Definition of Molecular Determinants of Prostate Cancer CellBone Extravasation
Steven R. Barthel1,2, Danielle L. Hays1, Erika M. Yazawa1, Matthew Opperman1, Kempland C. Walley1,Leonardo Nimrichter3, Monica M. Burdick4, Bryan M. Gillard5, Michael T. Moser5, Klaus Pantel6,Barbara A. Foster5, Kenneth J. Pienta7, and Charles J. Dimitroff1,2
AbstractAdvanced prostate cancer commonly metastasizes to bone, but transit of malignant cells across the bone
marrow endothelium (BMEC) remains a poorly understood step in metastasis. Prostate cancer cells roll on E-selectinþBMEC throughE-selectin ligand-binding interactions under shearflow, and prostate cancer cells exhibitfirm adhesion to BMEC via b1, b4, and aVb3 integrins in static assays. However, whether these discrete prostatecancer cell–BMEC adhesive contacts culminate in cooperative, step-wise transendothelial migration into bone isnot known. Here, we describe how metastatic prostate cancer cells breach BMEC monolayers in a step-wisefashion under physiologic hemodynamic flow. Prostate cancer cells tethered and rolled on BMEC and then firmlyadhered to and traversed BMEC via sequential dependence on E-selectin ligands and b1 and aVb3 integrins.Expression analysis in human metastatic prostate cancer tissue revealed that b1 was markedly upregulatedcompared with expression of other b subunits. Prostate cancer cell breaching was regulated by Rac1 and Rap1GTPases and, notably, did not require exogenous chemokines as b1, aVb3, Rac1, and Rap1 were constitutivelyactive. In homing studies, prostate cancer cell trafficking tomurine femurswas dependent onE-selectin ligand,b1integrin, and Rac1. Moreover, eliminating E-selectin ligand-synthesizing a1,3 fucosyltransferases in transgenicadenoma of mouse prostate mice dramatically reduced prostate cancer incidence. These results unify therequirement for E-selectin ligands, a1,3 fucosyltransferases, b1 and aVb3 integrins, and Rac/Rap1 GTPasesin mediating prostate cancer cell homing and entry into bone and offer new insight into the role of a1,3fucosylation in prostate cancer development. Cancer Res; 73(2); 1–11. �2012 AACR.
IntroductionMetastatic prostate cancer killed 33,720men in 2011, and the
prognosis with bonemetastasis is poor (1). Unfortunately, howprostate cancer cells breach bone marrow microvessels is ill-defined. Accordingly, we established an adhesion paradigmwherein prostate cancer cells roll on bone marrow endothe-lium (BMEC) under physiologic hemodynamic flow by recog-nizing BMEC endothelial (E)-selectin (2–5). This led to thepremise that bone metastasis is conferred by prostate cancer
cell E-selectin ligand–BMEC E-selectin binding and has beensupported by several observations. First, human bone marrowmicrovessels express E-selectin constitutively (6). Second,E-selectin-binding sLeX membrane glycoproteins and glycoli-pids, defined operationally as E-selectin ligands, and regulatorsof E-selectin ligand synthesis, a1,3 fucosyltransferases (a1,3FT) 3, 6, and 7, are upregulated in bone-metastatic prostatecancer cells (4) and prostate cancer lesions in bone (5) and areassociated with elevated prostate cancer grade, progression,and bone metastasis (2, 3, 7–9). Third, E-selectin ligandþ
prostate cancer cells home more efficiently than E-selectinligand� cells to bone, and homing is neutralized with anti-E-selectin antibody (5).
The putative role of E-selectin ligands in prostate cancerbone metastasis was first conjectured by earlier evidence inmice that E-selectin ligandþ hematopoietic stem cells (HSC)bound and traversed bone marrow microvessels via rolling onE-selectinþ bone marrow microvessels (10–12). In fact, HSCextravasation into murine bone involved the E-selectin ligand,PSGL-1 (12, 13). Other evidence indicated that integrins a4b1and a4b7 contributed to initial HSC rolling on VCAM-1 andMAdCAM-1 (12, 14) expressed constitutively on BMEC (6, 14).HSCs then firmly adhere and traverse BMEC via contributionsfrom CD44, a4b1, a5b1, a6b1, aLb2, and SDF-1-CXCR4 sig-naling (11, 15–18). Homing to and engraftment in bonemarrowalso involves HSC Rho and Ras GTPases, Rac1 and Rap1
Authors' Affiliations: 1Department of Dermatology, Brigham andWomen's Hospital; 2Harvard Medical School, Boston, Massachusetts;3Instituto deMicrobiologia Professor Paulo de Goes, Universidade Federaldo Rio de Janeiro, Rio de Janeiro, Brazil; 4Department of Chemical andBiomolecular Engineering, Biomedical Engineering Program, Ohio Univer-sity, Athens, Ohio; 5Department of Pharmacology and Therapeutics, Ros-well Park Cancer Institute, Buffalo, New York; 6Institute of Tumor Biology,University Medical Center Hamburg-Eppendorf, Hamburg, Germany; and7Department of Urology, University of Michigan Comprehensive CancerCenter, Ann Arbor, Michigan
Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).
Corresponding Author: Charles J. Dimitroff, HIM, Rm. 662, 77 AvenueLouis Pasteur, Boston, MA 02115. Phone: 617-525-5693; Fax: 617-525-5571; E-mail: [email protected]
(19, 20). Accordingly, E-selectin ligands, integrins, andGTPasesare candidate molecules regulating prostate cancer extrava-sation into bone.
While a4b1, a4b7, and aLb2 integrins have not beenfound on prostate cancer cells, a2b1, a3b1, a5b1, a6b1,aVb1, aVb3, aIIbb3, and a6b4 have been identified (21). All8 integrins seem constitutively active, although they can beregulated partly by chemokines SDF-1 or CCL2, which mayenhance prostate cancer cell adherence and migration/inva-sion on endothelial and basement membrane integrinligands under static conditions (22–26). Prostate cancer celladhesion to BMEC, bone marrow stroma, or endotheliumunder static conditions, in fact, is blocked with anti-b1 or -b3monoclonal antibody (mAb) or with Arg-Gly-Asp (RGD)peptide (27–29), and prostate cancer cell transendothelialmigration (TEM) through vascular endothelium in staticassays can be attenuated by anti-b3 mAb (30). Whetherthese prostate cancer integrins cooperate with E-selectinligands to elicit step-wise rolling, adhesion, and TEM underhemodynamic conditions analogous to the HSC molecularparadigm is unknown.
