up-regulation of cxcr4 expression in pc-3 cells by stromal- … · up-regulation of cxcr4...

Up-regulation of CXCR4 Expression in PC-3 Cells by Stromal-

Derived Factor-1A (CXCL12) Increases Endothelial Adhesion

and Transendothelial Migration: Role of MEK/ERK Signaling

Pathway–Dependent NF-KB Activation

Promil Kukreja,1Asim B. Abdel-Mageed,

2Debasis Mondal,

1Kai Liu,

2and Krishna C. Agrawal

1

Departments of 1Pharmacology and 2Urology, Tulane University Health Sciences Center, New Orleans, Louisiana

Abstract

The chemokine stromal-derived factor-1A (SDF-1A/CXCL-12)and its receptor, CXCR4, play a crucial role in adhesion andtransendothelium migration (TEM) of prostate cancer cells.We tested the hypothesis that enhanced expression of CXCR4in prostate cancer cells is dependent upon SDF-1A-mediatedactivation of nuclear factor-KB (NF-KB). SDF-1A increasedthe CXCR4 mRNA and protein expression in PC-3 cells butnot in LNCaP cells. Similarly, SDF-1A enhanced the NF-KB-dependent transcriptional activity in PC-3 cells but not inLNCaP cells. SDF-1A increased PC-3 cell adhesion to the hu-man umbilical vein endothelial cell monolayer and enhancedTEM, which was abrogated with anti-CXCR4 monoclonalantibody (mAb). Suppression of NF-KB activity in PC-3 cellsby a mutant IKBA super-repressor adenoviral vector de-creased the CXCR4 mRNA expression and inhibited adhesionand TEM. Transient overexpression of p65 subunit of NF-KBin PC-3 cells up-regulated CXCR4 receptor expression andincreased the adhesion and TEM of these cells in responseto SDF-1A gradient. Treatment of PC-3 cells with SDF-1Aleads to nuclear translocation of NF-KB protein within 15 to30 minutes, which correlated with IKBA phosphorylation.A p42/44 mitogen-activated protein kinase [MAPK, extracel-lular signal regulated kinase-1/2 (ERK-1/2)] biphasic acti-vation pattern was observed in these cells at 15 minutesand 3 hours after SDF-1A treatment. Phosphorylation ofIKB kinase A was observed within 30 minutes, which wasblocked by PD98059 [MAPK kinase (MEK) inhibitor]. PD98059cotreatment significantly inhibited SDF-1A-induced NF-KBreporter activity and CXCR4 receptor expression as shownby flow cytometry. These data suggest that SDF-1A-inducedexpression of CXCR4 in PC-3 cells is dependent on MEK/ERKsignaling cascade and NF-KB activation. (Cancer Res 2005;65(21): 9891-8)

Introduction

Prostate cancer is the second leading cause of cancer-relateddeaths in American males and the incidence of prostate canceris projected to increase significantly. In the United States, it isestimated that >232,090 men will be diagnosed with prostatecancer in 2005, and approximately >30,350 afflicted men will die

of this disease (1). The symptomatic phase of prostate cancer islargely due to occurrence of metastasis to the bone, which is acomplex and multistaged process. For disseminated prostatecancer cells to grow into overt metastasis in the bone, they mustsurvive and proliferate in the vasculature or in the surroundingtissue after extravasation. To exit the vasculature, the prostatecancer cells must first adhere to the endothelium and subsequentlymove through the endothelial monolayer and underlying connec-tive tissues.It has been shown that SDF-1a (CXCL12) produced by bone

marrow stromal cells and microvessel endothelial cells canfacilitate the migration of hematopoietic progenitor cells (HPC)which express the chemokine receptor CXCR4 (2). CXCR4 is alsofound to be expressed on the surface of prostate cancer cells andmay be involved in the metastasis process as well (3). In vitroadhesion assays have shown that pretreatment of prostate cancercells with SDF-1a significantly increases their adhesion toosteosarcoma and bone marrow–derived endothelial cells (4). Inthis context, SDF-1a and CXCR4 seem critical molecular determi-nants for these events (5, 6). CXCR4 expression varies in differentprostate cancer cell lines, and high levels of CXCR4 are linked toaggressive phenotypes in prostate cancer cells (7). Because SDF-1ais constitutively produced by stromal compartments in differenttissues, CXCR4 expression on neoplastic cells may be important intumor metastasis to different organs, similar to the migrationpotential of HPCs in different tissue compartments (8, 9). Signalingmolecule associated with SDF-1a response is linked to theG-protein-coupled CXCR4 receptor, which activates the mitogen-activated protein kinases (MAPK) extracellular signal regulatedkinase-1/2 (ERK-1/2), as well as protein kinase B, phosphatidyli-nositol 3-kinase (PI3K; refs. 10–13).Although the convergence of signal transduction pathways

following CXCR4 activation and stimulation of transcriptionfactors in prostate cancer cells have not been delineated, thepossible role of downstream activation of nuclear factor-nB(NF-nB) has been implicated from previous studies. NF-nB isconstitutively active and expressed at high levels in metastaticprostate cancer cells. NF-nB is involved in the regulation of genesassociated with pathologic processes like inflammation andcancer (14). Specific kinases, InB kinase (IKK), phosphorylatethe inhibitory a subunit, InBa at serine residues 32 and 36, whichis then ubiquitinylated and targeted for proteolysis (15). Thisallows the NF-nB subunits to dimerize and translocate to thenucleus where it can regulate the prometastatic genes. Indeed,NF-nB has been shown to increase the expression of CXCR4 inMCF-7 and MDA-MB-231 breast cancer cells (16). It is alsoknown that during fetal life and during marrow transplantation,HPCs ‘‘home’’ to the bone (17, 18) and SDF-1a expression in the

Requests for reprints: Krishna C. Agrawal, Department of Pharmacology, SL-83,Tulane University Health Sciences Center, New Orleans, LA 70112. Phone: 504-988-5444; Fax: 504-988-5283; E-mail: [email protected].

I2005 American Association for Cancer Research.doi:10.1158/0008-5472.CAN-05-1293

www.aacrjournals.org 9891 Cancer Res 2005; 65: (21). November 1, 2005

Research Article

Research. on November 14, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

osteoblasts and marrow endothelial cells function as chemo-attractant (19–21). It is therefore likely that metastatic prostatecancer may also use a similar pathway to localize to the bonemarrow. In this study, we have shown that NF-nB activation isdirectly linked to up-regulation of the expression of CXCR4 inPC-3 cells. This report also delineates the signal transductionmechanism involved in SDF-1a-mediated activation of NF-nB andimplicates a role of differential signaling in two different prostatecancer cell lines, the bone metastatic PC-3 and lymph nodemetastatic LNCaP cells, which may help elucidate the metastaticpotential of certain prostate cancer cell clones.

