Breast Cancer Stem Cells Are Regulated by Mesenchymal Stem ... › content › canres › ... · Tumor and Stem Cell Biology Breast Cancer Stem Cells Are Regulated by Mesenchymal

Tumor and Stem Cell Biology

Breast Cancer Stem Cells Are Regulated by MesenchymalStem Cells through Cytokine Networks

Suling Liu1, Christophe Ginestier3, Sing J. Ou1, Shawn G. Clouthier1, Shivani H. Patel1,Florence Monville3, Hasan Korkaya1, Amber Heath1, Julie Dutcher1, Celina G. Kleer1,Younghun Jung2, Gabriela Dontu1, Russell Taichman2, and Max S. Wicha1

AbstractWe have used in vitro and mouse xenograft models to examine the interaction between breast cancer stem

cells (CSC) and bone marrow–derived mesenchymal stem cells (MSC). We show that both of these cellpopulations are organized in a cellular hierarchy in which primitive aldehyde dehydrogenase expressingmesenchymal cells regulate breast CSCs through cytokine loops involving IL6 and CXCL7. In NOD/SCID mice,labeled MSCs introduced into the tibia traffic to sites of growing breast tumor xenografts where they acceleratedtumor growth by increasing the breast CSC population. With immunochemistry, we identified MSC–CSC nichesin these tumor xenografts as well as in frozen sections from primary human breast cancers. Bone marrow–derived MSCs may accelerate human breast tumor growth by generating cytokine networks that regulate theCSC population. Cancer Res; 71(2); 614–24. �2011 AACR.

Introduction

Many human cancers, including breast cancer, may bedriven by a population of cells that display stem cell proper-ties. These properties include self-renewal, which drivestumorigenesis, and differentiation, which contributes to can-cer cell heterogeneity. There is increasing evidence that thesecancer stem cells (CSC) mediate tumor metastasis and, byvirtue of their relative resistance to chemotherapy and radia-tion therapy, may contribute to treatment resistance andrelapse following therapy (1).

Self-renewal and cell fate determination of normal stemcells are regulated by both cell intrinsic and cell extrinsicpathways. The dysregulation of these pathways resulting instem cell expansion may be a key event initiating carcino-genesis. Developmental pathways such as Notch, Hedgehog,and Wnt play an important role in normal stem cellfunction and are frequently deranged in cancers (2–5).Extrinsic signals that regulate stem cell behavior originatein the stem cell microenvironment or niche. This niche

contains extracellular components as well as multiple celltypes.

Although there is little information on the composition andfunction of CSC niches, it is clear that tumor growth andmetastasis is highly dependent on the tumor microenviron-ment. This microenvironment is composed of tumor-asso-ciated fibroblasts, endothelial cells, adipocytes, and immunecells, all of which have been shown to play a role in tumorgrowth and metastasis (6). Mesenchymal stem cells (MSC),which can be defined as multipotent mesenchymal stromalcells, are a heterogeneous subset of stromal stem cells that canbe isolated from many adult tissues; proliferate as adherentcells; have fibroblast-like morphology; form colonies in vitro;and can differentiate into adipocytes, osteocytes, and chon-drocytes (7). Recently, through the use of mouse breast cancermodels, it has been shown that bone marrow–derived MSCsmay be recruited to the sites of developing tumors, thus,influencing their metastatic potential (8). It has been shownthat MSCs can produce IL6 (9, 10) and can stimulate tumorgrowth through the paracrine production of secreted IL6 (11).Both IL6 and IL8 have been implicated in the regulation ofCSCs (12, 13).

We have previously shown that both normal and malignantmammary stem cells can be isolated by virtue of theirincreased expression of aldehyde dehydrogenase (ALDH), asassessed by the ALDEFLUOR assay. We have used this methodto isolate functional stem cells from primary breast xenograftsand established human breast cancer cell (BCC) lines andshown that these cells mediate tumor invasion and metastasis(14). In the present study, we examined the interactionbetween bone marrow–derived MSCs and CSCs with in vitrosystems and mouse models. We show that mesenchymal cells(MC), like CSCs, are organized in a cellular hierarchy andthat ALDEFLUOR-positive MCs regulate CSC self-renewal.

Authors' Affiliations: 1Comprehensive Cancer Center and 2School ofDentistry, University of Michigan, Ann Arbor, Michigan; and 3Centre deRecherche en Cancerologie de Marseille, Laboratoire d’Oncologie Mole-culaire, UMR891 Inserm/Institut Paoli-Calmettes, Marseille, France

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Corresponding Author: Max S. Wicha 1500 E. Medical Center Dr., 6303Cancer Center, Ann Arbor, MI 48109. Phone: 734-936-1831; Fax: 734-615-3947; E-mail: [email protected] or Suling Liu, 1500 E. MedicalCenter Dr., 7110 Cancer Center, Ann Arbor, MI 48109. Phone: 734-615-4651; Fax: 734-647-9480; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-10-0538

�2011 American Association for Cancer Research.

CancerResearch

Cancer Res; 71(2) January 15, 2011614

Research. on June 14, 2020. © 2011 American Association for Cancercancerres.aacrjournals.org Downloaded from

Published OnlineFirst January 11, 2011; DOI: 10.1158/0008-5472.CAN-10-0538





http://cancerres.aacrjournals.org/



Interaction between these cell types is mediated by a cytokinenetwork involving CXCL7 and IL6. Furthermore, we show thatlabeled human bone marrow MCs traffic from the bonemarrow to accelerate the growth of human breast cancerxenografts at distant sites by expanding the CSC population.These studies suggest that MSCs form an important compo-nent of the CSC niche in that they regulate the self-renewal ofbreast CSCs.

Materials and Methods

Cell cultureBCC lines (SUM159 and SUM149) obtained fromDr. Stephen

Ethier have been extensively characterized (http://www.aster-and.com/Asterand/human_tissues/hubrcelllines.htm;15). TheMCF-7 cell line was purchased from American Type CultureCollection. The cell lines were grown using the recom-mended culture conditions as described previously (14).Human bone marrow–derived MCs, which were cryopre-served at passage 1, were purchased from ScienCell ResearchLaboratories and grown and passaged in the recommendedmedium. These MCs were characterized by the expression ofthe MC markers CD29, CD90, CD44, and CD105 but notCD45, CD34, and CD11b at passages 2 (SupplementaryFig. S1A) and 10 (Supplementary Fig. S1B). All experimentswere done with subconfluent cells in the exponential phaseof growth. The primary MCs from bone marrow werepurchased from Texas A&M HSC COM and cultured inthe recommended culture medium.

Virus infectionA highly efficient lentiviral expression system (pLentiLox

3.7; http://www.med.umich.edu/vcore/) was used to generateDsRed-, GFP-, or luciferase-expressing lentiviruses in the UMVector Core Facility. The cell lines were infected with thelentiviruses as described previously (3).

