tumorcell-derivedperiostinregulatescytokines that maintain ...hamid m. abdolmaleky1, and sam...

Signal Transduction

TumorCell-DerivedPeriostinRegulatesCytokinesThat Maintain Breast Cancer Stem CellsArthur W. Lambert1, Chen Khuan Wong1,2, Sait Ozturk1, Panagiotis Papageorgis1,2,Rekha Raghunathan1, Yuriy Alekseyev3, Adam C. Gower4, Bj€orn M. Reinhard5,Hamid M. Abdolmaleky1, and Sam Thiagalingam1,2,3

Abstract

Basal-like breast cancer (BLBC) is an aggressive subtype ofbreast cancer which is often enriched with cancer stem cells(CSC), but the underlying molecular basis for this connectionremains elusive. We hypothesized that BLBC cells are able toestablish a niche permissive to the maintenance of CSCs andfound that tumor cell-derived periostin (POSTN), a componentof the extracellular matrix, as well as a corresponding cognatereceptor, integrin avb3, are highly expressed in a subset of BLBCcell lines as well as in CSC-enriched populations. Furthermore,we demonstrated that an intact periostin–integrin b3 signalingaxis is required for the maintenance of breast CSCs. POSTN

activates the ERK signaling pathway and regulates NF-kB–mediated transcription of key cytokines, namely IL6 and IL8,which in turn control downstream activation of STAT3. Insummary, these findings suggest that BLBC cells have an innateability to establish a microenvironmental niche supportive ofCSCs.

Implications: The findings reported here indicate that POSTNproduced by CSCs acts to reinforce the stem cell state through theactivation of integrin receptors and the production of key cyto-kines. Mol Cancer Res; 14(1); 103–13. �2015 AACR.

IntroductionCancer stem cells (CSC) represent a highly malignant subpop-

ulation of tumor cells that have been proposed to drive tumor-igenesis, metastasis, disease recurrence and resistance to chemo-therapy (1, 2). Breast CSCs (CD44high/CD24low/�) were firstidentified based on their ability to initiate tumor growth anddifferentiate intonon-CSCswhen injected at limiting dilution (3).Since then, various other phenotypic characteristics have beenassociated with breast CSCs including high aldehyde dehydroge-nase (ALDH) activity (4), an enhanced capacity to generatemammospheres, a surrogate marker for self-renewal and stemcells (5, 6), and activation of an epithelial–mesenchymal transi-

tion (EMT) program (7, 8). The relevance of CSCs to breast canceris supported by clinicalfindings that linkmarkers of their presence(i.e., expression of stem cell–associated genes) to a poor diseaseprognosis (9) and chemotherapeutic resistance (10).

Several signaling pathways have emerged as critical regulatorsof the CSC state. Many of these signals constitute importantdevelopmental programs that, in the case of breast cancer, mayact to sustain a population of CSCs (11, 12). Recently, cytokines,namely IL6 and IL8, have been implicated in the maintenance ofbreast CSCs (13–16). And conditions such as hypoxia have alsobeen shown to promote the appearance of breast CSCs (17).Therefore, in addition to defined geneticmutations, the surround-ing tumor microenvironment may play a major role in thepromotion of the CSC state.

Basal-like breast cancer (BLBC) is an aggressive molecularsubtype of the disease that is enriched for cells with stem-likefeatures such as a CD44high/CD24low surface marker profile (18),activation of a mesenchymal transcriptional program (19) andresistance to therapy (20). The relevance and underlying basis forthis enrichment are not well understood but it raises the possi-bility that BLBC pathogenesis could be linked to the creation of atumor microenvironment conducive to the formation or main-tenance of CSCs.

Here, we examined the hypothesis that BLBC cells secreteperiostin (encoded by the POSTN gene, also known as OSF-2),a mesenchymal extracellular matrix (ECM) protein, to support apopulation of CSCs. These studies highlight a role for tumor cell-derivedPOSTN,which signals through integrinavb3 to regulate theexpression of key cytokines and mediates stemness in BLBC cells.

Materials and MethodsCell culture

All M cell lines (MI, MII, MIII) were cultured in DMEM F/12media containing 5% horse serum, 10 mg/mL insulin, 20 ng/mL

1Department of Medicine, Molecular Medicine Program, Section ofBiomedical Genetics and Cancer Center, Boston University School ofMedicine, Boston,Massachusetts. 2DepartmentofGenetics andGeno-mics, Boston University School of Medicine, Boston, Massachusetts.3Department of Pathology and Laboratory Medicine, Boston Univer-sity School of Medicine, Boston, Massachusetts. 4Clinical and Trans-lational Science Institute, Boston University, Boston, Massachusetts.5Department of Chemistry, BostonUniversity, Boston,Massachusetts.

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

C.K. Wong and S. Ozturk contributed equally to this article.

Current address for A.W. Lambert:Whitehead Institute for Biomedical Research,Cambridge,MA; current address for S. Ozturk: IcahnSchool ofMedicine atMountSinai, Oncological Sciences, New York, NY; current address for P. Papageorgis:Department of Life Sciences, School of Sciences, European University Cyprus,Nicosia, Cyprus.

Corresponding Author: Sam Thiagalingam, Boston University School of Med-icine, 72 East Concord Street, Boston, MA 02118. Phone: 617-638-6013; Fax: 617-638-4275; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-15-0079

�2015 American Association for Cancer Research.

MolecularCancerResearch

www.aacrjournals.org 103

on February 20, 2021. © 2016 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 27, 2015; DOI: 10.1158/1541-7786.MCR-15-0079

http://mcr.aacrjournals.org/

EGF, 100 ng/mL cholera toxin, and 0.5 mg/mL hydrocortisone, aspreviously described (21). SUM159 cell lines were cultured inHam's F/12 supplemented with 5% FBS, 10 mg/mL insulin, and0.5 mg/mL hydrocortisone while Hs578T cells weremaintained inDMEM containing 10% FBS. All other cell lines were cultured asdescribed previously (22). Mammospheres were grown in Mam-moCult media (Stem Cell Technologies) containing 0.48 mg/mLhydrocortisone and 1% methylcellulose. All cells were grown inthe presence of penicillin/streptomycin (1%) and CO2 (5%) at37�C in a humidified incubator.

Antibodies and reagentsA detailed list of all antibodies and reagents used is provided in

the Supplementary Data.

Generation of knockdown linesStable knockdown of periostin or integrin b3 was achieved by

transduction of lentiviral vectors (pLKO.1) followed by selectionwith puromycin (2 mg/mL). shRNA sequences can be found in theSupplementary Data.

qRT-PCRTotal RNA was isolated with TRIzol (Life Technologies) and

cDNA synthesized using random primers and SuperScript IIReverse Transcriptase (Life Technologies). qRT-PCR was per-formed with Power SYBR Green PCR Master Mix (Life Technol-ogies) and run on an Applied Biosystems 7900HT SequenceDetection System. Relative RNA expression was calculated usingtheDDCtmethod and normalized to b-actin. Primer sequences arelisted in the Supplementary Data.

