stress-induced cxcr4 promotes migration and invasion...

Signal Transduction

Stress-Induced CXCR4 Promotes Migration and Invasionof Ewing Sarcoma

Melanie A. Krook1, Lauren A. Nicholls1, Christopher A. Scannell1, Rashmi Chugh2,Dafydd G. Thomas3, and Elizabeth R. Lawlor1,3

AbstractEwing sarcoma is the second most common bone cancer in pediatric patients. Although the primary cause of

death in Ewing sarcoma is metastasis, the mechanism underlying tumor spread needs to be elucidated. To thisend, the role of the CXCR4/SDF-1a chemokine axis as a mediator of Ewing sarcoma metastasis wasinvestigated. CXCR4 expression status was measured in primary tumor specimens by immunohistochemicalstaining and in multiple cell lines by quantitative reverse transcriptase PCR and flow cytometry. Migration andinvasion of CXCR4-positive Ewing sarcoma cells toward CXCL12/SDF-1a were also determined. Interest-ingly, while CXCR4 status was disparate among Ewing sarcoma cells, ranging from absent to high-levelexpression, its expression was found to be highly dynamic and responsive to changes in the microenvironment.In particular, upregulation of CXCR4 occurred in cells that were subjected to growth factor deprivation,hypoxia, and space constraints. This upregulation of CXCR4 was rapidly reversed upon removal of theoffending cellular stress conditions. Functionally, CXCR4-positive cells migrated and invaded toward an SDF-1a gradient and these aggressive properties were impeded by both the CXCR4 small-molecule inhibitorAMD3100, and by knockdown of CXCR4. In addition, CXCR4-dependent migration and invasion wereinhibited by small-molecule inhibitors of Cdc42 and Rac1, mechanistically implicating these Rho-GTPases asdownstream mediators of the CXCR4-dependent phenotype.

Implications:This study reveals the highly plastic and dynamic nature ofCXCR4 expression in Ewing sarcoma andsupports amodel inwhich stress-induced upregulation ofCXCR4 contributes to tumormetastasis to lung and bonemarrow, which express high levels of SDF-1a. Mol Cancer Res; 12(6); 953–64. �2014 AACR.

IntroductionEwing sarcoma is an aggressive bone and soft tissue

malignancy that primarily affects children and young adults(1). Over the past several decades, overall survival hasimproved dramatically for patients who present with localizeddisease. Multiagent systemic chemotherapy and aggressivelocal control measures have led to 5-year event-free survivalrates of 70% to 80% in these patients (1, 2).However, for theapproximately 25% of patients who present with metastaticdisease, the outcome is significantly worse. Event-free survivalfor these patients remains less than 25%, and intensificationof chemotherapeutic regimens has failed to improve outcome

(1). In addition, up to a third of patients who present withlocalized disease will relapse at distant sites following an initialclinical remission and outcomes for these patients are equallydismal. Innovative approaches to therapy and improvedunderstanding of the metastatic process are needed toimprove outcomes for patients with primary and relapsedmetastatic Ewing sarcoma.Despite its clinical importance, the biologic mechanisms

underlying Ewing sarcoma metastasis remain largelyunknown. Chemokine receptors are seven-transmembrane,G-protein–coupled cell surface proteins that are defined bytheir ability to induce chemotaxis through the binding ofsmall chemoattractant cytokines or chemokines (3). Che-mokine (C-X-C motif) receptor 4 (CXCR4) is the mostcommonly expressed chemokine receptor in human cancer,and increased expression of the CXCR4-encoding transcriptwas recently found to be associated withmetastatic disease inEwing sarcoma–derived cell lines and tumors (4). Signifi-cantly, highCXCR4 expression has also been associatedwithmetastatic disease and poor outcome in many other humancancers of both epithelial and nonepithelial origin (3, 5),including breast cancer (6), pancreatic cancer (7), leukemia(8), rhabdomyosarcoma (9–11), and osteosarcoma (12–14).Interestingly, the ligand for CXCR4, CXCL12 (SDF-1a), is

Authors' Affiliations: Departments of 1Pediatrics and CommunicableDiseases, 2Internal Medicine, and 3Pathology, University of Michigan, AnnArbor, Michigan

M.A. Krook and L.A. Nicholls contributed equally to this work.

Corresponding Author: Elizabeth R. Lawlor, University of Michigan, 1600Huron Parkway, NCRCBuilding 520, Rm 1352, Ann Arbor, MI 48109-2800.Phone: 734-615-4814; Fax: 734-764-9017; E-mail:[email protected]

doi: 10.1158/1541-7786.MCR-13-0668

�2014 American Association for Cancer Research.

MolecularCancer

Research

www.aacrjournals.org 953

on July 27, 2018. © 2014 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst March 20, 2014; DOI: 10.1158/1541-7786.MCR-13-0668

http://mcr.aacrjournals.org/

highly expressed in common sites of Ewing sarcoma metas-tasis, including lung, bone, and bone marrow, furtherimplicating the potential role of this axis in mediating thedistant spread of primary tumor cells.In this study, we evaluated the expression characteristics of

CXCR4 in Ewing sarcoma primary tumors and cell lines, andspecifically addressed whether the CXCR4/SDF-1a axis pro-motes tumor cell migration and invasion. Our findings dem-onstrate that expression of CXCR4 is both highly variable inEwing sarcoma and highly dynamic, being reversibly inducedin response to microenvironmental stresses, including growthfactor deprivation, hypoxia, and space constraints. Moreover,our studies confirm that Ewing sarcoma cells that express highlevels of CXCR4 display increased chemotacticmigration andinvasion,which ismediated, at least in part by activation of theRho-GTPases, Rac1, and Cdc42. Importantly, inhibition ofthe CXCR4/SDF-1a axis inhibits the aggressive cellularphenotype, thus revealing the potential contribution ofCXCR4 signaling to Ewing sarcoma metastasis.

Materials and MethodsCell cultureEwing sarcoma cell lines were kindly provided by Dr.

Timothy Triche Children's Hospital Los Angeles (CHLA,Los Angeles, CA) and the Children's Oncology Group(COG) cell bank (www.cogcell.org) and identities con-firmed by short tandem repeat profiling (courtesy of Dr.Patrick Reynolds, Texas Tech University, Lubbock, TX).Cells were maintained in RPMI-1640 media (Gibco) sup-plemented with 10% FBS (Atlas Biologicals, Inc.) and 6mmol/L L-glutamine (Life Technologies) in an incubator at37�C in 5%CO2. For CHLA-25 studies, plates were coatedwith 0.2% gelatin before cell seeding. For serum-starvedconditions, cells were cultured in the same conditionswithout the addition of FBS. For hypoxia studies, cells wereincubated in 1%O2 in an xVivo system (Biospherix) at 37�Cand 5% CO2. For growth constraint conditions, cells werecultured under standard culture conditions and CXCR4analyzed when cells reached 100% confluence.

