the role of versican in modulating breast cancer cell self-renewal · tion of cells that function...

Cell Cycle and Senescence

The Role of Versican in Modulating Breast Cancer CellSelf-renewal

William Weidong Du1,3, Ling Fang1,2, Xiangling Yang1,2, Wang Sheng1,2, Bing L. Yang1, Arun Seth1,2,Yaou Zhang4, Burton B. Yang1,2, and Albert J. Yee1,3

AbstractVersican is highly expressed during the early stages of tissuedevelopment and its expression is elevated duringwound

repair and tumor growth. There is little literature on the potential role of breast cancer stem cells on the cellular–extracellular matrix interactions involving versican. An anti-versican short hairpin RNA (shRNA) was used to observethe effect of reduction of versican on breast cancer self-renewal. A versicanG3 construct was exogenously expressed inbreast cancer cell lines. Colony formation and mammosphere formation assays were conducted; flow cytometry wasapplied to analyze the prevalence of side population cells. The versican G3- and vector-transfected 66c14 cells wereinjected transdermally intoBALB/cmice as a 10-fold dilution series from1� 105 to 1� 102 cells permouse. VersicanG3 domain enhanced breast cancer self-renewal in both experimental in vitro and in vivo models. VersicanG3–transfected cells contained high levels of side population cells, formed more mammospheres when culturedin the serum-free medium, and formed a greater number and larger colonies. Reduction of versican's functionalitythrough anti-versican shRNA or knocking out the EGF-like motifs reduced the effect of versican on enhancingmammosphere and colony formation. Versican-enhanced self-renewal played a role in enhanced chemotherapeuticdrug resistance, relating partly to the upregulated expression of EGF receptor (EGFR) signaling. Versican ishighly expressed in breast cancer progenitor cells and was maintained at high levels before cell differentiation.Overexpression of versican enhanced breast cancer self-renewal through EGFR/AKT/GSK-3b (S9P) signalingand conferred resistant to chemotherapeutic drugs tested. Mol Cancer Res; 11(5); 443–55. �2013 AACR.

IntroductionExperimental and clinical evidence support the under-

standing that tumorigenesis is sustained by a small popula-tion of cells that function as tumor stem/progenitor cells(1, 2). In breast cancer, these cells arise from mutatedmammary stem/progenitor cells, which have been charac-terized by the expression of key cell surface markers (e.g.,CD24, CD44, CD29, or Sca-1) as well as dye efflux assaysbased on the differential release of incorporated Hoechstdyes or rhodamine-121 through overexpressed or overacti-vated drug transporters in the cell membrane (3–5). These

tumor progenitor cells are characterized by their ability toform new serially transplantable tumors in mice and todisplay stem/progenitor cell properties such as competencefor self-renewal and the capacity to reestablish tumor het-erogeneity (6, 7). In addition to flawed regulation of the self-renewal pathways, the number of cells within a tumor thathave the ability to self-renew is constantly expanding,resulting in a continuous expansion of tumorigenic cancercells (8). Thus, the identification of mechanisms in whichcancer cells regenerate through self-renewal is critical to ourunderstanding of tumorigenesis (9). A number of develop-mental signaling pathways, such asWnt,Notch, andHedge-hog, have been observed to play pivotal roles in governingboth self-renewal and the process of malignant transforma-tion (10, 11).Tumor stem cells, as well as other cells in the tissue, do not

function autonomously as independent self decision-makingentities. Increasing evidence shows the influence of thecellular microenvironment on tumor development and pro-gression (2, 12). Cellular activity is closely regulated throughinteractions with adjacent cells that create well-definedmicroenvironments in which tumor stem cells reside. How-ever, tumor cells also possess the ability to interact with andinfluence their surrounding environment. Examples of theseproperties include neo-angiogenesis, recruitment of epithe-lial cells, and modification of tissue architecture. Interest-ingly, certain extracellular matrix (ECM) and cell-surface

Authors' Affiliations: 1Sunnybrook Research Institute; 2Department ofLaboratory Medicine and Pathobiology; 3Centre for the Study of BoneMetastasis, Odette Cancer Centre, and HollandMusculoskeletal Program,Sunnybrook Health Sciences Centre, Division of Orthopaedic Surgery,Department of Surgery, University of Toronto, Toronto, Ontario, Canada;and 4Division of Life Science, Graduate School at Shenzhen, TsinghuaUniversity, Shenzhen, China

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

Corresponding Author: Albert J. Yee, Sunnybrook Health Sciences Cen-tre and Centre for the Study of Bone Metastasis, Odette Cancer Centre,Department of Surgery, University of Toronto, 2075 Bayview Avenue, Rm.MG 371-B, Toronto, Ontario M4N 3M5, Canada. Phone: 416-480-6815;Fax: 416-480-5886; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-12-0461

�2013 American Association for Cancer Research.

MolecularCancer

Research

www.aacrjournals.org 443

on June 14, 2020. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 28, 2013; DOI: 10.1158/1541-7786.MCR-12-0461

http://mcr.aacrjournals.org/

receptor molecules, such as Wnt, Notch, and TGF-b havebeen shown to be involved in stem cell self-renewal andtumor development (2, 10, 11, 13).Versican, a chondroitin sulfate (CS) proteoglycan, is one

of the main components of the ECM and provides a looseand hydrated matrix during key events in development anddisease (14). To date, 4 isoforms of versican (V0, V1, V2,andV3) have been identified in various tissues (15). Versicanis structurally composed of an N-terminal G1 domain, a CSattachment region, and a C terminus (or G3) selectin-likedomain. The G3 domain interacts with different ECMproteins and binds to certain cell surface proteins includingEGF receptor (EGFR; ref. 16). The G3 domain is composedof 2 EGF-like repeats, a lectin-like motif, and a complementbinding protein-like motif.Versican is highly expressed in the early stages of tissue

development, and its expression decreases after tissue mat-uration. It is also highly expressed in the interstitial tissues atthe invasive margins of breast carcinoma (17). Expressionlevels of versican seem to be related to the metastaticpotential of breast cancer and are predictive of relapse andoverall survival (18). Relapse in women with node-negativebreast cancer seems related to the level of versican depositedin peritumoral stroma by mammary fibroblasts (18, 19).Neoplastic remodeling of the ECM through increased ver-sican depositionmay facilitate local invasion and subsequentmetastasis (20). The G3 domain of versican has beenpreviously shown to be important in both local and systemictumor invasiveness of breast cancer and may enhance theinteractions between tumor cells and their surroundingstromal components. Furthermore, the G3 domain of ver-sican has been shown to enhance neo-vascularizationthrough interactions with VEGF and fibronectin(17, 21). Thus, versican participates in cell adhesion, pro-liferation, migration, aggregation, and angiogenesis andseems to play a central role in normal tissue morphogenesis(14, 22–24). Given recent scientific interest in the biologyof cancer stem cells, the role of versican in the regulationof breast cancer cell self-renewal has not been wellcharacterized.To investigate the effects of versican expression on breast

cancer cell self-renewal, we exogenously expressed a versicanG3 construct in mouse mammary tumor cell lines 4T1,66c14, 4T07, and human breast cancer cell lines MT-1,MB-468. To determine the expression of versican in breastcancer stem/progenitor cells and the role of the versican G3domain on breast cancer cell self-renewal, both in vitro andin vivomodels were used. Finally, we aimed to determine theeffect of commonly prescribed chemotherapeutic drugs onbreast cancer cell sensitivity/resistance to versican G3–enhanced tumor progenitor cell activity.