ELISAsSecreted periostin was measured with an ELISA kit (Adipo-

Gen) according to the manufacturer's protocol. Protein wasconcentrated using Amicon centrifugation filters (30K mem-brane, Millipore) before analysis. To measure cytokine produc-tion, cells were starved in serum-reduced media (0.1% serum)and conditioned media collected 48 hours later. IL8 proteinwas quantified using a Quantikine ELISA Kit (R&D Systems),whereas IL-6 quantification was performed with a CyPlex(CyVek, Inc.).

Western blottingProtein was isolated from cells on ice with RIPA buffer contain-

ing protease and phosphatase inhibitor cocktails (Roche). Forexamination of signaling events, cells were grown in serum-reduced media (0.1% serum) for 24 or 48 hours before proteinisolation. Proteinswere separatedby SDS-PAGEand immobilizedon PVDF membranes. Secondary antibodies were conjugated toperoxidase and detected by chemiluminescence with ECL solu-tion (PerkinElmer).

Flow cytometryFor examination of cell surface proteins, cells were trypsi-

nized, washed in FACS buffer (PBS with 0.2% BSA, 0.09%sodium azide and 1 mmol/L EDTA) and stained with antibodyfor 30 minutes at 4�C. To quantify the percentage of ALDH-positive cells, the ALDEFLUOR assay was used (Stem CellTechnologies). Briefly, cells were trypsinized, suspended inassay buffer, and treated for 30 minutes with the ALDEFLUORreagent (BODIPY-aminoacetaldehyde, BAAA; 3 mmol/L), in the

presence or absence of the ALDH inhibitor diethylaminoben-zaldehyde (DEAB; 37.5 mmol/L). Samples were analyzed onFACScan, FACSCalibur, LSR II or Accuri C6 (BD) cytometers.Cells were sorted using a different buffer (PBS, 2% horse serum,1 mmol/L ETDA) with a MoFlo (Beckman Coulter).

Animal studiesTumor xenograft experiments were conducted under the

approval of the Institutional Animal Care and Use Committeeat Boston University School of Medicine (Boston, MA). Biolumi-nescent cells were generated using a pMSCV-Luc-PGK-hygro plas-mid (Addgene-8782). Fifty thousand cells were injected subcu-taneously, in the absence ofMatrigel (in 0.1mLof serum-free F/12media), into 6- to 8 week-old NOD/SCID mice (The JacksonLaboratory). Animals were monitored on a weekly basis using anIVIS Spectrum Imaging System (Caliper) for a period of 6months,unless tumor growth necessitated euthanasia.

Gene expression microarray and bioinformatic analysisTotal RNA was isolated using TRizol (Life Technologies) and

cleanedupwith theRNeasyMini kit (Qiagen). Biotin labelingwasperformed using the Ambion WT Expression Kit (Life Technolo-gies) and the GeneChip WT Terminal Labeling and Controls Kit(Affymetrix), followed by hybridization to GeneChip HumanGene 2.0 ST arrays (Affymetrix). Raw CEL files were normalizedusing the Robust Multiarray Average [PubMed ID 14960456].Differential gene expression between shPN and shGFP wasassessed using the moderated t test implemented in the "limma"package. Raw CEL files from GEO Series GSE21653 were normal-ized in the same manner, and differential expression betweenmolecular subtypes was assessed using the Welch t test. Allmicroarray analyses were performed using the R environmentfor statistical computing.

Luciferase reporter assayCells were cotransfected with the 3� KB-L NF-kB luciferase

reporter plasmid (Addgene #26699) and a Renilla-luciferaseplasmid using X-tremeGENE HP DNA transfection reagent(Roche). Twenty-four hours after transfection, cells were starvedin reduced serum medium (0.1% FBS) for 16 hours. TNFatreatment (20 ng/mL, R&D Systems) served as a positive control.For experiments that involved inhibition of the ERK pathway,inhibitors were added 24 hours after transfection and lumines-cence was measured 24 hours later. Luminescence was quantifiedwith the Dual-Luciferase Reporter Assay (Promega) using a Glo-Maxmicroplate reader. Firefly luciferase signal was normalized tothe Renilla luciferase signal to determine relative luminescenceunits (RLU).

Statistical analysisStatistical analysis was performed using a two-tailed paired

Student t test. A P value less than 0.05 was considered statisticallysignificant. Error bars represent � SEM.

ResultsCharacterization of cancer stem cell traits in M cells

To identify tumor cell-derived factors that could support themaintenance of breast CSCs, we examined a well-established cellline model of breast cancer progression (23, 24). This modelsystem consists of three cell lines: MCF10A (MI) and two

Lambert et al.

Mol Cancer Res; 14(1) January 2016 Molecular Cancer Research104




derivatives of this line, MCF10AT1k.cl2 (MII) and MCF10CA1h(MIII). These cell lines exhibit increasing degrees of malignancy(24). Previous characterization by our laboratory has indicatedthat while MI and MII cells display numerous epithelial features,MIII cells clearly exhibit a phenotype associated with activationof an EMT program (21).

Given the connection between EMT and CSCs (7, 8), wewondered whether MIII cells also exhibit features associated withthe stem cell state. In support of this notion, MIII cells predom-inantly displayed the CD44high/CD24low surface phenotype (Fig.1A) and preferentially formed mammospheres, doing so withnearly a 9-fold greater efficiency than MII cells (Fig. 1B). Becausethe MIII cell line displayed various traits reminiscent of breastCSCs, we reasoned that it could serve as a good model to identifyregulators of the CSC state.

Periostin is highly expressed in basal-like breast cancer stemcells

Because all of the M model cell lines share the same geneticlineage, we hypothesized that altered expression of secreted micro-environmental factors may contribute to the CSC-like features ofMIII cells. Our previous gene expression profiling studies (21)revealed that periostin (POSTN) was overexpressed in the MIII cellline. Thiswasparticularly interesting becausePOSTN is regulatedbyTwist (25) and TGFb (26), two factors that promote passagethrough an EMT and acquisition of stem cell phenotypes (7). Wevalidated thatPOSTN is a transcriptional targetof theTGFbpathwayin this model of breast cancer, as treatment ofMII cells with TGFb1resulted in a strong increase in POSTN transcription (Supplemen-tary Fig. S1A). Conversely, active TGFb signaling was necessary tosustain POSTN expression in MIII cells (Supplementary Fig. S1B).

