the her2- and heregulin b1 (hrg) inducible tnfr...

Oncogenes and Tumor Suppressors

The HER2- and Heregulin b1 (HRG)–Inducible TNFRSuperfamily Member Fn14 Promotes HRG-Driven BreastCancer Cell Migration, Invasion, and MMP9 Expression

Kaushal Asrani1,2, Ruth A. Keri3, Rebeca Galisteo1,2, Sharron A.N. Brown1,2, Sarah J. Morgan1,2,Arundhati Ghosh1,2, Nhan L. Tran4, and Jeffrey A. Winkles1,2

AbstractHER2 overexpression occurs in 15% to 20% of all breast cancers and is associated with increased metastatic

potential and poor patient survival. Abnormal HER2 activation, either throughHER2 overexpression or heregulin(HRG):HER3 binding, elicits the formation of potent HER2–HER3 heterodimers and drives breast cancer cellgrowth and metastasis. In a previous study, we found that fibroblast growth factor-inducible 14 (Fn14), a memberof the TNF receptor superfamily, was frequently overexpressed in human HER2þ breast tumors. We report herethat HER2 and Fn14 are also coexpressed in mammary tumors that develop in two different transgenic mousemodels of breast cancer. In consideration of these findings, we investigated whether HER2 activation in breastcancer cells could directly induce Fn14 gene expression. We found that transient or stable transfection of MCF7cells with a HER2 expression plasmid increased Fn14 protein levels. Also, HRG1-b1 treatment of MCF7 cellstransiently induced Fn14 mRNA and protein expression. Both the HER2- and HRG1-b1–induced increase inFn14 expression in MCF7 cells as well as basal Fn14 expression in HER2 gene-amplified AU565 cells could bereduced byHER2 kinase inhibition with lapatinib or combinedHER2 andHER3 depletion using siRNA.We alsoreport that Fn14-depleted, HER2-overexpressing MCF7 cells have reduced basal cell migration capacity andreduced HRG1-b1–stimulated cell migration, invasion, and matrix metalloproteinase (MMP)-9 expression.Together, these results indicate that Fn14 may be an important downstream regulator of HER2/HER3–drivenbreast cancer cell migration and invasion. Mol Cancer Res; 11(4); 393–404. �2013 AACR.

IntroductionBreast cancer is the most common malignancy in women

world-wide and in the United States approximately 40,000women are predicted to die from this disease in 2012 (1).HER2, a member of the EGF receptor (EGFR) family, isoverexpressed in 15% to 20% of invasive breast cancers andis associated with aggressive tumor behavior, decreased timeto relapse, and poor clinical outcome (2). HER3, a kinase-inactive member of the EGFR family (3), is often expressedin HER2þ breast tumors (4) and it is the preferred hetero-dimerization partner for HER2 (5). Indeed, HER3 has beenshown to play a pivotal role in mediating both HER2

oncogenesis (6) and the resistance of breast cancers toHER2-targeted therapies (7). Heregulin-1 (HRG1; alsocalled neuregulin-1), a member of the EGF family (8), isa HER3 and HER4 ligand that elicits the formation ofpotent HER2–HER3 heterodimers (9). HRG1 is expressedin breast tumors (8) and has been implicated in modulatingbreast cancer cell invasion (10–12) and metastasis (13).Thus, the HER2–HER3 receptor pair forms a very potentmitogenic and transforming unit that promotes breasttumorigenesis.Cancer cell migration and invasion is controlled by

complex signaling events that regulate cytoskeletal changes,cell–cell and cell–extracellularmatrix interactions, and extra-cellular proteolytic activity (14).HER2/HER3heterodimer-driven cellular migration and invasion has been previouslyshown to require phosphoinositide 3-kinase (PI3K) activity(15) and a variety of downstream mediators have beenimplicated in these cellular processes [e.g., FAK-Src (16)and matrix metalloproteinase (MMP)-9 (17)]. In regard tothe MMPs, they are known to play an important role in thetumor microenvironment by mediating effects such asremodeling of the extracellular matrix, tissue invasion andintravasation, inflammation, and angiogenesis (18). HER2/HER3 signaling has been reported to upregulate MMPproduction and activity (17, 19, 20), but the biologic basisof this effect remains poorly understood.

Authors' Affiliations: 1Department of Surgery, Center for Vascular andInflammatory Diseases; 2Marlene and Stewart Greenebaum Cancer Cen-ter, University of Maryland School of Medicine, Baltimore, Maryland;3Department of Pharmacology, School ofMedicine, CaseWesternReserveUniversity, Cleveland, Ohio; and 4Cancer and Cell Biology Division, Trans-lational Genomics Research Institute, Phoenix, Arizona

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

Corresponding Author: Jeffrey A. Winkles, Department of Surgery, Uni-versity ofMarylandSchool ofMedicine, 800West Baltimore St., Room320,Baltimore, MD 21201. Phone: 410-706-8172; Fax: 410-706-8234; E-mail:[email protected]

doi: 10.1158/1541-7786.MCR-12-0542

�2013 American Association for Cancer Research.

MolecularCancer

Research

www.aacrjournals.org 393

on June 28, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 1, 2013; DOI: 10.1158/1541-7786.MCR-12-0542

http://mcr.aacrjournals.org/

TNF-like weak inducer of apoptosis (TWEAK) andfibroblast growth factor (FGF)-inducible 14 (Fn14) are aTNF superfamily ligand–receptor pair involved in manycellular processes associated with wound repair, includinginflammation and angiogenesis (21, 22). TWEAK:Fn14binding promotes Fn14 association with adaptor moleculescalled TNF receptor-associated factors (TRAF), whichcouple to various signaling pathways such as the NF-kB,mitogen-activated protein kinase (MAPK), and PI3K/Aktpathways (21). The TWEAK-Fn14 signaling axis playsan important role in regulating various aspects of tumorbehavior such as growth, survival, invasion, and angio-genesis (23–29). Importantly, Fn14 is highly expressed inmany solid tumors, and in some tumor types elevatedFn14 levels have been shown to correlate with diseaseprogression and poor patient outcome (23, 25, 28). Thereis also evidence that Fn14 overexpression in tumorcells can activate signaling pathways and stimulatecellular processes; for example, ectopic Fn14 expressionpromotes glioma cell invasion by activating Rac1 and NF-kB (25).In a previous study we showed that elevated Fn14 expres-

sion strongly correlated with the HER2þ/ER� intrinsicsubtype of breast cancer and with clinical indicators ofpoor prognosis (26). However, the mechanisms underlyingpreferential Fn14 expression in HER2þ breast tumors, aswell as the potential contributions of Fn14 to HER2-mediated disease, have yet to be elucidated. Here, we reportthat Fn14 expression is increased in mouse mammary tumorvirus (MMTV)-c-Neu and MMTV-polyoma middle Tantigen (PyMT) transgenic mouse breast tumors with ele-vated Neu (HER2) levels. Also, both HER2 overexpressionin MCF7 breast cancer cells and HRG1-b1 treatment ofMCF7 cells induces Fn14 gene expression, and these effectsare dependent on HER2/HER3 signaling. Finally, we showthat stable knockdown of Fn14 in HER2-overexpressingMCF7 cells decreases basal cell migration capacity andHRG1-b1–stimulated migration, invasion, and MMP-9expression.

Materials and MethodsTransgenic mouse modelsMMTV-c-Neu mice (FVB/N-Tg(MMTV-neu)202

Mul/J; ref. 30) were purchased from Jackson Laboratories.These mice were bred and mammary tissue samples isolatedas previously described (31). All MMTV-c-Neu animalstudies were approved by the Case Western Reserve Uni-versity (Cleveland, OH) Institutional Animal Care and UseCommittee. The MMTV-PyMT mice (FVB/N-Tg(MMTV-PyVT)634 Mul/J; refs. 32, 33) were also pur-chased from Jackson Laboratories. Male hemizygous trans-genic mice were bred to FVB/N females and at various timepoints wild-type and hemizygous littermates were selected,euthanized, and 5 mammary fat pad pairs were isolated, andthen frozen until use. All MMTV-PyMT animal studieswere approved by the University of Maryland School ofMedicine (Baltimore, MD) Institutional Animal Care andUse Committee.

