chemokine signaling facilitates early-stage breast cancer ... · model inhibited tumor invasion and...

Oncogenes and Tumor Suppressors

Chemokine Signaling Facilitates Early-StageBreast Cancer Survival and Invasion throughFibroblast-Dependent MechanismsGage Brummer, Diana S. Acevedo, Qingting Hu, Mike Portsche,Wei Bin Fang,Min Yao, Brandon Zinda, Megan Myers, Nehemiah Alvarez, Patrick Fields,Yan Hong, Fariba Behbod, and Nikki Cheng

Abstract

Ductal carcinoma in situ (DCIS) is the most common form ofbreast cancer, with 50,000 cases diagnosed every year in theUnited States. Overtreatment and undertreatment remain sig-nificant clinical challenges in patient care. Identifying keymechanisms associated with DCIS progression could uncovernew biomarkers to better predict patient prognosis andimprove guided treatment. Chemokines are small solublemolecules that regulate cellular homing through moleculargradients. CCL2-mediated recruitment of CCR2þ macrophagesare a well-established mechanism for metastatic progression.Although the CCL2/CCR2 pathway is a therapeutic target ofinterest, little is known about the role of CCR2 expression inbreast cancer. Here, using a mammary intraductal injection(MIND) model to mimic DCIS formation, the role of CCR2 wasexplored in minimally invasive SUM225 and highly invasiveDCIS.com breast cancer cells. CCR2 overexpression increasedSUM225 breast cancer survival and invasion associated with

accumulation of CCL2 expressing fibroblasts. CCR2-deficientDCIS.com breast cancer cells formed fewer invasive lesionswith fewer CCL2þ fibroblasts. Cografting CCL2-deficient fibro-blasts with DCIS.com breast cancer cells in the subrenal capsulemodel inhibited tumor invasion and survival associated withdecreased expression of aldehyde dehydrogenase (ALDH1), aproinvasive factor, and decreased expression of HTRA2, aproapoptotic serine protease. Through data mining analysis,high expression of CCR2 and ALDH1 and low HTRA2 expres-sion were correlated with poor prognosis of breast cancerpatients.

Implications: This study demonstrates that CCR2 overexpres-sion in breast cancer drives early-stage breast cancer progres-sion through stromal-dependent expression of CCL2 withimportant insight into prognosis and treatment of DCIS.Mol Cancer Res; 16(2); 296–308. �2017 AACR.

IntroductionDuctal carcinoma in situ (DCIS) is the most common form of

preinvasive breast cancer in the United States, with over 50,000cases diagnosed every year. Standard treatment for DCIS involvesa combination of lumpectomy and radiotherapy (1, 2). Yet,10%–35%of patients experience disease recurrence, often accom-panied by invasive ductal carcinoma (IDC; refs. 3, 4), indicatingthat undertreatment and overtreatment remain significant con-cerns in patient care. Fewapproaches exist to evaluate prognosis ofDCIS. Comparedwith IDC, the use of biomarkers inDCIS has notbeen well studied. Small- or low-grade lesions may still becomeinvasive (4, 5). Estrogen receptor (ER), Her2, Ki67, p16, andCox2are associated with disease recurrence but not with development

of invasive breast cancer (6). Identifying key mechanisms asso-ciated with DCIS progression could lead to better prognosticfactors and tailored treatments for patients with DCIS.

Chemokines are small soluble molecules (8 kDa), which formmolecular gradients tomediate homing of immune cells to tissuesduring inflammation. Chemokines signal to seven transmem-brane receptors that couple to G protein–dependent and -inde-pendent pathways to promote cell migration (7, 8). CCL2/CCR2chemokine signaling is a critical regulator of macrophage recruit-ment during wound healing and infection (9). CCL2 and CCR2are overexpressed in multiple cancer types including: pancreatic,prostate, and colon cancers and breast cancer correlating withpoor patient prognosis (10, 11). In breast cancer, animal modelshave demonstrated that CCL2 recruits CCR2þ macrophages topromote tumor growth andmetastasis (11, 12). The CCL2/CCR2pathway is a current therapeutic target of interest (11), but little isknown about mechanisms of this pathway in cancer beyondsignaling in immune cells.

We recently found that CCR2 is overexpressed in breast cancercells and regulatesCCL2-induced cell survival andmigration (13),indicating a macrophage-independent role for CCL2 in breastcancer. Using a novel Mammary Intraductal injection (MIND)model of DCIS, we demonstrate that CCR2 overexpression inDCIS lesions enhances invasive progression associated with accu-mulation of CCL2-expressing fibroblasts. Using the subrenalcapsule model, we demonstrate that fibroblasts derived from

Department of Pathology and Laboratory Medicine, University of KansasMedical Center, Kansas City, Kansas.

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

G. Brummer and D.S. Acevedo contributed equally to this article.

Corresponding Author: Nikki Cheng, University of Kansas Medical Center, 3901Rainbow Boulevard, Wahl Hall East 1020 Mailstop 3003, Kansas City, KS 66160.Phone: 913-945-6773; Fax: 913-945-6650; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-17-0308

�2017 American Association for Cancer Research.

MolecularCancerResearch

Mol Cancer Res; 16(2) February 2018296

on March 6, 2020. © 2018 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst November 13, 2017; DOI: 10.1158/1541-7786.MCR-17-0308

http://crossmark.crossref.org/dialog/?doi=10.1158/1541-7786.MCR-17-0308&domain=pdf&date_stamp=2018-1-20

http://mcr.aacrjournals.org/

DCIS promote breast cancer survival and invasion through CCL2-dependent mechanisms. Furthermore, increased CCL2/CCR2signaling in DCIS is associated with increased expression ofALDH1, a pro-invasive factor, and decreased expression ofHTRA2, a pro-apoptotic serine protease, factors associated withpoor prognosis of breast cancer patients (14, 15). These studiesidentify a key mechanism of DCIS progression involving CCL2/CCR2 signaling between fibroblasts and breast epithelial cells,with important clinical implications.

Materials and MethodsCell culture

Human fibroblasts were isolated from reduction mammo-plasty or DCIS tissues obtained from the Biospecimen CoreFacility at the University of Kansas Medical Center (KUMC), andimmortalized by expression of human teleromerase reverse tran-scriptase as described (16). Fibroblasts were authenticatedby expression of: platelet-derived growth factor receptor-a(PDGFR-a), fibroblast-specific protein 1 (Fsp1), and a-smoothmuscle actin (a-sma) and absence of pan-cytokeratin. DCIS.comcells originated from Dr. Fred Miller's laboratory (17). These celllines were cultured in DMEM containing 10% FBS (Atlas Biolo-gicals, catalog no. FR-0500-A), 2 mmol/L L-glutamine (Cellgro,catalog no. 25-005-CI), 100 IU/mL penicillin, and 100 mg/mLstreptomycin (Cellgro, catalog no. 10-080). SUM225 cells orig-inated from Dr. Steven Ethier's laboratory, Medical University ofSouth Carolina, Charleston, SC (18), and were cultured in HamF12 media containing 10% FBS, 5 mg/mL insulin, 1 mg/mLcortisone, and antibiotics. Cells were passaged no longer than6 months, and tested for mycoplasma after thawing using theMycoAlert Plus Kit (Lonza, catalog no. LT07-701).

Lentiviral transductionFor CCR2 overexpression, full-length CCR2 cDNA was

obtained from University of Missouri-Rolla cDNA Resource Cen-ter (clone ID no. CCR200000), and subcloned into pHAGE–CMV-MCS-IRES-zsgreen lentiviral plasmid (PLASMID, HarvardUniversity) using NHEI and XbaI restriction sites. pHAGE emptyvector was used as a vehicle control. CCR2 and nonsilencingcontrol shRNAs in pGFP-c-shlenti lentivirus vectors were pur-chased from OriGene (Catalog No. TL321181). The CCR2 target-ing sequence was: 50-TATTGTCATTCTCCTGAACACCTTCCAGG-30. CCL2 and nonsilencing control shRNAs in pGFP-c-shlentilentivirus vectors were obtained from OriGene (catalogno. TL316716). The CCL2 targeting sequence (OriGene) was:50-ACTTCACCAATAGGAAGATCTCAGTGCAG-30. CCL2 andnonsilencing control shRNAs in GIPZ shRNA lentivirus vectorswere obtained from Dharmacon (catalog no. V2LHS31298). TheCCL2 targeting sequence was: 50-TAAGTTAGCTGCAGATTCT-30.

