β-Catenin and NF-κB cooperate to regulate the uPA/uPAR system in cancer cells

Abstract

Expression of the urokinase plasminogen activator (uPA) and urokinase plasminogen activator receptor (uPAR) has recently been shown to be directly regulated by the Wnt/β-catenin signaling pathway in colon cancer cells, through β-catenin binding to T-cell factor binding element motifs present in their gene promoters. In our study, we present evidence that inhibition of β-catenin causes upregulation of uPA/uPAR gene expression enhancing invasive potential. Using MCF-7, MDA-MB-231 (breast cancer cells) and SW480 (colon cancer cells), we found that siRNA-mediated silencing of β-catenin increased uPA, uPAR and plasminogen activator inhibitor-1 (PAI-1) expression at the mRNA and protein levels. This increase was responsible for the observed enhanced invasive capacity of MDA-MB-231 and SW480 cancer cells. In addition, β-catenin stabilization and accumulation by lithium chloride treatment, a well-known inhibitor of glycogen synthase kinase-3β (GSK-3β), or by β-catenin/T-cell factor-4 expression vectors transfection led to a decrease in uPA, uPAR and PAI-1 mRNA expression in the studied cancer models. Treatment of β-catenin siRNA-transfected cells with a specific inhibitor of nuclear factor-kappaB (NF-κB), SN50, significantly reduced enhancement of uPA, uPAR and PAI-1 expression and cancer cell invasion, observed in β-catenin siRNA-transfected cells. Furthermore, β-catenin siRNA-treated cells exhibited NF-κB nuclear accumulation. These data suggest that β-catenin regulates the uPA/uPAR system in cooperation with NF-κB transcription factor, which constitutes a novel mechanism of regulation.

A major feature of cancer cells is their ability to migrate and to invade and develop in surrounding or distant tissues. Cancer metastasis is a complex multistep process involving tumor cell proliferation, migration across the extracellular matrix barrier, angiogenesis and cell implantation in distant organs.1 Proteolysis of the extracellular matrix is a crucial event for tumor cells invasion through the surrounding tissue.2

Among the multiple protease-mediated events, the urokinase plasminogen activator (uPA), its specific receptor (uPAR) and its primary inhibitor, plasminogen activator inhibitor-1 (PAI-1), have been shown to play a critical role in the regulation of cancer cell motility, invasion and metastasis.3, 4 Serine protease uPA converts the proenzyme plasminogen to plasmin able to degrade extracellular matrix directly or indirectly through activation of matrix metalloproteinases. Although uPA is produced by both normal and tumor cells, an upregulation of uPA/uPAR system partners has been observed in patients with invasive cancers.5 Expression of genes of the human uPA/uPAR system has been reported to be under the regulation of extracellular mediators as growth factors and cytokines and intracellular signaling through β-catenin and transcription factors such as nuclear factor-kappaB (NF-κB).6–8

β-catenin is a multifunctional protein involved in normal embryonic development and cell polarity as well as in abnormal cell function such as carcinogenesis.9 β-catenin participates in cadherin-mediated epithelial cell–cell adhesion and in cell signaling through the canonical Wnt/β-catenin pathway and noncanonical pathways.10, 11 In the absence of Wnt/β-catenin activation, β-catenin is maintained in a multiprotein complex including adenomatous polyposis coli (APC), glycogen synthase kinase-3β (GSK-3β) and axin.9, 12 GSK-3β phosphorylates several specific serine and threonine residues in the NH2-terminal region of β-catenin that is necessary for its subsequent ubiquitin-mediated proteasomal degradation. Mutations in either APC or β-catenin genes or inhibition of GSK-3β disrupt formation and inhibit the activity of the multiprotein complex, leading to stabilization and accumulation of β-catenin.13, 14 Stabilization and accumulation of β-catenin leads to its translocation in the nucleus and to the formation of a complex with the nuclear transcription factors, T-cell factor (Tcf)/lymphocyte enhancer factor (Lef).13, 14 The β-catenin/Tcf complex binds to the Tcf binding elements (TBE) and activates transcription of specific target genes such as cyclin D1,15c-myc,16 matrix metalloproteinase (MMP)-717 and vascular endothelial growth factor (VEGF).18

Little is known about the regulation of the uPA/uPAR system by β-catenin in cancer cells. To our knowledge, only two studies, concerning colon cancer cells, have reported a direct regulation of uPA8 and uPAR19 gene expression by β-catenin, through its binding to the TBE motifs present in their promoters. This prompted us to explore whether β-catenin regulates the uPA/uPAR system in breast cancer cell lines through a similar mechanism.

