Cancer Genes and Networks

Therapeutic Targeting of CD146/MCAM ReducesBone Metastasis in Prostate CancerEugenio Zoni1, Letizia Astrologo1, Charlotte K.Y. Ng2,3, Salvatore Piscuoglio3,Janine Melsen4, Jo€el Grosjean1, Irena Klima1, Lanpeng Chen5,Ewa B. Snaar-Jagalska5, Kenneth Flanagan6, Gabri van der Pluijm4, Peter Kloen7,Marco G. Cecchini†1, Marianna Kruithof-de Julio1, and George N. Thalmann8

Abstract

Prostate Cancer is the most common cancer and the secondleading cause of cancer-related death in males. When prostatecancer acquires castration resistance, incurable metastases,primarily in the bone, occur. The aim of this study is to testthe applicability of targeting melanoma cell adhesion mole-cule (MCAM; CD146) with a mAb for the treatment of lyticprostate cancer bone metastasis. We evaluated the effect oftargeting MCAM using in vivo preclinical bone metastasismodels and an in vitro bone niche coculture system. Weutilized FACS, cell proliferation assays, and gene expressionprofiling to study the phenotype and function of MCAMknockdown in vitro and in vivo. To demonstrate the impact ofMCAM targeting and therapeutic applicability, we employedan anti-MCAM mAb in vivo. MCAM is elevated in prostatecancer metastases resistant to androgen ablation. Treatmentwith DHT showed MCAM upregulation upon castration. Weinvestigated the function ofMCAM in a direct coculturemodel

of human prostate cancer cells with human osteoblasts andfound that there is a reduced influence of human osteoblastson human prostate cancer cells in which MCAM has beenknocked down. Furthermore, we observed a strongly reducedformation of osteolytic lesions upon bone inoculation ofMCAM-depleted human prostate cancer cells in animalmodelof prostate cancer bone metastasis. This phenotype is sup-ported by RNA sequencing (RNA-seq) analysis. Importantly,in vivo administration of an anti-MCAM human mAb reducedthe tumor growth and lytic lesions. These results highlight thefunctional role for MCAM in the development of lytic bonemetastasis and suggest that MCAM is a potential therapeutictarget in prostate cancer bone metastasis.

Implications: This study highlights the functional applicationof an anti-MCAM mAb to target prostate cancer bonemetastasis.

IntroductionProstate cancer is the most common cancer and the second

leading cause cancer-related deaths in men (1). Current treat-ments such as radiation therapy and androgen deprivation ther-apies (ADT) are effective when the cancer is still confined at the

primary site (2).However, after the tumor progresses and acquirescastration resistance, the development of incurable metastases isalmost inevitable (3).

Prostate cancer metastases occur at specific sites with one of themost common locations being the bone, where the occurrence ofboth lytic and blastic lesions has been documented (4). Previousstudies demonstrated that prostate cancer cells colonize the bonemicroenvironment within the so called "hematopoietic stem cell(HSC) niche" (3, 5). Prostate cancer metastasis–initiating cells(MIC) compete with the bone marrow cells for the occupancy ofthismicroenvironment and induce an ectopic epithelial tissue-of-origin niche, referred as a "developmental prostate niche" (6).

Melanoma cell adhesion molecule (MCAM/CD146) is a cell-surface glycoprotein composed of five immunoglobulin-likedomains, one transmembrane region, and a short cytoplasmictail, which interacts with the cytoskeleton (7). Growing evidencesupports the notion that high MCAM expression in a variety ofcarcinomas positively correlates with poor prognosis in prostatecancer (8), melanoma (9), pulmonary adenocarcinomas (10),epithelial ovarian cancer (11), and breast cancer (12). Overex-pression of MCAM has been shown to increase tumorigenicity ofhuman osteoblastic prostate cancer cells (LNCaP) in vivo (13). Inadition, MCAM was shown to induce epithelial-to-mesenchymaltransition (EMT) inbreast cancer, a process thought tobe involvedin the initiation of metastasis (14).

Upon transplantation, MCAM-expressing subendothelial cellsin human bone marrow stroma are capable of recapitulating

1Department for BioMedical Research, Urology Research Laboratory, Universityof Bern, Bern, Switzerland. 2Department of Biomedicine, University HospitalBasel, University of Basel, Basel, Switzerland. 3Institute of Pathology, UniversityHospital Basel, University of Basel, Basel, Switzerland. 4Department of Urology,Urology Research Laboratory Leiden University Medical Center, Leiden, theNetherlands. 5Institue of Biology, University of Leiden, Leiden, the Netherlands.6Prothena Biosciences, 331 Oyster Point Blvd, South San Francisco, California.7Department of Orthopedic Trauma Surgery, Academic Medical Center,Amsterdam, the Netherlands. 8Department of Urology, Inselspital, BernUniversity Hospital, University of Bern, Bern, Switzerland.

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

M. Kruithof-de Julio and G. N. Thalmann contributed equally to and are co-lastauthors of this article.

†Deceased.

Corresponding Authors: Marianna Kruithof-de Julio, Univeristy of Bern, Bern3008, Switzerland. Phone: 413-2632-0931; Fax: 413-1632-0551; E-mail:marianna.kruithofdejulio@dbmr.unibe.ch; and George N. Thalmann,george.thalmann@insel.ch

doi: 10.1158/1541-7786.MCR-18-1220

�2019 American Association for Cancer Research.

MolecularCancerResearch

www.aacrjournals.org 1049

on January 13, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 11, 2019; DOI: 10.1158/1541-7786.MCR-18-1220

hematopoiesis in heterotopic sites (15). This established thenotion that MCAM-positive subendothelial cells are relevant tohematopoiesis and play key roles in the maintenance of the HSCniche. Previously, we identified a gene signature for the effect thatprostate cancer cells exert on the bone stroma in a bonemetastasisxenograft mouse model (6). This included MCAM as a potentialmediator of the metastatic colonization and growth of prostatecancer cells in the bone in lytic and blastic bone metastases (6).The extracellular domain of MCAM can engage in heterophilicbinding with various ligands (e.g., Laminin-411). It has also beenproposed that homophilic binding with other MCAM mole-cules (16) might be occurring, suggesting that targeting MCAMon either tumor cells or stromal cells might have a functionaleffect on prostate cancer cell behavior.

In this study, we show that administration of an anti-MCAMmAb reduces intraosseous growth and lytic lesions in preclinicalmodels of prostate cancer bone metastasis. We provide evidencethat MCAM is strongly increased in androgen ablation–resistantmetastases derived from prostate cancer and show that MCAMknockdown reduces prostate cancer cell proliferation and osteo-blast-mediated induction of ALDH activity. Taken together,these findings suggest that MCAM supports the metastatic lyticphenotype in human prostate cancer cells and represents a pos-sible therapeutic target in patients with prostate cancer bonemetastasis.

Materials and MethodsCell lines and culture conditions

All the human cell lines employed in this study have beenauthenticated using highly polymorphic short tandem repeat(STR) loci. In addition, all the human cell lines have beenpreviously validated in vitro and in vivo (17–19). PC-3M-Pro4were cultured in DMEM (GibcoBRL) containing 4.5 g glucose/Lsupplemented with 10% FCII (Thermo Fisher Scientific),1% penicillin–streptomycin (PS, Life Technologies). PC-3M-Pro4Luc2 and PC-3M-Pro4Luc2dTomato cells were cultured inthe same medium supplemented with 0.8 mg/mL Neomycin(SantaCruzBiotechnology) orNeomycin and1mg/mLBlasticidin(Sigma-Aldrich), respectively. C4-2B cells were cultured inT-medium DMEM (Sigma-Aldrich) supplemented with 20%F-12K nutrient mixture Kaighn's modification (GibcoBRL),10% FCS, 0.125 mg/mL biotin, 1% insulin-transferrin-selenium(ITS), 6.825ng/mLT3, 12.5mg/mLadenine, 1%PS. ThedTomatoclones were supplemented with 1 mg/mL Blasticidin. Osteoblastswere derived and differentiated as we described previously (17).Culture wasmaintained inDMEM supplemented with 1%PS and1% ITS. All cells were maintained at 37�C and 5% CO2.