To generate lentivirus, 3.33 mg of PMD2G (Addgene catalogno.12260), 6.66 mg PDPAX2 (Addgene catalog no. 12259), and10 mg target vectors were cotransfected in HEK-293T cells usingLipofectamine 2000 (Thermo Fisher Scientific catalog no.11668027). Medium was removed 48 hours later and used totransduce cells, which were sorted for GFP expression by FACS.

Gene deletion by clustered regularly interspaced shortpalindromic repeats

The CCR2 guide RNA was cloned into the pSpCAs9(BB)-2A-GFP(PX458) vector (Addgene catalog no. 48318) using BsmBI

enzyme. The CCR2 guide RNA sequence was: 50-TTCA-CAGGGCTGTATCACATCGG-30, which targeted the exon encod-ing the extracellular loop between the second and third trans-membrane domains of human CCR2. The vector was transfectedintoDCIS.combreast cancer cells using jetPei transfection reagent(Polyplus, catalog no. 101-01), with N/P 7.5. Forty-eight hourslater, GFP-positive cells were FACS sorted, cultured as single-cell clones in 96-well plates, and expanded into 6-well plates.Genomic DNA of individual colonies was screened by PCRto detect mutant colonies. The detection primer pair spanningthe CCR2 targeting site was: 50-ACATGCTGGTCGTCCTCATC,30-AAACCAGCCGAGACTTCCTG. The PCR product of the wild-type (WT) gene was 901 bp, and contained one DdeI enzymedigestion site to yield 231- and 670-bp fragments. Exon excisionintroduced an additional DdeI restriction site resulting in frag-ment sizes of 181, 231, and 468 bp upon DdeI digestion.

ELISAA total of 40,000 cells/well were seeded in 24-well plates in

DMEM/10% FBS for 24 hours, washed in PBS, and incubatedin serum-free DMEM for 24 hours in 500 mL/well. Conditionedmedia were assayed for human CCL2 by ELISA (PeproTech,catalog no. 900-M31). Reactions were catalyzed using tetra-methylbenzidine substrate (catalog no. 34028, Pierce). Absor-bance was read at OD450 nmol/L using a BioTek MicroplateReader.

MIND modelNOD-SCID IL receptor-g2 null femalemice 8 to 10 weeks of age

were purchased from The Jackson Laboratory. MIND injectionswere performed as described (19). Briefly, 4,000 cells/mL breastepithelial cells were prepared in 50 mL PBS containing 0.1%Trypan blue. A Y incision was made on the abdomen of miceanesthetized with ketamine/xylazine [100mg/kg (K)þ 10mg/kg(X)] to expose the 4–5 and 9–10 inguinal glands. The inguinalnipples were snipped. A 30-gauge Hamilton syringe with a blunt-ended 0.5-inch needle was used to deliver 5 mL (20,000) cells/nipple. Skin flaps were closed with wound clips. Mice werepalpated for lesions twice weekly. SUM225 injected mice weresacrificed 7 weeks postinjection. DCIS.com-injected mice weresacrificed 4 weeks postinjection.

Subrenal graftTransplantation into subrenal capsules of NOD-SCID female

mice (6–8 weeks old) was performed as described (20). Briefly,250,000 fibroblasts were resuspended with 100,000 DCIS.comcells in 50 mL rat tail collagen I (BD Pharmingen), and cultured inDMEM/10% FBS for 24 hours. Mice were anesthetized by keta-mine/xylazine, a 1 to 1.5 cm midline incision was made in theback 3 cm from the base of the tail, and the lateral or contralateralkidney was exposed. A small incision was made in the capsulelayer using forceps and small spring-loaded scissors. The graft wasinserted using a glass pipette. The body wall was closed with gutabsorbable sutures and the skin was closed with wound clips.Mice were monitored twice weekly and sacrificed 3 weeksposttransplantation.

DAB immunostainingTissues were fixed in 10% neutral formalin buffer and embed-

ded in wax as described (21). For DAB immunostaining, 5 mm

CCL2/CCR2 Expression in DCIS Progression

www.aacrjournals.org Mol Cancer Res; 16(2) February 2018 297




sections were dewaxed and heated in 10 mmol/L sodium citratebuffer pH 6.0 for 2 minutes. Endogenous peroxidases werequenched in PBS/60% methanol/3% H2O2, blocked in PBS/3% FBS, and incubated with primary antibodies (1:100) over-night at 4�C: collagen IV (Novus Biologicals NB120-6586SS),cleaved caspase-3 Asp175 (Cell Signaling Technology, catalog no.9579), Von Willebrand Factor 8 (VWF8; Millipore, catalog no.Ab7356), PDGFR-a (Cell Signaling Technology, catalog no.5241), KI67 (Santa Cruz Biotechnology, catalog no. 1307),HTRA2 (Cell Signaling Technology, catalog no. 2176), CCL2(Santa Cruz Biotechnology, catalog no. 1304), or F4/80 (Abcam,catalog no. ab6640). Fsp1 antibodies (Abcam, catalog no. 27427)were diluted 1:3. Slides were incubated for 2 hours at 1:1,000with: anti-rabbit-biotinylated (Vector Laboratories, catalog no.BA-5000), anti-goat biotinylated (Vector Laboratories, catalog no.BA-5000), or anti-rat-biotinylated (catalog no. BA-9401, VectorLaboratories). For laminin staining, slides were treated with20 mg/mL Proteinase K for 1 hour at 37�C prior to incubationwith 1:100 pan-specific antibodies (Novus Biologicals, catalogno. NB300-144AF700). Slides were incubated with streptavidinperoxidase (Vector Laboratories, catalog no. PK-4000), developedwith 3,3'-diaminobenzidine (DAB) substrate (Dako, catalog no.K346711), counterstainedwithMayer hematoxylin andmountedwithCytoseal. ProliferatingCellNuclear Antigen (PCNA) (catalogno. sc25280, Santa Cruz Biotechnology) and ALDH1A1 (R & DSystems, catalog no. MAB5869) proteins were detected usingthe Mouse on Mouse (MOM) Kit (Vector Laboratories, catalogno. BMK-2202).

ImmunofluorescenceFor CK/a-sma costaining, slides were heated in 10 mmol/L

sodium citrate buffer pH 6.0 for 2minutes. Slides were incubatedwith antibodies 1:100 overnight at 4�C to: a-sma (Spring Bios-ciences, catalog no. SP171) and CK5 (Thermo Fisher Scientific,catalog no. MA5-12596) or CK19 (Thermo Fisher Scientific,catalog no. MS198). Slides were incubated for 2 hours at 1:200with anti-rabbit-IgG-Alexa Fluor 568 (Thermo Fisher Scientific,catalog no. A10042) and anti-mouse IgG-Alexa Fluor 488(Thermo Fisher Scientific, catalog no. A-11001). For pan-cytoker-atin/phalloidin costaining, slides were heated in 10 mmol/Lsodium citrate pH 6.8 for 5 minutes. Slides were incubated with1:100 Alexa Fluor 488-phalloidin (Thermo Fisher, catalog no.A12379) and anti-pan-cytokeratin (Santa Cruz Biotechnology,catalog no. 8018) overnight at 4�C, and incubatedwith secondaryanti-mouse-Alexa Fluor 647 (Thermo Fisher Scientific, catalogno. 31571) using the MOM Kit. Sections were counterstainedwith 40,6-diamidino-2-phenylindole, dihydrochloride (DAPI)and mounted with PBS/glycerol.

Image quantificationFive random fields/sectionwere captured at 10�magnification

using the FL-Auto EVOS System (Invitrogen). DAB staining wasquantified as described previously (10). Briefly, images wereimported intoAdobe Photoshop,DAB stainingwas selected usingthe Magic Wand tool, copied, and saved a separate file. Imageswere opened in ImageJ (NIH, Bethesda, MD), and converted tograyscale. Background pixels were removed by threshold adjust-ment. Images were subject to particle analysis. Positive DABvalues were normalized to total area values, expressed as arbitraryunits. To quantify stromal staining, epithelial tissues werecropped out in Adobe Photoshop. DAB staining was selected in

stroma, copied to a new window and saved as a separate file.Images were opened in Image J and quantified. Stromal DABvalues were normalized to total stromal values.