Quantitative real-time PCR and Western blottingRNA was isolated from cell lines (RNeasy Mini; Qiagen)

and cDNA was generated (iScript; Bio-Rad). Quantitativereal-time PCR (qRT-PCR) was performed using validatedTaqman primers (CXCR4, 18S, and B2M; Life Technolo-gies). Analysis was performed in triplicate using the Light-Cycler 480 System (Roche Applied Science) and average Cpvalues were normalized relative to reference genes (18S andB2M) within each sample using DD Cp method. Levels ofphospho-ERK [Cell Signaling; Phospho-p44/42 MAPK(Erk1/2) (Thr202/Tyr204) (D13.14.4E) XP Rabbit mAb#4370], phospho-Akt [Cell Signaling; Phospho-Akt(Ser473) (D9E) XP Rabbit mAb #4060], Akt [Cell Signal-ing; Akt (pan) (C67E7 Rabbit mAb #4691), Erk (CellSignaling; p44/42 MAPK (Erk 1/2) #9102], and ACTIN[Abcam; Anti-beta Actin antibody (HRP) (ab20272)] weredetermined in whole cell lysates using standardWestern blotassays as previously described (15).

Cell sorting and assessment of Rac1 activation in sortedpopulationsCells were dissociated with Accutase (EMD Millipore

Corporation) and resuspended in staining media (L-15media, 0.1% bovine serum albumin, 10 mmol/L HEPES;Life Technologies), then blocked for 15minutes at 4�Cwithagitation (in 0.5% FBS; Atlas Biologicals, Inc.). After block-ing, human CXCR4 Alexa Fluor 488 monoclonal antibody(R&DSystems; clone 44717)was added (5mLper 1.0� 106

cells) and incubated for 30 minutes at 4�C with agitation.After two washes, cells were resuspended in staining mediaand passed through a 0.40-mm sterile nylon mesh strainer(Thermo Fisher Scientific). Flow cytometry analysis wasperformed using a BD Accuri C6 Flow Cytometer (BDBiosciences). Fluorescence-activated cell sorting (FACS) ofcells into CXCR4-positive and CXCR4-negative fractions(top 10% and bottom 10%) was done using a BeckmanCoulter MoFlo Astrios (Flow Core, University of Michigan,Ann Arbor, MI) with gating determined by analysis ofunstained controls.For evaluation of Rac1 activation, FACS-sorted TC-32

cells were serum-starved overnight in the presence or absenceof SDF-1a (200 ng/mL; R&D Systems). Levels of Rac1activation were determined using a G-LISA kit (Cytoskel-eton) according to the manufacturer's instructions.

ImmunohistochemistryFor tumor immunohistochemistry, formalin-fixed, paraf-

fin-embedded tumor microarray slides were deparaffinized,hydrated, epitope retrieved, and stained with an antibodyagainst CXCR4 (dilution 1:500; Abcam; AB-2074) aspreviously validated and described (16). Specificity of theantibody was confirmed in our hands by immunostaining ofcell pellets collected fromCXCR4-high, CXCR4-low as wellas control and CXCR4 knockdown TC-32 cells. Adjacenttumor microarray slides were incubated with CD99 (Mousemonoclonal antibody; clone 12E7; DAKO; Cat # M3601;1:100) and hematoxylin and eosin to identify tumor cells.Sections were scored for the presence of CXCR4 using theAllred schema (17). The proportion of tumor cells wasassigned a score between 0 and 5, and the staining intensitywas assigned a score between 0 and 3. These 2 values wereadded to produce a staining score. Given recent studiesdescribing nuclear localization of CXCR4 in some cancers(18, 19), both cytoplasmic and nuclear stainingwere assessedand equally weighted. Nuclear staining of CXCR4 wasevident in >10% of nuclei in 35% of cases.

CXCR4 knockdownFor CXCR4 knockdown studies, cell lines were trans-

duced with pLKO.1 puro vectors that contained one of twoindependent short hairpin RNAs targeted to CXCR4:shCXCR4–1: 50-TGGAGGGGATCAGTATATACA-30and shCXCR4-2: 50-GTTTTCACTCCAGCTAACACA-30 (Addgene; plasmid 12271 and 12272; ref. 20) or an inertnonsilencing sequence: shNS: 50-CAACAAGATGAA-GAGCACCAA-30. Cells were selected in puromycin (2mg/mL; Sigma) for 72 hours before subsequent experiments.

Krook et al.

Mol Cancer Res; 12(6) June 2014 Molecular Cancer Research954




In vitro migration and invasionReal-time cell analysis (RTCA) of cell migration and

invasion was monitored using a CIM-plate 16 and xCELLi-gence DP System (Acea Bioscience, Inc.). Cells were serum-starved overnight in RPMI-1640 with 0.2% Media Grade(K) Probumin (Millipore). Before cell seeding, electrodeswere coated with 0.2% gelatin and RPMI-1640 containing0.2% Probumin was placed in the upper chamber, andmedia containing SDF-1a (100 ng/mL; R&D Systems)were added to lower chambers. The CIM-plate was allowedto equilibrate for 1 hour in an incubator at 37�C in 5%CO2.For migration studies, 1 � 105 cells/well were placed in theupper chamber of a CIM-16 plate and then the plate wasequilibrated for 30 minutes at room temperature. Formigration assays done with combination of stresses, cellswere serum-starved and placed in either normoxic or hypoxicconditions overnight before evaluation of migration. Forinvasion studies, 1� 105 cells/well were plated in the upperchamber of wells that had been previously coated with 5%(v/v) Growth Factor ReducedMatrigel Matrix (diluted 1:20in basal RPMI media; BD BioSciences). Matrigel-coatedplates were allowed to equilibrate for 4 hours in an incubatorat 37�C in 5% CO2 before addition of cells. For compoundassays, cells were pretreated overnight with either 2.5 mg/mLAMD3100 (Sigma-Aldrich), 30 mmol/L Rac1 inhibitor[NSC 23766 (hydrochloride); Cayman Chemical), or 7mmol/L Cdc42 inhibitor (ML 141; EMD Millipore) andthen seeded in CIM-16 plates as above. Parallel migrationassays were performed with 2 � 105 cells on 0.8 mm cellculture inserts (Thermo Fisher Scientific) for 24 hours. Afterincubation, noninvading cells were removed from the uppersurface and inserts were stained (Crystal Violet Stain; 0.5%crystal violet, 20% methanol) and migratory cells wereimaged by light microscopy.

Statistical analysisData are reported as mean � SEM from three indepen-

dent experiments, and P values were calculated using theStudent t test.