Knockdown of MCAM with shRNA transfectionShort hairpin RNA (shRNA) constructs were obtained from

Sigma's MISSION library [MCAM clone# TCRN0000151337(shRNA#1), TCRN0000155692 (shRNA#2), TCRN0000154854(shRNA#3)]. As a negative control, scrambled shRNA (SHC002,pLKO.1, shRNA-NT) with a lack of homology with any mamma-lian mRNA sequence was used.

RNA isolation and RT-qPCRTotal RNA was extracted using Trizol (Invitrogen) and cDNA

synthesized by reverse transcription according to manufacturer'sinstructions (Promega). Real-time qPCR was performed with

QuantStudio3 system (Thermo Fisher Scientific). ACTIN, HPRT,and GAPDH housekeeping genes were included for normaliza-tion. Primer sequences are reported in Supplementary Table S1.Data are displayed as 2�DCt when relative expression is indicatedon the y-axis.

Western blot analysisAnti-vimentin (Ab8979, Abcam) was diluted 1:1,000; anti–e-

cadherin was diluted 1:1,000 (AF648, R&D Systems); anti-criptowas diluted 1:1,000 (clone no. PBL6900; ref. 17). Detectionwas performed with 1:10,000 secondary horseradish peroxidase(hrp) antibody (NA931VS, NA934VS, Sigma-Aldrich). Actin wasdetected with 1:20,000 hrp antibody (A3854, Sigma-Aldrich).

Flow cytometry, ALDEFLUOR, and viable cell sortingFunctionalMCAMprotein expressionwas determined by FACS

withmouse IgG1anti-humanMCAM-Alexa647 cloneP1H12 (BDBiosciences). Nonspecific binding was excluded by staining withan isotype control antibody (mouse IgG1, BDBiosciences). ALDHactivity of the tumor cellswasmeasuredby theALDEFLUORAssayKit (Stemcell Technologies) according to the manufacturer'sinstructions (17). Gating is obtained by acquisition of a controltube for each sample. Therefore, the percentage of ALDHhigh cellsin the control gate correspond always to 0.01% of total. Subse-quently the same gate is applied to the sample to assess thepercentage of ALDHhigh cells in each experimental condition.After sorting, samples were controlled to assess purity of thesorted cell populations.

Proliferation assayCells were seeded at a density of 1,500 cells per well and growth

monitored for 24, 48, 72, or 96 hours. For each time point,AU 490nm was measured 2 hours after incubation with 20 mLof 3-(4,5 dimethylthiazol- 2-yl)- 5 -(3 -carboxymethoxyphenyl)-2 -(4 -sulfophenyl)- 2 Htetrazolium (MTS, Promega) at 37�Caccording to manufacturer's protocol. Data were normalized forthe number of cells seeded. N ¼ 5 per condition, performed atleast in biological triplicates.

Animal experimentsCB17 SCID male mice, 5 to 6 weeks old were intraosseous

injected with 50,000 PC-3M-Pro4Luc2dTomato cells bearing thestable shRNA#1 to knockdownMCAM expression or nontargeted(NT) shRNA control sequence. Sham-operated mice were includ-ed as additional control (data not shown). Body weight measure,bioluminescent imaging (BLI; NightOwl, Berthold) and X-rayassessment (Faxitron Bioptics), were conducted to monitor thehealthy status of the animals, the growth of the tumor cells, andthe progression of the lesions respectively. Bone morphometrywas conducted as described by Bassett and colleagues (20). A doseof anti-MCAM mAb (10 mg/kg) was used (for both the rat anti-mouse and rat anti-humanmolecule) and animals injected intra-peritoneally. Samedosewas applied to the control IgGmolecules.Monoclonal rat IgG1 anti-human MCAM (clone 2107) and ratanti-mouse MCAM mAb (clone 15) were kindly provided byProthena Biosciences (21, 22). The anti-human MCAM antibodyhas previously been tested in two clinical trials (ClinicalTrials.govIdentifier: NCT02630901 and NCT02458677). Control IgG(rat Fc and human Fc) were purchased from BioXCell (BE0096and BE0094). For zebrafish experiment, Tg(fli1:GFP)i114 zebra-fish line (23) was handled according to local animal welfare

Zoni et al.

Mol Cancer Res; 17(5) May 2019 Molecular Cancer Research1050

on January 13, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 11, 2019; DOI: 10.1158/1541-7786.MCR-18-1220

regulations to standard protocols (http://www.ZFIN.org). Twodays post fertilization, dechorionized zebrafish embryos wereanesthetized and injected with PC-3M-Pro4Luc2dTomato cellsas described previously (17). Data are representative of at leasttwo independent and blind experiments with �30 embryos pergroup. Survival rate of control group lower than 80% was used asdiscard cut-off. Images were acquired with Leica SP8 confocal(Leica Microsystems).

RNA sequencingA biological triplicate for MCAM knockdown cells (shRNA#1)

and control samples (shRNA-NT) of the cell line employed for thein vivo study was generated and total RNA extracted using RNeasyMini Kit (Qiagen). Samples were measured with NextSeq500(Illumina). Image analysis, base calling, and quality check wasperformed with the Illumina data analysis pipeline RTA v2.4.11andBcl2fastq v2.17. Sequence readswere alignedusing STAR two-pass (24) to the human reference genome GRCh37 and genecounts quantified using the "GeneCounts" option. Per-genecounts-per-million (CPM) were computed and log2-transformedadding a pseudo-count of 1 to avoid transforming 0. Genes withlog2-CPM <1 in more than three samples were removed. Unsu-pervised clusteringwas performedusing the top500most variablegenes, Euclidean distance as the distance metric and the Wardclustering algorithm, using the ConsensusClusterPlus (25) Rpackage. Differential expression analysis between MCAM knock-down cells and control samples was performed using the edg-eR (26) R package. Normalization was performed using the"TMM" (weighted trimmed mean) method and differentialexpression was assessed using the quasi-likelihood F-test. Geneswith FDR <0.05 were considered differentially expressed. Genesdifferentially expressed by>2-foldwere reported.Gene Set Enrich-ment Analysis (GSEA) was performed using the Prerankedtool (27) for the, C2 (canonical pathways) and C5 (biologicalprocesses; ref. 28).Geneswere rankedbasedon the F-statistic fromthe differential expression analysis, multiplying F-statistic ofdownregulated genes by �1. Pathways with FDR <0.25 wereconsidered significant. As an alternative, pathway analysis wasalso performed for the set of differentially expressed genes using g:Profiler (29). Enrichment maps for GSEA and g:Profiler data weregenerated with Cytoscape (30). Sets of genes with P value cutoff0.05 were included and similarity coefficient of 0.5 was applied.The RNA sequencing (RNA-seq) data have been deposited at theNCBI Sequence Read Archive under the accession SRP151808.

Analysis of publicly available dataset for transcriptomic andgenomic evaluation and for gene expression–based survivalcalculation

mRNA data for MCAM expression were extracted with Shiny-GEO (31) from the GSE6919 (32, 33), GSE6752 (32) andGSE101607 (34) dataset and analyzed with R. Gene expres-sion–based survival analysis for MCAM was conducted withPROGgene (35) on GSE40272 (36).