Cell culture and treatmentsCell lines were obtained from the following sources:

MCF7, BT474, SKBR3, MDA-MB-453, AU565, andNIH3T3 (American Type Culture Collection), MCF7/HER2 (Dr. Dihua Yu, University of Texas MD AndersonCancer Center, Houston, TX),MCF7/HER2-18 (Dr. AnneHamburger, University of Maryland School of Medicine),NIH3T3/HER2 (Dr. Peter Choyke, NIH, Bethesda, MD),and MCF7 Ca/LTLT-Ca (Dr. Angela Brodie, University ofMaryland School of Medicine). MCF7, MCF7/HER2,BT474, SKBR3, and MDA-MB-453 cells were maintainedin Dulbecco's modified Eagle's medium (Cellgro) andAU565, NIH3T3, NIH3T3/HER2, and MCF7/HER2-18 cells were maintained in RPMI-1640 (Cellgro). Both cellmediums were supplemented with 10% FBS (HyClone), 2mmol/L L-glutamine and 1% penicillin–streptomycin.MCF7/HER2 andMCF7/HER2-18 cells were additionallymaintained in 750 or 500 mg/mL G418 (Cellgro), respec-tively. Lentivirus-infected MCF7/HER2-18 cells wereadditionally maintained in 0.5 mg/mL puromycin (Cellgro).Fn14 short hairpin RNA (shRNA)-448 cells expressingmyc epitope-tagged Fn14 were additionally maintained in1mg/mL blasticidine (Sigma).MCF7Ca and LTLT-Ca cellswere grown as previously described (34). Cells were treatedwith the indicated concentrations of U0126, wortmannin(both from Cell Signaling Technology), lapatinib (LC Lab-oratories), MMP-2/MMP-9 inhibitor IV (SB-3CT; Calbio-chem), MK-2206 (Alexis Corporation), EGF, HB-EGF,BTC, HRG1-a, or HRG1-b1 (all from R&D Systems).

Western blot analysisWestern blotting was carried out as previously described

(35). The following primary antibodies were used: Fn14, p-HER2 (Tyr1248), p-HER3 (Tyr1289), p-Erk1/2 (Thr202/Tyr204), Erk1/2, p-Akt (S473), Akt, p-p90RSK (Ser380),p90RSK, p-p70 S6 kinase (Thr389), p70 S6 kinase, glyc-eraldehyde-3-phosphate dehydrogenase (GAPDH; all fromCell Signaling Technology), Neu, ErbB3, ErbB4 (all fromSanta Cruz Biotechnology), EGFR, Myc, and tubulin (allfrom Millipore).

FACS analysisFlow cytometry was conducted using phycoerythrin- anti-

Fn14 mAb ITEM-4 and immunoglobulin G (IgG)3 isotypecontrol (eBioscience Inc.) as previously described (26).

RNA isolation and quantitative real-time RT-PCR assaysTotal cellular RNA was extracted using the RNeasy kit

(Qiagen) as previously described (36). RNA was convertedto cDNA using the ProtoScript AMVLongAmpTaq reversetranscription-PCR (RT-PCR) Kit (New England Biolabs)according to manufacturer's instructions. Fn14 andGAPDH mRNA levels were quantified using an ABIPrism7900HTReal-time PCR system (Applied Biosystems)and the following primers and probes: Fn14, Cat.#Hs00171993_m1; MMP-9, Cat.# Hs00234579_m1;GAPDH, Cat.# Hs99999905_m1 (TaqMan Gene Expres-sion Assay, Applied Biosystems).

Asrani et al.

Mol Cancer Res; 11(4) April 2013 Molecular Cancer Research394




Plasmid DNA and siRNA transfectionsMCF7 cells were transiently transfected with 2 or 4 mg of

the pcDNA6-HER2 expression plasmid [provided by Dr.Mien-Chie Hung (University of Texas MD AndersonCancer Center)] per 60-mm dish using the Effectene Trans-fection Kit (Qiagen) according to the manufacturer'sinstructions. Cells were harvested 24 hours posttransfectionfor Western blot analysis. In other experiments, Fn14shRNA-448 cells were transfected with pcDNA6 vector orpcDNA6-Fn14-myc plasmid (37) as above, and drug-resis-tant cell colonies were selected using blasticidine. Positivelytransfected clones were expanded and screened for Myclevels by Western blot analysis. For siRNA experiments,MCF7, MCF7/HER2, MCF7/HER2-18, or AU565 cellswere plated at a density of 2� 105 in 60-mm dishes, and 24hours later the cells were transfected with either Allstarsnonsilencing control, luciferase control,HER4,HER2, and/or HER3 siRNA duplexes (Qiagen) at a concentration of 25nmol/L using the Lipofectamine 2000 transfection reagentas previously described (26). Cells were harvested 48 hoursposttransfection for Western blot analysis.

Lentiviral transduction and isolation of Fn14-depletedcell linesLentiviral constructs encoding control, nontarget shRNA

or shRNAs targeting 2 different regions of the Fn14 tran-script (Fn14 shRNA#448, Cat.# TRCN0000072448 andFn14 shRNA#562, Cat.# TRCN0000222562) were pur-chased from Sigma. Lentiviral packaging was conducted aspreviously described (27).MCF7/HER2-18 cells were trans-duced with the various lentiviruses and drug-resistant cellcolonies were selected using puromycin. Positively trans-duced clones were expanded and screened for Fn14 levels byWestern blot analysis.

Scratch wound assaysParental MCF7, MCF7/HER2-18, MCF7/HER2-18

control shRNA, and MCF7/HER2-18 Fn14 shRNA cellswere plated in triplicate in 6-well cluster dishes and allowedto grow to confluence in normal RPMI growth medium(10%FBS). The confluent cell monolayers were scratched inthe shape of a cross using a 200 mL pipette tip and normalRPMI growth medium (10% FBS) or low serum medium(0.5%) containing HRG1-b1 (50 or 200 ng/mL as indi-cated) was added to the cells. Wound closure was monitoredover a 24-hour time period and photographs were taken atthe 4 intersecting edges of the cross. Wound width at 0 and24 hours was measured and the difference plotted as per-centage wound closure.

Invasion assaysParental MCF7, MCF7/HER2-18, MCF7/HER2-18

control shRNA, MCF7/HER2-18 Fn14 shRNA-562, andMCF7/HER2-18 Fn14 shRNA-448 cells (2 � 105), andMCF7/HER2-18 Fn14 shRNA-448 cells transfected withvector or Fn14-myc plasmid (1 � 105) were suspended inRPMI containing 0.5% FBS and seeded in the upperchamber of 24-well BD Biocoat Matrigel Invasion Cham-

bers (BD Biosciences). Lower chambers contained RPMIwith 0.5% FBS and HRG1-b1 (50 or 200 ng/mL asindicated). After 48 hours of incubation, the cells were fixedand stained as previously described (27). Cells were countedfrom 25 high-power fields per filter.

Gelatin zymographyMCF7/HER2-18 control shRNA cells, Fn14 shRNA-

448 cells, and Fn14 shRNA-448 cells transfected with vectoror Fn14-myc expression plasmid were plated in 100-mmdishes, serum-starved for 24 hours in serum-free, phenol red-free RPMI medium, and then treated with 50 ng/mLHRG1-b1. Conditioned media was collected at 0, 12, and24 hours following treatment and concentrated using Ami-con Ultra-15 Centrifugal Filter Units (Millipore). Proteinconcentrations were determined using the BCA proteinassay (Thermo Fisher Scientific). Equal amounts of proteinwere subjected to gelatin zymography using Novex 10%Zymogram gels (Invitrogen) according to manufacturer'sinstructions.