ALDEFLUOR assay and flow cytometryThe ALDEFLUOR kit (StemCell Technologies Inc.) was used

as described previously (16). The antibodies for IL6R andgp130 were purchased from Immunotech and Pharmingen,and flow cytometry with antibody staining was done asdescribed previously (16). To assess cell viability, 0.5 mg/mL40,6-diamidino-2-phenylindole (DAPI; Sigma) was used.

In vitro differentiationIn vitro differentiation of hMSC-bm (unseparated popula-

tion, ALDEFLUOR-positive population, and ALDEFLUOR-negative population) was evaluated in triplicates, and thedetailed procedure for the differentiation assay is describedin the Supplementary Material.

RNA extractionTotal RNA was isolated using the RNeasy Micro Kit accord-

ing to the manufacturer's instructions (Qiagen).

Gene expression profiling with DNA microarraysAffymetrix human U133 Plus 2.0 was used. Preparation of

the cRNA, hybridizations, washes, and detection were done as

recommended by the supplier (http://www.affymetrix.com/index.affx). Expression data were analyzed by the RMA(Robust Multichip Average) method in R with Bioconductorand associated packages (17).

Real-time quantitative RT-PCROne nanogram of total RNA from the mammospheres or

differentiated cells on collagen-coated plates was used forreal-time quantitative RT-PCR (qRT-PCR), as described pre-viously (18).

Conditioned medium, antibody array, and Luminexbead assay

To prepare conditioned media, BCC lines (SUM159 orSUM149) alone, MSC alone, or the coculture of BCC andMSC (1:1 mixture) were plated in 100-mm tissue culture dishesin the mixture of IHM and MSCM (1:1 mixture). The detailedprocedure is described in the Supplementary Material.

Invasion assayAssays were done in triplicate in invasion chambers pre-

coated with reduced growth factor matrix from BD Bios-ciences. Cells were added to the upper chamber in 200 mLof serum-free medium. For the invasion assay, 20,000 MCswere seeded on the coated chamber, and the lower chamberwas filled with 600 mL of medium (Cambrex) with or without100 ng/mL IL6; 100 mg/mL IL6 blocking antibody; and 20,000preseeded SUM159, SUM149, or MCF-7 cells. After 27 hours ofincubation, the cells on the underside of the upper chamberswere stained with the blue stain in the Cell Invasion Assay Kit(Chemicon; cat. #ECM550) and counted using light micro-scopy.

Tumorigenicity in NOD/SCID miceSix-week-old female NOD/SCID mice were purchased from

Jackson Laboratories. Tumorigenicity of 1,000 of ALDEL-FUOR-positive, -negative, and unseparated SUM159-DsRedin the absence or presence of an equal number of MSCs inthe mammary fat pads of NOD/SCID mice was assessed. Sixmice were assayed for each group. The animals were eutha-nized when the tumors were approximately 1–1.5 cm at theirlargest diameter, in compliance with regulations for the use ofvertebrate animals in research. A portion of each fat pad wasfixed in formalin and embedded in paraffin for histologicanalysis. Another portion was analyzed by the ALDEFLUORassay. In addition to the established cell lines, MC1 primaryxenografts were developed and used as previously described(16).

Intratibial injectionThe mouse preparation and the intramedullary injection

into the tibia shaft was carried out according to previouslypublished methods (19, 20). Briefly, a few hours before trans-plantation, mice were irradiated with 300 cGy from an X-rayirradiator (Mark I, Model 25; J.L. Shepherd). All procedureswere approved by the Animal Care Committee of the Uni-versity of Michigan. The detailed procedure is described in theSupplementary Material.

MSCs Regulate Breast Cancer Stem Cells

www.aacrjournals.org Cancer Res; 71(2) January 15, 2011 615




Bioluminescence imagingBaseline bioluminescence was assessed before inoculation

and each week thereafter. For photon flux counting, we used acharge-coupled device camera system (Xenogen) with a nose-cone isofluorane delivery system and heated stage for main-taining body temperature. Results were analyzed after 2–12minutes of exposure using the Living Image software providedwith the Xenogen imaging system.

ImmunostainingFor ALDH1, DsRed, GFP, and DAPI quadruple fluorescent

staining, paraffin-embedded sections of breast tumors fromxenografts were deparaffinized in xylene and rehydrated ingraded alcohol. Antigen enhancement was done by incubatingthe sections in citrate buffer pH 6.0 (Dakocytomation) asrecommended. ALDH1 antibody (BD biosciences), DsRedantibody (Santa Cruz), and GFP antibody (Neomarker) wereused at a 1:25 dilution and incubated for 1 hour. PE-, FITC-,and Cy5-labeled secondary antibodies (Jackson Labs) wereused at a 1:250 dilution and incubated for 20 minutes. PE-conjugated secondary antibody (red color in the staining) wasused to detect the ALDH1 primary antibody, FITC-conjugatedsecondary antibody (green color in the staining) was used todetect the DsRed primary antibody, and Cy5-conjugated sec-ondary antibody (purple color in the staining) was used todetect the GFP primary antibody. Nuclei were counterstainedwith DAPI/antifade (Invitrogen; blue color in the staining) andcover-slipped. Sections were examined with a fluorescentmicroscope (Olympus FV-500 Confocal).

Statistical analysisResults are presented as the mean � standard deviation

(STDEV) for at least 3 repeated individual experiments foreach group. Mean and STDEV was determined on the basis ofan analysis of at least 3 replicates using Microsoft Excel.Statistical differences were determined by using ANOVAand Student's t test for independent samples. A value of P< 0.05 was considered statistically significant.

Results

Bone marrow–derived MSCs can expand the breastcancer stem cell population

To assess the ability of bone marrow MSCs to affect breastCSC functionality, we cocultured DsRed-labeled SUM159BCCs with human bone marrow–derived MCs. Followingcoculture, cell populations were separated by flow cytometry,and ALDH-expressing populations were assessed by theALDEFLUOR assay. SUM159 cells cultured alone containedapproximately 4% ALDEFLUOR-positive cells. Coculture withMCs increased the proportion of ALDEFLUOR-positive cellsover 3-fold to 14% without affecting the total cell numbers asdetermined by MTT assay (Fig. 1A, right), suggesting that thisinteraction leads to an increase in CSC self-renewal (Fig. 1A).To find whether this increase required contact betweencancer cells and MCs, each of these cell types were culturedin transwells, which precluded direct cell–cell contact butallowed communication via soluble factors. As shown in

Figure 1A, transwell culture recapitulated the effects of directcell–cell contact. Furthermore, expansion of the CSC popula-tion was reproduced by the addition of conditioned mediumobtained from either coculture or transwell culture of both cellcompartments but not by conditioned medium obtained froma culture of MCs alone (Fig. 1A). Similar results were seen in 2other BCC lines, SUM149 and MCF-7, representing basal andluminal subtypes (Supplementary Fig. S2B and C). The cocul-ture of tumor cells with MCs also increased the percentage oftumor cells expressing the breast CSC markers CD24�CD44þ

(ref. 21; Supplementary Fig. S2). This suggests that the CSCcompartment is regulated by soluble factors that are gener-ated as a result of the interaction between MCs and cancercells.