Here, we describe a step-wise molecular dependency ofprostate cancer cells to adhere and breach BMEC monolayersunder physiologic blood flow. Using microfluidics models, weobserved E-selectin ligand-dependent prostate cancer celltethering and rolling on E-selectinþ BMEC, transition to firmadhesion, and breaching BMECmonolayers via dependency onb1 and aVb3 integrins. However, prostate cancer cell breach-ing did not require exogenous chemokines or G-proteinreceptor signaling, which was likely circumvented throughconstitutively active b1 and aVb3 integrins and GTPases, Rac1and Rap1. Moreover, aLb2 and a4b1 integrins, characteristi-cally involved in HSC TEM, were not required for prostatecancer cell breaching. In addition, E-selectin ligandþ prostatecancer cells homed more efficiently to murine femurs than E-selectin ligand� cells in a b1 integrin- and Rac1-dependentmanner. Finally, transgenic adenoma of mouse prostate(TRAMP) mice deficient in E-selectin ligand-synthesizinga1,3 FT genes, FT4 and FT7, exhibited reduced prostate cancerincidence, implicating a1,3 fucosylation in transforming activ-ity of prostate epithelia and/or in tumorigenic regulation viatumor–tumor or tumor–host/stroma interactions. Theseresults unify requisite factors required for prostate cancer cellextravasation and offer new insight on the role of a1,3 FTs inprostate cancer development.
Materials and MethodsCells and tissues
Human prostate cancer cells, primary normal prostateepithelial cells, and leukemic KG1a cells were cultured asdescribed (2–4), and human bone-metastatic PC-E1 and PC-R1 cells were maintained in RPMI-1640/glutamine/10% FBS/1% penicillin/streptomycin (Life Technologies). All cell lineswere tested and authenticated by morphologic observationand by quantitative real-time PCR (qRT-PCR), Western blot-ting, and flow cytometry of glycosyltransferase genes/geneproducts as previously described (2–5).
Flow cytometryFlow cytometry was conducted as described (4), antibody
concentrations were 10 mg/mL, and conformation anti-b1antibodies, N29, HUTS-21, and 9EG7 were incubated 1 hourat 37�C.
Real-time PCRReal-time PCRwas conducted (4), and primer sequences are
shown in Supplementary Table S1.
Western blottingWestern blots were conducted as described (4, 5).
Prostate cancer rolling and breaching in flowRolling frequency and velocity was measured as described
(2, 5, 31). For breaching through endothelial cells (EC), prostatecancer cells (1� 106 cells in RPMI or endothelial growth media)were infused at 1.5 mL/min for 35 seconds, and a physiologicshear stress of 0.5 Dynes/cm2 was maintained for 4 hours.Photographs were taken every second for the first 4 minutesto capture rolling and then every 5 minutes to capture adhesionand breaching. Temperaturewasmaintained at 37�C via heat fanand water bath. Breaching, defined as a prostate cancer cellpiercing and penetrating the EC layer, was also investigated withBioFlux microfluidics technologies (Fluxion Biosciences). Whereindicated, prostate cancer cells and ECs were loaded with 0.5mmol/L CellTracker Green or CellTracker Red CMTPX (LifeTechnologies), respectively, and imagedby confocalfluorescencemicroscopy (Dana-Farber Cancer Institute, Boston, MA).
Statistical analysisResults were analyzed by 2-tailed t test, one-way ANOVA
with Dunnett posttest, or contingency table on GraphPadPrism (GraphPad Software).
ResultsProstate cancer PC-E1 and PC-R1 cells traverse BMECand prominently express b1 and aVb3 integrins
To identify prostate cancer cells with robust TEMactivity forsubsequent molecular analysis, we first examined the relativeefficiency of TEMusing a number ofmetastatic prostate cancercell lines. Using a static approach and a 16-hour incubationperiod, we found that bone metastatic PC-E1 and PC-R1 cellsshowed a higher capacity than other prostate cancer cells tobreach confluent human umbilical vein endothelial cell(HUVEC) andHBMEC-60monolayers, traverse 8micron pores,and attach to insert undersides (Fig. 1A and B).
To identify integrin–integrin ligand interactions controllingTEMof PC-E1 and PC-R1 cells, we surveyed expression ofb anda integrin subunits and found that b1 and b3 were mostprominent (Fig. 1C). In fact, qRT-PCR analysis of b chains inPC-E1 and PC-R1 cells and in prostate cancer tissue confirmedthat b1 transcript level was markedly higher than all other bsubunits and, at minimum, several 100-fold higher thaneven the next most highly expressed b subunit (Fig. S1A–B).We also found that b3 was elevated on PC-E1 and PC-R1 cells(Fig. 1C) relative to normal prostate epithelial (NPE) cells
Barthel et al.
Cancer Res; 73(2) January 15, 2013 Cancer ResearchOF2
(Supplementary Fig. S1C). Candidate a partners for b1 on PC-E1 and PC-R1 included a2, a5, a6, and aV, and for b3, aV (Fig.1C andD).aVb3Heterodimerwas present onPC-E1 and PC-R1cells, whereas b2, b4, b5, b6, b7, and b8 were undetectable onPC-E1 and PC-R1 cells (Fig. 1C) despite prominent b4 and b6expression on NPE (Supplementary Fig. S1C). Of note, a4 wasupregulated on PC-E1 cells, the first evidence of a4b1 on aprostate cancer cell (Fig. 1C). Data underscored 4 putative b1heterodimers, a2b1, a5b1, a6b1 and aVb1, and a b3 hetero-dimer, aVb3, as most highly expressed on PC-E1 and PC-R1cells (Fig. 1D). To strengthen these findings, we also assessedintegrin expression on a number of other prostate cancer cellsand found that b1 and aVb3 were prominently expressed onbone-metastatic prostate cancer PC-3M LN4, PC-3, and MDAprostate cancer 2b cell lines and on the brain-metastaticprostate cancer DU-145 cell line, though reduced on lymphnode-metastatic LNCaP and LNCaP C4-2b cells (Supplemen-tary Fig. S1C). These results suggested that high expression ofb1 and upregulation of aVb3 correlated with prostate cancerbone and brain metastasis.