Figure 1.Overexpression of POSTN and ITGB3 inBLBC cells suggests a functional role forthis signaling axis. A, flow-cytometricanalysis of CD44 and CD24 surfaceexpression in MI, MII, and MIII cells. Thegraph shows the percentage ofCD44high/CD24low cells in each line,n¼ 3. B, mammosphere formation assaywith MII and MIII cells, n ¼ 3. C,expression of POSTN in the M cell lineswas measured using qRT-PCR. D,Western blot analysis for POSTN in theconditioned media of the indicated celllines. Ponceau stainingwasused to showprotein loading. E, POSTN mRNA wasmeasured by qRT-PCR in MIII cellsbefore and after growth asmammospheres. F, POSTN mRNA wasmeasured by qRT-PCR in MII cells sortedas either CD44high/CD24low or CD44high

/CD24high. G, flow-cytometric analysis ofsurface levels of integrin avb3 in the Mcell lines. Gray shading indicates isotypecontrol. Numbers in parenthesesindicate the percentage of positive cellsin each line. H, expression of POSTN andITGB3 was assessed by qRT-PCR. Redindicates luminal cell lines while blackindicates basal-like cell lines. The circleshows the subset of cells that expresshigh levels of both genes. � , P < 0.05;�� , P < 0.01; error bars represent SEM.

Periostin-ITGB3 Signaling Maintains Breast CSCs

www.aacrjournals.org Mol Cancer Res; 14(1) January 2016 105




Using qRT-PCR, we confirmed that the level of POSTN mRNAwas significantly increased in MIII cells compared with MI or MIIcells (Fig. 1C). As expected, this increase in transcription wascorrelated with enhanced secretion of POSTN into the surround-ing media (Fig. 1D). Moreover, we found that POSTN was highlyexpressed in populations enriched for CSCs. For instance, MIIIcells grown as mammospheres, which are predominately formedby CSCs (6), expressed nearly 10 times more POSTN than theiradherent counterparts (Fig. 1E). In a second approach, when MIIcells were fractionated based on the surface expression of CD44and CD24, we found that cells within the CD44high/CD24low

(CSC) population expressed significantly more POSTN than cellswithin the CD44high/CD24high (non-CSC) fraction (Fig. 1F).Overall, in theMCF10A breast cancer model, high POSTN expres-sion directly correlated with multiple phenotypes attributed toCSCs.

Basal-like breast cancer cells express a periostin–integrin b3signaling axis

In examining gene expression profiles of the M cell lines, wenoted that, in addition to POSTN, MIII cells also expressed highlevels of integrin b3 (ITGB3), a subunit of the integrin avb3complex that is a known receptor for POSTN (27). Indeed, wefound that the surface levels of the functional integrin avb3complex were abundant on MIII cells but nearly absent from MIand MII cells (Fig. 1G). The coordinated expression of the recep-tor-ligand combination required for the activity of an intactsignaling axis suggested that, in MIII cells, POSTN could poten-tially signal in an autocrine fashion.

Wishing to expand on this observation, we examined whethersuch an axis was present in other breast cancer cells. To this end,we measured the mRNA levels of both POSTN and ITGB3 in apanel of nine breast cancer cell lines. A high level of expression ofboth genes was detected in four cell lines: BT549, SUM1315,SUM159, andHs578T cells (Fig. 1H).Notably, and similar toMIIIcells (21), all four of these cell lines fall under the molecularclassification of BLBC cells (22). In a subset of these lines, weverified that the increase in POSTN and ITGB3 expression corre-lated with enhanced POSTN secretion (Supplementary Fig. S2A)and increased surface levels of the integrin avb3 receptor complex(Supplementary Fig. S2B). These data indicated that a periostin–integrin b3 signaling axis is preserved, and potentially operative,in a subset of BLBC cells.

Tumor cell–derived POSTN and ITGB3 maintain breast cancerstem cells

To test the relevance of POSTN expression to CSCs, we opted tofocus on the SUM159 line because it has a well-defined CSCpopulation (28). We generated SUM159 cells with stable knock-down of POSTN (shPN) or ITGB3 (shBeta3) through expressionof corresponding shRNA constructs (Supplementary Fig. S3A andS3B). Knockdown of POSTN or ITGB3 in SUM159 cells did nothave an effect on cell proliferation or apoptosis (SupplementaryFig. S3C and S3D) and did not lead to any morphologic changesor reversion of the EMT program. (Supplementary Fig. S3E andS3F). However, POSTN knockdown resulted in a significantreduction in the ability of SUM159 cells to generate mammo-spheres, with shPN cells forming 72% fewermammospheres thancontrol (shGFP) cells (Fig. 2A). Similarly, SUM159 shBeta3 cellsalso exhibited a significant reduction inmammosphere formationpotential (Fig. 2A). Notably, the most dramatic effect of POSTN

knockdown was observed in the well-established (28), ALDH-positive CSC population of SUM159 cells. Under conditions ofreduced serum (0.1% serum), which are often used to analyze theeffect of ECMproteins and integrin signaling (29), close to 15%ofSUM159 shGFPwere ALDH-positive.On the other hand, only 2%of the cells were ALDH positive in SUM159 shPN cells, represent-ing a 7.4-fold reduction in this highly tumorigenic cell population(Fig. 2B). Furthermore, SUM159 shBeta3 cells had a 2.9-foldreduction in the ALDH-positive CSCpopulation (Fig. 2B). Impor-tantly, we verified these findings with an additional shRNAhairpin directed at either POSTN or ITGB3 (Supplementary Fig.S4A–S4C). These results suggest that BLBC cells rely on POSTNand ITGB3 to sustain a population of CSCs.

Additional evidence for this notion was obtained from experi-ments with the MIII cell line. Here too, knockdown of POSTN(Supplementary Fig. S5A) led to a significant reduction in mam-mosphere-forming potential; MIII shPN cells generated 52%fewer mammospheres than control cells (Supplementary Fig.S5B). Notably, when MIII cells were sorted based on the surfacelevel of integrin avb3, there was a dramatic difference in thecapacity to initiate mammosphere growth, where only MIII cellswith high surface levels of avb3 could robustly form mammo-spheres (Fig. 2C). The overlapping phenotypes observed follow-ing perturbation of POSTN or ITGB3 imply that these proteinsfunction in a closely related or overlapping pathway to regulatethe CSC state in BLBC.

Consistentwith the abovefindings,when shGFP and shPNcellswere seeded as mammospheres at limiting dilution we detected aclear reduction in the estimated stem cell frequency, with thesphere formation efficiency of 100 shGFP control cells beingroughly equal to that of 1000 shPN cells (Fig. 2D). Importantly,knockdown of POSTN impaired the ability of SUM159 cells toinitiate tumors when injected at limiting numbers (5� 104 cells).Six months after injection, 50% of mice had developed tumors inthe control group while no tumors were detected in animalsinjected with SUM159 shPN cells (Fig. 2E). This result is consis-tent with our in vitro findings and lend support to the idea thattumor cell-derived periostin is crucial for tumor initiation, thesalient characteristic of CSCs.

