sign in sign up
<<
Home , GLI2, GLI3

ANTICANCER RESEARCH 28 : 731-740 (2008)

Novel Link between Estrogen α and Hedgehog Pathway in Breast Cancer KENICHIRO KOGA 1, MASAFUMI NAKAMURA 1, HIROSHI NAKASHIMA 1, TAKASHI AKIYOSHI 1, MAKOTO KUBO 2, NORIHIRO SATO 1, SYOJI KUROKI 2, MASATOSHI NOMURA 3, MASAO TANAKA 2 and MITSUO KATANO 1

Departments of 1Cancer Therapy and Research, 2Surgery and Oncology, 3Medicine and Bioregulatory Science, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan

Abstract. Ligand-dependent constitutive activation of the signaling may represent a new therapeutic target for hedgehog (Hh) pathway is important in the development of carcinomas (12). In mouse breast tissue, the Hh pathway various carcinomas including breast cancer. A link between plays a critical role during ductal development (13), and α (ERα) and the Hh pathway in human disruption of Patched1 , a Hh ligand receptor and suppressor breast cancer is shown here for the first time. In ERα- of the Hh pathway, results in severe defects in ductal positive cells, estrogen depletion decreased the expression of such as ductal hyperplasia, that closely (Shh), a ligand of the Hh pathway, while resembles some hyperplasia in (14). Consistent estrogen supplementation triggered Shh up-regulation. This with these data, Hh signaling is activated in human breast estrogen-induced Shh expression activated the Hh pathway cancer tissues, and inhibition of the signaling activity in a ligand-dependent manner, and increased cell attenuates tumor cell proliferation (15). Thus, it is important proliferation. These effects were suppressed by ERα to elucidate the molecules that regulate Hh pathway inhibitors, including ICI 182,780 (ICI), the dominant activation in breast cancer; however, the detailed negative form of ERα and small interfering RNA (siRNA) mechanisms that underlie Hh pathway activation in breast against ERα. Consistent with the in vitro data, a positive cancer remain unclear. correlation between ERα and Shh expression was found in Estrogen is a crucial factor for physiological proliferation breast cancer tissues. These data suggest that ERα regulates and differentiation of the normal mammary gland. On the the Hh pathway through Shh induction, and promotes breast other hand, estrogen is also considered as a stimulant for cancer development. initiation and promotion of breast tumors. The biological effects of estrogen are mediated through two distinct The hedgehog (Hh) signaling pathway is a key mediator of intracellular receptors, estrogen receptor α (ERα) and ERβ, many fundamental processes in (1), which belong to the superfamily of nuclear receptors that and Hh signaling is crucial in some adult organs for stem share a similar structure and mode of action. ERα plays cell maintenance and tissue repair (2, 3). The Hh pathway important roles in the proliferation of ERα-positive breast has also been found to be constitutively activated in many cancer cells (16, 17), and ERα status is essential information human cancer tissues (4-7), while Hh pathway activation in making decisions for endocrine therapy (18). increases proliferation of various tumor cells in a ligand- Thus, it has been generally considered that both Hh dependent manner (5, 8-11). Thus, the factors which lead pathway activation and ERα activation participate in the to overexpression of the Hh ligand and activation of Hh and growth of breast tumors. There are two representative mechanisms of Hh pathway activation, ligand-dependent and ligand-independent activation (1). Since aberrant Hh-ligand production has been seen in Correspondence to: Prof. Mitsuo Katano, Department of Cancer several carcinomas including breast cancer (8, 9, 15, 19), Therapy and Research, Graduate School of Medical Sciences, and of Hh pathway components has been Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812- identified at low frequency in breast cancer (20-22), it is 8582, Japan. Tel: +81 92 642 6941, Fax: +81 92 642 6221, e-mail: [email protected] proposed that Hh pathway activation is mainly caused in a ligand-dependent manner in breast cancer. Thus, in the Key Words: Breast cancer, estrogen receptor α, hedgehog signaling present study we focused on the mechanism of up- pathway, proliferation, sonic hedgehog. regulation of sonic hedgehog (Shh), a ligand of the Hh