ResultsCXCR4 expression is highly heterogeneous in EwingsarcomaRecent studies of gene expression showed that expression

of the CXCR4 transcript varies among Ewing sarcoma celllines and tumors (4). To determine if expression of theCXCR4 protein is equally heterogeneous, we assessed apanel of four well-established Ewing sarcoma cell lines.qRT-PCR analyses corroborated earlier studies and dem-onstrated a wide range of CXCR4 expression (Fig. 1A). Thevariability in transcript expression was mirrored by flowcytometry studies of protein expression, with relatively lowlevels of CXCR4 detected in TC-71 and A673 cells andhigh-level expression evident in CHLA-25 and TC-32 cells(Fig. 1B). Analysis at the level of individual cells showedthat the variation in CXCR4 signal intensity between thedifferent cell lines was a result of different frequenciesof CXCR4-positive cells within each culture (Fig. 1B).

Specifically, in the two low-expressing cell lines, fewer than5% of cells expressed CXCR4. Conversely, 20% to 40% ofcells in CHLA-25 and TC-32 expressed detectable levels ofthe receptor at the cell surface. In addition, the level ofexpression in CXCR4-positive populations ranged fromweak to robust, as demonstrated by the continuum offluorescence intensities displayed by CXCR4-positive cells(Fig. 1C). To evaluate whether this same heterogeneity inCXCR4 protein expression exists in primary tumors, weevaluated a tissue microarray comprised of 64 Ewingsarcoma samples. Sufficient viable tumor was present toscore 43 tumor samples from 32 unique patients. Consis-tent with cell line data, CXCR4 staining showed markedintertumor variability, ranging from absent (N ¼ 13) tostrongly positive in the majority of tumor cells (N ¼ 13).The remainder of the samples (N ¼ 17) showed anintermediate staining pattern in which both CXCR4-pos-itive and CXCR4-negative tumor cells were identified inthe same core specimen (Fig. 1D). No difference in stainingpattern was identified between 28 samples that wereobtained from primary tumor specimens and 15 that wereisolated at the time of disease recurrence (Fig. 1E). Theaverage CXCR4 score was 5.0 in 4 diagnostic samples thatwere obtained from patients with metastatic disease and 3.8in 17 localized tumor samples. Although this analysisshowed a trend to increased expression in primary tumorsof patients who present with metastatic disease, the samplesize is inadequate to draw conclusions about associationsbetween CXCR4 expression and clinical stage. Thus, likecell lines, CXCR4 protein expression is highly heteroge-neous in Ewing sarcoma tumors, and individual cells withinthe same tumor also vary in CXCR4 expression.

CXCR4 expression is dynamic and induced in responseto growth factor deprivationTumor cell heterogeneity is a key factor that contributes

to drug resistance and tumor progression. We observedsignificant interexperiment heterogeneity in CXCR4expression in our in vitro studies of Ewing sarcoma celllines (Fig. 1A and B). In particular, we noted that therelative proportion of CXCR4-positive cells varied sub-stantially between replicate experiments, particularly inthe two high-expressing cell lines. This observation,together with the highly variable nature of expression intumor samples, led us to hypothesize that expression ofCXCR4 may be dynamic in Ewing sarcoma and subject toregulation in response to changes in the local microenvi-ronment. To begin to address this possibility, we testedwhether the variability in expression might be a conse-quence of the relative availability of growth factors. Toachieve this, we measured CXCR4 expression in cells thathad been deprived of serum. As shown, serum deprivationled to an increased frequency of CXCR4-positive cells inthree of the four cell lines (Fig. 2A). Only TC-71 cellsremained unchanged with fewer than 2% of cells expres-sing CXCR4 in both serum-rich and serum-deprivedconditions. To determine if the upregulation of CXCR4protein expression was a consequence of increased CXCR4

CXCR4 Promotes Ewing Sarcoma Invasion

www.aacrjournals.org Mol Cancer Res; 12(6) June 2014 955




transcription, we compared mRNA levels in the twoconditions. Consistent with transcriptional upregulation,CXCR4 mRNA levels increased in all four cell linesfollowing serum deprivation (Fig. 2B). In addition, thedegree of transcriptional induction corresponded to that ofincreased protein expression. TC-71 showed the least andTC-32 cells showed the most robust upregulation oftranscript (Fig. 2B). We next evaluated whether restora-tion of growth factor availability would reverse the induc-tion of CXCR4. To achieve this, serum was added to themedia of cells that had been starved for 24 hours. Fol-lowing the addition of serum, rapid downregulation ofCXCR4 expression was observed with levels being restoredto baseline within 24 hours (Fig. 2C).Next, we addressed whether the reversible changes in

CXCR4 expression seen in heterogeneous cell populationsreflected dynamic regulation at the level of individual cells.TC-32 cells were FACS-sorted into pure populations ofCXCR4-positive and CXCR4-negative cells, and then mon-

itored over 3 weeks in ambient culture conditions to deter-mine if positive cells would become negative and vice versa.Consistent with dynamic and bidirectional regulation ofCXCR4, both populations of FACS-sorted TC-32 cellsgradually reverted to their basal pattern of CXCR4 expres-sion (Fig. 2D). Specifically, the initial CXCR4-positivepopulation generated CXCR4-negative cells and the initialCXCR4-negative population generated CXCR4-positivecells with both cultures reestablishing the baseline equilib-rium state of approximately 30% to 40% CXCR4-positivecells within 3 weeks.Thus, CXCR4 expression in Ewing sarcoma cells is

dynamic and is rapidly and reversibly induced in responseto growth factor deprivation.Moreover, Ewing sarcoma cellsin standard tissue culture transition back and forth betweenCXCR4-negative and CXCR4-positive cell states inresponse to changes in the microenvironment, ultimatelymaintaining a basal equilibrium state that is specific for eachcell line and condition.

Figure 1. Heterogeneous expression of CXCR4 in Ewing sarcoma. A, qRT-PCR ofCXCR4 expression in Ewing sarcoma cell lines. Expression was normalizedto the housekeeping 18S rRNA in each sample. Individual replicates are shown and the horizontal line represents average expression of replicateexperiments. B, flow cytometry of CXCR4 cell surface expression in Ewing sarcoma cell lines. Data are expressed as the percentage of positive cells. Gatingwas determined based on unstained control cells in the same experiment. Individual replicates are shown and the horizontal line represents averageexpression of replicate experiments. C, representative dot plots for each of the 4 cell lines are shown, demonstrating the heterogeneity of CXCR4 expressionboth between cell lines and also showing the range of positivity within the CXCR4-positive cell in each cell line. D, immunohistochemical (IHC) stainingof 43 Ewing sarcoma tumor samples showed marked variability ranging from complete absence to robust staining in all tumor cells. Representativeexamplesof robust (score 6–8), intermediate (score2–5), andnegative (score 0) tumors are shownalongwith the total number of tumors in each category. BothCXCR4-positive and -negative tumor cells were evident in intermediate tumors. E, summary of IHC scores for 28 primary tumor samples and15 relapse samples shows no difference in CXCR4 expression between the two categories.

