CXCR4,butnotCXCR7,DiscriminatesMetastaticBehaviorin ......exclusively as a decoy receptor (8),...

Cell Cycle and Senescence

CXCR4, but notCXCR7,DiscriminatesMetastatic Behavior inNon–Small Cell Lung Cancer Cells

Young H. Choi, Marie D. Burdick, Brett A. Strieter, Borna Mehrad, and Robert M. Strieter

AbstractChemokines have been implicated as key contributors of non–small cell lung cancer (NSCLC) metastasis.

However, the role of CXCR7, a recently discovered receptor for CXCL12 ligand, in the pathogenesis of NSCLC isunknown. To define the relative contribution of chemokine receptors to migration and metastasis, we generatedhuman lung A549 andH157 cell lines with stable knockdown of CXCR4, CXCR7, or both. Cancer cells exhibitedchemotaxis to CXCL12 that was enhanced under hypoxic conditions, associated with a parallel induction ofCXCR4, but not CXCR7. Interestingly, neither knockdown cell line differed in the rate of proliferation, apoptosis,or cell adherence; however, in both cell lines, CXCL12-induced migration was abolished when CXCR4 signalingwas abrogated. In contrast, inhibition of CXCR7 signaling did not alter cellular migration toCXCL12. In an in vivoheterotropic xenograft model using A549 cells, expression of CXCR4, but not CXCR7, on cancer cells wasnecessary for the development of metastases. In addition, cancer cells knocked down for CXCR4 (or both CXCR4and CXCR7) produced larger and more vascular tumors as compared with wild-type or CXCR7 knockdowntumors, an effect that was attributable to cancer cell–derived CXCR4 out competing endothelial cells for availableCXCL12 in the tumor microenvironment. These results indicate that CXCR4, not CXCR7, expression engagesCXCL12 to mediate NSCLC metastatic behavior.

Implications: Targeting CXCR4-mediated migration and metastasis may be a viable therapeutic option inNSCLC. Mol Cancer Res; 12(1); 38–47. �2013 AACR.

IntroductionNon–small cell lung cancer (NSCLC) is the most com-

monly diagnosed cancer worldwide and is the leading cause ofcancer death (1, 2). Most patients with NSCLC die ofmetastatic disease, typically involving the mediastinal lymphnodes, other areas of the lungs, adrenal glands, brain, liver, andthe bone marrow. Understanding the biology that underliesNSCLC metastasis formation is therefore important.The CXCL12–CXCR4 biological axis is an evolutionarily

preserved mechanism essential to homing of progenitor cellsduring organogenesis, retention of precursor cells in the bonemarrow, and their recruitment to sites of tissue repair. Inaddition to its physiologic roles, this pathway has beenimplicated in metastasis formation in diverse cancers: manyhuman cancer cells, including NSCLC squamous cell andadenocarcinoma, expressCXCR4. In addition, the expression

of CXCR4 on cancer cells is upregulated in hypoxia, as mightbe encountered in the tumor microenvironment (3, 4).CXCL12 is constitutively expressed in tissues that are targetsofmetastases, and a substantial body of literature supports thenotion that interrupting this ligand–receptor interactioninhibits cancer cell migration and metastasis formation (5,6). CXCR4 and CXCL12 were thought to be an exclusiveligand–receptor pair until the discovery of CXCR7, analternative receptor for CXCL12 that also binds CXCL11(7). CXCR7 is expressed on subsets of T and B cells,activated endothelial cells, fetal hepatocytes, as well asmany cancer cells (6). In the context of malignancy, therole of CXCR7 and its functional relationship to CXCR4,CXCL12, and CXCL11 are incompletely defined. Indiverse experimental settings and using different cancermodels, some studies have reported that CXCR7 actsexclusively as a decoy receptor (8), others report that itmediates migration to CXCL12 or CXCL11 (9–11), andstill others indicating that it mediates cancer proliferation,apoptosis or adhesion (12–15), or angiogenesis (15, 16).We sought to define the relative contribution of CXCR4

and CXCR7 in NSCLC migration in vitro and metastasisformation in vivo by generating human A549 and H157 celllines that are stably knocked down for CXCR4, CXCR7, orboth. We report that migration and metastasis of NSCLCcells is dependent on the interaction of CXCL12 withCXCR4, but is independent of the interaction of CXCL11or CXCL12 with CXCR7.

Authors' Affiliation: Division of Pulmonary and Critical Care Medicine,Department of Medicine, University of Virginia School of Medicine, Char-lottesville, Virginia

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

Corresponding Author: Robert M. Strieter, Division of Pulmonary andCritical Care, Department of Medicine, University of Virginia School ofMedicine, Charlottesville, VA 22908. Phone: 434-243-7363; Fax: 434-924-2824; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-12-0334

�2013 American Association for Cancer Research.

MolecularCancer

Research

Mol Cancer Res; 12(1) January 201438

on August 21, 2020. © 2014 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst September 11, 2013; DOI: 10.1158/1541-7786.MCR-12-0334

http://mcr.aacrjournals.org/

Materials and MethodsCell cultureA549 and H157 cell lines were maintained in RPMI 1640

media (Invitrogen) supplementedwith1mmol/L L-glutamine,25 mmol/L HEPES, 1� Pen/Strep, and 10% FBS, here-after referred to as RPMI complete media. In some experi-ments, cells were transferred to RPMI with 1% FBS(hereafter referred to as starvation media) and incubatedin modular incubator chambers (Billups-Rothenberg)under normoxic (21% O2, 74% N2, 5% CO2) or hypoxic(1% O2, 94% N2, 5% CO2) culture conditions. In someexperiments, 10 mg/mL anti-human CXCR4 (MAB170;R&D systems) or CXCR7 polyclonal antibodies wereadded to the media. Parental cell lines were purchasedfrom American Type Culture Collection at the beginningof the project. At the conclusion of the experiments, allparental andmodified cell lines were authenticated by shorttandem repeat DNAprofiling and comparison to publishedATCC profiles (Genetica).

ReagentsPolyclonal rabbit anti-human CXCR7 was produced by

the immunization of a rabbit with a 21-mer peptide(MDLHLFDYAEPGNFSDISWPC) constituting the ami-no terminus of human CXCR7 (hCXCR7). The rabbit wasimmunized in multiple intradermal sites with CFA followedby 3 boosts in IFA. Direct ELISA was used to evaluateantisera titers, and serum was used for Western blot, andneutralization assays when titers had reached greater than1:1,000,000. The specificity of anti-hCXCR7 Ab wasassessed by Western blot analysis against cells expressinghCXCR7 and a panel of other human and murine recom-binant cytokines. Polyclonal goat anti-human CXCL12antibodies were produced as previously described (17).