Periostin regulates cytokine production via ERK signaling andNF-kB

Tounderstand themolecular events downstreamof POSTN,weperformed genome-wide expression profiling of SUM159 shGFPand shPN cells and used Gene Set Enrichment Analysis (GSEA) toidentify the major pathways that were altered upon POSTNknockdown in SUM159 cells. This analysis revealed six gene sets,containing a preponderance of cytokines, that were significantlyrepressed in SUM159 shPN cells (Supplementary Fig. S6A). Giventhe connection between cytokines and breast CSCs (13), wewished to investigate this further. Close examination of the twentygenes that comprise the cytokine pathway gene set highlightedfive cytokines that were clearly repressed in SUM159 shPN cellscompared with shGFP cells: IL1A, IL8, IL6, IL18, and IL16 (Fig.3A). Of these five cytokines, IL1A, IL8, and IL6 were found at theleading edge of other downregulated gene sets (SupplementaryFig. S6B), implying that theymay represent particularly importantmediators downstream of POSTN.

We used qRT-PCR to confirm the repression of IL1A, IL6, andIL8 in SUM159 shPN cells; importantly, the expression of all threegenes was also significantly reduced in SUM159 shBeta3 cells

Lambert et al.





(Fig. 3B). Further, conditioned media from shPN and shBeta3cells contained significantly less IL8 and IL6 protein than mediaconditioned from shGFP control cells (Fig. 3C and D). Of thesethree cytokines, IL6 expression was most closely correlated withthe expression of POSTN in a panel of six BLBC cell lines (Fig. 3E),suggesting that the regulation of IL6 expression by POSTNmay becommon in BLBC cells.

In light of the reduced cytokine levels, and because STAT3 is atransducer of cytokine signaling that has been linked to themaintenance of breast CSCs (15), we were prompted to investi-gate the STAT3 pathway in cells with knockdown of POSTN orITGB3. Indeed, although SUM159 shGFP cells exhibited relativelyhigh levels of active, phosphorylated STAT3, both SUM159 shPN

and shBeta3 cells hadmarkedly reduced levels of phosphorylatedSTAT3 (Fig. 3F). The impaired transcription of IL6 and IL8, alongwith reduced activation of STAT3, was also verified using anadditional hairpin directed at POSTN or ITGB3 (SupplementaryFig. S4D and S4E).

To test whether impaired production of secreted factors wasresponsible for this effect, we collected conditioned media fromboth SUM159 shGFP and shPN cells, and then treated the threecell lines with this or serum-reduced media (mock) for 60 min-utes. The diminished activation of STAT3 in the knockdown cellscould be rescued by transient exposure to media conditionedfrom shGFP cells, but not media conditioned from shPN cells(Fig. 3G). These data indicate that impaired production of

Figure 2.POSTN signaling is required to maintainbreast CSCs. A, SUM159 control (shGFP)and periostin (shPN) or integrin b3(shBeta3) knockdown cells were grownas mammospheres and quantified,n ¼ 2. Representative images ofmammospheres are shown on the right.B, quantification of the percentage ofALDH-positive cells in SUM159 shGFP,shPN, and shBeta3 cells after 48hours ofgrowth in serum-reduced (0.1% serum)media, n ¼ 3. Representative ALDHplots are shown on the right. DEAB, aninhibitor of ALDH, was used as a gatingcontrol. C, MIII cells were sorted by flowcytometry based on the surfaceexpression of the integrin avb3 complex(LM609) before seeding an equalnumber of cells for mammospheregrowth, n ¼ 3. D, SUM159 shGFP andshPN cells were seeded asmammospheres at limiting dilution,n ¼ 3. E, the percentage of animals thatdeveloped tumors 6 months aftersubcutaneous injection of the indicatedcell lines (5 � 104 cells), n ¼ 4 animalsper group. � , P < 0.05; �� , P < 0.01; errorbars represent SEM.






secreted factors from SUM159 shPN and shBeta3 cells, ostensiblythe cytokines mentioned above, is responsible for the observedreduction in STAT3 signaling, rather than any inherent intracel-lular defects. In addition, we found that treatment of shPN cellswithmedia conditioned from control shGFP cells led to a 1.5-foldincrease in the ALDH-positive subpopulation (SupplementaryFig. S7A). This effect was not observed in shBeta3 cells, suggestingthat both POSTN and cytokine signaling were required to main-tain this population.

Next, wewished todetermine the proximal signals downstreamof the POSTN–ITGB3 pathway that were responsible for control-ling cytokine production. Our approach to achieve this aimwas toexamine the expression profiles of SUM159 shGFP and shPN cellsusing Ingenuity Pathway Analysis (IPA) software to look for

upstream regulatory networks that might help to explain thedifferential gene expression profiles between these two cell lines.Interestingly, we noted that, in comparison with the controlshGFP cells, SUM159 shPN cells exhibited significant activationof gene expression profiles characteristic of cells treated with twoERK pathway inhibitors (Supplementary Fig. S7B and S7C). Thissuggested that cells with disrupted POSTN signaling might showimpaired activation of the ERK pathway. Consistent with this,cells with knockdown of POSTN or ITGB3 were found to havedramatically reduced levels of phosphorylated, active ERK1/2(Fig. 4A). This defect in ERK signaling appears to be upstreamof cytokine transcription, as pharmacologic inhibition of the ERKpathway led to significantly reduced expression of IL6, IL8, andIL1A (Fig. 4B and Supplementary Fig. S7D). In line with a recent

Figure 3.POSTN signaling regulates cytokineproduction. A, heatmap depictingexpression of the 20 genes contained inBioCarta Cytokine Network gene set inshGFP and shPN cells. Gene expressionvalues are zero-centered by row andarranged in order by the moderated tstatistic. Colors represent: blue, belowmean expression; red, above meanexpression. B, expression of IL1A, IL6,and IL8 was measured in the indicatedSUM159 cell lines by qRT-PCR. RNAwasisolated from cells grown in serum-reduced media for 48 hours. C, IL8protein was quantified from theconditioned media of the indicated celllines after 48 hours of growth in serum-reduced (0.1% serum) media. D, IL6protein was quantified from theconditioned media of the indicated celllines after 48 hours of growth in serum-reduced (0.1% serum) media. E,expression of POSTN and IL6 wasassessed by qRT-PCR in a panel of sixbasal-like breast cancer cell lines(Fig. 1H). F, Western blot analysis ofSTAT3 phosphorylation (tyrosine 705)and total STAT3 in the indicatedSUM159 cell lines. Protein was isolatedfrom cells grown in serum-reducedmedia (0.1% serum) for 48 hours. G,rescue of STAT3 signaling withconditioned media. Conditioned mediawere collected from shGFP and shPNcells and used to treat the indicatedSUM159 cell lines following growth inserum-reduced media for 48 hours.After starvation, the cells weretransiently exposed (60 minutes) toconditioned media (CM) from eithershGFP or shPN cells; serum-reducedmedia served as a mock control.Western blot analysis shows the levelsof phosphorylated STAT3 (tyrosine705), total STAT3, and b-actin.�� , P < 0.01; error bars represent SEM.

