Glucocorticoid Down-regulation of Fascin Protein Expression IsRequired for the Steroid-induced Formation of Tight Junctions andCell-Cell Interactions in Rat Mammary Epithelial Tumor Cells*

(Received for publication, May 29, 1998, and in revised form, November 16, 1998)

Vivian Wong‡, Dixie Ching‡, Pierre D. McCrea§, and Gary L. Firestone‡¶

From the ‡Department of Molecular and Cell Biology and the Cancer Research Laboratory, University of California,Berkeley, California 94720-3200 and the §Department of Biochemistry and Molecular Biology, University of Texas M.D.Anderson Cancer Center, Houston, Texas 77030-4095

Glucocorticoid hormones, which are physiologicalregulators of mammary epithelium development, inducethe formation of tight junctions in rat Con8 mammaryepithelial tumor cells. We have discovered that, as partof this process, the synthetic glucocorticoid dexametha-sone strongly and reversibly down-regulated the expres-sion of fascin, an actin-bundling protein that also inter-acts with the adherens junction component b-catenin.Ectopic constitutive expression of full-length mouse fas-cin containing a Myc epitope tag (Myc-fascin) in Con8cells inhibited the dexamethasone stimulation of tran-sepithelial electrical resistance, disrupted the inducedlocalization of the tight junction protein occludin andthe adherens junction protein b-catenin to the cell pe-riphery, and prevented the rearrangement of the actincytoskeleton. Ectopic expression of either the carboxyl-terminal 213 amino acids of fascin, which includes theactin and b-catenin-binding sites, or the amino-terminal313 amino acids of fascin failed to disrupt the glucocor-ticoid induction of tight junction formation. Mammarytumor cells expressing the full-length Myc-fascin re-mained generally glucocorticoid responsive and dis-played no changes in the levels or protein-protein inter-actions of junctional proteins or the amount ofcytoskeletal associated actin filaments. However, a cellaggregation assay demonstrated that the expression ofMyc-fascin abrogated the dexamethasone induction ofcell-cell adhesion. Our results implicate the down-regu-lation of fascin as a key intermediate step that directlylinks glucocorticoid receptor signaling to the coordi-nate control of junctional complex formation and cell-cell interactions in mammary tumor epithelial cells.

The apical junctional complex of epithelial cells is a special-ized set of cell-cell contacts consisting of the tight junction andthe adherens junction (1). The more apically located tight junc-tion serves as a selective barrier to the paracellular diffusion ofsolutes and thereby separates the apical and basolateral fluidcompartments of the epithelial cell monolayer (2–4). By limit-ing the lateral movement of plasma membrane lipids and pro-teins, the tight junction also contributes to the maintenance ofcellular polarity, physically defines the border between the

apical and basolateral plasma membrane surfaces (2–4), and isessential for the polarized functions of differentiated epithelialcells such as secretion, absorption, and responsiveness to ex-tracellular signals. At the molecular level, the tight junction isformed by the interaction between extracellular domains of thetransmembrane protein occludin from neighboring cells (5–8)and an intracellular protein complex that includes occludin andthe cytoplasmic proteins ZO-1 (9), ZO-2 (10, 11), cingulin (12,13), 7H6 antigen (12–15), symplekin (16), and rab13 (17). Theadherens junction, which lies immediately below the tight junc-tion toward the basal side of the cell (1), is formed by thecalcium-dependent, homophilic binding of E-cadherin extracel-lular domains from adjacent cells (18, 19). The adherens junc-tion is responsible for intercellular adhesive function (20), is arequirement for the generation of epithelial cell polarity bytight junctions (21–24), and plays a key role in regulatingcell-cell interactions involved in tissue morphogenesis and dif-ferentiation (25). The E-cadherin-mediated adherens junctionis formed by the specific association of a number of cytoplasmicproteins with the cytoplasmic tail of E-cadherin (26, 27): theseinclude a-catenin (28, 29), b-catenin (30), plakoglobin (31–33),and p120CTN (34–36). Other cytoplasmic molecular compo-nents of the adherens junction include actin, a-actinin (37),vinculin (38), radixin (39), tenuin (40), Src (41), and fascin (42).

The tight junction and the adherens junction structures arehighly dynamic in that their function, assembly, and/or disas-sembly can be controlled by a variety of intracellular signalsthat ultimately influence cell-cell interactions in a physiologi-cally appropriate manner (43–48). For example, the permeabil-ity properties of the tight junction and/or adhesive properties ofthe adherens junction can be altered by intracellular calcium,protein kinase C activity, and certain lipid-mediated cell sig-naling cascades (43, 49). The Rac and Rho members of the Rassuperfamily of small GTPases appear to provide another com-mon regulatory connection between tight junctions and adhe-rens junctions because these signaling molecules can controlintercellular adhesion, permeability, and apical junction as-sembly (50–53). The actin cytoskeleton, which forms the char-acteristic perijunctional ring underlying the adherens junction(54) and has been shown to associate with the tight junction(55), represents a structural link between the tight junctionand the adherens junction. One potential molecular connectionbetween the actin cytoskeleton and the apical junction may bethrough the fascin actin-bundling protein (56, 57) which alsobinds b-catenin to form a biochemically distinct non-cadherincomplex (42), although a functional role for fascin in thehormonal control of apical junction dynamics has not beencharacterized.

Mammary-derived cells represent a biologically useful sys-tem to explore hormone-activated signaling pathways that po-

* This work was supported by National Institutes of Health GrantDK-42799 (to G. L. F.). The costs of publication of this article weredefrayed in part by the payment of page charges. This article musttherefore be hereby marked “advertisement” in accordance with 18U.S.C. Section 1734 solely to indicate this fact.

¶ To whom correspondence and reprint requests should be addressed:Dept. of Molecular and Cell Biology, 591 LSA, University of Californiaat Berkeley, Berkeley, CA 94720-3200. Tel.: 510-642-8319; Fax: 510-643-6791; E-mail: glfire@uclink4.berkeley.edu.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 274, No. 9, Issue of February 26, pp. 5443–5453, 1999© 1999 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org 5443

by guest on October 9, 2020

http://ww

w.jbc.org/

Dow

nloaded from

tentially influence and mediate cell-cell interactions. We haveestablished that glucocorticoids can induce tight junction for-mation in a receptor-dependent manner in both nontumori-genic and tumorigenic rodent mammary epithelial cells (58–62), which implicates this lactogenic steroid (63, 64) as a keyphysiological signal for the increased junctional complex for-mation and mammary intercellular contacts that occur duringthe onset of lactation (65). In the rat Con8 mammary tumorepithelial cell line, which was derived from a 7,12-dimethyl-benz(a)anthracene-induced rat mammary adenocarcinoma (66,67), glucocorticoids induce the remodeling of the apical junctionand a polarized phenotype that results in the localization of theZO-1 tight junction protein from the cytoplasm to the cellperiphery at the site of cell-cell contacts. This regulated re-sponse induces the barrier function of the tight junction (60,61), which decreases paracellular permeability and stimulatesthe transepithelial electrical resistance (TER).1 Given theknown transcriptional mechanism of action of glucocorticoidreceptors (68–73), we propose that the regulated expression ofa specific set of glucocorticoid responsive gene products act asmolecular switches and/or structural components that func-tionally control the formation of intercellular junctions to in-duce a epithelial-like phenotype. Although, in general, rela-tively little is known about the regulated expression and/oractivity of apical junction-associated structural or regulatoryproteins by extracellular stimuli. In this study, we demonstratethat the glucocorticoid down-regulation of fascin protein ex-pression is a critical event for this steroid to induce the orga-nization of the apical junctional complex which provides, forthe first time, a direct functional link between a steroid regu-lated gene and the control of cell-cell interactions.

EXPERIMENTAL PROCEDURES

Cell Culture, Glucocorticoid Treatment, and Measurement of Trans-epithelial Electrical Resistance—Con8 rat mammary epithelial cells(61) were grown on standard tissue culture plates or Anopore mem-brane filters (0.2 mm, Nalgene Nunc International, Naperville, IL) inDulbecco’s modified Eagle’s medium/F-12 supplemented with 10% calfserum and penicillin/streptomycin (BioWhittaker, Walkersville, MD),and maintained at 37 °C and 5% CO2. Treatment of Con8 cells withglucocorticoids was done by the addition of the glucocorticoid agonist,dexamethasone (Sigma), at a final concentration of 1026 M (prepared asa 1023 M stock in ethanol) to normal growth media unless otherwiseindicated. The formation of tight junctions was monitored by measuringTER on filter grown cells using an EVOM Epithelial Voltohmmeter(World Precision Instruments, Sarasota, FL) as described previously(60, 61). The EVOM provides an alternating square wave current of620 mA through the monolayer. The TER was calculated from themeasured resistance and normalized by the area of the monolayer. Thebackground TER of blank Transwell filters was subtracted from theTER of cell monolayers.