Materials and MethodsMaterials and cell culturesMonoclonal antibodies against phosphorylated extracel-

lular signal–regulated kinase (perk), integrin-b1, and thepolyclonal antibodies against pEGFR and pAKT were

obtained from Santa Cruz Biotechnology. EGF, selectiveEGFR inhibitor AG 1478, selective mitogen-activated pro-tein/extracellular signal–regulated kinase (MEK) inhibitorPD 98059, hydroxyurea, and the monoclonal antibodyagainst b-actin were obtained from Sigma. Polyclonal anti-bodies against versican V1 isoform (ab19345), GSK-3b(S9P), and CD44 were obtained from Abcam. Monoclonalantibody against Sca-1 came from Bio-Rad. Monoclonalantibody against ALDH1 and Sox2 were obtained from BD.Horseradish peroxidase (HRP)–conjugated goat anti-mouseimmunoglobulin G (IgG) and HRP-conjugated goat anti-rabbit IgG were obtained from Bio-Rad. Immunoblottingwas carried out using the ECL Western Blot Detection Kit.Mouse mammary tumor cell lines 67NR, 66c14, 4T07,

4T1, and human breast cancer cell line MDA-MB-231 werecultured in Dulbecco's modified Eagle's medium (DMEM)media. Human breast cancer cell lines MT-1, MCF-7, andMDA-MB-468 were cultured in RPMI-1640 media. Cellswere supplemented with 10% fetal calf serum (FCS), pen-icillin (100 U/mL), streptomycin (100 mg/mL), and main-tained at 37�C in a humidified atmosphere of 5% CO2. ApcDNA1-G3 (G3) construct and pcDNA1-G3 fragmentlacking the EGF-like motifs (G3DEGF) construct werestably expressed in 66c14, 4T07, 4T1, and MT-1 cells asdescribed previously (25, 26).A monoclonal antibody 4B6 that recognizes an epitope in

the leading peptide engineered to the G3 and G3DEGFconstruct was used to confirm expression of the G3 products(25, 26).A short hairpin RNA (shRNA) construct targeting versi-

can was generated by inserting the sequence 50-ccca-gatctccgtggcagcgagaattcttgttcaagagacaagaattctcgctgc-cactttttggaagagctcatcga into the plasmid pSuper. This con-struct thus targeted the sequence of versican at nucleotides9453-9471 (50-gtggcagcgagaattcttg, NM 004385.4).

Side population cell analysisCells were removed from tissue culture dishes with trypsin

and EDTA, washed, suspended at 106 cells/mL in DMEMcontaining 2% FCS, and preincubated at 37�C for 10minutes, then cultured for 90 minutes with 2.5 mg/mLHoechst 33342 dye, either alone or in combination with 50mmol/L verapamil. Cells were analyzed by flow cytometryusing a dual-wavelength analysis (blue, 424–444 nm; red,675 nm) after excitation with 350-nm UV light. Propidiumiodide–positive dead cells (<15%) were excluded from theanalysis.

Western blot analysisSamples were subjected to SDS-PAGE using 7% to 10%

acrylamide. Separated proteins were transblotted onto anitrocellulose membrane in 1� Tris/glycine buffer contain-ing 20% methanol at 60 V for 2 hours in a cold room. Themembrane was blocked in TBST (10 mmol/L Tris–Cl, pH8.0, 150 mmol/L NaCl, and 0.05% Tween 20) containing5% nonfat dry milk powder (TBSTM) for 1 hour at roomtemperature, and then incubated with primary antibodiesat 4�C overnight. The membranes were washed with TBST

Du et al.

Mol Cancer Res; 11(5) May 2013 Molecular Cancer Research444




(3 � 30 minutes) and then incubated with appropriateHRP-conjugated secondary antibodies in TBSTM for 1hour. After washing, the bound antibodies were visualizedby ECL detection.

Mammosphere cultureHarvested G3- and vector-transfected cells were mechan-

ically and enzymatically dissociated in 0.05% Trypsin–EDTA for 10 minutes at 37�C, followed by resuspensionin serum-free DMEM/F12 medium containing 10 ng/mLbasic fibroblast growth factor (bFGF), 20 ng/mL EGF,L-glutamine (1:400), and B27 supplement (1:50). Singlecell suspensions were plated onto 60-mm Petri dishes at adensity of 1,000 cells/mL. After culturing for 3 days, spherecells were collected by gentle centrifugation, dissociated tosingle cells in 0.05% Trypsin–EDTA as described previ-ously, and cultured to generate mammospheres. Complexescontaining 30 or more cells were considered "large" spheres,whereas small spheres contained less than 30 cells persphere.

Colony formation in soft agarose gelA total of 103 cells were mixed in 0.3% low-melting

agarose in DMEM supplemented with 10% FBS and platedon 0.66% agarose-coated 6-well tissue culture plates. Fourweeks after cell inoculation, colonies were examined andphotographed. Complexes containing 100 cells were con-sidered large colonies, whereas small colonies contained 30–100 cells per colony.

Cell viability assaysThe G3- and vector-transfected 66c14 cells (2 � 105)

were cultured in 10% FBS/DMEM medium in culturedishes and maintained at 37�C for 12 hours. After cellattachment, culture media was changed into serum-freeDMEMmedium or 10% FBS/DMEMmedium containingdifferent concentrations of chemotherapeutical chemicalsincluding C2-ceramide, docetaxel, doxorubicin, or epirubi-cin. Cells were harvested daily and cell number was analyzedby Coulter counter. Cell survival assays were conductedusing a colorimetric proliferation assay (Cell ProliferationReagent WST-1; Basel, Switzerland). Versican G3- andcontrol vector-transfected breast cancer cells (1 � 104

cells/well) were inoculated and cultured in 10% FBS/DMEM medium in 96-well culture dishes for 12 hours.After cell attachment, culture media was changed to serum-free DMEM medium or 10% FBS/DMEM medium con-taining differing concentrations of C2-ceramide, docetaxel,doxorubicin, or epirubicin, and then cultured with 10 mLWST-1 reagent for 4 hours. The absorbance of the samplesread against a blank control was measured by a microplate(ELISA) reader.

RT-PCRA total of 2� 106 66c14 orMT-1 cells were harvested and

total RNAwas extracted using the Qiagen RNeasyMini Kit.Two micrograms of total RNA were used to synthesizecDNA, a portion of which (0.2 mg RNA) was used in a

PCR with 2 appropriate primers. PCR products weremeasured using quantitative real-time PCR (qRT-PCR)incorporating TaqMan chemistry (Applied Biosystems).Sequences of primers used in this study were: mouse integ-rin-b1: forward, 50-tgaatttgcaactggtttcctgg; reverse, 50-ctatg-tcacttgctggcaaccc; mouse Sca-1: forward, 50-tggacacttctc-acactacaaagtc; reverse, 50-cagagcaaggtctgcaggag; humanOct-4: forward, 50-gcgccttccttccccatggcggg; reverse, 50-gtacgccatccccccacataactc; human GapdH: forward, 50-aaggctggggctcatttgcag; reverse, 50-gatgttctggagagcccccgcg;mouse GapdH: forward, 50-atggtgaaggtctgtgtgatcatc;reverse, 50-tgggttctcactcctggaagaagg

Immunofluorescence stainingCultured cells were fixed with cold 3.5% paraformalde-

hyde for 15 minutes, washed 3 times with PBS and per-meabilized with 1% Triton X-100 in PBS for 30 minutes.After washing 3 times with PBS, cells were blocked in 1%bovine serum albumin (BSA) in PBS for 30 minutes, andthen incubated with primary antibodies (1:200) at 4�Covernight. After washing, the cells were incubated withfluorescein isothiocyanate (FITC)-conjugated secondaryantibodies (1:400) for 1 hour. Control samples were treatedwith either secondary antibody alone or preimmunemouse serum. The samples were examined under a NikonEclipse E600 upright microscope equipped with fluorescentdevices.