Mesenchymal cells are organized in a cellular hierarchyin which ALDEFLUOR-positive MCs regulate CSCexpansion

To determine whether MCs are organized in a cellularhierarchy, we used the ALDEFLUOR assay to isolate ALDH-expressing subpopulations from human bone marrow–derived MCs. As indicated in Figure 1B and SupplementaryFigure S1, MCs contain approximately 5%–6% ALDEFLUOR-positive cells. In both coculture and transwell culture, thisALDEFLUOR-positive subfraction is unaffected by the pre-sence of SUM159 cells (Fig. 1B). Defining characteristics ofstem cells include their ability to self-renew and to undergomultilineage differentiation. When placed in an appropriateinduction medium, ALDEFLUOR-positive MCs (mesenchymalstem cells) displayed adipogenic or osteogenic differentiation,whereas ALDEFLUOR-negative MCs did not (Fig. 1C). Further-more, we found that ALDEFLUOR-positive MCs can regener-ate both ALDEFLUOR-positive and ALDEFLUOR-negativeMCs, but ALDEFLUOR-negative MCs cannot regenerateALDEFLUOR-positive MCs (Supplementary Fig. S3A)—a prop-erty that also applies to the ALDEFLUOR-positive and -nega-tive BCCs (Supplementary Fig. S3B). This indicates that MCsare organized in a hierarchy in which ALDEFLUOR-positiveMCs can undergo self-renewal and multilineage differentia-tion. To find whether the differentiation of MCs affected theirability to regulate breast CSCs, we determined the effect ofALDEFLUOR-positive and ALDEFLUOR-negative MCs onSUM159 cells in coculture. As shown in Figure 1D, the effectof MCs on the SUM159 CSC population was mediated by theALDEFLUOR-positive MC population whereas ALDEFLUOR-negative MCs had no effect. These results were confirmed byfreshly isolated and well-characterized MCs from bone mar-row (Supplementary Fig. S4). These experiments show thatinteractions between MSC and CSC populations regulate theproportion of CSCs.

A cytokine network mediates interaction between BCCsand MCs

To determine the molecular mechanisms mediating theinteraction between BCCs and MCs, we examined the effect ofcoculture on the global gene expression profile of each cellpopulation, as assessed by an Affymetrix microarray. Wecompared gene expression patterns of SUM159 cells and

Liu et al.

Cancer Res; 71(2) January 15, 2011 Cancer Research616




MCs cultured alone or in coculture conditions. Among thegene families induced by coculture were cytokine genes. Asshown in Supplementary Table 1A, coculture induced mRNAexpression of CXCL5 (ENA78), CXCL6 (GCP2), and CXCL1(Gro-a), as well as IL6 and IL8 in both the SUM159 cells and inMCs. These results were confirmed by a real-time RT-PCR inthe SUM159 cell line (Supplementary Table 1B) as well as inthe primary xenograft MC1 (Supplementary Fig. S15). Inaddition, we detected a 6-fold increase in CXCL7 (NAP2)expression only in MCs induced by coculture with BCCs. Bothantibody arrays and the Luminex bead assay were used toquantitatively assess the effects of coculture on cytokineprotein expression. As shown in Figure 2 and SupplementaryFigure S5A, the levels of CXCL1, CXCL5, CXCL6, and CXCL7were all substantially increased by coculture. In addition,coculture substantially increased IL6 and IL8 production.Similar results were seen in other BCC lines: SUM149 (Sup-plementary Figs. S5B and S6) and MCF-7 (SupplementaryFigure S5C). To determine which of the cytokines inducedby coculture was responsible for affecting the CSC population,

we used the ALDEFLUOR assay to assess the effect of each ofthese cytokines on the proportion of CSCs. Dose–responsecurves were obtained for each cytokine (SupplementaryFig. S7). The effect of optimized concentrations of eachcytokine on the ALDEFLUOR-positive population ofSUM159 cells is shown in Figure 3A. IL6 and IL8 inducedsubstantial increases in the ALDEFLUOR-positive population,consistent with previous reports (14, 24, 25). CXCL1 had nosignificant effect, and CXCL5 and CXCL6 generated smallerbut statistically significant increases in ALDEFLUOR-positiveSUM159 cells. In contrast, the addition of CXCL7 producedalmost a 3-fold increase in ALDEFLUOR-positive SUM159cells—a level comparable with that produced by coculturewith MCs. Similar results were seen in other BCC lines:SUM149 (Supplementary Fig. S8A) and MCF-7 (Supplemen-tary Fig. S8C).

To determine the contribution of these cytokines inmediating the interaction of MSCs with CSCs, we usedcytokine-blocking antibodies. Dose–response curves for anti-body-blocking were obtained (Supplementary Fig. S7). The

0

1

2

3

4

5

6

MSC alone MC-DsRed co-cultured with

ALDH+ SUM159

MC-DsRed cocultured withALDH- SUM159

ALD

EFLU

OR

-pos

itive

MC

(%)

02468

10121416

Coculture C.M. from MC

C.M. fromcoculture

C.M. fromtranswell-

culture

AL

DE

FL

UO

R-p

osi

tive

SU

M15

9 ce

lls (

%)

Alone Transwell-culture

P < 10-6 P < 10-6 P < 10-6

P < 10-6

B

Ost

eog

enic

in

du

ctio

nA

dip

og

enic

in

du

ctio

nC

on

tro

l

ALDH+ ALDH-

D

C

Coculture

02468

1012141618

SUM159-DsRedalone

SUM159-DsRedCo-cultured w.

ALDH MC+

SUM159-DsRedCo-cultured w.

ALDH MC -

AL

DE

FL

UO

R-p

osi

tive

SU

M15

9 ce

lls (

%)

P < 10-7

A

Co

cult

ure

C.M

. fro

mM

C

C.M

. fro

mco

cult

ure

C.M

. fro

mtr

answ

ell

-cu

ltu

re

Alo

ne

3

Day 1Day 3Day 5

Day 7

SUM159

2

1

0

Cel

l via

bili

ty (

No

rmal

ized

w

ith

co

ntr

ol)

Tra

nsw

ell

-cu

ltu

re

Figure 1. ALDEFLUOR-positive MCs regulate breast CSC populations and are capable of osteogenic and adipogenic differentiation. A, the ALDEFLUOR-positive population of SUM159 is increased by coculture with MCs, transwell culture with MCs, or culture with a conditioned medium (CM) derived fromeither coculture or transwell culture but not by culture with a CM derived from MCs alone. The total cell number of SUM159 cells remains constantunder all of the treatment conditions. B, the ALDEFLUOR-positive population of MCs is approximately 5% and is not changed by coculture with eitherALDEFLUOR-positive or ALDEFLUOR-negative SUM159 cells. C, the ALDEFLUOR-positive but not the ALDEFLUOR-negative population of MCs iscapable of osteogenic and adipogenic differentiation. D, ALDEFLUOR-positive MCs increase the proportion of ALDEFLUOR-positive SUM159 cells.The P values refer to the significant difference of sample groups compared with the "alone" group. Error bar, SD.






concentration of antibody producing maximal inhibitionwas used and are shown in Figure 3B. The interaction ofCSCs with MSCs was partially inhibited by anti-IL6 and, morecompletely, by anti-CXCL7, suggesting an important role forthese 2 cytokines in mediating MSC–CSC interactions. Similarresults were seen in other BCC lines: SUM149 (SupplementaryFig. S7B) and MCF-7 (Supplementary Fig. S8D).