Flow cytometryCells were stained with anti-hCXCR4 or anti-hCXCR7

antibodies or appropriate controls followed by APC-conju-gated secondary antibodies (A-21463; BD Bioscience). Insome experiments, cells were also stained with CD45-PerCP(557513; BD Bioscience) and rabbit anti-Factor VIII–relat-ed antigen or appropriate control (F3520 and I-8140;Sigma), followed by secondary Alexa Fluor 610-R-phycoer-ythin goat anti-rabbit immunoglobulin G (IgG; A20981;Invitrogen). To quantify metastases, organs were digestedas previously described (3, 18–20). Organs from animalsinoculated with Green fluorescence protein (GFP)-trans-fected cancer cells were compared with organs from healthyanimals, and the proportions of GFP-expressing cells inorgan digests were quantified. Samples were analyzed on afluorescence-activated cell sorting (FACS) Canto II flowcytometer (Becton Dickinson) using FACS Diva software.

Establishment of stable knockdowns of CXCR4,CXCR7, and both CXCR4 and CXCR7 in A549 andH157 NSCLC cell linesGFPwas transfected intoparentalA549andH157cell lines

usingEmGFPvector(K493600;Invitrogen).GFP-transfected

cell lines had comparable migratory capacity to CXCL12 ascomparedwith control cells that were not transfected with thevector. Block-iT Pol II miR RNAi Expression Vector Kit(K494300; Invitrogen)was used to generate knockdowncells.For CXCR4 knockdown cells, humanCXCR4miRNA oligowasdesigned(topstrand,50-TGCTGTTGACTGTGTAGATGA CAT GGG TTT TGG CCA CTG ACT GAC CCATGT CAT ACA CAG TCA A-30; bottom strand, 50-CCTGTT GAC TGT GTA TGA CAT GGG TCA GTC AGTGGCCAAAACCCATGTCATCTACACAGTCAAC-30)using Invitrogen miRNA Designer software. These 2 RNAioligos were annealed (20 mmol/L condition, 95�C for 4minutes) followed by ligation into pcDNA 6.2-GW/miRvector. Ligated product was transformed into TOP10 compe-tentEscherichiacoli.Spectinomycin(50mg/mL)-selectedtrans-formant was grown in Lysogeny broth (LB) media to purifyCXCR4miRNA-containing pcDNA6.2-GW/miR (pcDNA6.2-GW/CXCR4)vector.PurifiedpcDNA6.2-GW/CXCR4vector was mixed with pDONR 221 vector and pLenti6/V5-DEST vector to generate pLenti6/V5-GW/CXCR4 expres-sion construct. Lentiviral stock, containing the packagedpLenti6/V5-GW/CXCR4 expression construct, was pro-duced by cotransfecting the ViraPower Packaging Mix andpLenti6/V5-GW/CXCR4 expression construct into 293FTProducercell line.Transductionofthis lentiviralconstruct intoA549 and H157 cell line was performed in the presenceof polybrene, followed by selection with blasticidin. ForCXCR7knockdowncells,CXCR7miRNAoligowasdesigned(top strand, 50-TGCTGAAGATGAGGTGTGTGACTTTGG TTT TGG CCA CTG ACT GAC CAA AGT CACACC TCA TCT T-30; bottom strand, 50-CCT GAA GATGAG GTG TGA CTT TGG TCA GTC AGT GGC CAAAAC CAA AGT CAC ACA CCT CAT CTT C-30) andprocessed for lentiviral transduction as described earlier. Toproduce cell lines with knockdown of both CXCR4 andCXCR7, CXCR4 knockdown cells were further transfectedwith CXCR7 miRNA followed by selection by reverse tran-scriptase (RT)-PCR. The resulting knockdown cell lines hadgreatlydiminishedexpressionofCXCR4andCXCR7mRNAandproteinascomparedwithnontransfectedparentalcell lines(Supplementary Fig. S1A–S1D). On flow cytometry, bothA549 and H157 cell lines displayed cell surface expression ofCXCR4andCXCR7.CXCR4expressionwasenhancedunderhypoxicconditions aspreviously reported (3)whereasCXCR7expression was unaffected. In both A549 and H157 CXCR4knockdown lines, cell surface expression of CXCR4 wasabolished without detectable effect on CXCR7 under bothnormoxic and hypoxic conditions; similarly, A549 andH157CXCR7 knockdown lines had little detectable cell surfaceexpression of CXCR7 but intact CXCR4 expression. A549and H157 lines with both CXCR4 and CXCR7 knockdownhad little detectable cell surface expression of either moleculeunder normoxic and hypoxic culture conditions (Supplemen-tary Fig. S1E–S1H).

Chemotaxis assaysChemotaxis assays were performed using 12-well chemo-

taxis chambers and 5 mm pore size filters (Neuroprobe)

CXCR4 and CXCR7 in Lung Cancer Metastasis

www.aacrjournals.org Mol Cancer Res; 12(1) January 2014 39




coated with 5 mg/mL fibronectin. Human CXCL12 orCXCL11 (300-46 and 300-28B; Peprotech) were added tothe lower wells and 105 cells added to the upper wells andincubated for 6 hours at 37�C. Filters were then removed,fixed in methanol, stained in 2% Toluidine blue, and thecells that had migrated through to the underside of the filterswas counted in 5 separate fields of view under �400 mag-nification. For transendothelial migration assays, humanlung microvascular endothelial cells were first seeded ontotranswell plate inserts with 8 mm pore size, CXCL12 wasadded to the lower well, and 105 A549 or H157 cells addedto upper wells. After incubation for 6 hours at 37�C,migrated cells were counted under fluorescence microscope.For the competitive chemotaxis assay, 105 human lungmicrovascular endothelial cells were seeded onto 8 mm poresize transwell inserts. The cells were allowed to adhere overnight, the media was changed to starvation media, and 30ng/mL of CXCL12 was added to the lower wells containingthe A549 cells. The inserts were placed in the lower wells andincubated for 6 hours at 37�C. The inserts were thenremoved, fixed in methanol, stained in 2% Toluidine blue,and the cells that had migrated through to the underside ofthe insert was counted in 5 separate fields of view under�400 magnification.

Western blottingImmunoblotting for CXCR4, CXCR7, and factor VIII–

related antigen were performed as previously described (17).

