norepinephrine promotes the b1-integrin mediated adhesion ... · process is an essential...

Angiogenesis, Metastasis, and the Cellular Microenvironment

Norepinephrine Promotes the b1-Integrin–MediatedAdhesion of MDA-MB-231 Cells to Vascular Endothelium bythe Induction of a GROa Release

Carina Strell1, Bernd Niggemann1, Melanie J. Voss1, Desmond G. Powe2,3, Kurt S. Z€anker1, andFrank Entschladen1

AbstractThe migratory activity of tumor cells and their ability to extravasate from the blood stream through the vascular

endothelium are important steps within the metastasis cascade.We have shown previously that norepinephrine is apotent inducer of the migration ofMDA-MB-468 human breast carcinoma cells and therefore investigated herein,whether the interaction of these cells as well as MDA-MB-231 andMDA-MB-435S human breast carcinoma cellswith the vascular endothelium is affected by this neurotransmitter as well. By means of a flow-through assay underphysiologic flow conditions, we show that norepinephrine induces an increase of the adhesion of the MDA-MB-231 cells, but not of MDA-MB-468 and MDA-MB-435S cells to human pulmonary microvascular endothelialcells (HMVEC). The adhesion of MDA-MB-231 cells was based on a norepinephrine-mediated release of GROafrom HMVECs. GROa caused a b1-integrin–mediated increase of the adhesion of MDA-MB-231 cells. Mostinterestingly, this effect of norepinephrine, similar to the aforementioned induction ofmigration inMDA-MB-468cells, wasmediated byb-adrenergic receptors and therefore abrogated byb-blockers. In conclusion, norepinephrinehas cell line–specific effects with regard to certain steps of the metastasis cascade, which are conjointly inhibited byclinically established b-blockers. Therefore, these results may deliver a molecular explanation for our recentlypublished retrospective data analysis of patients with breast cancer which shows that b-blockers significantly reducethe development of metastases. Mol Cancer Res; 10(2); 197–207. �2011 AACR.

Introduction

In his article on the war against cancer, Sporn pointed outalready in 1996 that "it is local invasion and distant metas-tasis that kill rather than excessive cell proliferation per se."(1). Accordingly, today more than 90% of breast cancer–related deaths are because of the development of metastases.Thus, future therapeutic strategies for cancer treatmentmustinclude the inhibition of metastasis formation, and thereforethe understanding of the molecular mechanisms of thisprocess is an essential prerequisite. The metastasis cascadeis a multistep process of highly regulated events includingtumor cell migration as well as the intra- and extravasationinto and out of the blood stream. Each step of the metastasis

cascade comprises the interaction of the tumor cells withtheir environment in cell–cell and cell–matrix contacts andthe susceptibility of these interactions to the regulation bysoluble signal substances. We have reported previously thatthe stress hormone norepinephrine is the most potentphysiologic inducer of the migration of breast carcinomacells (2), as well as of colon (3), and prostate carcinoma cells(4) in vitro. The promigratory effect is mediated byb-adrenergic receptors because it could be blocked signifi-cantly by clinically established b-blockers (i.e., propranolol;refs. 2–4). These findings were confirmed in a xenograftmouse model with PC-3 prostate carcinoma cells. In thismodel, parallel application of propranolol decreased lumbarlymph node metastases formation induced by norepineph-rine (5). Furthermore, other groups have shown that nor-epinephrine increases the migration and invasion of pan-creatic (6) and ovarian (7) cancer cell lines. Besides studiesusing the direct application of norepinephrine, there areincreasing numbers of studies showing that social stress, forexample, by isolation, is associated with an increased tumorgrowth (8) and metastasis (9) in mouse models.Recently, 3 independent epidemiologic studies provided

evidence of the clinical relevance of these experimentalfindings: retrospective data analyses of patients with breastcancer showed that those women who received b-blockersbecause of hypertension developed metastases significantly

Authors' Affiliations: 1Institute for Immunology, ZBAF, Witten/HerdeckeUniversity, Witten, Germany; 2Department of Cellular Pathology, Queen'sMedical Centre, Nottingham University Hospitals NHS Trust; and 3TheJohn van Geest Centre, The Nottingham Trent University, Nottingham,United Kingdom

Corresponding Author: Frank Entschladen, Institute of Immunology,ZBAF, Witten/Herdecke University, Stockumer Str. 10, 58448 Witten,Germany. Phone: 49-2302-926-187; Fax: 49-2302-926-158; E-mail:[email protected]

doi: 10.1158/1541-7786.MCR-11-0130

�2011 American Association for Cancer Research.

MolecularCancer

Research

www.aacrjournals.org 197

on September 20, 2020. © 2012 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst November 29, 2011; DOI: 10.1158/1541-7786.MCR-11-0130

http://mcr.aacrjournals.org/

later than untreated patients or patients treated with otheranti-hypertensive drugs and had consequently an increasedcancer-specific survival time (10–12). Immunohistochemi-cal analysis of tissue microarrays from patients with breastcancer showed that a- and b-adrenergic receptors are dif-ferentially overexpressed in distinct subtypes of breast can-cer. Especially, the a1b- and a2c-adrenergic receptors wereassociated with a poor prognosis in basal-like tumors (13).However, not only direct effects of the adrenergic

signaling on the tumor cells but also indirect effects oncells of the tumor environment might promote metastasisformation, as was recently shown by Sloan and colleagues(14): they provide evidence that the sympathetic (norad-renergic) activation of macrophages promotes metastasisformation in an orthotopic mouse model. This putsforward the hypothesis of parallels between a tumor andinflammation (15, 16). Proinflammatory chemokines, asdelivered by macrophages, regulate the site at whichleukocytes leave the blood stream by the activation ofintegrin-mediated adhesion (17, 18), and a similar role hasbeen attributed to chemokines for the localization ofmetastases in several types of cancer (19). CXC chemo-kines, such as CXCL8/IL-8 and CXCL1/GROa, bind tothe G-protein–coupled receptors, CXCR1 and CXCR2,which are described to play an important role with regardto disease progression and metastasis formation (20). Incolorectal cancers, GROa expression was shown to bepositively associated with tumor size, stage, invasion,lymph node metastasis, and patient survival (21).In the present study, we now question whether norepi-

nephrine does not only directly affect the migratory activityof breast carcinoma cells but also other parts of themetastasiscascade in a direct or indirect manner. Therefore, weinvestigated the ability of tumor cells to adhere to vascularendothelium, a further important step of the hematogenoustumormetastasis cascade during extravasation. The rationalefor these experiments is based on the observation that themicrovasculature of organ sites, especially of lymphoid organsites, is surrounded by nerve fibers that release catechola-mines at high local doses thereby affecting the endotheliumitself as well as circulating leukocytes (22).With regard to theinvolved receptors in the adhesion of the tumor cells to theendothelium, we especially focused on the role of integrins,as (i) they are known to be key receptors in the extravasationprocess (23) and (ii) their affinity to the ligand is regulated bythe so-called "inside-out-signaling" which is mediated bychemokines (24). In addition, we investigated the involve-ment of chemokines in the interaction of breast cancer cellswith endothelial cells and the regulation of their expressionby the transcription factor NF-kB.