Krook et al.





CXCR4 is induced in Ewing sarcoma cells that areexposed to hypoxia and growth constraintsHaving established that growth factor deprivation leads to

induction of CXCR4, we next questioned whether otherstresses that might be encountered by a growing Ewingsarcoma, such as hypoxia and space constraints, would alsoaffect CXCR4 expression. CXCR4 is induced by hypoxia-inducible factor 1-a (HIF1-a) in mesenchymal stem cellsand cancer cells that are exposed to hypoxic environments(21, 22). Consistent with these observations, we discoveredthat exposure of Ewing sarcoma cells to hypoxia resulted inan increase in CXCR4 transcript (Fig. 3A) and an increasedfrequency of CXCR4þ cells (Fig. 3B). Removal of thehypoxic insult resulted in a return to basal levels within48 hours (Fig. 3B). Interestingly, in direct contrast to growthfactor deprivation, TC-71 cells were more susceptible tohypoxia-induced changes than were TC-32 cells, indicatingthat the inherent plasticity of CXCR4 expression in responseto different stimuli varies among the different cell lines.

Finally, subjecting cells to space constraints, by growingthem to confluence, also resulted in reproducible upregula-tion of CXCR4 transcript (Fig. 3C) and protein expression(Fig. 3D) that was reversed when cells were returned tosubconfluent, log-phase growth conditions (Fig. 3D).Thus, like growth factor deprivation, exposing Ewing

sarcoma cells to hypoxia and space constraints also resultsin upregulation of CXCR4 transcription and an increasedfrequency of CXCR4-positive cells. These changes arereversed when thesemicroenvironment stresses are removed,demonstrating the highly plastic and dynamic nature ofCXCR4 regulation in Ewing sarcoma cells.

CXCR4 promotes Ewing sarcoma cell migration andinvasionGiven its well-established role as a mediator of metastasis

in numerous other cancers, we next investigated whetherCXCR4 might also contribute to an invasive phenotype inEwing sarcoma. First, we assessed whether Ewing sarcoma

Figure 2. CXCR4 expression is reversibly induced in response to growth factor deprivation. A, surface expression of CXCR4was determined by flowcytometryas in Fig. 1 for Ewing sarcoma cells plated under standard (10% FBS) and serum-deprived (serum-free media, SFM) conditions. Exposure of cells toSFM for 24 hours resulted in upregulation ofCXCR4. Each line andpair of data points represents the data for an independent experiment. B, qRT-PCRanalysisof CXCR4 expression in Ewing sarcoma cells grown in SFM conditions for 24 hours. Expression in each sample was normalized to the housekeeping b 2microglobulin (B2M) and expressed as fold change relative to expression in standard 10% FBS conditions. Results are shown as mean � SEMfrom three independent experiments. C, flow cytometry of CXCR4 expression in serum-starved Ewing sarcoma cells (SFM) after being returned to standardculture conditions (10%) shows reversion of expression to baseline state. Each line and pair of data points represents the data for an independentexperiment. D, TC32 cells were FACS-sorted into CXCR4-high (top 10%) and CXCR4-low (bottom 10%) populations and then both populations weremaintained in standard culture conditions for 3 weeks. CXCR4 expression wasmonitored by flow cytometry on days 5, 12, 16, and 21 after sorting, revealingreversion over time to baseline heterogeneity. Results are shown as mean � SEM from three independent experiments.






cells demonstrate chemotactic migration toward SDF-1a.As expected, given the very low frequency of CXCR4-positive cells, neither TC-71 nor A673 cells migrated towardSDF-1a (data not shown). In contrast, the CXCR4-high celllines, CHLA-25 and TC-32 both demonstrated substantialand rapid migration toward SDF-1a (Fig. 4A and B).Moreover, exposure of the cells to AMD3100, a small-molecule inhibitor of CXCR4, significantly inhibited thischemotactic migration (Fig. 4A and B). To further validatethese findings, we induced stable knockdown of CXCR4 inboth CHLA-25 and TC-32 cell lines using two differentshort hairpin RNA constructs (Fig. 4C and D). Consistent

with pharmacologic inhibitor studies, knockdown ofCXCR4 significantly impaired the migration of bothCHLA-25 and TC-32 cells toward SDF-1a (Fig. 4E and F).Invasion of cancer cells through basement membranes

comprised of extracellular matrix proteins is a critical step inthe metastatic cascade (23). To model this process in vitro,we used Matrigel, a gelatinous protein mixture mimickingextracellular components found in tumors (24). BothCHLA-25 and TC-32 cells invaded through the Matrigellayer toward SDF-1a, and invasion was abrogated by bothAMD3100 (Fig. 5A and B) and by CXCR4 knockdown(Fig. 5C and D). In contrast, SDF-1a had no effect on the

Figure 3. CXCR4 expression is reversibly induced in response to hypoxia and cell confluence. A, qRT-PCR analysis of CXCR4 expression in Ewing sarcomacells grown in hypoxic conditions for 24 hours shows upregulation of the transcript. Gene expression calculated in each sample was normalizedto the housekeepingb2microglobulin (B2M) and expressed as fold change in hypoxia relative to expression in normoxia (control). Histograms representmeanfold change � SEM for three independent experiments. B, flow cytometry of CXCR4 expression in Ewing sarcoma cells before (21%) and after (1%)exposure to hypoxia for 24 hours shows upregulation of CXCR4 expression in hypoxic conditions. The CXCR4-positive cell frequency reverted to baseline 48hours after cells were returned to ambient (21%) conditions. C, qRT-PCR analysis of CXCR4 expression in Ewing sarcoma cells grown in log phase,low density (low) compared with confluent, high-density (high) conditions for 48 hours. Gene expression calculated as in Fig. 3A and expressed as mean foldchange � SEM in high-density cells relative to low-density (control) cells. D, flow cytometry of CXCR4 expression in log-phase (low) and confluent(high) conditions shows upregulation ofCXCR4expression that is then reversedwhen cells are returned to low-density growth conditions after 48 hours. For Aand C, results are shown as mean � SEM from three independent experiments. For B and D, each line and pair of data points represents the data for anindependent experiment.

Krook et al.





invasive potential of A673 cells (data not shown). Thus,CXCR4-positive Ewing sarcoma cells are stimulated tomigrate and invade toward SDF-1a, but modulation of theCXCR4/SDF-1a axis by pharmacologic or genetic meanscan profoundly inhibit this response.