Heterotopic human NSCLC-SCID mouse chimerasWe used a heterotopic model of human NSCLC in severe

combined immunodeficient (SCID)-beige mice, as previous-ly described (17). Age-matched female 6- to 8-week-oldC57BL/6 mice were purchased from Jackson Laboratoriesand were maintained under pathogen-free conditions and incompliance with institutional animal care regulations at theUniversity of Virginia animal care facility. Briefly, 106 A549cells in 100mLwere injected subcutaneously into one flank of6-week-old female CB17-SCID beige mice (Charles River)and tumors size monitored weekly. In some experiments,tumor-bearing SCID mice were injected intraperitoneallywith 500 mL of neutralizing anti-CXCL12 or control serum,containing 10 mg goat IgG, 3 times per week, starting at thetime of xeno-engraftment. This amount of anti-CXCL12serum was found to specifically neutralize 1 mg of murineCXCL12 protein in bioassays without cross-reacting to apanel of other chemokines (17). Animals were euthanized at adesignated times, and organs and blood samples were pro-cessed as described previously (3, 18–20). At the time of tissueharvest, primary tumors were removed from animals, theirorthogonal diameters measured using Thorpe calipers (Bio-medical Research Instruments), and tumor volume calculatedas [volume¼ (d1� d2� d3)� 0.5236], where dn representsorthogonal diameter measurements.

ImmunolocalizationImmunohistochemistry of factor VIII–related antigen,

Ki67 and GFP were performed as previously described

(18, 21). Briefly, paraffin-embedded tumor sections weredeparaffinized, subjected to antigen retrieval (HK080-9K;Biogenex), fixed for 20 minutes in 1:1 absolute methanoland 3% H2O2, rinsed in PBS, and then blocked (HK085-5K; Biogenex) at room temperature for 30 minutes followedby Avidin/Biotin blocking (SP-2001; Vector Laboratories).Then, a 1:800 dilution of anti-factor VIII–related antibody,anti-Ki67 Ab-4 rabbit polyclonal antibody (RB-1510;Thermo Scientific), or control rabbit IgG or goat anti-GFP(Vector Laboratories; SP-0702), or control goat IgG (Sigma;I5356) were added and slides were incubated for 30minutes.Slides were then rinsed with PBS, overlaid with biotinylatedgoat anti-rabbit IgG (PK-6101; Vector Laboratories), andincubated for an additional 30 minutes. Slides were rinsedtwice with PBS, treated with streptavidin-conjugated per-oxidase for 30 minutes, washed 3 times with PBS, incubatedfor 5 to 10 minutes in the substrate chromogen 3,30-diaminobenzidine solution (SK-4100; Vector Laboratories)to allow color development and then rinsed with water.Finally, the slides were counterstained with hematoxylin.For GFP immunofluorescent staining, a 1:50 dilution ofeither mouse IgG or anti-GFP (SC-9996; Santa Cruz) wasadded and slides were incubated for 30 minutes. Slides werethen rinsed with PBS, overlaid with Alexa Fluor 488 anti-mouse IgG (A11001; Invitrogen), and incubated for anadditional 30 minutes. Slides were again washed 2 timeswith PBS, and then coversliped using aqueous mountingmedia. For TUNEL staining, DeadEnd ColorimetricTUNEL System (G7130; Promega) was used, according tomanufacturer specifications. For image analysis, we prepared5 histological sections per tumor, from 5mice per condition,and photographed 10 different fields from each section at�200 magnification using a Zeiss Image A1 microscopewith an inline AxioCam HRc camera interfaced with a PC.Images were captured with AxioVision software (v.4.8) andanalyzed usingNIH ImageJ software. Data were expressed aspercent of positive pixels per field.

Statistical analysisThe animal studies involved 5 to 10 mice bearing A549

tumors for each group. Data were analyzed on a personalcomputer using the Statview 5.0 statistical package (AbacusConcepts). Comparisons were evaluated by theANOVA testwith the post hoc analysis Bonferroni/Dunn. Data wereexpressed as mean � SEM and differences were consideredstatistically significant if P values were less than 0.05.

ResultsCXCR4mediates chemotaxis ofNSCLC cells toCXCL12We began by examining the relative contribution of

CXCR4 and CXCR7 to chemotaxis of A549 and H157cells to varying concentrations ofCXCL12.Under normoxicculture conditions, both cell lines displayed chemotaxis toconcentration of CXCL12 between 10 to 300 ng/mL; thiseffect was significantly attenuated in both cell lines withCXCR4 knockdown and combined CXCR4/7 knockdownbut was unaffected inCXCR7 knockdown lines (Fig. 1A and

Choi et al.

Mol Cancer Res; 12(1) January 2014 Molecular Cancer Research40




D). Under hypoxic culture conditions, chemotaxis of celllines with unmanipulated CXCR4/7 expression to CXCL12was enhanced to concentrations as low as 1 ng/mL; this effectwas again greatly diminished in cell lines knocked down forCXCR4 and CXCR4/7 but was unaffected in lines knockeddown for CXCR7 alone (Fig. 1B and E).To confirm that CXCR4 is necessary for migration of

NSCLC cell lines, we tested the effect of Ab-mediateneutralization of CXCR4 on chemotaxis of the cancer cellsto CXCL12. As observed previously, both A549 and H157lines exhibited brisk chemotaxis to 10 ng/mL CXCL12under hypoxic culture conditions; this effect was diminishedin cell lines knocked down for CXCR4 and CXCR4/7, butnot lines knocked down for CXCR7 alone (Fig. 1C and F).In addition, chemotaxis of A549 and H157 lines withunmanipulated CXCR4/7 expression and those of linesknocked down for CXCR7 was abolished in the presenceof neutralizing anti-CXCR4 antibodies, again demonstrat-ing the critical role of CXCR4 in migration of these cells toCXCL12.