Lambert et al.





report that linked POSTN signaling to activation of the NF-kBpathway (30), we found that knockdown of POSTN or ITGB3impaired the transcriptional activity of NF-kB as assessed by aluciferase reporter assay (Fig. 4C). This result was verified using anindependent shRNA hairpin directed at POSTN or ITGB3. (Sup-plementary Fig. S4F). Furthermore, pharmacologic disruption ofERK signaling also caused a decrease in NF-kB transcriptionalactivity (Fig. 4D).

To test whether the connection between POSTN-controlledcytokine expression and CSCs is applicable outside of theSUM159 line, we also knocked down POSTN in Hs578T cells(Supplementary Fig. S8A), another BLBC cell line (22) that coex-presses high levels of both POSTN and integrin avb3 (Fig. 1H andSupplementary Fig. S2A and S2B). Consistent with our previousobservations, Hs578T cells with knockdown of POSTN formed57.9% fewer mammospheres (Supplementary Fig. S8B) andshowed a 3-fold reduction in the ALDH-positive subpopulation(Supplementary Fig. S8C). The loss of these CSC characteristicswas also associated with reduced IL6 expression (SupplementaryFig. S8D) and impaired activation of the STAT3 pathway (Sup-plementary Fig. S8E). This suggests that the functional relevanceof POSTN-regulated cytokine pathways may be more broadlyapplicable to other basal-like CSCs.

Periostin is associated with a poor prognosis in basal-likebreast cancer

We reasoned that cancers with an operative POSTN-ITGB3signaling axis might be enriched with CSCs and therefore asso-ciated with a worse clinical prognosis. On the basis of the fact thatITGB3 marks a population of luminal progenitor cells, the pre-sumed cell-of-origin for basal-like cancers (31), onewouldpredictthat thiswouldbe limited to a subset of POSTN-expressingBLBCs,many of which are likely to be positive for ITGB3. In accordancewith this notion, in silico Kaplan–Meier analysis of a public breast

cancer dataset (32) revealed that high POSTN expression wasassociated with reduced relapse-free survival in basal-like (HR ¼1.42, P¼ 0.0084), but not luminal A (HR¼ 1.08, P¼ 0.4), breastcancers (Fig. 5A). As POSTN has been suggested to promoteangiogenesis through activation of integrin avb3 on endothelialcells (33), it is possible that high POSTN expression could berelated to neoangiogenesis, which is also a prognostic factor inbreast cancer (34). However, the effect of angiogenesis wouldseem to be independent ofmolecular subtype, suggesting that theassociation with relapse-free survival could be based on analternative explanation. Indeed, our findings are in line with arecent study that found a correlation between POSTN expression,breast CSC content (as assessed by CD44high/CD24low cells), anda worse clinical prognosis (35).

In addition, we examined the expression of POSTN, IL6, andIL8 in another independent microarray dataset of 226 primarybreast cancers (GEO Series GSE21653; ref. 36). In this dataset,POSTNwas highly expressed (median percentile¼ 99.6%) acrossall molecular subtypes (Supplementary Fig. S9), but IL6 and IL8levels were significantly elevated in BLBC (Fig. 5B), in agreementwith previous findings that have implicated these cytokines in thegrowth and tumorigenicity of BLBC stem cells (15, 16, 37).Overall, these clinical data are consistent with a model whereinPOSTN can control the production of cytokines in the context ofBLBC.

DiscussionEMT-induced cancer cell lines have been successfully used to

screen for selective inhibitors of CSCs (38) and we reasoned thatMIII cells—which clearly underwent EMT and display numerousother CSC traits—could represent a similar cell line model inwhich to study regulators of the CSC state. We hypothesized thatproteins secreted into the local microenvironment would play a

Figure 4.POSTN controls cytokine productionthrough the ERK and NF-kB pathway.A, the indicated SUM159 cells weregrown in serum-reduced media (0.1%serum) for 24 hours before proteinisolation. Western blot analysis showsthe levels of phosphorylated ERK1/2(threonine 202/tyrosine 204) and totalERK1/2. B, SUM159 shGFP cells weretreated with DMSO (mock) or an ERKpathway inhibitor (PD184325) for 24hours before RNA isolation. IL1A, IL6,and IL8mRNA levelsweremeasured byqRT-PCR. C, the activity of atransfected NF-kB luciferase reporterwasmeasured in the indicated cell linesafter 16 hours of growth in serum-reduced media (0.1% serum). Relativeluminescence units (RLU) werecalculated based on cotransfection of aRenilla luciferase construct. TNFatreatment served as a positive control,n ¼ 3. D, the activity of a transfectedNF-kB luciferase reporter wasmeasured in SUM159 shGFP cells24 hours after treatment with theindicated inhibitors, n ¼ 3. � , P < 0.05;�� , P < 0.01; error bars represent SEM.






particularly important role in the maintenance of breast CSCs.Consistent with this notion, our previous gene expression pro-filing study (21) found a correlation between TGFb-mediatedEMT and POSTN expression, and here we focused our efforts onthe function of this protein in regulation of the CSC state.

As a matricellular protein, POSTN can potentially serve as acrucial mediator between CSCs and their surrounding microen-vironment (39). Yet there is limited mechanistic detail as to howPOSTN may function in BLBC cells. We found a POSTN–ITGB3signaling axis is operational in a number of breast cancer cell lines,and in each case, the cells could be classified molecularly underthe basal-like subtype. This is notable for a number of reasons.First, numerous lines of evidence support the idea that BLBCarisesfrom transformation of luminal progenitor cells (40–42), whichare specifically marked by expression of ITGB3 (43). Second,BLBC cells share numerous phenotypic similarities with CSCs(18–20), which may be especially true of a related disease sub-type, claudin-low breast cancers (44, 45). In this context, we

suggest that tumor cell-derivedPOSTNmaybeparticularly impor-tant for the maintenance of a population of CSCs in basal-likedisease.

In support of this hypothesis, our results from knockdownexperiments, carried out in three different BLBC cell lines, showedthat POSTN expression was required to maintain CSC pheno-types. These findings complement work from others, whichshowed that ectopic expression of POSTN promotes stem-likeproperties in humanmammary epithelial cells (46).Our data alsosupport a major role for the integrin avb3 receptor in sustainingmammosphere growth and an ALDH-positive subpopulation.Another study has addressed the relationship between integrinb3 andCSCs and reported that, while ITGB3 is highly expressed inbreast CSCs, knockdown of this gene resulted in minimal effectson mammosphere growth (47). Interestingly, these experimentswere performed in derivatives of SK-BR-3 cells, a luminal cell lineand, as argued above, this specific integrin may be more relevantin the context of BLBC.