To further define the role of CXCL7 and IL6 in mediating theinteraction of MSCs with CSCs, we examined the effects ofCXCL7 and IL6 on cytokine production by each cellularsubcomponent. The addition of CXCL7 to SUM159 cellsinduced a cytokine expression pattern similar to that observedin coculture (Fig. 3C). Furthermore, the addition of CXCL7-blocking antibody to the coculture completely blocked theexpression of induced cytokines CXCL1, CXCL5, CXCL6, IL6,and IL8 (Fig. 3C). These experiments suggest that cytokineproduction by SUM159 cells in coculture is largely due toCXCL7 produced by MSCs. Similar results were seen inanother BCC line: SUM149 (Supplementary Fig. S9).

CXCL7 production by MSCs is regulated by IL6The previous experiments suggest a critical role for MSC-

derived CXCL7 in mediating signaling between the epithelialand mesenchymal components of tumors. To find how thiscytokine was regulated, we examined the effect of individualcytokines on CXCL7 production by MSCs. As shown in Figure4A, the addition of IL6 induced a >10-fold increase in CXCL7production by MCs. The effects of IL6 are mediated by areceptor complex composed of the IL6 receptor IL6R andgp130. We assessed the expression of these receptors onALDEFLUOR-positive and ALDEFLUOR-negative MCs. Weseparated the ALDEFLUOR-positive and ALDEFLUOR-nega-tive MCs by flow cytometry, immunostained the separatedpopulations with IL6R and gp130 antibodies, and reanalyzedthe cells by flow cytometry. As indicated in Figure 4B, both theIL6R and gp130 were primarily expressed in ALDEFLUOR-positive MCs (MSCs). Furthermore, although the CXCL7

mRNA level was undetectable in the MC control group, theaddition of 100 ng/mL IL6 dramatically increased the expres-sion of CXCL7 mRNA in ALDEFLUOR-positive MCs (Fig. 4C).

IL6 mediates MSCs chemotaxisTo address the functional significance of increased IL6R

expression in MSCs, we examined whether this cytokinemediated the chemotaxis of these cells. Recombinant IL6mediated chemotaxis of ALDEFLUOR-positive but not ALDE-FLUOR-negative MCs, an effect inhibited by the IL6-blockingantibody (Fig. 4D). In addition, ALDEFLUOR-positive but notALDEFLUOR-negative MCs were chemotactic toward breasttumor cells SUM159, SUM149, and MCF-7, an effect that wassubstantially inhibited by the IL6-blocking antibody (Fig. 4D).These results suggest that the chemotaxis of MSCs towardBCCs is primarily mediated by IL6.

These experiments suggest the existence of a cytokinenetwork that mediates the interaction between MCs andcancer cells in which IL6 produced by cancer cells interactswith IL6R/gp130 expressed onMSCs, which produce CXCL7 inresponse to this IL6 stimulation. CXCL7, in turn, induces thesecretion of a number of cytokines from both SUM159 andMCs, including IL6, IL8, CXCL6, and CXCL5. All of thesecytokines are capable of expanding the ALDEFLUOR-positiveCSC population. Furthermore, increased IL6 interacts withMSCs, forming a positive feedback loop. This model of cyto-kine networks mediating the interaction between MCs, BCCs,and breast CSCs is illustrated in Figure 4E.

MSCs stimulate the growth of breast tumor xenograftsin NOD/SCID mice by affecting the CSC population

We have previously shown that ALDEFLUOR-positive butnot ALDEFLUOR-negative SUM159 cells are tumorigenic inNOD/SCID mice (14). To assess the contribution of differentsubpopulations of MCs on tumor growth, ALDEFLUOR-posi-tive, ALDEFLUOR-negative, and unsortedMCsmixed in a ratioof 1:1 with SUM159 cells were orthotopically implanted in

Figure 2. The induction ofcytokine protein by coculture ofMCs and SUM159 cells asdetermined by antibody array.Protein level of CXCL5 (ENA78),CXCL6 (GCP2), CXCL1 (Gro-a),CXCL7, IL6, and IL8 is increasedby the coculture of breast tumorcells and MCs. *, P < 0.05 refers tothe significant difference of theindividual cytokine from thecoculture group compared withthe corresponding cytokine fromthe "A þ B" group. Error bar, SD.

Liu et al.





A

B

C

Figure 3. CXCL7 and IL6 play key roles in mediating the interaction of MCs and SUM159 cells. A, recombinant cytokines increase the ALDEFLUOR-positivepopulation of SUM159 cells in the order of CXCL7 > IL6 > IL8 > CXCL6 > CXCL5. B, blocking antibodies decrease the ALDEFLUOR-positivepopulation of SUM159 cells in the coculture in the order of anti-CXCL7 > anti- IL6 > anti-IL8 > anti-CXCL6 > anti-CXCL5. NT, "no treatments." TheP values refer to the significant difference of sample groups compared with the NT group. C, the addition of CXCL7 to SUM159 cells induces acytokine expression pattern, which recapitulated that of the coculture. Alternatively, the addition of CXCL7 blocking antibody to the coculture completelyblocked the expression of induced cytokines. *, P < 0.05 refers to the significant difference of the individual cytokine from the "coculture treatedwith 20 mg/ml anti-CXCL7" group compared with the corresponding cytokine from the coculture group or the significant difference of the individual cytokinefrom the "SUM159 treated with 10 ng/mL CXCL7" group compared with the corresponding cytokine from "SUM159-alone" group. Error bar, SD.