Rac1 and Cdc42 mediate CXCR4-dependent migrationand invasionThe mechanisms by which the CXCR4/SDF-1a axis

contributes to tumor growth and metastasis are pleiotropic,and cell type and context dependent (17). Activation of themitogen-activated protein kinase (MAPK) and phosphoi-nositide 3-kinase (PI3K) cascades are both observed down-stream of CXCR4 activation (17). In addition, studies ofbreast and liver cancer have shown that the small GTPases,Rho, Rac1, and Cdc42 are activated in these tumors fol-lowing SDF-1a engagement of CXCR4, and that Rho-GTPase signaling is, at least in part, responsible for medi-ating the invasive/metastatic phenotype (25, 26). Interest-ingly, recent studies of Ewing sarcoma have also implicatedRac1 as a keymediator of tumor metastasis (27). To begin toaddress the mechanisms by which CXCR4 promotes theinvasive cellular phenotype in Ewing sarcoma, we assessedthe effects of SDF-1a treatment on the MAPK and PI3K

pathways by evaluating phosphorylation of extracellularsignal–regulated kinase (ERK) and AKT. As shown, SDF-1a treatment for 24 hours, which promoted cell migrationand invasion, had no significant impact on activation ofeither kinase in CHLA-25 or TC32 cells (Fig. 6A). Next weinvestigated whether SDF-1a–dependent chemotacticmigration and invasion were dependent on Rac1 and/orCdc42. Exposure of Ewing sarcoma cells to either NSC23766 or ML 141, small-molecule inhibitors of Rac1 andCdc42, respectively, resulted in significant inhibition ofboth migration (Fig. 6B–D) and invasion (Fig. 6E). Inparticular, inhibition of Rac1 nearly completely abrogatedthe chemotactic invasion ofCXCR4-positive Ewing sarcomacells. To determine if Rac1 activation is induced by SDF-1a,TC-32 cells were FACS-sorted on the basis of CXCR4 andRac1 activity measured in the different populations in thepresence or absence of SDF-1a. As shown, CXCR4-highcells displayed higher Rac1 activity than CXCR4-low cells,even in unstimulated conditions (Fig. 6F). Exposure to SDF-1a potentiated Rac1 activity in both cell populations butactivation of Rac1 was reproducibly most pronounced inSDF-1a–stimulated CXCR4-high cells. Together, thesestudies demonstrate that the invasive cellular phenotypeimparted to CXCR4-positive Ewing sarcoma cells following

Figure 4. CXCR4 promoteschemotactic migration of Ewingsarcoma cells. A and B, migrationof CHLA-25 (A) and TC32 (B) cellstoward SDF-1a (100 ng/mL) wasmeasured using real-time cellanalysis (xCELLigence CIM-Plate16) in the presence and absence ofthe CXCR4 inhibitor AMD3100.AMD3100 significantly inhibitedchemotaxis. C and D, knockdownof CXCR4 was effectively achievedin CHLA-25 (C) and TC32 cells (D)using lentiviral transduction of 2different shRNA sequencesdirected against CXCR4 (sh1 andsh2). Control cellswere transducedwith an inert nonsilencing shRNAvector (shNS). Successfulknockdown was confirmed byqRT-PCR (left) and flow cytometry(right). E and F, migration of CHLA-25 (E) and TC32 (F) cells towardSDF-1a (100 ng/mL) was inhibitedfollowing knockdown of CXCR4. Inall plots, graphs represent mean �SEM of three independentexperiments with four replicatesper condition. ��, P < 0.01;��, P < 0.001; and ��, P < 0.0001as compared with controls.






SDF-1a engagement is, at least in part, mediated by down-stream activation of Rac1 and Cdc42 Rho-GTPases, inparticular Rac1.

CXCR4-dependent migration is increased in Ewingsarcoma cells that are exposed to multiple stressesCells in the center of rapidly growing tumors are subjected

to a diminished blood supply and must simultaneouslyendure conditions of both growth factor and oxygen dep-rivation. Given our findings that CXCR4 and CXCR4-dependent migration are induced by each of these stressesindependently, we next investigated whether chemotacticmigration of Ewing sarcoma cells would be further enhancedin cells that were simultaneously exposed to both serumstarvation and hypoxia. As predicted, migration of serum-starved (and thus CXCR4-upregulated) CHLA-25 andTC32 cells toward SDF-1a was increased under hypoxicas compared with normoxic conditions (Fig. 7A). Together,these studies suggest an additive role of microenvironmentalstresses in promoting CXCR4-mediated Ewing sarcoma cellmigration.

DiscussionIn these studies, we have shown that expression of CXCR4

is heterogeneous, both in Ewing sarcoma cell lines andprimary tumors, and that expression is also highly dynamic.In particular, CXCR4 transcript and protein expression arereversibly increased when cells are exposed to serum depri-vation, hypoxia, and confluent growth conditions. All ofthese stresses are encountered by a growing tumor in vivo as itoutstrips its blood supply and expands to abut surrounding

adjacent tissues, resulting in growth factor and oxygendeprivation and space constraints. Using both pharmaco-logic and genetic tools, we have also demonstrated thatCXCR4-positive Ewing sarcoma cells display a highlymigra-tory and invasive chemotactic phenotype when exposed tothe CXCR4 ligand, SDF-1a/CXLC12. Our finding thatEwing sarcoma cells dynamically regulate CXCR4 leads us topropose a new model of Ewing sarcoma tumor cell invasionin which local microenvironment-induced cell stress resultsin upregulation of CXCR4, promoting chemotactic migra-tion and invasion of CXCR4-positive Ewing sarcoma cells todistant sites of metastasis. In particular, this model proposesa mechanistic basis for the preferential metastasis of Ewingsarcoma cells to lungs and bonemarrow,microenvironmentsrich in SDF-1a/CXLC12 (Fig. 7B).Studies of Ewing sarcoma tumors and cell lines have

previously identified a potential role for the CXCR4/SDF-1a axis in Ewing sarcoma pathogenesis (4, 16, 28, 29). Inparticular, interrogation of gene expression databasesidentified an association between high levels of the CXCR4transcript and metastatic disease (4). In addition, con-comitant clinical correlative studies in the same studysuggested that Ewing sarcoma tumors that express highlevels of CXCR4 and a related chemokine receptor,CXCR7, which also binds SDF-1a, are associated withworse overall survival (4). More recently, an immunohis-tochemical study of 30 Ewing sarcoma tumors revealedrobust CXCR4 staining in approximately one third ofcases, and these investigators also reported an associationbetween CXCR4 expression and poor outcome, althoughno correlation with metastatic disease was identified (16).In our own study, we also detected robust expression of

Figure5. CXCR4promotes invasionof Ewing sarcoma cells. A andB, invasion of CHLA-25 (A)and TC32 (B) cells toward SDF-1a(100 ng/mL) through a Matrigellayer was monitored byreal-time cell assays as in Fig. 4.AMD3100 inhibited migration ofboth cell lines. C and D,knockdown of CXCR4, as in Fig. 4,resulted in significant inhibition ofinvasion of CHLA-25 (C) andTC32 (D) cells. Graphs representmean � SEM of three independentexperiments with four replicatesper condition. ��, P < 0.01 and��, P < 0.001 as compared withcontrols.