Flow cytometryCells were washed once with PBS and harvested in 0.05%

trypsin/0.025% EDTA. Detached cells were washed withPBS containing 1% FCS and 1% penicillin–streptomycin(wash buffer) and resuspended in wash buffer (106 cells/100mL). Cells were then incubated with Sca-1 monoclonalantibodies and FITC-conjugated secondary antibodies for30 minutes on ice. The labeled cells were washed in washbuffer and then analyzed by flow cytometry.

In vivo tumorigenicity in BALB/c miceThe G3- and vector-transfected 66c14 cells were cultured

in 10% FBS/DMEMmedia at 37�Cwith 5%CO2. At 70%to 80% subconfluency, the cells were re-fed with fresh 10%FBS/DMEM media 24 hours before inoculation into mice.Cell viability was determined by Trypan blue exclusion andcells were suspended with greater than 95% viability withoutcell clumping. Ethics approval was received from the insti-tutional animal care committee. Four-week-old BALB/cmice were injected transdermally with the G3- and vec-tor-transfected 66c14 cells (1� 102, 1� 103, 1� 104, and 1� 105 cells in 150 mL 10% FBS/DMEM medium). Injec-tion was administered into the fourth (inguinal) mammaryfat pad using a 1 mL syringe with a 26 G needle. Eachexperimental group contained 8 randomly chosen mice.Tumors were measured weekly with a vernier caliper. Allmice were sacrificed when the tumor measured 1 cm, whenacceptable tumor endpoints in animal care where reached,or at 21 weeks posttransplantation. Primary tumors wereharvested and fixed in 10% formalin.

Versican Modulates Breast Cancer Cell Self-renewal

www.aacrjournals.org Mol Cancer Res; 11(5) May 2013 445




A

B

0 50K 100K 150K 200K 250K<Indo blue-A>

0

50K

100K

150K

200K

250K

<D

AP

I/H

O342-A

>

5.25%

0 50K 100K 150K 200K 250K<Indo blue-A>

0

50K

100K

150K

200K

250K

0

50K

100K

150K

200K

250K

0

50K

100K

150K

200K

250K

<D

AP

I/H

O342-A

>

2.5%

0 50K 100K 150K 200K 250K<Indo blue-A>

0

50K

100K

150K

200K

250K

<D

AP

I/H

O342-A

>

9.93%

0 50K 100K 150K 200K 250K<Indo blue-A>

0

50K

100K

150K

200K

250K

<D

AP

I/H

O342-A

>

2.2%

0 50K 100K 150K 200K 250K<Indo blue-A>

<D

AP

I/H

O342-A

>

14.7%

Differentiated

0 50K 100K 150K 200K 250K<Indo blue-A>

<D

AP

I/H

O342-A

>

4.34%

In sphere 4 h

16 h 24 h 48 h

Ho

ech

st

blu

e

Hoechst red

Ho

ech

st

blu

e

Hoechst red

D

C

*

**

**

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

67NR 66c14 4T07 4T1

SP

ce

lls (

%)

β-Actin

Versican V1

(250 kDa)

Serum Serum free

3 3 7 7 14 14

0.61 1.01 1.29 1.45 1.32 1.37 1.35

E

β-Actin

Versican V1

(250 kDa)

Serum Serum free Serum recovery

4 8 16 24 48 72

0.57 1.61 1.72 1.25 1.18 1.04 0.49 0.51

67NR 66c14 4T07 4T1

Versican V1

β-Actin

0.58 0.66 0.72 1.31

170

80

58

46

35

Figure 1. Expression of versican inmammary tumor stem/progenitor cells. A, prevalence of Hoechstlow side population (SP) cells in the 4mousemammary celllines. Comparedwith 66c14 cells; �,P<0.05; ��,P<0.01;n¼4; analyzedwith t test. B, cells of the 4mousemammary cell lineswere cultured inDMEMmediumwith 10% FBS and cell lysates were prepared and subjected to immunoblotting with antibodies to versican and b-actin. C, 4T1 cells were cultured in DMEMmedium with 10% FBS or cultured as mammospheres in suspension with serum-free medium for 3, 7, and 14 days. Cell lysates were prepared andsubjected to immunoblottingwith antibodies to versican and b-actin. D, 4T1 cells were cultured asmammospheres in suspensionwith serum-freemedium for14 days. Themammosphere cells were also recoveredwith 10%FBS/DMEM for 4, 8, 16, 24, 48, and 72 hours. The differentiated cells were culturedwith 10%

Du et al.





Tissue slide H&E staining and immunohistochemistryFreshly excised primary tumors were fixed in 10% for-

malin overnight, immersed in 70% ethanol, embedded inparaffin, and sectioned. Sections were deparaffinized withxylene and ethanol and boiled in a pressure cooker. Afterwashing with TBS containing 0.025% Triton X-100, thesections were blocked with 10% goat serum and incubatedwith primary antibody in TBS containing 1% BSA over-night. The sections were washed and labeled with biotiny-lated secondary antibody, followed by avidin-conjugatedHRP provided by the Vectastain ABC Kit (Vector, PK-4000). The slides were then counterstained with Mayer'sHematoxylin and slide-mounted.

ResultsVersican is highly expressed in mammary tumor stem/progenitor cellsThe Hoechst 33342 dye-based side population technique

facilitated isolation of mouse mammary stem/progenitorcells. To investigate the prevalence of side population cellsin breast cancer cells, we stained the 4 mouse mammarytumor cell lines with Hoechst 33342 dye and identified theside population by its characteristic fluorescent profile indual-wavelength analysis. The 4mousemammary tumor celllines contained side population cells ranging from 0.4% to3.2% of the total viable cells (67NR, 0.62% � 0.12%;66c14, 0.42% � 0.05%; 4T07, 0.96% � 0.11%; 4T1,2.73%� 0.53%). The proportion of side population cells in4T1 cell lines were substantially higher than observed in theother 3 cell lines, with a median of 2.56% and a range of1.98% to 3.34% (Fig. 1A). Immunoblotting also showedthat 4T1 cells expressed higher levels of the versican V1isoform than other cell lines (Fig. 1B). We compared theexpression of versican in mammary/progenitor cells withthat of differentiated mammary tumor cells by using mam-mosphere-derived cells grown in suspension culture withserum-free medium versus regularly cultured 4T1 cells with10% FBS medium. Immunoblotting showed that theexpression of versican is greater in stem/progenitor cellscontained within mammospheres than in differentiated cellscultured in serum medium (Fig. 1C). The mammosphere4T1 cells expressed higher levels of versican compared withthe differentiated cells and approached that of differentiatedcells once cultured in serummedium for 72 hours (Fig. 1D).The mammosphere 4T1 cells also contained a high percent-age of side population cells when compared with differen-tiated cell cultures. The prevalence of side populationcells declined rapidly once cultured with serum mediumand cells were also observed to be differentiating (Fig. 1E).Cells containing a high percentage of side population cells,possessing stem cell-like properties, also produced high levelsof versican before cell differentiation. Immunostaining

showed that mammosphere 4T-1 cells not only expressedhigh levels of versican, but also expressed high levels ofcharacteristic breast cancer stem cell markers Sca-1, Sox2,and ALDH1 (Supplementary Fig. S1). These results indi-cated that versican was highly expressed in mammary tumorstem/progenitor cells and its expression seemed to be down-regulated during cell differentiation.