ResultsFn14 and HER2 are coexpressed in MMTV-c-Neu andMMTV-PyMT transgenic mouse breast tumorsPrevious reports have indicated that Fn14 is highly

expressed in human breast tumors (26, 38, 39); and inparticular, it has been shown that Fn14 is most frequentlyoverexpressed in theHER2þ breast cancer subtype (26, 39).To investigate this further, we analyzed Fn14 levels in theMMTV-c-Neu transgenic mouse model of Neu (HER2)-induced tumorigenesis. These mice express the rat c-Neuproto-oncogene under transcriptional control of theMMTVpromoter/enhancer and develop spontaneous mammarytumors (30, 31). Fn14 expression was detected in all of theHER2þ mammary tumors examined with no expressionseen in age-matched wild-type mammary glands (Fig. 1A).HER3 expression was also upregulated in the tumors,consistent with a previous report (40). We also examinedFn14 levels in theMMTV-PyMTmodel of breast cancer. Inthis model, mammary hyperplasia is detected at approxi-mately 4 weeks of age and advanced carcinoma at approx-imately 12 weeks (32, 33). It has been reported that Neu(HER2) expression is activated in PyMT-driven mammarytumors, with the highest HER2 expression levels in carci-noma stage tumors (33). We detected Fn14 expression inboth wild-type mammary glands and PyMT tumors isolatedfrom 4- to 10-week-oldmice, but Fn14 levels were highest inthe 8- and 10-week-old HER2þ mammary tumor samples(Fig. 1B).

HER2 overexpression in human breast cancer cell linesand murine fibroblasts increases Fn14 levelsTo investigate whether HER2 overexpression in breast

cancer cells could directly upregulate Fn14 gene expression,we used MCF7 cells as they normally express relatively lowlevels of HER1 and HER2, moderate levels of Fn14 andHER4, and high levels of HER3 (Supplementary Fig. S1).We found that transient transfection of these cells with a

Fn14 Regulates HRG-Driven Cell Invasion

www.aacrjournals.org Mol Cancer Res; 11(4) April 2013 395




HER2 expression plasmid increased Fn14 protein levels(Fig. 2A). We then compared Fn14 levels in parentalMCF7 cells and a HER2-overexpressing stably transfectedMCF7 cell line (MCF7/HER2). HER2 overexpression inthis cell line activated both HER2 and HER3 as shown byWestern blotting with phospho(p)-HER2 and p-HER3antibodies (Fig. 2B). Fn14 expression was elevated in theMCF7/HER2 cells, as measured by both Western blotanalysis (Fig. 2C) and fluorescence-activated cell sorting(FACS; Fig. 2D) analysis. In this Western blot analysisand in several others to follow, 2 Fn14 species can bedetected. These species represent full-length Fn14 and anN-terminal truncated Fn14 isoform (unpublished data).We determined whether HER2 overexpression in MCF7cells was increasing Fn14 mRNA expression by quanti-

tative real-time RT-PCR (qRT-PCR) analysis, and foundthat Fn14 mRNA levels were not significantly increased inMCF7/HER2 cells compared with MCF7 cells (Supple-mentary Fig. S2). HER2 overexpression also increasedFn14 levels in another stably transfected MCF7 cell line(MCF7/HER2-18; Fig. 2E). Finally, we compared Fn14levels in MCF7 cells stably transfected with an aromataseexpression plasmid versus their HER2-overexpressingletrozole-resistant counterpart (34) and in parental murineNIH3T3 fibroblasts versus a HER2-overexpressing stablytransfected NIH3T3 cell line. In both cases, HER2 over-expression increased Fn14 levels (Supplementary Fig. S3).

Inhibition of HER2 kinase activity, combinatorialHER2–HER3 depletion, and inhibition of MEK1/2 andPI3K signaling downregulates Fn14 expression inHER2-overexpressing MCF7 and AU565 breast cancercellsWe tested whether HER2 tyrosine kinase activity was

critical for HER2-mediated Fn14 upregulation using lapa-tinib, a tyrosine kinase inhibitor that targets bothHER1 andHER2 (41). Lapatinib treatment of MCF7/HER2 cellsinhibited HER2 and HER3 phosphorylation and signifi-cantly downregulated Fn14 expression (Fig. 3A). We alsoused RNA interference to deplete HER2 and HER3 expres-sion inMCF7/HER2 cells. Cells were transiently transfectedwith control siRNA or HER2 and HER3 siRNA alone or incombination. Combined HER2–HER3 depletion, but notHER2 or HER3 depletion alone, markedly inhibited Fn14expression (Fig. 3B). Lapatinib treatment and combinedHER2–HER3 depletion also decreased Fn14 expression inthe MCF7/HER2-18 cell line (Supplementary Fig. S4).AU565 is a breast cancer cell line with HER2 gene ampli-fication that normally expresses high levels of HER2 andHER3 (Supplementary Fig. S1) and constitutive HER2 andHER3 tyrosine phosphorylation can be detected in thesecells (Fig. 3C). Lapatinib treatment and HER2, HER3, orboth HER2 and HER3 depletion significantly decreasedFn14 expression in AU565 cells (Fig. 3C and D). HER4depletion in MCF7/HER2 or AU565 cells did not alterFn14 expression (data not shown).HER2 overexpression in MCF7 (Supplementary Fig.

S5A and Fig. 3B) and AU565 (Fig. 3D) cells leads toconstitutive activation of the MEK/ERK and PI3K/Aktsignaling cascades. Therefore, we analyzed the effects ofHER2 and/or HER3 depletion on p-Erk and p-Akt levelsto investigate why HER2 or HER3 depletion alonedecreased Fn14 levels in AU565 cells but not MCF7/HER2 or MCF7/HER2-18 cells. We found that HER2or HER3 depletion alone had a much greater effect on p-Akt levels in the AU565 cells compared with the MCF7/HER2 cells (Fig. 3B and D). To determine the relativecontribution of the MEK/ERK and PI3K/Akt pathwaysin regulating Fn14 expression in these 2 cell lines wetreated cells with either the MAP/extracellular signal–regulated kinase (MEK)1/2 inhibitor U0126 or the PI3Kinhibitor wortmannin. Each of these inhibitors reducedFn14 expression levels in both cell lines, but Fn14

Wild-type

Wild

-type

mammary

gland

MMTV-c-Neu

MM

TV-P

yMT

Wild

-type

MM

TV-P

yMT

Wild

-type

MM

TV-P

yMT

Wild

-type

MM

TV-P

yMT

mammary

tumor

Tubulin

Fn14

HER2

HER3

A

HER2

GAPDH

Fn14

B

Age (wks): 4 4 6 6 8 8 10 10

Figure 1. Fn14 is expressed inMMTV-c-Neu andMMTV-PyMT transgenicmouse breast tumors. A, 5 normal mammary glands and 3 mammarytumors isolated from12-month-oldwild-typeorMMTV-c-Neu transgenicmice, respectively, were analyzed for Fn14, HER2, HER3, and tubulinexpression by Western blotting. B, normal mammary gland andmammary tumor samples isolated from age-matched wild-type orMMTV-PyMT transgenic mice, respectively, were analyzed for Fn14,HER2, and GAPDH expression by Western blotting.

Asrani et al.





expression in AU565 cells was more sensitive to wort-mannin treatment (Supplementary Fig. S5B and S5C).Treatment of AU565 cells with the Akt-specific inhibitorMK-2206 also reduced Fn14 expression levels (data notshown).