Material and Methods

Cell culture and treatment

Human breast cancer cells MCF-7 and MDA-MB-231 and human colon cancer cells SW480 were obtained from the American Type Culture Collection (Manassas, VA). The MCF-7 cells were maintained in Dulbecco's minimal essential medium (DMEM: F12) supplemented with 1% of nonessential amino acids (Invitrogen, Cergy-Pontoise, France) and 10% fetal bovine serum (FBS). The MDA-MB-231 cells were grown in DMEM with 10% FBS, 1% sodium pyruvate and 1% L-glutamine (Invitrogen). SW480 cells were maintained in DMEM containing 10% FBS. Culture media were supplemented with 100 U/ml penicillin and 100 μg/ml streptomycin. Cells were cultured at 37°C in a humidified incubator with 5% CO₂. Cells were cultured in presence or absence of the following reagents: uPA inhibitor amiloride (Sigma, St Louis, MO), uPA blocking antibody (American Diagnostica, Greenwich, CT), inhibitor of NF-κB SN50 (Santa Cruz Biotechnology, San Diego, CA) and lithium chloride (LiCl, Sigma).

SiRNA transfection

We tested Dharmacon SMARTpool (Dharmacon, Lafayette, CO) containing four siRNA β-catenin (ON-TARGETplus SiRNA) and we selected the two most efficient siRNA. Scramble control siRNA (Silencer FAM-labeled negative control) was from Ambion (Austin, TX). SiRNAs (10 nmol/L) were transfected with Oligofectamine Reagent (Invitrogen) according to the manufacturer's instructions. Briefly, cells (2.5 × 10⁵) were plated in six-well plates and incubated overnight in growth medium before transfection. For transfection, siRNA and Oligofectamine were mixed with serum-free opti-MEM medium (Invitrogen) and added into each well, which contained serum-free opti-MEM medium. Four hours later, medium containing 20% FBS was added into each well. Cells were then incubated for 24 hr for RNA extraction or 48 hr for Western blot and zymography analysis. When siRNA-transfected cells were treated with NF-κB inhibitor SN50, it was added 4 hr after siRNA.

Expression vectors transfection

Cells (2.5 × 10⁵) seeded in six-well plates were transient transfected by stable β-cateninS45 (1 μg) and of Tcf-4 (1 μg) cloned in pcDNA3 expression vectors (kindly provided by Paul Polakis, Genentech, USA) using Lipofectamine 2000 transfection reagent (Invitrogen) as recommended by the manufacturer. Controls were generated by transfecting the cells with 2 μg of pcDNA3 vector. Six hours later, medium was changed for the corresponding growth medium. Cells were then incubated for 3, 24 and 48 hr for RNA analysis.

LiCl treatment

For LiCl treatment, cells (2.5 × 10⁵) were cultured in growth medium (six-well plate) and treated at 60% confluence for 6 hr at the concentrations of 50 mM for MCF-7 and MDA-MB-231 cells or 10 mM for SW480 cells. The NF-κB inhibitor SN50 (100 μg/ml) was added 1 hr after LiCl. At the end of treatment, cells were harvested for RNA and protein analysis.

RNA extraction and reverse transcription

RNA was isolated using TRIzol reagent according to the manufacturer's instructions (Invitrogen). Reverse transcription of 1 μg RNA was performed using RevertAid H Minus M-MuLV Reverse Transcriptase (Fermentas, Hanover, MD).

Real-time quantitative PCR

β-catenin, cyclin D1, c-myc, uPA, uPAR, PAI-1, MMP-1 and MMP-9 transcript levels were measured by PCR using Perfect Master Mix-Probe (AnyGenes, Creteil France) on LightCycler 2.0 (Roche, Indianapolis, IN). The expression levels of interesting transcripts were normalized to the housekeeping PPIA (peptidylprolyl isomerase A) and TBP (TATA-box binding protein) gene transcripts. As there was no difference between control genes, results are presented as copies of target gene per copy of PPIA. Gene expression levels were determined using standard calibration curves prepared from gene-specific PCR products. All PCRs were done in duplicate.