NOD/SCID mice. Unsorted MCs alone were implanted ascontrol. The addition of ALDEFLUOR-positive MCs greatlyaccelerated tumor growth, whereas ALDEFLUOR-negativeMCs had no effect on the growth of SUM159 tumors. UnsortedMCs were intermediate in their ability to stimulate tumorgrowth (Fig. 5A and Supplementary Fig. S11A). MCs alone werenot tumorigenic (Fig. 5A). The introduction of the GFP labelintoMCs and theDsRed label into SUM159 cells allowed for the

separation of these cells from established tumors and theassessment of the effects of added MCs on the CSC populationin vivo. The introduction of MSCs increased the proportion ofALDEFLUOR-positive SUM159 cells almost 4-fold comparedwith tumors grown from SUM159 cells alone. This increase inALDEFLUOR-positive SUM159 cells was seen in tumors grownin the presence of ALDEFLUOR-positive (ALDH-positive) butnot ALDEFLUOR-negative (ALDH-negative) MCs (Fig. 5B and

0

0.0005

0.001

0.0015

0.002

0.0025

Control 100 ng/mL IL6

Treatments

Rel

ativ

e C

XC

L7

mR

NA

leve

lIL

6R-p

osi

tive

MC

(%

)

ALDH+

ALDH-

0

2

4

6

8

10

12

No

rmal

ized

CX

CL

7 p

rote

in le

vel (

%)

N.T. 2 ug/mlCXCL1

1 ug/mlCXCL5

10 ng/mlCXCL6

100 ng/mlIL8

100 ng/mlIL6

Mesenchymal cells

P < 10-4

40

30

20

10

00

1

2

3

4

IL6R gp130

ALDH-

ALDH+

gp

130-po

sitive MC

(%)

Mesenchymal cells

P < 10-4P < 10-4

0

100

200

300

400

500

600

700

800

Control

100 n

g/mL IL

6

SUM159

SUM149

MCF-7

MCF-7 +

100µ

g/mL an

ti-IL

6

SUM149 +

100µ

g/mL an

ti-IL

6

SUM159 +

100µ

g/mL an

ti-IL

6

100 n

g/mL IL

6 + 10

0µg/m

L anti-

IL6

Bottom chamber (SFM)

Nu

mb

er o

f m

igra

tin

g c

ells

/ 10

000

cells

(27

h)

ALDH+ MC

ALDH- MC

A

B

C

D

E

P < 10-4

P < 10-3

P < 10-2

Bulk tumor cellsMesenchymal

CXCL7

IL6

CXCL5

CXCL1

CXCR1

CXCL6IL8IL6

CancerStem

renewal

IL6Rgp130

ALDH+

IL6R

ALDH-

ALDH-

ALDH- gp130

CXCR2

stem cell

cell self-

cells

Figure 4. IL6 regulates CXCL7 production by MSCs and mediates MSC chemotaxis, and the interaction between mesenchymal cells and tumorcells is regulated by a positive feedback loop including CXCL7 and IL6. A, the effects of the optimized concentration of cytokine on the induction ofCXCL7 byMCs. IL6 induces more than a 10-fold increase of CXCL7 production by MCs. NT, "no treatments." The P value refers to the significant difference ofthe sample group compared with the NT group. B, both the IL6R and GP130 were primarily expressed in the ALDEFLUOR-positive populations of MCs.The P values refer to the significant difference of the ALDHþ group compared with the ALDH� group. C, MCs were treated with vehicle control or 100 ng/mL ofIL6 for 4 days; both ALDEFLUOR-positive and ALDEFLUOR-negative cells were sorted out from both groups, and total RNA was isolated. Real-timeRT-PCR showed that CXCL7mRNA level was undetectable in the MC control group but elevated in ALDEFLUOR-positive MCs with IL6 treatment. TheP valuerefers to the significant difference of the ALDHþ cells compared with the ALDH� cells in the same group. D, ALDEFLUOR-positive MCs, but not ALDEFLUOR-negative MCs, have chemotactic capability, which is increased by IL6. Chemotaxis of ALDEFLUOR-positive MCs toward tumor cells is inhibitedby IL6 blocking antibody. The P values refer to the significant difference of sample groups compared with the control group. E, a model of cytokinenetworks mediating the interaction between MSCs, BCCs, and breast CSCs. Error bar, SD.

Liu et al.





Supplementary Fig. S11B). To provide functional data confirm-ing the ALDEFLUOR results, we examined the ability of serialdilutions of cells obtained from primary tumors to formsecondary tumors in NOD/SCID mice. As shown in Figure5D, cells obtained from primary tumors grown in the presenceof MCs had significantly greater tumor-generating capacity insecondary mice than cells obtained from primary tumorsgrown in the absence of MCs. These results confirm theALDEFLUOR data, suggesting that MSCs have the capacityto increase the breast CSC population. Furthermore, the per-centage of CSCs assessed by the ALDEFLUOR assay or by theexpression of CD24�CD44þ in the secondary tumors wassimilar in secondary tumors generated from MC-supplemen-ted or control primary tumors. This suggests that the accel-erated growth of primary tumors by MCs results from anincrease inCSCs rather than froman alteration of the biologicalproperties of these cells (Supplementary Fig. S11C). Similar invivo results were in other BCC lines: SUM149 (SupplementaryFig. S12) andMCF-7 (Supplementary Fig. S13). To show that the

effect of MSCs on breast CSCs was not limited to establishedcell line–generated xenografts, we used MC1, a breast xeno-graft established directly from a human breast tumor tissue.We have previously shown that ALDEFLUOR-positive cells inthis tumor display CSC properties (14). As was the case withestablished cell lines, the addition of MSCs accelerated tumorgrowth by increasing the proportion of ALDEFLUOR-positivetumor cells in this model (Supplementary Fig. S14).

To document physical interactions between MSCs andCSCs in growing tumors, we used 4-color fluorescence inwhich SUM159 cells were identified by DsRed expression(green), MCs by GFP expression (purple), and stem cells(MSCs and CSCs) by the expression of the stem cell markerby ALDH1 expression (red). Nuclei were identified by DAPIstaining (blue). We have previously reported that the ALDE-FLUOR-positive SUM159 cells can be identified in situ by usingan ALDH1 monoclonal antibody (14). Four-color fluorescencerevealed close apposition between ALDH1-positive MCs(MSCs) and ALDH1-positive SUM159 cells (CSCs; Fig. 5C).

1,000

1,200

1,400SUM159-DsRed ALDH+ alone (1k)

SUM159-DsRed ALDH+ + HMSC-bm Total (1k)

SUM159-DsRed ALDH+

A

3

4

5

P < 10-3

P < 10-3

B

*

** P < 0.05

200

400

600

800

Tum

or

volu

me

( (m

m3 )

+ HMSC-bm ALDH+ (1k)

SUM159-DsRed ALDH+ (1k)+ HMSC-bm ALDH–

HMSC-bm total alone (1k)

HMSC-bm ALDH+ alone (1k)

AL

DE

FL

UO

R-p

osi

tive

0

1

2

SU

M15

9 ce

lls (

%)

SUM159alone

SUM159 + MC total

SUM159 + ALDH– MC

SUM159 + ALDH+ MC

**

* *

**

0

847770635649423528211470

Days after injection

DAPI GPF (MC) ALDH1 MergeDsRed (SUM159)

alone HMSC-bm ALDH-

(1k)

C

250

300 1k SUM159-DsRed from alone

1k SUM159-DsRed from mix

10k SUM159-DsRed from alone

D*

SU

M15

9 9-

DsR

edal

on

e

150

200

or

volu

me

(mm

3 )