HRG1-b1 stimulation of MCF7 cells induces Fn14 geneexpressionHER2 signaling in breast cancer cells can be activated

by both HER2 overexpression and by ligand engagementof other EGFR family members; and in particular, byHRG1 binding to HER3 (5, 42). Therefore, we nextevaluated the effects of HRG1 and other EGF familyligands on Fn14 expression in MCF7 cells. Cells weretreated with either the HER1 ligand EGF, the HER1/HER4 ligands heparin-binding EGF-like growth factor(HB-EGF), and betacellulin (BTC), or the HER3/HER4ligands HRG1-a and HRG1-b1. All ligands increasedFn14 expression, but the maximal effect was seen withHRG1-b1 (Fig. 4A). The HRG1-b1 stimulatory effectwas both dose-dependent (Fig. 4B) and time-dependent(Fig. 4C). Also, HRG1-b1 treatment of MCF7 cellstransiently increased Fn14 mRNA levels (SupplementaryFig. S6). HRG1-b1–mediated Fn14 upregulation requiresHER3 binding and/or signaling as HER3 depletion inMCF7 cells attenuated this effect (Fig. 4D). Lapatinibpretreatment of MCF7 cells also inhibited the HRG1-b1stimulatory effect, indicating that HER2 signaling, most

likely via HER2–HER3 heterodimer formation, isrequired for Fn14 gene induction (Fig. 4E). HRG1-b1stimulation of AU565 cells also upregulated Fn14 expres-sion levels (data not shown).

Fn14 depletion decreases MCF7/HER2-18 cell basalmigration capacityWe compared the migratory potential of parental

MCF7 cells and MCF7/HER2-18 cells using scratchwound assays. The HER2-overexpressing cells had anapproximately 2.2-fold greater basal migration capacity(Fig. 5A and B). To determine the functional significanceof Fn14 expression in HER2-overexpressing or HRG1-b1–stimulated breast cancer cells, we established MCF7/HER2-18 clonal cell lines stably transduced with either acontrol shRNA or 2 shRNAs targeting different regions ofthe Fn14 transcript. Effective Fn14 knockdown wasconfirmed by Western blot analysis (Fig. 5C). We ana-lyzed the migratory potential of the 4 cell lines using thescratch wound assay. Fn14 depletion significantlydecreased the ability of MCF7/HER2-18 cells to closethe scratch wound (Fig. 5D).

MCF7/HER2-18 cells are poorly invasive in Matrigelassays using serum as the stimulusWe next tried to determine the effects of Fn14 depletion

on MCF7/HER2-18 cell basal invasive capacity using mod-ified Boyden chambers coated with basement membrane

Figure 2. Both transient and stableHER2 overexpression in MCF7 breastcancer cells increases Fn14 proteinexpression. A, MCF7 cells weretransiently transfected with theindicated amount of a HER2expression plasmid. Cells wereharvested at 24 hoursposttransfection and Fn14, HER2, andtubulin expression analyzed byWestern blotting. B and C, MCF7 andMCF7/HER2 cells were analyzed forHER2, HER3, p-HER2, p-HER3, Fn14,and tubulin levels byWestern blotting.D, MCF7 and MCF7/HER2 cells weresubjected to FACS analysis using ananti-Fn14 mAb (MCF7 cells: darkdashed line; MCF7/HER2 cells: darksolid line) or isotype control IgG (MCF7cells: light dashed line; MCF7/HER2cells: light dotted line). E, MCF7 andMCF7/HER2-18 cells were analyzedfor Fn14, HER2, and tubulinexpression by Western blotting.

C

Fn14

HER2

Tubulin

E

Fn14

HER2

Tubulin

100 101 102 103 104

Fluorescence intensity

0

20

40

60

80

100

% M

axim

um

D

HER2

Tubulin

p-HER2

p-HER3

B

HER3

A

Fn14

Tubulin

HER2

MCF7

MCF7

MCF7/

HER

2

MCF7

MCF7/

HER

2

MCF7

MCF7/

HER

2-18

0 μg 2 μg 4 μgHER2:






extract (Matrigel). The MCF7/HER2-18 control shRNAcell lines did not invade through Matrigel when either 10%or 20% serum-containing medium was used in the bottomchamber as the chemoattractant and the incubation periodwas as long as 72 hours (data not shown); consequently, wecould not evaluate if Fn14 depletion in these cells wouldreduce basal invasion capacity.

HRG1-b1–stimulated cell migration and invasion isattenuated in Fn14-depleted MCF7/HER2-18 cellsHRG1-b1 treatment of MCF7 cells increases Fn14

gene expression (shown earlier) and HRG1-b1 is a potentpromigratory and proinvasive factor for these cells (12).We found that HRG1-b1 treatment of MCF7/HER2-18

cells also stimulated migration and invasion, and thiseffect was approximately 2.9- and 4.9-fold greater, respec-tively, than the effect on parental MCF7 cells (Supple-mentary Fig. S7). Because HRG1-b1 treatment of controlshRNA, but not Fn14 shRNA MCF7/HER2-18 cells,increases Fn14 expression (Fig. 6A and data not shown),we examined whether Fn14 upregulation was required forHRG1-b1–driven migration and invasion using our 4 celllines. We found that Fn14 depletion attenuated bothHRG1-b1–stimulated cell migration (Fig. 6B) and inva-sion (Fig. 6C). We next investigated whether reintroduc-tion of Fn14 into Fn14-depleted MCF7/HER2-18cells could rescue the invasion deficit. Fn14 shRNA-448 cells were stably transfected with vector control or

Fn14

Tubulin

A

p-HER2

p-HER3

HER3

HER2

Fn14

Tubulin

p-HER3

p-HER2

C

HER3

HER2

B siRNA

Fn14

HER2

HER3

Tubulin

p-Erk

p-Akt

Erk

Akt

D siRNA

Fn14

HER2

HER3

Tubulin

p-Erk

p-Akt

Erk

Akt

Figure 3. Lapatinib treatment or HER2/HER3 siRNA cotransfection ofMCF7/HER2 and AU565 cells decreases Fn14 expression levels. A,MCF7/HER2 cells were serum-starved overnight (0.5% FBS) and theneither left untreated (NT, no treatment) or treated with dimethyl sulfoxide(DMSO) vehicle or lapatinib (100 nmol/L) for 8 hours. Cellswere harvestedand Fn14, p-HER2, HER2, p-HER3, HER3, and tubulin levels analyzed byWestern blotting. B, MCF7/HER2 cells were either left untreated (NT) ortransiently transfected with Allstars control, HER2, and/or HER3 siRNAduplexes. Cells were harvested 48 hours later and analyzed for Fn14,HER2, HER3, p-Erk, Erk, p-Akt, Akt, and tubulin levels by Westernblotting. C and D, AU565 cells were treated in the same manner as theMCF7/HER2 cells as described above in (A) and (B).