b1 andaVb3 integrins onprostate cancer PC-R1 cells areconstitutively activeUsing b1þ and aVb3þ PC-R1 cells to conduct molecular
modeling of TEM, we first validated b1- and aVb3-dependentPC-R1 cell adhesion and found that PC-R1 cells bound to b1and aVb3 ligands, fibronectin (FN), laminin (LN), collagen I
(COL), and vitronectin (VN) without exogenous chemokinedependency (Fig. 2A). b1 and aVb3 Ligand binding was alsoobserved with PC-3, PC-3M LN4, PC-E1, LNCaP, LNCaP C4-2b,DU-145 prostate cancer cell lines (Supplementary Fig. S2).Unexpectedly, PC-R1 and PC-E1 cells also adhered stronglyto VCAM-1, which is the first demonstration of VCAM-1–dependent prostate cancer cell adhesion (Fig. 2A and Fig.S2A). As expected, PC-R1 cells, including all other prostatecancer cell lines, did not adhere to b2 ligand, ICAM-1 (Fig. 2Aand Supplementary Fig. S2A); and no adhesion to PECAM-1 orosteopontin was found (data not shown).
Given high avidity of PC-R1 cell adhesion to b1 and aVb3ligands independent of chemokine addition, we assessed theactivation state of b1 and aVb3 integrins using conformation-sensitive and blocking mAbs. All b1 conformation-sensitivemAbs recognizing increasingly active b1 structures, N29 <HUTS-21 < 9EG7 (32), reacted strongly with PC-R1 cells inbuffer devoid of exogenous chemokines (Fig. 2B). These qui-escent conditions typically do not permit for HUTS-21 and9EG7 mAb reactivity to b1 on resting leukocytes (33). Thesedata were confirmed with other prostate cancer cell lines andtheir reactivity to these conformation-sensitive antibodiesclosely correlated with their adhesiveness to b1 and aVb3integrin ligands (Supplementary Fig. S2B). In agreement withthis finding, PC-R1 cell adhesion to VCAM-1 and to COL wascompletely blocked by anti-b1mAb, whereas adhesion to FNorVN was significantly blocked with both anti-b1 and anti-aVb3
Figure 1. Prostate cancer PC-E1 and PC-R1 cells traverse BMEC and prominently express b1 and aVb3. A–B, number of cells that underwent TEM(n ¼ 9 � SEM; �, P < 0.05; ��, P < 0.01, vs. PC-3M LN4; one-way ANOVA with Dunnett posttest). C, anti-integrin antibody (open histogram) or isotype(shaded histogram); representative of n ¼ 3. D, predicted integrin heterodimers on PC-E1 and PC-R1 cells based on flow cytometric data of integrinsubunits.
Molecular Determinants for PCa Cell Breaching of BMEC
www.aacrjournals.org Cancer Res; 73(2) January 15, 2013 OF3
mAbs (Fig. 2C–F). Inhibition of adhesion to FNandVNwas alsoevidenced by pretreating PC-E1 cells with anti-b1 and anti-aVb3 mAbs (Supplementary Fig. S2C and D). Similar toinhibitory efficacy on purified ligands, we found that PC-R1and PC-E1 cell adhesion to HUVEC and HBMEC-60 cells wasreduced by 20% to 75% with anti-b1 and/or anti-aVb3 mAbs(Fig. 2G and H and Supplementary Fig. S2E).
Because Rap1 and Rac1 GTPases are known regulators of b1and b3 integrin-mediated adhesion on bone-homing HSCs,leukocytes, mesenchymal stem cells (MSC), and platelets (19,20, 34), we examined their role in b1 and b3-dependentprostate cancer cell adhesion. We found that b1-dependentPC-R1 cell adhesion to VCAM-1 was reduced by 75% with aRap1 antagonist, whereas no changewas observed using a Rac1antagonist (Fig. 2I). Importantly, Rap1 and Rac1 inhibitors didnot reduce PC-R1 cell adhesion to FN, COL, or VN, nor change
b1 or aVb3 expression or b1 structure as determined by N29,HUTS-21 or 9EG7mAb reactivity (data not shown). These datacollectively indicated that PC-R1 cell adhesion involved con-stitutively active b1 and aVb3 integrins, whereas a4b1 func-tion seemed to be partially regulated by Rap1-GTPase activity.
Development of a model for molecular analysis ofprostate cancer cell adhesion and migration underphysiologic shear flow
To examine the role of b1 and aVb3 integrins in prostatecancer cell adhesion in the more physiologic context,we needed to generate PC-R1 cells that could stably expressE-selectin ligands for initiating adhesion under shear flow. Inthat, prostate cancer cells require E-selectin ligands tobind endothelium in blood flow and characteristically loseE-selectin ligand expression along with a1,3 FTs required for
Figure2. b1andaVb3 integrins are constitutively active andmediatePC-R1cell adhesion toBMEC.A,%PC-R1cell adhesion to integrin ligands (n¼9�SEM).B, anti-b1 activation–sensitive antibodies (open histogram) or isotype (shaded histogram); n ¼ 3. C–I, PC-R1 cell adhesion to integrin ligands or ECmonolayers in the presence of blocking or isotype antibodies (n¼ 9� SEM; �, P < 0.05; ��, P < 0.01, vs. isotype or untreated; one-way ANOVA with Dunnettposttest).
Barthel et al.
Cancer Res; 73(2) January 15, 2013 Cancer ResearchOF4
synthesizing E-selectin–binding determinants (2–5), we firstgenerated PC-R1 cells stably expressing FT3, 6, or 7. Our priorstudies on FT3, 6, and 7 in prostate cancer tissue and in E-selectin ligandþ MDA prostate cancer 2b cells and PC-3 celltransfectants suggested that E-selectin ligandþ PC-R1 cellsthat express functionally active integrins would be an idealmodel for studying adhesion to and migration through vascu-lar endothelium under physiologic shear stress (4, 5). As such,FT3, 6, or 7were stably expressed in PC-R1 cells (PC-R1 FT) andresulted in upregulated sLeX levels (Fig. 3A). FT expression didnot change adhesion to integrin ligands, integrin expression, or
E-selectin glycoprotein ligand transcription (SupplementaryFig. S3A–E). Overexpression of FT3, FT6, or FT7 in PC-3M LN4and DU-145 cells also resulted in high sLeX expression (Sup-plementary Fig. S4A).