Scoring of tumor invasionTissues were sectioned at three depths approximately 50 mm

apart. Two to three serial sections per depth were stained. Imageswere captured at 4� and 10� magnification and scored in ablinded fashion: 1 (non-invasive), 2 (lowly invasive), or 3 (highlyinvasive). For CK/a-sma CO-IF, 1 indicated no invasion, withintacta-smaþmyoepitheliumand confinement of epithelial cellswithin the duct; 2 indicated 50% or less disappearance of thea-sma surrounding the duct and/or appearance of three or fewercells invading through the duct; 3 indicatedmore than50%a-smadisappearance, with appearance of more than three cells invadedthrough the duct and making contact with the periductal stroma.For collagen IV and laminin immunostaining: 1 indicated welldefined expression in the basement membrane; 2 indicated addi-tional low-level expression in lesion; 3 indicated higher expres-sion in epithelium, with poor definition between epithelium andstroma. For phalloidin/pan-cytokeratin staining, 1 indicated awell-defined border between tumor and kidney, with a few tumorcells invaded into kidney tissue; 2 indicated some tumor cellinvasion, characterized by viable tumor cells present in kidneytissue; the border between kidney and tumor tissue was lessdefined; and 3 indicated high invasion characterized by extensivenumber of tumor cells in kidney tissue; tumor was embedded inkidney, and the border between kidney tissue and tumor wereundefined.

Flow cytometryFlow cytometry staining for CCR2 expression was conducted as

described (13). Briefly, adherent cells were detached from plasticby Accutase (Thermo Fisher Scientific, catalog no. A1110501),washed in PBS, and incubated with anti-CCR2-PE for 1 hour onice. Samples were washed in PBS three times and analyzed on aLSRII Flow cytometer, normalized to unstained controls.

Fibroblast proliferation assayFibroblasts were seeded 30,000/well in 24-well plates over-

night. DCIS.com cells (500,000) were seeded in 10-cm dishes,and incubated with 5-mL serum-free DMEM for 24 hours. Fibro-blasts were treated with 500 mL of DMEM or tumor conditionedmedium for 24 or 48 hours. Fibroblasts were detached throughtrypsinization, quenched in DMEM/10% FBS and pelleted bymicrocentrifugation. Fibroblasts were resuspended in 50-mL PBSand counted by hemocytometer.

Statistical analysisCell culture experiments were repeated a minimum of three

times. Data are expressed as mean� SEM. Statistical analysis wasdetermined using two-tailed t test or ANOVAwith Bonferroni posthoc comparisons for normal distributions and Kruskall–Wallistest with Dunn post hoc comparison for non-Gaussian distribu-tions. Statistical analysis was performed using GraphPadSoftware. Significance was determined by P < 0.05 (�, P < 0.05;��, P < 0.01; ��, P < 0.0001; n.s., not significant or P > 0.05).

Ethics approval and consent to participateAll animal experiments were performed at KUMC according to

guidelines from the Association for Assessment and Accreditation

Brummer et al.

Mol Cancer Res; 16(2) February 2018 Molecular Cancer Research298




of Laboratory Animal Care. Experiments were approved by theInstitutional Animal Care and Use Committee. Patient sampleswere collected under approval by Institutional Review Board(IRB) at KUMC. All sampleswere deidentified by the BiospecimenCore, an IRB-approved facility, prior to distribution.

ResultsCCR2 overexpression in SUM225 cells enhances DCISprogression

InDCIS, cancer cells grow, but remainwithin the boundaries ofducts and lobules, which are lined by a-smaþ myoepithelial cellsand basement membrane, structural barriers between the stromaand duct. Progression from DCIS to IDC is characterized bydisappearance of themyoepithelium and appearance of invadingductal carcinoma cells into the surrounding stroma (22). Toclarify the role of epithelial CCR2 expression inDCIS progression,we utilized MIND models established through injection ofSUM225 andDCIS.combreast cancer cells. SUM225breast cancercells are lowly invasive, of a luminal/Her2þ subtype (23). DCIS.com breast cancer cells, a basal-like subtype, are more highly-invasive (19). By flow cytometry, CCR2 expression was signifi-cantly lower in SUM225 cells compared with DCIS.com breastcancer cells (Fig. 1A). We first examined the effects of CCR2overexpression on progression of SUM225 lesions. By lentivirustransduction, two different SUM225 cell lines were generated tooverexpress CCR2 (CCR2-L and CCR2-H), and compared withSUM225 cells expressing pHAGE vehicle control (Fig. 1A). Thesecells were MIND injected into NOD-SCID mice, and examined7 weeks postinjection, when lesions were palpable. CCR2-overexpressing xenografts showed no significant changes inmam-mary tissue mass compared to pHAGE control (Fig. 1B).

Extent of epithelial invasion in the mammary gland or breasttissue has been determined by evaluating myoepithelial integritythrough a-sma expression, and examining for presence of carci-noma cells contacting the surrounding stroma (19, 24–26). Toevaluate the effects of CCR2 overexpression on ductal invasion,we costained for a-sma to define ductal myoepithelium, and forhuman specific CK19 to define SUM225 cells. Lesions werescored for invasiveness. Noninvasive lesions had intact a-smaþmyoepithelium, lowly-invasive lesions showed reduced a-smaexpression, lining the breast duct, and a few invasive cancer cells.Highly-invasive lesions showed minimal a-sma expression andmultiple invasive cancer cells. In the pHAGE controls, 21% werenoninvasive, 51% were lowly invasive and 28% were highlyinvasive. Of CCR2-L MIND lesions, 12% were noninvasive,57% were lowly invasive, and 31% were highly invasive. OfCCR2-H MIND lesions, 8% were noninvasive, 66% were lowlyinvasive, and 26% were highly invasive (Fig. 1C). The decrease inlowly invasive lesions and increase in lowly invasive lesionsindicate a shift towards invasion.

To further characterize invasion through basement membrane,mammary tissues were stained for laminin and collagen IV, whichare basement membrane proteins associated with invasiveness inbreast cancer (27, 28). In noninvasive lesions, laminin andcollagen IV were expressed in the basement membrane andsurrounding stroma. In lowly invasive lesions, laminin and col-lagen IV were also detected in the epithelium, correlating with afew invading epithelial cells. In highly-invasive lesions, lamininand collagen IV expression in lesions resulted in poorlydefined borders between stroma and epithelium. Consistent with

a-sma/CK19 CO-IF, laminin and collagen IV staining revealedthat CCR2 overexpression in SUM225 cells decreased thenumber of lowly-invasive lesions and increased the number ofhighly-invasive lesions (Supplementary Fig. S1A and S1B). CCR2overexpression was also associated with increased cell prolifera-tion and decreased apoptosis as indicated by Ki67 and cleavedcaspase-3 staining (Fig. 1D and E). These data indicate that CCR2overexpression in SUM225 cells enhances the progression ofMIND xenografts.

Knockdown or knockout of CCR2 in DCIS.com cells inhibitsDCIS progression

We next examined the effects of CCR2 deficiency on progres-sion of DCIS.com lesions. Of the four shRNA sequences tested,one induced significant knockdown of CCR2 expression in DCIS.com cells (Fig. 2A). MIND injection of DCIS.com cells in NOD-SCID mice resulted in palpable mammary lesions at 4 weeks.CCR2 knockdown (CCR2-KD) decreased mammary tissue tumorgrowth compared with control shRNA expressing xenografts(Fig. 2B). To examine for changes in ductal invasion, sectionswere CO-IF stained for a-sma and human-specific CK5 to identifyDCIS.com cells, and scored. Although the percentage of lowlyinvasive lesions (75%) was higher in the CCR2-KD group com-pared with control (68%), the percentage of highly invasivelesions dropped from 20% in the control shRNA group to 6%in the CCR2-KD group. The percentage of noninvasive lesionsincreased from 12% in controls to 19% in the CCR2-KD group.These data indicated a shift from high to less invasive lesions withCCR2-KD (Fig. 2C). This trend was also observed in analysis ofcollagen IV and laminin expression in DCIS.com MIND lesions(Supplementary Fig. S2A and S2B). CCR2 KDwas also associatedwith decreased tumor cell proliferation and increased apoptosis(Fig. 2D and E).