Immunofluorescence Staining and Western Blotting—For immuno-fluorescence, Con8 cells were grown on Anopore membranes and theTER was measured before and after dexamethasone treatment. Toprepare cells for immunofluorescence, cells were fixed with 100% meth-anol at 220 °C for 30 min, and subsequently dried with 100% acetone at220 °C for 5 min. Filters were incubated with a blocking buffer (1%non-fat dry milk in 0.5% Triton X-100, 5 mM EDTA, 0.15 M NaCl, and 20mM HEPES, pH 7) before incubation with the primary antibodies.Rabbit anti-occludin antibodies (8) and anti-b-catenin (gifts of Barry M.Gumbiner, Sloan Kettering), rabbit anti-fascin antibodies (42), rabbitanti-actin antibodies (Sigma), and mouse anti-c-Myc epitope antibodies(Zymed Laboratories Inc., San Francisco, CA) were all used at a 1:1000dilution. Secondary fluorescein isothiocyanate-conjugated anti-rabbitand Texas Red-conjugated anti-mouse antibodies were purchased fromMolecular Probes, Inc. (Eugene, OR) and used at a 1:100 dilution.

Stained cells were mounted with SlowFadeTM Light Antifade reagent(Molecular Probes, Inc.) and stored at 4 °C before visualization.

For Western blot analyses, cells were rinsed twice in phosphate-buffered saline (BioWhittaker), and extracted directly in SDS-PAGE(polyacrylamide gel electrophoresis) sample buffer (50 mM Tris-HPO4,pH 6.8, 2.5 mM EDTA, 15% sucrose, 2% SDS, and 50 mM dithiothreitol)containing protease inhibitors (5 mM phenylmethylsulfonyl fluoride, 5mg/ml pepstatin A, 1 mg/ml TLCK, 10 mg/ml leupeptin, 20 mg/ml apro-tinin, 50 mg/ml antipain, 2 mM benzamidine, 50 mg/ml soybean trypsininhibitor, and 2.5 mM iodoactamide). Samples were boiled for 15 minand cooled to room temperature before the addition of 1 M iodoaceticacid to a final concentration of 125 mM. SDS-PAGE was performedusing the Mini-protein II cell electrophoresis unit (Bio-Rad) accordingto the manufacturer’s guidelines. After electrophoresis, proteins wereelectrophoretically transferred to nitrocellulose membranes (MicronSeparations Inc., Westboro, MA) using the Mini Trans-Blot electro-phoretic transfer cell (Bio-Rad) according to manufacturer’s guidelines.The blots were incubated with blotting buffer (5% non-fat dry milk, 0.15M NaCl, 0.2% Triton X-100) before probing with a 1:1000 dilution ofprimary antibodies (anti-occludin, anti-b-catenin, anti-E-cadherin, an-ti-a-catenin, anti-fascin, or anti-c-Myc epitope). The mouse monoclonalanti-fascin antibody (gift of DAKO Corp., Carpinteria, CA) was ob-tained as a crude supernatant of hybridoma media and was not dilutedprior to use. Horseradish peroxidase-conjugated anti-rabbit and anti-mouse antibodies (Bio-Rad) were used as secondary probes, and theblots were developed by an ECL kit (Amersham Life Sciences, Inc.).

Immunoprecipitation of Junctional Complex Proteins—For the exam-ination of co-immunoprecipitated proteins, each 100-mm2 plate of con-fluent Con8 cells was extracted in 5 ml of 1% Tween 20 buffer (1%Tween 20, 0.15 M NaCl, 5 mM EDTA, 20 mM HEPES, pH 7.4) in thepresence of protease inhibitors (5 mM phenylmethylsulfonyl fluoride, 5mg/ml pepstatin A, 1 mg/ml TLCK, 10 mg/ml leupeptin, 20 mg/ml apro-tinin, 50 mg/ml antipain, 2 mM benzamidine, 50 mg/ml soybean trypsininhibitor, and 2.5 mM iodoactamide). Extraction was carried out on icefor 30 min and all samples were precleared by centrifugation at12,000 3 g for 15 min before immunoprecipitation. Co-immunoprecipi-tation was performed with Protein A-Sepharose CL-4B (PharmaciaBiotech Inc.) in the presence of rabbit anti-b-catenin or preimmuneserum at 4 °C for 4 h. Immunoprecipitates were washed five times inthe same immunoprecipitation buffer, followed by a final wash withphosphate-buffered saline. Co-immunoprecipitated proteins were solu-bilized immediately in SDS-PAGE sample buffer and prepared for SDS-PAGE as described above for Western blotting.

Assay for Cytoskeletal-associated Actin Filaments—The Triton X-100extraction method (74) was employed to determine level of F-actin andthe amount of cytoskeleton-associated fascin and b-catenin. The mam-mary tumor cells were grown to confluency and treated with or without1 mM dexamethasone for 96 h. The cells were then trypsinized into asingle-cell suspension, counted using a hemacytometer, and then 2million cells collected into a pellet by centrifugation. After aspiratingthe media, each cell pellet was detergent extracted by suspension in 1ml of lysis buffer (0.5% Triton X-100, 20 mM PIPES, pH 6.8, 2 mM

MgCl2, 50 mM KCl, 5 mM EGTA, 5 mM dithiothreitol, 1 mM ATP, and 1mg/ml each of leupeptin, pepstatin, and aprotinin) and then immedi-ately centrifuged at 8,700 3 g in a microcentrifuge for 3 min. Theresulting pellets (containing the Triton X-100-resistant proteins) weresuspended in 100 ml of 20 mM phosphate buffer (14.8 mM NaH2PO4 and5.2 mM K2HPO4, pH 7.0, and 1 mg/ml each of leupeptin, pepstatin, andaprotinin). A parallel set of cells was centrifuged and then directlysuspended in 100 ml of 20 mM phosphate buffer and represents the totalcell extracts. For the Western blots, the Triton X-100-resistant proteinsand total cell extracts were dissolved in an equal volume of 2 3 SDS-PAGE sample buffer (described above) and the proteins electrophoreti-cally fractionated by SDS-PAGE.

Stable Con8 Cell Lines Expressing Full-length, NH2-terminal Portionand COOH-terminal Portion Myc-Fascin—The full-length mouse fascintagged with a 6X c-Myc epitope at its NH2 terminus was cloned into themammalian expression vector pcDNA3 (Invitrogen), which drives theinserted cDNA by the CMV promoter and contains the neomycin resist-ance gene. Mammalian expression vectors for the NH2-terminal frag-ment of mouse fascin containing amino acids 1–313 and the COOH-terminal fragment of mouse fascin containing amino acids 281 to 493each tagged with a 6X c-Myc epitope at the NH2 terminus were clonedinto the pCS21MT mammalian expression vector. The fascin se-quences in this expression vector are driven by the simian CMV IE94promoter and the vector backbone of pCS2 is pBluescript KS1 (75,76). Plasmid DNA was expanded in XL-1 Escherichia coli cells andpurified with a plasmid purification kit (Qiagen, Inc., Chatsworth, CA).

1 The abbreviations used are: TER, transepithelial electrical resist-ance; PAGE, polyacrylamide gel electrophoresis; SGK, serum and glu-cocorticoid inducible protein kinase; TLCK, Na-p-tosyl-L-lysine chlo-romethyl ketone; PIPES, 1,4-piperazinediethanesulfonic acid; CMV,cytomegalovirus.

Steroid Control of Cell-Cell Interactions5444

by guest on October 9, 2020

http://ww

w.jbc.org/

Dow

nloaded from

The pcDNA3-fascin, pCS21MT NH2-terminal fascin, and the pCS21MTCOOH-terminal fascin DNA were introduced into Con8 cells at 24 hafter plating of cells at 30% confluency. One set of cells was alsotransfected with the pCS21MT vector alone. In transfections with thepCS21MT based vectors, the cells were co-transfected with the neomy-cin resistance gene driven by the CMV promoter. Transfection of cellswas performed using LipofectAMINE reagent according to the manu-facturer’s protocol (Life Technologies, Inc., Gaithersburg, MD). Poten-tial expressors were selected at 600 mg/ml G418 (Life Technologies,Inc.), which was cytotoxic to .99% of parental Con8 cells. Single cloneswere isolated, expanded, and Myc-fascin expression was assessed byantigen blotting with either the anti-fascin or anti-Myc epitope anti-bodies. Multiple fascin expressing clonal cell lines derived from theparental Con8 cells were analyzed in parallel to control for clonalvariation. The mammary tumor cells expressing high levels of full-length Myc-fascin are f/3.1, f/3.8, and f/3.11, of the NH2-terminal por-tion of Myc-fascin are FN12 and FN36, and of the COOH-terminalportion of Myc-fascin are FC10 and FC12. The f/3.4, f/3.3, and f/3.2 cellsdo not express Myc-fascin although the original transfected containedthe full-length Myc-fascin gene. The C1 cells were transfected with thepCS21MT expression vector without any fascin sequences.