In cell-binding assays, PC-R1 FT cells and control E-selectinligandþ KG1a and MDA prostate cancer 2b cells adheredstrongly to E-selectin (Fig. 3B). Western blot and thin layerchromatographic analysis revealed candidate E-selectin–bind-ing determinants on PC-R1 FT glycoproteins and glycolipids(Supplementary Fig. S4B–C). E-selectin–binding glycoproteinswere identified previously in related prostate cancer FTþ cells
Figure 3. b1 andaVb3 integrins andRac/Rap1GTPases regulate TEMof E-selectin ligandþPC-R1cells throughBMEC.A, anti-sLeX antibody (openhistogram)or isotype (shaded histogram); representative of n¼ 3. B, adhesion of PC-R1 empty or FT cells along with positive control prostate cancer (PCa)MDAPCa 2bcells andKG1a cells to E- or P-selectin chimeras or to Fc control (n¼ 9�SEM). C andD, TEM in the presence or absence of blocking anti-b1 oraVb3mAbs orof isotype control antibodies (n¼ 16� SEM, 4 experiments; ��, P < 0.01; ���, P < 0.001, vs. isotype; one-way ANOVAwith Dunnett posttest). E, fluorescence-activated cell sorting analysis of b1 in scrambled (Scr) or b1 knockdown (KD) PC-R1 FT7 cells. Scr (open histogram, bold), b1KD (open histogram, dotted), andisotype antibody (shaded histogram). Mean fluorescence (n¼ 3�SEM; ���,P < 0.001, t test). F, TEMof PC-R1 FT7 (Scr) or b1KD (n¼ 3�SEM; ���,P < 0.001,t test). G, TEM of untreated or inhibitor-treated PC-R1 FT7 cells (n ¼ 9 � SEM; ��, P < 0.01; ���, P < 0.001, one-way ANOVA with Dunnett posttest).
Molecular Determinants for PCa Cell Breaching of BMEC
www.aacrjournals.org Cancer Res; 73(2) January 15, 2013 OF5
as CD44 (HCELL), carcinoembryonic antigen, melanoma celladhesion molecule, and podocalyxin-like protein (5). More-over, there were distinct PC-R1 FT glycosphingolipids (GSL)bearing sialofucosylatedmoieties reactive to E-selectin/Fc thatcomigrated with GM1 and GD3 gangliosides (SupplementaryFig. S4C). Of note, FT3 and FT6 cell transductants alsoexpressed an E-selectin–reactive GSL that comigrated withGD1b, which was absent in FT7 cells (Supplementary Fig. S4C).When glycoproteins were proteolytically removed as evi-denced by complete loss of CD44 (Supplementary Fig. S4D),remaining glycolipids bearing sLeX-E-selectin ligands weredetected and accounted for roughly half of total E-selectinbinding activity (Supplementary Fig. S4D). In fact, protease-digested PC-R1 FT cells rolled in equal number and velocity onE-selectin (Supplementary Fig. S4E and S4F). In all, these dataaffirm: (i) our intent to use PC-R1 FT cells as a model foradhesion and TEM analysis under blood flow conditions, (ii)highlight the key roles of FT3, 6, and 7 in regulating E-selectinligand expression, and (iii) provide direct evidence of theimportance of GSLs in E-selectin–dependent adhesion.
E-selectin ligandþ prostate cancer cells breachE-selectinþ HBMEC in a b1, aVb3, and Rac/Rap1-dependent manner
Using E-selectin ligandþ PC-R1 FT7 cells, we first analyzedwhether they could traverse HUVEC or HBMEC-60 cell mono-layers stimulated with IL-1b, which induces E-selectin expres-
sion (Supplementary Fig. S4G). We found that TEM of PC-R1FT7 cell pretreated with neutralizing anti-b1 or anti-aVb3mAbs was blocked by 40% to 80% (Fig. 3C and D). Moreover,RNA silencing of b1 protein (Fig. 3E) reduced TEM of PC-R1FT7 cells by 50% (Fig. 3F). Inhibitors of Rac1 and Rap1 blockedTEM through HBMEC-60 cells by 30% and 80%, respectively, incontrast to antagonists of CXCR4 and CCR2 or pertussis toxinpretreatment, which did not have any inhibitory effects (Fig.3G). In agreement with lack of inhibitory effect, expression ofCXCR4 and related chemokine receptor, CXCR7, along withCCR2 was undetected or minimal in PC-R1 cells and in anumber of prostate cancer cell lines (Supplementary Fig. S5A),and mRNA levels of CXCR4, CXCR7, and CCR2 along withanother known prostate cancer chemokine receptor, CCR4,were variable and/or downregulated in primary andmetastaticprostate cancer tissue when compared with normal prostatetissue (Supplementary Fig. S5B–S5E). Thus, PC-R1 FT cell TEMunder static conditions was dependent on b1 and aVb3 andRac1 and Rap1 GTPases, although not on G-protein signaling,through CXCR4 or CCR2.
We then examined the role of b1 and aVb3 integrins inadhesion and TEM of PC-R1 FT cells under flow conditions inthe parallel-plate flow chamber. In studies using IL-1b–stim-ulated HBMEC-60 monolayers, flow data revealed that PC-R1FT3, 6, and 7 tethered and rolled on E-selectinþHBMEC-60 cellmonolayers, whereas E-selectin ligand� PC-R1 empty cells didnot (Fig. 4A andB). This rolling activity was completely blocked
Figure 4. E-selectin–E-selectin ligands, b1 and aVb3 integrins, and Rac1/Rap1 GTPases cooperate in PC-R1 cell breaching of BMEC under flow conditions.A andB,meanPC-R1FTcell rollingevents (A) and rolling velocity (B) on IL-1b–stimulatedHBMEC-60 cells at 1Dyne/cm2�anti-E-selectinmAb from4 fields at�100 magnification (n¼ 9� SEM; ND, not detected; ���, P < 0.001 vs. isotype, one-way ANOVA with Dunnett posttest). C–F,% PC-R1 FT andMDA PCa 2bcell breaching of HBMEC-60 cell monolayers � anti-E-selectin mAb, anti-integrin mAb, Rap1 inh. or Rac1 inh. normalized to isotype or untreated control(n ¼ 4 � SEM; ��, P < 0.01; ���, P < 0.001, t test).
Barthel et al.