Materials and Methods

Reagents. Recombinant human SDF-1a was obtained from Research

Diagnostics, Inc. (Flanders, NJ). PD98059 was purchased from Calbiochem(La Jolla, CA) and LY-294002 was purchased from Cell Signaling Technology,

Inc. (Beverly, MA). Cycloheximide was obtained from Sigma (St. Louis, MO).

Antibody to human CXCR4 was obtained from Santa Cruz Biotechnology

(Santa Cruz, CA). Rabbit anti-p65NF-nB, anti-InBa, anti-IKK, anti-p42/44MAPK, anti-phospho-IKK, and anti-phospho-InBa were purchased from

Cell Signaling Technology. Calcein AM dye was obtained from Molecular

Probes (Eugene, OR). The reverse transcription-PCR (RT-PCR) primers forCXCR4 (12) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were

synthesized from Midland Certified Reagent (Midland, TX). The mutant

InB (mInB) super-repressor adenoviral construct was a generous gift of

Dr. Inder Verma (Salk Institute, CA). The human p65 and p50 expressionplasmids subcloned in pcDNA and NF-nB luciferase (pNF-nB-Luc) and

control luciferase (pFR-Luc) reporter plasmids were purchased from

Stratagene (La Jolla, CA).

Cell culture. PC-3 and LNCaP cells were obtained from the AmericanType Culture Collection (Manassas, VA). PC-3 cells were grown in Ham’s

F-12 and LNCaP cells were maintained in RPMI 1640. Media were

supplemented with 10% fetal bovine serum, 1% penicillin-streptomycin,

and 1% l-glutamine. The human umbilical vein endothelial cells (HUVEC)were obtained from Clonetics (BioWhittaker, Inc., Walkersville, MD) and

were grown and subcultured in the specified medium (EBM-2) according to

the manufacturer’s protocol. All cells were grown in a humidified incubatorat 37jC and 5% CO2.

Fluorescent labeling of prostate cancer cells. The transient

fluorescent labeling of prostate cancer cells with calcein AM was done

by using the Vybrant cell adhesion assay protocol (Molecular Probes).Briefly, prostate cancer cells were washed with PBS and resuspended

in RPMI without serum at 5 � 106 cells/mL. Cells were incubated

with the calcein AM solution (5 AL/mL) for 30 minutes at 37jC. Cellswere washed with RPMI, and fluorescence (250 AL) was determined usinga Flx-800 microplate fluorescence reader (Bio-Tek Instruments, Inc.,

Winooski, VT), with wavelengths set at absorbance maximum of

494 nm and emission maximum of 517 nm. The fluorescent-labeledprostate cancer cells were used in adhesion and TEM experiments with

the HUVECs.

Adhesion assay. The adhesion of prostate cancer cells to HUVECs was

monitored in two-cell coculture model in Petri plates. The HUVECs weregrown in 24-well culture plates until they reached 90% confluency. The

confluency of monolayer was verified by Dif-quik staining of the cells. The

adhesion assays were carried out by addition of fluorescent-labeled (calcein

AM) prostate cancer cells at a 1:1 ratio to HUVECs in quadruplicatecultures. Where stated, the prostate cancer cells were pretreated with SDF-

1a (0-200 ng/mL) for 6 hours and adhesion was allowed for 30 minutes at

37jC. The nonadherent prostate cancer cells were thoroughly washed off

using four washes with PBS, and the plates containing adhered prostatecancer cells were quantified

Transendothelial migration assay. The TEM assay was conducted

in 12-well transwell culture plates containing microporous (3.0 Am)membranes (Corning Costar Corp., Cambridge, MA). Briefly, HUVECs

were seeded onto transwell culture plates and confluency of monolayer

was verified by Dif-quik staining. Prostate cancer cells were added tothe upper chamber and transmigration towards an SDF-1a gradient in

the lower chamber was monitored using calcein-labeled prostate cancer

cells. In some experiments, prostate cancer cells were pretreated with

SDF-1a (100 ng/mL) for 6 hours or transfected with NF-nB expressionplasmids.

Reverse transcription-PCR. Total RNAs were isolated using the Trizol

method (Life Technologies, Carlsbad, CA). PCR for CXCR4 was done using

an assay kit from Promega (Madison, WI), and a thermal cycler from

Perkin-Elmer (Shelton, CT; model 9600). Total RNAs (f2 Ag) were used ineach reverse transcription reaction mixture (50 AL), containing MMLV-RT

(0.5 unit), RNase inhibitor (0.1 unit), oligo-dT (1 Ag), and deoxynucleotides

(deoxynucleotide triphosphates, 1 mmol/L each), and incubated for 30 to

60 minutes at 42jC. The PCR amplifications were carried out at the

conditions specified for each primer pair, by using Taq DNA polymerase

(0.5 unit/100 AL). The cycling conditions of 94jC for 1 minute, 55jC for1 minute, and 72jC for 1 minute for 30 cycles followed by a 10-minute

extension at 72jC were used for CXCR4. The PCR products were viewed by

electrophoresis on a 2.0% agarose gel containing ethidium bromide, along

with a DNA molecular weight marker (Boehringer Mannheim, Mannheim,

Germany). Band intensities were monitored by a Bio-Rad (Hercules, CA)UV Gel Documentation system, and a semiquantitative analysis was done

by comparing relative band intensities of GAPDH.