Materials and Methods

Cell lines and cell cultureThe human breast carcinoma cell lines MDA-MB-468,

MDA-MB-435S, and MDA-MB-231 (all from the Amer-ican Type Culture Collection) were cultured in Dulbecco'sModified Eagle's Media (DMEM; PAA) containing 10%

fetal calf serum (PAA) at 37�C humidified atmosphere with5% CO2. Confirmatory experiments were carried out withPC-3 human prostate carcinoma cells (DSMZ). These cellswere cultured in HAM's F-12 medium and RPMI-1640medium (1:1; both PAA) containing 10% heat-inactivatedfetal calf serum (PAA) under the same conditions as thebreast carcinoma cells. All cell lines were authenticated by thesupplier and used no longer than 6 months after resuscita-tion. For experiments, cells were collected by short accutase(PAA) treatment at a subconfluent (75% confluency) stageof growth.Human pulmonary microvascular endothelial cells

(HMVEC; Lonza) were cultured in EBM-2MVmedia withsupplements (Lonza) in a humidified incubator with 5%CO2. Endothelial cells were only used within 6 passages androutinely screened for Mycoplasma infection (MycoAlert;Lonza).

Migration experimentsFor the investigation of the migratory activity, migra-

tion experiments were carried out as described in detailpreviously (2). In brief, a cell suspension of 8� 104 tumorcells was mixed with a carbonate-buffered collagen solu-tion (1.63 mg/mL collagen; PureCol, Inamed) containingminimal essential medium. In test samples, norepineph-rine (Sigma-Aldrich) was added to a final concentration of10 mmol/L. The suspension was filled into self-con-structed migration chambers consisting of a microscopicslide, wax walls on 3 sides and a coverslip on top, andpolymerized within 30 minutes at 37�C humidified atmo-sphere with 5% CO2. After polymerization, the residualchamber volume was filled with PBS with or withoutnorepinephrine, and the fourth open side of the chamberwas sealed with wax. The migratory activity of the cellswithin the collagen matrix was monitored for 15 hours bytime lapse video microscopy. After recording, the paths of30 randomly selected cells were digitized by computer-assisted cell tracking. The migratory activity was calculatedin 15-minute intervals as the portion of cells (in percent-age), which was locomotory active, for example, when 15of 30 cells moved, the migratory activity at this certaintime point was 50%.

Flow-through adhesion assayThe flow-through adhesion assay was conducted as

described previously (25). HMVECs were seeded oncollagen IV–coated flow chambers (m-slides I, IBIDI) andgrown to confluency for 2 days. Tumor cells (1.5 � 105/mL in DMEM with 2% fetal calf serum) were drawn overthe endothelial monolayer by a perfusion pump (PerfusorIV, B. Braun Melsungen AG) at a flow rate of 12.2 mL/h,which results in a shear stress of 0.25 dyne/cm2 andrepresents physiologic blood flow conditions in smallvessels. The suspension flow was recorded by videomicroscopy for 20 minutes using a 2.5-fold objective. Toinvestigate the impact of norepinephrine and/or theb-blocker propranolol (both Sigma-Aldrich), the endo-thelial monolayer was preincubated with norepinephrine

Strell et al.

Mol Cancer Res; 10(2) February 2012 Molecular Cancer Research198




and/or propranolol for 2 hours at the indicated concen-trations. Furthermore, the substances were also added tothe tumor cell suspension drawn over the endothelium.The NF-kB activation inhibitor (Merck Biosciences) wasused in the same manner at a concentration of 10 nmol/L.This concentration was used because of previous studieson cell migration (unpublished results) and corresponds tothe highest concentration that showed no effect on thespontaneous migratory activity of tumor cells.For blocking experiments, MDA-MB231 were preincu-

bated for 10 minutes with monoclonal mouse anti-CXCR1and anti-CXCR2 (each 5 mg/mL; R&D Systems), mouseIgG2a (10 mg/mL; R&D Systems), mouse anti-b1-integrinclone 4B4 (5 mg/mL; Beckman Coulter), or mouse IgG1 (5mg/mL; Beckman Coulter). The highest possible concen-tration of IgG that showed no effect on the adhesion of thetumor cells (data not shown) was chosen as final concen-tration for blocking antibodies.For chemokine-dependent adhesion experiments, the

HMVEC monolayer was overlaid for 3 minutes at roomtemperature with GROa and/or interleukin (IL)-8 each at 1mg/mL (both R&D Systems). Denaturated GROa and IL-8(10 minutes at 95�C) served as negative control. TheHMVEC monolayer was washed 3 times with PBS andused directly in the flow-through assay.MDA-MB-231 cells were found to adhere to the endo-

thelium mostly in clusters. Cluster formation was analyzedusing ImageJ Software (NIH, Bethesda, MD). Thus, thearea covered by MDA-MB-231 clusters was calculated inpixels (Fig. 1), which used to quantify the intensity ofadhesion. At least 2 different areas were analyzed perrecorded video.

RNA isolation and cDNA synthesisTotal RNA from 5 � 105 HMVECs or 3 � 106 tumor

cells was isolated using the Total RNA Isolation Kit fromMacherey Nagel according to the manufacturer's instruc-tions. The cDNA was synthesized using the First StrandcDNA Synthesis Kit with RevertAid HMinus M-MuL V asreverse transcriptase and random hexamer primers(Fermentas).

Quantitative real-time PCRReal-time PCR was carried out on an ABI StepOnePlus

(Applied Biosystems) system using "MESA Blue qPCRMaster Mix Plus" (Eurogentec) to determine target ampli-fication. The expression level of b-actin was used as referencefor normalization, and fold expression changes were calcu-lated using the standard DDCT method. The followingprogram was used: 95�C for 10 minutes followed by 45cycles of 95�C for 15 seconds and 60�C for 60 seconds; thiswas followed by melting curve analysis to verify singleproduct amplification. The primers were (50!30): b-actin:fwd-GGA CTT CGA GCA AGA GAT GG and rev-AGGAAG GAA GGC TGG AAG AG; GROa: fwd-ATT CACCCC AAG AAC ATC CA and rev-TAA CTA TGG GGGATGCAGGA; and IL-8: fwd-GTCTGCTAGCCAGGATCC AC and rev-GCT TCCACA TGT CCT CAC AA.