0250-7005/2008 $2.00+.40 731 ANTICANCER RESEARCH 28 : 731-740 (2008) pathway in several carcinomas (8). Our previous study has Real-time reverse -PCR. The total RNA was extracted also indicated a positive correlation between ERα status and by RNeasy mini kit (Qiagen, Valencia, CA, USA) and quantified by Gli1 nuclear translocation, which is a marker of Hh pathway spectrophotometry (Ultrospec 2100 Pro; Amersham Pharmacia Biotech, Cambridge, UK). The RNA (700 ng) was treated with activation, suggesting a link between ERα status and Hh DNase, and reverse transcribed to cDNA with the Quantitect pathway activation (15). To prove this potential link, we Reverse Transcription Kit (Qiagen) according to the manufacturer’s selected two breast cancer cell lines, ERα-positive MCF-7 protocol. The reactions were run with SYBR Premix Ex Taq (Takara cells and ERα-negative MDA-MB-231 cells, which are Bio. Inc., Otsu, Japan) on a DNA Engine Opticon 2 System (MJ representative cell lines and well-characterized at the Research, Waltham, MA, USA). The pShh-GFP or pGli1 were molecular level were selected. serially diluted in 10-fold increments and amplified with the primer pairs to generate a standard curve for Shh or Gli1 . Standard curves Materials and Methods for pS2 and β-actin were generated using cDNA from MCF-7 cells treated with E2 for 16 h. Each sample was run in triplicate. All the Cell culture, reagents, and antibodies. The MCF-7 cells and primer sets amplified fragments less than 200 bp long. Sequences MDA-MB-231cells were purchased from American Type Culture of the primers used were β-actin , forward, 5’-TTG CCG ACA GGA Collection and were maintained at 37˚C under a humidified TGC AGA AGG A-3’, reverse, 5’-AGG TGG ACA GCG AGG CCA GGA T-3’; Shh , forward, 5’-GTG TAC TAC GAG TCC AAG GCA atmosphere of 5% CO 2 and 95% air in RPMI1640 medium (Life Technologies, Grand Island, NY, USA) supplemented with 10% C-3’, reverse, 5’-AGG AAG TCG CTG TAG AGC AGC-3’; Gli1 , fetal bovine serum (FBS, Life Technologies) and antibiotics (100 forward, 5’-GGT TCA AGA GCC TGG GCT GTG T-3’, reverse, units/mL penicillin and 100 μg/mL streptomycin, Meijiseika, 5’-GGC AGC ATT CTC AGT GAT GCT G-3’; ERα, forward, 5’- Tokyo, Japan). The cells were incubated for 24 h in phenol-red GGA GGG CAG GGG TGA A-3’, reverse, 5’-GGC CAG GCT free minimum essential medium (MEM; Invitrogen, Carlsbad, CA, GTT CTT CTT AG-3’and pS2 , forward, 5’-CAT GGA CGT CCC USA) without FBS prior to all experiments (termed estrogen TCC AGA AGA G-3’, reverse, 5’-CTC TGG GAC TAA TCA CCG starvation). Thereafter, cells were principally cultured in MEM TGC TG-3’. The amount of each target in a given sample was supplemented with 5% dextran-coated charcoal-treated FBS normalized to the level of β-actin in that sample. (DCC-FBS-MEM) in a humidified 95% air and 5% CO 2 atmosphere. Stock solutions of 17-β-estradiol (E2) purchased Immunoblotting. Whole-cell extraction was performed with M- from Sigma (Deisenhofen, Germany), ICI 182,780 (ICI), a pure PER Reagents (Pierce Biotechnology, Rockford, IL, USA) anti-estrogen from Tocris Cookson Ltd. (Ellisville, MO, USA) and according to the manufacturer’s instructions. The cyclopamine (Cyc), a Hh pathway inhibitor from Toronto concentration was determined with Bio-Rad Protein Assay (Bio- Research Chemicals (North York, Ontario, Canada) diluted