CXCR7 is dispensable for migration of NSCLC cell linesto CXCL11 and CXCL12We used several strategies to assess the contribution of

CXCR7 to migration of NSCLC cells. First, we tested theeffect of a neutralizing anti-CXCR7 antibody on migrationof NSCLC cells in chemotaxis studies. Neutralization ofCXCR7 did not influence the chemotaxis of A549 or H157cells with unmanipulatedCXCR4/7 expression toCXCL12,whereas cell lines knocked down for CXCR4had diminishedchemotaxis to CXCL12, again unaffected by CXCR7 neu-

tralization (Supplementary Fig. S2). We next used a tran-sient siRNA approach to attenuate CXCR7 expression andassess its role in cell migration. Nontransfected A459 andH157 cells and cells with stable CXCR4 knockdown weresubjected to CXCR7 siRNA transfection and were found tohave efficient inhibition of CXCR7 expression (Supplemen-tary Fig. S2B), but showed no change in their migratoryactivities toward CXCL12 (Supplementary Fig. S2C).Because CXCL11 is another ligand for CXCR7, we next

tested the effect of CXCL11 on chemotaxis of NSCLC cells.For both A549 and H157 lines, CXCL11 had no detectableinfluence on the migration of cells under normoxic andhypoxic culture conditions; this finding was unaffected incells with stable knockdowns of CXCR4, CXCR7, or both.Furthermore, concentrations of CXCL11 between 1 and300 ng/mL did not interfere with CXCR4-dependentmigration of A549 and H157 cells induced by 10 ng/mLCXCL12 (Fig. 2). These results suggest that CXCR7 doesnot contribute to the migration of NSCLC cells under ourexperimental conditions.

CXCR4 but not CXCR7 mediates the migration ofNSCLC cells across endothelial barriersA key feature of metastasis formation is the ability of

cancer cells to cross endothelial barriers. We therefore testedthe role of CXCR4 and CXCR7 in migration of NSCLCcells in modified in vitro chemotaxis assays that incorporatemigration across a human microvascular endothelial cellbarrier. As expected, A549 and H157 cells migrated acrossthe endothelial barrier to CXCL12 under normoxic condi-tions and trans-endothelial migration was enhanced under

Figure 1. CXCR4 mediatesmigration ofNSCLCcells. A549 (A–C) and H157 (D–E) cells wereserum-starved and cultured innormoxic or hypoxic cultureconditions for 24 hours beforeassessing chemotaxis to indicatedconcentrations of CXCL12 for 6hours under normoxic (A and D) orhypoxic (B and C and E and F)culture conditions. Results arerepresentative of 3 experiments;data shown representmean�SEMof cells per high-power field ofn ¼ 3 for each group. WT,nontransfected cells; R4KD, cellswith stable knockdown of CXCR4;R7KD, cells with stable knockdownof CXCR7; DbKD, cells with stableknockdown of CXCR4 andCXCR7;�,P < 0.05 in comparison of CXCR4knockdown and CXCR4/CXCR7knockdown cells as comparedwithnontransfected cells; #, P < 0.05compared with chemotaxis toCXCL12 without anti-CXCR4exposure.






hypoxic culture conditions (Fig. 3). In addition,migration ofA549 and H157 cell lines knocked down for CXCR4 andCXCR4/7 toward CXCL12 was significantly diminished,but lines knocked down for CXCR7 alone had comparablemigration to CXCL12 as nontransfected cells (Fig. 3).Similar to findings of chemotaxis in the absence of endo-thelial cells, Ab-mediated neutralization of CXCR4 inhib-ited the migration of A549 and H157 cells with intactCXCR4 expression to CXCL12 (Fig. 3C and F).

CXCR4-deficient NSCLC lines produce large vascularprimary tumors with reduced metastases in vivoTo assess the biological relevance of our in vitro findings,

we tested GFP-transfected A549 cells and A549 cells withstable knockdowns of CXCR4, CXCR7, and CXCR4 andCXCR7 in a heterotropic xenotransplantation animal mod-el.Mice bearing CXCR4 andCXCR4/7 knockdown cancerswere unexpectedly found to have dramatically larger primarytumors after only 4 weeks, necessitating animal euthanasiaper animal welfare guidelines. In contrast, cells with intactCXCR4/CXCR7 expression and CXCR7 knockdown cellsproduced tumors that did not differ significantly in size; thetumors were at the limit of in vivo detection after 4 weeksand, after 9 weeks, were approximately a quarter of the size ofCXCR4 andCXCR4/7 knockdowns tumors at 4weeks (Fig.4). We reasoned that the observed differences in tumor sizemay be the result of differences in the rate of proliferation orapoptosis of cancer cells with disrupted CXCR4 or CXCR7expression. However, knockdown of CXCR4 or CXCR7 inA549 andH157 lines did not result in altered proliferation ofthe cell lines as compared with respective control cell lines,

irrespective of the presence of CXCL12 in the media (Sup-plementary Fig. S3A and S3B). Similarly, rate of apoptosis oftumor cells in response to staurosporine and methotrexatewas unaffected by knockdown of CXCR4, CXCR7, or bothin A549 or H157 cells and was unrelated to the presence ofCXCL12 (Supplementary Fig. S3C and S3D).We next assessed the composition of the in vivo primary

tumors. Because of the dramatic differences in the growth rateof the tumors in different cell types, it was not possible toexamineCXCR4 andCXCR4/7 knockdown cancers beyondweek 4; conversely, primary tumors caused by CXCR7knockdown cells and cells with unmanipulated CXCR4/7expression were at the limit of detection on week 4. Wetherefore harvested the tumors on week 4 from animals withCXCR4 knockdown and CXCR4/7 knockdown tumors andon week 9 from animals with CXCR7 knockdown tumorsand tumors with unmanipulated CXCR4/7 expression.Grossly, tumors produced by CXCR4 andCXCR4/7 knock-down cells were friable and hemorrhagic as compared withthemore compact tumors produced by A549 cells with intactCXCR4/CXCR7 expression and CXCR7 knockdown cells.We also foundCXCR4andCXCR4/7 knockdown tumors tohave a lower proportion of cancer cells (detected on the basisof expression of GFP), but increased proportion of host-derived endothelial cells (defined as CD45-negative cellsexpressing factor VIII-related antigen; Fig. 5A and B). Con-sistent with this finding, CXCR4 and CXCR4/7 knockdowntumors had higher factor VIII–related antigen protein con-tent and lower GFP protein content as compared with cellswith intact CXCR4/CXCR7 expression or CXCR7 knock-down tumors (Fig. 5C). To better assess the vascular mass of

Figure 2. CXCL11 does not interferewith the CXCL12/CXCR4chemotactic behavior of NSCLCcells. A549 (A and B) and H157(C and D) cells were serum-starvedand cultured in normoxic orhypoxic culture conditions for 24hours before chemotaxis assays.Cells were then subjected tochemotaxis assays in response tothe indicated concentrations ofCXCL11 (1–300 ng/mL) with orwithout 10 ng/mL CXCL12 for 6hours under normoxic (A and C) orhypoxic (B and D) cultureconditions. Results arerepresentative of 3 experiments;data shown representmean�SEMof cells per high-power field ofn¼3for each group. �, P < 0.05 incomparison of CXCR4 knockdownand CXCR4/CXCR7 knockdowncells as compared withnontransfected cells.