Western blotting

Cells were harvested, washed twice in PBS 1× and lysed in a lysis buffer (50 mM Tris-HCl, pH 7.4, 137 mM NaCl, EDTA 2 mM and 1% nonidet P-40) containing protease inhibitors (Roche Diagnostic, Meylan, France). For β-catenin Western blotting, 1% sodium deoxycholate and 0.1% sodium dodecyl sulfate were added in the lysis buffer. The cleared lysates were collected by centrifugation at 12,000g for 10 min at 4°C. The protein concentrations in lysates were measured by BCA™ protein assay kit (Pierce, MA). Equal amounts of protein from cell lysates (10 or 30 μg of proteins) were electrophoresed under reducing conditions on 4–12% acrylamide gels (Invitrogen) and transferred onto nitrocellulose membranes. Membranes were incubated overnight with primary antibody using the following antibodies and dilutions: β-catenin 1/2,000, actin 1/4,000 (C 2206, A 2066, respectively, Sigma), uPA 1/500 and uPAR 1/500 (389, 399R, respectively, American Diagnostica). After washing and incubation with a secondary antibody (horseradish peroxidase-conjugated anti-mouse or anti-rabbit IgG 1/5,000, Dako, Carpentaria, CA), protein bands were visualized by using the Western blotting Luminol Reagent (Santa Cruz Biotechnology) according to the manufacturer's guidelines.

uPA and PAI-1 ELISA

uPA and PAI-1 concentrations were determined in cell culture media by a uPA ELISA kit from American Diagnostica and a PAI-1 ELISA kit from Diagnostica Stago (Asnieres, France) according to the manufacturer's protocols. Briefly, β-catenin siRNA or scrambled siRNA-treated cells were serum-deprived for 24 hr, after which medium samples were harvested and centrifugated. uPA and PAI-1 levels in each sample were related to the number of cells in the corresponding well at the time of sample collection. Receptor-bound uPA and uPA complexed with PAI-1 and PAI-2 are all recognized by this assay.

Casein zymography

Casein zymography was done as described previously.20 Briefly, treated or untreated cells were serum-deprived for 24 hr, after which medium samples were harvested and centrifugated. After normalization for correspondent lysate, conditioned media were electrophoresed on 10% SDS-PAGE gels containing 2 mg/ml casein (Sigma) and 10 μg/ml plasminogen (Calbiochem, San Diego, SA) for uPA activity. The gels were washed two times for 20 min at room temperature in distilled water containing 2.5% (v/v) Triton-X 100 before incubation at 37°C overnight (MDA-MB-231 cells) or 48 hr (MCF-7 and SW480 cells) in a reaction buffer (Tris/HCl 50 mM and CaCl₂ 2 mM). The gels were then stained in 0.5% Coomassie brilliant blue R-250 solution and destained with distilled water containing acetic acid (10%, v/v) and methanol (40%, v/v). Clear zones against the blue background indicated the presence of caseinase activity.

Invasion assay

The ability of the cells to migrate through Matrigel (10 μg, BD Biosciences, San Jose, CA) polycarbonate-coated filters (8 μm) was measured by using Transwell chambers (BD Biosciences). Cells were detached with Versene (Invitrogen) and added to each transwell (1 × 10⁵ cells) in DMEM containing 0.1% bovine serum albumin (BSA) and 10% FBS being used as chemoattractant in the bottom chamber. Amiloride (20 μM), uPA blocking antibody (100 μg/ml) or SN50 (100 μg/ml) were added to the cells previously transfected with siRNA β-catenin or siRNA control. After 24 hr (MDA-MB-231 cells) or 48 hr (SW480 cells), cells on the upper filter surface were removed by scrubbing with a cotton swab. Invasive cells on the downside face of the membrane were fixed and stained with Diff-Quik (Dade Behring, Deerfield, IL) and counted.