Statistical analysisStatistical analysis was performed using GraphPad Prism 6.0

using t test for comparison between two groups or ANOVA forcomparison between more groups. Two-way ANOVA was usedto examine the influence of knockdown and control cells on thedependent variable (time) in the in vivo study. Data are pre-sented as mean � SEM or mean � SD. P values �0.05 are

considered to be statistical significant (�, P � 0.05; ��, P < 0.01;���, P < 0.001).

Study approvalAnimal experiment was approved by the local ethical commit-

tee ofCanton of Bern, Switzerland (permit number BE 55/16) andcarried out in accordance with Swiss Guidelines for the Care andUse of Laboratory Animals.

ResultsMCAM is highly expressed in prostate cancer and associatedmetastasis

MCAM expression correlates with poor prognosis in severaltypes of cancer, including prostate cancer (37). We investigatedthe levels of MCAM in normal (N ¼ 18), tumor (N ¼ 65), andtumor adjacent tissues (N ¼ 63) and prostate cancer metastasis(N ¼ 20) in the GSE6919 and GSE6752 datasets (32, 33), whichinclude transcriptional data for primary site and androgen abla-tion–resistant metastases. We found thatMCAMwas increased intumor (P ¼ 0.056) and tumor-adjacent tissue (P ¼ 0.043) com-pared with normal samples (Fig. 1A and B). MCAM levels weresimilar in tumor and tumor adjacent tissues (Fig. 1C). Moreover,MCAM was strongly increased in prostate cancer metastases(N ¼ 20) compared with primary tumors (N ¼ 10; P <0.001; Fig. 1D). Subsequent fractionation of MCAM expressionin metastases from various soft tissues (lymph node, adrenal,liver, and lung) also revealed a significant increase in MCAMexpression in these sites compared with the primary tumor site(P¼ 0.001; Fig. 1E). The relation betweenMCAM expression andprostate cancer recurrence in GSE40272 (36) suggested aninvolvement of high MCAM expression in disease relapse,although this did not reach significance (Supplementary Fig.S1A and S1B).

Previously, it was demonstrated that a subpopulation ofALDHhigh cells isolated from the aggressive PC-3M-Pro4 pros-tate cancer cell line displays high clonogenicity in vitro andpossesses the ability to generate bone metastases in preclinicalmouse models, compared with nontumorigenic nonmetastaticALDHlow (38). Therefore, we measured the expression ofMCAM in selected subpopulations of human prostate cancercells from the PC-3M-Pro4 cell line and tested whether it waspossible to identify a subset of MCAMhigh cells with aggressivefeatures. Using viable cell sorting, we identified four subsetsof cells: ALDHhighMCAMhigh; ALDHhighMCAMlow; ALDHlow

MCAMhigh; and ALDHlowMCAMlow (Fig. 1F; SupplementaryFig. S1C). Analysis ofMCAM expression by RT-qPCR confirmedthat we successfully isolated a subset of MCAMhigh cells by flowcytometry (Fig. 1G). However, when tested for colony formingcapacity, the four subpopulations of cells displayed similarbehavior (Fig. 1H and I).

Characterization of MCAM knockdown cell lines andextravasation ability

We studied the functional role of MCAM in androgen receptor(AR)-negative and lytic prostate cancer PC-3M-Pro4 and PC-3M-Pro4 luc2dTomato cells, and in AR-positive and blastic C4-2B andC4-2BdTomato cells (17).

To investigate the function of MCAM, we used lentiviraldelivery of three independent MCAM-targeting shRNAs(MCAM KD) and one control NT shRNA. RT-qPCR showed

Targeting MCAM Reduces Bone Metastasis in Prostate Cancer

www.aacrjournals.org Mol Cancer Res; 17(5) May 2019 1051

on January 13, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 11, 2019; DOI: 10.1158/1541-7786.MCR-18-1220

significant reduction of MCAM mRNA with each of the 3MCAM shRNAs (shRNA#1; shRNA#2; shRNA#3) comparedwith the nontargeted control (shRNA-NT; Supplementary Fig.

S2A). Measurement of functionally active protein by flowcytometry showed marked reduction of MCAM levels withshRNA#1 (approximately 30% reduction vs. NT control cells),

Figure 1.

High expression ofMCAM is associated with prostate cancer and prostate cancer metastasis. A–C,MCAM expression in primary normal, primary prostate cancer,and tumor adjacent tissue (GSE6919). Expression levels are presented as boxplots. D and E,MCAM expression in primary prostate cancer site compared withdistant metastasis in samples from prostate cancer patients (GSE6752). F, Schematic representation of viable cell sorting; a fraction of MCAMHigh andMCAMLow cells (�20% of parental gate) is isolated from the ALDHHigh and from the ALDHLow subpopulation, respectively.G, RT-qPCR on the sortedsubpopulations forMCAM expression. H and I, Clonogenic assay and quantification of the colony forming capacity of the selected subpopulations after sorting.N¼ 3 technical replicates. Data are presented as mean� SD (� , P < 0.05; �� , P < 0.01; ��� , P < 0.001).

Zoni et al.

Mol Cancer Res; 17(5) May 2019 Molecular Cancer Research1052

on January 13, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 11, 2019; DOI: 10.1158/1541-7786.MCR-18-1220

whereas similar protein levels were measured for shRNA#2 andshRNA#3 (Fig. 2A). In C4-2B cells, RT-qPCR displayed strongreduction of MCAM mRNA by each of the 3 shRNAs compared

with NT control (Supplementary Fig. S2B). Analysis of func-tionally active protein expression by flow cytometry showedhigher reduction with shRNA#1 and shRNA#3 versus NT

Figure 2.

MCAM knockdown in PC-3M-Pro4Luc2dTomato and C4-2BdTomato human prostate cancer cell lines. Functionally active protein expression is evaluated byflowcytometry and displayed in panel (A) for PC-3M-Pro4Luc2dTomato cells and (B) for C4-2BdTomato. Data are representative of at least three independentexperiments. C, Evaluation of effect of MCAM knockdown on cell proliferation over 96 hours in PC-3M-Pro4Luc2dTomato cells and control (sh-NT) cells withthree independent shRNAs. Data are representative of at least three independent experiments. �� , P < 0.01 and ��� , P < 0.001 with Bonferroni multiplecomparison test.D, Proliferation assay on C4-2BdTomato cells with MCAM knockdown by three independent shRNAs compared with control nontargeted(sh-NT) cells. Data are represented as mean� SD. ���, P < 0.001 with Bonferroni multiple comparison test. E, ALDEFLUOR assay on MCAM knockdown andnontargeted (NT) PC-3M-Pro4 cells to assess the percentage of ALDHhigh cells upon MCAM knockdown and data quantification (F). Data are representative ofthree independent experiments and represented as mean� SD. G, ALDEFLUOR assay on MCAM knockdown and nontargeted (NT) C4-2BdTomato cells toassess the percentage of ALDHhigh cells upon MCAM knockdown and data quantification (H). Data are representative of three independent experiments andrepresented as mean� SD.

Targeting MCAM Reduces Bone Metastasis in Prostate Cancer

www.aacrjournals.org Mol Cancer Res; 17(5) May 2019 1053

on January 13, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 11, 2019; DOI: 10.1158/1541-7786.MCR-18-1220

control, compared with shRNA#2 (Fig. 2B), in line with thetranscriptional data.

We found that MCAM knockdown caused a reduction in cellproliferation in vitro in PC-3M-Pro4 cells (P < 0.01 for shRNA#1and shRNA#3 at 72 hours and P < 0.001 for shRNA#1 at 96 hourswith Bonferroni multiple comparison test, compared with NTcontrol; Fig. 2C). Similarly, in C4-2B cells, MCAM knockdownreduced proliferation in vitro (P < 0.001 for shRNA#1, shRNA#2,and shRNA#3 at 72 hours and P < 0.001 for shRNA#3 at 96 hourswith Bonferroni multiple comparison test; Fig. 2D). On the basisof these results and on the knockdown validation at the proteinlevel, we selected the shRNA#1 to continue with subsequent invitro and in vivo characterization.