Cell Adhesion Assay—Parental Con8 and mouse fascin-expressingand control transfected Con8 cell lines that were treated with or with-out dexamethasone (1026 M for 4 days) were evaluated using an aggre-gation assay. To obtain cells for aggregation assays, each 100-mm2 plateof confluent cells was rinsed twice in phosphate-buffered saline, andsubsequently incubated with 0.25% trypsin (Sigma) in Dulbecco’s mod-ified Eagle’s medium/F-12 for 1 h at 37 °C in a 5% CO2 atmosphere.After an hour of trypsinization, all cells were detached from the cultureplate and subsequently collected by centrifugation at 10 3 g for 15 minin 45 ml of fresh Dulbecco’s modified Eagle’s medium/F-12. Cells wereresuspended gently by pipetting in 5 ml of fresh growth media (Dulbec-co’s modified Eagle’s medium/F-12 containing 10% calf serum). Thepresence of 10% calf serum in the cell suspension inhibits the activity ofresidual trypsin. Cells were either examined immediately after resus-pension or allowed to rest on the bottom of a 50-ml conical tube for 30min with 5-min intervals of gentle swirling. Cell aggregates were ex-amined by light microscopy using a 3 10 objective lens and a dark fieldcondenser setting.

RESULTS

Time Course and Dose Response of the Glucocorticoid Down-regulation of Fascin Protein in Rat Mammary Epithelial TumorCells—We have previously established that glucocorticoidsstimulate tight junction formation in Con8 mammary epithe-lial tumor cells (61). Given the well characterized effects ofglucocorticoids on gene expression (68–73), it seemed likelythat this steroid induces the formation of tight junctions byaltering the expression of a select subset of proteins that areassociated with and that may regulate the overall junctional

complex. Furthermore, the formation of intercellular junctionsis intimately linked to the cytoskeleton, and we have observedthat dexamethasone causes a rearrangement of Con8 mam-mary tumor cells from round-shaped cells that grow on top ofeach other to cuboidal shapes characteristic of an epithelialmonolayer (60, 61). Thus, potential targets of glucocorticoidsignaling may include proteins that interact with both the tightjunction and the adherens junction as well as the cytoskeleton.Recently, fascin, an actin-bundling protein (56, 57), was foundto interact with the adherens junction protein b-catenin (42),and therefore provided a candidate regulatory protein involvedin the glucocorticoid control of junctional complex formation. Toexplore this possibility, the level of fascin protein produced wascompared with the kinetics of tight junction formation in Con8mammary tumor cells treated with the synthetic glucocorticoiddexamethasone over a 4-day time course. At the indicatedtimes in dexamethasone, the generation of TER in cell mono-layers grown on Anopore membrane filters was used to monitortight junction formation, and a parallel set of Western blotswas probed for fascin protein or for the tight junction proteinoccludin. As shown in Fig. 1a, dexamethasone induced a sig-nificant increase in the TER that continued to rise throughoutthe time course. In contrast, the TER of cells not treated withglucocorticoids remained essentially unchanged. Western blotsshowed that total cellular fascin protein began to decrease atapproximately 48 h and was markedly reduced (.80%) by 72 hof dexamethasone treatment (Fig. 1a). During this time periodin which fascin levels are reduced, occludin protein levels re-mained unchanged. The overall time frame of induced TER andfascin protein down-regulation is consistent with a functionalrelationship between these processes.

As a complementary approach to confirm the correlationbetween the glucocorticoid down-regulation of fascin with theformation of tight junctions, the production of fascin and theinduction of TER was analyzed in mammary tumor cellstreated with various doses of steroid for 4 days. As shown inFig. 1b, half-maximal down-regulation of fascin protein and thehalf-maximal stimulation of TER was observed at approxi-mately 1028 M dexamethasone, which correlates with the Kd ofglucocorticoid-receptor occupancy for these cells (67). The max-imal effect of dexamethasone on both fascin down-regulationand the stimulation of TER was observed at 1027 M (Fig. 1b),with higher concentrations of dexamethasone having no addi-tional effects (data not shown). A similar dose-dependent re-

FIG. 1. Time course and dose re-sponse of the glucocorticoid down-regulation of fascin protein. a, conflu-ent monolayers of Con8 cells grown onAnopore membrane filter supports weretreated with 1 mM dexamethasone (Dex),and at the indicated days in culture, theformation of tight junctions was moni-tored in triplicate sets of cell cultures bymeasurements of the TER. Western blotsof a parallel set of cell cultures wereprobed for occludin and fascin protein ex-pression. b, Con8 cells grown on Anoporemembrane filter supports were treatedwith the indicated concentration of dexa-methasone for 4 days and the TER wasdetermined in triplicate sets of cell cul-tures. Western blots of electrophoreticallyfractionated total cell extracts from a par-allel set of cell cultures treated 4 dayswith 1029 M, 1028 M, or 1027 M dexa-methasone were probed for fascin proteinexpression.

Steroid Control of Cell-Cell Interactions 5445

by guest on October 9, 2020

http://ww

w.jbc.org/

Dow

nloaded from

sponse on fascin down-regulation and stimulation of TER wasobserved in mammary tumor cells treated for shorter durationswith dexamethasone. Thus, the down-regulation of the 45-kDafascin expressed in mammary tumor cells correlated with theformation of tight junctions in a dose- and time-dependentmanner that is consistent with a glucocorticoid receptor-de-pendent process.

Glucocorticoid Withdrawal Reverses the Rearrangement ofthe Apical Junction and the Down-regulation of Fascin Pro-tein—To test whether glucocorticoids are required for themaintenance of the organized apical junction and for the con-tinued down-regulation of fascin, we examined the effects ofdexamethasone withdrawal on the overall rearrangement ofjunctional proteins and on fascin protein production. Con8mammary tumor cells were treated with dexamethasone for 6days, at which time the TER is strongly induced. Subsequently,the culture medium was replaced with fresh medium withoutadded glucocorticoids for 4 days and the TER returned to basallevels. The subcellular localization of the tight junction protein,occludin, and the adherens junction protein, b-catenin, wereexamined by indirect immunofluorescence microscopy. Dexa-methasone withdrawal effectively reversed the glucocorticoid-induced rearrangement of both occludin and b-catenin (Fig. 2a,panels 6 versus 1). Both occludin and b-catenin redistributedfrom the junctional complex at the cell periphery back to anunorganized pattern indistinguishable from the untreatedcells upon withdrawal of dexamethasone (Fig. 2a, panels 6versus 2).

A parallel set of Con8 mammary tumor cells were examinedfor the protein expression of fascin, a-catenin, b-catenin, andcingulin. As shown in Fig. 2b, dexamethasone withdrawal (low-er panel, lane 6) reversed the dexamethasone-mediated down-regulation of fascin protein under conditions in which the ex-pression of tight junction and adherens junction proteinsremained constant. Fascin expression was almost restored topretreatment levels by glucocorticoid withdrawal. It is worthmentioning that the lane for 6 was slightly under loaded whichis noticeable on the b-catenin blot. Therefore, it appears thatthe glucocorticoid regulation of junctions assembly and disas-sembly correlated closely with the expression levels of fascinprotein. In addition, the reversibility of the glucocorticoid-in-duced junctions formation implicates this steroid as a mainte-nance factor rather than a terminal differentiation signal.

Ectopic Expression of Full-length Fascin Prevents the Glu-cocorticoid Induction of Transepithelial Electrical Resistance inMammary Tumor Cell Monolayers—Conceivably, the down-regulation of fascin is a critical functional event in the glucocor-ticoid signaling pathway in that fascin may inhibit key cellularevents that lead to the formation of tight junctions. To directly

investigate this possibility, we tested whether the ectopic con-stitutive expression of fascin would prevent the dexamethasoneinduction of tight junction formation. Con8 mammary tumorcells were stably transfected with cDNA of full-length mousefascin containing a Myc epitope tag at the NH2 terminus (77).This fascin mammalian expression vector contains the neomy-cin resistance gene (Fig. 3a) and stable clonal cell lines thatexpress different levels of Myc-fascin were obtained by selec-tion using a lethal dosage of G418. Based on the open readingframe, the Myc-fascin cDNA is expected to encode an approxi-mate 65-kDa protein, taking into account the 10-kDa increasein molecular weight due to the Myc epitope. However, duringour initial screening for expressing clones, we consistently ob-served that the Con8 mammary tumor cells expressed Myc-fascin as an 80-kDa protein species. Fig. 3a shows a Westernblot characterizing the ectopically expressed Myc-fascin fromone of the transfected cell lines (f/3.1). Cell extracts from f/3.1cells were electrophoretically fractionated and Western blotswere probed with c-Myc monoclonal antibodies, fascin poly-clonal antibodies, or fascin monoclonal antibodies. As a control,a Western blot of a cell extract isolated from non-transfectedCon8 cells was also incubated with c-Myc monoclonal antibod-ies. As shown in Fig. 3a, the c-Myc antibody recognized an80-kDa protein that is not present in the non-transfected Con8cells. Similarly migrating forms of Myc-fascin were detected inthe f/3.1 cells using the polyclonal or monoclonal fascin anti-bodies (Fig. 3a). Endogenous fascin is not apparent in the blotsbecause of the short time of exposure to the chemiluminescencereagents before development of the x-ray film. Taken together,these results verify that the 80-kDa protein observed in trans-fected Con8 cells is the ectopically expressed, Myc-taggedmouse fascin. In this regard, transfection of this same Myc-fascin expression vector, or a fascin expression vector with aCOOH-terminal Myc epitope, also produced an 80-kDa proteinspecies in a human tumor cell line.2