β-Catenin and NF-κB staining by immunofluorescence

Cells were fixed in 4% paraformaldehyde solution and permeabilized in 50 mM Tris, pH 7.4, 150 mM NaCl and 0.2% Tween 20 buffer supplemented with 0.4% Triton X-100 for 10 min. Cells were successively blocked in 3% BSA and incubated with the purified mouse anti-β-catenin 1/50 (610153, BD Transduction Laboratories) and with an Alexa Fluor 488 goat anti-mouse 1/50 (A21121, Invitrogen) or with the rabbit anti-NF-κB p65 1/150 (sc-372, Santa Cruz Biotechnology) and with an Alexa Fluor 568 goat anti-rabbit 1/50. The nuclei were stained with 4,6-diamidino-2-phenylindole (DAPI) from Sigma. Images were acquired by immunofluorescence microscopy on a Zeiss Axiovert 200M microscope (Zeiss, Oberkochen, Germany) and a Axiocam MRM camera (Zeiss) using the Axiovision v4.5.0.0 software (Zeiss). Immunofluorescence pictures were acquired by confocal microscopy on a Zeiss LSM 510 META confocal laser microscope (Zeiss, Oberkochen, Germany) with a Plan Apochromat 63× N.A.1.4 oil-immersion objective using the LSM510 software v4.0 (Zeiss).

Statistical analysis

The presented data are the mean of the results of at least three independent experiments performed in duplicate and presented as mean ± SEM. For statistical analysis, the nonparametric Wilcoxon test was used.

APC: adenomatous polyposis coli; GSK-3β: glycogen synthase kinase-3β; Lef: lymphocyte enhancer factor; LiCl: lithium chloride; NF-κB: nuclear factor-kappaB; PAI-1: plasminogen activator inhibitor-1; TBE: Tcf binding element; Tcf: T-cell factor; uPA: urokinase plasminogen activator; uPAR: urokinase plasminogen activator receptor

Results

β-Catenin inhibition upregulates uPA/uPAR system in breast and colon cancer cells

First, we examined the invasive capacity and the expression patterns of β-catenin and uPA/uPAR system in human breast cancer cell lines, MCF-7 and MDA-MB-321, and colon cancer cell line, SW480. MDA-MB-231 and SW480 cells exhibit invasive capacity (Fig. 1a) associated with a high expression and activity of the urokinase plasminogen activator system (uPA, uPAR and PAI-1) as measured by q-RT-PCR, Western blot (Figs. 1b and 1e) and uPA zymography (Fig. 1c). MCF-7 cells have no invasive capacity and low levels of uPA, uPAR and PAI-1 expression and weak uPA activity. β-Catenin expression analyzed by Western blot was not associated with cell invasive capacity (Fig. 1e). However, β-catenin subcellular localization analyzed by immunofluorescence was different in the three cancer models. In the noninvasive MCF-7 cell line, β-catenin was mainly localized in membranes and cytoplasm, whereas β-catenin was localized in the cytoplasm and nucleus in the invasive MDA-MB-231 and SW480 cancer cell lines (Fig. 1d).

Figure 1. β-Catenin and uPA/uPAR expression by MCF-7, MDA-MB-231 and SW480 cancer cell lines. (a) Invasive capacity of MCF-7, MDA-MB-231 and SW480 cells (1 × 10⁵ cells) measured after an incubation of 24 hr for MDA-MB-231 and 48 hr for MCF-7 and SW480 cells. (b) Expression of uPA, uPAR and PAI-1 transcripts quantified using q-RT-PCR was higher in MDA-MB-231 and in SW480 cells when compared to that observed in MCF-7 cells. The columns show the mean of relative expression to PPIA housekeeping gene of three independent experiments (mean ± SEM). (c) Casein zymography showed higher activity of uPA in MDA-MB-231 and SW480 invasive cell lines than in MCF-7 cells. (d) β-Catenin immunofluorescence analysis in the three cell lines. DAPI stains the nuclei in blue and β-catenin stains in green. Pictures are representative of three experiments (×40). (e) Western blot analysis of the expression of β-catenin in the three cell lines. The uPA protein expression was higher in MDA-MB-231 and in SW480 cell lines compared to MCF-7 cell line. Actin was used as a loading control. A representative experiment of three independent experiments is shown. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