To validate the effects of CCR2-KD on DCIS.com progression,the CCR2 gene was knocked out by CRISPR. Two WT clones andone homozygous knockout clone (CCR2-KO) were identifiedfrom 70 clones (Supplementary Fig. S3A). By flow cytometry,WT clones showed similar CCR2 expression levels to parentalcells, whereas CCR2-KO cells showed a significant reduction inCCR2 expression (Supplementary Fig. S3B). MIND injection ofCCR2-KO cells resulted in a significant reduction in mammarytissue mass and fewer invasive lesions (Supplementary Fig. S3Cand S3D), compared withWT xenografts. These data indicate thatCCR2-KD or -KO inhibits the progression from DCIS to IDC inDCIS.com MIND xenografts.

Increased angiogenesis, fibrosis, and macrophage recruitmentare associated with invasive breast cancer (22). To determine howepithelial CCR2 expression affected the surrounding mammarystroma, immunostaining was performed to analyze expression ofbiomarkers for macrophages (F4/80) and endothelial cells(VWF8). To account for fibroblast heterogeneity, we immunos-tained for two different markers: Fsp1 and PDGFR-a (29, 30).DAB expression of stromal biomarkers was quantified by pixeldensity analysis and normalized to total stromal area, using anImageJ protocol described previously (10). There were no signif-icant changes in VWF8 or F4/80 expression with CCR2 over-expression or knockdown (Supplementary Fig. S4A and S2B).Fsp1 and PDGR-a were expressed in fibroblastic stroma and inepithelial cells, consistent with studies showing mesenchymalmarker expression in breast cancer cells (10, 31). CCR2-overexpressing SUM225 xenografts showed increased stromal






expression of Fsp1 and PDGFR-a (Fig. 3A and B), associated withstromal CCL2 expression (Fig. 3C). Conversely, CCR2-deficientDCIS.com MIND xenografts showed a significant decrease infibroblastic cells and decreased CCL2 expression in the stroma(Fig. 4A and B). These data indicate that CCR2 overexpressionor knockdown is associated with changes in CCL2 expressingfibroblasts in the DCIS stroma.

CCL2 from DCIS fibroblasts is important for CCR2-mediatedbreast cancer survival and invasion

To further characterize the expression of CCL2 in DCIS stroma,fibroblasts were isolated frompatient samples of normal breast orDCIS tissues and analyzed for CCL2 expression. By ELISA, two of

three DCIS fibroblast lines (1213-249, 80H) expressed higherlevels of CCL2 compared with normal fibroblasts (hNAF2525,hNAF8727) and DCIS.com cells (Fig. 5A). Although previousstudies have established an important role for carcinoma-associ-ated fibroblasts from invasive breast cancers (32, 33), the role offibroblasts derived from DCIS tissues remain poorly understood.To determine the functional role of CCL2 derived from DCISfibroblasts in CCR2-mediated breast cancer progression, we uti-lized the subrenal capsule model. The advantage of this model isthat is devoid of fibroblasts, enabling us to determine the relativecontribution of cografted fibroblasts without interference fromhost stroma. Mammary carcinoma cells grafted in the subrenalcapsule form tumors similarly to orthotopic injection (34, 35).

A BC

ount

CCR2-PE Intensity

pHAGE

Hig

hLo

wN

onin

vasi

ve

No.

of l

esio

ns s

core

d

P < 0.001180160140120100806040200

pHAGE

pHAGE

Noninvasive Low High

CCR2-L

CCR2-LpHAGE

pHAGE

CCR2-L

CCR2-HpHAGE

CCR2-L

CCR2-H

CCR2-H 0.2 0.1

0.08

0.06

0.04

0.02

0.15

0.1

0.05

0

Ki6

7 ex

pres

sion

Cle

aved

cas

pase

-3ex

pres

sion

CCR2-H

CCR2-L

CCR2-H

CCR2-L CCR2-H Secondary controls

CCR2-LCCR2-H

800.5

n.s.0.4

0.3

0.2

0.1

0

60

40

20

DCIS.comSuM225

pHAGE

pHAGECCR2-L

CCR2-LCCR2-H

CCR2-H

% C

CR

2-po

sitiv

e ce

lls

Mam

mar

y tis

sue

mas

s (g

)

0

UnstainedpHAGE

101 102 103 104 105

C

D E

Figure 1.

CCR2 overexpression in SUM225breast cancer cells enhances invasiveprogression. A, Flow cytometryanalysis for CCR2 expression inparental DCIS.com or SUM225parental cells or SUM225 cellsexpressing vehicle pHAGE control orCCR2 (CCR2-L, CCR2-H). Histogramanalysis shown on left. Graph showspercentages of positive cells. B, Tissuemass of SUM225 MIND injectedmammary glands C, SUM225 lesionswere costained for CK-19 (red) anda-sma (green), and scored for thenumber of invasive lesions n ¼ 252lesions for control pHAGE, 272 lesionsfor CCR2-L, and 231 for CCR2-H group.Representative images are shownwithsecondary antibody control panel ofanti-rabbit-Alexa Fluor 488/anti-mouse-Alexa Fluor 568/DAPI overlay.Arrows indicate invasive foci. D andE, ImageJ quantification ofimmunostaining for Ki67 (D) orcleaved caspase-3 (E) in SUM225lesions (arbitrary units). Arrows pointto examples of positive staining.Statistical analysis was performedusing one-way ANOVA withBonferonni post hoc comparison(B, D, E) or c2 test (C). Statisticalsignificance was determined byP < 0.05. � , P <0.05; ns, not significant.Mean � SEM values are shown. Scalebar ¼ 200 mm.

Brummer et al.





1213-249 DCIS-derived fibroblasts, which showed the highestlevel of CCL2, were cografted with DCIS.com breast cancercells in the renal capsule of NOD-SCID mice, and analyzed forchanges in tumor growth and invasion. Fibroblasts cograftedwithparental DCIS.com cells showed increased tumormass comparedwithDCIS.com cells grafted alone (Fig. 5B). CCR2-deficientDCIS.com cells cografted with 1213-249 fibroblasts showed a signifi-cant 20% decrease in tumor mass compared with fibroblastscografted with control DCIS.com cells (Fig. 5B). To examine forchanges in tumor invasion into normal kidney tissue, we per-formed CO-IF staining for pan-cytokeratin and phalloidin todistinguish tumor cells from kidney tissues. In the subrenalcapsule model, pan-cytokeratin antibodies stained DCIS.comtumors more clearly than CK5 antibodies used in the MIND

model. Pan-cytokeratin antibodies recognized CK: 4, 5, 6, 8,10, 13, and 18, and preferentially stained breast cancer cellsover kidney tissues, which expressed fewer of the cytokeratins(36). Using this approach, tumor invasion was characterized by alack of defined border between tumor and kidney tissues, andscattering of tumor cells throughout the kidney, as observed incontrol lesions (Fig. 5C). CCR2-deficient cells cografted withfibroblasts showed a reduction in tumor invasion, characterizedbymore cohesive tumors and a clearer delineation between tumorand kidney tissues (Fig. 5C). Decreased invasiveness was associ-ated with decreased tumor cell proliferation and increased apo-ptosis as indicated by PCNA and cleaved caspase-3 immunostain-ing (Fig. 5D and E). CCR2-KD DCIS.com cells grafted alone didnot show significant differences in tumor growth or invasion

A

Par

Noninvasive Low High200

Con CCR2-KD

150

100

50

0

CC

R2-

KD

Con

P < 0.001

Con CCR2-KD

% C

CR

2

Mam

mar

y tis

sue

mas

s (g

)

60

0.8

0.6

0.4

0.2

Con CCR2-KD

No.

of l

esio

ns s

core

d

Noninvasive

0.50.8

0.6

0.4

0.2

Con CCR2-KD0

0.4

0.3

0.2

Ki6

7 E

xpre

ssio

n

Cle

aved

cas

pase

-3 e

xpre

ssio

n

0.1

0

Con CCR2-KD Con CCR2-KD

Low High

0

50

40

30

20

10

0

C

D E

B

Figure 2.

shRNA-mediated CCR2-KD in DCIS.com breast cancer cells inhibitsinvasive progression. A, Flowcytometry analysis for CCR2expression in parental (Par) or DCIS.com cells expressing control (Con) orCCR2 shRNA (CCR2-KD). B, Tissuemass of DCIS.com MIND–injectedmammary glands C, DCIS.com MINDlesions were costained for CK5 (red)and a-sma (green) andcounterstained with DAPI (blue).Representative images are shownwitharrows pointing to invading tumorcells. Lesions were scored forinvasiveness. n ¼ 152 lesions forcontrol shRNA group, 193 lesions forCCR2-KD group. n ¼ 8 mice/group.D and E, ImageJ quantification of Ki67(D) or cleaved caspase-3 (E)immunostaining in DCIS.com lesions.Arbitrary units are shown. Arrowspoint to examples of positive staining.Statistical analysis was performedusing one-way ANOVA withBonferonni post hoc comparison(B, D, E) or c2 test (C). Statisticalsignificance was determined byP < 0.05. � , P < 0.05; �� , P < 0.01.Scale bar ¼ 200 mm.






comparedwith control shRNA cells grafted alone (SupplementaryFig. S5A–S5C). These studies indicate that CCR2-KD inDCIS.combreast cancer cells inhibits fibroblast-mediated tumor growthand invasion.