To assess the role ofMCAM in themodulation of ALDHactivityin PC-3M-Pro4 cells, we employed the ALDEFLUOR assay(Fig. 2E). MCAM knockdown cells displayed a lower percentageof ALDHhigh cells compared with NT control, although this didnot reach statistical significance (Fig. 2F). Similarly, no changewasobserved in C4-2B prostate cancer cells (Fig. 2G and H). Thissuggests that knockdown of MCAM does not cause depletion ofthe ALDHhigh population of human prostate cancer cells. Analysisof a panel of EMT markers (E-CAD, N-CAD, VIM, ZEB1, ZEB2,SNAIL1, SNAIL2 andTWIST) in the presence or absence ofMCAMknockdown suggests that suppressing MCAM expression pro-motes an epithelial transcriptional phenotype as shown by thesignificant increase in the ratio E-cadherin/vimentin in both celllines (P < 0.001 for PC-3M-Pro4 and P < 0.01 for C4-2B; Fig. 3Aand B) and by the increase in the ratio E-Cad/N-Cad in PC-3M-Pro4 (P < 0.001 Supplementary Fig. S2C–S2E). However, evalu-ation of protein expression by Western blot analysis revealed anonsignificant alteration in E-cadherin and vimentin expression(left, Fig. 3C–F) and in the ratio E-cadherin/vimentin(right, Fig. 3F). We used zebrafish to assess the impact of MCAMknockdown on the ability of prostate cancer cells to extravasate,which is an EMT feature, and to grow at distant sites (Supple-mentary Fig. S2F). Quantification of disseminated cells at 1 daypostinjection and 4 days postinjection revealed similar behaviorin knockdown and control (Supplementary Fig. S2G), supportingour analysis on E-cadherin and vimentin expression.

MCAM is required for osteoblast-mediated induction of ALDHactivity in prostate cancer cells and is increased upon castration

The osteoblastic microenvironment in bone functions as pre-metastatic niche by attracting bone-metastasizing prostate cancercells (5). We have previously shown that coculture of humanprostate cancer cells and mature human osteoblasts leads toincrease in the percentage of highly metastatic ALDHhigh prostatecancer cells (17).

To investigate the role of MCAM in the context of the osteo-blastic niche, we performed direct coculture of MCAM knock-down prostate cancer cells and human osteoblasts. After 48 hoursof coculture, the dTomato-labelled PC-3M-Pro4 prostate cancercells were separated from the osteoblasts by viable cell sorting(Fig. 3G). Immediately after sorting, we measured the ALDHactivity of the sorted PC-3M-Pro4MCAMknockdown (shRNA#1)and control prostate cancer cells (Sh-NT) that were either cocul-tured (Sh-NTþOB; shRNA#1þOB; Fig. 3H; each conditionincludes its own gating control, small insert in each panel) ornot cocultured (Sh-NT-OB; shRNA#1-OB; Fig. 3H) with osteo-blasts. In the absence of osteoblast coculture, NT and MCAMknockdown cells each displayed similar percentages of ALDHhigh

cells (CTRL bars, Fig. 3I). However, when cocultured with osteo-blasts (OB), the MCAM knockdown prostate cancer cells dis-played a significant reduction in the percentage of ALDHhigh cells,when compared with NT prostate cancer cells cocultured withosteoblasts (P < 0.01; Fig. 3I, OB bars). This result indicates thatMCAM plays a role in maintaining levels of ALDH activity inprostate cancer cells in the presence of osteoblast cells in cocultureand suggests that it may play a similar role within the osteoblasticmicroenvironment. To test whether there is cross-talk betweenMCAM and the androgen signaling in the context of prostatecancer bone metastasis, we evaluated the expression level ofMCAM and its extracellular matrix interaction partner lamininalpha 4 (LAMA4) in the GSE101607 dataset (34), which containsfresh-frozen bone metastasis samples from AR-driven (N ¼ 32)and non-AR–driven (N ¼ 8) tumors. We found a significantincrease in MCAM (P ¼ 0.044) and LAMA4 (P ¼ 0.045) innon–AR-driven conditions compared with AR-driven (Fig. 4Aand B). This reinforces the hypothesis that MCAM levels increaseupon disease progression. Furthermore, we evaluated the effect ofDHT administration in C4-2B cells under high-castration condi-tions in medium with Charcoal Stripped Serum (CSS) for 72hours (ref. 39; Fig. 4C). The androgen response was compared incontrol andMCAMknockdown cells by evaluating the expressionof 4 androgen responsive genes uponDHT stimulation (10 nmol/L; ref. 40). C4-2B cells displayed an intact AR machinery asindicated by the strong upregulation of FKBP5, KLK3, andTMPRSS2 and by the downregulation ofOPRK1 (Fig. 4D; ref. 40).No differencewas detected between control and knockdown cells,indicating that MCAM has no influence on the androgen respon-siveness in C4-2B cells. Similarly, administration of the antian-drogen MDV3100 (enzalutamide) revealed no difference in themodulationof theAR responsive genes in control andknockdowncells (Fig. 4E). Interestingly, we observed a consistent down-regulation of OPRK1 in MCAM knockdown cells compared withcontrol. Finally, although no change was observed on MCAMexpression upon MDV3100 treatment compared with DMSOcontrol, DHT administration resulted in a strong reduction ofMCAM expression compared with castration (EtOH control inCSS; Fig. 4F). This supports our dataset analysis and indicates thatMCAM levels are higher upon castration.

MCAM knockdown reduces prostate cancer lytic bonemetastasis in a preclinical mouse model

To test the impact of MCAM knockdown on bonemetastasis ofprostate cancer cells, we performed intraosseous inoculation ofPC-3M-Pro4MCAMknockdown cells andNT control cells expres-sing luciferase 2 (Luc2) in male mice (sham-operated mice wereincluded as an additional control). Animals were monitored byBLI during the course of the experiment (4 weeks; Fig. 5A; allanimals, at day 7 and day 28 displayed in Supplementary Fig.S3A). Luciferase activity of MCAM knockdown and NT controlcells was assessed by serial cell dilution to confirm that there wasno external influence of the shRNAs employed on BLI signaldetection (Supplementary Fig. S3B). Body weight of the animalswas monitored along the course of the experiment (Supplemen-tary Fig. S3C) and the development of bone lesions was assessedbyX-raymeasurements (Fig. 5B; data for allmice, including sham,at day 7 and day 28 are displayed in Supplementary Fig. S3D).Wefound that MCAM knockdown strongly reduced the lytic pheno-type of PC-3M-Pro4 cells compared with NT control. This wasconfirmed by bone morphometric analysis (Fig. 5C;

Zoni et al.

Mol Cancer Res; 17(5) May 2019 Molecular Cancer Research1054

on January 13, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 11, 2019; DOI: 10.1158/1541-7786.MCR-18-1220

Supplementary Fig. S3E; analysis at day 28) and histologic eval-uation (Fig. 5D; Supplementary Fig. S3F). Bone morphometryrevealed that bone area in knockdown was similar to sham andsignificantly higher inmice injectedwithMCAMknockdown cellscompared with those injected with NT control cells (P < 0.05;Fig. 5E) despite the fact that BLI measurement revealed similartumor burden between MCAM knockdown and NT control cells(Fig. 5F). Finally, we used RT-qPCR tomeasure the expression of apanel of genes previously identified as regulators of multiplesteps of the bone metastatic cascade (TDGF1/CRIPTO, PMEPA1,COL1a, VEGFa, DKK1, PTHLH, MSF; refs. 17, 41). We found astrong inhibitionof the oncogeneCRIPTO/TDGF1 (42) inMCAMknockdown cells compared with NT control (P < 0.05) and a

general modulation of molecules involved in bone remodeling(e.g., PTHLH and DKK1; Fig. 5G). The downmodulation ofCRIPTO/TDGF1 in MCAM knockdown cells was confirmed alsoat protein level (Fig. 5H). Taken together, our data suggest thatMCAM influences the expression of molecules that are importantmodulators of bone remodeling and bone metastasis.