To study the role of β-catenin in the regulation of uPA/uPAR system, we used a RNA silencing strategy to downregulate endogenous β-catenin expression. We first used a SMARTpool from Dharmacon that inhibited efficiently β-catenin expression in the three studied cell lines (data not shown). After that the four siRNA of the SMARTpool were used separately. The two most efficient siRNA downregulated β-catenin expression at the mRNA level by ∼75% for MCF-7, ∼85% for MDA-MB-231 and ∼85% for SW480 cells (Fig. 2a) and at the protein level, analyzed by Western blot and immunofluorescence (Figs. 2a and 2b) when compared to the control siRNA. The inhibition of β-catenin was associated with an increase in uPA/uPAR system partner expression. Indeed, the β-catenin siRNA upregulated uPA, uPAR and PAI-1 expression in the three cancer models at both the protein and the mRNA levels (Fig. 3). Indeed, a statistically significant increase of uPA, uPAR and PAI-1 mRNA levels of 2.6-, 3.5- and 9.5-fold in MCF7 cells, 3.3- and 2.1-fold in MDA-MB-231 and 11.7-, 2.9- and 4.5-fold in SW480 cells was observed, respectively, compared to the control siRNA. Similarly, uPA and PAI-1 levels measured by ELISA were increased between twofold and fivefold in the cell supernatants after siRNA treatment. Moreover, uPA activity was increased in β-catenin siRNA-transfected cells supernatants (Fig. 3b).

Figure 2. Efficiency of β-catenin siRNA transfection. MCF-7, MDA-MB-231 and SW480 cells were transfected with two different β-catenin siRNA or with the control siRNA at 10 nmol/l. (a) Analysis by q-RT-PCR showed a decreased expression of β-catenin 24 hr after transfection in the three cancer cell models. The columns show the mean of relative expression to PPIA housekeeping gene of three independent experiments (mean ± SEM, *p < 0.05). Decrease in β-catenin protein expression was evaluated by Western blot analysis of 10 μg cell lysates 48 hr after transfection. Actin was used as a loading control. A representative blot of three independent experiments is shown. (b) Immunofluorescence confirmed that β-catenin expression decreased 48 hr after transfection. Nuclei are stained in blue with DAPI and β-catenin in green (×40). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Figure 3. β-Catenin siRNA upregulates uPA/uPAR system expression in the three cancer cell lines. (a) Western blot analysis showed that transfection of MCF-7, MDA-MB-231 and SW480 cancer cell lines with β-catenin siRNA led to an increase in uPA, uPAR and PAI-1 protein expression compared to control siRNA. Actin was used as a loading control. Representative blot of three independent experiments is shown. (b) Enzymatic activity of uPA, analyzed by casein zymography, was enhanced by β-catenin siRNA in the three cancer models. uPA and PAI-1 protein levels in cell supernatants were quantified by ELISA. The columns represent the mean of three independent experiments (mean ± SEM, *p < 0.05). (c) uPA, uPAR, PAI-1, cyclin D1 and c-myc and (d) MMP-1, MMP-9 and TBP mRNA expression as analyzed by q-RT-PCR is shown. The columns show the mean of relative expression to PPIA housekeeping gene of three independent experiments (mean ± SEM, *p < 0.05).

Two other proteolytic systems have also been explored in these models. We observed a statistically significant increase of MMP-1 and MMP-9 expression in the three cell lines after β-catenin siRNA treatment compared to control cells (Fig. 3d), whereas the expression of tissue-type plasminogen activator (tPA) was not affected (data not shown).

We next investigated the expression of β-catenin target genes cyclin D1 and c-myc. As expected, we observed a decrease in their expression in MDA-MB-231 and SW480 cells transfected with β-catenin siRNA compared to control cells. Their expression was not affected in MCF7 cells. TBP mRNA expression, used as a negative control, was not affected by this transfection (Fig. 3d).

β-Catenin inhibition promotes invasive capacity of MDA-MB-231 and SW480 cells

The uPA/uPAR system has previously been implicated in cancer invasion.3 We therefore evaluated whether upregulation of the uPA/uPAR system by β-catenin siRNA affected the invasive capacity of MDA-MB-231 and SW480 cancer cells. Using a Matrigel invasion assay, we observed that β-catenin siRNA significantly enhanced the invasive capacity of MDA-MB-231 and SW480 cell lines by 23 and 59%, respectively, compared to control cells (Fig. 4). This observed effect of β-catenin siRNA was reduced in the presence of the uPA inhibitor amiloride (by 28% for MDA-MB-231 and by 31% for SW480 cells) or in the presence of a uPA blocking antibody (by 29% for MDA-MB-231 and 40% for SW480 cells) (Figs. 4a and 4b). These data suggest that β-catenin inhibition promotes the invasive capacity of MDA-MB-231 and SW480 tumor cells through the upregulation of uPA expression.