Reverse transcriptase PCRChemokine and b-adrenergic receptor expression was

analyzed by reverse transcriptase PCR. Again, b-actin wasused as reference for normalization.The cDNA samples wereanalyzed usingGoTaqDNApolymerase (Promega) with 1.5mmol/L MgCl2 (2.5 mmol/L MgCl2 for CXCR2) and thefollowing primers (50!30): b-actin: fwd-GTG GGG CGCCCC AGG CAC CA and rev-CTC CTT AAT GTC ACGCAC GAT TTC (annealing temperature, 65�C; ref. 26);CXCR1: fwd-TTT GTT TGT CTT GGC TGC TG andrev-GCC AAG AAC TCC TTG CTG AC (annealingtemperature, 57�C); CXCR2: fwd-CAG TTA CAG CTCTAC CCT GCC and rev-CCAGGAGCAAGGACA-GACCCC (annealing temperature, 65�C; ref. 27); b1-adrenergic receptor: fwd-TCA TCC GAGGCA AAG AGAAAA and rev-TCCCATCCCTTCCCAAACTT (anneal-ing temperature, 58�C); and b2-adrenergic receptor: fwd-TCA CCT TTC AAG TAC CAG AGC CT and rev-ACACAA TCC ACA CCA TCA GAA TG (annealing temper-ature, 58�C).

10 µmol/L Norepinephrine

Control 11,562 pixels marked

40,666 pixels marked

Figure 1. Screenshots showing the method used for the quantificationof the adhesion of the tumor cells to the endothelium under flowconditions. The area on the endothelium covered by MDA-MB-231cells was marked, and the total covered area was calculated in pixels.

Tumor Cell Adhesion to Vascular Endothelium

www.aacrjournals.org Mol Cancer Res; 10(2) February 2012 199




Amplification programs were as follows: b-actin: 94�C for5 minutes, 30 cycles of 94�C for 45 seconds, 65�C for 30seconds, 72�C for 90 seconds, finally followed by 72�C for 5minutes; CXCR1: 94�C for 5minutes, 30 cycles of 94�C for30 seconds, 57�C for 30 seconds, 72�C for 60 seconds,finally followed by 72�C for 5 minutes; CXCR2: 94�C for 2minutes, 35 cycles of 94�C for 30 seconds, 65�C for 30seconds, 72�C for 20 seconds, finally followed by 72�C for 5minutes; and b1- and b2-adrenergic receptors: 94�C for 5minutes, 35 cycles of 94�C for 30 seconds, 58�C for 30seconds, 72�C for 30 seconds, finally followed by 72�C for 5minutes.Primers for chemokine receptors and b-actin were chosen

to span the template for at least one intron so that anygenomic DNA contamination would result in a largerproduct band. Because primers for the b-adrenergic recep-tors were not exon–exon junction spanning and not sepa-rated by at least one intron, we included in these analyses, a"no amplification control" to ensure that the product is notdue to genomic DNA contamination. PCR products werevisualized on a 1.0% (1.5% for b-adrenergic receptors)ethidium bromide agarose gel.

Enzyme-linked immunoassayHMVECs were incubated for 2 hours with 10 mmol/L

norepinephrine and propranolol alone or in combination; allsamples were prepared as doublets. Supernatants were col-lected, and the concentrations of IL-8 and GROa weremeasured by an enzyme-linked immunoassay using the"Quantakine kit" (R&D Systems) according to the manu-facturer's protocol.

Flow cytometrySamples of 5 � 105 endothelial cells or tumor cells were

fixed with 2% paraformaldehyde for 10 minutes, permea-bilized with 0.5% Triton X-100/PBS for 10 minutes, andthen stained in 2% goat-serum/PBS with 10 mg/mL of theprimary antibody (V-19 for the b1-adrenoceptor and H-20 for the b2-adrenoceptor; both from Santa Cruz Bio-technology) or with a rabbit isotypic control antibody(Santa Cruz Biotechnology) for 15 minutes. After wash-ing, the cells were incubated with 10 mg/mL of a secondaryanti-rabbit fluorescein isothiocyanate (FITC)-conjugatedantibody (Santa Cruz Biotechnology) in 2% goat-serum/PBS for further 10 minutes. All incubation steps wereconducted at room temperature. The mean fluorescenceintensity was measured using a FACSCalibur flow cyt-ometer (Becton Dickinson).

Western blot analysisThe expression of the b1- and b2-adrenoceptors was

confirmed by immunoblotting using the same primaryantibodies that were used for flow cytometry. Cells (3 �105) were lysed in Laemmli sample buffer and incubated at95�C for 10 minutes. Proteins were separated using an 8%PAGE (28) and then transferred to an Immobilon-P transfermembrane, with a pore size of 0.45 mm (Millipore Corpo-ration). After transfer, the membranes were blocked for 1

hour at room temperature in TBS containing 5% non-fatmilk powder, followed by 3 washing steps. The membraneswere incubated overnight at 4�Cwith the primary antibodiesand a rabbit anti-eIF4E-antibody (Cell Signalling) as controlfor the applied protein amounts (all antibodies were used in a1:1,000 dilution). Subsequently, the membranes werewashed vigorously with TBS containing 0.1% Tween 20(AppliChem) and then incubated at 1:1,000 dilution with ahorseradish peroxidase (HRP)-linked secondary goat anti-rabbit IgG antibody (Cell Signalling) at room temperaturefor 2 hours. The luminescence signals were detected using aHamamatsu C 4742-98 system (Hamamatsu).

StatisticsStatistical analyses were conducted using Student t test

(two-tailed, unpaired) if not stated differently. A P value lessthan 0.05 was considered statistically significant.