Versican G3 enhances breast cancer cell mammosphereand colony formationTo investigate the effect of versican on breast cancer self-

renewal, we exogenously expressed a versican G3 constructin MT-1, 66c14, 4T07, and 4T1 cell lines. Flow cytometryassays showed that the proportions of side population cellswere higher in versican G3-transfected MT1, 66c14, and4T07 cells than in vector control groups (Fig. 2A and B).Cultured in suspension with serum-free medium, versicanG3-transfected cells formed larger mammospheres thanvector control cells (Fig. 2C). The number of mammo-spheres reflected the quantity of cells capable of in vitro self-renewal, whereas the number of cells/sphere indicated theself-renewal capacity of each sphere-generating cell. Disso-ciated G3-transfected cells from primary mammospheresgenerated an equivalent proportion of secondary and tertiaryspheres, which highlighted their potential for in vitro self-renewal (Fig. 2D and E). Versican G3-expressing breastcancer cells showed enhanced colony formation capacity onsoft agarose gel (Supplementary Fig. S2A and S2B).

Versican enhances breast cancer cell resistance tochemotherapy and influences markers associated withself-renewalWe treated versican G3- and vector-transfected 66c14

cells with 8mmol/L doxorubicin or 10mmol/L epirubicin for24 hours. We observed that versican G3-transfected 66c14cells showed resistance to doxorubicin and epirubicin treat-ment (27). To investigate whether these remaining viablecells were of tumor stem cell origin, we carried out immu-nofluorescence staining with antibodies to Sca-1, which washighly expressed inmousemammary tumor progenitor/stemcells. Remaining viable cells showed a high level of Sca-1expression (Supplementary Fig. S3A). An immunofluores-cence staining experiment carried out on 66c14 cells cul-tured in 10%FBS/DMEM medium also showed that versi-can G3-transfected cells expressed higher levels of Sca-1 thanthe vector cells (Supplementary Fig. S3B). In studying therelationship between versican-enhanced breast cancer cellself-renewal and chemotherapy resistance, we also culturedthe 4mousemammary tumor cell lines 67NR, 66c14, 4T01,and 4T1 (containing different proportions of side popula-tion cells and differential versican expression) in 8 mmol/Ldoxorubicin for 4 days. We observed that although the 4T1

FBSmedium on tissue culture dishes for 3 days. Cell lysates were prepared and subjected to immunoblotting with antibodies to versican and b-actin. E, 4T1cells were cultured asmammospheres in suspension in Petri dishes containing serum-freemedium for 14 days. Themammosphere cells were also recoveredwith 10%FBS/DMEM for 4, 8, 16, 24, 48, and 72 hours. All the sampleswere harvested, stainedwithHoechst 33342 dye, and subjected to analysismeasuringthe prevalence of side population cells. The differentiated cells were cultured with 10% FBS medium on tissue culture dishes for 3 days.






MT-1 4T166c14 4T07

0 50K 100K150K200K250K0

50K

100K

150K

200K

250K

2.72

0 50K 100K150K200K250K0

50K

100K

150K

200K

250K

1.69

0 50K 100K150K200K250K0

50K

100K

150K

200K

250K

2.59

0 50K 100K150K200K250K0

50K

100K

150K

200K

250K

2.98

0 50K 100K150K200K250K

0

50K

100K

150K

200K

250K

0.18

0 50K 100K150K200K250K0

50K

100K

150K

200K

250K

0.59

0 50K 100K150K200K250K0

50K

100K

150K

200K

250K

2.5

0 50K 100K150K200K250K0

50K

100K

150K

200K

250K

1.44

Hoechst red

Hoech

st b

lue

Vector

G3

A

**

**

**

0

1

2

3

4

MT-1 66c14 4T07 4T1

SP

cells

(%

)

Vector

G3

1 d

3 d

5 d

10 d

15 d

Vector G3C

Vector G3

Primary

Secondary

Teritary

****

**

0

5

10

15

20

25

30

35

Vector G3

Sphere

s/1

,000 c

ells

Primary

Secondary

Tertiary

B

D

E

Figure2. Expressionof versicanG3domain enhancedbreast cancer cellmammosphere formation.A, the typical prevalenceofHoechstlow sidepopulation (SP)cells in versican G3- and vector-transfected MT-1, 66c14, 4T07, and 4T1 cell lines as shown by flow cytometry. B, prevalence of Hoechstlow side populationcells in versicanG3-andvector-transfectedMT-1, 66c14, 4T07, and4T1cell lines.Comparedwith vector control group; �,P<0.05; ��,P<0.01;n¼4; analyzedwith t test. C, a typical mammosphere of versican G3- and vector-transfected 66c14 cells cultured in suspension with serum-free medium. D, typicalmammospheres of versican G3- and vector-transfected 66c14 cells formed primarily, secondarily (generated from dissociated primary spheres), and tertiary.E, numbers of primary, secondary, and tertiary mammospheres. Compared with vector control group; �, P < 0.05; ��, P < 0.01; n ¼ 6; analyzed with t test.

Du et al.





cells, which had a high proportion of progenitor cells andexpressed higher levels of versican compared with othermouse mammary tumor cell lines, showed evidence ofcellular survival (Supplementary Fig. S3C).

Versican-expressing cells expressed high levels of breastcancer self-renewal markersRT-PCR and immunoblotting experiments showed that

the versican G3-transfected mouse mammary tumor cellsexpressed high levels of Sca-1 and integrin-b1 (Supplemen-tary Fig. S3D–S3F). Immunoblotting showed that G3-expressing human breast cancer cells expressed increasedALDH1, CD44, and integrin-b1 (Supplementary Fig.S3G). These cells expressed high levels of Oct-1 as evaluatedby RT-PCR (Supplementary Fig. S3H). The evaluatedbreast cancer cell self-renewal markers seemed upregulatedin versican G3-transfected breast cancer cells, supportingG30s influence on breast cancer self-renewal.

Effect of versican on promoting breast cancer cell self-renewal is related to its EGF-like motifsThe 4T07 cells were stably transfected with a G3 con-

struct, a G3 fragment lacking the EGF-like motifs(G3DEGF), and a vector control. We observed thatG3DEGF expression did not enhance mammosphere for-mation to the degree observedwithG3-transfected cells (Fig.3A and D). Colony formation assays showed that both G3and G3DEGF-transfected cells formed more colonies thanvector control cells. However, G3DEGF-expressing cellsformed smaller colonies when compared withG3-expressingcells (Fig. 3B and E). The prevalence of side population cellsin G3DEGF-expressing cells was lower than in G3-expres-sing cells (Fig. 3F). Immunoblotting showed that G3DEGF-expressing cells did not show enhanced pEGFR, Sca-1, Sox2,ALDH1, and integrin-b1 as was observed with G3-trans-fected cells (Fig. 3C). Flow cytometry indicated that withoutthe EGF-like motifs, G3-expressing 4T07 cells did notsignificantly enhance Sca-1 expression (Fig. 3G).

Versican promoted breast cancer self-renewal throughenhanced EGFR signalingTo investigate the role of EGFR signaling inG3-enhanced

self-renewal, the G3- and vector-transfected 4T07 cells werecultured in suspension culture conditions using serum-freestem cell medium. G3 cells were also cultured with orwithout EGF, with selective EGFR inhibitor AG 1478,selective ERK inhibitor PD 98059, selective c-jun-NH2-kinase (JNK) inhibitor side population 600125, or selectiveAKT inhibitor triciribine for 14 days. The G3 cells formedmore mammospheres than the vector control cells, however,the G3 cells were observed to form very few mammosphereswithout EGF in themedium (Fig. 4A). Both selective EGFRinhibitor AG 1478 and selective AKT inhibitor triciribineprevented versican G3-enhanced mammosphere formation,which indicated that observed G3-enhanced breast cancerself-renewal effects were due, at least in part, to its effect oninfluencing the EGFR/AKT pathway (Fig. 4A). Immuno-blotting showed that the versican G3-expressing cells

expressed high levels of pEGFR, pAKT, GSK-3b (S9P),Sca-1, Sox2, and ALDH1. These effects were blocked byinhibitor AG 1478 and selective AKT inhibitor triciribine(Fig. 4B). To observe whether versican enhanced breastcancer cell self-renewal through EGFR-AKT-GSK-3b(S9P)–mediated pathways, we cultured the versican G3-expressing cells in 40 mmol/L c2-ceramide or 2 mmol/Ldocetaxel. We observed inhibited expression of GSK-3b(S9P) in breast cancer cells (27). Immunoblotting alsoshowed decreased expression of GSK-3b (S9P) with c2-ceramide and docetaxel. Previously observed effects of ver-sican G3-expressing cells on enhanced expression of stemcell markers—Sca-1, Sox2, and ALDH1—were blocked(Fig. 4D).Overexpression of versican in breast cancer cells enhanced

expression of pEGFR and its downstream effects on pERK,pSAPK/JNK, and pAKT. pAKT seemed to enhance theexpression of GSK-3b (S9P) and some stem cell markers(Sca-1, Sox2, and ALDH1) resulting in enhanced tumor cellresistance to chemotherapy, enhanced cellular survival, andenhanced self-renewal (Fig. 4C). On the other hand,enhanced expression of pEGFR/pJNK in cancer cells seemedto enhance cell sensitivity to the chemotherapeutic drugsevaluated (27).