Several individual single cell-derived subclones of trans-fected Con8 cells were screened for expression of Myc-fascinand for the ability of glucocorticoids to induce tight junctionformation as assessed by the development of TER. In the threeclonal transfected cell lines that express high levels of Myc-fascin (f/3.1, f/3.8, and f/3.11), dexamethasone failed to stimu-late the monolayer TER under conditions in which the level ofectopically expressed fascin remained constant (Fig. 3b, upperpanel versus middle panel). In contrast, in three recoveredtransfected cell lines that express little or no Myc-fascin (f/3.4,f/3.3, and f/3.2), the TER was glucocorticoid inducible indicat-

2 P. McCrea, unpublished observation.

FIG. 2. Reversibility of glucocorti-coid-induced translocation of occlu-din and b-catenin and fascin down-regulation by dexamethasone with-drawal. a, Con8 cells were treated with-out (2) or with (1) 1 mM dexamethasoneor were first treated with dexamethasoneand then the steroid removed by replacingthe culture medium (6). The localizationof occludin or b-catenin was examined byindirect immunofluorescence microscopy.The scale bar represents 50 mm. b, West-ern blots of Con8 cells treated without (2)or with (1) 1 mM dexamethasone or ofcells first treated with dexamethasoneand then the steroid removed (6) wereprobed for expression of a-catenin, b-cate-nin, cingulin, or fascin.

Steroid Control of Cell-Cell Interactions5446

by guest on October 9, 2020

http://ww

w.jbc.org/

Dow

nloaded from

ing the proper formation of tight junctions. Overall, the level ofMyc-fascin expression correlated with the degree of inhibitionon glucocorticoid-induced TER. To confirm that the transfectedcell clones remained generally glucocorticoid responsive, thedexamethasone stimulated expression of the endogenous se-rum-glucocorticoid inducible protein kinase (SGK) was exam-ined in the recovered transfected cell lines. We have previouslyestablished that SGK is under direct glucocorticoid receptortranscriptional control due to a glucocorticoid response elementin its promoter (78, 79). As shown in Fig. 3b (lower panel), ineach of the recovered clonal cell lines, SGK expression wasinduced by glucocorticoids. Therefore, the inability of glucocor-ticoids to induce tight junction formation in the high Myc-fascin expressors was not due to a general defect in glucocor-ticoid receptor signaling.

For the remainder of the experiments, the f/3.1 clone wasutilized because this cell line produced the highest constitutivelevel of Myc-fascin. The f/3.2 clone expressed an insignificantlevel of Myc-fascin and therefore was employed as one of thetransfection control cells for comparison to the f/3.1 clone. Asdiscussed below, the C1 cell clone represents the other trans-fection control cell line used in certain experiments. As shownin Fig. 4, ectopic expression of Myc-fascin in the f/3.1 cloneinhibited the dexamethasone-stimulated TER over a 4-daytime course, whereas the TER was rapidly stimulated in f/3.2cells in a manner similar to that of parental Con8 mammarytumor cells (see Fig. 1a). Indirect immunofluorescence micros-copy using Myc antibodies showed the localization of Myc-fascin throughout the cells in both 4-day dexamethasone-treated and untreated f/3.1 cells (Fig. 4). In contrast, Myc-fascin was not detected by immunofluorescence staining ineither untreated (Fig. 4) or dexamethasone-treated (data notshown) f/3.2 cells. Taken together, these results demonstratedthat ectopic expression of the full-length fascin gene inhibitedthe glucocorticoid-stimulated formation of tight junctions andfurther implicates the down-regulation of fascin as an essentialintermediate step in the glucocorticoid signaling pathway thatcontrols the junctional complex in mammary tumor cells.

Ectopic Expression of Amino-terminal or Carboxyl-terminalFragments of Fascin Fail to Prevent the Glucocorticoid Induc-tion of Transepithelial Electrical Resistance—Previous studieshave shown that the COOH-terminal 216 amino acids of fascin(residues 277 to 493) can bind in vitro to actin (56, 57, 77),whereas, analysis by the yeast two-hybrid assay revealed thatthe fascin coding region corresponding to amino acids 324–453functionally interact with b-catenin (42). Although the precisestructural boundaries involved in each interaction have notbeen defined, the fascin molecule can be approximately dis-sected into an NH2-terminal fragment that does not bind eitheractin or b-catenin and a COOH-terminal fragment that bindsto both actin and b-catenin. Therefore, Con8 mammary tumorcells were transfected with expression vectors that encode ei-ther the amino-terminal 313 amino acids (FN construct) or thecarboxyl-terminal 213 amino acids (FC construct) of the fascincoding region that contains both the actin and the b-catenin-binding sites. These constructs overlap by 32 amino acids be-tween residues 281 and 313 and each contain six copies of theMyc epitope tag at the amino terminus (Fig. 5a). The Con8 cellswere stably transfected with each construct as well as with anempty pSC21MT expression vector and individual cell coloniesisolated from each population of G418-resistant cells. Two sin-gle cell-derived subclones of cells transfected with either theNH2-terminal fascin fragment (FN12 and FN36 cells) or theCOOH-terminal fascin fragment (FC10 and FC12 cells) as wellas one of the vector-transfected cell lines (C1 cells) were exam-ined for expression of the Myc-tagged fascin domains by West-ern blots using the Myc antibodies and for the glucocorticoidstimulation of TER. As shown in Fig. 5a, the transfected FC10and FC12 cell lines produced a 34-kDa Myc-tagged COOH-terminal fascin protein and the transfected FN12 and FN36cell lines produced a 45-kDa Myc-tagged NH2-terminal fascinprotein, compared with the 80-kDa full-length Myc-fascin ex-pressed in the f/3.1 cells. The ectopically expressed COOH-terminal and NH2-terminal fascin proteins correspond to thepredicted sizes based on the fascin amino acid sequence ineach construct and the 11-kDa size of the Myc epitope tag. As

FIG. 3. Ectopic expression of a Myc epitope-tagged form of full-length mouse fascin in Con8 cells prevents the glucocorticoidinduction of transepithelial electrical resistance. a, the Myc-fascin expression vector encodes the full-length mouse fascin with anamino-terminal Myc epitope tag linked to the pcDNA3.1 expression vector. Western blots of electrophoretically fractionated cell extracts fromnontransfected Con8 cells or a Myc-fascin transfected Con8 subcloned denoted f/3.1 were probed with anti-Myc primary antibodies (myc). ParallelWestern blots of the f/3.1 cell extracts were also probed with polyclonal (poly) or monoclonal (mono) antibodies to mouse fascin. b, individualsubclones of transfected Con8 mammary tumor cells were grown on Anopore membrane filter supports and treated with or without 1 mM

dexamethasone for 4 days. The f/3.1, f/3.8, and f/3.11 subclones produce varying levels of constitutively expressed Myc-fascin, whereas, the f/3.4,f/3.3, and f/3.2 subclones do not produce Myc-fascin and represent transfection controls. The monolayer TER was monitored in each cell culture andWestern blots of parallel sets of cells were probed using Myc epitope or SGK-specific primary antibodies.

Steroid Control of Cell-Cell Interactions 5447

by guest on October 9, 2020

http://ww

w.jbc.org/

Dow

nloaded from

expected, the vector-transfected control cells, C1, do not ex-press any Myc epitope-tagged fascin proteins.

The ability of glucocorticoids to stimulate the TER was mon-itored in each of these transfected mammary tumor cells after96 h treatment with or without 1 mM dexamethasone. As shown

in Fig. 5b, in contrast to inhibitory effects of full-length Myc-fascin (Fig. 3), the ectopic expression of either the COOH-terminal or NH2-terminal fragments of fascin, in two differenttransfected cell lines, failed to disrupt the dexamethasone stim-ulation of TER (Fig. 5b). The overall range of the glucocorticoid-

FIG. 4. Ectopic expression of Myc-fascin inhibits the dexamethasone in-duction of tight junction formation.Top panel, the f/3.1 and f/3.2 subclones oftransfected Con8 mammary tumor cellswere grown on Anopore membrane filtersupports, treated with or without 1 mM

dexamethasone (Dex), and the monolayerTER was measured at the indicated daysin culture. Lower panel, the localization ofMyc epitope-fascin protein was examinedin f/3.1 and f/3.2 cells treated with orwithout 1 mM dexamethasone for 4 daysby indirect immunofluorescence micros-copy using Myc epitope antibodies. Thescale bar represents 25 mm.