Figure 4. β-catenin siRNA promotes tumor cell invasion: implication of uPA activity. β-Catenin siRNA transfection increased the invasive capacity of MDA-MB-231 cells (a) and SW480 cells (b) compared to the control siRNA. Transfected cells were treated with uPA inhibitor, amiloride (20 μM) or uPA blocking antibody (100 μg/ml). Amiloride or uPA blocking antibody was added to the upper compartment with the cells during 24 hr for MDA-MB-231 cells or 48 hr for SW480 cells and decreased significantly the invasive capacity of β-catenin siRNA-transfected cells. The values are mean ± SEM, *p < 0.05.

NF-κB cooperates with β-catenin in the regulation of uPA/uPAR system in breast and colon cancer cell lines

NF-κB has been reported to upregulate uPA,21uPAR22 and PAI-123 gene expression. Furthermore, cytoplasmic β-catenin can interact directly with NF-κB, thus inhibiting its translocation into the nucleus.24, 25 We therefore investigated whether NF-κB participates in the regulation of uPA, uPAR and PAI-1 expression by β-catenin. MDA-MB-231 and SW480 tumor cells were treated with SN50, an inhibitor of NF-κB nuclear translocation. uPA, uPAR and PAI-1 mRNA levels decreased after SN50 treatment. The increase in uPA, uPAR and PAI-1 mRNA levels induced by β-catenin siRNA was significantly reduced after SN50 treatment (Fig. 5a). Indeed, a statistically significant decrease in uPA, uPAR (∼50%) and PAI-1 (∼70%) levels in MCF-7 cells, uPA, uPAR and PAI-1 levels (∼45%) in MDA-MB-231 cells and uPAR (∼24%), uPA and PAI-1 (∼38%) levels in SW480 cells was observed in the presence of SN50, compared to transfected cells untreated with SN50. In addition, treatment of β-catenin siRNA-transfected cells with SN50 reduced the invasive capacity of MDA-MB-231 and SW480 cells compared to untreated siRNA-transfected cells by 78 and 58%, respectively. Invasive capacity of these cell lines was reduced after treatment with SN50 (Fig. 5b).

Figure 5. Inhibition of NF-κB by SN50 in β-catenin siRNA-transfected cells decreases their mRNA expression of uPA/uPAR system and cancer cell invasion. (a) Analysis by q-RT-PCR showed that uPA, uPAR and PAI-1 mRNA levels in the three β-catenin siRNA-transfected cells were decreased by SN50 (100 μg/ml) treatment. SN50 was added 4 hr after transfection. The columns show the mean of relative expression to PPIA housekeeping gene of three independent experiments (mean ± SEM, *p < 0.05). (b) Treatment with SN50 reduced the invasive capacity of β-catenin siRNA-transfected cells (mean ± SEM, *p < 0.05). (c) NF-κB immunofluorescence analysis of the three cell lines using confocal microscopy. Representative pictures show that β-catenin siRNA-transfected cells exhibited a nuclear accumulation of NF-κB 6 hr after transfection compared to scramble siRNA-transfected cells. Nuclei are stained in blue with DAPI and NF-κB in red (×63). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

As NF-κB activity is known to necessitate translocation to the nucleus, we examined the cell distribution of NF-κB within the three types of cells after transfection with β-catenin siRNA by immunofluorescence. As is shown in Figure 5c, siRNA-treated cells exhibited an increase of NF-κB staining in the nucleus with a number of cells showing its nuclear accumulation, not observed in control cells. These data are in favor of NF-κB activation in knock down β-catenin cells. Therefore, our results suggest that NF-κB can cooperate with β-catenin in the regulation of the uPA/uPAR system.