Western immunoblot. For Western blot analysis, cells were starved in

serum-free Kaighn’s modified Ham’s F-12 for 12 hours and stimulated with100 ng/mL SDF-1a for the indicated period of time at 37jC. The treated

cells were lysed in 200 AL of lysis buffer and fractionated by a 10% SDS-

PAGE gel. Proteins were transferred onto a nitrocellulose membrane by

electrotransfer for 1 hour. After soaking in blocking buffer (1� TBS, 0.1%Tween 20, and 5% casein), membranes were incubated with primary

antibody overnight at 4jC. Blots were developed using horseradish

peroxidase–linked secondary antibody and a chemiluminescent detection

system (LumiGLO, KPL, Gaithersburg, MD).Flow cytometric analysis. To monitor the CXCR4 protein levels, PC-3

and LNCaP cells were fixed in 4% formaldehyde diluted in PBS and

permeabilized using 0.1% Triton X-100 in PBS. FITC-conjugated anti-CXCR4antibody (1:400) was added and incubated at 4jC for 1 hour. Cells were

washed in PBS and resuspended in 500 AL PBS before analysis using a

Beckman Coulter FACSCalibur (Miami, FL). A total of 10,000 events were

analyzed.Transient transfection and adenoviral transduction. Cells were

transiently transfected with p65, p50, or pCMV-CXCR4 using electro-

poration technique (Gene Pulser, Bio-Rad). At 48 hours after transfection,

RNA was isolated for RT-PCR and flow cytometric was carried forcell surface CXCR4 expression. NF-nB activity was assessed by cotransfect-

ing pNF-nB-Luc and pCMV h-galactosidase. The luciferase activity was

normalized to the h-galactosidase activity. For adenoviral transduction,cells were incubated for 24 hours with either an adenovirus expression

vector for the dominant-negative InB super-repressor (pAxCAmInB-M) or

the wild-type (pAxCAInBa) construct (multiplicity of infection = 100).

Transduction efficiency was monitored by an Adv-h-galactosidase re-porter construct followed by in situ staining of the percentage of blue

cells.

Statistical analysis. All experiments were done in duplicate or triplicate

cultures. The results obtained from at least three independent experimentsare expressed as FSE. Statistical analysis was done using the Statview

software. The two-tailed Student’s t test was used to determine whether the

samples are representative of a known experimental condition, and one-way

ANOVA was used to compare among three or more groups. P < 0.05 wasconsidered significant.

Results

Stromal-derived factor-1A stimulation induced CXCR4expression in PC-3 cells. We monitored the levels of CXCR4transcripts in PC-3 and LNCaP cells with or without SDF-1astimulation by semiquantitative RT-PCR assays. The data showed

Cancer Research

Cancer Res 2005; 65: (21). November 1, 2005 9892 www.aacrjournals.org



that PC-3 cells constitutively expressed 2- to 3-fold higher levelsof CXCR4 mRNA compared with LNCaP cells (Fig. 1A). Within6 hours following SDF-1a treatment, CXCR4 transcript levelsincreased by 4- to 6-fold in PC-3 cells but not in LNCaP cells. Theenhanced CXCR4 gene expression returned to basal level within24 hours after stimulation. However, even at this time point, ahigher level of CXCR4 expression was observed in PC-3 cellscompared with LNCaP cells. These findings suggested thatincreased CXCR4 expression in PC-3 cells, both constitutive andinducible, may play a direct role in the enhanced levels of CXCR4expression observed in PC-3 cells. The protein expression ofCXCR4 receptor was monitored using flow cytometry (Fig. 1B).Similar to the RT-PCR data, PC-3 cells expressed a higher (6- to8-fold) constitutive level of CXCR4 protein compared with LNCaPcells. Exposure to SDF-1a for 12 hours showed a 4- to 6-foldincrease in CXCR4 protein levels in PC-3 cells compared withuntreated cells, whereas in LNCaP cells, there was no significantchange in CXCR4 expression.Stromal-derived factor-1A–induced CXCR4 expression

increases human umbilical vein endothelial cells adhesionand transendothelial migration of PC-3 cells. For the adhesionstudies, PC-3 and LNCaP cells were treated with increasingconcentrations of SDF-1a for 6 hours and cocultured withHUVECs grown in 24-well culture plates for 30 minutes. Non-adherent prostate cancer cells were removed and percent changein prostate cancer cell adhesion was monitored by fluorimetricanalysis to determine the effects of SDF-1a on adhesion ofprostate cancer cells to HUVECs. We observed a concentration-dependent inductive effect of SDF-1a (0-200 ng/mL) uponexposure of PC-3 cells but not of LNCaP cells (Fig. 1C). Even inthe absence of stimulation, PC-3 cells were more adherent toHUVECs compared with LNCaP cells. Exposure to SDF-1a showeda significant (P < 0.05) increase (175-200%) in adhesion of PC-3cells; however, adhesion of LNCaP cells was not significantlychanged following their SDF-1a stimulation. The SDF-1a-induced(100 ng/mL) adhesion could be significantly blocked by coincu-bation of PC-3 cells with mAb to CXCR4. A slight decrease in basallevel of PC-3 adhesion was also observed in the presence ofantibody to CXCR4.We monitored the effects of SDF-1a-induced CXCR4 expression

in PC-3 cell TEM via a confluent HUVEC monolayer (Fig. 1D). Theunstimulated or SDF-1a prestimulated (6 hours) PC-3 cells werefluorescent labeled and cultured with HUVEC in transwell cultureplates, and percent change in TEM were monitored after 6 hoursof coincubation. To observe the transmigratory potential of achemokine gradient, SDF-1a was added to the lower chamber. TheSDF-1a-stimulated cells showed a slight increase in TEM. Increasein the percentage of unstimulated PC-3 cells transmigratingtowards an SDF-1a concentration gradient was also observed.However, the combined exposure of PC-3 cells to SDF-1a andpresence of the chemokine in the lower chamber significantlyincreased TEM. Interestingly, a proportional increase in transmi-gration was not observed when SDF-1a was present in the upperchamber of the transwell. Furthermore, the increase in TEM wassuppressed upon incubation of the coculture with CXCR4 anti-bodies added to the upper chamber.Suppression of nuclear factor-KB activity decreased both

CXCR4 expression and endothelial adhesion of PC-3 cells.NF-nB activity is closely related to regulation of critical genesinvolved in cancer metastases (22). Overexpression of the trans–dominant-negative mutant subunit of the inhibitory component

of NF-nB (InBa) suppresses nuclear translocation of the DNA-binding subunits of NF-nB and abrogates NF-nB-mediatedsignaling in different cell types (23). To monitor the role ofNF-nB in CXCR4 expression in prostate cancer cells, weintroduced adenoviral vectors coding for either the wild type