To determine the relevance of CCL2 expression in DCIS fibro-blasts, fibroblasts were immortalized by hTERT expression toenable stable shRNA expression. Two CCL2-deficient fibroblastlines were generated from two different shRNA systems. A 47%decrease in CCL2 expression was observed using the GFP-c-shLenti OriGene system (CCL2-pLenti). A 30% decrease inCCL2 expressionusing theGIPZDharmacon system(CCL2/GIPZ;Fig. 6A). CCL2-deficient or control fibroblasts were cograftedwithDCIS.combreast cancer cells in the subrenal capsule and analyzedfor changes in tumor progression. CCL2-deficient fibroblastscografted with DCIS.com cells resulted in smaller tumors andreduced tumor invasion, associated with decreased tumor cellproliferation and increased apoptosis (Figs. 6B-C). These studiesindicate that CCL2 derived from DCIS fibroblasts enhances pro-gression of DCIS.com breast tumors.

CCL2/CCR2–mediated invasion is associated with increasedALDH1A1 and decreased HTRA2 expression

Finally, we analyzed the relationship between expression ofdownstream CCL2/CCR2 signaling proteins and increased breastcancer survival and invasion. Through candidate screening offactors related to breast cancer survival and invasion, we foundthat CCL2 treatment of DCIS.com breast cancer cells over timeincreased expression of ALDH1A1, a stem cell and proinvasivefactor (14), and reduced expression of HTRA2, a proapoptoticmitochondrial serine protease (ref. 15; Supplementary Fig. S6Aand S6B). CCR2-KD in DCIS.com breast cancer cells decreasedexpression of ALDH1A1 and increased HTRA2 in MIND xeno-grafts by IHC staining (Fig. 7A and B). Conversely, CCR2 over-expression in Sum255 MIND xenografts enhanced ALDH1A1expression and decreased HTRA2 expression (SupplementaryFig. S7A and S7B), indicating that epithelial CCR2 can regulateALDH1A1 and HTRA2 expression. Furthermore, CCL2-KD infibroblasts increased HTRA2 expression and decreased ALDH1expression in DCIS.com cells in the subrenal capsule model

A pHAGE CCR2-H

Magnified image

Magnified image

Magnified image

CCR2-L Anti-rabbit

0.6

0.04

0.03

0.02

0.2

0.15

0.1

0.05

0.0

0.01

0

0.4

0.2

0

Fsp

1 E

xpre

ssio

nP

DG

FR

-α E

xpre

ssio

nC

CL2

Exp

ress

ion

CC

R2-

H

CC

R2-

L

pHA

GE

CC

R2-

H

CC

R2-

L

pHA

GE

CC

R2-

H

CC

R2-

L

pHA

GE

pHAGE CCR2-H CCR2-L Anti-rabbit

B

C

Figure 3.

CCR2 overexpression in SUM225MINDxenografts increases the levels ofCCL2-expressing fibroblasts Sum 225MIND lesions were immunostained forfibroblast-specific protein 1 (Fsp1; A)and progesterone growth factorreceptor-a (PDGFR-a; B); or CCL2expression (C). Representativeimages are shown with magnifiedimage underneath. The stroma ismarked with an asterisk in themagnified image. Expression in thestroma was quantified by ImageJ, inarbitrary units. Statistical analysis wasperformed using one-way ANOVAwith Bonferonni post hoc comparison.Statistical significance wasdetermined by P < 0.05. � , P < 0.05;�� , P < 0.001. Mean � SEM values areshown. Scale bar ¼ 400 mm.

Brummer et al.





(Fig. 7C andD), indicating that paracrineCCL2 signaling from thefibroblastic stroma was important for regulating ALDH1A1 andHTRA2 expression. Through Kaplan–Meier Plotter analysis (37),increased CCR2 and ALDH1A1 and decreased HTRA2 expressionwere significantly associated with decreased metastasis-freesurvival of breast cancer patients (Fig. 7E). These data demonstratea clinical relevance for CCL2/CCR2 signaling proteins in breastcancer.

DiscussionThe role of fibroblasts in DCIS progression is poorly under-

stood. Fibroblasts derived from invasive breast ductal carcinomaspromote tumor growth, invasion, metastasis, and chemoresis-

tance (32, 33). One study showed that fibroblasts from normal,IDC, or arthritic tissues enhanced progression of MCF10A celllines in a subcutaneous injection model through transforminggrowth factor-b and Hedgehog-dependent mechanisms (38). Forthe first time, we show that fibroblasts derived from DCIS patientsamples accelerate progression from DCIS to IDC throughCCR2-dependentmechanisms.Moreover, CCL2/CCR2-mediatedbreast cancer progression is associated with increased expressionof clinical relevant proinvasive factors (ALDH1A1) and decreasedexpression of proapoptotic factors (HTRA2).

Here, we noted some complementary and conflicting pheno-types through CCR2 overexpression and knockdown. CCR2 over-expression in SUM225 cells enhanced formation of invasivelesions and increased the presence of CCL2þ fibroblasts,

A Con CCR2-KD

Magnified image

Magnified image

Magnified image

Anti-rabbit

Con CCR2-KD Anti-goat

B

C

0.4

0.3

0.1

0.2

0

0.4

0.2

0.15

0.1

0.05

0

0.5

0.3

0.1

0.2

0

Fsp

1 E

xpre

ssio

nP

DG

FR

-α E

xpre

ssio

nC

CL2

Exp

ress

ion

Con CCR2-KD

Con CCR2-KD

Figure 4.

CCR2 shRNA knockdown in DCIS.comMIND xenografts reduces thelevels of CCL2-expressing fibroblasts.DCIS.com MIND lesions wereimmunostained for fibroblast-specific protein 1 (Fsp1; A),progesterone growth factorreceptor-a (PDGFR-a; B), or CCL2expression (C). Representativeimages are shown with magnifiedimage underneath. The stroma ismarked with an asterisk in themagnified image. Expression in thestroma was quantified by ImageJ, inarbitrary units. Statistical analysis wasperformed using two-tailed T test(B). Statistical significance wasdetermined by P < 0.05. � , P < 0.05.Mean � SEM values are shown.Scale bar ¼ 400 mm.






associated with increased ALDH1 and decreased HTRA2. CCR2-KD and knockout in DCIS.com cells inhibited invasive progres-sion and decreased the presence of CCL2þ fibroblasts, associatedwith decreased ALDH1 and increased HTRA2 expression. How-ever, although CCR2-KD significantly affected mammary tumormass, CCR2 overexpression did not. Although CCR2 overexpres-sion increased tumor cell proliferation and survival of SUM225lesions, these levels were still lower than the cell proliferation andsurvival detected in DCIS.com MIND xenografts. CCR2 expres-sion levels in overexpressing cells did not reach the levels detectedin DCIS.com breast cancer cells. Therefore, it is possible that theincrease in cell proliferation and survival in CCR2-overexpressingcells was not sufficient to affect overallmammary tissuemass. Thelevels of CCR2 expression in DCIS.com cells are consistent withprevious studies showing that CCR2 expression levels are higherin basal-like breast cancer cells compared to luminal breast cancercells (13). Because SUM225 cells are luminal/Her2þ, additional

oncogenic pathways may be important to DCIS progression ofthis subtype. Regardless of subtype, by analyzing the effects ofCCR2 overexpression in SUM225 cells with CCR2-KD in DCIS.com cells, we demonstrate a critical role for epithelial CCR2receptor expression in DCIS progression.