A

10 10010 100 10 100 10 100 10 1000

Fn14

Tubulin

0 1 3 10 30 100

HRG1-β1 (ng/mL)

Fn14

Tubulin

B C

Fn14

Tubulin

0 1 h 2 h 4 h 8 h 12 h 24 h

HRG1-β1

Fn14

Tubulin

– – +

p-HER2

p-HER3

HRG1-β1:

Lapatinib:

DMSO:

– + + +

– – –

–

+

E

Fn14

HER3

Tubulin

Luc HER3

HRG1-β1: – + – +

D

Figure 4. HRG1-b1 treatment of MCF7 cells induces Fn14 geneexpression. A,MCF7 cells were serum-starved overnight (0.5%FBS) andeither left untreated or treated with 10 or 100 ng/mL of the indicatedgrowth factors for 4 hours. Cells were harvested and Fn14 and tubulinexpression analyzed by Western blotting. B, serum-starved MCF7 cellswere either left untreated or treated with the indicated doses of HRG1-b1for 4 hours. Cells were harvested and Fn14 and tubulin expressionanalyzed by Western blotting. C, serum-starved MCF7 cells were eitherleft untreated or treated with HRG1-b1 (10 ng/mL) for the indicated timeperiods. Cells were harvested and Fn14 and tubulin expression analyzedby Western blotting. D, MCF7 cells were transiently transfected witheither luciferase control siRNA or HER3 siRNA. After 48 hours ofincubation, the cells were serum-starved overnight (0.5% FBS) and theneither left untreated or treatedwith HRG1-b1 (10 ng/mL) for 4 hours. Cellswere harvested and analyzed for Fn14, HER3, and tubulin expression byWestern blotting. E, serum-starved MCF7 cells were either left untreatedor treated with HRG1-b1 (10 ng/mL) for 4 hours. In some cases, the cellswere first pretreated for 30 minutes with DMSO vehicle or lapatinib (100nmol/L). Cells were harvested and Fn14, p-HER2, p-HER3, and tubulinexpression analyzed by Western blotting.

Asrani et al.





a plasmid encoding myc epitope-tagged human Fn14.This plasmid did not include the Fn14 30-untranslatedregion (UTR), thus ectopic Fn14 expression should not beinhibited by Fn14 shRNA coexpression, and this wasconfirmed by Western blot analysis (Fig. 6D). EctopicFn14 expression in the Fn14-depleted MCF7/HER2-18cells resulted in an approximately 1.8-fold increase inHRG1-b1–stimulated cell invasion (Fig. 6E).

HRG1-b1–stimulated MMP-9 expression is attenuatedin Fn14-depleted MCF7/HER2-18 cellsIt has been previously reported that HRG1-b1 treatment

of MCF7 cells stimulates MMP-9 expression and activity(17), and we found that HRG1-b1–stimulated MCF7/HER2-18 cell invasion is reduced by approximately 50%when cells are treated with SB-3CT, a specific inhibitor ofMMP-2/MMP-9 activity (ref. 43; Fig. 7A and B). There-fore, we determined whether Fn14 depletion had an effecton basal or HRG1-b1–stimulated MMP-9 gene expression.Real-time qRT-PCR analysis indicated that basal as well asHRG1-b1–stimulated MMP-9 mRNA expression was low-er in the Fn14 shRNA-448 cells compared with controlshRNA cells (Fig. 7C). Gelatin zymography assays were thenconducted using conditionedmedia obtained from the 2 celllines to examine MMP-9 protein levels. We found that theFn14 shRNA-448 cells had reduced basal as well as HRG1-b1–stimulated MMP-9 gelatinolytic activity (Fig. 7D).

Furthermore, forced expression of the Fn14 protein in theFn14 shRNA-448 cells was able to increase HRG1-b1–stimulated MMP-9 expression (Fig. 7E).

DiscussionOverexpression of the EGFR family member HER2

occurs in 15% to 20% of all breast cancer cases and someHER2þ tumors also express the HER2-preferred bindingpartner HER3 (4). HRG1, a ligand for HER3, is alsoexpressed in breast tumors (8). Both HER2 overexpression-and HRG1-mediated HER2/HER3 signaling likely contri-butes to breast cancer pathophysiology. In this study, we firstextended our earlier findings showing a correlation betweenHER2 and Fn14 expression in human breast tumors (26) byshowing their coexpression in mammary tumor specimensderived fromMMTV-c-Neu andMMTV-PyMT transgenicmice. We then showed that (i) HER2 overexpression,achieved via plasmid transfection of MCF7 breast cancercells or endogenous HER2 gene amplification in AU565cells, activates HER2–HER3 signaling and increases Fn14expression levels, (ii) HRG1-b1 treatment of MCF7cells, which also activates HER2–HER3 signaling, caninduce Fn14 gene expression, and (iii) HRG1-b1 treatmentof HER2-overexpressing MCF7/HER2-18 cells furtherupregulates Fn14 levels, and this "super-induction" isrequired for maximal cell migration, invasion, and MMP-9 expression.

Figure 5. Fn14 depletion reducesMCF7/HER2-18 cell basal migrationcapacity. A, confluent monolayers ofparental MCF7 and MCF7/HER2-18cells were wounded using a pipettetip. Wound closure was monitoredby microscopy and representativephotomicrographs (�20magnification) are shown. B, woundwidth was calculated at 0 and 24hours and the difference plotted aspercentage wound closure. Thevalues shown are mean � SEM of 3experiments. ��, P < 0.01 comparedwith MCF7 cells by Student t test. C,MCF7/HER2-18 clonal cell linesstably transduced with a controlshRNA lentivirus (1A and B)or 2 different Fn14 mRNA-targetedshRNA lentiviruses (448 and 562)were analyzed for Fn14 and tubulinexpression by Western blotting. D,confluent monolayers of the 4 celllines described earlier werewounded using a pipette tip. Woundwidth was calculated at 0 and 24hours and the difference plotted aspercentage wound closure. Thevalues shown are mean � SEM of 3experiments. ��, P < 0.01 comparedwith control shRNA-1A cells byStudent t test.

Control

shRNA

1A 1B 448

Fn14

Tubulin

Fn14

shRNA

562

C D

Wound c

losure

(%

)

**

0

10

20

30

40

50

60

70

80

90

100

1A 1B 448 562

**

Cntrl shRNA Fn14 shRNA

A

MCF7

HER2-18

0 h 24 h

Wound c

losure

(%

)

**

0

10

20

30

40

50

60

70

80

90B

MCF7 HER2-18






We identified a direct link between HER2 overexpressionand elevated Fn14 levels in both MCF7 and AU565 breastcancer cells. First, we found that both transient and stableexpression of HER2 in MCF7 cells increased Fn14 levels.Second, inhibition of HER2–HER3 signaling in MCF7/HER2 and AU565 cells using lapatinib significantly reducedFn14 expression. Third, HER2 or HER3 depletiondecreased Fn14 levels in AU565 cells, whereas combinedHER2/HER3depletionwas required to see a similar effect inMCF7/HER2 and MCF7/HER2-18 cells. This cell linedifference may be due to a more complete reduction of Akt

activation in the AU565 cells following HER2 or HER3depletion alone. This possibility is supported by our MEK,PI3K, and Akt inhibitor results, indicating that Fn14expression in AU565 cells is predominantly driven byPI3K/Akt signaling.HER2 overexpression can activate gene transcription via

several signaling pathways, including theMAPK (44), PI3K/Akt (44), and NF-kB (45) pathways, and the Fn14 genepromoter has potential transcription factor–binding sites forboth AP-1 and NF-kB (25). Therefore, we postulated thatthe increase in Fn14 expression in HER2-overexpressing

****

No

. o

f ce

lls in

va

de

d

0

20

40

60

80

100

120

140

160

1A 1B 448 562

C

0

10

20

30

40

50

60

1A 1B 448 562

****


Wound c

losure

(%

)

B

Myc

Tubulin

D EN

o. o

f ce

lls in

va

de

d **

0

50

100

150

200

250

Vector Fn14


Fn14

Tubulin

HRG1-β1: 0 1 h 2 h 4 h 8 h

Control-1A Fn14-448A

0 1 h 2 h 4 h 8 h

Figure6. Fn14depletion inMCF7/HER2-18cells reducesHRG1-b1–stimulated cellmigration and invasion. A, theMCF7/HER2-18control shRNA-1AandFn14shRNA-448 cell lines were serum-starved overnight (0.5% FBS) and then either left untreated or treated with HRG1-b1 (50 ng/mL) for the indicatedtime periods. Cells were harvested and Fn14 and tubulin expression analyzed by Western blotting. B, MCF7/HER2-18 control and Fn14 shRNA celllines were treated with HRG1-b1 (200 ng/mL) andmigration capacity was compared using the scratch wound assay.Woundwidth was calculated at 0 and 24hours and the difference plotted as percentage wound closure. The values shown are mean � SEM of 3 experiments. ��, P < 0.01 compared withcontrol shRNA-1A cells by Student t test. C,MCF7/HER2-18 control and Fn14 shRNA cell lineswere treatedwith HRG1-b1 (200 ng/mL) and invasive capacitywas compared using the Matrigel invasion assay. HRG1-b1–stimulated cell invasion was quantitated as described in the Materials and Methods.The values shown are mean � SEM of 3 experiments. ��, P < 0.01 compared with control shRNA-1B cells by Student t test. D, Fn14 shRNA-448 cells,stably transfected with vector or Fn14-myc plasmid, were analyzed for Fn14-myc and tubulin expression by Western blotting. E, vector or Fn14-transfectedFn14 shRNA-448 cells were treated with HRG1-b1 (200 ng/mL) and invasive capacity was quantitated as described in theMaterials andMethods. The valuesshown are mean � SEM of 3 experiments. ��, P < 0.01 by Student t test.