Figure 1. A, SDF-1a-induced CXCR4 gene expression in PC-3 and LNCaPcells. Cells were incubated with SDF-1a for the indicated periods of time. Theprostate cancer cells were serum starved for 12 hours and treated with SDF-1a(100 ng/mL) for 6 and 24 hours. Representative of three independentexperiments. B, flow cytometric analysis of CXCR4 receptor expression inprostate cancer cell lines. Staining with isotype control mouse IgG2b (1 and 2)and FITC-labeled anti-CXCR4 mAb (3-6 ) is shown in histograms for PC-3 andLNCaP cell lines. The prostate cancer cells were treated with SDF-1a for12 hours (5-6). No treatment group under similar conditions (3-4). Numbersrepresent percent of cells tagged with CXCR4 mAb. Representative of twoindependent experiments. C, effect of SDF-1a treatment on adhesion of prostatecancer cells to endothelial monolayer (HUVECs). PC-3 and LNCaP cells werelabeled with fluorescent calcein dye after treatment with SDF-1a and usedfor adhesion assay as described. D, effect of SDF-1a pretreatment on TEMof PC-3 cells using transwell chambers. Cells were pretreated with SDF-1a(100 ng/mL) for 6 hours, and then TEM was monitored in the presence ofSDF-1a (10 ng/mL) in the lower chamber. Adhesion and TEM in differenttreatment groups were compared as percent control. Columns, mean ofthree independent experiments; bars, FSE. *, P < 0.05 [SDF-1a (100 ng/mL)versus control]; **, P < 0.01 [SDF-1a (100 ng/mL) versus SDF-1a + CXCR4antibody].

SDF-1a Increases CXCR4 Expression in PC-3 Cells




or the trans–dominant-negative mInBa into prostate cancer cells.We then monitored CXCR4 gene expression by RT-PCR (Fig. 2A)and change in endothelial adhesion of the prostate cancer cells(Fig. 2B). The introduction of trans–dominant-negative mutant ofInBa suppressed (2- to 3-fold) CXCR4 gene expression in PC-3cells but not in LNCaP cells (Fig. 2A). Furthermore, a significantdecrease (30-40%) in endothelial adhesion of PC-3 cells express-ing the trans–dominant-negative mutant was also observed(Fig. 2B), whereas mInBa did not significantly affect endothelialadhesion of LNCaP cells. These findings suggested a direct role ofNF-nB signaling pathway in regulating CXCR4 expressionspecifically in the bone-metastasized PC-3 cells but not in thelymph node–metastasized LNCaP cells.Overexpression of nuclear factor-KB (p65 subunit)

increases both CXCR4 expression and adhesion of PC-3 cells.The transcriptional activation of NF-nB occurs via heterodime-rization of its subunits, p50 and p65 (24). Because our studysuggested a crucial role of the NF-nB pathway in CXCR4expression in PC-3 cells, we further investigated whether one orboth of the NF-nB subunits are important in this process. PC-3cells were transfected with either control vector or the p65 or p50expression vectors. As positive controls, parallel groups were alsotransfected with the CXCR4 expression vector. CXCR4 mRNAexpression was monitored in the transfected cells by RT-PCR(Fig. 2C). The RT-PCR data showed that overexpression of thep65 (Rel-A) subunit of NF-nB significantly up-regulated (2- to3-fold) CXCR4 gene expression in PC-3 cells. Flow cytometricanalysis of cell surface CXCR4 expression in PC-3 cells alsoshowed increased levels of CXCR4 protein in p65-transfected cells(data not shown). Thus, p65 overexpression enhanced CXCR4expression in PC-3 cells. However, introduction of the p50 sub-unit of NF-nB did not significantly affect CXCR4 expression. Thetransfected PC-3 cells were also used for adhesion assays(Fig. 2D). The PC-3 cells transfected with either the CXCR4 orthe NF-nB p65 expression plasmids showed a significant increase

(160-175%) in adhesion to HUVECs. In contrast, no detectablepercent change in adhesion was observed following introductionof the p50 subunit. These findings suggested that increase inNF-nB p65-mediated signaling in PC-3 cells enhanced CXCR4levels, which may be directly linked to their increased adhesiveproperties towards HUVECs.Stromal-derived factor-1A stimulation regulates temporal

activation of nuclear factor-KB in PC-3 cells. We monitored theSDF-1a-induced NF-nB activation in PC-3 cells by Westernimmunodetection of p65 and InBa and by transient transfectionassays using an NF-nB reporter plasmid (Fig. 3). SDF-1a stim-ulation increased nuclear levels of NF-nB p65 in a biphasic manner(Fig. 3A). The p65 subunit increased within 15 to 30 minutesfollowing SDF-1a stimulation, which was followed by a decreaseto basal levels within 2 hours and a second phase of up-regulationat 3 to 4 hours following stimulation (Fig. 3A).To determine SDF-1a-induced activation of NF-nB, kinetic

analysis of total InBa and phospho-InBa was conducted byWestern blot analysis (Fig. 3B). Activation of NF-nB requiressequential phosphorylation, ubiquitination, and degradation ofInBa. Phospho-InBa levels increased by 4-fold at 30 minutes afterSDF-1a stimulation, with maximal increase observed between 2and 3 hours, suggesting that the activation occurred withinminutes following addition of the ligand to the culture. Asimultaneous decrease in total InBa levels was also observedwithin 2 hours. SDF-1a-induced InBa activation correlated withactivation of NF-nB as early as 15 minutes (Fig. 3B).We also examined the effect of SDF-1a (100 ng/mL) treatment of

PC-3 cells at different time intervals on NF-nB-dependenttranscriptional activity (Fig. 3C). PC-3 cells were transfected withNF-nB luciferase reporter construct and treated with SDF-1afor 1, 4, 6, and 18 hours before harvesting. The normalized relativeluciferase activity values showed that SDF-1a treatment signifi-cantly increased (8- to 10-fold) NF-nB activity within 4 hours(Fig. 3C). The relative fold induction in NF-nB activity showed a

Figure 2. A, blockade of NF-nB signaling suppressed CXCR4mRNA expression in PC-3 cells. PC-3 and LNCaP cells weretransduced with adenoviral vector encoding trans-dominantmInB. Transduction efficiency was monitored by a reporterAdv-h-galactosidase construct followed by in situ staining.GAPDH mRNA levels were monitored as internal control forRT-PCR studies. B, effect of overexpression of mInB on adhesionof the prostate cancer cells to HUVEC monolayer. In thisexperiment, the PC-3 (Adv WT) group was included as a control.Columns, mean of three independent experiments usingStudent’s t test; bars, FSE. **, P < 0.01 (AdvWT and AdvmInB-a).C, effect of overexpression of p65NF-nB on CXCR4 mRNAexpression. PC-3 and LNCaP cells were transfected transientlywith p65, p50, and CXCR4 expression vector constructs and RNAwas isolated 48 hours after transfection. Effect of pCMV-CXCR4transfection was also measured, which served as a positivecontrol. Numbers represent fold change based on densitometricanalysis of the bands. D, effect of overexpression of p65NF-nBon adhesion of PC-3 and LNCaP cells. Adhesion assays weredone as described, and effects of overexpression of theplasmids were monitored. Columns, mean of three independentexperiments; bars, FSE. *, P < 0.05 (p65 versus control);**, P < 0.01 (p65 versus p65 + CXCR4 antibody).