Despite a 2-fold increase in the number of CCR2þ cells in theCCR2-H SUM225 cell line, CCR2-H cells did not show increasedinvasion, proliferation, or survival comparedwith CCR2-L cells. Itis possible a threshold of receptor expression modulates cellularactivity. Such a threshold has been detected in T cells whereby8,000 T-cell receptors/cell are needed for a commitment toproliferate (39, 40). A threshold also exists for EGFR levels inregulating Cbl- and Grb2-dependent signaling in epithelial cells(41). In our studies, CCR2-L cells may have reached a thresholdfor CCR2 expression in determining cellular invasion. Althoughmore cells expressed CCR2 in the CCR2-H cell line, the level ofexpression may not have been sufficient to commit these cells to

0 0

Con/Fbs CCR2-KD/Fbs

Con/Fbs

Con/Fbs

CCR2-KD/Fbs

CCR2-KD/FbsCon/Fbs

Anti-mouse Anti-rabbit

0 0

0.05

0.1

0.15

0.2

Cle

aved

cas

pase

-3

0.05

0.1

0.15

PC

NA

Exp

ress

ion

CCR2-KD/Fbs

Con/Fbs

CCR2-KD/Fbs

Con/Fbs

CCR2-KD/Fbs

0

2

4

6

8

10

12

14 P < 0.01 Noninvasive

High

Low

No.

of s

ectio

ns s

core

d

Par alone Par/Fbs Con/Fbs CCR2-KD/Fbs

0.05

0.1

0.15

0.2

0.25

0.3

DCIS.co

m

hNAF25

25

hNAF87

27

1213

-249 80

H

HPO7021

3

5,000

10,000

15,000

20,000

25,000

A

C

D E

BC

CL2

Con

cent

ratio

n (p

g/m

L)

Mea

n K

idne

y/tu

mor

mas

s (g

)

Figure 5.

CCR2-KD in DCIS.com cells inhibitsfibroblast mediated cancerprogression. A, CCL2 ELISA ofconditioned medium from fibroblastsderived from normal breast(hNAF2525, hNAF8727) or DCIStissues (1213–249, 80H, HPO70213),in comparison with DCIS.com breastcancer cells.B–E. 1213–249 fibroblasts(Fbs) were cografted with parental(Par) DCIS.com cells or DCIS.comcells expressing control (Con) orCCR2 shRNAs (CCR2-KD) in thesubrenal capsule of NOD-SCID micefor 21 days. Kidney tissues weremeasured for tumor mass (B), scoredfor tumor invasion into normal kidneyby pan-cytokeratin (red) andphalloidin (green) staining (C), tumorcell proliferation by PCNAimmunostaining (D), tumor cellapoptosis by cleaved caspase-3immunostaining (E). Scale bar ¼400 mm. Arrows point to examples ofpositive staining. Expression intissues was quantified by ImageJ.n ¼ 7 mice per group. Statisticalanalysis was performed using oneway ANOVA with Bonferonni posthoc comparison (B) or two tailedT test (D, E). Statistical significancewas determined by P < 0.05. � ,P < 0.05; �� , P < 0.001. Mean � SEMare shown.

Brummer et al.





invade. Histogram analysis revealed that although more cellsoverexpressed CCR2 in the CCR2-H cell line, expression levelsdid not vary highly between CCR2-L and CCR2-H cells. Inaddition to a receptor threshold, heterogeneity in expression ofintracellular signaling components in breast cancer (42)may alsoexplain why CCR2-H cells did not result in further DCIS progres-sion. As we are unable to control which SUM225 cells expressCCR2, it is possible some CCR2 overexpressing cells did notexhibit the necessary downstream signaling components toinduce invasion. As CCR2 overexpression in SUM225 cells didnot reach the levels of invasion detected in DCIS.com cells, it islikely that other oncogenic factors would be required to furtherenhance carcinoma invasion. Several oncogenic signaling path-ways including Notch, EGF, and HGF signaling are associatedwith DCIS progression (43, 44). It would be of interest to furtherunderstand how CCL2/CCR2 coordinates DCIS progression withother oncogenic factors.

We also observed that CCR2 overexpression and knockdownaffected the levels of fibroblasts in DCIS stroma. We expectedthat CCR2 signaling in breast cancers modulated fibroblastgrowth through expression of soluble growth factors such as

PDGF and WNT5A, positive regulators of fibroblast prolifer-ation (45, 46). However, cultured DCIS fibroblasts treatedwith conditioned medium from CCR2-deficient or controlDCIS.com control cells showed no significant changes in cellgrowth (Supplementary Fig. S8). Furthermore, there were nochanges in blood vessel density or macrophage recruitment,indicating that epithelial CCR2 would not regulate fibroblastaccumulation indirectly through these stromal cell types. It ispossible that epithelial CCR2 acts on other stromal compo-nents to indirectly modulate fibroblast growth, includingadipocytes or granulocytes. Another possibility may involvethe extracellular matrix. Hyaluronan and fibronectin increasefibroblast cell growth, whereas collagen suppresses fibroblastgrowth through mechano-signal transduction mechanisms(47, 48). These factors would be present in mammary tissues,but not in conditioned medium. Studies are currentlyunderway to understand how CCL2/CCR2 signalingbreast cancer cells modulate the surrounding breast tumormicroenvironment.

We show that increased ALDH1 and decreased HTRA2expression are associated with CCL2/CCR2-mediated DCIS

0 0 0

0.1

0.2

0.3

0.4

0.5

0.6

0.050.1

0.150.2

0.250.3

0.350.4

Tum

or m

ass

(g)

pLen

ti-Con

pLen

ti-CCL2

KD

pLen

ti-Con

pLen

ti-CCL2

KD

GIPZ-C

on

GIPZ-C

CL2KD

pLenti-Con pLenti-CCL2KDGIPZ-Con

P < 0.01P < 0.05

0

2

4

6

8

10

12

14

16

0

2

4

6

8

10

12

14

16

No.

of s

ectio

ns s

core

d

GIPZ-CCL2KD

pLenti-Con pLenti-CCL2KDGIPZ-Control

Noninvasive Low High

GIPZ-CCL2KD

GIPZ-C

ontro

l

GIPZ-C

CL2KD

500

1,000

1,500

2,000

2,500A

C

B

CC

L2 (

pg/m

L)

Figure 6.

CCL2 derived from DCIS fibroblasts isimportant for progression of DCIS.combreast cancer cells. A, CCL2 ELISA of1213–249 fibroblasts expressing controlshRNA (Con) or CCL2 shRNAs frompLenti or GIPZ lentivirus systems.B and C, Control or CCL2 deficient 1213–249 fibroblasts were co-grafted withDCIS.com breast cancer cells in thesubrenal capsule and analyzed forchanges in tumor growth (B), and scoredfor tumor invasion into normal kidneytissue by pan-cytokeratin (red) andphalloidin (green) staining (C). n ¼ 7mice/group. Scale bar ¼ 400 mm. K,kidney; T, tumor. Statistical analysis wasperformed using One way ANOVA withBonferonni post hoc comparison.Statistical significance was determinedby P < 0.05. �� , P < 0.01. Mean � SEMare shown.






progression. Previous studies have implicated ALDH1 expres-sion in cancer stem cell renewal, invasion, and drug resis-tance (49). Emerging studies indicate an important role forHTRA2 in positively regulating mitochondrial-dependent apo-ptosis (50). The increased expression of HTRA2 in CCR2-deficient lesions is consistent with the increased expressionof cleaved caspase-3, as an indicator of apoptosis. CCL2/CCR2signaling in breast cancer cells may promote DCIS progressionby enhancing ALDH1þ tumor initiating cells, or activatinginvasive pathways through ALDH1 activity in breast cancercells. CCL2/CCR2 signaling may facilitate survival of DCISlesions through suppression of HTRA2-mediated apoptosispathways.

In summary, these studies identify a novel role for CCL2/CCR2signaling in cancer progression, identify potentially new prog-nostic factors for DCIS, and potentially newmolecular targets forthe prevention of invasive breast cancer.