MCAM knockdown impacts expression of genes that regulatehematopoiesis and bone remodeling

To identify the putative mechanisms of action of MCAM, weperformed RNA-seq on MCAM knockdown cells. Unsupervisedhierarchical clustering demonstrated a separation betweenMCAM knockdown and control NT samples (Fig. 6A; heatmap

Figure 3.

Effects of MCAM knockdown on EMTmarkers and characterization of coculture with human osteoblasts. A and B,MCAM knockdown results in increased ratio ofE-cadherin/vimentin and E-cadherin/N-cadherin. Data are representative of three independent samples and normalized with housekeeping gene. Data arerepresented as mean� SD. C and D,Western blot analysis for E-cadherin expression and quantification. Calculation is normalized to expression of actin. Data arerepresentative of three independent samples. Data are represented as mean� SD. E and F,Western blot analysis for vimentin expression, quantification, andprotein ratio E-cadherin/vimentin. Calculation is normalized to expression of housekeeping (actin). Data are representative of three independent samples. Dataare represented as mean� SD. G, Schematic representation of the coculture experiment. PC-3M-Pro4 cells expressing dTomato fluorescent protein are culturedin direct contact with differentiated human osteoblasts for 48 hours prior to viable cells sorting. H, Representative images of FACS analysis on ALDH activity onPC-3M-Pro4dTomato cells with nontargeted (sh-NT) without osteoblasts (�OB) or with osteoblasts (þOB). Assay was performed immediately after sorting.Same conditions are shown for the MCAM knockdown cells (two right panels, H). In each panel the small insert represents the control used for gating accordingfrommanufacturer's protocol. Quantification is displayed in I. Data are represented as mean� SEM (�� , P < 0.01 with Bonferroni Multiple Comparison Test).

Targeting MCAM Reduces Bone Metastasis in Prostate Cancer

www.aacrjournals.org Mol Cancer Res; 17(5) May 2019 1055

on January 13, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 11, 2019; DOI: 10.1158/1541-7786.MCR-18-1220

and volcano plot in top and bottom, respectively). Differentialexpression analysis between MCAM knockdown and NT controlrevealed that 55 different genes were significantly upregulated(>2-fold increase, FDR<0.05) and99 significantly downregulated(>2-fold decrease, FDR < 0.05; Supplementary Table S2 andSupplementary Table S3). We performed GSEA (27) to identifybiological processes and pathways modulated by MCAM knock-down (Fig. 6B). We found that the upregulated genes in MCAMknockdown cells were enriched for the process related to negativeregulation of hematopoiesis [normalized enrichment score(NES) ¼ 1.68, P ¼ 0.02; Fig. 6C; Supplementary Table S6, forthe list of 14 genes involved in the pathway). This supports ourprevious findings (6) and the notion that increasing the numberofHSCniches increasesmetastatic growth in thebonemarrow(5).In addition, we found an enrichment of GeneOntology processes

of biomineral tissue development (NES ¼ 1.8, P ¼ 0.002) andbone mineralization (NES ¼ 1.68, P ¼ 0.008) among the genesupregulated uponMCAM knockdown (Fig. 6D–F). This supportsthe hypothesis that MCAM knockdown disrupts signaling path-ways associated to bone remodeling and are in accordance withour in vivo experiment.

Among the genes downregulated in MCAM knockdown cells,we found an enrichment of genes involved in G0 and early G1

(NES ¼ �1.5, P ¼ 0.02) and in E2F-mediated regulation of DNAreplication (NES ¼ �1.5, P ¼ 0.01; Supplementary Fig. S4A andS4B; Supplementary Table S5 and S6). These data reinforce ourfinding that MCAM knockdown cells displayed reduced prolifer-ation in vitro. Similar results were obtained using an alternativegene set analysis method g:Profiler (Supplementary Fig. S4Cand S4D).

Figure 4.

Expression of MCAM in non-AR–driven prostate cancer and in vitro functional effect of DHT treatment. A and B,MCAM and LAMA4 expression in prostate cancerbone metastasis from AR-driven and non-AR–driven disease (in GSE101607). C, Schematic representation of in vitro castration and AR stimulation. N¼ 3technical replicates. CSS; Charcoal Stripped Serum. D, Effect of DHT on AR-responsive genes. Results are expressed as the mean� SD. E, Effect of MDV3100on AR-responsive genes. F, Effect of AR stimulation or inhibition on MCAM expression. Results are expressed as the mean� SD, � , P < 0.05; �� , P < 0.01; and��� , P < 0.001 with t test; $$$, P < 0.001 between EtOH control and DHT in the respective cell clone.

Zoni et al.

Mol Cancer Res; 17(5) May 2019 Molecular Cancer Research1056

on January 13, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 11, 2019; DOI: 10.1158/1541-7786.MCR-18-1220

Targeting MCAM with a mAb reduces intraosseous growth anddiminishes the extension of lytic lesions in an intraosseousmodel

To test the effect of targetingMCAMon both the tumor and thestroma in apreclinical intraosseousmodel of prostate cancer bonemetastasis, we employed an anti-human and anti-mouse MCAMmAb. Five animals per experimental group (anti-human, anti-mouse, anti-human þ mouse, control IgG rat, control IgGhuman) were pretreated with 10 mg/kg of mAbs and controlsIgG intraperitoneally the day prior to intrabone injection ofprostate cancer cells. After intrabone injection, all animalsreceived administration of themAbs or controls IgG every secondday at a dose of 10 mg/kg intraperitoneally. Animals were mon-itored by BLI during the course of the experiment (5 weeks,Fig. 7A, all animals at end of experiment displayed in Supple-mentary Fig. S5A). Administration of anti-human MCAM mAb

resulted in significantly smaller intraosseous growth (P < 0.05;Fig. 7B). Body weight was evaluated, and development of bonelesions was monitored during the experiment with X-ray (Fig. 7Cand D, all animals displayed in Supplementary Fig. S5B andstatistic in Supplementary Fig. S5C). In addition, luciferase activ-ity of prostate cancer cells was evaluated prior to in vivo experi-mentation (Supplementary Fig. S5D). Bone morphometric anal-ysis and histologic evaluation revealed a higher bone area inanimals treated with anti-human mAbs compared with otherexperimental groups (P<0.05; Fig. 7E and F, all animals displayedin Supplementary Fig. S6A–S6D; statistical evaluation displayedin Supplementary Fig. S6B and S6C). Evaluation of the kidneyhistology revealed normal and comparable histologic character-istics in all the experimental groups (Supplementary Fig. S7A).Contralateral bones as control of normal histologic features werecollected and displayed in Supplementary Fig. S7B.

Figure 5.