FIG. 5. Stable ectopic expression of amino-terminal and carboxyl-terminal Myc-fascin fragments and effects on the transepithelialelectrical resistance. a, the FN expression construct encodes the NH2-terminal 313 amino acids of fascin with a 6X Myc-epitote tag on the aminoterminus and driven by the CMV promoter. The FC expression construct encodes the COOH-terminal 213 amino acids of fascin with a 6X Mycepitope tag on the amino terminus and driven by the CMV promoter. Con8 mammary tumor cells were co-transfected with a CMV driven neomycinresistance gene together with the FN construct, FC construct, or a pCS21MT empty vector and individual single cell-derived colonies werecollected. FC10 and FC12 cells express COOH-terminal Myc-fascin, FN12, and FN36 express NH2-terminal Myc-fascin, f/3.1 cells producefull-length Myc-fascin (see Fig. 3), and the C1 cells are a vector-transfected control cell line. Western blots of electrophoretically fractionated cellextracts from each of these transfected cells were probed with anti-Myc primary antibodies. b, individual subclones of transfected Con8 cells weregrown on Anopore membrane filter supports and treated with or without 1 mM dexamethasone (DEX) for 4 days and the monolayer TER wasmonitored in each cell culture as described under “Experimental Procedures.”

Steroid Control of Cell-Cell Interactions5448

by guest on October 9, 2020

http://ww

w.jbc.org/

Dow

nloaded from

induced TERs in the transfected cells approximated that ob-served in the vector control C1 cells (Fig. 5b) and other stabletransfected control cell lines (Fig. 3). Thus, the ability of con-stitutively expressed fascin to inhibit the glucocorticoid regu-lation of tight junction function does not simply require theportion of the protein containing both the actin and b-catenin-binding sites.

Ectopic Fascin Expression Prevents the Glucocorticoid-in-duced Rearrangement of b-Catenin and Occludin to the CellPeriphery and Disrupts the Regulated Reorganization of theActin Cytoskeleton—To determine whether the ectopic expres-sion of fascin disrupts the formation of both tight junctions andadherens junctions, the glucocorticoid-regulated rearrange-ment of the adherens junction protein b-catenin and the tightjunction protein occludin was examined in the parental andtransfected mammary tumor cell clones. Myc-fascin expressingf/3.1 cells, control f/3.2 cells, and untransfected Con8 cells weretreated with or without dexamethasone for a 72-h time courseand the localization of the apical junction proteins was ana-lyzed by indirect immunofluorescence microscopy. Glucocorti-coids induce b-catenin to redistribute from a disorganizedstaining pattern mostly away from the cell periphery to adistinct “cobblestone” junctional-like pattern in parental mam-mary tumor cells (Fig. 6, left column) and in the transfectioncontrol f/3.2 cell clone (Fig. 6, right column). In contrast, con-stitutive expression of Myc-fascin in the f/3.1 cell clone pre-vented this reorganization of b-catenin to the cell periphery(Fig. 6, middle column). Similar to the disruptive effects onadherens junction proteins, ectopic expression of Myc-fascinalso prevented the localization of tight junction proteins. Asshown in Fig. 7 (panels a and a9), the dexamethasone stimu-lated rearrangement of occludin to the cell periphery observed

in f/3.2 cells was inhibited in f/3.1 cells which produce highlevels of Myc-fascin (Fig. 7, panels b and b9). Taken together,these results demonstrated that ectopic expression of fascininterfered with the ability of glucocorticoids to induce a globalrearrangement of the tight and adherens junction proteins, andhence the formation of the junctional complex.

The effect of ectopic fascin expression on the organization ofthe actin cytoskeleton was examined by indirect immunofluo-rescence staining for actin. As shown in Fig. 7 (panels c versusd), constitutive expression of Myc-fascin did not alter the actincytoskeleton staining pattern in cells that were not treatedwith glucocorticoids. However, in f/3.1 cells (Fig. 7, panel d9),the expression of Myc-fascin prevented the dexamethasone-induced rearrangement of the actin cytoskeleton that was ob-served in both parental and f/3.2 cells (Fig. 7, panels c9 and e9).A similar result was obtained with phalloidin staining whichconfirms that the observed differences are due to polymerizedactin (data not shown). Thus, ectopic expression of fascin didnot change the basal actin cytoskeleton organization, butrather it disrupted the glucocorticoid-induced signal that con-trols the membrane rearrangement that leads to junctionalcomplex formation.

Because fascin binds and bundles actin filaments as well asbinds to b-catenin in a non-E-cadherin complex (56, 57, 77), thelevel of cytoskeletal-associated F-actin, fascin, and b-cateninwere biochemically examined in the Myc-fascin transfectedf/3.1 cells compared with the C1 vector-transfected controlcells. Cells were treated with or without dexamethasone for 4days and level of total Myc-fascin, actin, or b-catenin in thewhole cell extracts were compared with the level of each proteindetected in a Triton X-100-resistant cell pellet, which repre-sents the cytoskeletal associated actin filaments (74). As shown

FIG. 6. Ectopic expression of fascin prevents the glucocorti-coid-induced rearrangement of b-catenin to the junctional com-plex. Transfected f/3.1 high Myc-fascin expressing cells (b-b999) or f/3.2transfection control cells (c-c999) as well as the parental Con8 mammarytumor cells (a-a999) were treated with 1 mM dexamethasone for theindicated times and the localization of b-catenin was examined byindirect immunofluorescence microscopy. The scale bar represents 50mm.

FIG. 7. Ectopic expression of fascin prevents the glucocorti-coid-induced rearrangement of occludin and actin to the junc-tional complex. Upper panels, transfected f/3.1 high Myc-fascin ex-pressing (b, b9) and f/3.2 transfection control (a, a9) mammary tumorcells were treated with or without 1 mM dexamethasone for 3 days. Thelocalization of occludin was examined by indirect immunofluorescencemicroscopy. Lower panels, transfected f/3.1 high Myc-fascin expressing(d, d9) or f/3.2 transfection control (e, e9) mammary tumor cells as wellas the parental Con8 cells (c, c9) were treated with or without 1 mM

dexamethasone for 3 days and actin localization examined by indirectimmunofluorescence microscopy. The scale bar represents 25 mm.

Steroid Control of Cell-Cell Interactions 5449

by guest on October 9, 2020

http://ww

w.jbc.org/

Dow

nloaded from

in Fig. 8, most of the actin expressed in the presence or absenceof glucocorticoids can be detected in the Triton X-100-resistantcell pellet and the ectopic expression of Myc-fascin did not alterthe actin content of this fraction. Also, the same amount ofb-catenin was detected in the Triton X-100-resistant fractionregardless of the level of Myc-fascin or the steroid treatment(Fig. 8). Approximately 50% of the expressed Myc-fascin frac-tionated with the detergent-resistant cell pellets which likelyresults from its binding to actin. These results suggest that theconstitutive expression of fascin did not disrupt the apicaljunctions or the glucocorticoids regulated organization of theactin cytoskeleton by an alteration in the cytoskeletal associ-ated actin filaments or the interaction of b-catenin with thecytoskeleton.

The Levels and Complex Formation of Junctional ProteinsAre Not Altered by the Ectopic Expression of Fascin—The ex-pression levels and complex formation of junctional proteinswere examined in the transfected mammary tumor cells. Cellsthat express high (f/3.1) or insignificant (f/3.2) levels of ectopicMyc-fascin were treated with dexamethasone over a 7-day timecourse and the total cellular proteins were analyzed by West-ern blotting. As shown in Fig. 9a (top panel, anti-Myc blot),Myc-fascin levels in f/3.1 cells remained constant at a high levelthroughout the 7-day dexamethasone treatment, while f/3.2cells expressed virtually no Myc-fascin. During the same 7-daytime course, the expression levels of occludin, E-cadherin,

b-catenin, and actin remained unchanged in both f/3.1 and f/3.2cell clones. Under these conditions, the cells maintained theirglucocorticoid responsiveness as indicated by the strong induc-tion of SGK protein expression (Fig. 9a, bottom panel). Theformation of b-catenin protein complexes with E-cadherin,a-catenin, and fascin protein was examined in the transfectedcell clones by Western blot analysis of coimmunoprecipitatedproteins. Cell extracts of dexamethasone-treated and untreatedcells were first immunoprecipitated with b-catenin antibodies in1% Tween 20 to preserve the protein-protein interactions ofb-catenin. Western blot analysis of the electrophoretically frac-tionated immunoprecipitates revealed that ectopic expressionof Myc-fascin did not alter the ability of b-catenin to complexwith E-cadherin or a-catenin (Fig. 9b, upper three panels).Similar amounts of E-cadherin and a-catenin were observed toco-immunoprecipitate with b-catenin in both dexamethasone-treated and untreated f/3.1 and f/3.2 cells. Furthermore, anti-Myc blots showed that the Myc-fascin protein co-immunopre-cipitated with b-catenin from f/3.1 cells, but not from f/3.2 cells(Fig. 9b, bottom panel). These results demonstrate that theconstitutive expression of Myc-fascin does not alter the expres-sion levels or complex formation of the adherens junctionproteins.