100k SUM159-DsRed from alone


*

* P < 0.05

SU

M15

9-D

sRed

+ M

C-G

FP

0

50

100

28211470

Tum *

*


Figure 5. ALDEFLUOR-positive MCs induce CSCs stimulating breast tumor growth in NOD/SCID mice. A, ALDEFLUOR-positive (ALDHþ) but notALDEFLUOR-negative (ALDH�) MCs stimulate tumor growth of SUM159 cells in NOD/SCID mice. *, P < 0.05 refers to the significant difference of the samplegroups compared with the "SUM159–DsRed ALDHþ alone (1k)" group. B, ALDEFLUOR-positive but not ALDEFLUOR-negative MCs increase theALDEFLUOR-positive population of SUM159 cells in NOD/SCID mice. The P values refer to the significant difference of sample groups compared with theSUM159-alone group. C, 4-color immunofluorescence staining reveals close apposition between ALDH1-positiveMCs and ALDH1-positive SUM159 cells. Toquantitate this association, we counted 9 fields to determine the frequency of association of ALDH-positive CSCs and MSCs. There are 14 CSC–MSCdirect-contacts out of a total of 38 MSCs and 35 CSCs. This statistically significant positive association shows the nonrandom distribution of these2 cell types in tumor xenografts. SUM159–DsRed: DsRed-labeled SUM159; MC-GFP: GFP-labeled MC; blue: DAPI staining; green: DsRed staining;purple: GFP staining; red: ALDH1 staining. White arrow, ALDH1þ SUM159; white triangle, ALDH1þ MC. D, serial dilutions of tumor cells isolated from primarytumors of SUM159-alone or SUM159 mixed with MCs were engrafted into secondary NOD/SCID mice. Tumor latency was reduced, and tumor growthwas increased in the tumors generated from primary tumors of SUM159 mixed with MCs. *, P < 0.05 refers to the significant difference of the "SUM159–DsRedfrom mix" compared with the "SUM159-DsRed from alone" group injected with the same number of cells. Error bar, SD.






MSCs traffic from bone marrow to primary breasttumor sites in xenografts and are detected in primaryhuman breast cancers

To determine whether bone marrow MSCs are capable oftrafficking to primary breast tumor sites, we labeled MCs withluciferase and DsRed. The growth of implanted MCs at thetibial site was shown by bioluminescence (Fig. 6A, left). Oneweek after the MC tibial implantation, we implanted SUM159cells in the mammary fat pads. Tumor size was monitoredweekly. As shown in Figure 6A (right), breast tumor growth

was significantly accelerated by human MCs introduced intothe mouse tibia. Five weeks after tumor implantation, animalswere sacrificed, and the presence of MSCs and CSCs wasassessed by immunochemistry and immunofluorescence. Asshown in Figure 6B, DsRed immunochemistry revealed thepresence of MCs in tumors grown in animals with tibial MCinoculation but not in control tumors grown in the absence ofMC introduction. To localize MSCs and CSCs in situ, we usedimmunofluorescence to identify MCs (green) and the stemcells marker ALDH1 (red). Merged images show adjacent

A3.957 x 1061.812 x 106 3.525 x 106

60

80

100

120

140

160

volu

me

(mm

3 )

Without MC intratibia injection

With MC intratibia injection

*

** P < 0.05

0

20

40

60

2 9 16 23 30

Tu

mo

r


*

BDsRed (MC) ALDH1DAPI Merge

Wit

ho

ut

MC

DsRed (MC)

Wit

h M

CIn

trat

ibia

inje

ctio

nIn

trat

ibia

inje

ctio

n

Ca b dc e

ALDH1 ALDH1 CD105 CD31 Pan-CK

700

600

500

400

x 103

300

200

100

Figure 6. MSCs traffic from bone marrow to primary breast tumor sites in xenografts and are detected in primary human breast cancers. A, MCslabeled with luciferase and DsRed were introduced via intratibial injection. The tumor growth was facilitated by introducing MCs into mouse tibia.*, P < 0.05 refers to the significant difference of the "with MC intratibial injection" group compared with the "without MC intratibial injection" group.B, in situ localization of ALDH1þ MCs and ALDH1þ SUM159 cells identifies a CSC niche. Light imaging reveals the presence of DsRed-labeled MC inmice with intratibial MC inoculation but not in controls. Brown, DsRed. Immunofluorescence detects DsRed-labeled MC (green), ALDH1-positive MC (yellow),and ALDH1-positive SUM159 cells (red) in the merged image. C, immunolocalization of ALDH1-positive tumor cells, MCs, and endothelial cells in breastcancer clinical specimens. We counted 6 fields to determine the frequency of association of ALDH-positive CSCs and MSCs. There are 21 CSC–MSC direct-contacts out of a total of 115 MSCs and 33 CSCs. Stain ALDH1 (a stem/progenitor marker), CD105 (a MSC marker and endothelial cell marker), CD31 (anendothelial cell marker), and Pan-CK AE1/AE3 (Pan-CK, an epithelial cell marker) in consecutive serial sections from a primary breast tumor. Onerepresentative sample from 3 independent samples is shown. Red arrow, CSCs (ALDH1þCD105�CD31�Pan-CKþ); yellow arrows, MSCs(ALDH1þCD105þCD31�Pan-CK�); green arrows, endothelial progenitors (ALDH1þCD105þCD31þ). Error bar, SD.

Liu et al.





ALDH1-positive MCs (yellow) and ALDH1-positive SUM159cells (red), suggesting the existence of a CSC niche character-ized by ALDH1-positive CSCs and ALDH1-positive MCs.To find whether similar MSC–CSC niches are present in

primary human breast cancers, we used immunochemistry toidentify ALDH1-positive MCs and ALDH1-positive cancer cellsin frozen sections of 3 independent ALDH1-positive primarybreast cancers. We stained serial sections for CD105 and CD31to identifyMCs and distinguish them fromendothelial cells, andwe identified the stemcells in each population, based onALDH1expression. Pan-cytokeratin (Pan-CK) was used as an epithelialcell marker. As shown in Figure 6C (1 representative samplefrom 6 independent samples is shown), ALDH1-positive MCs(ALDH1þCD105þCD31�Pan-CK�) are found in apposition toALDH1-positive cancer cells (ALDH1þCD105�CD31�Pan-CKþ). This histology closely resembles that seen in the mousexenografts.