Cancer Res; 73(2) January 15, 2013 Cancer ResearchOF6
with anti-E-selectin mAb, indicating that E-selectin ligand–E-selectin interactions are critical for initiating prostate cancercell adhesion under shear flow (Fig. 4A). Following the rollingactivity, we then observed a transition to firm adhesion, thencellular piercing of tight EC junctions and incorporation intothe endothelial plane, which peaked after a 4-hour period(Supplementary Fig. S6A). This activity is defined here as a"Breaching" process. Phase contrast and confocal fluores-cence imaging in 3 dimensions confirmed breached prostatecancer cells (green) adjacent to endothelial cells [red, indi-cating paracellular prostate cancer-endothelial breaching(Supplementary Fig. S6B)]. Interestingly, PC-R1 FT7–ECbreaching occurred in RPMI medium without serum orexogenous growth factors/chemokines. Notably, all E-selec-tin ligandþ PC-R1 FT and control MDA prostate cancer 2bcells tethered, rolled, firmly adhered, and breached E-selectinþ HBMEC-60 monolayers (Fig. 4C and Supplemen-tary Movie S1). Conversely, E-selectin ligand� PC-R1 emptyor E-selectinþHBMEC-60 cells neutralized with anti-E-selec-tin mAb showed no breaching, which underscored E-selectinligand dependency (Fig. 4C and Supplementary Movie S2).When anti-b1 and aVb3 mAbs were added to control MDAprostate cancer 2b and PCR1 FT7 cells in these breachingexperiments, there was a significant reduction in breachingactivity (P < 0.01; Fig. 4D and E). Furthermore, incubatingRap1 or Rac1 inhibitors with PC-R1 FT7 cells also signifi-cantly blocked breaching activity (P < 0.01; Fig. 4F). Theseresults suggested that prostate cancer cells sequentiallyrolled on HBMEC, transitioned to firm adhesion, and brea-
ched EC monolayers in an E-selectin ligand-, b1 and aVb3integrin-, and Rac1 and Rap1-dependent manner.
E-selectin ligandþ PC-R1 FT cells traffic to bone marrowvia dependency on E-selectin, b1 integrin, and Rac1
To analyze the role of b1 and aVb3 integrins and Rap1 andRac1 GTPases in prostate cancer homing to bone, we generatedPC-R1 FT cells stably expressing luciferase and injected theminto Rag2/Janus kinase(Jak)-3-null mice deficient in T, B, andnatural killer (NK) cells (5). Following prostate cancer cellinjection, we assayed for presence of luciferaseþ prostate cancercells in normalizing control tissues (blood and spleen) and inbone by quantitative PCR (5). After 4 hours postinjection, wefound that allmice injectedwithE-selectin ligandþPC-R1FT3, 6,or 7 cells contained luciferaseþ cells in bone,whereasonly25%ofmice contained E-selectin ligand� PC-R1 empty cells (Fig. 5A).By pretreating mice with neutralizing anti-E-selectin mAb, PC-R1 FT cell homing to bone was reduced by 40% to 88% (Fig. 5A).Moreover, after 24 hours postinjection, 95% of femurs containedPC-R1 FT7, 67%PC-R1FT6, 30%PC-R1 FT3, andonly 13%PC-R1empty cells; and bone retention was ablated by pretreatingmicewith anti-E-selectin mAb (Fig. 5B). We also found that pretreat-ing PC-R1 cells with anti-b1 mAb blocked PC-R1 FT7 cellretention in bone by 88%, whereas anti-aVb3 mAb blockedretention by 20% (Fig. 5C). Pretreatment with a Rac1 inhibitorblocked bone retention by 78%, whereas Rap1 inhibitor blockedby 25% (Fig. 5D). These data showed that E-selectin ligandþ PC-R1 FT cells homed to bone, which was dependent on E-selectinligands, b1 integrin and Rac1 with minor contributions from
Figure 5. E-selectin ligandþ PC-R1FT cell dissemination to bone isdependent on E-selectin ligands, b1and Rac1. A–D, % incidence ofLuciferaseþ PC-R1 cells in bone at4 or 24 hours postintracardiacinjection as determined by PCRanalysis of luciferase; �, P < 0.05;��, P < 0.01; ���, P < 0.001,contingency table with 2-tailedFisher test.
Molecular Determinants for PCa Cell Breaching of BMEC
www.aacrjournals.org Cancer Res; 73(2) January 15, 2013 OF7
aVb3 integrin and Rap1. Also, FT7 was most influential of alla1,3 FTs in bone homing activity of prostate cancer cells.
a1,3 FTs promote prostate cancer progressionWe and others have shown a key role for a1,3 FT expression
in prostate cancer growth and bone homing activity in vivo(2, 5). To explore the role of a1,3 FTs in spontaneous prostatecancer formation and progression within the prostate gland, wegenerated TRAMPmice, which develop prostate adenocarcino-ma, that were deficient in a1,3 FTs, FT4 and FT7, by targetedgene disruption. In that mice do not express FT3 and FT6 (35),and FT4 does not contribute to sLeX or E-selectin ligandformation in prostate cancer cells, analysis of these mutantmice in terms of sLeX or E-selectin ligand formation was relianton FT7.We found that TRAMPmice deficient ina1,3 FT activityexhibiteda lower incidence ofprostate cancer formation (Fig. 6Aand B) and lower rate of tumor progression as evidenced bysignificantly smaller prostate weights (Fig. 6C and D). Unfortu-nately, observationsonmetastatic activity inFT4and7-deficientTRAMP mice were not possible due to lack of primary tumorformation. As such, data indicated a key role for a1,3 FT inprimary prostate cancer development in the prostate gland.