Disclosure of Potential Conflicts of InterestN. Alvarez is a President at DeNovo Genomics and has ownership interest

(including patents) in DeNovo Genomics. No potential conflicts of interestwere disclosed by the other authors.

Authors' ContributionsConception and design: G. Brummer, F. Behbod, N. ChengDevelopment of methodology: G. Brummer, Q. Hu, M. Yao, N. Alvarez,P. Fields, N. Cheng

0

0

0

0.025

0.05

0.075

0.1

0

Low CCR2High CCR2

0.00 50

Time (months)

Low

Number at risk Number at risk Number at risk

High00

11

1314

7986

286304

584573

815794

00

20

189

7293

269321

545612

806803

00

20

1611

9075

306284

554603

805804

Time (months) Time (months)100 150 200 250 300 0 50 100 150 200 250 300 0 50 100 150 200 250 300

0.2

0.4

0.6

0.8

1.0

0.0

0.2

0.4

0.6

0.8

1.0

0.0

HR = 0.8 (0.66–0.98)

Logrank P = 0.033

HR = 0.65 (0.53–0.8)Logrank P = 2.9e–05

HR = 0.79 (0.65–0.97)Logrank P = 0.024

0.2

0.4

0.6

0.8

1.0

Pro

babi

lity

DM

FS

Low ALDH1A1High ALDH1A1

Low HTRA2High HTRA2

0.025

0.05

0.075

0.1

0.05

0.1

0.15

Con CCR2-KD

Con CCR2-KD

Con CCR2-KD

Anti-mouse

Anti-rabbit

GIPZ-Con GIPZ-CCR2-KD Anti-mouse

GIPZ-Con GIPZ-CCR2-KD Anti-rabbit

Con CCR2-KD

GIPZ-Con

GIPZ-CCR2-KD

GIPZ-Con

GIPZ-CCR2-KD

0.1

0.2

0.3A

B

C

D

E

ALD

H1

Exp

ress

ion

ALD

H 1

Exp

ress

ion

HT

RA

2 E

xpre

ssio

nH

TR

A2

Exp

ress

ion

Figure 7.

CCL2/CCR2-mediated DCIS progressionis associated with increased ALDH1and decreased HTRA2 expression.A–D, ALDH1 and HTRA2 expression wasexamined by immunostaining of tumortissues in the DCIS.com MIND Model(A and B) and subrenal capsule model(C and D). Expression in tissues wasquantified by ImageJ. K, kidney tissue;T, tumor. Scale bar ¼ 400 mm. E, RNAexpression of CCR2 (Affyid 207794_at),ALDH1A1 (Affyid 212224_at), and HTRA2(Afftyid 2030809_s_at) were analyzedfor associations with distant metastasis–free survival (DMFS) through Kaplan–Meier plotter. Statistical analysis wasperformed using two tailed T test(A and B) or log-rank test (C). Statisticalsignificancewas determined by P <0.05.� , P < 0.05; �� , P < 0.01. Mean � SEM areshown.

Brummer et al.





Acquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): G. Brummer, D.S. Acevedo, Q. Hu, M. Portsche,W.B. Fang, M. Yao, M. Myers, Y. Hong, N. ChengAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): G. Brummer, D.S. Acevedo, Q. Hu, M. Portsche,W.B. Fang, M. Myers, Y. Hong, N. ChengWriting, review, and/or revision of the manuscript: G. Brummer,D.S. Acevedo, Q. Hu, M. Portsche, W.B. Fang, M. Yao, M. Myers, N. ChengAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): G. Brummer, D.S. Acevedo, M. Portsche,W.B. Fang, M. Yao, B. Zinda, P. Fields, N. ChengStudy supervision: G. Brummer, D.S. Acevedo, N. Cheng

AcknowledgmentsThis study was supported by the Susan G. Komen Foundation

(CCR13261859; to N. Cheng) and National Cancer Institute (R01CA172764;to F. Behbod and N. Cheng).

The costs of publication of this articlewere defrayed inpart 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 June 13, 2017; revised September 20, 2017; accepted November 2,2017; published OnlineFirst November 13, 2017.

References1. Pang JM, Gorringe KL, Fox SB. Ductal carcinoma in situ—update on risk

assessment and management. Histopathology 2016;68:96–109.2. Park TS, Hwang ES. Current trends in themanagement of ductal carcinoma

in situ. Oncology 2016;30:823–31.3. Virnig BA, Tuttle TM, Shamliyan T, Kane RL. Ductal carcinoma in situ of the

breast: a systematic review of incidence, treatment, and outcomes. J NatlCancer Inst 2010;102:170–8.

4. Khan S, Epstein M, Lagios MD, Silverstein MJ. Are we overtreating ductalcarcinoma insitu (DCIS)? Ann Surg Oncol 2016;24:59–63.

5. Goldstein SN. Controversies in pathology in early-stage breast cancer.Semin Rad Oncol 2011;21:20–5.

6. Kerlikowske K, Molinaro AM, Gauthier ML, Berman HK, Waldman F,Bennington J, et al. Biomarker expression and risk of subsequent tumorsafter initial ductal carcinoma in situ diagnosis. J Natl Cancer Inst 2010;102:627–37.

7. Palomino DC, Marti LC. Chemokines and immunity. Einstein 2015;13:469–73.

8. Balkwill FR. The chemokine system and cancer. J Pathol 2012;226:148–57.9. Ansari AW, Kamarulzaman A, Schmidt RE.Multifaceted impact of host C-C

Chemokine CCL2 in the immuno-pathogenesis of HIV-1/M. tuberculosisco-infection. Front Immunol 2013;4:312.

10. Yao M, Yu E, Staggs V, Fan F, Cheng N. Elevated expression of chemokineC-C ligand 2 in stroma is associatedwith recurrent basal-like breast cancers.Mod Pathol 2016;29:810–23.

11. Lim SY, Yuzhalin AE, Gordon-Weeks AN, Muschel RJ. Targeting the CCL2-CCR2 signaling axis in cancer metastasis. Oncotarget 2016;7:28697–710.

12. Borsig L, Wolf MJ, Roblek M, Lorentzen A, Heikenwalder M. Inflammatorychemokines and metastasis–tracing the accessory. Oncogene 2014;33:3217–24.

13. Fang WB, Jokar I, Zou A, Lambert D, Dendukuri P, Cheng N. CCL2/CCR2chemokine signaling coordinates survival and motility of breast cancercells through Smad3protein- andp42/44mitogen-activated protein kinase(MAPK)-dependent mechanisms. J Biol Chem 2012;287:36593–608.

14. Ginestier C, Hur MH, Charafe-Jauffet E, Monville F, Dutcher J, Brown M,et al. ALDH1 is a marker of normal andmalignant humanmammary stemcells and a predictor of poor clinical outcome. Cell Stem Cell 2007;1:555–67.

15. Jeyaraju DV, Cisbani G, De Brito OM, Koonin EV, Pellegrini L. Hax1 lacksBH modules and is peripherally associated to heavy membranes: implica-tions for Omi/HtrA2 and PARL activity in the regulation of mitochondrialstress and apoptosis. Cell Death Differ 2009;16:1622–9.

16. O'Hare MJ, Bond J, Clarke C, Takeuchi Y, Atherton AJ, Berry C, et al.Conditional immortalization of freshly isolated human mammary fibro-blasts and endothelial cells. Proc Natl Acad Sci U S A 2001;98:646–51.

17. Miller FR, Santner SJ, Tait L, Dawson PJ.MCF10DCIS.comxenograftmodelof human comedo ductal carcinoma in situ. J Natl Cancer Inst 2000;92:1185–6.

18. Ethier SP, Ammerman CA, Dziubinski ML. Isolation and culture ofhuman breast cancer cells from primary tumors and metastases. In:Margot M, Ip BBA, editors. Methods in mammary gland biology andbreast cancer research. New York, NY: Kluwer Academic/PlenumPublishers; 2000.

19. Behbod F, Kittrell FS, LaMarcaH, EdwardsD, Kerbawy S,Heestand JC, et al.An intraductal human-in-mouse transplantation model mimics the sub-types of ductal carcinoma in situ. Breast Cancer Res 2009;11:R66.

20. Cheng N, Lambert DL. Mammary transplantation of stromal cells andcarcinoma cells in C57BL/6J mice. J Vis Exp 2011;2716:1–6.