CB17 SCIDmale mice inoculated with PC-3M-Pro4Luc2 human prostate cancer cells with nontargeting shRNA (sh-NT) or MCAM-targeting shRNA (shRNA#1) orsham animals. 50,000 cells/animal. See Supplementary information for representation of all the single animals. A–C, Representative BLI images and bonemeasurements with X-ray (B) and bone morphometric analysis (C). D, Representative histologic analysis (hematoxylin and eosin staining) of sections of tibia(T, tumor; B, bone; BM, bone marrow). E,Quantification of the bone area at day 28. Data are represented as mean� SD. F,Quantification of tumor burden by BLIimaging. G, Expression of genes related to bone metastasis in MCAM KD and nontargeted control (NT) prostate cancer cells. N¼ 3 independent experiments,represented as� SD. H,Western blotting for Cripto expression. Data are representative of three independent protein isolations.

Targeting MCAM Reduces Bone Metastasis in Prostate Cancer

www.aacrjournals.org Mol Cancer Res; 17(5) May 2019 1057

on January 13, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 11, 2019; DOI: 10.1158/1541-7786.MCR-18-1220

Figure 6.

RNA-seq and GSEA.A, Heatmap illustrating hierarchical clustering of the samples, based on the 500most variable genes and Volcano plot of�log10(P) againstlog2(fold change). Red dots are those with at least 2-fold difference and FDR� 0.05 (i.e., differentially expressed genes). B, GSEA performed on C5_biologicalprocesses. Red bars, upregulated pathways; blue bars, downregulated pathways. Numbers on the left (of the blue bars) or on the right (of the red bars) indicatethe amount of genes enriched for each gene sets. Data are sorted for NES (x-axis). C, GSEA plot for negative regulators of hematopoiesis. NES and P value areindicated in the plot. D,GSEA enrichment plot for biomineral tissue development. E, GSEA enrichment plot for bone mineralization. The plots C–E illustrateprofiles of the NES and positions of the genes on the rank-ordered list in GSEA. Statistical P value is also indicated. F, Enrichment map, for C5 (biologicalprocesses) displays sets of genes upregulated (red) or downregulated (blue) in MCAM knockdown cells.

Zoni et al.

Mol Cancer Res; 17(5) May 2019 Molecular Cancer Research1058

on January 13, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 11, 2019; DOI: 10.1158/1541-7786.MCR-18-1220

DiscussionIn this study, we provide evidence that MCAM drives lytic

metastatic human prostate cancer and has potential as a thera-peutic target in metastatic prostate cancer.

Our in vivo experiments revealed that MCAM knockdown inprostate cancer cells reduces their ability to generate lytic lesions

upon inoculation into the bone. We found that MCAM knock-down cells and NT control cells displayed similar tumor burdenin vivo despite showing different cell proliferation rates in vitro.This supports the notion that, in the presence of the bonemicroenvironment, MCAM has a prominent role in modulatingthe process of bone remodeling and suggests that, in this context,other mechanisms sustain the proliferation of tumor cells.

Figure 7.

CB17 SCIDmale mice inoculated with PC-3M-Pro4Luc2 human prostate cancer cells intrabone and treated with anti-MCAMmAb. A, Representative BLI images ofseparate experimental groups (5 animal/group, see Supplementary Materials and Methods for representation of all the single animals. 50,000 cells/animal; in thelabel descriptions of the group names; H, human; M, mouse). B,Quantification of tumor burden by BLI imaging. Data are represented as mean� SEM. C,Representative X-ray images of separate experimental groups. See Supplementary Information for representation of all the single animals. D, Body weight ofanimals measured over the course of the experiment. Data are represented as mean� SD. E, Representative images of bone morphometric analysis. SeeSupplementary information for representation of all the single animals and image processing. F,Quantification of the bone area at the end of experiment. Dataare represented as mean� SD (� , P < 0.05 with one way ANOVA and Bonferroni multiple comparison test).

Targeting MCAM Reduces Bone Metastasis in Prostate Cancer

www.aacrjournals.org Mol Cancer Res; 17(5) May 2019 1059

on January 13, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 11, 2019; DOI: 10.1158/1541-7786.MCR-18-1220

Treatment with an anti-MCAM mAb revealed significantimpact on lytic lesions in the mice bearing prostate cancercells in the bone. This finding is in line with the results of ourin vivo experiment with MCAM knockdown. However, theadministration of anti-MCAM mAb also resulted in lowertumor burden in the intraosseous model. This might be relatedto the residual level of MCAM in our knockdown line (approx-imately 30% knockdown on the protein level) compared withthe administered mAb. We showed here that the reduction inMCAM expression is sufficient to influence the expression ofthe oncogenic driver CRIPTO/TDGF1 (43). This suggests thatin MCAM knockdown in prostate cancer cells, the proliferationin vitro might be supported by TDGF1. Consistent with thispossibility, we have previously shown that TDGF1 knockdowndecreases cell proliferation in vitro and bone metastases in vivoand that TDGF1 mRNA increases upon direct coculture ofhuman prostate cancer cells with human osteoblasts (17).Although MCAM knockdown does not prevent prostate cancergrowth in bone, it does impact on the lytic phenotype. Thisrole of MCAM is supported by our RNA-seq data and tran-scriptional analysis on a set of genes that was previouslyshown to regulate bone metastasis in lytic prostate cancercells (41). MCAM knockdown reduces the mRNA of theosteolytic factor PTHLH (44) and the Wnt inhibitor dick-kopf-1 (DKK-1; ref. 45). DKK-1 expression was proposed asan early event in skeletal metastasis, thus favoring osteolysis atthe metastatic site (45). In addition, the concurrent down-regulation of PTHLH upon MCAM knockdown might explainthe effect on cell proliferation measured in vitro, given thatPTHLH was shown to play a role in prostate cancer cells'proliferation (46).

We found that in human patients with prostate cancer, MCAMis significantly elevated in androgen ablation–resistantmetastasesrelative to primary tumors. This is in line with our findings thatMCAM is elevated upon castration. Moreover, this matches pre-vious clinical literature reporting thatMCAMexpression is strong-ly related to poor prognosis in a variety of carcinomas includingprostate cancer (37).

We have shown that although it is possible to isolate afraction of MCAMhigh and MCAMlow cells, their molecularfeatures do not necessarily overlap with those of "bulk"ALDHhigh and ALDHlow cells as shown by our clonogenicassay. This suggests that further subpopulations of cells mightdisplay independent phenotypic characteristics as we showedhere for the MCAMhigh fraction.

Our findings also suggest that the modulation of MCAMexpression might impact on the transcriptional program ofEMT-related genes. This is in line with previous studies showingthat MCAM promotes EMT and EMT-related drug resistance (47).However, our data on protein expression highlight a lack ofsignificant effect on the detection of E-cadherin and vimentin atthe protein level, possibly due to residual MCAM protein. Extrav-asation is one of the features of EMT and MCAM knockdownprostate cancer cells behaved similarly to control cells wheninjected into zebrafish. Treatment with DHT and MDV3100revealed no difference in the response of control and knockdowncells to androgen stimulation or blockade. However, administra-tion of DHT resulted in a strong decrease in MCAM expressioncompared with castration. This finding and our analysis of twoprostate cancer bone metastasis subsets of AR-driven and non-AR–driven disease supports the involvement of MCAM during

disease progression and reinforce the hypothesis of targetingMCAM in advanced disease.

In conclusion, we have analyzed the role of MCAM in prostatecancer cells by investigating its biological function with modelsthat recapitulate the presence and absence of androgens and theostoblastic niche in vitro and the context of the bone microenvi-ronment in vivo. Although additional studies are required todissect the molecular function of MCAM, our data indicate thatMCAM is required for producing the lytic phenotype in prostatecancer bonemetastasis. Moreover, we showed here that treatmentwith anti-MCAM mAb has a significant impact on prostatecancer cells' growth intrabone and diminishes the extension oflytic lesions. Our findings affirm the potential utility of MCAM-targeting agents that are able to interfere with its biologicalfunction for use in treating metastatic disease to the bone. Thecombination of these agents with currently available drugs thattarget prostate cancer growth might lead to better treatmentfor patients with prostate cancer, especially those with life-threatening metastatic disease.