Silencing versican's functionality reduces its observedeffects on breast cancer cell self-renewalTo investigate the effect of silencing versican's function on

breast tumor cell self-renewal, we first transfected the humanbreast cancer cell line MDA-MB-231 with anti-versicanshRNA. These transfected cells formed fewer and smallermammospheres when cultured in suspension culture con-ditions using serum-free medium (Fig. 5A and B). Flowcytometry showed that anti-versican shRNA cells possessedlower proportions of tumor progenitor/stem cells comparedwith vector control cells (Fig. 5C). Immunoblotting showedthat anti-versican shRNA-expressing MDA-MB-231 cellsdecreased expression levels of versican V1, pEGFR, pAKT,GSK-3b (S9P), and stem cell markers Sca-1, Sox2, andALDH1 (Fig. 4D). Anti-versican shRNA-expressing cellsalso showed reduced colony formation ability when com-pared with the vector control cells (Fig. 5E and F).

Expression of versican promoted breast cancer self-renewal in vivoWe injected versican G3- and vector-transfected 66c14

cells into the fourth mammary fat pad of BALB/c mice as a10-fold dilution series from 1 � 105 to 1 � 102 cells permouse.With as few as 1,000 inoculated cells, G3-transfected66c14 cells could form tumors in BALB/c mice, whereas thevector control cells were unable to show the same ability atthe same dilution (Table 1). Although tumors could form ifgreater than 10,000 vector control cells were injected, theincidence of tumors in the vector control group was muchlower than that observed of the G3 group using the samenumber of injected cells.Immunohistochemical staining showed that primary

tumor tissues of versican G3-treated mice not only expressed






Vector

G3

G3ΔEGF

100 101 102 103 104FL1-H: NL493

0

10

20

30

# C

ells

9.0191

100 101 102 103 104FL1-H: NL493

0

10

20

30

40

50

# C

ells

0.2199.8

FSC-H, SSC-H subset

vector

Event count: 1989

FSC-H, SSC-H subset

G3

Event count: 1720

100 101 102 103 104

FL1-H: NL493

0

30

60

90

120

# C

ells

0.043100

FSC-H, SSC-H subset

G3ΔEGF

Event count: 1966

BA

E

C

G

Vector

G3

G3ΔEGF

D

F

**

**

**

0

10

20

30

40

50

Small colony Large colony

Colo

nie

s/1

,000 c

ells

**

**

** *

0

2

4

6

8

10

12

14

16

7 d 14 d 21 d

Sphere

s/1

,000 c

ells

Vector

G3

G3ΔEGF

*

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Vector G3 G3 EGF

SP

cells

(%

)

Vector G3 G3 EGF

4B6

pEGFR

Sca-1

Integrin-β1

β-Actin

ALDH1

SOX2

0.15 1.95 2.16

0.65 1.31 0.74

0.85 1.45 0.79

0.47 1.10 0.52

0.62 1.70 0.59

0.53 1.15 0.66

Figure 3. Knocking out the EGF-like motifs in the versican G3 construct (G3DEGF) resulted in reduced effects on breast cancer cell self-renewal. A, typicalmammospheres of vector-, versican G3-, and G3DEGF-transfected 4T07 cells formed in serum-free DMEM medium after culture for 14 days. B, typicalcolonies of vector-, versican G3-, and G3DEGF-transfected 4T07 cells formed in soft agarose gel after culture for 4 weeks. C, vector-, versican G3-, andG3DEGF-transfected 4T07 cells were cultured in 10% FBS/DMEM medium. Cell lysates were prepared and subjected to immunoblotting with antibodies toversican G3, pEGFR, Sca-1, Sox2, ALDH1, integrin-b1, and b-actin. D, the numbers of mammospheres of vector-, versican G3-, and G3DEGF-transfected4T07 cells counted at 7, 14, and 21 days. Compared with vector control group; �, P < 0.05; ��, P < 0.01; n ¼ 6; analyzed with t test. E, the numbers ofcolonies of vector-, versican G3-, and G3DEGF-transfected 4T07 cells formed on soft agarose gel. Compared with vector control group; �, P < 0.05;��,P < 0.01; n¼ 3; analyzedwith t test. F, prevalence of Hoechstlow side population (SP) cells of vector-, G3-, andG3DEGF-transfected 4T07 cells. Comparedwith 4T07 cells; �, P < 0.05; ��, P < 0.01; n ¼ 4; analyzed with t test. G, typical flow cytometry of vector-, G3-, and G3DEGF-transfected 4T07 cells stainedwith antibodies against Sca-1 and FITC-conjugated secondary antibodies.

Du et al.





high levels of 4B6, pEGFR, pAKT, andGSK-3b (S9P), all ofwhich were related with tumor invasiveness, but alsoexpressed high levels of tumor stem cell markers Sox2,Sca-1, and ALDH1. Our in vivo studies complementedobserved in vitro effects on breast cancer self-renewal (Fig. 6).

DiscussionIn human breast carcinoma, versican is detected in the

interstitial tissues at the invasive margins and in the elastictissues associated with tumor invasion (25, 28). The highexpression of versican in human breast tumors is prognostic,being predictive of relapse, and negatively impacting overallsurvival rates (18, 19). The mechanisms by which versican

facilitates tumor growth and metastatic transformation inbreast cancer are still not clear. Our study highlights theunderstanding that versican expression is greater in moreinvasive breast cancer cell lines, such as 4T1. In our experi-ments using the Hoechst 33342 dye-based side populationtechnique, we observed that invasive breast cancer cell linescontained a high proportion of side population cells, anenriched source of stem cells or primitive andundifferentiatedcells (29, 30).Thiswas concordantwith high levels of versicanexpression. Mammary stem/progenitor cells were reported inthis study as Hoechst 33342–negative and Sca-1–positivecells (31). To confirm the relationship between versicanexpression and tumor progenitor cell activity, we cultured4T1 cells on a nonadherent surface. Mammospheres, which

Figure 4. Expression of versicanpromoted breast cancer cell self-renewal through enhanced EGFR/AKT/GSK-3b (S9P) signals. A,numbers of mammospheres of G3-transfected 4T07 cells cultured insuspension with serum-free mediumwith or without EGF or with selectiveEGFR inhibitor AG1478 (2 mmol/L),selective MEK inhibitor PD 98059 (50mmol/L), selective JNK inhibitor sidepopulation (SP) 600125 (100 nmol/L),or selective AKT inhibitor triciribine (2mmol/L) for 14 days. The vector cellswerealso cultured in suspensionwithserum-free medium. Compared withG3 control cells; �, P < 0.05; ��, P <0.01; n¼ 3; analyzed with t test. B, allthe suspendedcellswere cultured for2 days, lysised, and subjected toimmunoblotting with antibodies topEGFR, pERK,GSK-3b (S9P), Sca-1,Sox2, ALDH1, and b-actin. C, theeffects of expression of versican onbreast cancer cell invasion,metastasis, and tumorigenesis viaEGFR signaling. Overexpression ofversican in breast caner cellsenhances expression of pEGFR andits down-pathway pERK, pAKT.Expression of ERK enhances tumorcell migration, invasion, growth, andmetastasis. Expression of pAKTenhances GSK-3b (S9P), and somestem cell markers—Sca-1, Sox2,ALDH1, resulting in tumor cellenhanced chemical resistance, cellsurvival, and cell self-renewal. D, thevector- and G3-transfected 4T07cells were cultured in 10%FBS/DMEM with 40 mmol/L c2-ceramideor 2 mmol/L docetaxel for 24 hours,lysised, and subjected toimmunoblotting with antibodies toGSK-3b (S9P), Sca-1, Sox2, ALDH1,and b-actin.