Krook et al.





CXCR4 in approximately one third of cases and anabsence of CXCR4-positive cells in another third. How-ever, CXCR4-positive cells were also identified in theremaining third of cases, but tumor cells were found tobe heterogeneously positive. Consistent with the study byBerghuis and colleagues (16), the pattern of CXCR4expression in our tumor cohort did not correlate withthe source of the tumor sample. Samples from bothprimary and recurrent lesions showed equally heteroge-neous expression patterns. Together, these studies confirmthe heterogeneous nature of CXCR4 protein expression inprimary Ewing sarcoma tumors and support further inves-tigation of the contribution of CXCR4 signaling to Ewingsarcoma progression. Whether or not high-level expressionor an increased frequency of CXCR4-positive cells at thetime of diagnosis portends a worse prognosis for patientsstill requires further investigation. Specifically, given the

complexities of prognostic biomarker discovery, it iscritical that this question next be addressed prospectivelyin a large cohort of equivalently treated patients (30).Moreover, given the highly heterogeneous nature ofCXCR4 expression, a single core-needle biopsy samplemay or may not be representative of CXCR4 expression inother areas of the tumor. Ideally, multiple cores should beassessed when a dynamically regulated and heterogeneousprotein like CXCR4 is being evaluated as a potentialprognostic biomarker.Berghuis and colleagues identified a role for CXCR4/

SDF-1a in promoting cell proliferation, rather than metas-tasis (16). Given the pleiotropic nature and cell context-specific response of CXCR4-dependent signaling, it is notsurprising that different experimental designs have uncov-ered different results and elucidated different functions forthe CXCR4/SDF-1a axis in Ewing sarcoma pathogenesis.

Figure 6. CXCR4-mediatedchemotaxis is dependent on Rac1and Cdc42. A, Western blot ofCHLA-25 and TC32 cells shows nosignificant induction of either P-ERK (left) or P-AKT (right) following24-hour exposure of serum-starved cells (SFM) to SDF-1a (100ng/mL). B and C, Endpoint analysisof cell migration toward SDF-1a inthe presence or absence of Rac1(NSC 23766) or Cdc42 (ML141)inhibitors was performed asdescribed in Materials andMethods using transwell assaysand crystal violet staining.Inhibition of Rac1 and Cdc42 bothimpeded CXCR4-dependent cellmigration. D and E, pharmacologicinhibition of Rac1 (NSC 23766) andCdc42 (ML141) inhibits CXCR4-dependent migration (D) andinvasion (E) of CHLA-25 and TC32cells. Summary histograms showmean � SEM of three independentexperiments with four replicatesper condition. �, P < 0.05; ��, P <0.01; and ��, P < 0.001 ascompared with controls. F, Rac1activity was measured in TC32cells sorted on the basis of CXCR4.Absorbance values are normalizedto control condition (0% inCXCR4-low) and summary histogramsshow mean � SEM of twoindependent sorts with threereplicates per condition. �,P <0.05.






We have shown that exposure of CXCR4-positive Ewingsarcoma cells to SDF-1a results in robust induction ofchemotaxis, and that both migration and invasion arepromoted by activation of CXCR4 signaling. In addition,studies with small-molecule inhibitors AMD3100, NSC23766, and ML 141 showed that migration and invasiontoward SDF-1a are dependent on CXCR4 and its down-stream effectors, Rac1 and Cdc42, respectively. Interesting-ly, our studies also indicated that the basal activity of Rac1 ishigher in CXCR4-positive Ewing sarcoma cells thanCXCR4-negative cells, even in the absence of ligand andthat Rac1 was maximally activated by SDF-1a in theCXCR4-positive population. Moreover, we have also foundthat inhibiting Rac1 blocks SDF-1a –independent invasionof serum-starved Ewing cells that do not express high levelsof CXCR4 (data not shown). In addition, Rac1 was alsorecently implicated as a key mediator of Ewing sarcoma cellinvasion and metastasis downstream of the tyrosine kinasereceptor ERBB4 (27). Thus, activation of Rac1 is implicatedin both nonchemotactic and SDF-1a–mediated Ewingsarcoma migration and invasion, downstream of and inparallel to CXCR4-dependent signaling, suggesting that thisRho-GTPase may be a critical downstream hub, present atthe convergence of multiple Ewing sarcoma metastaticpathways.

The origins of tumor heterogeneity are multifactorial, andcontributing factors include genetic variation, stochasticprocesses, different microenvironments, and cell plasticity(31). Indeed, dynamic regulation of metastasis-inducinggenes in response to exogenous cues is a hallmark of epithelialcancer cell plasticity, resulting in epithelial–mesenchymaltransition (EMT), a critical initiating event in the onset ofcarcinoma metastasis (32). Unlike most adult solid tumors,pediatric solid tumors mainly arise from nonepithelial tis-sues, predominantly neural and mesenchymal lineages, thusobviating a role for EMT. We have discovered that, likeEMT genes in epithelial cancers, CXCR4 expression inEwing sarcoma is highly plastic and this phenotypic plas-ticity results in functional changes that can contribute to cellinvasion and metastatic dissemination. In particular,CXCR4 expression is highly responsive to stresses in thelocal microenvironment, reverting to its basal state when thestressor is removed. Consistent with this observation,dynamic regulation of CXCR4 has also been observed inneuroblastoma, a neural crest–derived solid tumor (33, 34),demonstrating that plasticity of CXCR4 is not limited toEwing sarcoma. Interestingly, high levels of CXCR4 havealso been identified in tumor- and metastasis-initiatingcancer stem cell populations (7, 35, 36), suggesting thatdynamic regulation of CXCR4 may contribute to the

Figure 7. Hypothetical model ofstress-induced, CXCR4-dependent invasion andmetastasis. A, migration of CHLA-25 and TC32 cells toward SDF-1a(100 ng/mL) was measured usingreal-time cell analysis(xCELLigence CIM-Plate 16) innormoxic (21% O2) or hypoxicconditions (1% O2). Chemotacticmigration of each cell line wasfurther increased in hypoxia relativeto normoxia. Graphs representmean � SEM of three independentexperiments with four replicatesper condition. Cell index wasnormalized to migration innormoxic conditions for each cellline. �, P < 0.05 as compared withcontrols. B, a growing tumorbegins to deplete its resources,including growth factors andoxygen. Continued tumor growthleads to space constraint at theprimary site. Upregulation ofCXCR4 in response to thesemicroenvironmental stressespromotes invasion of Ewingsarcoma cells through basementmembranes and extracellularmatrix and chemotaxis towardSDF-1a–rich secondary sites suchas lung and bone marrow.