Figure 5.High POSTN expression is associatedwith a poor prognosis in basal-likebreast cancer. A, in silico Kaplan–Meieranalysis of relapse-free survival in basal-like (n ¼ 581) and Luminal A (n ¼ 1,678)breast cancers. Patients were dividedinto two groups based on the medianexpression level of POSTN: highexpression (red) or low expression(black). High POSTN expression wasassociated with reduced relapse-freesurvival in basal-like (left panel, HR ¼1.42, P¼0.0084), but notwith luminal A(right panel, HR ¼ 1.08, P ¼ 0.4), breastcancers. B, IL6 and IL8 expression in apublicly available dataset of primarybreast cancer samples (GSE21653).Differential expression betweengroups was assessed by Welch t test.� , P < 0.05; �� , P < 0.001.

Lambert et al.





We also sought to identify the pathways downstream of thePOSTN-ITGB3 signaling axis that are relevant in the context ofCSC regulation. Here, we focused on a defined cytokine network,consisting of IL6, IL8, and IL1A, which was repressed followingPOSTN knockdown. Interestingly, other gene sets significantlyrepressed in shPN cells included the hematopoietic lineage andstem cell gene sets. As POSTN has been found to localize to stemcell niches in the bone marrow (48), this suggests that POSTN-mediated regulation of cytokine production might also havephysiologic relevance to hematopoietic stem cells. Furthermore,our observations are also consistent with a report that demon-strated that POSTN was required for the production of multipleproinflammatory cytokines during chronic skin inflammation(30). Thus, in the context of cancer, breast CSCs may activate asimilar program downstream of POSTN. We also found that cellswith defective POSTN signaling showed reduced activation ofERK signaling and, consequently, impaired cytokine transcrip-tion. This result is in accordance with a recent study, whichreported that activation of the MAPK pathway due to DUSP4loss promoted the formation of CSCs and increased the expres-sion of IL6 and IL8, in this case through the transcription factorsETS-1 and c-JUN (49). Our data suggest that, at least downstreamof POSTN, the NF-kB pathway is likely to be responsible fordriving cytokine expression, similar to what has been observedduring skin inflammation (30). Nevertheless, NF-kB transcrip-tional activity appears to lie downstream of the ERK pathway,raising the possibility that POSTN-mediated regulation of theCSC state may be susceptible to interference by MAPK inhibitors.

Overall, this leads to a model in which the POSTN-ITGB3signaling axis regulates the production of cytokines, which inturn activate STAT3 signaling. This model is compelling becauseof the rapidly accumulating evidence that links cytokines, espe-cially IL6 and IL8, to the regulation of breast CSCs (15, 16, 28, 50).These alterations may be further understood when consideringrecent studies that have reported on the cellular plasticity thatexists between non-CSCs and CSCs (14, 51). For instance, it hasbeen shown that the IL6-STAT3pathway can shift this equilibriumin favor of CSCs (14). Thus, one possibilitymay be that POSTN, atleast in part through control of cytokine production, is able to tiltthis equilibrium to promote the formation and/or maintenanceof CSCs.

In addition to the experiments presented here, others haverecently found that POSTN plays a role in the regulation of breastCSCs during the process of metastatic colonization (48). It hasalso been shown to act as a factor derived from neovascular tipcells in the perivascular niche, where it can serve to break tumorcell dormancy (52). Our work, while consistent with the centralidea that POSTN is required for themaintenanceofCSCs at sites oftumor cell dissemination, also differs in important aspects. First,we found that, in a subset of basal-like cancer lines, POSTN can betumor cell derived, potentially allowing these cells to generatetheir own functional niche and rendering themmore independentfrom their surroundingmicroenvironment. Similarly, POSTNhasrecently been reported to be secreted by glioblastoma stem cells(53). Second, our data highlight a major role for integrin avb3 infacilitating POSTN signaling, which represents an importantregulatory pathway for the maintenance of breast CSCs. Indeed,a recent study found integrin b3 to be a marker of CSCs and amediator of resistance to targeted therapy, not only in breastcancer, but also in lung and pancreatic cancers (54). Therefore,it is conceivable that periostin could contribute to CSC mainte-

nance and resistance to therapy in breast as well as other carci-nomas. Finally, our study implies that a POSTN signaling networkcan, in principle, act in the primary tumor as well, helping tomaintain a fraction of CSCs at earlier stages of tumor progression(Fig. 6). Therefore, we suggest that a subset of BLBCs can establishtheir own microenvironmental niche supportive of breast CSCsthrough the production of tumor cell-derived POSTN, whichmight be clinically relevant as a biomarker or therapeutic target.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: A.W. Lambert, S. Ozturk, C.K. Wong, P. Papageorgis,S. ThiagalingamDevelopment of methodology: A.W. Lambert, C.K. Wong, S. Ozturk,R. Raghunathan, B.M. Reinhard, H.M. Abdolmaleky, S. ThiagalingamAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): A.W. Lambert, C.K. Wong, S. Ozturk, P. Papageorgis,R. Raghunathan, Y. Alekseyev, S. ThiagalingamAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): A.W. Lambert, C.K. Wong, S. Ozturk, P. Papageorgis,A.C. Gower, B.M. Reinhard, S. ThiagalingamWriting, review, and/or revision of themanuscript: A.W. Lambert, C.K. Wong,S. Ozturk, P. Papageorgis, B.M. Reinhard, H.M. Abdolmaleky, S. ThiagalingamAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): A.C. GowerStudy supervision: S. Thiagalingam

AcknowledgmentsThe authors thank Drs. David Sherr and Ramon Parsons for generously

providing reagents.

Grant SupportThis work was supported by grants from Susan G. Komen for the Cure

(KG081435) and NIH/NCI (CA165707) (to S. Thiagalingam) and NIH/NCI(CA138509) (to B.M. Reinhard and S. Thiagalingam). A.W. Lambert is arecipient of a predoctoral traineeship award from the Department of Defense,Breast Cancer Research Program (W81XWH-11-1-006). P. Papageorgis was

Figure 6.Working model of POSTN signaling in basal-like breast cancer. Cancer cellswhich coexpress POSTN (in response to TGFb) and integrin avb3 are able toactivate downstream ERK signaling, which regulates the production ofcytokines via NF-kB. POSTNmay also be able to activate NF-kB directly (30).These soluble signals act to maintain a population of CSCs that may beresponsible for the recurrence of cancers associated with a worse clinicalprognosis. We suggest that this signaling network is most likely to beoperative in a subset of basal-like breast cancers. Note, POSTN has also beenshown to interact with Wnt1 and Wnt3A to enhance CSC maintenance (48).






supported with a postdoctoral fellowship from the Research Promotion Foun-dation, Cyprus (DIDAKTOR/0609/24). This work was also supported by a seedgrant from the BostonUniversityGenome Science Institute aswell as the BostonUniversity Flow Cytometry Core Facility, and the core facilities of the BostonUniversity Clinical and Translational Science Institute (CTSA award UL1-TR000157).