in Rad, Hercules, CA, USA), and the whole-cell extract (100 μg) 99.5% ethanol were stored at –30˚C. The recombinant human Shh was separated by electrophoresis on SDS-polyacrylamide gel, and N-terminal peptide (rhShh) and the rat anti-Shh NH 2-terminal transferred to Protran nitrocellulose membranes (Dassel, peptide antibody (αShh-Ab) were purchased from R&D Systems Germany). The blots were then incubated with anti-Shh (1:100), (Minneapolis, MN, USA). The anti-GAPDH (sc-25778), anti-Shh Gli1 (1:200), GAPDH (1:1000), ERα (1:500), or phospho ERα (sc-9024), anti-Gli1 (sc-6153), and anti-ER α (sc-8002) antibodies (1:500) primary antibody overnight at 4˚C followed by incubation were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, in horse radish peroxidase (HRP)-linked secondary antibody USA) and the mouse anti-phospho ER α (Ser118) from Cell (Amersham Biosciences, Piscataway, NJ, USA) at room Signaling Technology, Inc. (Danvers, MA, USA). The expression temperature for 1 h. The immunocomplexes were detected with the level of Gli1 , a transcriptional factor and one of the target of ECL plus Western Blotting Detection System (Amersham the Hh pathway, was used as a marker of Hh signaling activity Biosciences) and visualized with a Molecular Imager FX (Bio- (23, 24). It has also been demonstrated that phosphorylation of Rad). GAPDH was used for protein loading control. The Shh ER α protein at Ser 118 is crucial for the transcriptional activity of protein level was quantified with the 19 kDa fragment, as it is the ER α in the presence of estrogen (25). Therefore phosphorylated physiologic active form of Shh (31). ER α protein at Ser 118 (P-ER α) expression and mRNA expression of pS2 , a primary target gene of ERα (26) were Dual luciferase assay. The MCF-7 cells in 24-well plates were considered markers of ERα activation in this study. transfected with plasmids with TransFast transfection reagent according to the manufacturer’s instructions. The cells in each well Expression vectors. cDNAs encoding human ERα or dominant were co-transfected with 10 ng of pRL-SV40 (Promega) and 1 μg negative human ERα (S554fs) were cloned into the pSG5 ERE-tk-Luc. After estrogen starvation, E2 and ICI or Cyc were expression vector as described previously (27, 28). These cDNAs added to each well, and luciferase assays were performed 6 h later and ERE-tk-luc, a consensus estrogen response element upstream with the dual luciferase assay kit (Promega) according to the of luciferase, were kindly provided by Dr Norio Wake (Kyushu manufacturer’s instructions. The luciferase activities were University, Fukuoka, Japan). The pRL-SV40 was purchased from normalized to the Renilla luciferase activity. Promega (Madison, WI, USA). The pIRES2-hShh-EGFP (pShh- GFP) and pcDNA3.1/His-hGli1 (pGli1) were kindly provided by Immunostaining of cell line. The MCF-7 cells (2 x10 4/well) were Dr. Aubie Shaw (Division of Urology, Department of Surgery, seeded onto pre-underlaid poly-L-lysine coated cover glass University of Wisconsin, Madison, WI, USA) (29) and Dr. H. (Asahi Techno Glass Corporation, Chiba, Japan) in 24-well Sasaki (Center for Developmental Biology, Riken, Kobe, Japan) plates, and were incubated overnight in 10% FBS-RPMI. After (30), respectively. estrogen starvation, the cells were treated with control vehicle