The Glucocorticoid Induction of Intercellular Adhesion IsInhibited by the Ectopic Expression of Fascin—To test whetherthe ectopic expression of Myc-fascin alters intercellular adhe-sion through the adherens junction, a cell aggregation assaywas performed on parental and transfected mammary tumorcells. Dexamethasone-treated (1 mM for 4 days) or untreatedcells were collected by trypsinization in the presence of cal-cium, which preserves E-cadherin function, and subjected tosheer force by gently pipetting through a round-tip pipetteuntil only small cell aggregates remained. As shown in Fig. 10,in the absence of dexamethasone, each of the three tested celllines formed small aggregates of 50 cells or less. The cellaggregates were loosely attached to each other, the cell borderswere rounded and intercellular space was prominently seen asbright lines under the dark field microscope setting. In dexa-methasone-treated cultures, parental cells and the f/3.2 trans-fection control cells formed large compact aggregates (200–1000 cells) in which the cell borders were not readily observedeven at the periphery of the cell mass. There were no observ-able differences in cell aggregate size and characteristics be-tween the parental and f/3.2 cells (Fig. 10, panels a9 and c9). Inaddition, when the parental and f/3.2 cells were allowed tofurther aggregate for 15 min after the initial cell adhesionassay, cell aggregates consisting of thousands of cells wereformed (data not shown). In contrast, glucocorticoids failed to

FIG. 8. Characterization of the cytoskeletal-associated actinfilaments and association of b-catenin with the cytoskeleton inMyc-fascin expressing and control transfected cells. The f/3.1cells expressing Myc-fascin and the vector control C1 mammary tumorcells were grown to confluency and treated with or without 1 mM dexa-methasone (DEX) for 96 h. After trypsinization into single-cell suspen-sions, 2 million cells per condition were collected by centrifugation,extracted with a Triton X-100 containing buffer, and the detergent-resistant cell pellet (Triton X-100 resistant) fractionated by SDS-poly-acrylamide gel electrophoresis. A second 2 million cell pellet collectedfrom each condition was electrophoretically fractionated without deter-gent extraction (total cell extracts). The Western blots were probed forMyc-fascin (using Myc antibodies), actin, and b-catenin using the ap-propriate primary antibodies.

FIG. 9. Ectopic expression of fascindoes not alter the expression levelsand complex formation of junctionalproteins. a, transfected f/3.1 Myc-fascinexpressing and f/3.2 transfection controlmammary tumor cells were treated with 1mM dexamethasone for 0, 3, 5, and 7 days.Parallel sets of Western blots of total cellextracts were probed for the production ofthe Myc epitope, occludin, E-cadherin,b-catenin, actin, and SGK. b, the trans-fected f/3.1 and f/3.2 mammary tumorcells were treated with or without 1 mM

dexamethasone for 4 days and b-cateninimmunoprecipitates isolated from the cor-responding cell extracts. The Westernblots of electrophoretically fractionatedimmunoprecipitated protein were probedfor E-cadherin, a-catenin, b-catenin, andthe Myc epitope.

Steroid Control of Cell-Cell Interactions5450

by guest on October 9, 2020

http://ww

w.jbc.org/

Dow

nloaded from

stimulate intercellular adhesion in f/3.1 cells, which constitu-tively produce Myc-fascin (Fig. 10, panel b9). Dexamethasone-treated f/3.1 cells remained as small, loosely attached aggre-gates that were indistinguishable from cell aggregrates of f/3.1cells not treated with steroid and thereby demonstrates thatthe ectopic expression of Myc-fascin disrupted the glucocorti-coid activation of cell-cell adhesion.

DISCUSSION

The regulation of cell-cell interactions and junctional com-plex dynamics is essential for epithelia to reversibly adapt toproliferative signals and to respond to physiological controlsduring tissue growth and differentiation. In this study, we havediscovered that glucocorticoids, a key hormonal regulator ofmammary gland differentiation, down-regulated the produc-tion of fascin protein, an actin-bundling protein that interactswith b-catenin, concomitant with the induction of tight junc-tions formation and cell-cell adhesion in Con8 rat mammaryepithelial tumor cells. The constitutive expression of fascindisrupted the ability of glucocorticoids to trigger the formationof the apical junctional complex and induce intercellular adhe-sion. These results implicate the down-regulation of fascin as akey intermediate step in the glucocorticoid signaling pathwaythat controls mammary cell-cell interactions and provides, forthe first time, a direct functional link between a steroid-regu-lated gene and the control of the apical junctions.

We propose that the selective regulation of fascin productionin mammary tumor cells is necessary for the coordinate regu-lation of tight and adherens junctions formation by glucocorti-coids. It is tempting to speculate that fascin acts as an inhibitorof junctional complex formation, and thus the glucocorticoiddown-regulation of fascin eliminates an inhibitory signal. Theglucocorticoid-mediated decrease in total cellular fascin proteinlevels temporally and dose-dependently correlated with thegeneration of functional tight junctions as indicated by thedevelopment of transepithelial electrical resistance, as well aswith the localization of tight junction protein occludin and theadherens junction protein b-catenin to the cell periphery. Un-der these conditions, glucocorticoids did not significantly alterthe total cellular protein levels of tight junction and adherensjunction proteins such as occludin, cingulin, ZO-1 (61), b-cate-nin, and E-cadherin. Thus, glucocorticoids appear to induce theassembly of both the tight and adherens junctions from pre-existing junctional components. The role of fascin down-regu-lation in the glucocorticoid-regulated formation of epithelialjunctions was directly tested by the ectopic expression of full-

length mouse fascin in Con8 rat mammary tumor cells. Char-acterization of stable clonal cell lines that produce variouslevels of ectopically expressed full-length Myc-fascin showed adirect correlation between the expression level and the degreeof inhibition of glucocorticoid-induced transepithelial electricalresistance. These results show that the inhibition of junctionalcomplex formation is specific to fascin expression and is not dueto random clonal variation. The constitutive production of ex-ogenous fascin was sufficient to prevent the glucocorticoid-induced development of transepithelial electrical resistanceand the reorganization of occludin and b-catenin to the tightand adherens junctions, respectively. In transfected cells thatproduce a high level of ectopic fascin, expression of junctionalproteins and formation of the E-cadherinza-cateninzb-catenincomplex was unaltered. Therefore, the ectopic expression offascin appears to specifically disrupt the overall assembly proc-ess of the junctional complex that is usually triggered by glu-cocorticoids and suggests that fascin acts as a negative regula-tor of the glucocorticoid-mediated formation of epithelialjunctions.

The glucocorticoid induction of junctions assembly is accom-panied by an activation of cell-cell adhesion that resulted in theformation of compact cell aggregates. In contrast, the Myc-fascin expressing cells remained loosely packed in the presenceof dexamethasone. Thus, the constitutive expression of fascinprotein not only blocks the glucocorticoid-induced formation ofjunctional complexes, but also interferes with the glucocorti-coid activation of cell-cell adhesion. Fascin might regulate in-tercellular adhesion directly by modulating the function ofE-cadherin or indirectly via its effect on actin which has beenimplicated to contribute to the adhesive strength of E-cadherin(80, 81). Fascin, which has been shown to bundle actin fila-ments (56, 82, 83), could potentially influence cell-cell adhesionand the assembly of intercellular junctions by modulating theactin cytoskeleton, a key player in the generation of epithelialcellular architecture and polarity (24). However, this possibil-ity seems unlikely because the ectopic expression of fascin didnot disrupt the cytoskeletal-associated actin filaments or theinteraction of b-catenin with the cytoskeleton. Although theimmediate mechanism of action of fascin is not clear, the dis-covery of a role for fascin in the regulation of cell-cell adhesionand intercellular junctions formation provides both structuraland functional links between the actin cytoskeleton and thecontrol of intercellular junctions.

Different molecular weight forms of fascin have been re-

FIG. 10. Ectopic expression of fascin prevents the glucocorticoid activation of cell-cell adhesion. Transfected f/3.1 high Myc-fascinexpressing (b, b9) or f/3.2 transfection control (c, c9) mammary tumor cells as well as the parental Con8 cells (a, a9) were treated with or without1 mM dexamethasone for 4 days and intercellular adhesion assessed as described in the text. Representative views of the cell adhesion assay areshown and the scale bar represents 100 mm.