DiscussionDissemination, entry, and growth of cancer cells in distal
tissues causes 90% of cancer-related deaths and remains a
major unsolved problem in prostate cancer mortality (36).Here, we identified functional regulators of prostate cancerextravasation, including tethering, firm adhesion, and move-ment into bone marrow endothelium under physiologic bloodflow conditions. We described key mechanistic roles for pros-tate cancer cell a1,3 FT activity and related E-selectin ligandexpression, for b1 and aVb3 integrins, and for Rac1/Rap1GTPases in prostate cancer cell extravasation (Fig. 7A). Wealso identified a new role fora1,3 FT activity in prostate cancerformation (Fig. 7B). Interestingly, contrary to evidence on thehallmark role of chemokine receptors in integrin activation, wefound that integrin-mediated prostate cancer cell adhesionand migration across BMEC monolayers did not require che-mokine(s) as b1 and aVb3 and GTPases were constitutivelyactive (23–25, 37–39). Our data also confirmed earlier reportswhereby a1,3 FT3, 6, and 7 were critical for forming sLeX andcorresponding E-selectin ligands and bone-homing activity ofmetastatic prostate cancer cells (5). Considering our observa-tion that a1,3 FTs, FT4 and FT7, promoted prostate cancerformation in TRAMP mice and FT3 promotion of humanprostate cancer growth (40), the collective role of a1,3 FTs,FT3, 6, and 7,may be to aid the exit of prostate cancer cells fromcirculation through E-selectin ligands and also to generatea1,3fucose residues that may play a role in intrinsic transformingactivity and/or tumor cell–host/stroma interactions promot-ing tumorigenicity. Analysis of prostate cancer bone metasta-sis beyond a 24-hour assessment still needs to be conducted to
Figure 6. a1,3 FT4 and 7 areprotumorigenic in TRAMP mice.TRAMP mice wt (þ/þ),heterozygous (þ/�), andhomozygous null (�/�) for FT4 andFT7 expression were generatedand evaluated for primary tumorincidence and size (prostateweight) at 18 and 23 weeks. A andB, �, P ¼ 0.0361; ��, P ¼ 0.0051,contingency table with 2-tailedFisher test. C and D, �, P < 0.05;��, P < 0.01, one-way ANOVA withDunnett posttest.
Barthel et al.
Cancer Res; 73(2) January 15, 2013 Cancer ResearchOF8
further address the role of a1,3 FTs in prostate cancergrowth in bone. This is the first report describing pleotropicroles of a1,3 fucosylation in malignant progression and meta-stasis of prostate cancer.In all, our data parallel the molecular circuitry required for
osteotropic activity of HSCs and MSCs, wherein E-selectinligandþ cells display a greater osteotropism than E-selectinligand� cells (34, 41, 42). In fact, considering recent data thatbone-homing MSCs exhibit a Step-2 bypass pathway wherebychemokine-mediated integrin activation is not required foroptimal integrin avidity and TEM activity (34, 41), our dataindicated a similar chemokine receptor-independent mechan-osignaling circuit for robust prostate cancer cell firm adhesionand vascular breaching. Moreover, prostate cancer cell linescapable of extravasating BMEC, and metastatic prostate can-cer tissues, in general, showed variably low or downregulatedlevels of chemokine receptors thought involved in prostate
cancer metastasis, CXCR4, CXCR7, CCR2 and CCR4. Extrava-sating prostate cancer cells relied on E-selectin ligands, b1 andaVb3 integrins and Rac1/Rap1 GTPase activity without che-mokine-mediated integrin activation. While we did not findevidence of E-selectin ligand–dependent activation of b1 and/oraVb3 integrins or Step-2 bypass pathway, our findings showthat seeding and exiting of circulating prostate cancer cells intobone require the cooperation of a1,3 fucosylation-dependentE-selectin ligands and b1 and aVb3 integrins for efficientrolling, firm adhesion, and TEM activity. Future studies needto address whether these adhesion molecules coassociate onprostate cancer cells to regulate adhesion and movement asobserved for a4b1 and CD44 on T cells and MSCs and alsowhether such molecules regulate extravasation of fresh-isolat-ed, native-circulating prostate cancer cells from patients tohelp rationalize pharmacologic targeting and treatment strat-egies (34, 43). Moreover, as these studies were conducted in
Figure 7. Model of prostatecancer progression andextravasation to bone. A, model ofprostate cancer cell extravasationinto bone. Step 1, a1,3 FTs, FT3, 6,and 7 catalyze the synthesis of sLeX
on membrane glycoproteins andneolactosphingolipids to promotecorresponding E-selectin ligandactivity on PCa cells. E-selectinligandþ PCa cells roll on BMECE-selectin. Constitutively active b1due partly to Rap1-GTPase activityand active aVb3 integrins mediate(Step 2) firm adhesion to putativeintegrin ligands FN, VN, andVCAM-1on BMEC. Step 3, prostate cancercells traverse BMEC junctions drivenby Rap1 and Rac1 GTPase activityalong with b1 and aVb3 binding toputative surface and basementmembrane integrin ligands, FN, VN,VCAM-1, LN, and COL. B,development of prostate cancerlesions in the prostate secretory andneuroendocrine cell layer ispromoted by FT4 and FT7expression.
Molecular Determinants for PCa Cell Breaching of BMEC
www.aacrjournals.org Cancer Res; 73(2) January 15, 2013 OF9
immunocompromised mice, the role of immune cells in pros-tate cancer trafficking requires further investigation.
Interestingly, lymph node-metastatic LNCaP and LNCaPC4-2b cells were least migratory and adhesive due to less activeand lower b1 and aVb3 levels and lacked sialo-LacNAc, aprecursor for sLeX and E-selectin ligand formation. Prior data,in fact, show that lymph node prostate cancer had low levels ofFT3, 6, and 7 and corresponding sLeX structures comparedwith prostate cancer in other tissues, notably bone and liver (5),indicating that b1 and aVb3 integrins and E-selectin ligandsmay be less important in lymph node metastasis. Our molec-ular paradigm outlined here may be more applicable for bonemetastasis. Surprisingly, while dermal postcapillary venulesexpress E-selectin, prostate cancer metastasis in skin is rare,suggesting that E-selectin ligandþ prostate cancer cells maynot survive or proliferate within skin (44). Thus, while E-selectin ligandþ circulating prostate cancer cells efficientlybind and breach BMEC, growth-related events needed forcolonization in bone may be a more critical and less efficientstep of bone metastasis that could be investigated over timeperiods longer than assessed here.
Collectively, we identified multiple adhesion molecules forpreventing extravasation of circulating prostate cancer cellsinto distant tissues. Antagonizing E-selectin ligands, a1,3 FT,b1 and aVb3 integrins and/or Rap1/Rac1 GTPases couldcurtail prostate cancer cell homing and extravasation intobone (45). To avert ancillary alteration of homeostatic traf-ficking activity of HSCs or MSCs, strategic pharmacologicefforts could focus on antagonizing FT3 and FT6 in epitheli-al-derived tumors or selective blockade of hyperactive b1variants on prostate cancer cells. Because FT4 and FT7 arepredominanta1,3 FTs in leukocytes, targeting FT3 and 6wouldspare altered leukocytic trafficking and tissue distributionpatterns. Such a regimen could be further fine tuned byinclusion of antagonists against activated a2, a5, a6, andaVb3 structures, which would be selectively expressed oncirculating prostate cancer cells not on circulating leukocytes.