732 Koga et al : Cross-talk OF ERα and Hh Pathway IN Breast Cancer

(0.1% ethanol), or indicated doses of E2, ICI, or Cyc for 24 h in 5% DCC-FBS-MEM. The cells were fixed in 4% paraformaldehyde followed by permeabilization with 0.2% Triton X-100, and then incubated with primary followed by secondary antibodies. The cells were counterstained with 4’,6- diamidino-2-phenylindole (DAPI, Sigma-Aldrich Corp., St. Louis, MO, USA). After mounting in Vectorshield Mounting Medium (Vector Laboratories, Burlingame, CA, USA), the samples were examined by fluorescence microscopy (BX 50; Olympus Corp., Tokyo, Japan). The exposure time for recording was manually fixed at 2 msec, 1.8 msec, and 0.2 msec for Shh, P-ERα, and DAPI, respectively. The antibodies and dilutions used were as follows: rabbit anti-Shh (1:200), mouse anti- phospho ERα (1:100), AlexaFluor 488 goat anti-rabbit IgG and AlexaFluor 594 goat anti-mouse IgG (1:400; Molecular Probes, Eugene, OR, USA).

Transient cell transfection. The MCF-7 cells (1x10 6/well) seeded in a 25 cm 2 flask were transfected with 9 μg of dominant negative ERα (DNER) or pSG5 plasmid control, with TransFast Reagent (Promega). The transfected cells were incubated in 10% FBS-RPMI for 16 h. After estrogen starvation, the transfected cells were treated with the indicated reagents in 5% DCC-FBS-MEM for 24 h, and Figure 1. ERα and Hh pathway activation in estrogen-depleted medium. (A) then used for immunoblotting. pS2 mRNA and P-ERα protein expression, and (B) Gli1 mRNA and Gli1 protein expression by MCF-7 cells in indicated medium. MCF-7 cells were cultured in 5% FBS-MEM (FBS) or 5% DCC-FBS-MEM (DCC-FBS) Small interfering RNA (siRNA) against ERα. The MCF-7 cells 6 (estrogen depletion) for 16 h to measure pS2 or Gli1 mRNA expression, or (2.0x10 cells) were transfected with siRNA (100 nM) against for 24 h to measure P-ERα or Gli1 protein expression. Relative pS2 or Gli1 ERα by nucleofection with NucleoFector II (Amaxa GmbH, mRNA expression is shown after normalization to the corresponding β-actin Koeln, Germany) as per the manufacturer’s instructions, and mRNA expression. Results are expressed as mean±S.D. *p<0.05; **p<0.01. 2 plated in a 25 cm -flask for 12 h in 10% FBS-RPMI. After GAPDH is shown as protein loading control. estrogen starvation, the cells were treated with E2 for 24 h, and then used for immunoblotting. The Validate stealth RNAi against ERα, and the stealth RNAi-negative control (Invitrogen) were used. Results Proliferation assay. The cells (2 x10 4/well) seeded in 48-well plates in complete culture medium were incubated overnight. Hh pathway activation. Hh activity with or without estrogen After estrogen starvation the medium was changed to 5% DCC- in the breast cancer cell lines was studied. Minimum essential FBS-MEM containing the indicated doses of reagents. After 72 h medium supplemented with dextran-coated charcoal-treated incubation, the cells were harvested by trypsinization, and cells were counted with a Coulter counter (Beckman Coulter, fetal bovine serum (DCC-FBS-MEM) was used as the Fullerton, CA, USA). estrogen-depleted medium. When the ERα-positive MCF-7 cells were cultured with DCC-FBS-MEM instead of Clinical samples. Samples from 14 patients with primary breast conventional FBS-MEM, pS2 mRNA and P-ERα expression carcinoma selected on the basis of ERα status who underwent decreased (Figure 1A) and Gli1 mRNA and protein expression resection at the Department of Surgery and Oncology, Kyushu also decreased (Figure 1B). However, the Hh pathway activity University (Fukuoka, Japan), between February 2003 and August did not change in response to estrogen depletion in ERα- 2003 were included. Eight specimens were ERα-positive and 6 were ERα-negative. All 14 patients gave informed consent before surgical negative MDA-MB-231 cells (data not shown). treatment. All the surgical specimens were frozen at –80˚C, examined histopathologically, and classified using the tumor-node- Effect of estrogen on Hh pathway activation. To confirm metastasis (TNM) classification system of the International Union whether E2 could influence Hh pathway activation in the Against Cancer (UICC, Fifth edition, 1997). The total mRNA of MCF-7 cells, they were incubated with 5% DCC-FBS-MEM these specimens was extracted using the RNeasy mini kit (Qiagen) containing E2, which significantly increased both pS2 as per the manufacturer’s recommendation. mRNA expression and ERα responsive reporter activity Statistical analysis. The Student’s t-test was used for statistical (Figure 2A). Gli1 mRNA expression in the MCF-7 cells was analysis unless otherwise indicated. All the calculations were carried also increased, in sharp contrast to the MDA-MB-231 cells out with StatView 5.0 J software (Abacus Concepts, Berkeley, CA, (Figure 2B). The E2-induced activation of ERα and the Hh USA). P-values less than 0.05 were considered significant. pathway were almost completely inhibited by ICI (Figure