Choi et al.





the tumors, we also examined the resected tissue histologi-cally. Immunohistochemical localization of endothelial cellsrevealed greatly increased vascularitywithin tumors generatedby CXCR4 knockdown, as compared with nontransfected,A549 cells (Fig. 5D).Histology also revealed that, in additionto being much larger and more vascular, the CXCR4 andCXCR4/7 knockdown tumors did contain many more

necrotic cells thanwild-type andCXCR7knockdown tumorsin vivo.To confirm that the increased size of primary tumors in

CXCR4-knockdown cancers is attributable to the interac-tion of CXCR4 with CXCL12, we also assessed the effect ofCXCL12 immunoneutralization on the size and composi-tion of primary tumors. CXCL12 neutralization resulted inreduced tumor size in CXCR4 and CXCR4/7 knockdowntumors associated with reduction of CD45� F8RAþ endo-thelial cells, but did no affect the size or vascularity ofCXCR7 knockdown tumors or tumors with intact CXCR4andCXCR7 expression (Supplementary Fig. S4). These dataindicate that primary tumors generated by CXCR4 knock-down A549 cells have greatly increased vascular mass, whichcontributes to their increased size; conversely, CXCR7expression has no detectable effect on the size, cellularity,or vascularity of primary tumors.

CXCR4-deficient NSCLC tumors have reducedmetastatic potential in vivoWenext assessed extent of metastases from different cancer

cells by evaluating the proportion of GFP-expressing cells inthe lungs, liver, bone marrow, brain, adrenal glands, andperipheral blood. As expected, the extent of metastasesincreased between 4 and 9 weeks in both A549 cells withintact CXCR4/CXCR7 expression and in CXCR7

Figure 4. Role of CXCR4 and CXCR7 expression on primary A549 tumorsize in a heterotopic xenograft murine cancer model. Data represent themean � SEM of n ¼ 7 to 9 mice per group. �P < 0.05 in comparison ofCXCR4 knockdown and CXCR4/CXCR7 knockdown cells to cells withintact expression of CXCR4 and CXCR7; NS, no statistically significantdifference.

Figure 3. CXCR4 mediates migration of NSCLC cells across endothelial cell barriers. A549 (A–C) and H157 (D–F) cells were serum-starved and cultured innormoxic or hypoxic culture conditions for 24 hours before assessing trans-endothelial migration to indicated concentrations of CXCL12 for 6 hoursunder normoxic (A andD) or hypoxic (B–CandE–F) culture conditions. Results are representative of 3 experiments; data shown representmean�SEMof cellsper high-power field of n ¼ 3 for each group. �P < 0.05 in comparison of CXCR4 knockdown and CXCR4/CXCR7 knockdown cells to cells with intactexpression of CXCR4 and CXCR7; #P < 0.05 compared with chemotaxis to CXCL12 without anti-CXCR4 exposure.






knockdown cells (Fig. 6 and Supplementary Fig. S5). Despitetheir larger size, we found significantly fewer metastatic cellsin cancers caused by A549 cells knocked down for CXCR4and CXCR4/7 as compared with A549 cells knocked downfor CXCR7 alone or cells with intact CXCR4 and CXCR7expression 4 weeks after cancer implantation. Similar to theobservations in vitro (Supplementary Figs. S3A–S3D), wefound no difference in the proportion of proliferating cells inthe tumors based on Ki67 staining (22.5 � 1.9% in A549cancer cells, 20.8� 1.7% in CXCR4 knockdown cells, 19.1� 3.1% in CXCR7-knockdown cells, and 22.8 � 2% inCXCR4 and CXCR7 knockdown cells) or the proportion ofapoptotic cells in the tumors based on TUNEL staining (5.8� 1.2% in A549 cancer cells, 4.7 � 1.0% in CXCR4knockdown cells, 7.0� 0.85% in CXCR7-knockdown cells,and 4.5� 0.92% in CXCR4 and CXCR7 knockdown cells).Given that cancer cell adhesion to matrix is critical tometastasis formation and that CXCR7 had previously beenimplicated in cell adhesion (7, 22), we also examined the roleof CXCR4 and CXCR7 in adhesion of A549 and H157 cellsto plastic in vitro. We found that neither the presence of

CXCL12 nor the expression of CXCR4 or CXCR7 had anydetectable effect on adhesion of either NSCLC line (Supple-mentary Fig. S3E and S3F). Thus, CXCR4 but not CXCR7seems to be critical to metastasis formation in NSCLC.

CXCR4 expression by A549 cells inhibits endothelial cellchemotaxis to CXCL12To explain the unexpected finding of larger and more

vascular primary tumors in the absence of CXCR4, wehypothesized that NSCLC CXCR4 competes with endo-thelial cells for CXCL12 in the niche of primary tumor.Because a major means of clearing chemokines, includingCXCL12, is receptor-mediated cellular uptake, we firstassessed the role of CXCR4 andCXCR7 onA549 in clearingexogenous CXCL12 from the media. We found that tumorcells with intact expression of CXCR4/7 and cells knockeddown for CXCR7 showed rapid clearance of exogenousCXCL12,whereas cells that were knocked down forCXCR4alone or both CXCR4/7 had little detectable clearance ofCXCL12 (Fig. 7A). We next assessed the effect of thisprocess on the in vitro chemotaxis of endothelial cells to

Figure 5. Role of CXCR4 and CXCR7 expression on primary A549 tumor composition and vascularity. GFP-expressing cancer cells (A) and CD45-factor VIIIrelatedantigen (F8RAþ) endothelial cells (B)were quantified in tumor cell suspensionsby flowcytometry. Tumor F8RAandGFPprotein contentwasquantifiedby Western blot and intensity of the corresponding band was quantified by densitometer and normalized to sample b-actin protein (C). Panel D showsrepresentative immunohistochemical data (original magnification�100 for main panels,�400 for insets) of primary tumors of CXCR7 knockdown A549 cellsand cells with intact CXCR4 andCXCR7expression, stainedwith anti-F8RA antibody (right) or the control IgG (left). Bar graphs representmean�SEMof n¼ 5mice per group (A and B) and 3 replicates representative of 3 separate experiments (C). In all panels, R4KD and DbKD were obtained on week 4 andWT and R7KD samples were obtained on week 9. �P < 0.05 in comparison of CXCR4 knockdown and CXCR4/CXCR7 knockdown cells to cells with intactexpression of CXCR4 and CXCR7; NS, no statistically significant difference.