Cancer Research




peak level at 3 to 4 hours which returned to basal levels within18 hours following SDF-1a stimulation.Stromal-derived factor-1A–induced nuclear factor-KB

activation is dependent on IKB kinase phosphorylationvia mitogen-activated protein kinase/extracellular signal-regulated kinase signaling cascade. NF-nB activation requiresactivation of IKK via its phosphorylation (P-IKK). SDF-1a inducedP-IKK-a levels within 10 to 15 minutes (data not shown) andreached peak at 30 minutes following treatment of PC-3 cellswith SDF-1a (Fig. 4A). Preincubation of PC-3 cells with a specificinhibitor of MAPK kinase (MEK), PD-98059 (50 Amol/L), abro-gated SDF-1a-induced IKK-a phosphorylation (Fig. 4A). TheIKK-a phosphorylation and NF-nB activation may also occur viathe MAPK pathway. A similar but more rapid biphasic effect ofSDF-1a on MAPK-1/2 (ERK-1/2) levels was indeed observed(Fig. 4B). Western analysis of the MAPK p42 and p44 levels inSDF-1a-treated PC-3 cells showed an increase within 15 minutes,which waned to basal levels within 60 minutes and wereup-regulated again 2 to 3 hours following stimulation with SDF-1a. In PC-3 cells transfected with NF-nB-luc, SDF-1a-inducedreporter activation was also blocked by PD98059 (50 Amol/L),

which suggested that early activation of NF-nB by SDF-1a isdependent on the MEK/ERK pathway (Fig. 4C).Suppression ofmitogen-activatedprotein kinase/extracellular

signal-regulated kinase signaling decreased stromal-derivedfactor-1A–induced CXCR4 expression and transendothelialmigration of PC-3 cells. Flow cytometric analysis showed thatSDF-1a-induced CXCR4 protein expression was significantlyinhibited by cotreatment of PC3 cells with PD98059 but notLY-294002 (Fig. 5A). SDF-1a-induced CXCR4 protein expression(33.2 F 2.8) was significantly enhanced compared with untreatedgroup (8.2 F 3.6). PD98059 cotreatment suppressed the CXCR4expression (18.6 F 2.2) in PC-3 cells; in contrast, LY-294002(PI3K inhibitor) cotreatment did not change the CXCR4expression (Fig. 5A). We also monitored TEM of SDF-1a-treatedPC-3 cells in the presence and absence of PD98059 in responseto SDF-1a (10 ng/mL). SDF-1a pretreatment for 6 hours

Figure 4. A, effect of SDF-1a on IKKa protein levels and its phosphorylationin PC-3 cells. Whole cell extracts were prepared from SDF-1a-treated PC-3cells for the indicated period of time. IKKa phosphorylation was then analysedby anti-phospho-IKKa (pIKKa ) Western blot. The effect of PD98059 onSDF-1a-induced IKKa phosphorylation was also monitored. B, PC-3 cells werepreincubated with PD98059 (50 Amol/L) for 1 hour and treated with SDF-1afor 30 minutes. Kinetic analysis of SDF-1a-induced ERK-1/2 (p42/p44 MAPK)expression. Aliquots from same whole cell extracts were prepared and subjectedto anti-p42/44 MAPK Western blotting. Numbers represent fold change basedon densitometric analysis of bands, with vehicle-treated control samples beingset at 1.0, and are an average of duplicate experiments of the same clone.C, PD98059 cotreatment suppressed SDF-1a-stimulated NF-nB activation. PC-3cells were cotransfected with NF-nB-luc constructs and treated with SDF-1a(100 ng/mL) for 4 hours in the presence or absence of PD98059 (50 Amol/L).Relative luciferase activity was normalized to h-galactosidase activity to correctfor variability in transfection efficiency. Columns, mean of three independentexperiments; bars, FSE. **, P < 0.01 (SDF-1a compared with SDF-1a +PD98059).

Figure 3. A, SDF-1a-induced NF-nB nuclear translocation in PC-3 cells.PC-3 cells were serum starved for 12 hours and incubated with or withoutSDF-1a for the indicated periods of time. Whole cell and nuclear extractswere then prepared and subjected to Western blot analysis using p65NF-nBantibodies. B, additional aliquots of the same whole cell extracts were subjectedto anti-IkBa Western blot analysis. Numbers represent fold change based ondensitometric analysis of bands, with vehicle-treated control samples beingset at 1.0, and are an average of duplicate experiments of the same clone.C, temporal effect of SDF-1a on NF-nB transcriptional activity in PC-3 cells.PC-3 cells were cotransfected with NF-nB-luc and pFR-luc constructs andtreated with SDF-1a (100 ng/mL) for 1, 4, 6, and 18 hours. Relative luciferaseactivity was normalized to h-galactosidase activity to correct for variability intransfection efficiency. The results were analysed as relative fold induction.Columns, mean of triplicate experiments; bars, FSE. **, P < 0.01 [SDF-1a(4 hours) compared with untreated group].





increased TEM of the PC-3 cells (60-80%), which was completelyblocked by PD89059 cotreatment (Fig. 5B). However, PD98059treatment alone in the absence of SDF-1a did not affect TEM ofPC-3 cells.

Discussion

Recent evidence suggests that SDF-1/CXCR4 axis plays apivotal role in triggering bone metastasis. The chemotacticeffect of SDF-1a is known to enhance the TEM of both tumorcells and hematopoietic cells (25). The local release of SDF-1afrom bone marrow endothelial cells as well as a concentrationgradient generated via their production by the marrow stromalcells may act in conjunction to enhance the TEM of prostatecancer cells. The chemokine, SDF-1a is constitutively producedby endothelial cells, fibroblasts, and osteoblasts in the bonemarrow (26) and may play an important role in bone metastasisof prostate cancer cells. The current study shows that there is asignificant increase in PC-3 cell adhesion to HUVECs followingstimulation with SDF-1a. Two different phenotypes of prostatecancer cells, the bone metastasis–derived PC-3 cells and lymphnode metastatic LNCaP cells were investigated to delineate themolecular mechanisms that may be involved in their ability tometastasize to the bone. The findings implicate a crucial roleof the CXCR4/SDF-1a signaling in both recruitment and TEMof PC-3 cells towards the underlying marrow stroma andsuggest a preferential adhesion of PC-3 cells to the endothelialcells and their enhanced transmigration in response to thechemokine gradient.SDF-1a has been shown to induce NF-nB activation in primary