Results

As introduced above, we have reported previously thatseveral carcinoma cells lines from different tissue originsignificantly increase their locomotory activity in responseto norepinephrine (2–4). This neurotransmitter or so-calledstress hormone is to our knowledge the strongest physiologicinducer of the migratory activity of MDA-MB-468 breastcarcinoma cells (2). Likewise, MDA-MB-231 and MDA-MB-435S both present the relevant b1- and b2-adrenergicreceptors as was shown by flow cytometry and immuno-blotting using the same antibodies for both techniques (Fig.2A and B). However, as the Western blot analysis showedonly weak bands, we additionally confirmed the receptorexpression by PCR (Fig. 2C). In addition to the tumor cellslines, HMVECs were tested for b-adrenergic receptorexpression. A human embryonic stem cell line served aspositive control (29). Human pulmonary smooth musclecells were used as positive control for the b2-adrenergicreceptor and negative control for the b1-adrenergic receptor.Human peripheral blood mononuclear cells (i.e., the Tlymphocytes of this fraction) and human prostate carcinomacells of the cell line PC-3 were investigated for b-adrenergicreceptor expression by flow cytometry previously and servedas further positive controls (4, 30). However, although therelevant receptors were expressed on MDA-MB-231 andMDA-MB-435S cells, norepinephrine did in contrast toMDA-MB-468 cells not increase the migratory activity ofthese cells (MDA-MB-231, Fig. 3A: control 52.4%� 7.0%vs. norepinephrine 40.9% � 10.8% locomoting cells;MDA-MB-435S, Fig. 3B: control 43.7% � 5.0% vs.norepinephrine 38.5% � 5.9% locomoting cells; MDA-MB-468, Fig. 3C: control 50.4%� 11.1% vs. norepineph-rine 62.5% � 8.7% locomoting cells).However, the effect of norepinephrine on the adhesion of

the breast carcinoma cells to endothelium under flow con-ditions revealed a completely different picture (Fig. 4). Theadhesion of MDA-MB-468 cells and MDA-MB-435Scells was not influenced by norepinephrine (MDA-MB-468; Fig. 4A: control 7,646 � 2,418 vs. norepinephrine

Strell et al.





8,428� 1,824 pixels, andMDA-MB-435S; Fig. 4B: control8,651 � 3,598 vs. norepinephrine 8,595 � 2,854 pixels),but in contrast, norepinephrine treatment led to a significantand dose-dependent increase of the adhesion of MDA-MB-231 cells with 10 mmol/L being most effective (10,111 �12,825 to 39,906 � 19,927 pixels; P ¼ 0.019; Fig. 4C).This increase of adhesion was mediated by b-adrenergicreceptors, as the b-blocker propranolol abolished the effectof norepinephrine (control 10,822 � 2,979 to norepineph-rine 30,121 � 10,501 pixels; P ¼ 0.012, and norepineph-

rine in combination with propranolol 12,063 � 6,123pixels; P ¼ 0.008; Fig. 4D).As already shown by PCR (Fig. 2C), not only the tumor

cells but also the HMVECs express both the b1- and b2-adrenergic receptor. This was confirmed by flow cytometryto show that the receptors are present on the surface of theHMVECs (Fig. 5A). In response to norepinephrine, theHMVECs significantly increased the release of the chemo-kines GROa (50.2 � 19.3 to 137.5 � 44.3 pg/h/100,000cells; P¼ 0.011) and IL-8 (53.3� 11.7 to 139.2� 9.9 pg/h/

Figure 2. Expression ofb-adrenergic receptors in MDA-MB-231, MDA-MB-435S, and MDA-MB-468 human breast carcinoma celllines. A, measurement of the b1- andb2-adrenergic receptor expressionby flow cytometry. Gray field,receptor-specific antibody; blackline, isotypic (ISO) control antibody;MFI, mean fluorescence intensity. B,Western blot analysis of the b1- andb2-adrenergic receptor (AR)expression. Confirmation of theexpression was conducted with thesame antibodies that were used forflow cytometry. The protein eIF4Ewas used as protein loading control.C, confirmation of b1- and b2-adrenergic receptor expression byPCR.ESCL, embryonic stemcell line;PBMC, peripheral bloodmononuclear cells; PSMC,pulmonary smooth muscle cells;NAC, no amplification control. Thestandard shown for the PCRproducts is a 100-bp ladder.

MDA-MB-231

Flourescence intensity

Eve

nts

Eve

nts

01

28

01

28

01

28

01

28

01

28

01

28

Eve

nts

Eve

nts

100 101 102 103 104 100 101 102 103 104 100 101 102 103 104

100 101 102 103 104100 101 102 103 104100 101 102 103 104

Flourescence intensity Flourescence intensity

Flourescence intensityFlourescence intensityFlourescence intensity

MFIISO = 3.36MFIβ1AR = 16.81




MDA-MB-435SA MDA-MB-468


MFIISO = 9.16MFIβ2AR= 27.57

B72 kDa

β2AR

231 435S 468

eIF4E

25 kDa

β1ARC

231 435S 468

72 kDa

eIF4E

25 kDa

β1AR

β2AR

β-Actin

ES

CL

500 bp

100 bp

500 bp

100 bp

500 bp

100 bp

PB

MC

PS

MC

23

1

46

8

435S

PC

-3

NA

C

HM

VE

C

ES

CL

PB

MC

PS

MC

23

1

46

8

435S

PC

-3

NA

C

HM

VE

C

ES

CL

PB

MC

PS

MC

23

1

46

8

435S

PC

-3

NA

C

HM

VE

C






100,000 cells; P ¼ 0.016; Fig. 5B, left). Accordingly, therelative expression of the relevant genes wasmore than 4-foldincreased (GROa: 4.67 and IL-8: 4.41), as was assessed byquantitative real-time PCR (Fig. 5B, right). The release ofboth chemokines wasmediated by b-adrenergic receptors, asa blockade of these receptors using propranolol stronglyreduced the release (GROa: 72.1 � 2.4 pg/h/100,000 cellsand IL-8: 100.5� 1.1; P¼ 0.032). With regard to GROa,this reduction was not significant due to the high SD of thenorepinephrine sample.GROa and IL-8 bind to the same chemokine receptors,

which are CXCR1 and CXCR2. PCR analysis revealed thatthe RNA for both receptors is expressed in high amounts inMDA-MB-231 cells, whereas the MDA-MB-435S and

MDA-MB-468 cells showed considerable less amounts(Fig. 6A). Confirmatory experiments were carried out withPC-3 human prostate carcinoma cells, which showed overallless adhesion to the endothelium, but a similar increase inresponse to norepinephrine from 5,004� 1,785 to 8,663�2,363 pixels (Fig. 6B, left). Furthermore, these cells expresshigh amounts of both chemokine receptors CXCR1 andCXCR2 too (Fig. 6B right). We thus concluded that theMDA-MB-231 cells should be susceptible to the chemo-kines' signaling and hypothesized that the proadhesive effectof norepinephrine might be transmitted by these chemo-kines. Therefore, we inhibited the function of these recep-tors on the MDA-MB-231 cells by preincubation withblocking antibodies and subsequently subjected the cells

AMDA-MB-231

0

20

40

60

80

100

9007205403601800

Time (min)

Lo

co

mo

tin

g c

ells

(%

)

Control 10 µmol/L norepinephrine

0

20

40

60

80

100

9007205403601800

Time (min)

Lo

co

mo

tin

g c

ells

(%

)


0

20

40

60

80

100

NorControl

0

20

40

60

80

100

NorControl

BMDA-MB-435S

CMDA-MB-468

0

20

40

60

80

100

9007205403601800

Time (min)

Lo

co

mo

tin

g c

ells

(%

)


0

20

40

60

80

100

NorControl

Figure 3. Migration of (A) MDA-MB-231, (B) MDA-MB-435S, and (C)MDA-MB-468 cells in response tonorepinephrine. Each graph showson the left side, the time course ofthe mean migratory activity and onthe right side, the time average ofthe migratory activity as meanvalue and SD of the sameexperiments. Three independentexperiments were carried out witheach cell line (in total, 90 cells wereanalyzed per sample). Nor,norepinephrine.