Steroid Control of Cell-Cell Interactions 5451

by guest on October 9, 2020

http://ww

w.jbc.org/

Dow

nloaded from

ported in a variety of rodent and human cells, although to date,most of the characterized cell types express fascin with molec-ular masses in the range of 54–58 kDa (56, 82, 84). We haveobserved that the rat Con8 mammary tumor epithelial cell lineexpressed a fascin-like protein at an apparent molecular massof 45 kDa, which we have shown to be antigenically related tomouse fascin by competition binding to anti-mouse fascin an-tibodies. The polyclonal anti-fascin antibody used in our exper-iments was raised against a thioredoxin fusion protein thatcontains full-length mouse fascin (42, 77) and the primaryantibody binding to the 45-kDa fascin protein produced in themammary tumor cells was completely competed off by theaddition of 100 mg of the fusion protein (data not shown).Conceivably, the processing of fascin protein into a smallerform may be important in the regulation of cell-cell adhesionand junctional complex formation. In this regard, it has beenshown that fascin can be proteolytically cleaved to a 30-kDaform that binds actin filaments in vitro (56). We have alsoobserved that ectopic expression of the full-length mouse fascin(tagged with the Myc epitope) yielded a form of mouse fascinthat electrophoretically migrated as a 80-kDa protein. Al-though an unusually high molecular weight, this 80-kDa Myc-fascin species is recognized by both polyclonal and monoclonalanti-fascin antibodies. Moreover, our preliminary evidence sug-gests that expression of this full-length Myc-fascin cDNA, aswell as a fascin cDNA with the Myc epitope tag on the COOHterminus, in human epithelial tumor cells also generates asimilar 80-kDa species.2 The mechanism by which this highmolecular weight form of Myc-fascin is produced is unknown,although stable protein-protein interactions may potentiallyaccount for the unexpected size. A dimeric form of the trypticfragments of purified recombinant fascin has been shown tomigrate as a stable complex in SDS-PAGE, suggesting thatfascin indeed forms stable dimers under denaturing conditions(83).

It is tempting to consider that the interaction of fascin withcomponents of the apical junction, such as b-catenin, repre-sents a potential regulatory mechanism by which the expres-sion of high levels of exogenous fascin might inhibit glucocor-ticoid-induced cell-cell adhesion and intercellular junctionsformation. Several studies have shown that the regulation ofadherens junction assembly correlated with a reduced bindingof b-catenin to E-cadherin (44–46). However, our resultsshowed that under conditions in which the ectopic expression offull-length Myc-fascin disrupted the glucocorticoid induction ofcell-cell adhesion and intercellular junctions formation, thesame level of b-catenin co-precipitated with the E-cadherinimmune complex. A small fraction of Myc-fascin coprecipitateswith E-cadherin, likely through its interaction with b-catenin.It is not clear whether Myc-fascin and E-cadherin bind tob-catenin in the same complex or in mutually exclusive com-plexes, although it has been shown that fascin competes withE-cadherin for b-catenin binding in an in vitro assay (42). Theectopic expression of the carboxyl-terminal 213 amino acids offascin, which includes the actin and b-catenin-binding sites,failed to disrupt the glucocorticoid induction of tight junctionformation, which suggests that the ability of fascin to disruptthe glucocorticoid-induced cell-cell interactions involves poten-tially complicated structure/function relationships within thefull-length fascin protein. Conceivably, the inhibition of cell-cell adhesion by exogenous fascin could result from an effect offascin on the E-cadherinzb-catenin protein complex, such as toinduce conformational changes that attenuate the adhesivefunction of E-cadherin. Alternatively, fascin might indirectlyregulate the function of E-cadherin via the modulation of thevarious functions of b-catenin or of the cortical actin cytoskel-

eton. Since b-catenin is known to interact with the LEF-1transcription factor (85) and the adenomatous polyposis coliprotein (86), the binding of exogenous fascin to endogenousb-catenin, and perhaps other cellular components, may poten-tially affect the normal activities of b-catenin in the regulationof its interacting proteins, which might in turn affect intercel-lular adhesion and junctions formation.

The coordinate induction of tight junction formation andcell-cell adhesion through fascin down-regulation might reflectan important biological switch in mammary cell differentiation.Our results, in which the ectopic expression of fascin disruptedthe glucocorticoid stimulation of tight junctions and adherensjunctions, implicate fascin as a negative regulator of cell-cellinteractions. Thus, the glucocorticoid down-regulation of en-dogenous fascin expression is necessary for the induced forma-tion of the apical junctional complex in rodent mammary tumorcells. Glucocorticoid receptors can stimulate or inhibit genetranscription by their selective DNA binding to glucocorticoidresponse elements (68–70), and by their ability to directly bindto and attenuate the function of certain transcription factors(68, 71–73). If the fascin gene is a direct target of glucocorticoidsignaling, it is likely that the glucocorticoid receptor inhibitsfascin gene transcription by interfering with specific transcrip-tion factors that act on the fascin gene promoter or by inducingthe expression of transcriptional inhibitors that target the fas-cin gene. Unraveling the signaling pathway by which glucocor-ticoid down-regulates fascin protein expression would be animportant step toward the understanding of glucocorticoid reg-ulation of epithelial junctions formation, and perhaps the roleof intercellular junctions in the regulation of cell growth anddifferentiation.

Acknowledgments—We express our appreciation to Paul Woo forcritical evaluation of this manuscript and helpful experimental sugges-tions. We thank Tran Van, Minnie Wu, and Kenneth Oh for technicalassistance. We are also grateful to Jerry Kapler for excellent photogra-phy and both Anna Fung and Tran Van for help in the preparation ofthis manuscript.

REFERENCES

1. Farquhar, M. G., and Palade, G. E. (1963) J. Cell Biol. 17, 375–4122. Stevenson, B. R., and Paul, D. L. (1989) Curr. Opin. Cell Biol. 1, 884–8913. Gumbiner, B. (1990) Curr. Opin. Cell Biol. 2, 881–8874. Gumbiner, B. (1987) Am. J. Physiol. 253, C749–C7585. Balda, M. S., Whitney, J. A., Flores, C., Gonzalez, S., Cereijido, M., and Matter,

K. (1996) J. Cell Biol. 134, 1031–10496. Furuse, M., Hirase, T., Itoh, M., Nagafuchi, A., Yonemura, S., Tsukita, S., and

Tsukita, S. (1993) J. Cell Biol. 123, 1777–17887. McCarthy, K. M., Skare, I. B., Stankewich, M. C., Furuse, M., Tsukita, S.,

Rogers, R. A., Lynch, R. D., and Schneeberger, E. E. (1996) J. Cell Sci. 109,2287–2298

8. Wong, V., and Gumbiner, B. M. (1997) J. Cell Biol. 136, 399–4099. Stevenson, B. R., Siliciano, J. D., Mooseker, M. S., and Goodenough, D. A.

(1986) J. Cell Biol. 103, 755–76610. Gumbiner, B., Lowenkopf, T., and Apatira, D. (1991) Proc. Natl. Acad. Sci.

U. S. A. 88, 3460–346411. Jesaitis, L. A., and Goodenough, D. A. (1994) J. Cell Biol. 124, 949–96112. Citi, S., Sabanay, H., Jakes, R., Geiger, B., and Kendrick-Jones, J. (1988)

Nature 333, 272–27613. Citi, S., Sabanay, H., Kendrick-Jones, J., and Geiger, B. (1989) J. Cell Sci. 93,

107–12214. Zhong, Y., Saitoh, T., Minase, T., Sawada, N., Enomoto, K., and Mori, M.

(1993) J. Cell Biol. 120, 477–48315. Zhong, Y., Enomoto, K., Isomura, H., Sawada, N., Minase, T., Oyamadaa, M.,

Konishi, Y., and Mori, M. (1994) Exp. Cell Res. 214, 614–62016. Keon, B. H., Schafer, S., Kuhn, C., Grund, C., and Franke, W. W. (1996) J. Cell

Biol. 134, 1003–101817. Zahraoui, A., Joberty, G., Arpin, M., Fontaine, J. J., Hellio, R., Tavitian, A.,

and Louvard, D. (1994) J. Cell Biol. 124, 101–11518. Gumbiner, B., Stevenson, B., and Grimaldi, A. (1988) J. Cell Biol. 107,

1575–158719. Behrens, J., Birchmeier, W., Goodman, S. L., and Imhof, B. A. (1985) J. Cell

Biol. 101, 1307–131520. Kemler, R. (1993) Trends Genet. 9, 317–32121. Eaton, S., and Simons, K. (1995) Cell 82, 5–822. Rodriguez-Boulan, E., and Nelson, W. J. (1989) Science 245, 719–72523. McNeill, H., Ozawa, M., Kemler, R., and Nelson, W. J. (1990) Cell 62, 309–31624. Drubin, D. G., and Nelson, W. J. (1996) Cell 84, 335–34425. Gumbiner, B. M. (1996) Cell 84, 345–35726. Ozawa, M., Ringwald, M., and Kemler, R. (1990) Proc. Natl. Acad. Sci. U. S. A.

Steroid Control of Cell-Cell Interactions5452

by guest on October 9, 2020

http://ww

w.jbc.org/

Dow

nloaded from

87, 4246–425027. Nagafuchi, A., and Takeichi, M. (1988) EMBO J. 7, 3679–368428. Herrenknecht, K., Ozawa, M., Eckerskorn, C., Lottspeich, F., Lentner, M., and

Kemler, R. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 9156–916029. Nagafuchi, A., Takeichi, M., and Tsukita, S. (1991) Cell 65, 849–85730. McCrea, P. D., Turck, C. W., and Gumbiner, B. (1991) Science 254, 1359–136131. Knudsen, K. A., and Wheelock, M. (1992) J. Cell Biol. 118, 671–67932. Cowin, P., Kapprell, H.-P., Franke, W. W., Tamkun, J., and Hynes, R. O. (1986)

Cell 46, 1063–107333. Peifer, M., McCrea, P. D., Green, K. J., Wieschaus, E., and Gumbiner, B. M.