Choi et al.





CXCL12 is the presence of A549 cancer cells. We foundchemotaxis of endothelial cells to CXCL12 was markedlyattenuated in the presence of nontransfected A549 cells andCXCR7 knockdown A549 cells (Fig. 7B). In contrast,chemotaxis of endothelial cells to CXCL12 was not influ-enced by the presence of A549 cells that were knocked downfor CXCR4 or CXCR4/7. These data suggest that cancercell CXCR4 outcompetes endothelial cells for availableCXCL12; as a result, CXCR4 promotes angiogenesis onlyin the absence of cancer cell CXCR4.

DiscussionIn this study, we used NSCLC cell lines with stable

knockdowns of CXCR4, CXCR7, or both receptors toinvestigate the contribution of CXCR4 and CXCR7 toNSCLC cell migration. CXCR4, but not CXCR7, wasfound to dictate the migration of NSCLC cells towardCXCL12 in vitro and to mediate metastases in vivo. Wealso unexpectedly discovered that tumor growth was accel-erated in vivo in the absence of cancer cell CXCR4 expres-sion, which was associated with increased tumor vascularitythat was dependent on CXCL12. Finally, CXCR4 on A549cells gave the cells the ability to outcompete endothelial cellsin binding to CXCL12, indicating that CXCR4 expressionin NSCLC has important roles in cell migration and metas-tasis but paradoxically attenuates vascular mass in the tumormicroenvironment.In the context of cancer biology, the CXCL12/CXCR4

molecular pathway has been shown to regulate themigrationof tumor cells tometastatic sites inmany cancers (23–29). Inaddition, several reports have suggested that a decrease inCXCR4 expression inhibits CXCL12-induced migration ofprostate cancer or lung cancer cells (30–32). The discovery

of CXCR7 as a promiscuous receptor for CXCL11 andCXCL12 ended the previously accepted exclusive associa-tion between CXCL12 and CXCR4. Although CXCR4 andCXCR7 have been reported to play diverse roles in themigration, proliferation and apoptosis of different cell types,our data confirm the exclusive association of CXCL12 withCXCR4 in migration and metastasis of NSCLC cells, andthat CXCR7 or CXCL11 were not involved in this process.In this context, the literature suggests that CXCR7 isinvolved in the migration of some but not other cells:CXCR7 is crucial for proper migration of primordial germcells or of interneurons toward their targets by attainingregulation of CXCL12 distribution or CXCR4 expressionlevel (33, 34). However, in MCF breast cancer cells and inengrafted neural stem cells, CXCR7 does not mediatemigration of these cells (7, 35). In addition, CXCR7 func-tions as a decoy receptor that does not activate G protein-mediated signaling in breast cancer cells and vascular smoothmuscle cells (7, 33). Furthermore, expression of CXCR4 orCXCR7 did not influence apoptosis, proliferation, or adhe-sion, in agreement with our previous report that CXCL12did not significantly affect proliferation or apoptosis ofNSCLC cells in vitro (17).This work also provides new insights into the biological

significance of the CXCL12–CXCR4 axis in NSCLC. Astriking finding in our study was that knockdown of CXCR4in A549 cells resulted in greatly accelerated development ofprimary tumors in SCID mice that was associated withgreatly increased tumor vasculature, suggesting that theabsence of tumor CXCR4 expression resulted in enhancedrecruitment or differentiation of endothelial cells or pro-genitors in the tumor. Indeed, substantial evidence supportsthe contention that CXCL12 is chemotactic to endothelialcells in vitro and is angiogenic in some circumstances

Figure 6. Role of CXCR4 andCXCR7 expression on A549 tumormetastasis. GFP-expressingcancer cells were quantified invarious organ suspensions by flowcytometry. GFP-expressingmetastatic cells were alsodetectable in the brainhistologically (Supplementary Fig.S5). Data represent the mean �SEM of n ¼ 5 mice per group.�, P < 0.05 in comparison to cellswith intact expression of CXCR4and CXCR7 at 4 weeks; #, P < 0.05in comparison to cells with intactexpression of CXCR4 and CXCR7at 9 weeks; NS, no statisticallysignificant difference.






(36–38). In contrast, previous work has demonstrated thatCXCL12 concentrations are very low in CXCR4-expressingcancers and that immunoneutralization of CXCL12 in thissetting attenuates metastases and prolonged survival inanimal models without influencing primary tumor angio-genesis (17, 24, 39). In contrast, this article shows that, whenCXCR4 is absent only on cancer cells but is present onendothelial cells, the abundantCXCL12 in the tumormilieuresults in potent angiogenesis and accelerated tumor growth.Based on these observations and our data showing thatcancer cell CXCR4 is sufficient to clear extracellular

CXCL12 and to inhibit endothelial cell chemotaxis toCXCL12, we propose the following model: In the contextof normal host where CXCR4 is expressed by both malig-nant and noncancerous cells, the tumor microenvironmentis devoid of extracellular CXCL12 because of uptake of theligand by CXCR4 on the cancer cells; as a result, theCXCL12–CXCR4 axis does not participate in recruitmentof endothelial cells and angiogenesis in the tumor, but plays akey role in recruiting cancer cells that have escaped intocirculation to tissues with high levels of CXCL12 expression,such as the lungs, meninges, and bonemarrow. The selectivelack ofCXCR4on cancer cells, however, results in high levelsof CXCL12 in the tumor microniche, resulting in robustrecruitment of endothelial cells, enhanced angiogenesis, anddevelopment of large, vascular tumors. Interestingly, inaddition to being larger and more vascular, CXCR4 knock-down tumors were found to containmore areas of necrosis invivo. Because CXCR4 knockdown cancer cells did notdisplay altered proliferation or death in vitro, we speculatethat increased necrosis may be the result of formation ofabnormal microvasculature. Alternatively, it is possible thatthe increased necrosis is an indirect consequence of alteredtumor microenvironment in vivo.In summary, our results demonstrate that CXCL12 and

CXCR4 have a single ligand–single receptor relationship inmediating NSCLC migration and metastasis, and in thecontext of NSCLC that expresses CXCR4, the CXCR4–CXCL12 axis is not involved in angiogenesis of the primarytumor.