A

C

**

**

****

0

5

10

15

20

25

30

35

40

45

WithoutG3Vector

EGF

AG

1478

PD

98059

SP

600125

TCN

Sphere

s/1

,000 c

ells

D Vector G3

Control C2 Doc

0.34 1.47 0.72 0.51

0.71 1.32 0.65 0.83

0.41 1.09 0.69 0.58

0.49 1.62 0.79 0.70

β-Actin

pGSK-3β(S9P)

Sca-1

ALDH1

SOX2

Vector G3

Control AG PD SP TCN

1478 98059 600125

B

pAKT

pEGFR

β-Actin

0.72 2.59 0.92 1.84 2.41 3.61

0.61 1.75 0.46 1.28 1.10 0.53

0.49 1.49 0.64 1.80 1.26 0.28

0.49 1.36 0.32 1.10 1.34 0.52

0.33 1.04 0.29 0.98 0.93 0.27

0.71 3.79 0.104 3.57 3.60 1.34

GSK-3β

(S9P)

Sca-1

ALDH1

SOX2

Expression of versican G3

EGFR

GSK-3β (S9P)

AG 1478

pAKT

Sca-1ALDH1SOX2

Chemical resistenceCell survivalCell self-renewal

TCN






are spherical colonies formed by epithelial cells when culturedon nonadherent surfaces, were enriched in stem/progenitorcells (30). Increasing expression of versican, as well as Sca-1,Sox2, and ALDH1, was observed in cultured mammo-spheres. Sca-1, the first identified mouse stem cell marker,is an anchored membrane protein expressed by differentprogenitor populations including the mammary gland pro-genitors (32). Sox2 and ALDH1 are also reported to be stemcell markers in human breast cancer cells. Themammosphere4T1 cells contained a high percentage of side population cellswhen compared with differentially cultured cells. An inter-esting observation included the high expression of versican inmammospheres. Both the expression of versican as well as theprevalence of side population cells were reduced as cellulardifferentiation occurred over time in serummedium culture.Structurally, all versican splice forms include an N-ter-

minal G1 domain and a C-terminus containing a selectin-

Vector siRNA

3 d

7 d

7 d

**

**

0

5

10

15

20

25

30

35

40

Small spheres Large

spheres

Sphere

s/1

,000 c

ells Vector

siRNA

0 50K 100K 150K200K 250K

<Indo blue-A>

0

50K

100K

150K

200K

250K

<D

AP

I/H

O342-A

>

1.12

0 50K 100K 150K 200K250K

<Indo blue-A>

0

50K

100K

150K

200K

250K

<D

AP

I/H

O342-A

>

1.58

Vector siRNA

Hoechst red

Hoechst

blu

e

BA

C

E

Vector

siRNA

**

0

10

20

30

40

50

60

70

80

Vector siRNA

Co

lon

ies/1

,00

0 c

ells

D

F

pAKT

pEGFR

β-Actin

pGSK-3 β

(S9P)

Sca-1

ALDH1

SOX2

Versican V1 (250 kDa)

1.59 0.72

1.06 0.57

2.11 0.48

0.94 0.51

1.05 0.66

1.60 0.71

1.12 0.38

Control siRNA

Figure 5. Expression of anti-versican shRNA decreased breastcancer cell mammosphere andcolony formation. A, typicalmammospheres of vector- andanti-versican shRNA-transfectedMDA-MB-231 cells cultured insuspension using serum-freemedium. B, numbers of small andlarge mammospheres. Comparedwith vector control group; �, P <0.05; ��, P < 0.01; n ¼ 5; analyzedwith t test. C, the typical prevalenceof Hoechstlow side population (SP)cells of vector- and anti-versicanshRNA-transfected MDA-MB-231cells shown by flow cytometry. D,the vector- and anti-versicanshRNA-transfected MDA-MB-231cells were cultured in 10%FBS/DMEM medium, lysised, andsubjected to immunoblotting withantibodies to versican V1, pEGFR,pAKT, GSK-3b (S9P), Sca-1, Sox2,ALDH1, and b-actin. E, typicalcolonies of vector- and anti-versican shRNA-transfected MDA-MB-231 cells formed in softagarose gel after culture for 4weeks. F, the numbers of coloniesformed on soft agarose gel.Compared with vector controlgroup; �, P < 0.05; ��, P < 0.01;n ¼ 3; analyzed with t test.

Table 1. Tumor formation of versican G3- andvector-transfected 66c14 cells in vivo

Cell number injected and tumors formed

1 � 102 1 � 103 1 � 104 1 � 105

Vector 0/8 0/8 3/8 8/8G3 0/8 4/8� 7/8� 8/8

NOTE: Versican G3- and vector-transfected 66c14 cellswere injected into the fourth mammary fat pad of BALB/cmice as a 10-fold dilution series from 1� 105 to 1� 102 cellspermouse.Allmicewere sacrificedafter 6weeks (�,P<0.05).The overall test for differences in stem cell frequenciesbetween the groups were calculated the website: http://bioinf.wehi.edu.au/software/elda/, P ¼ 0.00132).

Du et al.





like (G3) domain (33, 34). Given their ubiquitousness andhigh degree of conservation, it is likely that G1 and G3 playvital roles in proteoglycan functionality (35). The effects ofversican were greatly reduced when the G3 domain or EGF-like motifs within the G3 domain were deleted (36–38).This study showed that the versican G3 domain not onlyenhanced breast cancer cell mammosphere formation andcolony formation but also promoted tumor formation invivo, consistent with observations from our in vitro experi-ments.With as few as 1,000 cells, G3-transfected 66c14 cellscould form tumors in BALB/c mice, whereas the vectorcontrol cells required more than 10,000 cells before tumorformation. Versican G3-expressing human breast cancercells expressed higher levels of CD44, Sox2, ALDH1,Sca-1, and integrin-b1, whereas the G3-expressing mouse

mammary tumor cells also showed increased expression ofintegrin-b1 and Sca-1. Increased expression of breast cancercell self-renewal markers was associated with the increasedexpression of versican G3. This was supported by studiessilencing versican or the versican G3 domain's functionality.Using shRNA or G3 untranslated region (UTR) reduced theeffects on mammosphere and colony formation corroborat-ing versican G3 domain's role in enhancing breast cancer cellself-renewal.Given the frequency at which abnormalities in EGFR

signaling are present in human breast cancer, EGFRhas beenan attractive target for therapeutic manipulation (39–42).Versican-promoted breast tumor invasion is influenced byits enhanced expression of EGFR signaling (25). In our studyof stem cell renewal, we observed that negating the effects of

Figure 6. Expression of versicanpromoted breast cancer cell tumorformation and self-renewalin vivo. Shown are hematoxylinand eosin (H&E) staining andimmunohistochemistry of primarytumors arising from G3-transfected66c14 cell inoculations tested withantibodies to versican G3, pEGFR,pAKT, GSK-3b (S9P), Sca-1, Sox2,ALDH1.

