nat Rev Mol Cell Biol. 2011;12:235–245. [PubMed]
31. Sehgal A, Briggs J, Rinehart-Kim J, Basso J, Bos TJ. The chicken c-Jun 5′ untranslated region directs translation by internal initiation. Oncogene. 2000;19:2836–2845. [PubMed]
32. Stoneley M, Willis AE. Aberrant regulation of translation initiation in tumorigenesis. Curr Mol Med. 2003;3:597–603. [PubMed]
33. Coyle SM, Gilbert WV, Doudna JA. Direct link between RACK1 function and localization at the ribosome in vivo. Mol Cell Biol. 2009;29:1626–1634. [PMC free article] [PubMed]
34. Nilsson J, Sengupta J, Frank J, Nissen P. Regulation of eukaryotic translation by the RACK1 protein: a platform for signalling molecules on the ribosome. EMBO Rep. 2004;5:1137–1141. [PMC free article] [PubMed]
35. Evdokimova V, Ruzanov P, Anglesio MS, Sorokin AV, Ovchinnikov LP, Buckley J, Triche TJ, Sonenberg N, Sorensen PH. Akt-mediated YB-1 phosphorylation activates translation of silent mRNA species. Mol Cell Biol. 2006;26:277–292. [PMC free article] [PubMed]
36. Kang YK, Hong SW, Lee H, Kim WH. Prognostic implications of ezrin expression in human hepatocellular carcinoma. Mol Carcinog. 2010;49:798–804. [PubMed]
37. Wei YC, Li CF, Yu SC, Chou FF, Fang FM, Eng HL, Uen YH, Tian YF, Wu JM, Li SH, et al. Ezrin overexpression in gastrointestinal stromal tumors: an independent adverse prognosticator associated with the non-gastric location. Mod Pathol. 2009;22:1351–1360. [PubMed]
38. Elzagheid A, Korkeila E, Bendardaf R, Buhmeida A, Heikkila S, Vaheri A, Syrjanen K, Pyrhonen S, Carpen O. Intense cytoplasmic ezrin immunoreactivity predicts poor survival in colorectal cancer. Hum Pathol. 2008;39:1737–1743. [PubMed]
39. Madan R, Brandwein-Gensler M, Schlecht NF, Elias K, Gorbovitsky E, Belbin TJ, Mahmood R, Breining D, Qian H, Childs G, et al. Differential tissue and subcellular expression of ERM proteins in normal and malignant tissues: cytoplasmic ezrin expression has prognostic significance for head and neck squamous cell carcinoma. Head Neck. 2006;28:1018–1027. [PubMed]
40. Kobel M, Gradhand E, Zeng K, Schmitt WD, Kriese K, Lantzsch T, Wolters M, Dittmer J, Strauss HG, Thomssen C, et al. Ezrin promotes ovarian carcinoma cell invasion and its retained expression predicts poor prognosis in ovarian carcinoma. Int J Gynecol Pathol. 2006;25:121–130. [PubMed]
41. Wong CW, Lee A, Shientag L, Yu J, Dong Y, Kao G, Al-Mehdi AB, Bernhard EJ, Muschel RJ. Apoptosis: an early event in metastatic inefficiency. Cancer Res. 2001;61:333–338. [PubMed]
42. Chicurel ME, Singer RH, Meyer CJ, Ingber DE. Integrin binding and mechanical tension induce movement of mRNA and ribosomes to focal adhesions. Nature. 1998;392:730–733. [PubMed]
43. Kim S, Coulombe PA. Emerging role for the cytoskeleton as an organizer and regulator of translation. Nat Rev Mol Cell Biol. 2010;11:75–81. [PubMed]
44. Polak P, Oren A, Ben-Dror I, Steinberg D, Sapoznik S, Arditi-Duvdevany A, Vardimon L. The cytoskeletal network controls c-Jun translation in a UTR-dependent manner. Oncogene. 2006;25:665–676. [PubMed]
45. Van Horck FP, Holt CE. A cytoskeletal platform for local translation in axons. Sci Signal. 2008;1:pe11. [PMC free article] [PubMed]
46. Willett M, Flint SA, Morley SJ, Pain VM. Compartmentalisation and localisation of the translation initiation factor (eIF) 4F complex in normally growing fibroblasts. Exp Cell Res. 2006;312:2942–2953. [PubMed]
47. De Benedetti A, Graff JR. eIF-4E expression and its role in malignancies and metastases. Oncogene. 2004;23:3189–3199. [PubMed]
48. Livingstone M, Atas E, Meller A, Sonenberg N. Mechanisms governing the control of mRNA translation. Phys Biol. 2010;7:021001. [PubMed]
49. Hsieh AC, Ruggero D. Targeting eukaryotic translation initiation factor 4E (eIF4E) in cancer. Clin Cancer Res. 2010;16:4914–4920. [PubMed]
50. Woods AJ, Roberts MS, Choudhary J, Barry ST, Mazaki Y, Sabe H, Morley SJ, Critchley DR, Norman JC. Paxillin associates with poly(A)-binding protein 1 at the dense endoplasmic reticulum and the leading edge of migrating cells. J Biol Chem. 2002;277:6428–6437. [PubMed]
51. Woods AJ, Kantidakis T, Sabe H, Critchley DR, Norman JC. Interaction of paxillin with poly(A)-binding protein 1 and its role in focal adhesion turnover and cell migration. Mol Cell Biol. 2005;25:3763–3773. [PMC free article] [PubMed]
52. Di Cristofano C, Leopizzi M, Miraglia A, Sardella B, Moretti V, Ferrara A, Petrozza V, Della Rocca C. Phosphorylated ezrin is located in the nucleus of the osteosarcoma cell. Mod Pathol. 2010;23:1012–1020. [PubMed]
53. Batchelor CL, Woodward AM, Crouch DH. Nuclear ERM (ezrin, radixin, moesin) proteins: regulation by cell density and nuclear import. Exp Cell Res. 2004;296:208–222. [PubMed]
54. Rabl J, Leibundgut M, Ataide SF, Haag A, Ban N. Crystal structure of the eukaryotic 40S ribosomal subunit in complex with initiation factor 1. Science. 2011;331:730–736. [PubMed]
55. Sengupta J, Nilsson J, Gursky R, Spahn CM, Nissen P, Frank J. Identification of the versatile scaffold protein RACK1 on the eukaryotic ribosome by cryo-EM. Nat Struct Mol Biol. 2004;11:957–962. [PubMed]
56. Angenstein F, Evans AM, Settlage RE, Moran ST, Ling SC, Klintsova AY, Shabanowitz J, Hunt DF, Greenough WT. A receptor for activated C kinase is part of messenger ribonucleoprotein complexes associated with polyA-mRNAs in neurons. J Neurosci. 2002;22:8827–8837. [PubMed]

Articles from Neoplasia (New York, N.Y.) are provided here courtesy of Neoplasia Press
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...