21. Cheng N, Bhowmick NA, Chytil A, Gorksa AE, Brown KA, Muraoka R,et al. Loss of TGF-beta type II receptor in fibroblasts promotesmammary carcinoma growth and invasion through upregulation ofTGF-alpha-, MSP- and HGF-mediated signaling networks. Oncogene2005;24:5053–68.

22. Mardekian SK, Bombonati A, Palazzo JP. Ductal carcinoma in situ of thebreast: the importance of morphologic and molecular interactions.Hum Pathol 2016;49:114–23.

23. Sambade MJ, Kimple RJ, Camp JT, Peters E, Livasy CA, Sartor CI, et al.Lapatinib in combination with radiation diminishes tumor regrowth inHER2þ and basal-like/EGFRþ breast tumor xenografts. Int J Radiat OncolBiol Phys 2010;77:575–81.

24. Valdez KE, Fan F, Smith W, Allred DC, Medina D, Behbod F. Humanprimary ductal carcinoma in situ (DCIS) subtype-specific pathology ispreserved in a mouse intraductal (MIND) xenograft model. J Pathol 2011;225:565–73.

25. Damiani S, Ludvikova M, Tomasic G, Bianchi S, Gown AM, Eusebi V.Myoepithelial cells and basal lamina in poorly differentiated in situ ductcarcinoma of the breast. An immunocytochemical study. Virchows Arch1999;434:227–34.

26. Lazard D, Sastre X, Frid MG, Glukhova MA, Thiery JP, Koteliansky VE.Expression of smooth muscle-specific proteins in myoepithelium andstromal myofibroblasts of normal and malignant human breast tissue.Proc Natl Acad Sci U S A 1993;90:999–1003.

27. Patarroyo M, Tryggvason K, Virtanen I. Laminin isoforms in tumor inva-sion, angiogenesis and metastasis. Semin Cancer Biol 2002;12:197–207.

28. Ioachim E, Charchanti A, Briasoulis E, Karavasilis V, Tsanou H, ArvanitisDL, et al. Immunohistochemical expression of extracellular matrix com-ponents tenascin, fibronectin, collagen type IV and laminin in breastcancer: their prognostic value and role in tumour invasion andprogression.Eur J Cancer 2002;38:2362–70.

29. Park SY, Kim HM, Koo JS. Differential expression of cancer-associatedfibroblast-related proteins according to molecular subtype and stromalhistology in breast cancer. Breast Cancer Res Treat 2015;149:727–41.

30. SugimotoH,Mundel TM, KieranMW, Kalluri R. Identification of fibroblastheterogeneity in the tumor microenvironment. Cancer Biol Ther 2006;5:1640–6.

31. Carvalho I, Milanezi F, Martins A, Reis RM, Schmitt F. Overexpression ofplatelet-derived growth factor receptor alpha in breast cancer is associatedwith tumour progression. Breast Cancer Res 2005;7:R788–95.

32. Bussard KM, Mutkus L, Stumpf K, Gomez-Manzano C, Marini FC. Tumor-associated stromal cells as key contributors to the tumor microenviron-ment. Breast Cancer Res 2016;18:84.

33. Fang WB, Yao M, Cheng N. Priming cancer cells for drug resistance: role ofthe fibroblast niche. Front Biol 2014;9:114–26.

34. Zhang C, Yan Z, Arango ME, Painter CL, Anderes K. Advancing biolumi-nescence imaging technology for the evaluation of anticancer agents in theMDA-MB-435-HAL-Luc mammary fat pad and subrenal capsule tumormodels. Clin Cancer Res 2009;15:238–46.

35. Cheng N, Chytil A, Shyr Y, Joly A, Moses HL. Enhanced hepatocyte growthfactor signaling by type II transforming growth factor-beta receptor knock-out fibroblasts promotes mammary tumorigenesis. Cancer Res 2007;67:4869–77.






36. LivasyCA, KaracaG,NandaR, TretiakovaMS,OlopadeOI,MooreDT, et al.Phenotypic evaluation of the basal-like subtype of invasive breast carci-noma. Modern Pathol 2006;19:264–71.

37. Szasz AM, Lanczky A, Nagy A, Forster S, Hark K, Green JE, et al. Cross-validation of survival associated biomarkers in gastric cancer using tran-scriptomic data of 1,065 patients. Oncotarget 2016;7:49322–33.

38. Hu M, Yao J, Carroll DK, Weremowicz S, Chen H, Carrasco D, et al.Regulation of in situ to invasive breast carcinoma transition. Cancer Cell2008;13:394–406.

39. Kimachi K, Croft M, Grey HM. The minimal number of antigen-majorhistocompatibility complex class II complexes required for activation ofnaive and primed T cells. Eur J Immunol 1997;27:3310–7.

40. Viola A, Lanzavecchia A. T cell activation determined by T cell receptornumber and tunable thresholds. Science 1996;273:104–6.

41. Capuani F, Conte A, Argenzio E, Marchetti L, Priami C, Polo S, et al.Quantitative analysis reveals how EGFR activation and downregulation arecoupled in normal but not in cancer cells. Nat Commun 2015;6:7999.

42. Norton N, Advani PP, Serie DJ, Geiger XJ, Necela BM, Axenfeld BC, et al.Assessment of tumor heterogeneity, as evidenced by gene expressionprofiles, pathway activation, and gene copy number, in patients withmultifocal invasive lobular breast tumors. PLoS One 2016;11:e0153411.

43. Jedeszko C, Victor BC, Podgorski I, Sloane BF. Fibroblast hepatocytegrowth factor promotes invasion of human mammary ductal carcinomain situ. Cancer Res 2009;69:9148–55.

44. Farnie G, Clarke RB, Spence K, Pinnock N, Brennan K, Anderson NG, et al.Novel cell culture technique for primary ductal carcinoma in situ: role ofNotch and epidermal growth factor receptor signaling pathways. J NatlCancer Inst 2007;99:616–27.

45. Grotendorst GR, Duncan MR. Individual domains of connective tissuegrowth factor regulate fibroblast proliferation and myofibroblast differ-entiation. FASEB J 2005;19:729–38.

46. Vuga LJ, Ben-Yehudah A, Kovkarova-Naumovski E, Oriss T, Gibson KF,Feghali-Bostwick C, et al. WNT5A is a regulator of fibroblast prolifer-ation and resistance to apoptosis. Am J Respir Cell Mol Biol 2009;41:583–9.

47. Meran S, Thomas DW, Stephens P, Enoch S, Martin J, Steadman R, et al.Hyaluronan facilitates transforming growth factor-beta1-mediated fibro-blast proliferation. J Biol Chem 2008;283:6530–45.

48. Mariggio MA, Cassano A, Vinella A, Vincenti A, Fumarulo R, Lo Muzio L,et al. Enhancement of fibroblast proliferation, collagen biosynthesisand production of growth factors as a result of combining sodiumhyaluronate and amino acids. Int J Immunopathol Pharmacol 2009;22:485–92.

49. Rodriguez-Torres M, Allan AL. Aldehyde dehydrogenase as a marker andfunctional mediator of metastasis in solid tumors. Clin Exp Metastasis2016;33:97–113.

50. Kilbride SM,Prehn JH.Central roles of apoptotic proteins inmitochondrialfunction. Oncogene 2013;32:2703–11.


Brummer et al.




2018;16:296-308. Published OnlineFirst November 13, 2017.Mol Cancer Res Gage Brummer, Diana S. Acevedo, Qingting Hu, et al. Survival and Invasion through Fibroblast-Dependent MechanismsChemokine Signaling Facilitates Early-Stage Breast Cancer

Updated version

10.1158/1541-7786.MCR-17-0308doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2017/11/11/1541-7786.MCR-17-0308.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/16/2/296.full#ref-list-1

This article cites 49 articles, 8 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/16/2/296.full#related-urls

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

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://mcr.aacrjournals.org/content/16/2/296To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-17-0308

http://mcr.aacrjournals.org/content/suppl/2017/11/11/1541-7786.MCR-17-0308.DC1

http://mcr.aacrjournals.org/content/16/2/296.full#ref-list-1

http://mcr.aacrjournals.org/content/16/2/296.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/16/2/296


chemokine signaling facilitates early-stage breast cancer ... · model inhibited tumor invasion and...

Documents