astrocytes and Kaposi’s sarcoma spindle cells (27). We hypothe-sized that SDF-1a-induced CXCR4 gene up-regulation is mediatedvia activation of NF-nB in bone metastatic PC-3 cells. Weobserved that there is a concentration-dependent effect of SDF-1atreatment of PC-3 cells but not of LNCaP cells in increasing theiradhesion to HUVECs. The semiquantitative RT-PCR assay and

flow cytometric data clearly showed that PC-3 cells constitutivelyexpress CXCR4. Our observation that within 6 hours followingSDF-1a treatment there was a 4- to 6-fold increase in CXCR4message in PC-3 cells but not in the LNCaP cells suggest thata specific signaling in bone metastatic prostate cancer cells maylead to their enhanced CXCR4 expression showing responsivenessto the chemokine.SDF-1a and CXCR4 are overexpressed in several neoplasms that

metastasize to the bone marrow, such as breast cancer,neuroblastoma, and several leukemias (28, 29). The transcriptionfactors that regulate SDF-1a and CXCR4 expression in prostatecancer cells are yet to be confirmed. NF-nB has been shown toregulate several prometastatic and antimetastatic factors (30). Inthis study, we explored whether NF-nB has a direct role inregulation of CXCR4 expression in two different prostate cancercell lines. The present data suggest that overexpression of super-repressor I nBamutant suppressed CXCR4 gene expression in PC-3cells but not in LNCaP cells. This was also supported by thefindings that transient transfection of NF-nB expression plasmid(p65 subunit) up-regulated CXCR4 expression in PC-3 cells. Ourdata clearly show that SDF-1a enhances transcriptional activity ofNF-nB in an InBa-dependent manner, which in turn would up-regulate expression of CXCR4, an effect that was reversed by adominant-negative mutant InBa. These data clearly suggest thatsuppression of NF-nB results in inhibition of SDF-1a-stimulatedexpression of CXCR4.The differential response to SDF-1a treatment may be

attributed to constitutively active NF-nB in PC-3 cells. It isreported that constitutive activation of NF-nB is inversely relatedto the androgen receptor status of the cells (31). Prominentconstitutive NF-nB activation was observed in prostate cancer celllines lacking androgen receptor expression (PC-3), whereas onlyvery low levels of NF-nB activity was seen in androgen responsiveLNCaP cell line (32). The detailed mechanisms responsible forconstitutive NF-nB activation in prostate cancer cells remainunclear, but some studies have implicated constitutive activationof IKK complex as a possible mechanism (33). Thus, the half-life ofInBa protein is greatly reduced in prostate cancer cells withconstitutive NF-nB activity compared with the cells with basalactivity. Thus, despite the expression of CXCR4 receptor inandrogen responsive LNCaP cells, the resistance to SDF-1atreatment is most likely associated with suppression of NF-nBactivity in this cell line.We have explored the regulatory mechanism of NF-nB

activation in response to SDF-1a in bone metastatic PC-3 cells.Our data have shown that expression of p65 NF-nB increasedwithin 15 to 30 minutes following SDF-1a treatment. Interestingly,p65 expression showed a biphasic pattern, initial up-regulationdecreased to basal level within 2 hours, and the second phase ofup-regulation was observed at 3 to 4 hours. The activation ofNF-nB requires sequential phosphorylation, ubiquitination, anddegradation of InBa (34). Comparison of the levels of total andphosphorylated forms of InBa in extracts from SDF-1a-treatedPC-3 cells showed that total InBa was reduced after 1 hour ofSDF-1a stimulation with maximal decay between 2 and 3 hours.At the same time, the levels of phospho-InBa were found to bemaximal at 2 to 3 hours of SDF-1a stimulation. These resultssuggest that SDF-1a-induced NF-nB expression correlated withdegradation of total InBa and induction of phospho-InBa.However, one of the key target genes regulated by NF-nB is itsinhibitor InBa. A feedback inhibition pathway for control of InBa

Figure 5. A, effect of PD98059 cotreatment on SDF-1a-induced CXCR4protein expression in PC-3 cells using flow cytometry. The PC-3 cells weretreated with SDF-1a in the presence or absence of inhibitors (PD 98059 andLY-294002). Cotreatment with PD 98059 (50 Amol/L) or LY-294002 (30 Amol/L)was carried out under similar conditions. Columns, average % CXCR4expression of two independent experiments. B, effect of PD98059 cotreatmenton the SDF-1a-induced TEM of PC-3 cells in response to a chemokinegradient of SDF-1a (10 ng/mL) in transwell chambers. TEM assay was donein 12-well plates with transwell chambers, and SDF-1a was placed in lowerchamber to develop a gradient. Columns, average of three experiments.**, P < 0.01 (SDF-1a with respect to untreated group); *, P < 0.05 (SDF-1awith respect to SDF-1a + PD98059 group).

Cancer Research




gene transcription and down-regulation of transient activationof NF-nB activity has been described (35–37). Additionally, ithas been shown that proinflammatory cytokines induce biphasicNF-nB activation, consisting of a transient phase mediatedthrough InBa followed by a persistent phase mediated throughInBh (38). Hoffmann et al. have shown that InB/NF-nB signalinghas biphasic characteristics: InBa mediates rapid NF-nB activa-tion and strong negative feedback regulation, resulting in anoscillatory NF-nB activation profile, whereas InBh and InBqrespond more slowly to IKK activation and act to dampen thelong-term oscillations of the NF-nB response (39). Based on theseobservations, it is possible that the bimodal induction of NF-nBactivation by SDF-1a in PC-3 cells is related to transientinhibition caused by InBa. Alternatively, it remains to be seenwhether the second phase of NF-nB activation by NF-SDF-1a isassociated with InBh and/or InB phosphorylation. Thus, thepossibility of involvement of another component in SDF-1a orNF-nB signaling for CXCR4 activation in PC-3 cells certainlywarrants further investigation.ERK-1/2 MAPK mediates signals for many growth factors and