Sampling and assaying for a1,3FT, E-selectin ligand, and b1expression on localized prostate cancer or circulating prostatecancer cells might help prognosticate metastatic potential andguide treatment intervention.
Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.
Authors' ContributionsConception and design: S.R. Barthel, M.T. Moser, B.A. Foster, C.J. DimitroffDevelopment of methodology: S.R. Barthel, E.M. Yazawa, L. Nimrichter, M.T.Moser, K. Pantel, B.A. Foster, C.J. DimitroffAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.):D.Hays,M.J. Opperman, K.C.Walley,M.M. Burdick, B.M. Gillard, M.T. Moser, K. Pantel, B.A. Foster, K.J. Pienta, C.J. DimitroffAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): S.R. Barthel, D. Hays, E.M. Yazawa,M.J. Opperman, K.C. Walley, M.T. Moser, K. Pantel, K.J. Pienta, C.J. DimitroffWriting, review, and/or revision of themanuscript: S.R. Barthel, D. Hays, E.M. Yazawa, K.C. Walley, L. Nimrichter, M.T. Moser, K. Pantel, B.A. Foster, K.J.Pienta, C.J. DimitroffAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): S.R. Barthel, E.M. Yazawa, K.C. Walley,C.J. DimitroffStudy supervision: S.R. Barthel, B.A. Foster, C.J. Dimitroff
AcknowledgmentsThe authors thank Dr. Ronald Schnaar for supporting efforts in prostate
cancer glycolipid analyses (Johns Hopkins University, Baltimore, MD).
Grant SupportThis work was supported by NIH/NCI grant (R01 CA118124; to C. Dimitroff),
NIH/NCCAM grant (R01 AT004268; to C. Dimitroff), American Cancer SocietyPostdoctoral Fellowship (10-227 to S. Barthel), NIH Kirschstein-NRSA Postdoc-toral Fellowship (F32 CA144219-01A1; to S. Barthel), Dermatology FoundationResearch Grant (A050422 to S. Barthel), Fulbright/CAPES and CNPq to L.Nimrichter, NIH/NCI grant (RO1 CA095367; to B. Foster), and Roswell ParkCancer Institute and NIH/NCI grant (P30 CA016056).
The costs of publication of this article were defrayed in part by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received August 16, 2012; revised October 4, 2012; accepted October 24, 2012;published OnlineFirst November 13, 2012.
References1. American Cancer Society CS. Available from: http://www.cancer.org.2. Dimitroff CJ, Lechpammer M, Long-Woodward D, Kutok JL. Rolling of
human bone-metastatic prostate tumor cells on human bone marrowendothelium under shear flow is mediated by E-selectin. Cancer Res2004;64:5261–9.
3. Dimitroff CJ, Descheny L, Trujillo N, Kim R, Nguyen V, Huang W, et al.Identification of leukocyte E-selectin ligands, P-selectin glycoproteinligand-1 and E-selectin ligand-1, on humanmetastatic prostate tumorcells. Cancer Res 2005;65:5750–60.
4. Barthel SR, Gavino JD,Wiese GK, Jaynes JM, Siddiqui J, Dimitroff CJ.Analysis of glycosyltransferase expression in metastatic prostatecancer cells capable of rolling activity on microvascular endothelial(E)-selectin. Glycobiology 2008;18:806–17.
5. Barthel SR, Wiese GK, Cho J, Opperman MJ, Hays DL, Siddiqui J,et al. Alpha 1,3 fucosyltransferases are master regulators of pros-tate cancer cell trafficking. Proc Natl Acad Sci U S A 2009;106:19491–6.
6. SchweitzerKM,Drager AM, vander ValkP, ThijsenSF,ZevenbergenA,Theijsmeijer AP, et al. Constitutive expression of E-selectin and vas-cular cell adhesion molecule-1 on endothelial cells of hematopoietictissues. Am J Pathol 1996;148:165–75.
7. Martensson S, Bigler SA, Brown M, Lange PH, Brawer MK, HakomoriS. Sialyl-Lewis(x) and related carbohydrate antigens in the prostate.Hum Pathol 1995;26:735–9.
8. Jorgensen T, Berner A, Kaalhus O, Tveter KJ, Danielsen HE, BryneM. Up-regulation of the oligosaccharide sialyl LewisX: a new prog-nostic parameter in metastatic prostate cancer. Cancer Res 1995;55:1817–9.
9. Idikio HA. Sialyl-Lewis-X, Gleason grade and stage in non-metastatichuman prostate cancer. Glycoconj J 1997;14:875–7.
10. Mazo IB, Gutierrez-Ramos JC, Frenette PS, Hynes RO, Wagner DD,von Andrian UH. Hematopoietic progenitor cell rolling in bone marrowmicrovessels: parallel contributions by endothelial selectins and vas-cular cell adhesion molecule 1. J Exp Med 1998;188:465–74.
11. Frenette PS, Subbarao S, Mazo IB, von Andrian UH, Wagner DD.Endothelial selectins and vascular cell adhesion molecule-1 promotehematopoietic progenitor homing to bone marrow. Proc Natl Acad SciU S A 1998;95:14423–8.
12. Katayama Y, Hidalgo A, Furie BC, Vestweber D, Furie B, Frenette PS.PSGL-1 participates in E-selectin-mediated progenitor homing tobone marrow: evidence for cooperation between E-selectin ligandsand alpha4 integrin. Blood 2003;102:2060–7.
Barthel et al.
Cancer Res; 73(2) January 15, 2013 Cancer ResearchOF10
13. Hidalgo A, Weiss LA, Frenette PS. Functional selectin ligands medi-ating human CD34(þ) cell interactions with bone marrow endotheliumare enhanced postnatally. J Clin Invest 2002;110:559–69.
14. Katayama Y, Hidalgo A, Peired A, Frenette PS. Integrin alpha4beta7and its counterreceptor MAdCAM-1 contribute to hematopoietic pro-genitor recruitment into bone marrow following transplantation. Blood2004;104:2020–6.
15. QianH,TryggvasonK, JacobsenSE,EkblomM.Contributionof alpha6integrins to hematopoietic stem and progenitor cell homing to bonemarrow and collaboration with alpha4 integrins. Blood 2006;107:3503–10.
16. Peled A, Kollet O, Ponomaryov T, Petit I, Franitza S, Grabovsky V, et al.The chemokineSDF-1 activates the integrins LFA-1, VLA-4, andVLA-5on immature human CD34(þ) cells: role in transendothelial/stromalmigration and engraftment of NOD/SCID mice. Blood 2000;95:3289–96.