Disclosure of Potential Conflicts of InterestK. Flanagan is the head of cell biology at Prothena Biosciences. No potential

conflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: E. Zoni, L. Astrologo, K. Flanagan, G. van der Pluijm,M.G. Cecchini, M. Kruithof-de Julio, G.N. ThalmannDevelopment of methodology: E. Zoni, S. Piscuoglio, L. Chen, E.B. Snaar-Jagalska, M. Kruithof-de Julio, G.N. ThalmannAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): E. Zoni, L. Astrologo, J. Melsen, J. Grosjean,I. Klima, L. Chen, E.B. Snaar-Jagalska, G. van der Pluijm, P. Kloen,M. Kruithof-de Julio, G.N. ThalmannAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): E. Zoni, L. Astrologo, C.K.Y. Ng, S. Piscuoglio,J. Melsen, M. Kruithof-de Julio, G.N. ThalmannWriting, review, and/or revision of the manuscript: E. Zoni, L. Astrologo,C.K.Y. Ng, S. Piscuoglio, E.B. Snaar-Jagalska, K. Flanagan, G. van der Pluijm,M. Kruithof-de Julio, G.N. ThalmannAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): E. Zoni, M. Kruithof-de Julio, G.N. ThalmannStudy supervision: E. Zoni, E.B. Snaar-Jagalska, M. Kruithof-de Julio,G.N. Thalmann

AcknowledgmentsThe authors thank Guido de Roo and Sabrina Veld from the Flow Cytometry

Facility (Department of Hematology, LeidenUniversityMedical Center, Leiden,theNetherlands).We alsowould like to thank StefanM€uller, BernadetteNyfeler,and Thomas Schaffer from the FACS laboratory (Department of BioMedicalResearch, University of Bern, Bern, Switzerland). Monoclonal anti-humanMCAMand rat anti-mouseMCAMantibodieswere kindly providedbyProthenaBiosciences. This project received support by the Swiss National Science Foun-dation (310030_156933 to G.N. Thalmann and 31003A_169352 toM. Kruithof-de Julio) and by the Dutch Cancer Society, grant no. UL2015-7599KWF toM.Kruithof-de Julio. Additionalfinancial supportwas providedbythe Swiss Cancer League (KFS-3995-08-2016 to S. Piscuoglio) and the SwissNational Science Foundation (Ambizione PZ00P3_168165 to S. Piscuoglio). Inmemory of Marco G. Cecchini.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received November 14, 2018; revised December 6, 2018; accepted February6, 2019; published first February 11, 2019.

Zoni et al.

Mol Cancer Res; 17(5) May 2019 Molecular Cancer Research1060

on January 13, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 11, 2019; DOI: 10.1158/1541-7786.MCR-18-1220

References1. Jemal A, Center MM, DeSantis C, Ward EM. Global patterns of cancer

incidence and mortality rates and trends. Cancer Epidemiol BiomarkersPrev 2010;19:1893–907.

2. BollaM, Van TienhovenG,Warde P, Dubois JB, Mirimanoff RO, StormeG,et al. External irradiation with or without long-term androgen suppressionfor prostate cancer with high metastatic risk: 10-year results of an EORTCrandomised study. Lancet Oncol 2010;11:1066–73.

3. La Manna F, Karkampouna S, Zoni E, De Menna M, Hensel J, ThalmannGN, et al. Metastases in prostate cancer. Cold Spring Harb Perspect Med2019;9:a033688.

4. Hensel J, Thalmann GN. Biology of bone metastases in prostate cancer.Urology 2016;92:6–13.

5. Shiozawa Y, Pedersen EA, Havens AM, Jung Y, Mishra A, Joseph J, et al.Human prostate cancermetastases target the hematopoietic stem cell nicheto establish footholds in mouse bone marrow. J Clin Invest 2011;121:1298–312.

6. €Ozdemir BC, Hensel J, Secondini C, Wetterwald A, Schwaninger R,Fleischmann A, et al. The molecular signature of the stroma response inprostate cancer-induced osteoblastic bone metastasis highlights expan-sion of hematopoietic and prostate epithelial stem cell niches.PLoS One 2014;9:e114530.

7. Sers C, Kirsch K, Rothbacher U, Riethmuller G, Johnson JP. Genomicorganization of the melanoma-associated glycoprotein MUC18: implica-tions for the evolutionof the immunoglobulin domains. ProcNatl Acad SciU S A 1993;90:8514–8.

8. Wu GJ, Wu MW, Wang SW, Liu Z, Qu P, Peng Q, et al. Isolation andcharacterization of the major form of human MUC18 cDNA gene andcorrelation of MUC18 over-expression in prostate cancer cell lines andtissues with malignant progression. Gene 2001;279:17–31.

9. Xie S, LucaM,Huang S, GutmanM, Reich R, Johnson JP, et al. Expression ofMCAM/MUC18 by human melanoma cells leads to increased tumorgrowth and metastasis. Cancer Res 1997;57:2295–303.

10. Kristiansen G, Yu Y, Schl€uns K, Sers C, Dietel M, Petersen I. Expression ofthe cell adhesion molecule CD146/MCAM in non-small cell lung cancer.Anal Cell Pathol 2003;25:77–81.

11. Aldovini D, Demichelis F, Doglioni C, Di VizioD, Galligioni E, Brugnara S,et al. M-CAM expression as marker of poor prognosis in epithelial ovariancancer. Int J Cancer 2006;119:1920–6.

12. Zabouo G, Imbert AM, Jacquemier J, Finetti P, Moreau T, Esterni B, et al.CD146 expression is associated with a poor prognosis in human breasttumors andwith enhancedmotility in breast cancer cell lines. Breast CancerRes 2009;11:R1.

13. Wu GJ, WuMW,Wang C, Liu Y. Enforced expression of METCAM/MUC18increases tumorigenesis of human prostate cancer LNCaP Cells in NudeMice. J Urol 2011;185:1504–12.

14. Zeng Q, Li W, Lu D, Wu Z, Duan H, Luo Y, et al. CD146, an epithelial-mesenchymal transition inducer, is associated with triple-negative breastcancer. Proc Natl Acad Sci U S A 2012;109:1127–32.

15. Sacchetti B, Funari A, Michienzi S, Di Cesare S, Piersanti S, Saggio I, et al.Self-renewing osteoprogenitors in bone marrow sinusoids can organize ahematopoietic microenvironment. Cell 2007;131:324–36.

16. Bardin N. Identification of CD146 as a component of the endothelialjunction involved in the control of cell-cell cohesion. Blood 2001;98:3677–84.

17. Zoni E, Chen L, Karkampouna S, Granchi Z, Verhoef EI, La Manna F,et al. CRIPTO and its signaling partner GRP78 drive the metastaticphenotype in human osteotropic prostate cancer. Oncogene 2017;36:4739–49.

18. Zoni E, van derHorst G, van deMerbel AF, Chen L, Rane JK, Pelger RC, et al.miR-25 modulates invasiveness and dissemination of human prostatecancer cells via regulation of alphav- and alpha6-Integrin Expression.Cancer Res 2015;75:2326–36.

19. Kroon J, in 't Veld LS, Buijs JT, CheungH, vanderHorstG, vander PluijmG.Glycogen synthase kinase-3b inhibition depletes the population of pros-tate cancer stem/progenitor-like cells and attenuates metastatic growth.Oncotarget 2013;5:8986–94.

20. Bassett JH, van der Spek A, Gogakos A, Williams GR. Quantitative X-rayimaging of rodent bone by Faxitron. Methods Mol Biol 2012;816:499–506.