733 ANTICANCER RESEARCH 28 : 731-740 (2008)

Figure 3 . Ligand-dependent Hh pathway activation in MCF-7 cells. (A) Figure 2. ERα and Hh pathway activation by treatment with E2 alone or Gli1 mRNA expression in MCF-7 cells treated with or without in combination with ICI 182,780 (ICI) or cyclopamine (Cyc) in breast recombinant human Shh N-terminal peptide (rhShh) for 8 h. rhShh was cancer cells. (A, left) Relative pS2 mRNA expression in MCF-7 cells resolved in PBS containing 0.1% bovine serum albumin (BSA) as per treated for 16 h and examined by real-time RT-PCR. (A, right) Dual manufacturer’s recommendation. Control contained the same amount of luciferase assay performed 6 h after treatment with indicated reagents. BSA. (B) Gli1 mRNA expression in MCF-7 cells treated for 16 h with (B) Relative Gli1 mRNA expression in MCF-7 or MDA-MB-231 cells anti-Shh antibody (αShh-Ab, 15 μg/ml) or isotype-matched control IgG after treatment with indicated reagents for 16 h. Results are expressed (Control IgG, 15 μg/ml) in the presence of E2. Results are expressed as as mean±S.D. *p<0.05; **p<0.01; n.s., not significant. mean±S.D. *p<0.05.

2A, B). Co-incubation with Cyc completely prevented the The correlation between E2 stimulation and Shh production E2-induced Gli1 mRNA elevation in the MCF-7 cells was then examined. When the MCF-7 cells were cultured in (Figure 2B), but did not affect the pS2 mRNA expression DCC-FBS-MEM instead of FBS-MEM, Shh mRNA and ERα transcriptional activity (Figure 2A). expression was significantly decreased, as was Shh and P-ERα Shh expression. The ligand dependency of Hh pathway protein expression (Figure 4A). Supplementation with E2 in activation in human breast cancer was confirmed by DCC-FBS-MEM significantly increased Shh mRNA culturing the MCF-7 cells in 5% DCC-FBS-MEM expression in the MCF-7 cells, in contrast to the MDA-MB- supplemented with or without recombinant human Shh 231 cells (Figure 4B). The elevation of Shh mRNA expression peptide (rhShh). Incubation with rhShh markedly increased in the MCF-7 cells induced by E2 was completely decreased the Gli1 mRNA expression compared to the rhShh-free to control levels by co-incubation with ICI (Figure 4B), while cultures (Figure 3A), suggesting that Hh activation is ligand- there was no effect with the addition of Cyc, on the increased dependent in MCF-7 cells. Next, the ligand-dependence of Shh mRNA expression. The E2-induced elevation of Shh E2-induced Hh pathway activation was demonstrated by expression and inhibition by ICI were also detected at the culturing the MCF-7 cells in 5% DCC-FBS-MEM with E2 protein level in the MCF-7 cells, but not in the MDA-MB-231 in the presence or absence of antibodies directed against Shh cells (Figure 4C). E2-induced elevation of Shh expression was (αShh-Ab). While the Gli1 mRNA expression significantly also visualized by immunofluorescence staining in MCF-7 increased with E2, the rise in Gli1 mRNA was significantly cells (Figure 4D). After E2 stimulation, Shh protein staining less in the presence of αShh-Ab (Figure 3B). was higher than that seen in control cells consistent with the