Krook et al.





dynamic regulation of stemness that has been described inhighly plastic cancer cell populations (37). We hypothesizethat dynamic regulation of CXCR4 in Ewing sarcoma, aswell as other pediatric solid tumors, contributes to cellularheterogeneity and supports the dynamic transition of cellsbetween nonmetastatic and metastatic states. Studies areongoing in our laboratory to determine the precisemolecularmechanisms that underlie the dynamic regulation ofCXCR4 expression and to define whether it is under thecontrol of epigenetic, transcriptional, and/or posttranscrip-tional regulatory pathways.Current systemic cytotoxic agents have reached the limit

of tolerability, and novel approaches to treatment, inparticular approaches that prevent metastatic relapse, aredesperately needed for Ewing sarcoma and other invasivesolid tumors (38). The CXCR4/SDF-1a axis is a well-established mediator of tumor metastasis, and it offers apotentially attractive therapeutic target for the treatmentand prevention of metastatic disease (17). Our currentwork, along with recent studies of other sarcomas andneuroblastoma (9, 10, 14, 34, 39), suggests that this axisrepresents a potential target for metastasis prevention inEwing sarcoma as well as other aggressive pediatric tumorsand should be further investigated in relevant preclinicaltherapeutic models of these cancers. In particular, studiesof spontaneous metastasis using orthotopic, patient-derived xenograft models will be most informative andshould be pursued for preclinical studies of CXCR4-targeted therapies.

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

Authors' ContributionsConception and design: M.A. Krook, L.A. Nicholls, E.R. LawlorDevelopment of methodology: L.A. Nicholls, D.G. Thomas, E.R. LawlorAcquisition of data (provided animals, acquired and managed patients, providedfacilities, etc.): M.A. Krook, L.A. Nicholls, C.A. Scannell, D.G. ThomasAnalysis and interpretation of data (e.g., statistical analysis, biostatistics, compu-tational analysis): M.A. Krook, L.A. Nicholls, C.A. Scannell, R. Chugh,D.G. Thomas, E.R. LawlorWriting, review, and/or revision of the manuscript: M.A. Krook, L.A. Nicholls,C.A. Scannell, R. Chugh, D.G. Thomas, E.R. LawlorAdministrative, technical, or material support (i.e., reporting or organizing data,constructing databases): L.A. NichollsStudy supervision: E.R. Lawlor

AcknowledgmentsThe authors thank Dr. Gary Luker and his laboratory and staff of the University of

Michigan FACS core for their technical assistance. They also thank current and pastmembers of the Lawlor laboratory for helpful discussion.

Grant SupportThis work was supported by grants from the NIH (SARC SPORE 1U01-

CA114757, to E.R. Lawlor and R. Chugh; NHLBI-5T32HL007622, to L.A.Nicholls; and NICHD-5T32HD007513, to L.A. Nicholls) and the University ofMichigan's Cancer Center Support Grant (P30CA046592) by the use of the followingCancer Center Core(s): Flow cytometry. Additional financial support from the LiddyShriver Sarcoma Initiative, the Russell G. Adderley Endowment, the Nancy NewtonLoeb Fund, the Evan Shapiro Fund to Combat Pediatric Cancer, and Go4theGoal isgratefully acknowledged.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be herebymarked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

Received December 19, 2013; revised February 24, 2014; accepted March 12,2014; published OnlineFirst March 20, 2014.

References1. Balamuth NJ, Womer RB. Ewing's sarcoma. Lancet Oncol 2010.

11:184–92.2. Gorlick R, Janeway K, Lessnick S, Randall RL, Marina N. COG

Bone Tumor Committee. Children's Oncology Group's 2013 blue-print for research: bone tumors. Pediatr Blood Cancer 2013;60:1009–15.

3. Domanska UM, Kruizinga RC, Nagengast WB, Timmer-Bosscha H,Huls G, de Vries EG, et al. A review on CXCR4/CXCL12 axis inoncology: no place to hide. Eur J Cancer 2013;49:219–30.

4. Bennani-Baiti IM, Cooper A, Lawlor ER, Kauer M, Ban J, Aryee DN,et al. Intercohort gene expression co-analysis reveals chemokinereceptors as prognostic indicators in Ewing's sarcoma. Clin CancerRes 2010;16:3769–78.

5. Cojoc M, Peitzsch C, Trautmann F, Polishchuk L, Telegeev GD,Dubrovska A. Emerging targets in cancer management: role of theCXCL12/CXCR4 axis. Onco Targets Ther 2013;6:1347–61.

6. Muller A, Homey B, Soto H, Ge N, Catron D, Buchanan ME, et al.Involvement of chemokine receptors in breast cancer metastasis.Nature 2001;410:50–6.

7. Hermann PC, Huber SL, Herrler T, Aicher A, Ellwart JW, Guba M, et al.Distinct populations of cancer stem cells determine tumor growth andmetastatic activity in human pancreatic cancer. Cell Stem Cell2007;1:313–23.

8. Mohle R, Failenschmid C, Bautz F, Kanz L. Overexpression of thechemokine receptor CXCR4 in B cell chronic lymphocytic leukemia isassociated with increased functional response to stromal cell-derivedfactor-1 (SDF-1). Leukemia 1999;13:1954–9.

9. Libura J, Drukala J, Majka M, Tomescu O, Navenot JM, Kucia M, et al.CXCR4-SDF-1 signaling is active in rhabdomyosarcoma cells andregulates locomotion, chemotaxis, and adhesion. Blood 2002;100:2597–606.

10. Strahm B, Durbin AD, Sexsmith E, Malkin D. The CXCR4-SDF1alphaaxis is a critical mediator of rhabdomyosarcoma metastatic signal-ing induced by bone marrow stroma. Clin Exp Metastasis 2008;5:1–10.

11. TarnowskiM,GrymulaK, RecaR, Jankowski K,MaksymR, TarnowskaJ, et al. Regulation of expressionof stromal-derived factor-1 receptors:CXCR4 and CXCR7 in human rhabdomyosarcomas. Mol Cancer Res2010;8:1–14.

12. Laverdiere C, Hoang BH, Yang R, Sowers R, Qin J, Meyers PA, et al.Messenger RNA expression levels of CXCR4 correlate with metastaticbehavior and outcome in patients with osteosarcoma. Clin Cancer Res2005;11:2561–7.

13. Lin F, Zheng SE, Shen Z, Tang LN, Chen P, Sun YJ , et al. Relation-ships between levels of CXCR4 and VEGF and blood-borne metas-tasis and survival in patients with osteosarcoma. Med Oncol2011;28:649–53.

14. Namlos HM, Kresse SH, M€uller CR, Henriksen J, Holdhus R, Sæter G,et al. Global gene expression profiling of human osteosarcomasreveals metastasis-associated chemokine pattern. Sarcoma 2012;2012:639038.

15. Lawlor ER, Scheel C, Irving J, Sorensen PH. Anchorage-independentmulti-cellular spheroids as an in vitro model of growth signaling inEwing tumors. Oncogene 2002;21:307–18.