Strell et al.





to the flow-through adhesion assay (Fig. 6C). Such areceptor blockade abrogated the norepinephrine effect. Inthe control, there was no difference between the cells treatedwith either an isotypic control antibody (IgG2a; 16,390 �3,243 pixels) or anti-CXCR1 and anti-CXCR2 antibodies(aCXCR1/2; 13,311 � 3,136 pixels). In contrast, norepi-nephrine treatment led to a significant increase of adhesionin IgG2a-treated cells (25,966 � 3,505 pixels; P < 0.001),which was significantly reduced down to the control level bythe chemokine receptor blockade (12,614 � 2,039 pixels;P< 0.001; Fig. 6C). In turn, overlay of the endotheliumwithGROa significantly (P < 0.001) induced adhesion (from10,158� 3,914 to 21,829� 4,173 pixels; Fig. 6D), whereas

IL-8 had only a very weak and nonsignificant effect (12,123� 3,589). Accordingly, the combination of both chemo-kines was of similar efficacy as GROa alone (19,866 �2,266; P ¼ 0.011; Fig. 6D).Subsequently, we investigated the molecular basis of this

interaction between MDA-MB-231 cells and HMVECs. Itis well known that integrins are important adhesion mole-cules for the extravasation of tumor cells (23). Pretreatmentof the MDA-MB-231 cells with the b1-integrin–blockingantibody 4B4 significantly (P ¼ 0.011) inhibited the nor-epinephrine-induced adhesion from 27,043 � 5,917 to18,383 � 3,299 pixels, but not the spontaneous adhesionof the control [15,028 � 5,033 (IgG1 isotype control) vs.16,110 � 5,320 pixels (4B4); Fig. 7A].

B(MDA-MB-435S)

A(MDA-MB-468)

Control

Control 0.1 1 10 100

Nor Control Nor

50

40

30

20

10

0

50

40

30

20

10

0

50

40

30

20

10

0

D (MDA-MB-231)

Norepinephrine (µmol/L)

**

*

* *

Pix

els

(x1,0

00)

Pix

els

(x1,0

00)

Pix

els

(x1,0

00)

Pix

els

(x1,0

00)

C (MDA-MB-231)

50

40

30

20

10

0

Control Nor Prop Nor + Prop

Figure 4. Adhesion of the tumor cell lines to endothelium under flowconditions. The field that was covered by adhesive tumor cells wascalculated in pixels as explained in the Materials and Methods section.Eachexperimentwascarried out independently 3 times;mean values andSDs are shown. �, statistically significant changes (P < 0.05). A, B, D,norepinephrine (Nor) was used at 10 mmol/L. D, propranolol (Prop) wasused at 10 mmol/L.

β1AR

MFIISO = 2.91

MFIβ1AR = 5.70

MFIISO = 2.91

MFIβ2AR = 6.66

β2ARA

B

* *

*

GROα geneControl Nor Prop Nor + Prop

12

10

8

6

4

2

0

200

150

100

50

0

200

150

100

50

0

12

10

8

6

4

2

0

Control Nor Prop Nor + Prop

GR

Oα

(pg/h

/100,0

00 c

ells

)IL

-8 (p

g/h

/100,0

00 c

ells

)

IL-8 gene

Rela

tive e

xpre

ssio

nR

ela

tive e

xpre

ssio

n


Eve

nts

0128

100 101 102 103 104


Eve

nts

0128

100 101 102 103 104

Figure 5. Expression of b-adrenergic receptors (A) and influence ofnorepinephrine on the chemokine release (B) of HMVECs. A, theexpression of the b1- and b2-adrenergic receptor (AR) in HMVECs wasanalyzed by flow cytometry. The shown graphs are representative for 3independent measurements. B, the release of the chemokines GROaand IL-8 was investigated by an enzyme-linked immunoassay.Norepinephrine (Nor) and propranolol (Prop) were used at 10 mmol/L. Inaddition, the expression level of the relevant genes was assessed byquantitative real-time PCR. The results of the immunoassay are shown asmean values and SDs of 3 independent experiments, the PCR analysis isshown as box plot of a single experiment. �, statistically significantchanges (P < 0.05). ISO, isotypic.






Furthermore, inhibition of endothelial NF-kB led to aslight, however not statistically significant reduction ofthe adhesion of the control cells from 16,872 � 3,643 to13,080� 798 pixels, but significantly (P ¼ 0.002) reducedthe norepinephrine-mediated adhesion from 28,285 �5,074 to 16,779 � 4,504 pixels (Fig. 7B).

Discussion

Proinflammatory chemokines are supposed to playan important role in tumor progression and metastasisformation. Several animal studies gave evidence thatblocking CXCR1 and especially CXCR2 signaling, eitherby use of neutralizing antibodies or small-molecule inhi-bitors, inhibits colon cancer metastasis to liver (31, 32).One earlier study showed similar results for a CXCR1/CXCR2 blockade on lung metastasis development in arenal cell carcinoma model (33). But in this study, theanti-metastatic effect is explained by inhibition of tumor-associated angiogenesis. Direct effects of the CXCR1/CXCR2 blocking the extravasation of tumor cells werenot investigated in this context. With our in vitro flow-through adhesion assay, we show that GROa, and to alesser extent, IL-8, can promote the adhesion of CXCR1/CXCR2-positive tumor cells to endothelium. Chemo-

kines are released by endothelial cells in response to certainstimuli and then become bound on the apical side of theendothelial cell layer by glycosaminoglycans (34, 35).By this action, they are presented to rolling cells leadingto the rapid activation of integrins (17, 36). Accordingly,our own results show that overlay of the endothelium withthe chemokines GROa and, to a much lesser extent,IL-8 lead to a b1-integrin–mediated increase in the adhe-sion of MDA-MB-231. Therefore, we provide in vitroarguments for a possible involvement of GROa in reg-ulating integrin-mediated adhesion of MDA-MB-231cells, which might have parallels to the known role ofproinflammatory chemokines in the extravasation ofleukocytes. Our work complements recent results by Yaoand colleagues showing that the inhibition of the a5b1-integrin prevents the extravasation of MDA-MB-231 cellsin the lungs of athymic nude mice (37). Similar effectsof a5b1-integrin blocking on lung metastasis were shownin other studies for a metastatic rat prostate cancer cell line(38) and a metastatic murine melanoma cell line (39).These and our data point toward an important role ofb1-integrin in lung metastasis.The following question is why or in response to which

stimuli these chemokines are released. It is well knownthat the generation of inflammatory conditions comprises