Discussion

Stem cells are regulated by the interplay between extrinsicfactors and cell intrinsic regulatory pathways. During normaldevelopment and tissue homeostasis, these extrinsic factorsare provided by cellular and extracellular elements that definethe stem cell niche (24, 25). There is increasing evidence thatmany tumors, including breast cancer, may be driven by acellular subcomponent that displays stem cell properties.Although it is clear that the tumor microenvironment influ-ences tumor growth and metastasis (26), it is unclear whetherthese effects are mediated by CSCs. In this study, we usedin vitro systems and mouse models to show an important rolefor bonemarrow–derivedMSCs in regulating breast CSCs. Theuse of MCs and cancer cells, both of human origin, facilitatedthe study of cytokine interaction obviating known speciesdifferences in these factors. In fact, the significant facilitationof breast tumor growth by human MCs introduced into themouse tibia may reflect these species differences. We showthat the interaction between MSCs and CSCs is mediated by apositive feedback cytokine loop in which IL6 and CXCL7 playpivotal roles. This loop requires the simultaneous presence ofboth cell types but does not require cell–cell contact as shownby transwell and conditioned medium experiments. Further-more, we show that MCs, like cancer cells, are organized in ahierarchy in which primitive ALDH-expressing MCs capable ofself-renewal and multilineage differentiation interact withcancer cells to regulate CSC self-renewal. IL6 produced bycancer cells interacts with IL6R and gp130 on ALDEFLUOR-positive MCs. An IL6-mediated chemotaxis may facilitate thehoming of MSCs to the sites of primary tumor growth, as wellas induce CXCL7 production by these cells. MSC-derivedCXCL7, in turn, interacts with cancer cells through the CXCR2receptor (27), where it induces the synthesis of a number ofcytokines, including IL6 and IL8. These pleiotropic effects ofCXCL7 are consistent with previous reports (27). The expres-sion of CXCL7 and its receptor CXCR2 has been shown to beincreased in breast carcinomas (28). Furthermore, CXCL7transfection increased the invasive capacity of BCCs (28),consistent with the previously shown increased invasive

and metastatic properties of CSCs (29). We have previouslyshown that IL8 interacts with the CXCR1 receptor on CSCs,triggering their self-renewal and invasive properties (14, 30).IL6 has also been reported to be capable of regulating breaststem cells (22) and colon CSCs (23). In addition to regulatingCSCs, IL6 produced by cancer cells interacts with MSCs,further increasing their CXCL7 production, generating a posi-tive feedback loop (Fig. 5A). We confirmed the functionalimportance of these interactions by showing that MSCsaccelerate breast tumor growth in NOD/SCID mice. Further-more, as was the case in vitro, in this mouse model, this effectwas mediated by ALDEFLUOR-positive MCs, which werecapable of increasing the CSC population in vivo. The closeapposition of ALDH1þ tumor cells and MSCs was also shownin frozen sections of primary human breast cancers.

It is clear that the tumor microenvironment plays animportant role in tumor growth and metastasis (8). Previouselegant studies have suggested a role for MSCs in tumormetastasis, which is mediated by CCL5 (8). In addition,researchers reported that IL1a produced by MSCs mediatessimilar effects (31) and that MSCs promote LTC-IC expansion(32). Our studies extend these previous findings by demon-strating that MSCs regulate cancer cell behavior through theireffect on CSCs. On the other hand, MSCs have been shown toactually inhibit tumor growth in some models (33). Thissuggests that the effects of MSCs on tumor growth arecomplex and may be context-dependent.

The homing of bone marrow–derived MSCs to sites oftumor growth may rely on similar mechanisms for the homingof these cells to sites of tissue injury. Developing tumors mayrecruit MSCs from the bone marrow where they interact withand regenerate CSCs. If this is the case, then the developmentof strategies aimed at interfering with these pathways mayprovide a means of targeting CSCs. Since these cells maymediate tumor growth and metastasis as well as contribute totreatment resistance, these strategies may lead to improvedclinical outcomes for patients with advanced breast cancers.

Disclosure of Potential Conflicts of Interest

M. Wicha has financial holdings in OncoMed Pharmaceuticals, which hasapplied for a patent on CSC technologies. The other authors disclosed nopotential conflicts of interest.

Acknowledgments

We thank Dr. Stephen Ethier for generously providing the BCC lines SUM159and SUM149 and Chris Neeley for technical support. We also thank theUniversity of Michigan core facilities.

Grant Support

This work was supported by NIH grants CA129765, CA101860 (M. Wicha),and 5 P 30 CA46592 (M. Wicha), R01CA125577, and R01 CA107469 (C.G. Kleer).The Taubman Research Institute (M. Wicha).

The costs of publication of this article were defrayed in part by the paymentof page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received February 13, 2010; revised September 8, 2010; accepted September29, 2010; published OnlineFirst Janaury 11, 2011.






References1. Kakarala M, Wicha MS. Implications of the cancer stem-cell hypo-

thesis for breast cancer prevention and therapy. J Clin Oncol2008;26:2813–20.

2. Dontu G, Jackson KW, McNicholas E, Kawamura MJ, Abdallah WM,WichaMS. Role of Notch signaling in cell-fate determination of humanmammary stem/progenitor cells. Breast Cancer Res 2004;6:R605–15.

3. Liu S, Dontu G, Mantle ID, Patel S, Ahn NS, Jackson KW, et al.Hedgehog signaling and Bmi-1 regulate self-renewal of normal andmalignant human mammary stem cells. Cancer Res 2006;66:6063–71.

4. Liu S, Dontu G, Wicha MS. Mammary stem cells, self-renewal path-ways, and carcinogenesis. Breast Cancer Res 2005;7:86–95.

5. Reya T, Clevers H. Wnt signalling in stem cells and cancer. Nature2005;434:843–50.

6. Albini A, Sporn MB. The tumour microenvironment as a target forchemoprevention. Nat Rev Cancer 2007;7:139–47.

7. Uccelli A, Moretta L, Pistoia V. Mesenchymal stem cells in health anddisease. Nat Rev Immunol 2008;8:726–36.

8. Karnoub AE, Dash AB, Vo AP, Sullivan A, Brook MW, Bell GW, et al.Mesenchymal stem cells within tumour stroma promote breast cancermetastasis. Nature 2007;449:557–63.

9. GunnWG, Conley A, Deininger L, Olson SD, Prockop DJ, Gregory CA.A crosstalk between myeloma cells and marrow stromal cells stimu-lates production of DKK1 and interleukin-6: a potential role in thedevelopment of lytic bone disease and tumor progression in multiplemyeloma. Stem Cells 2006;24:986–91.

10. Arnulf B, Lecourt S, Soulier J, Ternaux B, Lacossagne MN, CrinquetteA, et al. Phenotypic and functional characterization of bone marrowmesenchymal stem cells derived from patients with multiple myeloma.Leukemia 2007;21:158–63.

11. Spaeth EL, Dembinski JL, Sasser AK, et al. Mesenchymal stem celltransition to tumor-associated fibroblasts contributes to fibrovas-cular network expansion and tumor progression. PLoS ONE 2009;4:e4992.

12. Landi S, Moreno V, Gioia-Patricola L, Guino E, Navarro M, de Oca J,et al. Association of common polymorphisms in inflammatory genesinterleukin (IL)6, IL8, tumor necrosis factor alpha, NFKB1, and peroxi-some proliferator-activated receptor gamma with colorectal cancer.Cancer Res 2003;63:3560–6.