Disclosure of Potential Conflicts of InterestR.M. Strieter is employed as a Global Head of Translational Medicine for

Respiratory Diseases in Novartis Institutes of Biomedical Research. No potentialconflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: R.M. StrieterDevelopment of methodology: Y.H. Choi, M.D. Burdick, R.M. StrieterAcquisition of data (provided animals, acquired and managed patients, providedfacilities, etc.): Y.H. Choi, M.D. Burdick, B.A. Strieter, R.M. StrieterAnalysis and interpretation of data (e.g., statistical analysis, biostatistics, compu-tational analysis): Y.H. Choi, M.D. Burdick, B. Mehrad, R.M. StrieterWriting, review, and/or revision of the manuscript: Y.H. Choi, B. Mehrad,R.M. StrieterStudy supervision: R.M. Strieter

Grant SupportThis work is supported by the National Heart, Lung, and Blood Institute

HL073848 (B. Mehrad), HL098329 (B. Mehrad, R.M. Strieter), HL066027 andHL098526 (R.M. Strieter).

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 June 6, 2012; revised August 5, 2013; accepted August 6, 2013;published OnlineFirst September 11, 2013.

References1. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates

ofworldwide burden of cancer in 2008:GLOBOCAN2008. Int JCancer2010;127:2893–917.

2. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA CancerJ Clin 2012;62:10–29.

3. Phillips RJ, Mestas J, Gharaee-Kermani M, Burdick MD, Sica A,Belperio JA, et al. Epidermal growth factor and hypoxia-inducedexpression of CXC chemokine receptor 4 on non–small cell lungcancer cells is regulated by the phosphatidylinositol 3-kinase/PTEN/AKT/mammalian target of rapamycin signaling pathway and

Figure 7. Effect of NSCLC cell expression of CXCR4 on CXCL12. (A)concentrationofCXCL12 in the supernatantwasmeasuredafter 24hoursculture of A549 cells in hypoxic conditions in the presence of 30 ng/mLCXCL12 (represented by the dashed line). (B) after 24 hours culture inhypoxic conditions, chemotaxis of humanmicrovascular endothelial cells(placed in the upper chambers of transwell plates) to 10 ng/mLCXCL12 inthe lower chambers was quantified, in the presence or absence of A549cells in the lower chambers. Results fromeachpanel are representative of3 experiments; data shown represent mean � SEM of n ¼ 3 for eachgroup. �, P < 0.05 in comparison to R4- and double-knockdown cells (A)and compared with chemotaxis of endothelial cells in the absence ofA549 cells (B); NS, no statistically significant difference in comparison tochemotaxis of endothelial cells in the absence of A549 cells.

Choi et al.





activation of hypoxia inducible factor-1a. J Biol Chem 2005;280:22473–81.

4. Schioppa T, Uranchimeg B, Saccani A, Biswas SK, Doni A, RapisardaA, et al. Regulation of the chemokine receptor CXCR4 by hypoxia.J Exp Med 2003;198:1391–402.

5. Burger JA, Kipps TJ. CXCR4: a key receptor in the crosstalk betweentumor cells and their microenvironment. Blood 2006;107:1761–7.

6. Sun X, Cheng G, Hao M, Zheng J, Zhou X, Zhang J, et al. CXCL12/CXCR4/CXCR7 chemokine axis and cancer progression. CancerMetastasis Rev 2010;29:709–22.

7. Burns JM, Summers BC, Wang Y, Melikian A, Berahovich R, Miao Z,et al. A novel chemokine receptor for SDF-1 and I-TAC involved in cellsurvival, cell adhesion, and tumor development. J Exp Med 2006;203:2201–13.

8. Luker KE, Steele JM, Mihalko LA, Ray P, Luker GD. Constitutive andchemokine-dependent internalization and recycling of CXCR7 inbreast cancer cells todegrade chemokine ligands.Oncogene2010;29:4599–610.

9. Grymula K, Tarnowski M, Wysoczynski M, Drukala J, Barr FG, Ratajc-zak J, et al. Overlapping and distinct role of CXCR7-SDF-1/ITAC andCXCR4-SDF-1 axes in regulating metastatic behavior of human rhab-domyosarcomas. Int J Cancer 2010;127:2554–68.

10. Miekus K, Jarocha D, Trzyna E, Majka M. Role of I-TAC-bindingreceptorsCXCR3 andCXCR7 in proliferation, activation of intracellularsignaling pathways and migration of various tumor cell lines. FoliaHistochem Cytobiol 2010;48:104–11.

11. Tarnowski M, Liu R, Wysoczynski M, Ratajczak J, Kucia M, RatajczakMZ.CXCR7: a newSDF-1-binding receptor in contrast to normalCD34(þ) progenitors is functional and is expressed at higher level in humanmalignant hematopoietic cells. Eur J Haematol 2010;85:472–83.

12. Hattermann K, Held-Feindt J, Lucius R, M€uerk€oster SS, Penfold ME,Schall TJ, et al. The chemokine receptor CXCR7 is highly expressed inhuman glioma cells and mediates antiapoptotic effects. Cancer Res2010;70:3299–308.

13. Miao Z, Luker KE, Summers BC, Berahovich R, Bhojani MS, Rehem-tulla A, et al. CXCR7 (RDC1) promotes breast and lung tumor growth invivo and is expressed on tumor-associated vasculature. Proc NatlAcad Sci U S A 2007;104:15735–40.

14. Singh RK, Lokeshwar BL. The IL-8-regulated chemokine receptorCXCR7 stimulates EGFR signaling to promote prostate cancer growth.Cancer Res 2011;71:3268–77.

15. ZhengK, Li HY, SuXL,WangXY, Tian T, Li F, et al. Chemokine receptorCXCR7 regulates the invasion, angiogenesis and tumor growth ofhuman hepatocellular carcinoma cells. J Exp Clin Cancer Res 2010;29:31.

16. Kollmar O, Rupertus K, Scheuer C, Nickels RM, Haberl GC, Tilton B,et al. CXCR4 and CXCR7 regulate angiogenesis and CT26.WT tumorgrowth independent from SDF-1. Int J Cancer 2010;126:1302–15.

17. Phillips RJ, Burdick MD, Lutz M, Belperio JA, Keane MP, Strieter RM.The stromal derived factor-1/CXCL12-CXC chemokine receptor 4biological axis in non–small cell lung cancer metastases. Am J RespirCrit Care Med 2003;167:1676–86.

18. AddisonCL, Daniel TO, BurdickMD, Liu H, Ehlert JE, Xue YY, et al. TheCXC chemokine receptor 2, CXCR2, is the putative receptor for ELRþCXC chemokine-induced angiogenic activity. J Immunol 2000;165:5269–77.