Asrani et al.





cells was likely due to an increase in Fn14 gene transcription;however, we did not observe a significant difference inFn14 mRNA levels between the MCF7 and MCF7/HER2cells. Thus, it appears that HER2 overexpression-mediatedFn14 upregulation is occurring via a posttranscriptionalmechanism.The HER2 receptor has no known ligand, but it can be

activated following ligand engagement of other EGFR familymembers via receptor heterodimerization (2, 3); therefore,we examined the effects of various EGF family members onFn14 expression in MCF7 cells. HRG1-a and HRG1-b1,ligands for HER3 and HER4, were the 2 most potentinducers of Fn14 expression in MCF7 cells. HRG1-b1

treatment of MCF7 cells transiently increased Fn14 mRNAlevels, suggesting that ligand engagement was inducing Fn14gene transcription. Also, we found that both HER3 siRNAand lapatinib treatment prevented the HRG1-b1–stimulat-ed increase in Fn14 expression observed in these cells,indicating that HRG1-b1–induced Fn14 expression wasmost likely due to HER2–HER3 heterodimerization andsignaling.We generated MCF7/HER2-18 cell lines with reduced

Fn14 levels using Fn14 shRNA lentiviral constructs toinvestigate the potential role of Fn14 in HER2/HER3–activated breast cancer cells. First, we compared the basalmigration capacity of the control and Fn14-depletedMCF7/

0

20

40

60

80

100

120

140

DMSO SB-3CT

*

B

No

. o

f ce

lls in

va

de

d

DMSO SB-3CT

48 h

AHRG1-β1

mR

NA

fold

expre

ssio

n(M

MP

-9/G

AP

DH

)

HRG1-β1:

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

0 h 12 h 0 h 12 h

Cntrl-1A Fn14-448

*

C

E

MMP-9 -

HRG-1β1:

MMP-2 -

Fn14-448

0 12 h 24 h 0 12 h 24 h

Vector Fn14

D

0 12 h 24 h 0 12 h 24 h

MMP-9 -

HRG-1β1:

Cntrl-1A Fn14-448

MMP-2 -

Figure 7. Fn14 depletion in MCF7/HER2-18 cells reduces HRG1-b1-stimulated MMP-9 expression. A, MCF7/HER2-18 control shRNA-1A cells were platedinto invasion chambers and treated with DMSO vehicle or MMP-2/MMP-9 Inhibitor IV (SB-3CT; 20 mmol/L) in the presence of HRG1-b1 (200 ng/mL). After 48hours of incubation, the invasive cells were stained and representative photomicrographs (�20 magnification) are shown. B, HRG1-b1–stimulatedcell invasion of vehicle- or drug-treated cellswas quantitated asdescribed in theMaterials andMethods. The values shown aremean�SEMof 3 experiments.�, P < 0.05 by Student t test. C, the MCF7/HER2-18 control shRNA-1 and Fn14 shRNA-448 cell lines were serum-starved overnight (0.5% FBS) andtheneither left untreatedor treatedwithHRG1-b1 (50ng/mL) for 12hours. Cellswere harvested,RNAwas isolated, andMMP-9andGAPDHmRNAexpressionwasquantitated using real-timeRT-PCR.MMP-9 expressionwas normalized toGAPDHexpression. The values shownare themean�SEMof 3 experiments.�,P < 0.05 byStudent t test. D, serum-starvedMCF7/HER2-18 control shRNA-1 and Fn14 shRNA-448 cells were either left untreated or treatedwithHRG1-b1(50 ng/mL) for the indicated time periods. Conditioned media was collected and concentrated and MMP-2/MMP-9 activity analyzed using gelatinzymography. E, Fn14 shRNA-448 cells stably transfected with vector or Fn14-myc plasmid were serum-starved overnight (0.5% FBS) and then either leftuntreated or treatedwithHRG1-b1 (50 ng/mL) for the indicated time periods. Conditionedmediawas collected and concentrated andMMP-2/MMP-9 activityanalyzed using gelatin zymography.






HER2-18 cell lines and found that Fn14 depletionreduced migratory capacity. This finding is consistentwith previous studies showing that Fn14 depletionreduces lung (27) and prostate (28) cancer cell migration.Second, we attempted to examine the effect of Fn14depletion on basal invasion capacity but found that thecontrol MCF7/HER2-18 cell lines could not invadethrough a Matrigel barrier. Previous studies have alsofound that parental (26, 46) and HER2-overexpressing(47) MCF7 cells are poorly invasive. In our earlier report,we showed that ectopic Fn14 expression in MCF7 cellsincreases invasive capacity (26). Our finding that HER2overexpression in MCF7 cells increases Fn14 levels butnot invasive capacity indicates that this degree of endog-enous Fn14 upregulation is not sufficient to increaseMCF7 cell invasiveness under our experimentalconditions.Since HRG1-b1 treatment, like HER2 overexpression,

can activate HER2/HER3 signaling and upregulate Fn14gene expression in parental MCF7 cells, we investigatedwhether HRG1-b1 treatment of the control shRNA-expressing MCF7/HER2-18 cells might increase endog-enous Fn14 levels over and beyond the level observed withHER2 overexpression alone. We found that this was thecase; in addition, we showed that HRG1-b1 treatment ofcontrol MCF7/HER2-18 cells, like parental MCF7 cells(12), increased migration and invasion. These findingsallowed us to test whether Fn14 upregulation was requiredfor HRG1-b–stimulated MCF7/HER2-18 cell migrationand invasion, and this turned out to be the case. Weconfirmed that Fn14 levels could regulate invasion byshowing that ectopic expression of Fn14 in Fn14-depletedMCF7/HER2-18 cells stimulated HRG1-b1–mediatedcell invasion.It has been reported thatHRG1-b1 treatment andHER2/

HER3 signaling in breast cancer cells can upregulate MMP-9 production (17) and we found that maximal HRG1-b1–stimulated MCF7/HER2-18 cell invasion requires MMP-9activity. Since TWEAK:Fn14 binding can upregulateMMP-9 expression in various cell types, including gliomacells (48) and prostate cancer cells (28), we investigatedwhether Fn14 levels could also modulate MMP-9 expres-sion. We found that in comparison with the control MCF7/HER2-18 cells, the Fn14-depleted cells had lower basal aswell as HRG1-b1–stimulated MMP-9 expression. Thisfinding is consistent with a recent report showing thattransient Fn14 depletion in PC-3 prostate cancer cells candecrease basal MMP-9 expression (28).In conclusion, we have shown that ectopic HER2

overexpression in MCF7 cells, endogenous HER2 over-expression in AU565 breast cancer cells, and HRG1-b1

stimulation of parental as well as HER2-overexpressingMCF7 cells increases Fn14 gene expression. These effectsare most likely mediated via HER2/HER3 activation andstimulation of the extracellular signal–regulated kinase(ERK) and PI3K/Akt signaling pathways. We also foundthat HRG1-b1 treatment of HER2-overexpressing cellspromotes cell migration, invasion, and MMP-9 expres-sion, and these effects are attenuated in Fn14-depletedcells. Our finding that HRG1-b1–mediated Fn14 upre-gulation in MCF7/HER2-18 cells is required for maximalMMP-9 production may explain, at least in part, whyHRG1-b1–stimulated invasion is impaired in the Fn14-depleted cell lines. Taken together, our results indicatethat Fn14 may play a role in HER2/HER3–driven breastcancer cell migration and invasion. Furthermore, sinceFn14 and HER2 are frequently coexpressed in humanbreast tumors (26, 39), agents targeting the Fn14 proteincould be particularly beneficial to those HER2þ patientswith intrinsic or acquired resistance to the current ther-apeutic agents trastuzumab (49) and lapatinib (50).