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 February 17, 2015; revised October 12, 2015; accepted October 14,2015; published OnlineFirst October 27, 2015.

References1. Shackleton M, Quintana E, Fearon ER, Morrison SJ. Heterogeneity in

cancer: cancer stem cells versus clonal evolution. Cell 2009;138:822–9.2. Visvader JE, Lindeman GJ. Cancer stem cells in solid tumours: accumu-

lating evidence and unresolved questions. Nat Rev Cancer 2008;8:755–68.3. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF.

Prospective identification of tumorigenic breast cancer cells. Proc NatlAcad Sci U S A 2003;100:3983–8.

4. Ginestier C, Hur MH, Charafe-Jauffret 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.

5. Dontu G, Abdallah WM, Foley JM, Jackson KW, Clarke MF, Kawamura MJ,et al. In vitropropagation and transcriptional profiling of humanmammarystem/progenitor cells. Genes Dev 2003;17:1253–70.

6. Ponti D, Costa A, Zaffaroni N, Pratesi G, Petrangolini G, Coradini D, et al.Isolation and in vitro propagation of tumorigenic breast cancer cells withstem/progenitor cell properties. Cancer Res 2005;65:5506–11.

7. Mani SA, Guo W, Liao MJ, Eaton EN, Ayyanan A, Zhou AY, et al. Theepithelial-mesenchymal transition generates cells with properties of stemcells. Cell 2008;133:704–15.

8. Morel AP, LievreM, ThomasC,HinkalG, Ansieau S, Puisieux A.Generationof breast cancer stem cells through epithelial-mesenchymal transition.PLoS ONE 2008;3:e2888.

9. Liu R,Wang X, ChenGY,Dalerba P, Gurney A, Hoey T, et al. The prognosticrole of a gene signature from tumorigenic breast-cancer cells. N Engl J Med2007;356:217–26.

10. Li X, Lewis MT, Huang J, Gutierrez C, Osborne CK, Wu MF, et al. Intrinsicresistance of tumorigenic breast cancer cells to chemotherapy. JNatl CancerInst 2008;100:672–9.

11. Takebe N, Ivy SP. Controversies in cancer stem cells: targeting embryonicsignaling pathways. Clin Cancer Res 2010;16:3106–12.

12. Reya T,Morrison SJ, ClarkeMF,Weissman IL. Stem cells, cancer, and cancerstem cells. Nature 2001;414:105–11.

13. Korkaya H, Liu S, Wicha MS. Breast cancer stem cells, cytokine networks,and the tumor microenvironment. J Clin Invest 2011;121:3804–9.

14. Iliopoulos D, Hirsch HA, Wang G, Struhl K. Inducible formation of breastcancer stem cells and their dynamic equilibriumwith non-stem cancer cellsvia IL6 secretion. Proc Natl Acad Sci U S A 2011;108:1397–402.

15. Marotta LL, Almendro V, Marusyk A, Shipitsin M, Schemme J, Walker SR,et al. The JAK2/STAT3 signaling pathway is required for growth of CD44(þ)CD24(�) stem cell-like breast cancer cells in human tumors. J ClinInvest 2011;121:2723–35.

16. Ginestier C, Liu S, Diebel ME, Korkaya H, Luo M, Brown M, et al. CXCR1blockade selectively targets human breast cancer stem cells in vitro and inxenografts. J Clin Invest 2010;120:485–97.

17. Conley SJ, Gheordunescu E, Kakarala P, Newman B, Korkaya H, Heath AN,et al. Antiangiogenic agents increase breast cancer stem cells via thegeneration of tumor hypoxia. Proc Natl Acad Sci U S A 2012;109:2784–9.

18. Honeth G, Bendahl PO, Ringner M, Saal LH, Gruvberger-Saal SK, LovgrenK, et al. The CD44þ/CD24� phenotype is enriched in basal-like breasttumors. Breast Cancer Res 2008;10:R53.

19. Sarrio D, Rodriguez-Pinilla SM, Hardisson D, Cano A, Moreno-Bueno G,Palacios J. Epithelial-mesenchymal transition in breast cancer relates to thebasal-like phenotype. Cancer Res 2008;68:989–97.

20. Creighton CJ, Li X, Landis M, Dixon JM, Neumeister VM, Sjolund A, et al.Residual breast cancers after conventional therapy displaymesenchymal aswell as tumor-initiating features. Proc Natl Acad Sci U S A 2009;106:13820–5.

21. Papageorgis P, Lambert AW, Ozturk S, Gao F, Pan H, Manne U, et al. Smadsignaling is required to maintain epigenetic silencing during breast cancerprogression. Cancer Res 2010;70:968–78.

22. Neve RM,Chin K, Fridlyand J, Yeh J, Baehner FL, Fevr T, et al. A collection ofbreast cancer cell lines for the study of functionally distinct cancer subtypes.Cancer Cell 2006;10:515–27.

23. Santner SJ, Dawson PJ, Tait L, Soule HD, Eliason J, Mohamed AN, et al.Malignant MCF10CA1 cell lines derived from premalignant humanbreast epithelial MCF10AT cells. Breast Cancer Res Treat 2001;65:101–10.

24. Strickland LB, Dawson PJ, Santner SJ, Miller FR. Progression of premalig-nant MCF10AT generates heterogeneous malignant variants with charac-teristic histologic types and immunohistochemical markers. Breast CancerRes Treat 2000;64:235–40.

25. Oshima A, Tanabe H, Yan T, Lowe GN, Glackin CA, Kudo A. A novelmechanism for the regulation of osteoblast differentiation: transcription ofperiostin, a member of the fasciclin I family, is regulated by the bHLHtranscription factor, twist. J Cell Biochem 2002;86:792–804.

26. Horiuchi K, Amizuka N, Takeshita S, Takamatsu H, Katsuura M, Ozawa H,et al. Identification and characterization of a novel protein, periostin, withrestricted expression to periosteum and periodontal ligament andincreased expression by transforming growth factor beta. J Bone MinerRes 1999;14:1239–49.

27. Ruan K, Bao S, Ouyang G. The multifaceted role of periostin in tumori-genesis. Cell Mol Life Sci 2009;66:2219–30.

28. Charafe-Jauffret E, Ginestier C, Iovino F, Wicinski J, Cervera N, Finetti P,et al. Breast cancer cell lines contain functional cancer stem cells withmetastatic capacity and a distinctmolecular signature. Cancer Res 2009;69:1302–13.

29. King WG, Mattaliano MD, Chan TO, Tsichlis PN, Brugge JS. Phosphati-dylinositol 3-kinase is required for integrin-stimulated AKT and Raf-1/mitogen-activated protein kinase pathway activation. Mol Cell Biol1997;17:4406–18.