21. Breuer J, Korpos E, Hannocks MJ, Schneider-Hohendorf T, Song J, ZondlerL, et al. Blockade ofMCAM/CD146 impedes CNS infiltration of T cells overthe choroid plexus. J Neuroinflammation 2018;15:236.

22. Flanagan K, Fitzgerald K, Baker J, Regnstrom K, Gardai S, Bard F, et al.Laminin-411 is a vascular ligand for MCAM and facilitates TH17 cell entryinto the CNS. PLoS One 2012;7:e40443.

23. Stoletov K, Montel V, Lester RD, Gonias SL, Klemke R. High-resolutionimaging of the dynamic tumor cell vascular interface in transparentzebrafish. Proc Natl Acad Sci U S A 2007;104:17406–11.

24. Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, et al. STAR:ultrafast universal RNA-seq aligner. Bioinformatics 2013;29:15–21.

25. Wilkerson MD, Hayes DN. ConsensusClusterPlus: a class discovery toolwith confidence assessments and item tracking. Bioinformatics 2010;26:1572–3.

26. NikolayevaO, RobinsonMD. edgeR for differential RNA-seq andChIP-seqanalysis: an application to stem cell biology.MethodsMol Biol 2014;1150:45–79.

27. SubramanianA, TamayoP,Mootha VK,Mukherjee S, Ebert BL,GilletteMA,et al. Gene set enrichment analysis: a knowledge-based approach forinterpreting genome-wide expression profiles. Proc Natl Acad Sci U S A2005;102:15545–50.

28. Liberzon A, Subramanian A, Pinchback R, Thorvaldsdottir H, Tamayo P,Mesirov JP. Molecular signatures database (MSigDB) 3.0. Bioinformatics2011;27:1739–40.

29. Reimand J, Arak T, Adler P, Kolberg L, Reisberg S, Peterson H, et al. g:Profiler-a web server for functional interpretation of gene lists (2016update). Nucleic Acids Res 2016;44:W83–9.

30. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al.Cytoscape: a software environment for integrated models of biomolecularinteraction networks. Genome Res 2003;13:2498–504.

31. Dumas J, Gargano MA, Dancik GM. shinyGEO: a web-based applicationfor analyzing gene expression omnibus datasets. Bioinformatics 2016;32:3679–81.

32. Chandran UR, Ma C, Dhir R, Bisceglia M, Lyons-Weiler M, Liang W,et al. Gene expression profiles of prostate cancer reveal involvement ofmultiple molecular pathways in the metastatic process. BMC Cancer2007;7:64.

33. Yu YP, Landsittel D, Jing L, Nelson J, Ren B, Liu L, et al. Gene expressionalterations in prostate cancer predicting tumor aggression and precedingdevelopment of malignancy. J Clin Oncol 2004;22:2790–9.

34. Ylitalo EB, Thysell E, Jernberg E, Lundholm M, Crnalic S, Egevad L, et al.Subgroups of castration-resistant prostate cancer bone metastases definedthrough an inverse relationship between androgen receptor activity andimmune response. Eur Urol 2017;71:776–87.

35. Goswami CP, Nakshatri H. PROGgene: gene expression based survivalanalysis web application for multiple cancers. J Clin Bioinforma2013;3:22.

36. Gulzar ZG, McKenney JK, Brooks JD. Increased expression of NuSAP inrecurrent prostate cancer is mediated by E2F1. Oncogene 2013;32:70–7.

37. Wu GJ, Varma VA, WuMW,Wang SW, Qu P, Yang H, et al. Expression of ahuman cell adhesion molecule, MUC18, in prostate cancer cell lines andtissues. Prostate 2001;48:305–15.

38. van den Hoogen C, van der Horst G, Cheung H, Buijs JT, Lippitt JM,Guzman-Ramirez N, et al. High aldehyde dehydrogenase activity identifiestumor-initiating and metastasis-initiating cells in human prostate cancer.Cancer Res 2010;70:5163–73.

39. Sedelaar JP, Isaacs JT. Tissue culture media supplemented with 10% fetalcalf serum contains a castrate level of testosterone. Prostate 2009;69:1724–9.

40. Cai C,HeHH,Chen S, Coleman I,WangH, FangZ, et al. Androgen receptorgene expression in prostate cancer is directly suppressed by the androgenreceptor through recruitment of lysine-specific demethylase 1. Cancer Cell2011;20:457–71.

41. Fournier PG, Juarez P, Jiang G, Clines GA, Niewolna M, Kim HS, et al. TheTGF-beta signaling regulator PMEPA1 suppresses prostate cancer metas-tases to bone. Cancer Cell 2015;27:809–21.

42. KlauzinskaM, CastroNP, RangelMC, Spike BT, Gray PC, Bertolette D, et al.The multifaceted role of the embryonic gene Cripto-1 in cancer, stem cellsand epithelial-mesenchymal transition. Semin Cancer Biol 2014;29:51–8.

Targeting MCAM Reduces Bone Metastasis in Prostate Cancer

www.aacrjournals.org Mol Cancer Res; 17(5) May 2019 1061

on January 13, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 11, 2019; DOI: 10.1158/1541-7786.MCR-18-1220

43. Terry S, El-Sayed IY, Destouches D, Maille P, Nicolaiew N,Ploussard G, et al. CRIPTO overexpression promotes mesenchy-mal differentiation in prostate carcinoma cells through parallelregulation of AKT and FGFR activities. Oncotarget 2015;6:11994–2008.

44. Powell GJ, Southby J, Danks JA, Stillwell RG, Hayman JA, Henderson MA,et al. Localization of parathyroid hormone-related protein in breast cancermetastases: increased incidence in bone compared with other sites.Cancer Res 1991;51:3059–61.

45. Hall CL, Kang S,MacDougaldOA,Keller ET. Role ofWnts in prostate cancerbone metastases. J Cell Biochem 2006;97:661–72.

46. Asadi F, Faraj M, Malakouti S, Kukreja SC. Effect of parathyroid hormonerelated protein, and dihydrotestosterone on proliferation and ornithinedecarboxylasemRNA in human prostate cancer cell lines. Int Urol Nephrol2001;33:417–22.

47. Liu WF, Ji SR, Sun JJ, Zhang Y, Liu ZY, Liang AB, et al. CD146 expressioncorrelates with epithelial-mesenchymal transition markers and a poorprognosis in gastric cancer. Int J Mol Sci 2012;13:6399–406.

Mol Cancer Res; 17(5) May 2019 Molecular Cancer Research1062

Zoni et al.

on January 13, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 11, 2019; DOI: 10.1158/1541-7786.MCR-18-1220

2019;17:1049-1062. Published OnlineFirst February 11, 2019.Mol Cancer Res   Eugenio Zoni, Letizia Astrologo, Charlotte K.Y. Ng, et al.   in Prostate CancerTherapeutic Targeting of CD146/MCAM Reduces Bone Metastasis

  Updated version

  10.1158/1541-7786.MCR-18-1220doi:

Access the most recent version of this article at:

  Material

Supplementary

  http://mcr.aacrjournals.org/content/suppl/2019/02/09/1541-7786.MCR-18-1220.DC1

Access the most recent supplemental material at:

   

   

  Cited articles

  http://mcr.aacrjournals.org/content/17/5/1049.full#ref-list-1

This article cites 47 articles, 13 of which you can access for free at:

  Citing articles

  http://mcr.aacrjournals.org/content/17/5/1049.full#related-urls

This article has been cited by 1 HighWire-hosted articles. Access the articles at:

   

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

  Subscriptions

Reprints and

  .pubs@aacr.org

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/17/5/1049To request permission to re-use all or part of this article, use this link

on January 13, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 11, 2019; DOI: 10.1158/1541-7786.MCR-18-1220