17. Avigdor A, Goichberg P, Shivtiel S, Dar A, Peled A, Samira S, et al.CD44 and hyaluronic acid cooperate with SDF-1 in the trafficking ofhuman CD34þ stem/progenitor cells to bone marrow. Blood2004;103:2981–9.
18. Papayannopoulou T,CraddockC,NakamotoB, PriestleyGV,Wolf NS.The VLA4/VCAM-1 adhesion pathway defines contrasting mechan-isms of lodgement of transplanted murine hemopoietic progenitorsbetween bone marrow and spleen. Proc Natl Acad Sci U S A 1995;92:9647–51.
24. LobergRD,DayLL,Harwood J,YingC,St JohnLN,GilesR, et al. CCL2is a potent regulator of prostate cancer cell migration and proliferation.Neoplasia 2006;8:578–86.
25. Sun YX, Fang M, Wang J, Cooper CR, Pienta KJ, Taichman RS.Expression and activation of alpha v beta 3 integrins by SDF-1/CXC12increases the aggressiveness of prostate cancer cells. Prostate2007;67:61–73.
26. Kukreja P, Abdel-Mageed AB, Mondal D, Liu K, Agrawal KC. Up-regulation of CXCR4 expression in PC-3 cells by stromal-derivedfactor-1alpha (CXCL12) increases endothelial adhesion and transen-dothelial migration: role of MEK/ERK signaling pathway-dependentNF-kappaB activation. Cancer Res 2005;65:9891–8.
27. Romanov VI, Goligorsky MS. RGD-recognizing integrins mediateinteractions of human prostate carcinoma cells with endothelial cellsin vitro. Prostate 1999;39:108–18.
28. LangSH,ClarkeNW,GeorgeNJ, TestaNG.Primary prostatic epithelialcell binding to humanbonemarrow stromaand the role of alpha2beta1integrin. Clin Exp Metastasis 1997;15:218–27.
29. Scott LJ, Clarke NW, George NJ, Shanks JH, Testa NG, Lang SH.Interactions of human prostatic epithelial cells with bone marrowendothelium: binding and invasion. Br J Cancer 2001;84:1417–23.
30. Wang X, Ferreira AM, Shao Q, Laird DW, Sandig M. Beta3 integrinsfacilitate matrix interactions during transendothelial migration of PC3prostate tumor cells. Prostate 2005;63:65–80.
31. Wiese G, Barthel SR, Dimitroff CJ. Analysis of physiologic E-selectin-mediated leukocyte rolling on microvascular endothelium. J Vis Exp2009;24:1009.
32. Barthel SR, Johansson MW, McNamee DM, Mosher DF. Roles ofintegrin activation in eosinophil function and the eosinophilic inflam-mation of asthma. J Leukoc Biol 2008;83:1–12.
33. Barthel SR, Annis DS, Mosher DF, Johansson MW. Differentialengagement of modules 1 and 4 of vascular cell adhesion mole-cule-1 (CD106) by integrins alpha4beta1 (CD49d/29) and alphaMbeta2(CD11b/18) of eosinophils. J Biol Chem 2006;281:32175–87.
34. Thankamony SP, Sackstein R. Enforced hematopoietic cell E- and L-selectin ligand (HCELL) expression primes transendothelial migrationof human mesenchymal stem cells. Proc Natl Acad Sci U S A2011;108:2258–63.
35. Costache M, Apoil PA, Cailleau A, Elmgren A, Larson G, Henry S, et al.Evolution of fucosyltransferase genes in vertebrates. J Biol Chem1997;272:29721–8.
36. Weinberg RA. The Biology of Cancer. New York: Garland Science;2007.
37. TaichmanRS,CooperC,Keller ET,PientaKJ, TaichmanNS,McCauleyLK. Use of the stromal cell-derived factor-1/CXCR4 pathway in pros-tate cancer metastasis to bone. Cancer Res 2002;62:1832–7.
38. Sun YX, Wang J, Shelburne CE, Lopatin DE, Chinnaiyan AM, RubinMA, et al. Expression of CXCR4 and CXCL12 (SDF-1) in humanprostate cancers (PCa) in vivo. J Cell Biochem 2003;89:462–73.
39. Sun YX, Schneider A, Jung Y, Wang J, Dai J, Wang J, et al. Skeletallocalization and neutralization of the SDF-1(CXCL12)/CXCR4 axisblocks prostate cancer metastasis and growth in osseous sites invivo. J Bone Miner Res 2005;20:318–29.
40. Inaba Y, Ohyama C, Kato T, Satoh M, Saito H, Hagisawa S, et al. Genetransfer of alpha1,3-fucosyltransferase increases tumor growth of thePC-3 human prostate cancer cell line through enhanced adhesion toprostatic stromal cells. Int J Cancer 2003;107:949–57.
41. Sackstein R, Merzaban JS, Cain DW, Dagia NM, Spencer JA, LinCP, et al. Ex vivo glycan engineering of CD44 programs humanmultipotent mesenchymal stromal cell trafficking to bone. Nat Med2008;14:181–7.
42. Merzaban JS, Burdick MM, Gadhoum SZ, Dagia NM, Chu JT, Fuhl-brigge RC, et al. Analysis of glycoprotein E-selectin ligands on humanand mouse marrow cells enriched for hematopoietic stem/progenitorcells. Blood 2011;118:1774–83.
43. Nandi A, Estess P, SiegelmanM. Bimolecular complex between rollingand firm adhesion receptors required for cell arrest; CD44 associationwith VLA-4 in T cell extravasation. Immunity 2004;20:455–65.
44. Weninger W, Ulfman LH, Cheng G, Souchkova N, Quackenbush EJ,Lowe JB, et al. Specialized contributions by alpha(1,3)-fucosyltrans-ferase-IV and FucT-VII during leukocyte rolling in dermalmicrovessels.Immunity 2000;12:665–76.
Published OnlineFirst November 13, 2012.Cancer Res Steven R. Barthel, Danielle L. Hays, Erika M. Yazawa, et al. Bone ExtravasationDefinition of Molecular Determinants of Prostate Cancer Cell
Updated version
10.1158/0008-5472.CAN-12-3264doi:
Access the most recent version of this article 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/early/2013/01/10/0008-5472.CAN-12-3264To request permission to re-use all or part of this article, use this link