13. Levina V, Marrangoni AM, DeMarco R, Gorelik E, Lokshin AE. Drug-selected human lung cancer stem cells: cytokine network, tumori-genic and metastatic properties. PLoS ONE 2008;3:e3077.

14. Charafe-Jauffret E, Ginestier C, Iovino F, Wicinski J, Cervera N, FinettiP, et al. Breast cancer cell lines contain functional cancer stem cellswith metastatic capacity and a distinct molecular signature. CancerRes 2009;69:1302–13.

15. Neve RM, Chin K, Fridlyand J, Yeh J, Baehner FL, Fevr T, et al. Acollection of breast cancer cell lines for the study of functionallydistinct cancer subtypes. Cancer Cell 2006;10:515–27.

16. Ginestier C, Hur MH, Charafe-Jauffret E, Monville F, Dutcher J, BrownM, et al. ALDH1 is a marker of normal and malignant human mammarystem cells and a predictor of poor clinical outcome. Cell Stem Cell2007;1:555–67.

17. Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, AntonellisKJ, Scherf U, et al. Exploration, normalization, and summaries

of high density oligonucleotide array probe level data. Biostatistics2003;4:249–64.

18. Liu S, Ginestier C, Charafe-Jauffret E, Kleer CG, Merajver SD, DontuG, et al. BRCA1 regulates human mammary stem/progenitor cell fate.Proc Natl Acad Sci U S A 2008;105:1680–5.

19. Kushida T, InabaM, Hisha H, Ichioka N, Esumi T, Ogawa R, et al. Intra-bone marrow injection of allogeneic bone marrow cells: a powerfulnew strategy for treatment of intractable autoimmune diseases inMRL/lpr mice. Blood 2001;97:3292–9.

20. Muguruma Y, Yahata T, Miyatake H, Sato T, Uno T, Itoh J, et al.Reconstitution of the functional human hematopoietic microenviron-ment derived from humanmesenchymal stem cells in themurine bonemarrow compartment. Blood 2006;107:1878–87.

21. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF.Prospective identification of tumorigenic breast cancer cells. ProcNatl Acad Sci U S A 2003;100:3983–8.

22. Sansone P, Storci G, Tavolari S, Guarnieri T, Giovannini C, Taffurelli M,et al. IL-6 triggers malignant features in mammospheres from humanductal breast carcinoma and normal mammary gland. J Clin Invest2007;117:3988–4002.

23. Grivennikov S, Karin E, Terzic J, Mucida D, Yu GY, Vallabhapurapu S,et al. IL-6 and stat3 are required for survival of intestinal epithelialcells and development of colitis-associated cancer. Cancer Cell2009;15:103–13.

24. Xie T, Spradling AC. A niche maintaining germ line stem cells in theDrosophila ovary. Science 2000;290:328–30.

25. Martinez-Agosto JA, Mikkola HK, Hartenstein V, Banerjee U. Thehematopoietic stem cell and its niche: a comparative view. GenesDev 2007;21:3044–60.

26. Morrissey C, Vessella RL. The role of tumor microenvironment inprostate cancer bone metastasis. J Cell Biochem 2007;101:873–86.

27. Kalwitz G, Endres M, Neumann K, Skrinner K, Ringe J, Sezer O, et al.Gene expression profile of adult human bone marrow-derivedmesenchymal stem cells stimulated by the chemokine CXCL7. Int JBiochem Cell Biol 2009;41:649–58.

28. Tang Z, Yu M, Miller F, Berk RS, Tromp G, Kosir MA. Increasedinvasion through basement membrane by CXCL7-transfected breastcells. Am J Surg 2008;196:690–6.

29. Sheridan C, Kishimoto H, Fuchs RK, Mehrotra S, Bhat-Natshatri P,Turner CH, et al. CD44þ/CD24� breast cancer cells exhibit enhancedinvasive properties: an early step necessary for metastasis. BreastCancer Res 2006;8:R59.

30. Ginestier C, Liu S, Diebel ME, Korkaya H, Luo M, Brown M, et al.CXCR1 blockade selectively targets human breast cancer stem cellsin vitro and in xenografts. J Clin Invest 2010;120:485–97.

31. Haynesworth SE, Baber MA, Caplan AI. Cytokine expression byhuman marrow-derived mesenchymal progenitor cells in vitro: effectsof dexamethasone and IL-1 alpha. J Cell Physiol 1996;166:585–92.

32. Kadereit S, Deeds LS, Haynesworth SE, Koc ON, Kozik MM, SzekelyE, et al. Expansion of LTC-ICs and maintenance of p21 and BCL-2expression in cord blood CD34(þ)/CD38(�) early progenitors culturedover human MSCs as a feeder layer. Stem Cells 2002;20:573–82.

33. Zhu Y, Sun Z, Han Q, Liao L, Wang J, Bian C, et al. Human mesench-ymal stem cells inhibit cancer cell proliferation by secreting DKK-1.Leukemia 2009;23:925–33.

Liu et al.





Correction

Correction: Breast Cancer Stem Cells AreRegulated by Mesenchymal Stem Cells throughCytokine Networks

In this article (Cancer Res 2011;71:614–24), which was published in the January 15,2011 issue of Cancer Research (1), an incorrect version of Figure 4E was published. Thecorrect version of the figure is provided below.

Reference1. Liu S, Ginestier C, Ou SJ, Clouthier SG, Patel SH, Monville F, et al. Breast cancer stem cells are

regulated by mesenchymal stem cells through cytokine networks. Cancer Res 2011;71:614–24.

Published OnlineFirst March 8, 2011.�2011 American Association for Cancer Research.doi: 10.1158/0008-5472.CAN-11-0126

Figure 4.

CancerResearch

www.aacrjournals.org 2407

2011;71:614-624. Published OnlineFirst January 11, 2011.Cancer Res Suling Liu, Christophe Ginestier, Sing J. Ou, et al. Stem Cells through Cytokine NetworksBreast Cancer Stem Cells Are Regulated by Mesenchymal

Updated version

10.1158/0008-5472.CAN-10-0538doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2011/01/14/0008-5472.CAN-10-0538.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/71/2/614.full#ref-list-1

This article cites 33 articles, 10 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/71/2/614.full#related-urls

This article has been cited by 40 HighWire-hosted articles. Access the articles at:

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

SubscriptionsReprints and

[email protected] at

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

Permissions

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

.http://cancerres.aacrjournals.org/content/71/2/614To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-10-0538

http://cancerres.aacrjournals.org/content/suppl/2011/01/14/0008-5472.CAN-10-0538.DC1

http://cancerres.aacrjournals.org/content/71/2/614.full#ref-list-1

http://cancerres.aacrjournals.org/content/71/2/614.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/71/2/614


Documents

Breast Cancer Stem Cells Are Regulated by Mesenchymal Stem ... › content › canres › ... · Tumor and Stem Cell Biology Breast Cancer Stem Cells Are Regulated by Mesenchymal