G-protein-coupled receptors. Activation of ERK-1/2 MAPK byG-protein-coupled receptors occurs via Raf/MEK-1/2/ERK-1/2cascade, which can be blocked by a specific chemical inhibitor,PD98059 (40, 41). Consistent with our results, SDF-1a activatesERK-1/2 MAPK in several cell types (42), including epitheloid(43) and ovarian (44) carcinoma cells. However, we observed abiphasic activation of MAPK levels in PC-3 cells; the p42/p44MAPK levels showed an increase in expression within 15minutes, which waned to basal levels within 60 minutes andwere up-regulated again at 2 hours following SDF-1a stimulation.We have shown that SDF-1a stimulation of PC-3 cells inducedIKKa activation by up-regulating the phosphorylated form of

IKKa. However, unlike in hematopoietic progenitor cells (45),early SDF-1a activation of ERK-1/2 could be responsible foractivation of NF-nB in PC-3 cells. Activation of IKKa was foundto be sensitive to a pharmacologic inhibitor, PD98059, a specificinhibitor of MEK-1 and MEK-2 phosphorylation. Additionally,we also show that PD98059 effectively blocked SDF-1a-inducedNF-nB activation by reporter assays. Accordingly, our datasuggest that IKKa/NF-nB activation is a downstream target toMEK-1/2 induced signaling by SDF-1a in PC-3 cells.The role of SDF-1a-induced MEK/ERK signaling was investi-

gated by monitoring CXCR4 expression. Our data showed thatSDF-1a-induced CXCR4 protein expression was significantlysuppressed by PD98059 and that NF-nB p65 subunit regulatesCXCR4 gene expression in PC-3 cells. The functional role ofCXCR4 receptor in adhesion to endothelial cells and TEM ofPC-3 cells further attests to its vital role in bone metastasis.These data provide evidence of increased metastatic potential ofosteotropic PC-3 cells by enhancing expression of CXCR4 uponexposure to SDF-1a in a MEK/ERK/NF-nB-dependent manner.In summary, we have shown that following SDF-1a treatment ofPC-3 cells, MEK activation occurs within 15 minutes leading toIKK and InBa phosphorylation. This leads to nuclear localiza-tion of NF-nB followed by transcription and expression ofCXCR4. These findings suggest novel targets for therapeuticintervention to suppress bone metastasis of prostate cancercells.

Acknowledgments

Received 4/14/2005; revised 8/27/2005; accepted 9/1/2005.The costs of publication of this article were defrayed in part by the payment of page

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



References1. Jemal A, Murray T, Ward E, et al. Cancer statistics. CACancer J Clin 2005;55:10–30.

2. Netelenbos T, van den Born J, Kessler FL, et al.Proteoglycans on bone marrow endothelial cells bindand present SDF-1a towards hematopoietic progenitorcells. Leukemia 2003;17:175–84.

3. Mochizuki H, Matsubara A, Teishima J, et al.Interaction of ligand-receptor system between stro-mal-cell-derived factor-1 and C-X-C chemokine recep-tor-4 in human prostate cancer: a possible predictorof metastasis. Biochem Biophys Res Commum 2004;320:656–63.

4. Taichman RS, Cooper C, Keller ET, Pienta KJ, Taich-man NS, McCauley LK. Use of the stromal cell-derivedfactor/CXCR4 pathway in prostate cancer metastasis tobone. Cancer Res 2002;62:1832–7.

5. Sun YX, Wang J, Shelburne CE, et al. Expressionof CXCR4 and SDF-1a(CXCL-12) in human pros-tate cancers (PCa) in vivo . J Cell Biochem 2003;89:462–73.

6. Singh S, Singh UP, Grizzle WE, Lillard JW. SDF-1a-CXCR4 interactions modulate prostate cancer cellmigration, metalloproteinase expression and invasion.Lab Invest 2004;84:1666–76.

7. Darash-Yahana M, Pikarsky E, Zeira E, et al. Role ofhigh expression of CXCR4 in tumor growth, vasculari-zation, and metastasis. FASEB J 2004;18:1240–2.

8. Feng Y, Broder CC, Kennedy PE, Berger EA. HIV-1entry cofactor: functional cDNA cloning of a seven-transmembrane G protein-coupled receptor. Science1996;272:872–7.

9. McGrath KE, Koniski AD, Maltby KM, McGann JK,Palis J. Embryonic expression and function of the

chemokine SDF-1 and its receptor, CXCR4. Dev Biol1999;213:442–56.

10. Ganju RK, Brubaker SA, Meyer J, et al. The a

chemokine, stromal cell-derived factor-1a, binds to thetransmembrane G-protein-coupled CXCR4 receptor andactivates multiple signal transduction pathways. J BiolChem 1998;273:23169–75.

11. Han Y, Wang J, Ransohoff RM. TNF-a down regulatesCXCR4 expression in primary murine astrocytes. BrainRes 2001;888:1–10.

12. Popik W, Hesselgesser JE, Pitha PM. Binding ofhuman immunodeficiency virus type 1 to CD4 andCXCR4 receptors differentially regulates expression ofinflammatory genes and activates the MEK/ERKsignaling pathway. J Virol 1998;72:6406–13.

13. Sotsios Y, Whittaker GC, Westwick J, Ward SG. TheCXC chemokine stromal derived factor activates a Gi-coupled phosphoinositide 3-kinase in T lymphocytes.J Immunol 1999;163:5954–63.

14. Sun SC, Xiao G. Deregulation of NF-nB and itsupstream kinases in cancer. Cancer Metastasis Rev 2003;22:405–22.

15. Brown K, Gerstberger S, Carlson L, Franzoso G,Siebenlist U. Control of InB-a proteolysis by sitespecific, signal-induced phosphorylation. Science 1995;267:1485–8.

16. Helbig G, Christopherson KW II, Bhat-Nakshatri P,et al. NF-kB promotes breast cancer cell migration andmetastasis by inducing the expression of the chemokinereceptor CXCR4. J Biol Chem 2003;278:21631–8.

17. Hardy C. Homing of hematopoietic stem cells to themarrow. Am J Med Sci 1995;309:260–6.

18. Kim CH, Broxmeyer HE. SLC/exodus2/6Ckine/TCA4induces chemotaxis of hematopoietic progenitor cells:differential activity of ligands of CCR7, CXCR3, or

CXCR4 in chemotaxis vs. suppression of progenitorproliferation. J Leukoc Biol 1999;66:455–61.

19. Aiuti A, Tavian M, Cipponi A, et al. Expression ofCXCR4, the receptor for stromal cell-derived factor-1 onfetal and adult human lympho-hematopoietic progeni-tors. Eur J Immunol 1999;29:1823–31.