30. Masuoka M, Shiraishi H, Ohta S, Suzuki S, Arima K, Aoki S, et al. Periostinpromotes chronic allergic inflammation in response to Th2 cytokines.J Clin Invest 2012;122:2590–600.

31. Prat A, Perou CM. Mammary development meets cancer genomics. NatMed 2009;15:842–4.

32. Gyorffy B, Lanczky A, Eklund AC, Denkert C, Budczies J, Li Q, et al. Anonline survival analysis tool to rapidly assess the effect of 22,277 genes onbreast cancer prognosis using microarray data of 1,809 patients. BreastCancer Res Treat 2010;123:725–31.

33. Shao R, Bao S, Bai X, Blanchette C, Anderson RM, Dang T, et al. Acquiredexpression of periostin by human breast cancers promotes tumor angio-genesis through up-regulation of vascular endothelial growth factor recep-tor 2 expression. Mol Cell Biol 2004;24:3992–4003.

34. Weidner N, Semple JP, Welch WR, Folkman J. Tumor angiogenesis andmetastasis–correlation in invasive breast carcinoma. N Engl J Med1991;324:1–8.

35. Xu D, Xu H, Ren Y, Liu C, Wang X, Zhang H, et al. Cancer stem cell-relatedgene periostin: a novel prognostic marker for breast cancer. PLoS ONE2012;7:e46670.

36. Sabatier R, Finetti P, Cervera N, Lambaudie E, Esterni B, Mamessier E, et al.A gene expression signature identifies two prognostic subgroups of basalbreast cancer. Breast Cancer Res Treat 2011;126:407–20.

37. Hartman ZC, PoageGM, denHollander P, Tsimelzon A, Hill J, PanupinthuN, et al. Growth of triple-negative breast cancer cells relies upon coordinateautocrine expression of the proinflammatory cytokines IL-6 and IL-8.Cancer Res 2013;73:3470–80.

38. Gupta PB, Onder TT, Jiang G, Tao K, Kuperwasser C, Weinberg RA, et al.Identification of selective inhibitors of cancer stem cells by high-through-put screening. Cell 2009;138:645–59.

39. Bornstein P, Sage EH. Matricellular proteins: extracellular modulators ofcell function. Curr Opin Cell Biol 2002;14:608–16.


Lambert et al.




40. Lim E, Vaillant F, WuD, Forrest NC, Pal B, Hart AH, et al. Aberrant luminalprogenitors as the candidate target population for basal tumor develop-ment in BRCA1 mutation carriers. Nat Med 2009;15:907–13.

41. Molyneux G, Geyer FC, Magnay FA, McCarthy A, Kendrick H, NatrajanR, et al. BRCA1 basal-like breast cancers originate from luminal epi-thelial progenitors and not from basal stem cells. Cell Stem Cell2010;7:403–17.

42. Proia TA, Keller PJ, Gupta PB, Klebba I, Jones AD, Sedic M, et al. Geneticpredisposition directs breast cancer phenotype by dictating progenitor cellfate. Cell Stem Cell 2011;8:149–63.

43. Asselin-Labat ML, Sutherland KD, Barker H, Thomas R, Shackleton M,Forrest NC, et al. Gata-3 is an essential regulator of mammary-glandmorphogenesis and luminal-cell differentiation. Nat Cell Biol 2007;9:201–9.

44. Herschkowitz JI, Zhao W, Zhang M, Usary J, Murrow G, Edwards D,et al. Comparative oncogenomics identifies breast tumors enriched infunctional tumor-initiating cells. Proc Natl Acad Sci U S A 2012;109:2778–83.

45. Prat A, Parker JS, Karginova O, Fan C, Livasy C, Herschkowitz JI, et al.Phenotypic and molecular characterization of the claudin-low intrinsicsubtype of breast cancer. Breast Cancer Res 2010;12:R68.

46. Wang X, Liu J, Wang Z, Huang Y, Liu W, Zhu X, et al. Periostincontributes to the acquisition of multipotent stem cell-like propertiesin human mammary epithelial cells and breast cancer cells. PLoS ONE2013;8:e72962.

47. Yu F, Deng H, Yao H, Liu Q, Su F, Song E. Mir-30 reductionmaintains self-renewal and inhibits apoptosis in breast tumor-initiating cells. Oncogene2010;29:4194–204.

48. Malanchi I, Santamaria-Martinez A, Susanto E, Peng H, Lehr HA, DelaloyeJF, et al. Interactions between cancer stem cells and their niche governmetastatic colonization. Nature 2012;481:85–9.

49. Balko JM, Schwarz LJ, Bhola NE, Kurupi R, Owens P, Miller TW, et al.Activation of MAPK pathways due to DUSP4 loss promotes cancer stemcell-like phenotypes in basal-like breast cancer. Cancer Res 2013;73:6346–58.

50. Iliopoulos D, Hirsch HA, Struhl K. An epigenetic switch involving NF-kappaB, Lin28, Let-7 MicroRNA, and IL6 links inflammation to celltransformation. Cell 2009;139:693–706.

51. Gupta PB, Fillmore CM, Jiang G, Shapira SD, Tao K, Kuperwasser C, et al.Stochastic state transitions give rise to phenotypic equilibrium in popula-tions of cancer cells. Cell 2011;146:633–44.

52. Ghajar CM, Peinado H, Mori H, Matei IR, Evason KJ, Brazier H, et al. Theperivascular niche regulates breast tumour dormancy. Nat Cell Biol2013;15:807–17.

53. Zhou W, Ke SQ, Huang Z, Flavahan W, Fang X, Paul J, et al. Periostinsecreted by glioblastoma stem cells recruits M2 tumor-associated macro-phages and promotes malignant grwoth. Nat Cell Biol 2015;17:170–82.

54. Seguin L, Kato S, Franovic A, CamargoMF, Lesperance J, Elliott KC, et al. Anintegrin beta(3)-KRAS-RalB complex drives tumour stemness and resis-tance to EGFR inhibition. Nat Cell Biol 2014;16:457–68.






2016;14:103-113. Published OnlineFirst October 27, 2015.Mol Cancer Res Arthur W. Lambert, Chen Khuan Wong, Sait Ozturk, et al. Breast Cancer Stem CellsTumor Cell-Derived Periostin Regulates Cytokines That Maintain

Updated version

10.1158/1541-7786.MCR-15-0079doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2015/11/04/1541-7786.MCR-15-0079.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/14/1/103.full#ref-list-1

This article cites 54 articles, 15 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/14/1/103.full#related-urls

This article has been cited by 4 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/14/1/103To 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-15-0079

http://mcr.aacrjournals.org/content/suppl/2015/11/04/1541-7786.MCR-15-0079.DC1

http://mcr.aacrjournals.org/content/14/1/103.full#ref-list-1

http://mcr.aacrjournals.org/content/14/1/103.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/14/1/103


tumorcell-derivedperiostinregulatescytokines that maintain ...hamid m. abdolmaleky1, and sam...

Documents