Sca-1

100 μm

pEGFR

pAKT

pGSK-3 β (S9P)

ALDH1

Sox2

G3 tumorVector tumorG3 tumorVector tumor

Versican G3






EGF-like motifs using a versican G3 fragment (G3DEGF)reduced versican G3 effects on breast cancer cell mammo-sphere and colony formation. Mammosphere formationassays showed that versican G3-enhanced breast cancerself-renewal was related to enhanced expression of keymarkers in the EGFR/AKT/GSK-3b (S9P) pathway. Bothselective EGFR inhibitor AG 1478 and selective AKTinhibitor triciribine could block this signaling pathway andprevent the versican G3-observed effects on mammarycancer cell mammosphere formation and expression ofGSK-3b (S9P), Sox2, ALDH1, and Sca-1.Moreover, reduc-ing the expression of GSK-3b (S9P) using C2-ceramide ordocetaxel also inhibited versican-enhanced expression ofstem cell markers—Sox2, ALDH1, and Sca-1. G30s effecton in vivo tumor formation was associated with changes toboth EGFR/AKT/GSK-3b (S9P) signaling and the afore-mentioned tumor cell self-renewal markers. EGFR/AKT/GSK-3b (S9P) signaling and breast cancer cell self-renewalmarkers—Sox2, ALDH1, and Sca-1—were observed to bemuch higher in primary tumors of G3-treated mice ascompared with those of the vector control group.In this study, we also observed that versican G3-promoted

breast cancer cell self-renewal conferred enhanced resistanceto chemotherapeutic drug therapy using clinical agents suchas doxorubicin and epirubicin. Recent studies have proposedthat cancer-initiating cells may be more resistant to irradi-ation than other cells in the tumor population (1, 43, 44).There are also reports that cancer-initiating cells can bemoreresistant to chemotherapeutic drugs because of increasedexpression of antiapoptotic proteins or increased expressionof the ATPase drug efflux pumpABCG2/5 (1, 45–47). It hasbeen reported that when EGFR targeting therapy wascoadministered with cytotoxic chemotherapy agents orradiotherapy, additive or even synergistic antitumor effectswere observed (26, 27, 48). The study of versican as apotential treatment target merits further preclinical study.Versican is highly expressed in breast cancer progenitor

cells and its expression is maintained at high levels beforecellular differentiation. Silencing versican expression wasshown to inhibit breast cancer self-renewal. Overexpression

of the versican G3 domain enhanced breast cancer self-renewal through EGFR/AKT/GSK-3b (S9P) signaling andconferred enhanced resistance to the chemotherapeuticdrugs. Findings from our mechanistic study complementclinical literature recognizing the importance of versican incancer invasiveness and metastasis. Strategies designed totarget versican-mediated breast cancer self-renewal may leadto therapies and/or adjuncts benefiting patients with breastcancer.

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

Authors' ContributionsConception and design:W.W. Du, L. Fang, X. Yang, W. Sheng, B.L. Yang, A. Seth,Y. Zhang, B.B. Yang, A.J. YeeDevelopment of methodology:W.W.Du, L. Fang, A. Seth, Y. Zhang, B.B. Yang, A.J. YeeAcquisition of data (provided animals, acquired and managed patients, providedfacilities, etc.): W.W. Du, L. Fang, A. Seth, Y. Zhang, B.B. Yang, A.J. YeeAnalysis and interpretation of data (e.g., statistical analysis, biostatistics, compu-tational analysis): W.W. Du, L. Fang, A. Seth, Y. Zhang, B.B. Yang, A.J. YeeWriting, review, and/or revision of the manuscript:W.W. Du, L. Fang, A. Seth, Y.Zhang, B.B. Yang, A.J. YeeAdministrative, technical, or material support (i.e., reporting or organizing data,constructing databases): W.W. Du, A. Seth, Y. Zhang, B.B. Yang, A.J. YeeStudy supervision: L. Fang, X. Yang, W. Sheng, B.L. Yang, A. Seth, Y. Zhang, B.B.Yang, A.J. Yee

AcknowledgmentsThe authors thank Ms. Gisele Knowles (Cytometry and Microscopy Core Facility

of Sunnybrook Health Sciences Centre) for her assistance with the cell-cycle assays,Miss Sarah Davies and Mr. Weining Yang (Division of Orthopaedics, SunnybrookHealth Sciences Centre) for assistance with the article submission and editing.

Grant SupportThis work was partly supported by a grant fromNational Sciences and Engineering

ResearchCouncil ofCanada (227937-01) to B.B. Yang, who is the recipient of aCareerInvestigator Award (CI5958) from the Heart and Stroke Foundation of Ontario.Funding, in part, fort his work was also supported by a grant from the HollandMusculoskeletal Program, Sunnybrook Health Sciences Centre (to A.J. Yee).

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be herebymarked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

Received July 30, 2012; revised December 6, 2012; accepted December 10, 2012;published OnlineFirst February 28, 2013.

References1. Hill RP, Perris R. "Destemming" cancer stem cells. J Natl Cancer Inst

2007;99:1435–40.2. Ponti D, Zaffaroni N, Capelli C, DaidoneMG. Breast cancer stem cells:

an overview. Eur J Cancer 2006;42:1219–24.3. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF.

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

4. Smalley MJ, Clarke RB. The mammary gland "side population": aputative stem/progenitor cell marker? J Mammary Gland Biol Neo-plasia 2005;10:37–47.

5. Hadnagy A, Gaboury L, Beaulieu R, Balicki D. SP analysis may beused to identify cancer stem cell populations. Exp Cell Res 2006;312:3701–10.

6. Al-Hajj M, Clarke MF. Self-renewal and solid tumor stem cells. Onco-gene 2004;23:7274–82.

7. Rafii S, Lyden D. Therapeutic stem and progenitor cell transplantationfor organ vascularization and regeneration. Nat Med 2003;9:702–12.

8. Reya T,Morrison SJ, ClarkeMF,Weissman IL. Stem cells, cancer, andcancer stem cells. Nature 2001;414:105–11.

9. Clarke MF. A self-renewal assay for cancer stem cells. Cancer Che-mother Pharmacol 2005;56(Suppl 1):64–8.

10. Dontu G, Al-Hajj M, Abdallah WM, Clarke MF, WichaMS. Stem cells innormal breast development and breast cancer. Cell Prolif 2003;36(Suppl 1):59–72.

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

12. Bhowmick NA, Chytil A, Plieth D, Gorska AE, Dumont N, Shap-pell S, et al. TGF-beta signaling in fibroblasts modulates theoncogenic potential of adjacent epithelia. Science 2004;303:848–51.

13. Taipale J, Beachy PA. The Hedgehog and Wnt signalling pathways incancer. Nature 2001;411:349–54.

Du et al.





14. Rahmani M, Wong BW, Ang L, Cheung CC, Carthy JM, Walinski H,et al. Versican: signaling to transcriptional control pathways. Can JPhysiol Pharmacol 2006;84:77–92.

15. Dours-Zimmermann MT, Zimmermann DR. A novel glycosaminogly-can attachment domain identified in two alternative splice variants ofhuman versican. J Biol Chem 1994;269:32992–8.

16. Wu YJ, La Pierre DP, Wu J, Yee AJ, Yang BB. The interaction ofversican with its binding partners. Cell Res 2005;15:483–94.

17. Yee AJ, Akens M, Yang BL, Finkelstein J, Zheng PS, Deng Z, et al. Theeffect of versican G3 domain on local breast cancer invasiveness andbony metastasis. Breast Cancer Res 2007;9:R47.