C

**

αCXCR1/2αCXCR1/2IgG2a IgG2a

10 µmol/L norepinephrine

Pix

els

(x1,0

00)

50

40

30

20

10

0

Pix

els

(x1,0

00)

50

40

30

20

10

0

D

GROα GROα+IL-8IL-8Control

**

β-Actin

A

231 435 468

CXCR1

231 435 468231 435 468

CXCR2

B

500 bp

100 bp

0

5

10

15

Control Nor

Pix

el (x

1,0

00

) β-A

ctin

CX

CR

1

CX

CR

2

Figure 6. Expression and function of the chemokine receptors CXCR1 andCXCR2 in tumor cell adhesion to HMVECs under flow conditions. A, the expressionof the chemokine receptors CXCR1 andCXCR2was investigated by PCR. b-Actin served as loading control. The standard is a 100-bp ladder. B, confirmatoryexperiments were carried out with PC-3 human prostate carcinoma cells. Left, adhesion of the cells to HMVECs under flow conditions; right, PCR ofCXCR1 and CXCR2 expression. C and D, adhesion of MDA-MB-231 cells to HMVECs under flow conditions. C, the cells were pretreated with antibodiesblocking the chemokine receptors (aCXCR1/2) or an isotypic control antibody (IgG2a). D, the endotheliumwas overlaidwithGROa and IL-8 at a concentrationof 1 mg/mL for 3 minutes before the experiment. Denaturated chemokines were used as control. Graphs in (C) and (D) show mean values and SDs of3 independent experiments. �, statistically significant changes (P < 0.05). Nor, norepinephrine.

Strell et al.





the release and action of chemokines and that suchinflammatory conditions contribute to the progression ofcancer (15, 16). However, in these conditions, the chemo-kines are largely provided by leukocytes. It is well docu-mented that the vasculature is, at least in part, innervatedby noradrenergic sympathetic nerves (40–45), and asshown herein, norepinephrine induces the release ofGROa and IL-8 from the vascular endothelium. Thisinduction by norepinephrine is mediated via b-adrenergicreceptors, as the blockade of these receptors by theclinically established b-blocker propranolol significantlyreduced the chemokine release. We have also shown anupregulation of the gene expression level of IL-8 andGROa in response to norepinephrine incubation. Theexpression of the chemokines GROa, IL-8, and otherinflammatory chemokines is under the control of NF-kB(46), which also holds true for the herein used HMVECs.NF-kB mediates the PPAR-induced upregulation of IL-8in HUVECs (47). Several previous studies have shownfor distinct cell types that norepinephrine is an activator

of NF-kB (48) and thereby regulates the expression ofproinflammatory chemokines and cytokines (49, 50).In accordance with these findings, the addition of aNF-kB activation inhibitor reduced the norepinephrineinduced adhesion of MDA-MB231 to endothelium in ourflow-through adhesion assay. Therefore, our study addsfurther evidence of a direct effect of norepinephrine onendothelial cells, in addition to the recent observationthat b2-adrenoceptor–deficient mice show dysfunctionalangiogenesis (51), which possibly explains why b-blockersare effective in the treatment of hemangiomas (52).In summary, our study shows 3 possible points to

inhibit the norepinephrine-induced adhesion of MDA-MB231 breast carcinoma cells to lung endothelium: (i)inhibition of norepinephrine signaling by b-blockers, (ii)blocking of b1-integrins, and (iii) inhibition of chemokinesignaling/release. Among these possibilities, b-blockers areclinically well established in long-term hypertension ther-apy, and epidemiologic studies confirm that they have noeffect in causing or promoting cancer (53). Furthermore,we and others have reported previously by mice studies(5, 14), and retrospective patient data analysis (10–12),that the use of b-blockers inhibits metastasis formation.One molecular explanation for these observations wereour in vitro migration studies, in which the migratoryactivity of carcinoma cells from breast (2), prostate (4),and colonic (3) tissue origin increased by treatment withnorepinephrine; this effect was inhibited by b-blockers aswell. In addition, other groups have shown evidence thatpropranolol inhibits the norepinephrine-induced migra-tion and invasion of pancreatic (6) and ovarian (7) cancercells. However, metastasis formation is a multistep pro-cess, which requires not only the induction of migratoryactivity (54) but also the ability of tumor cells to enter theblood stream and extravasate at the site of metastasisdevelopment; at least in hematogenic dissemination(55). Here, we provide a further molecular explanationfor how b-blockers can inhibit metastasis formation by theattenuation of the b1-integrin–mediated adhesion oftumor cells to the vascular endothelium.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interests were disclosed.

Acknowledgments

The authors thank the B. Braun Melsungen AG for generously leaving theperfusion pump to us.

Grant Support

This work was supported by the Bruno and Helene J€oster Foundation (Cologne,Germany) and the Fritz Bender Foundation (Munich, Germany).

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 March 18, 2011; revised November 14, 2011; accepted November 21,2011; published OnlineFirst November 29, 2011.

A

**

10 µmol/L norepinephrine

IgG1 IgG1

Control NF-κB Inh NF-κB

Inh + Nor

Nor

4B4 4B4

B

**

Pix

els

(x1

,00

0)

Pix

els

(x1

,00

0)

50

40

30

20

10

0

50

40

30

20

10

0

Figure 7. Inhibition of b1-integrins and NF-kB abrogates the effect ofnorepinephrine on MDA-MB-231 cell adhesion. A, MDA-MB-231cells were pretreated with either the b1-integrin–blocking antibody4B4 or an isotypic control antibody (IgG1). B, HMVECs werepreincubated with an NF-kB activation inhibitor (NF-kB Inh);norepinephrine (Nor) was used at 10 mmol/L. Both graphs showmean values and SDs of 3 independent experiments. �, statisticallysignificant changes (P < 0.05).






References1. Sporn MB. The war on cancer. Lancet 1996;347:1377–81.2. Drell TL, Joseph J, Lang K, Niggemann B, Zaenker KS, Entschladen F.