20. Peled A, Petit I, Kollet O, et al. Dependence of humanstem cell engraftment and repopulation of NOD/SCIDmice on CXCR4. Science 1999;283:845–8.

21. Aiuti A, Webb IJ, Bleul C, Springer T, Gutierrez-Ramos JC. The chemokine SDF-1 is a chemoattractantfor human CD34+ hematopoietic progenitor cells andprovides a new mechanism to explain the mobilizationof CD34+ progenitors to peripheral blood. J Exp Med1997;185:111–20.

22. Andela VB, Gordon AH, Zotalis G, et al. NF-nB: apivotal transcription factor in prostate cancer metasta-sis to bone. Clin Orth and Rel Res 2003;S415:75–85.

23. Baeuerle PA, Baltimore D. InB: a specific inhibitorof the NF-nB transcription factor. Science 1998;242:540–6.

24. Karin M, Ben-Neriah Y. Phosphorylation meetsubiquitination: the control of NF-nB activity. Annu RevImmunol 2000;18:621–63.

25. Imai K, Kobayashi M, Wang J, et al. Selectivesecretion of chemoattractants for hemopoietic pro-genitor cells to bone marrow. Br J Haematol 1999;106:905–11.

26. Ponomaryov T, Peled A, Petit I, et al. Increasedproduction of SDF-1 following treatment with DNAdamaging agents: relevance for human stem cell homingand repopulation of NOD/SCID mice. J Clin Invest 2000;106:1331–9.

27. Han Y, Huang D, Pardo CA, Ransohoff RM. TNF-amediates SDF-1a-induced NF-nB activation and



Cancer Research


cytotoxic effects in primary astrocytes. J Clin Invest2001;108:425–35.

28. Geminder H, Sagi-Assif O, Goldberg L, et al. Apossible role for CXCR4 and its ligand, the CXCchemokine stromal cell-derived factor-1, in the devel-opment of bone metastasis in neuroblastoma. J Immunol2001;167:4747–57.

29. Mohle R, Failenschmid C, Bautz F, Kanz L. Over-expression of chemokine receptor CXCR4 in B cellchronic lymphocytic leukemia is associated withincreased functional response to SDF-1. Leukemia1999;13:1954–9.

30. Andela VB, Schwarz EM, Puzas JE, O’Keefe RJ, RosierRN. Tumor metastasis and the reciprocal regulation ofprometastatic and antimetastatic factors by nuclearfactor nB. Cancer Res 2000;60:6557–62.

31. Palayoor ST, Youmell MY, Calderwood SK, ColemanCN, Price BD. Constitutive activation of InB kinase a

and NF-nB in prostate cancer cells is inhibited byibuprofen. Oncogene 1999;18:7389–94.

32. Gasparian AV, Yao YJ, Kowalczyk D, et al. The role ofIKK in constitutive activation of NF-nB transcriptionfactor in prostate carcinoma cells. J Cell Sci 2002;115:141–51.

33. Suh J, Payvandi F, Edelstein LC, et al. Mechanisms ofconstitutive NF-nB activation in human prostate cancercells. Prostate 2002;52:183–200.

34. Suh J, Rabson AB. NF-nB activation in humanprostate cancer: important mediator or epiphenome-non? J Cell Biochem 2004;91:100–17.

35. Chen Q, Cant CA, Moll T, et al. NF-nB subunit-specific regulation of the InBa promoter. J Biol Chem1994;269:13551–7.

36. Chiao PJ, Miyamoto S, Verma IM. Autoregulationof InBa activity. Proc Natl Acad Sci U S A 1994;91:28–32.

37. Sun C, Ganchi PA, Ballard DW, Greene WC. NF-nBcontrols expression of inhibitor InBa: evidence for aninducible autoregulatory pathway. Science 1993;259:1912–5.

38. Thompson JE, Phillips RJ, Erdjument-Bromage H,Tempst P, Ghosh S. InBh regulates the persistentresponse in a biphasic activation of NF-nB. Cell 1995;80:573–82.

39. Hoffmann A, Levchenko A, Scott ML, Baltimore D.The InB-NF-nB signaling module: temporal control andselective gene activation. Science 2002;298:1241–5.

40. Lewis TS, Shapiro PS, Ahn NG. Signal transduction

through MAP kinase cascades. Adv Cancer Res 1998;74:49–139.

41. Liebmann C, Bohmer FD. Signal transduction path-ways of G-protein coupled receptors and their cross-talkwith receptor tyrosine kinases: lesson from bradykininsignaling. Curr Med Chem 2000;7:911–43.

42. Montes M, Tagieva NE, Hevekar N, Nahmias C,Baleux F, Trautmann A. SDF-1-induced activation ofERK enhances HIV-1 expression. Eur Cytokine Netw2000;11:470–7.

43. Peng SB, Peek V, Zhai Y, et al. Akt activation, butnot extracellular signal-regulated kinase activation, isrequired for SDF-1a/CXCR4-mediated migration ofepitheloid carcinoma cells. Mol Cancer Res 2005;3:227–36.

44. Porcile C, Bajetto A, Barbero S, Pirani P, Schettini G.CXCR4 activation induces epidermal growth factorreceptor transactivation in an ovarian cancer cell line.Ann N Y Acad Sci 2004;1030:162–9.

45. Lee Y, Gotoh A, Kwon HJ, et al. Enhancementof intracellular signaling associated with hematopoi-etic progenitor cell survival in response to SDF-1/CXCL12 in synergy with other cytokines. Blood 2002;99:4307–17.



2005;65:9891-9898. Cancer Res Promil Kukreja, Asim B. Abdel-Mageed, Debasis Mondal, et al.

B ActivationκDependent NF-−Signaling PathwayAdhesion and Transendothelial Migration: Role of MEK/ERK

(CXCL12) Increases Endothelial αStromal-Derived Factor-1Up-regulation of CXCR4 Expression in PC-3 Cells by

Updated version

http://cancerres.aacrjournals.org/content/65/21/9891

Access the most recent version of this article at:

Cited articles

http://cancerres.aacrjournals.org/content/65/21/9891.full#ref-list-1

This article cites 44 articles, 18 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/65/21/9891.full#related-urls

This article has been cited by 24 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] at

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

Permissions

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

.http://cancerres.aacrjournals.org/content/65/21/9891To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/content/65/21/9891.full#ref-list-1

http://cancerres.aacrjournals.org/content/65/21/9891.full#related-urls

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

mailto:[email protected]



up-regulation of cxcr4 expression in pc-3 cells by stromal- … · up-regulation of cxcr4...

Documents