β-Catenin overexpression by LiCl treatment or β-catenin/Tcf-4 expression vectors transfection downregulates uPA/uPAR system in cancer cells

Next, we sought to determine whether the stabilization and accumulation of the β-catenin protein using LiCl, a well-known GSK-3β inhibitor,26, 27 had a regulatory effect on uPA/uPAR system expression. LiCl (10–50 μM) induced a concentration-dependent accumulation of the β-catenin protein in all three cell lines (Fig. 6a). According to these results, MDA-MB-231, MCF-7 and SW480 tumor cells were treated with optimal concentrations of LiCl: 50 μM for MCF-7 and MDA-MB-231 cells and 10 μM for SW480 cells. As shown in Figure 6b, treatment with optimal concentrations of LiCl causes a statistically significant decrease in uPA and uPAR expression levels in the three LiCl-treated cell lines compared to untreated cells, of 33 and 52% for MCF-7 cells, 78 and 60% for MDA-MB-231 cells and 56 and 83% for SW480 cells, respectively. As our results suggest that NF-κB cooperates with β-catenin in the regulation of the uPA/uPAR system, we investigated the effect of SN50 on LiCl-treated cells. We observed no significant difference in uPA, uPAR and PAI-1 mRNA expression levels between cells treated with LiCl alone or with LiCl and SN50 (Fig. 6b). Similarly, no variation of uPA protein levels was observed between cells treated with LiCl and cells treated with LiCl and SN50 (Fig. 6c). These results confirm the involvement of NF-κB in the regulation of the uPA/uPAR system by β-catenin.

Figure 6. β-Catenin overexpression by LiCl treatment or β-catenin/Tcf-4 expression vectors transfection downregulates uPA/uPAR system in cancer cells. (a) Effect of serial concentrations of LiCl on β-catenin protein expression in the three cancer cell lines after 6 hr of treatment, evaluated by Western blot, actin being used as a loading control. The optimal LiCl concentrations were of 50 mM for MCF-7 and MDA-MB-231 cell lines and of 10 mM for SW480 cell line. (b) Q-RT-PCR analysis of uPA, uPAR and PAI-1 mRNA expression after LiCl treatment alone or LiCl and SN50. The columns show the mean of relative expression to PPIA housekeeping gene of three independent experiments (mean ± SEM, *p < 0.05). (c) Western blot analysis showed that expression of uPA protein was reduced similarly by LiCl treatment used alone or in the presence of SN50. Actin was used as a loading control. (d) Q-RT-PCR analysis of β-catenin, uPA, uPAR, PAI-1 and TBP mRNA expression after transfection of β-catenin/Tcf-4 expression vectors in MCF-7, MDA-MB-231 and SW480 cell lines. The columns show the mean of relative expression to PPIA housekeeping gene of three independent experiments (mean ± SEM, *p < 0.05).

Expression vectors β-catenin/Tcf-4 transient transfection was used as an alternative approach to overexpress stable β-catenin. Similarly to LiCl treatment, uPA, uPAR and PAI-1 expressions decreased in MCF-7, MDA-MB-231 and SW480 β-catenin/Tcf-4-transfected cells (Fig. 6c).

Discussion

The functional importance of β-catenin signaling during tumorigenesis has mainly been attributed to the promotion of tumor cell progression28 by stimulating expression of several target genes involved in cancer proliferation (c-myc and cyclin D1),15, 16 invasion (MMP-7 and MT1-MMP)17, 29 and angiogenesis (VEGF).18 Using three cancer cell models (breast and colon cancer cell lines), we demonstrate in our study that inhibition of β-catenin leads to an upregulation of the uPA/uPAR system through NF-κB cooperation, promoting cancer cell invasion. As these data were observed in three different cancer models, this suggests a more general mechanism and not one limited to a specific cell line.