Effects of neurotransmitters on the chemokinesis and chemotaxis ofMDA-MB-468 humanbreast carcinomacells. Breast Cancer Res Treat2003;80:63–70.

3. Masur K, Niggemann B, Zanker KS, Entschladen F. Norepinephrine-induced migration of SW 480 colon carcinoma cells is inhibited bybeta-blockers. Cancer Res 2001;61:2866–9.

4. Lang K, Drell TL, Lindecke A, Niggemann B, Kaltschmidt C, ZaenkerKS, et al. Induction of a metastatogenic tumor cell type by neuro-transmitters and its pharmacological inhibition by established drugs.Int J Cancer 2004;112:231–8.

5. Palm D, Lang K, Niggemann B, Drell TL IV, Masur K, Zaenker KS, et al.The norepinephrine-driven metastasis development of PC-3 humanprostate cancer cells in BALB/c nude mice is inhibited by beta-blockers. Int J Cancer 2006;118:2744–9.

6. Guo K, Ma Q, Wang L, Hu H, Li J, Zhang D, et al. Norepinephrine-induced invasion by pancreatic cancer cells is inhibited by propranolol.Oncol Rep 2009;22:825–30.

7. Sood AK, Bhatty R, Kamat AA, Landen CN, Han L, Thaker PH, et al.Stress hormone-mediated invasion of ovarian cancer cells. Clin Can-cer Res 2006;12:369–75.

8. Williams JB, Pang D, Delgado B, Kocherginsky M, Tretiakova M,Krausz T, et al. A model of gene-environment interaction revealsaltered mammary gland gene expression and increased tumor growthfollowing social isolation. Cancer Prev Res (Phila) 2009;2:850–61.

9. Wu W, Murata J, Hayashi K, Yamaura T, Mitani N, Saiki I. Socialisolation stress impairs the resistance of mice to experimental livermetastasis of murine colon 26-L5 carcinoma cells. Biol Pharm Bull2001;24:772–6.

10. PoweDG, VossMJ, Zaenker KS, HabashyHO,GreenAR, Ellis IO, et al.Beta-blocker drug therapy reduces secondary cancer formation inbreast cancer and improves cancer specific survival. Oncotarget2010;1:628–38.

11. Barron TI, Connolly RM, Sharp L, Bennett K, Visvanathan K. Betablockers and breast cancer mortality: a population-based study. J ClinOncol 2011;29:2635–44.

12. Melhem-Bertrandt A, Chavez-Macgregor M, Lei X, Brown EN, Lee RT,Meric-Bernstam F, et al. Beta-blocker use is associatedwith improvedrelapse-free survival in patients with triple-negative breast cancer. JClin Oncol 2011;29:2645–52.

13. Powe DG, Voss MJ, Habashy HO, Z€anker KS, Green AR, Ellis IO,et al. Alpha- and beta-adrenergic receptor (AR) protein expressionis associated with poor clinical outcome in breast cancer: animmunohistochemical study. Breast Cancer Res Treat 2011;130:457–63.

14. Sloan EK, PricemanSJ, CoxBF, YuS, PimentelMA, TangkanangnukulV, et al. The sympathetic nervous system induces a metastatic switchin primary breast cancer. Cancer Res 2010;70:7042–52.

15. Balkwill F, Mantovani A. Inflammation and cancer: back to Virchow?Lancet 2001;357:539–45.

16. Mantovani A, Allavena P, Sica A, Balkwill F. Cancer-related inflamma-tion. Nature 2008;454:436–44.

17. Laudanna C, Kim JY, Constantin G, Butcher E. Rapid leukocyteintegrin activation by chemokines. Immunol Rev 2002;186:37–46.

18. Shamri R, Grabovsky V, Gauguet JM, Feigelson S, Manevich E,Kolanus W, et al. Lymphocyte arrest requires instantaneous inductionof an extended LFA-1 conformation mediated by endothelium-boundchemokines. Nat Immunol 2005;6:497–506.

19. Kakinuma T, Hwang ST. Chemokines, chemokine receptors, andcancer metastasis. J Leukoc Biol 2006;79:639–51.

20. Singh S, Singh AP, Sharma B, Owen LB, Singh RK. CXCL8 and itscognate receptors in melanoma progression and metastasis. FutureOncol 2010;6:111–6.

21. OgataH,SekikawaA,YamagishiH, IchikawaK, TomitaS, Imura J, et al.GROalpha promotes invasion of colorectal cancer cells. Oncol Rep2010;24:1479–86.

22. Kin NW, Sanders VM. It takes nerve to tell T and B cells what to do. JLeukoc Biol 2006;79:1093–104.

23. Strell C, Entschladen F. Extravasation of leukocytes in comparison totumor cells. Cell Commun Signal 2008;6:10.

24. Alon R. Chemokine arrest signals to leukocyte integrins trigger bi-directional-occupancy of individual heterodimers by extracellular andcytoplasmic ligands. Cell Adh Migr 2010;4:211–4.

25. Strell C, Lang K, Niggemann B, Zaenker KS, Entschladen F. Surfacemolecules regulating rolling and adhesion to endothelium of neutrophilgranulocytes and MDA-MB-468 breast carcinoma cells and theirinteraction. Cell Mol Life Sci 2007;64:3306–16.

26. Sakkas LI, Scanzello C, JohansonN,Burkholder J,Mitra A, SalgameP,et al. T cells and T-cell cytokine transcripts in the synovial membrane inpatients with osteoarthritis. Clin Diagn Lab Immunol 1998;5:430–7.

27. Schmausser B, Josenhans C, Endrich S, Suerbaum S, Sitaru C,Andrulis M, et al. Downregulation of CXCR1 and CXCR2 expressionon human neutrophils by Helicobacter pylori: a new pathomechanismin H. pylori infection? Infect Immun 2004;72:6773–9.

28. LaemmliUK.Cleavageof structural proteinsduring the assembly of thehead of bacteriophage T4. Nature 1970;227:680–5.

29. Yan L, Jia Z, Cui J, Yang H, Zhang Y, Zhou C. Beta-adrenergic signalsregulate cardiac differentiation of mouse embryonic stem cells viamitogen-activated protein kinase pathways. Dev Growth Differ 2011;53:772–9.

30. Strell C, Sievers A, Bastian P, Lang K, Niggemann B, Z€anker KS, et al.Divergent effects of norepinephrine, dopamine and substancePon theactivation, differentiation and effector functions of human cytotoxic Tlymphocytes. BMC Immunol 2009;10:62.