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

Authors' ContributionsConception and design: K. Asrani, S.A.N. Brown, J.A. WinklesDevelopment of methodology: K. Asrani, S.A.N. BrownAcquisition of data (provided animals, acquired and managed patients, providedfacilities, etc.): K. Asrani, R.A. Keri, R. Galisteo, S.A.N. Brown, S.J. Morgan, A.Ghosh, N.L. TranAnalysis and interpretation of data (e.g., statistical analysis, biostatistics, compu-tational analysis): K. Asrani, R.A. Keri, S.J. Morgan, N.L. Tran, J.A. WinklesWriting, review, and/or revision of the manuscript: K. Asrani, S.J. Morgan, J.A.WinklesAdministrative, technical, or material support (i.e., reporting or organizing data,constructing databases): K. Asrani, S.A.N. Brown

AcknowledgmentsThe authors thank Dr. Mark Williams (University of Maryland School of

Medicine) for assistance with the FACS experiments. We also thank Dr. Mien-ChieHung (University of Texas MD Anderson Cancer Center) for the HER2 expressionplasmid, Dr. Dihua Yu (University of Texas MD Anderson Cancer Center) for theMCF7/HER2 cell line, Dr. Anne Hamburger (University of Maryland School ofMedicine) for the MCF7/HER2-18 cell line, Dr. Peter Choyke (NIH) for theNIH3T3/HER2 cell line, and Dr. Angela Brodie (University of Maryland Schoolof Medicine) for the MCF7-Ca and LTLT-Ca cell lines.

Grant SupportThis work was supported in part by Susan G. Komen for the Cure Research Grant

KG081095 (to J.A. Winkles), NIH grants R01 CA130940 (to N.L. Tran), and R01CA090398 (to R.A. Keri).

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 September 17, 2012; revised January 2, 2013; accepted January 22, 2013;published OnlineFirst February 1, 2013.

References1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA Cancer

J Clin 2012;62:10–29.2. Moasser MM. The oncogene HER2: its signaling and transforming

functions and its role in human cancer pathogenesis. Oncogene2007;26:6469–87.

3. Campbell MR, Amin D, Moasser MM. HER3 comes of age: newinsights into its functions and role in signaling, tumor biology, andcancer therapy. Clin Cancer Res 2010;16:1373–83.

4. Naidu R, Yadav M, Nair S, Kutty MK. Expression of c-erbB3 protein inprimary breast carcinomas. Br J Cancer 1998;78:1385–90.

Asrani et al.





5. Citri A, Skaria KB, Yarden Y. The deaf and the dumb: the biology ofErbB-2 and ErbB-3. Exp Cell Res 2003;284:54–65.

6. Lee-Hoeflich ST, Crocker L, Yao E, Pham T, Munroe X, HoeflichKP, et al. A central role for HER3 in HER2-amplified breastcancer: implications for targeted therapy. Cancer Res 2008;68:5878–87.

7. Sergina NV, Rausch M, Wang D, Blair J, Hann B, Shokat KM, et al.Escape from HER-family tyrosine kinase inhibitor therapy by thekinase-inactive HER3. Nature 2007;445:437–41.

8. Montero JC, Rodríguez-Barrueco R, Oca~na A, Díaz-Rodríguez E,Esparís-Ogando A, Pandiella A. Neuregulins and cancer. Clin CancerRes 2008;14:3237–41.

9. Agus DB, Akita RW, Fox WD, Lewis GD, Higgins B, Pisacane PI, et al.Targeting ligand-activated ErbB2 signaling inhibits breast and pros-tate tumor growth. Cancer Cell 2002;2:127–37.

10. Xue C, Liang F, Mahmood R, Vuolo M,Wyckoff J, Qian H, et al. ErbB3-dependent motility and intravasation in breast cancer metastasis.Cancer Res 2006;66:1418–26.

11. Hernandez L, Smirnova T, Kedrin D,Wyckoff J, Zhu L, Stanley ER, et al.The EGF/CSF-1 paracrine invasion loop can be triggered by heregulinbeta1 and CXCL12. Cancer Res 2009;69:3221–7.

12. Kushima Y, Iida K, Nagaoka Y, Kawaratani Y, Shirahama T, SakaguchiM, et al. Inhibitory effect of (�)-epigallocatechin and (�)-epigalloca-techin gallate against heregulin beta1-induced migration/invasion ofthe MCF-7 breast carcinoma cell line. Biol Pharm Bull 2009;32:899–904.

13. Atlas E, Cardillo M, Mehmi I, Zahedkargaran H, Tang C, Lupu R.Heregulin is sufficient for the promotion of tumorigenicity andmetastasis of breast cancer cells in vivo. Mol Cancer Res 2003;1:165–75.

14. Friedl P, Alexander S. Cancer invasion and the microenvironment:plasticity and reciprocity. Cell 2011;147:992–1009.

15. Adam L, Vadlamudi R, Kondapaka SB, Chernoff J, Mendelsohn J,Kumar R. Heregulin regulates cytoskeletal reorganization and cellmigration through the p21-activated kinase-1 via phosphatidylinosi-tol-3 kinase. J Biol Chem 1998;273:28238–46.

16. Benlimame N, He Q, Jie S, Xiao D, Xu YJ, Loignon M, et al. FAKsignaling is critical for ErbB-2/ErbB-3 receptor cooperation for onco-genic transformation and invasion. J Cell Biol 2005;171:505–16.

17. Yao J, Xiong S, Klos K, Nguyen N, Grijalva R, Li P, et al. Multiplesignaling pathways involved in activation of matrix metalloproteinase-9 (MMP-9) by heregulin-beta1 in humanbreast cancer cells. Oncogene2001;20:8066–74.

18. KessenbrockK,PlaksV,WerbZ.Matrixmetalloproteinases: regulatorsof the tumor microenvironment. Cell 2010;141:52–67.

19. Bosc DG, Goueli BS, Janknecht R. HER2/Neu-mediated activation ofthe ETS transcription factor ER81 and its target gene MMP-1. Onco-gene 2001;20:6215–24.

20. Yuan G, Qian L, Song L, Shi M, Li D, Yu M, et al. Heregulin-betapromotes matrix metalloproteinase-7 expression via HER2-mediatedAP-1 activation in MCF-7 cells. Mol Cell Biochem 2008;318:73–9.

21. Winkles JA. The TWEAK-Fn14 cytokine-receptor axis: discovery,biology and therapeutic targeting. Nat Rev Drug Discov 2008;7:411–25.

22. Burkly LC, Michaelson JS, Zheng TS. TWEAK/Fn14 pathway: animmunological switch for shaping tissue responses. Immunol Rev2011;244:99–114.

23. Kwon OH, Park SJ, Kang TW, Kim M, Kim JH, Noh SM, et al. Elevatedfibroblast growth factor-inducible 14 expression promotes gastriccancergrowth vianuclear factor-kBand is associatedwithpoor patientoutcome. Cancer Lett 2012;314:73–81.