16. Berghuis D, SchilhamMW, Santos SJ, Savola S, Knowles HJ, DirksenU, et al. The CXCR4-CXCL12 axis in Ewing sarcoma: promotion oftumor growth rather than metastatic disease. Clin Sarcoma Res 2012;2:24.

17. Harvey JM, Clark GM, Osborne CK, Allred DC. Estrogen receptorstatus by immunohistochemistry is superior to the ligand-bindingassay for predicting response to adjuvant endocrine therapy in breastcancer. J Clin Oncol 1999;7:1474–81.






18. Don-Salu-Hewage AS, Chan SY, McAndrews KM, Chetram MA,Dawson MR, Bethea DA, et al. Cysteine (C)-X-C receptor 4 undergoestransportin 1-dependent nuclear localization and remains functional atthe nucleus of metastatic prostate cancer cells. PLoS ONE 2013;8:e57194.

19. Xu TP, Shen H, Liu LX, Shu YQ. The impact of chemokine receptorCXCR4 on breast cancer prognosis: a meta-analysis. Cancer Epide-miol 2013;37:725–31.

20. Orimo A, Gupta PB, Sgroi DC, Arenzana-Seisdedos F, Delaunay T,Naeem R, et al. Stromal fibroblasts present in invasive human breastcarcinomaspromote tumorgrowthandangiogenesis throughelevatedSDF-1/CXCL12 secretion. Cell 2005;121:335–48.

21. Liu H, XueW, GeG, Luo X, Li Y, XiangH, et al. Hypoxic preconditioningadvances CXCR4 and CXCR7 expression by activating HIF-1a inMSCs. Biochem Biophys Res Commun 2010:401:509–15.

22. Oh YS, KimHY, Song IC, Yun HJ, Jo DY, Kim S, et al. Hypoxia inducesCXCR4 expression and biological activity in gastric cancer cellsthrough activation of hypoxia-inducible factor-1alpha. Oncol Rep2012;28:2239–46.

23. Valastyan S, Weinberg RA. Tumor metastasis: molecular insights andevolving paradigms. Cell 2011;147:275–92.

24. Benton G, Kleinman HK, George J, Arnaoutova I. Multiple uses ofbasement membrane-like matrix (BME/Matrigel) in vitro and in vivowith cancer cells. Int J Cancer 2011:128:1751–7.

25. Gassmann P, Haier J, Schl€uter K, Domikowsky B, Wendel C, WiesnerU, et al. CXCR4 regulates the early extravasation of metastatic tumorcells in vivo. Neoplasia 2009;11:651–61.

26. Yagi H, Tan W, Dillenburg-Pilla P, Armando S, Amornphimoltham P,Simaan M, et al. A synthetic biology approach reveals a CXCR4-G13-Rho signaling axis driving transendothelial migration of metastaticbreast cancer cells. Sci Signal 2011;4:ra60.

27. Mendoza-Naranjo A, El-Naggar A,Wai DH,Mistry P, LazicN, Ayala FR,et al. ERBB4 confers metastatic capacity in Ewing sarcoma. EMBOMol Med 2013;5:1019–34.

28. Chansky HA, Barahmand-Pour F, Mei Q, Kahn-Farooqi W, Zielinska-Kwiatkowska A, Blackburn M, et al. Targeting of EWS/FLI-1 by RNA

interference attenuates the tumor phenotype of Ewing's sarcoma cellsin vitro. J Orthop Res 2004;22:910–7.

29. Jin Z, Zhao C, Han X, Han Y. Wnt5a promotes Ewing sarcoma cellmigration through upregulating CXCR4 expression. BMC Cancer2012;12:480.

30. Shukla N, Schiffman J, ReedD, Davis IJ,Womer RB, Lessnick SL, et al.Biomarkers in Ewing sarcoma: the promise and challenge of person-alized medicine. a report from the Children's Oncology Group. FrontOncol 2013;3:141.

31. Saunders NA, Simpson F, Thompson EW, Hill MM, Endo-Munoz L,Leggatt G, et al. Role of intratumoural heterogeneity in cancer drugresistance: molecular and clinical perspectives. EMBO Mol Med2012;4:675–84.

32. DeCraeneB,BerxG. Regulatory networks defining EMTduring cancerinitiation and progression. Nat Rev Cancer 2013;13:97–110.

33. Carlisle AJ, Lyttle CA, Carlisle RY, Maris JM. CXCR4 expressionheterogeneity in neuroblastoma cells due to ligand-independent reg-ulation. Mol Cancer 2009;8:126.

34. Zhang L, Yeger H, Das B, Irwin MS, Baruchel S. Tissue microenviron-ment modulates CXCR4 expression and tumor metastasis in neuro-blastoma. Neoplasia 2007;9:36–46.

35. Duda DG, Kozin SV, Kirkpatrick ND, Xu L, Fukumura D, Jain RK.CXCL12 (SDF1alpha)-CXCR4/CXCR7 pathway inhibition: an emerg-ing sensitizer for anticancer therapies? Clin Cancer Res 2011;17:2074–80.

36. Ikegaki N, Shimada H, Fox AM, Regan PL, Jacobs JR, Hicks SL, et al.Transient treatment with epigenetic modifiers yields stable neuroblas-toma stem cells resembling aggressive large-cell neuroblastomas.Proc Natl Acad Sci U S A 2013;110:6097–102.

37. Gupta PB, Fillmore CM, Jiang G, Shapira SD, Tao K, Kuperwasser C,et al. Stochastic state transitions give rise to phenotypic equilibrium inpopulations of cancer cells. Cell 2011;146:633–44.

38. Steeg PS. Perspective: the right trials. Nature 2012;485:S58–9.39. MaM, Ye JY, Deng R, Dee CM, Chan GC. Mesenchymal stromal cells

may enhance metastasis of neuroblastoma via SDF-1/CXCR4 andSDF-1/CXCR7 signaling. Cancer Lett 2011;312:1–10.


Krook et al.




2014;12:953-964. Published OnlineFirst March 20, 2014.Mol Cancer Res Melanie A. Krook, Lauren A. Nicholls, Christopher A. Scannell, et al. SarcomaStress-Induced CXCR4 Promotes Migration and Invasion of Ewing

Updated version

10.1158/1541-7786.MCR-13-0668doi:

Access the most recent version of this article at:

Cited articles

http://mcr.aacrjournals.org/content/12/6/953.full#ref-list-1

This article cites 39 articles, 8 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/12/6/953.full#related-urls

This article has been cited by 3 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://mcr.aacrjournals.org/content/12/6/953To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-13-0668

http://mcr.aacrjournals.org/content/12/6/953.full#ref-list-1

http://mcr.aacrjournals.org/content/12/6/953.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/12/6/953


stress-induced cxcr4 promotes migration and invasion...

Documents