18. Ricciardelli C, Brooks JH, Suwiwat S, Sakko AJ, Mayne K, RaymondWA, et al. Regulation of stromal versican expression by breastcancer cells and importance to relapse-free survival in patientswith node-negative primary breast cancer. Clin Cancer Res 2002;8:1054–60.

19. Suwiwat S, Ricciardelli C, Tammi R, Tammi M, Auvinen P, Kosma VM,et al. Expression of extracellular matrix components versican, chon-droitin sulfate, tenascin, and hyaluronan, and their association withdisease outcome in node-negative breast cancer. Clin Cancer Res2004;10:2491–8.

20. Delpech B, Girard N, Olivier A, Maingonnat C, van Driessche G, vanBeeumen J, et al. The origin of hyaluronectin in human tumors. IntJ Cancer 1997;72:942–8.

21. Zheng PS, Wen J, Ang LC, Sheng W, Viloria-Petit A, Wang Y, et al.Versican/PG-MG3domain promotes tumor growth and angiogenesis.FASEB J 2004;18:754–6.

22. Theocharis AD. Versican in health and disease. Connect Tissue Res2008;49:230–4.

23. Wu Y, Chen L, Cao L, Sheng W, Yang BB. Overexpression of the C-terminal PG-M/versican domain impairs growth of tumor cells byintervening in the interaction between epidermal growth factor recep-tor and beta1-integrin. J Cell Sci 2004;117:2227–37.

24. Zheng PS, Vais D, Lapierre D, Liang YY, Lee V, Yang BL, et al. PG-M/versican binds to P-selectin glycoprotein ligand-1 and mediates leu-kocyte aggregation. J Cell Sci 2004;117:5887–95.

25. Wu Y, Zhang Y, Cao L, Chen L, Lee V, Zheng PS, et al. Identification ofthe motif in versican G3 domain that plays a dominant-negative effecton astrocytoma cell proliferation through inhibiting versican secretionand binding. J Biol Chem 2001;276:14178–86.

26. Arteaga C. Targeting HER1/EGFR: a molecular approach to cancertherapy. Semin Oncol 2003;30:3–14.

27. DuWW, Yang BB, Yang BL, Deng Z, Fang L, Shan SW, et al. VersicanG3 domain modulates breast cancer cell apoptosis: a mechanism forbreast cancer cell response to chemotherapy andEGFR therapy. PLoSONE 2011;6:e26396.

28. Paulus W, Baur I, Dours-Zimmermann MT, Zimmermann DR. Differ-ential expression of versican isoforms in brain tumors. J NeuropatholExp Neurol 1996;55:528–33.

29. Zucchi I, Sanzone S, Astigiano S, Pelucchi P, Scotti M, Valsecchi V,et al. The properties of a mammary gland cancer stem cell. Proc NatlAcad Sci U S A 2007;104:10476–81.

30. Dontu G, Abdallah WM, Foley JM, Jackson KW, Clarke MF, Kawa-mura MJ, et al. In vitro propagation and transcriptional profiling ofhuman mammary stem/progenitor cells. Genes Dev 2003;17:1253–70.

31. Welm B, Behbod F, Goodell MA, Rosen JM. Isolation and character-ization of functional mammary gland stem cells. Cell Prolif 2003;36(Suppl 1):17–32.

32. Grange C, Lanzardo S, Cavallo F, Camussi G, Bussolati B. Sca-1identifies the tumor-initiating cells in mammary tumors of BALB-neuTtransgenic mice. Neoplasia 2008;10:1433–43.

33. Zako M, Shinomura T, Kimata K. Alternative splicing of the unique"PLUS" domain of chicken PG-M/versican is developmentally regu-lated. J Biol Chem 1997;272:9325–31.

34. Sandell LJ. In situ expression of collagen and proteoglycan genes innotochord and during skeletal development and growth. Microsc ResTech 1994;28:470–82.

35. Yang BL, Cao L, Kiani C, Lee V, Zhang Y, Adams ME, et al. Tandemrepeats are involved inG1domain inhibitionof versican expression andsecretion and the G3 domain enhances glycosaminoglycan modifica-tion and product secretion via the complement-binding protein-likemotif. J Biol Chem 2000;275:21255–61.

36. Zhang Y, Cao L, Yang BL, Yang BB. The G3 domain of versicanenhances cell proliferation via epidermial growth factor-like motifs.J Biol Chem 1998;273:21342–51.

37. YangBL,YangBB,ErwinM,AngLC, Finkelstein J,YeeAJ. VersicanG3domain enhances cellular adhesion and proliferation of bovine inter-vertebral disc cells cultured in vitro. Life Sci 2003;73:3399–413.

38. Du WW, Yang BB, Shatseva TA, Yang BL, Deng Z, Shan SW, et al.Versican G3 promotes mouse mammary tumor cell growth, migration,and metastasis by influencing EGF receptor signaling. PLoS ONE2010;5:e13828.

39. Mendelsohn J, Baselga J. The EGF receptor family as targets forcancer therapy. Oncogene 2000;19:6550–65.

40. Prenzel N, Zwick E, Leserer M, Ullrich A. Tyrosine kinase signalling inbreast cancer. Epidermal growth factor receptor: convergence pointfor signal integration and diversification. Breast Cancer Res 2000;2:184–90.

41. O'Donovan N, Crown J. EGFR and HER-2 antagonists in breastcancer. Anticancer Res 2007;27:1285–94.

42. Hutcheson IR, Knowlden JM, Jones HE, Burmi RS, McClelland RA,Barrow D, et al. Inductive mechanisms limiting response to anti-epidermal growth factor receptor therapy. Endocr Relat Cancer2006;13(Suppl 1):S89–97.

43. BaoS,WuQ,McLendonRE,HaoY,ShiQ,HjelmelandAB, et al. Gliomastem cells promote radioresistance by preferential activation of theDNA damage response. Nature 2006;444:756–60.

44. Phillips TM, McBride WH, Pajonk F. The response of CD24(�/low)/CD44þ breast cancer-initiating cells to radiation. J Natl Cancer Inst2006;98:1777–85.

45. Frank NY, Margaryan A, Huang Y, Schatton T, Waaga-Gasser AM,Gasser M, et al. ABCB5-mediated doxorubicin transport and che-moresistance in human malignant melanoma. Cancer Res 2005;65:4320–33.

46. Liu G, Yuan X, Zeng Z, Tunici P, Ng H, Abdulkadir IR, et al. Analysis ofgene expressionandchemoresistanceofCD133þ cancer stemcells inglioblastoma. Mol Cancer 2006;5:67.

47. Doyle LA, Ross DD. Multidrug resistance mediated by the breast can-cer resistance protein BCRP (ABCG2). Oncogene 2003;22:7340–58.

48. Sirotnak FM. Studies with ZD1839 in preclinical models. Semin Oncol2003;30:12–20.






2013;11:443-455. Published OnlineFirst February 28, 2013.Mol Cancer Res William Weidong Du, Ling Fang, Xiangling Yang, et al. The Role of Versican in Modulating Breast Cancer Cell Self-renewal

Updated version

10.1158/1541-7786.MCR-12-0461doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2013/02/28/1541-7786.MCR-12-0461.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/11/5/443.full#ref-list-1

This article cites 48 articles, 17 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/11/5/443.full#related-urls

This article has been cited by 2 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/11/5/443To 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-12-0461

http://mcr.aacrjournals.org/content/suppl/2013/02/28/1541-7786.MCR-12-0461.DC1

http://mcr.aacrjournals.org/content/11/5/443.full#ref-list-1

http://mcr.aacrjournals.org/content/11/5/443.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/11/5/443


the role of versican in modulating breast cancer cell self-renewal · tion of cells that function...

Documents