24. Tran NL, McDonough WS, Savitch BA, Sawyer TF, Winkles JA,Berens ME. The tumor necrosis factor-like weak inducer of apo-ptosis (TWEAK)-fibroblast growth factor-inducible 14 (Fn14) signal-ing system regulates glioma cell survival via NF-kappaB pathwayactivation and BCL-XL/BCL-W expression. J Biol Chem 2005;280:3483–92.

25. Tran NL, McDonoughWS, Savitch BA, Fortin SP, Winkles JA, SymonsM, et al. Increased fibroblast growth factor-inducible 14 expressionlevels promote glioma cell invasion via Rac1 and nuclear factor-

kappaB and correlate with poor patient outcome. Cancer Res2006;66:9535–42.

26. Willis AL, Tran NL, Chatigny JM, Charlton N, Vu H, Brown SAN, et al.The fibroblast growth factor-inducible 14 receptor is highly expressedin HER2-positive breast tumors and regulates breast cancer cellinvasive capacity. Mol Cancer Res 2008;6:725–34.

27. Whitsett TG, Cheng E, Inge L, Asrani K, Nathan M, Hostetter G, et al.Elevated expression of Fn14 in non–small cell lung cancer correlateswith activated EGFR and promotes tumor cell migration and invasion.Am J Pathol 2012;181:111–20.

28. Huang M, Narita S, Tsuchiya N, Ma Z, Numakura K, Obara T, et al.Overexpression of Fn14 promotes androgen-independent prostatecancer progression through MMP-9 and correlates with poor treat-ment outcome. Carcinogenesis 2011;32:1589–96.

29. Ho DH, Vu H, Brown SAN, Donohue PJ, Hanscom HN, Winkles JA.Soluble tumor necrosis factor-like weak inducer of apoptosisoverexpression in HEK293 cells promotes tumor growth andangiogenesis in athymic nude mice. Cancer Res 2004;64:8968–72.

30. Guy CT, Webster MA, Schaller M, Parsons TJ, Cardiff RD, Muller WJ.Expression of the neu protooncogene in the mammary epithelium oftransgenic mice induces metastatic disease. Proc Natl Acad Sci U S A1992;89:10578–82.

31. Landis MD, Seachrist DD, Monta~nez-Wiscovich ME, Danielpour D,Keri RA. Gene expression profiling of cancer progression revealsintrinsic regulation of transforming growth factor-beta signaling inErbB2/Neu-induced tumors from transgenic mice. Oncogene 2005;24:5173–90.

32. Guy CT, Cardiff RD, Muller WJ. Induction of mammary tumors byexpression of polyomavirus middle T oncogene: a transgenic mousemodel for metastatic disease. Mol Cell Biol 1992;12:954–61.

33. Lin EY, Jones JG, Li P, ZhuL,WhitneyKD,MullerWJ, et al. Progressionto malignancy in the polyoma middle T oncoprotein mouse breastcancer model provides a reliable model for human diseases. Am JPathol 2003;163:2113–26.

34. JelovacD, SabnisG, LongBJ,Macedo L,GoloubevaOG,Brodie AMH.Activation of mitogen-activated protein kinase in xenografts and cellsduring prolonged treatment with aromatase inhibitor letrozole. CancerRes 2005;65:5380–9.

35. Brown SAN, Ghosh A, Winkles JA. Full-length, membrane-anchored TWEAK can function as a juxtacrine signaling moleculeand activate the NF-kappaB pathway. J Biol Chem 2010;285:17432–41.

36. Yepes M, Brown SAN, Moore EG, Smith EP, Lawrence DA, WinklesJA. A soluble Fn14-Fc decoy receptor reduces infarct volume ina murine model of cerebral ischemia. Am J Pathol 2005;166:511–20.

37. Brown SAN, Richards CM, Hanscom HN, Feng SL, Winkles JA. TheFn14 cytoplasmic tail binds tumour-necrosis-factor-receptor-associ-ated factors 1, 2, 3 and 5 and mediates nuclear factor-kB activation.Biochem J 2003;371:395–403.

38. Michaelson JS, Cho S, Browning B, Zheng TS, Lincecum JM, WangMZ, et al. TWEAK inducesmammary epithelial branchingmorphogen-esis. Oncogene 2005;24:2613–24.

39. ChaoDT, SuM, Tanlimco S, ShoM, Choi D, FoxM, et al. Expression ofTweakR in breast cancer and preclinical activity of enavatuzumab, ahumanized anti-TweakR mAb. J Cancer Res Clin Oncol 2013;139:315–25.

40. Siegel PM, Ryan ED, Cardiff RD, Muller WJ. Elevated expression ofactivated formsofNeu/ErbB-2andErbB-3are involved in the inductionof mammary tumors in transgenic mice: implications for human breastcancer. EMBO J 1999;18:2149–64.

41. Moy B, Kirkpatrick P, Kar S, Goss P. Lapatinib. Nat Rev Drug Discov2007;6:431–2.

42. Wallasch C, Weiss FU, Niederfellner G, Jallal B, Issing W, Ullrich A.Heregulin-dependent regulation of HER2/neu oncogenic signaling byheterodimerization with HER3. EMBO J 1995;14:4267–75.

43. Lin CW, Shen SC, Hou WC, Yang LY, Chen YC. Heme oxygenase-1inhibits breast cancer invasion via suppressing the expression ofmatrix metalloproteinase-9. Mol Cancer Ther 2008;7:1195–206.






44. Niu G, Carter WB. Human epidermal growth factor receptor 2regulates angiopoietin-2 expression in breast cancer via AKT andmitogen-activated protein kinase pathways. Cancer Res 2007;67:1487–93.

45. Merkhofer EC, Cogswell P, Baldwin AS. Her2 activates NF-kappaBand induces invasion through the canonical pathway involving IKKal-pha. Oncogene 2010;29:1238–48.

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

47. Mitra D, Brumlik MJ, Okamgba SU, Zhu Y, Duplessis TT, Parvani JG,et al. An oncogenic isoform of HER2 associated with locally dissem-inated breast cancer and trastuzumab resistance. Mol Cancer Ther2009;8:2152–62.

48. Winkles JA, Tran NL, Berens ME. TWEAK and Fn14: new moleculartargets for cancer therapy? Cancer Lett 2006;235:11–7.

49. Pohlmann PR, Mayer IA, Mernaugh R. Resistance to trastuzumab inbreast cancer. Clin Cancer Res 2009;15:7479–91.

50. Chen FL, Xia W, Spector NL. Acquired resistance to small moleculeErbB2 tyrosine kinase inhibitors. Clin Cancer Res 2008;14:6730–4.

Asrani et al.





2013;11:393-404. Published OnlineFirst February 1, 2013.Mol Cancer Res Kaushal Asrani, Ruth A. Keri, Rebeca Galisteo, et al. Invasion, and MMP9 ExpressionMember Fn14 Promotes HRG-Driven Breast Cancer Cell Migration,

Inducible TNFR Superfamily−1 (HRG)βThe HER2- and Heregulin

Updated version

10.1158/1541-7786.MCR-12-0542doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2013/02/01/1541-7786.MCR-12-0542.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/11/4/393.full#ref-list-1

This article cites 50 articles, 22 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/11/4/393.full#related-urls

This article has been cited by 5 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/11/4/393To 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-12-0542

http://mcr.aacrjournals.org/content/suppl/2013/02/01/1541-7786.MCR-12-0542.DC1

http://mcr.aacrjournals.org/content/11/4/393.full#ref-list-1

http://mcr.aacrjournals.org/content/11/4/393.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/11/4/393


the her2- and heregulin b1 (hrg) inducible tnfr...

Documents