31. Varney ML, Singh S, Li A, Mayer-Ezell R, Bond R, Singh RK. Smallmolecule antagonists for CXCR2 and CXCR1 inhibit human coloncancer liver metastases. Cancer Lett 2011;300:180–8.

32. Yamamoto M, Kikuchi H, Ohta M, Kawabata T, Hiramatsu Y, Kondo K,et al. TSU68 prevents liver metastasis of colon cancer xenografts bymodulating the premetastatic niche. Cancer Res 2008;68:9754–62.

33. Mestas J, Burdick MD, Reckamp K, Pantuck A, Figlin RA, Strieter RM.The role of CXCR2/CXCR2 ligand biological axis in renal cell carcino-ma. J Immunol 2005;175:5351–7.

34. Kawashima H. Roles of sulfated glycans in lymphocyte homing. BiolPharm Bull 2006;29:2343–9.

35. Cinamon G, Grabovsky V, Winter E, Franitza S, Feigelson S, Shamri R,et al. Novel chemokine functions in lymphocyte migration throughvascular endothelium under shear flow. J Leukoc Biol 2001;69:860–6.

36. Constantin G,MajeedM,Giagulli C, Piccio L, KimJY, Butcher EC, et al.Chemokines trigger immediate beta2 integrin affinity and mobilitychanges: differential regulation and roles in lymphocyte arrest underflow. Immunity 2000;13:759–69.

37. Yao H, Veine DM, Fay KS, Staszewski ED, Zeng ZZ, Livant DL. ThePHSCN dendrimer as a more potent inhibitor of human breast cancercell invasion, extravasation, and lung colony formation. Breast CancerRes Treat 2011;125:363–75.

38. Livant DL, BrabecRK, PientaKJ, AllenDL, Kurachi K,Markwart S, et al.Anti-invasive, antitumorigenic, and antimetastatic activities of thePHSCN sequence in prostate carcinoma. Cancer Res 2000;60:309–20.

39. Saiki I, Murata J, Matsuno K, Ogawa R, Nishi N, Tokura S, et al. Anti-metastatic and anti-invasive effects of polymeric Arg-Gly-Asp (RGD)peptide, poly(RGD), and its analogues. Jpn J Cancer Res 1990;81:660–7.

40. Joyner MJ, Casey DP. The catecholamines strike back.What NO doesnot do. Circ J 2009;73:1783–92.

41. Felten DL, Felten SY, Carlson SL, Olschowka JA, Livnat S. Noradren-ergic and peptidergic innervation of lymphoid tissue. J Immunol1985;135:755s–65s.

42. Head RJ. Hypernoradrenergic innervation and vascular smooth mus-cle hyperplastic change. Blood Vessels 1991;28:173–8.

43. Lincoln J. Innervation of cerebral arteries by nerves containing 5-hydroxytryptamine and noradrenaline. Pharmacol Ther 1995;68:473–501.

44. Rahn KH, Barenbrock M, Hausberg M. The sympathetic nervoussystem in the pathogenesis of hypertension. J Hypertens Suppl1999;17:S11–4.

Strell et al.





45. Wallin BG. Human sympathetic nerve activity and blood pressureregulation. Clin Exp Hypertens A 1989;11 Suppl 1:91–101.

46. Son DS, Parl AK, Rice VM, Khabele D. Keratinocyte chemoattractant(KC)/human growth-regulated oncogene (GRO) chemokines and pro-inflammatory chemokine networks in mouse and human ovarian epi-thelial cancer cells. Cancer Biol Ther 2007;6:1302–12.

47. Meissner M, Hrgovic I, Doll M, Naidenow J, Reichenbach G, Hailemar-iam-Jahn T, et al. Peroxisome proliferator-activated receptor {delta}activators induce IL-8 expression in nonstimulated endothelial cells ina transcriptional and posttranscriptional manner. J Biol Chem2010;285:33797–804.

48. Minneman KP, Lee D, Zhong H, Berts A, Abbott KL, Murphy TJ.Transcriptional responses to growth factor and G protein-coupledreceptors in PC12 cells: comparison of alpha(1)-adrenergic receptorsubtypes. J Neurochem 2000;74:2392–400.

49. Flierl MA, RittirschD, NadeauBA, Sarma JV, DayDE, Lentsch AB, et al.Upregulation of phagocyte-derived catecholamines augments theacute inflammatory response. PLoS One 2009;4:e4414.

50. Walsh DE, Greene CM, Carroll TP, Taggart CC, Gallagher PM, O'NeillSJ, et al. Interleukin-8 up-regulation by neutrophil elastase ismediated

by MyD88/IRAK/TRAF-6 in human bronchial epithelium. J Biol Chem2001;276:35494–9.

51. Ciccarelli M, Sorriento D, Cipolletta E, Santulli G, Fusco A, Zhou RH,et al. Impaired neoangiogenesis in beta-adrenoceptor gene-deficientmice: restoration by intravascular human beta-adrenoceptor genetransfer and role of NFkappaB and CREB transcription factors. Br JPharmacol 2011;162:712–21.

52. Leaute-Labreze C, Dumas de la Roque E, Hubiche T, Boralevi F,Thambo JB, Taieb A. Propranolol for severe hemangiomas of infancy.N Engl J Med 2008;358:2649–51.

53. Bangalore S, Kumar S, Kjeldsen SE, Makani H, Grossman E, Wetter-slev J, et al. Antihypertensive drugs and risk of cancer: network meta-analyses and trial sequential analyses of 324,168 participants fromrandomised trials. Lancet Oncol 2011;12:65–82.

54. Entschladen F, Drell TL, Lang K, Joseph J, Zaenker KS. Tumour-cellmigration, invasion, and metastasis: navigation by neurotransmitters.Lancet Oncol 2004;5:254–8.

55. Wong SY, Hynes RO. Lymphatic or hematogenous dissemination:how does a metastatic tumor cell decide? Cell Cycle 2006;5:812–7.






2012;10:197-207. Published OnlineFirst November 29, 2011.Mol Cancer Res Carina Strell, Bernd Niggemann, Melanie J. Voss, et al.

ReleaseαGROMDA-MB-231 Cells to Vascular Endothelium by the Induction of a

Mediated Adhesion of−1-IntegrinβNorepinephrine Promotes the

Updated version

10.1158/1541-7786.MCR-11-0130doi:

Access the most recent version of this article at:

Cited articles

http://mcr.aacrjournals.org/content/10/2/197.full#ref-list-1

This article cites 55 articles, 12 of which you can access for free 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/10/2/197To 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-11-0130

http://mcr.aacrjournals.org/content/10/2/197.full#ref-list-1

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/10/2/197


norepinephrine promotes the b1-integrin mediated adhesion ... · process is an essential...

Documents