(1992) J. Cell Biol. 118, 681–69134. Aghib, D. F., and McCrea, P. D. (1995) Exp. Cell Res. 218, 359–36935. Reynolds, A. B., Daniel, J., McCrea, P. D., Wheelock, M. J., Wu, J., and Zhang,

Z. (1994) Mol. Cell. Biol. 14, 8333–834236. Shibamoto, S., Hayakawa, M., Takeuchi, K., Hori, T., Miyazawa, K.,

Kitamura, N., Johnson, K. R., Wheelock, M. J., Matsuyoshi, N., Takeichi,M., and Ito, F. (1995) J. Cell Biol. 128, 949–958

37. Knudsen, K. A., Soler, A. P., Johnson, K. R., and Wheelock, M. J. (1995) J. CellBiol. 130, 67–77

38. Drenckhahn, D., and Franz, H. (1986) J. Cell Biol. 102, 1843–185239. Tsukita, S., Hieda, Y., and Tsukita, S. (1989) J. Cell Biol. 108, 2369–238240. Tsukita, S., Itoh, M., and Tsukita, S. (1989) J. Cell Biol. 109, 2905–291541. Stoker, A. W., Kellie, S., and Wyke, J. A. (1986) J. Virol. 58, 876–88342. Tao, Y. S., Edwards, R. A., Tubb, B., Wang, S., Bryan, J., and McCrea, P. D.

(1996) J. Cell Biol. 134, 1271–128143. Balda, M. S., Gonzalez-Mariscal, L., Contreras, R. G., Macias-Silva, M.,

Torres-Marquez, M. E., Garcıa Sainz, J. A., and Cereijido, M. (1991)J. Membr. Biol. 122, 193–202

44. Behrens, J., Vakaet, L., Friis, R., Winterhager, E., Van Roy, F., Mareel, M. M.,and Birchmeier, W. (1993) J. Cell Biol. 120, 757–766

45. Kinch, M. S., Clark, G. J., Der, C. J., and Burridge, K. (1995) J. Cell Biol. 130,461–471

46. Fialka, I., Schwarz, H., Reichmann, E., Oft, M., Busslinger, M., and Beug, H.(1996) J. Cell Biol. 132, 1115–1132

47. Nusrat, A., Giry, M., Turner, J. R., Colgan, S. P., Parkos, C. A., Carnes, D.,Lemichez, E., Boquet, P., and Madara, J. L. (1995) Proc. Natl. Acad. Sci.U. S. A. 92, 10629–10633

48. Stuart, R. O., and Nigam, S. K. (1995) Proc. Natl. Acad. Sci. U. S. A. 92,6072–6076

49. Madara, J. L., Parkos, C., Colgan, S., Nusrat, A., Atisook, K., and Kaoutzani,P. (1992) Ann. N. Y. Acad. Sci. 664, 47–60

50. Braga, V. M., Machesky, L. M., Hall, A., and Hotchin, N. A. (1997) J. Cell Biol.137, 1421–1431

51. Jou, T. S., and Nelson, W. J. (1998) J. Cell Biol. 142, 85–10052. Gopalakrishnan, S., Raman, N., Atkinson, S. J., and Marrs, J. A. (1998) Am. J.

Physiol. C798–C80953. Takaishi, K., Sasaki, T., Kotani, H., Nishioka, H., and Takai, Y. (1997) J. Cell

Biol. 139, 1047–105954. Owaribe, K., Kodama, R., and Eguchi, G. (1981) J. Cell Biol. 90, 507–51455. Madara, J. L. (1987) Am. J. Physiol. 253, C171–C17556. Edwards, R. A., and Bryan, J. (1995) Cell Motil. Cytoskeleton 32, 1–9

57. Otto, J. J. (1994) Curr. Opin. Cell Biol. 6, 105–10958. Zettl, K. S., Sjaastad, M. D., Riskin, P. M., Parry, G., Machen, T. E., and

Firestone, G. L. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 9069–907359. Singer, K. L., Stevenson, B. R., Woo, P. L., and Firestone, G. L. (1994) J. Biol.

Chem. 269, 16108–1611560. Buse, P., Woo, P. L., Alexander, D. B., Reza, A., and Firestone, G. L. (1995)

J. Biol. Chem. 270, 28223–2822761. Buse, P., Woo, P. L., Alexander, D. B., Cha, H. H., Reza, A., Sirota, N. D., and

Firestone, G. L. (1995) J. Biol. Chem. 270, 6505–651462. Woo, P. L., Cha, H. H., Singer, K. L., and Firestone, G. L. (1996) J. Biol. Chem.

271, 404–41263. Topper, Y. J., and Freeman, C. S. (1980) Physiol. Rev. 60, 1049–110664. Snedeker, S. M., Brown, C. F., and DiAugustine, R. P. (1991) Proc. Natl. Acad.

Sci. U. S. A. 88, 276–28065. Pitelka, D. R., Hamamoto, S. T., Duafala, J. G., and Nemanic, M. K. (1973)

J. Cell Biol. 56, 797–81866. Goya, L., Maiyar, A. C., Ge, Y., and Firestone, G. L. (1993) Mol. Endocrinol. 7,

1121–113267. Webster, M. K., Guthrie, J., and Firestone, G. L. (1990) J. Biol. Chem. 265,

4831–483868. Beato, M., Herrlich, P., and Schutz, G. (1995) Cell 83, 851–85769. Zilliacus, J., Wright, A. P., Carlstedt-Duke, J., and Gustafsson, J.-A. (1995)

Mol. Endocrinol. 9, 389–40070. Truss, M., and Beato, M. (1993) Endocr. Rev. 14, 459–47971. McEwan, I. J., Wright, A. P., and Gustafsson, J. A. (1997) Bioassays 19,

153–16072. Horwitz, K. B., Jackson, T. A., Bain, D. L., Richer, J. K., Takimoto, G. S., and

Tung, L. (1996) Mol. Endocrinol. 10, 1167–117773. Wahli, W., and Martinez, E. (1991) FASEB J. 5, 2243–224974. Dharmawardhane, S., Warren, V., Hall, A. L., and Condeelis, J. (1989) Cell

Motil. Cytoskeleton 13, 57–6375. Rupp, R. A., Snider, L., and Weintraub, H. (1994) Genes Dev. 8, 1311–132376. Turner, D. L., and Weintraub, H. (1994) Genes Dev. 8, 1434–144777. Edwards, R. A., Herrera-Sosa, H., Otto, J., and Bryan, J. (1995) J. Biol. Chem.

270, 10764–1077078. Webster, M. K., Goya, L., Ge, Y., Maiyar, A. C., and Firestone, G. L. (1993) Mol.

Cell. Biol. 13, 2031–204079. Maiyar, A. C., Phu, P. T., Huang, A. J., and Firestone, G. L. (1997) Mol.

Endocrinol. 11, 312–32980. Adams, C. L., Nelson, W. J., and Smith, S. J. (1996) J. Cell Biol. 135,

1899–191181. Angres, B., Barth, A., and Nelson, W. J. (1996) J. Cell Biol. 134, 549–55782. Yamakita, Y., Ono, S., Matsumura, F., and Yamashiro, S. (1996) J. Biol. Chem.

271, 12632–1263883. Ono, S., Yamakita, Y., Yamashiro, S., Matsudaira, P. T., Gnarra, J. R.,

Obinata, T., and Matsumura, F. (1997) J. Biol. Chem. 272, 2527–253384. Adams, J. C. (1997) Mol. Biol. Cell 8, 2345–236385. Behrens, J., von Kries, J. P., Kuhl, M., Bruhn, L., Wedlich, D., Grosschedl, R.,

and Birchmeier, W. (1996) Nature 382, 638–64286. Rubinfeld, B., Albert, I., Porfiri, E., Fiol, C., Munemitsu, S., and Polakis, P.

(1996) Science 272, 1023–1025

Steroid Control of Cell-Cell Interactions 5453

by guest on October 9, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Vivian Wong, Dixie Ching, Pierre D. McCrea and Gary L. FirestoneMammary Epithelial Tumor Cells

Steroid-induced Formation of Tight Junctions and Cell-Cell Interactions in Rat Glucocorticoid Down-regulation of Fascin Protein Expression Is Required for the

doi: 10.1074/jbc.274.9.54431999, 274:5443-5453.J. Biol. Chem. 

  http://www.jbc.org/content/274/9/5443Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/274/9/5443.full.html#ref-list-1

This article cites 85 references, 54 of which can be accessed free at

by guest on October 9, 2020

http://ww

w.jbc.org/

Dow

nloaded from