19. Arenberg DA, Kunkel SL, Polverini PJ, Glass M, Burdick MD, StrieterRM. Inhibition of interleukin-8 reduces tumorigenesis of human non–small cell lung cancer in SCID mice. J Clin Invest 1996;97:2792–802.

20. Arenberg DA, Kunkel SL, Polverini PJ, Morris SB, Burdick MD, GlassMC, et al. Interferon-gamma-inducible protein 10 (IP-10) is an angio-static factor that inhibits human non–small cell lung cancer (NSCLC)tumorigenesis and spontaneous metastases. J Exp Med 1996;184:981–92.

21. Keane MP, Arenberg DA, Lynch JP 3rd, Whyte RI, Iannettoni MD,Burdick MD, et al. The CXC chemokines, IL-8 and IP-10, regulateangiogenic activity in idiopathic pulmonary fibrosis. J Immunol1997;159:1437–43.

22. Dai X, Tan Y, Cai S, Xiong X, Wang L, Ye Q, et al. The role of CXCR7 onthe adhesion, proliferation and angiogenesis of endothelial progenitorcells. J Cell Mol Med 15:1299–309.

23. Geminder H, Sagi-Assif O, Goldberg L, Meshel T, Rechavi G, Witz IP,et al. A possible role for CXCR4 and its ligand, the CXC chemokinestromal cell-derived factor-1, in the development of bone marrowmetastases in neuroblastoma. J Immunol 2001;167:4747–57.

24. 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.

25. Pan J, Mestas J, Burdick MD, Phillips RJ, Thomas GV, Reckamp K,et al. Stromal derived factor-1 (SDF-1/CXCL12) and CXCR4 in renalcell carcinoma metastasis. Mol Cancer 2006;5:56.

26. ScottonCJ,Wilson JL, Scott K, StampG,WilbanksGD, Fricker S, et al.Multiple actions of the chemokine CXCL12 on epithelial tumor cells inhuman ovarian cancer. Cancer Res 2002;62:5930–8.

27. Smith MC, Luker KE, Garbow JR, Prior JL, Jackson E, Piwnica-WormsD, et al. CXCR4 regulates growthof both primary andmetastatic breastcancer. Cancer Res 2004;64:8604–12.

28. Taichman RS, Cooper C, Keller ET, Pienta KJ, Taichman NS,McCauley LK. Use of the stromal cell-derived factor-1/CXCR4pathway in prostate cancer metastasis to bone. Cancer Res 2002;62:1832–7.

29. Vandercappellen J, Van Damme J, Struyf S. The role of CXC chemo-kines and their receptors in cancer. Cancer Lett 2008;267:226–44.

30. Hong JS, Pai HK, Hong KO, Kim MA, Kim JH, Lee JI, et al. CXCR-4knockdown by small interfering RNA inhibits cell proliferation andinvasion of oral squamous cell carcinoma cells. J Oral Pathol Med2009;38:214–9.

31. Xing Y, Liu M, Du Y, Qu F, Li Y, Zhang Q, et al. Tumor cell-specificblockade of CXCR4/SDF-1 interactions in prostate cancer cells byhTERT promoter induced CXCR4 knockdown: a possible metastasispreventing and minimizing approach. Cancer Biol Ther 2008;7:1839–48.

32. Yang J, Zhang B, Lin Y, Yang Y, Liu X, Lu F. Breast cancer metastasissuppressor 1 inhibits SDF-1alpha-inducedmigration of non–small celllung cancer by decreasing CXCR4 expression. Cancer Lett 2008;269:46–56.

33. Boldajipour B, Mahabaleshwar H, Kardash E, Reichman-Fried M,Blaser H, Minina S, et al. Control of chemokine-guided cell migrationby ligand sequestration. Cell 2008;132:463–73.

34. Sanchez-Alcaniz JA, Haege S, Mueller W, Pla R, Mackay F, Schulz S,et al. Cxcr7 controls neuronal migration by regulating chemokineresponsiveness. Neuron 2011;69:77–90.

35. Carbajal KS, Schaumburg C, Strieter R, Kane J, Lane TE. Migration ofengrafted neural stem cells is mediated by CXCL12 signaling throughCXCR4 in a viral model of multiple sclerosis. Proc Natl Acad Sci U S A2010;107:11068–73.

36. Salvucci O, Yao L, Villalba S, Sajewicz A, Pittaluga S, Tosato G.Regulation of endothelial cell branching morphogenesis by endoge-nous chemokine stromal-derived factor-1. Blood 2002;99:2703–11.

37. Tachibana K, Hirota S, Iizasa H, Yoshida H, Kawabata K, Kataoka Y,et al. The chemokine receptor CXCR4 is essential for vascularization ofthe gastrointestinal tract. Nature 1998;393:591–4.

38. Takabatake Y, Sugiyama T, Kohara H, Matsusaka T, Kurihara H, KoniPA, et al. The CXCL12 (SDF-1)/CXCR4 axis is essential for the devel-opment of renal vasculature. J Am Soc Nephrol 2009;20:1714–23.

39. Schrader AJ, Lechner O, Templin M, Dittmar KE, Machtens S, MengelM, et al. CXCR4/CXCL12 expression and signalling in kidney cancer.Br J Cancer 2002;86:1250–6.






2014;12:38-47. Published OnlineFirst September 11, 2013.Mol Cancer Res Young H. Choi, Marie D. Burdick, Brett A. Strieter, et al. Small Cell Lung Cancer Cells

−CXCR4, but not CXCR7, Discriminates Metastatic Behavior in Non

Updated version

10.1158/1541-7786.MCR-12-0334doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2013/09/11/1541-7786.MCR-12-0334.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/12/1/38.full#ref-list-1

This article cites 38 articles, 17 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/12/1/38.full#related-urls

This article has been cited by 1 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/12/1/38To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-12-0334

http://mcr.aacrjournals.org/content/suppl/2013/09/11/1541-7786.MCR-12-0334.DC1

http://mcr.aacrjournals.org/content/12/1/38.full#ref-list-1

http://mcr.aacrjournals.org/content/12/1/38.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/12/1/38


CXCR4,butnotCXCR7,DiscriminatesMetastaticBehaviorin ......exclusively as a decoy receptor (8),...

Documents

Transcript of CXCR4,butnotCXCR7,DiscriminatesMetastaticBehaviorin ......exclusively as a decoy receptor (8),...