uPA has previously been reported to be a β-catenin target gene. Indeed, β-catenin transactivates human uPA promoter/enhancer directly via Tcf-4 and TBE and works in synergy with other transcription factors, namely AP1 and ets-1.8 β-catenin downregulation in colon cancer cell lines (SW480 and DLD-1 cells) partially decreases uPA mRNA expression. Taken together, these results indicate that β-catenin plays an important but not exclusive role in the regulation of the human uPA gene, involving several transcription factors. More recently, it was reported that β-catenin/Tcf-4 complex directly binds to the uPAR promoter, enhancing its activity in cooperation with two Sp motifs.19 Our results demonstrated that the inhibition of β-catenin using a siRNA strategy significantly upregulated uPA, uPAR and PAI-1 expression at the mRNA and protein levels in two human breast cancer models, MCF-7 and MDA-MB-231 cells, as well as in the colon cancer cell line, SW480. Furthermore, the inhibition of β-catenin expression promoted the invasive potential of cancer cells through the upregulation of uPA expression. Beside these observed results, β-catenin inhibition downregulated the expression of two well-known β-catenin target genes, cyclin D1 and c-myc, in the studied models. These unexpected results are in accordance with those of Takahashi et al.30 reporting that knockdown of β-catenin expression with shRNA in B16 melanoma cancer cells enhances in vitro cell migration potential and reduces cadherin protein expression. B16 cells stably transfected with sh-β-catenin, subcutaneously inoculated in mice, form more lung nodules compared to control cells. Moreover, Liu et al.31 have shown that knockdown of p120-catenin by siRNA reduces E-cadherin protein expression and β-catenin mRNA and protein expressions, promoting proliferation and invasive capacity of lung cancer cells in vitro and in vivo. These results raise concerns about the effect of gene silencing of β-catenin on cell migration, invasion and tumor growth. To further understand the mechanism of upregulation of the uPA/uPAR system by β-catenin, we investigated the involvement of NF-κB. Indeed, NF-κB plays an important role in cancer progression and metastasis and has been reported to physically interact with β-catenin, resulting in a reduction in NF-κB nucleus translocation, DNA binding and target gene expression.25 Furthermore, NF-κB-responsive elements are present in uPA,21uPAR22 and PAI-123 promoters. The treatment of β-catenin siRNA-transfected cells with a specific inhibitor of NF-κB (SN50) significantly reduced enhancement of uPA, uPAR and PAI-1 mRNA expression and cancer cell invasion, observed with β-catenin siRNA transfection alone. Furthermore, β-catenin siRNA-treated cells exhibited an increase of NF-κB staining in the nucleus. These data suggest that β-catenin regulates the uPA/uPAR system in cooperation with the NF-κB transcription factor, which constitutes a novel regulatory mechanism.

When β-catenin is not associated with cadherin at epithelial cell adhesion junctions, the multiprotein complex consisting of APC, Axin and GSK-3β phosphorylates β-catenin, which is degraded by the proteasome system. Inactivation of GSK-3β, particularly by LiCl treatment,26, 27 stabilizes β-catenin that then accumulates in the cytosol. Subsequently, β-catenin translocates to the nucleus and interacts with Tcf-4 to induce his target genes. In our study, we observed that LiCl treatment of breast and colon cancer cell lines led to a decrease in uPA and uPAR expression at the mRNA and protein levels. Similar results were observed in β-catenin/Tcf-4 expression vectors-transfected cells. These data support our findings of an increase in the uPA/uPAR system after β-catenin downregulation. Furthermore, the NF-κB inhibitor had no effect on uPA/uPAR expression after LiCl treatment of cancer cells, as can be expected in the presence of an excess of cellular β-catenin. This supports the idea that uPA/uPAR regulation by β-catenin is dependent on the NF-κB pathway. A relationship between β-catenin and NF-κB has been suggested by a study showing that LiCl treatment increases β-catenin cellular levels leading to the inhibition of the expression of Traf1, a NF-κB target gene, in colon RKO and breast MDA-MB-231 cancer cells.32 More recently, Du et al.33 have shown that β-catenin regulates TNFα-induced iNOS and Traf1 expression, both NF-κB target genes, by physical interaction with NF-κB in colon and liver cancer cells. Indeed, SW480 colon cancer cells stably transformed with wild-type APC have a decrease in β-catenin protein and an increase in TNFα-induced NF-κB DNA binding as well as iNOS and Traf1 expression. The authors suggest that inactivation of β-catenin restores NF-κB activity. In contrast, the overexpression and the stabilization of β-catenin produces the opposite effects by inhibiting NF-κB pathway signaling leading to a decrease in iNOS and Traf1 expression. These findings underscore the complex role of β-catenin, NF-κB and their target genes in tumor progression.

Our results strengthen the emerging view that β-catenin can exert different effects on tumor cells. Our results showed a novel crosstalk between β-catenin and uPA/uPAR system through NF-κB cooperation in breast and colon cancer cells. These findings raise serious concerns for the use of β-catenin inhibition as an anticancer therapy strategy because of the complex downstream responses.

Acknowledgements

The authors thank Ms. Christelle Doliger and Mr. Niclas Setterblad at the Imagery Department of the Institut Universitaire d'Hématologie-IFR105 for the confocal microscopy. The authors thank Ms. Marie-Pierre Podgorniak for technical assistance and Dr. Michele Goodhardt for helpful discussions.