734 Koga et al : Cross-talk OF ERα and Hh Pathway IN Breast Cancer

Figure 4. Shh expression in MCF-7 and MDA-MB-231 cells. (A) Shh mRNA or Shh and P-ERα protein expression in MCF-7 cells cultured in 5% FBS-MEM (FBS) or 5% DCC-FBS-MEM (DCC-FBS) (estrogen-depleted) medium. The cells were cultured for 16 h to measure Shh mRNA expression or for 24 h to detect Shh and P-ERα protein expression. (B) Shh mRNA expression in MCF-7 cells or MDA-MB-231 cells treated with indicated reagents for 16 h. (C) Shh protein expression in MCF-7 cells or MDA-MB-231 cells treated with indicated reagents for 24 h. (D) Immunofluorescent staining of MCF-7 cells treated with indicated reagents for 24 h. Magnification, 400x; Bars, 20 μm. (E) Shh or P-ERα protein expression in MCF-7 cells transfected with dominant negative ERα (DNER) or control plasmid (pSG5) in the presence or absence of E2. After transfection with DNER or pSG5, MCF-7 cells were treated with E2 for 24 h. (F) ERα mRNA expression in MCF-7 cells transfected with ERα-siRNA (#1 or #2) or control siRNA (Cont.) at 100 nM (F, left). ERα protein expression in MCF-7 cells was confirmed by immunoblotting (F, left inset). Shh or P-ERα protein expression in MCF-7 cells transfected with ERα-siRNAs or control siRNA and treated with or without E2 for 24 h (F, right). Results are expressed as mean±S.D. *p<0.05; **p<0.01; n.s., not significant. GAPDH was shown as protein loading control.

735 ANTICANCER RESEARCH 28 : 731-740 (2008)

Figure 5. ERα and Hh pathway activation and cell proliferation in MCF-7 cells. (A) Cell proliferation in the presence of E2 alone (left panel) or in combination with ICI (middle panel) or Cyc (right panel) in MCF-7 cells. (B) Cell proliferation following incubation with recombinant human Shh N-terminal peptide (rhShh) in MCF-7 cells (left panel). Each control contained same amount of bovine serum albumin (BSA). Cell proliferation with anti-Shh antibody (αShh-Ab, 15 μg/ml) or isotype-matched control IgG (Control IgG, 15 μg/ml) in the presence of E2 in MCF-7 cells (right panel). αShh-Ab or control IgG was added to medium at the same time as E2. Results are expressed as mean±S.D. *p<0.05; n.s., not significant.

observation that nuclear P-ERα staining (Red) showed a proliferation of MCF-7 cells in a dose-dependent manner higher signal after E2 stimulation than in the control cells. (Figure 5B, left). Thus, these data suggested that E2-induced These E2-induced elevations of P-ERα and Shh protein Shh production was biologically significant, and activated intensity were suppressed to control level by ICI. MCF-7 cell proliferation in a ligand-dependent manner. To further confirm that activated ERα increased Shh expression, the Shh protein levels in the dominant-negative ERα status correlation with Shh mRNA expression in breast form of ERα (DNER)-transfected MCF-7 cells was measured. cancer tissues. Real-time RT-PCR analysis was used for the The Shh protein level in the DNER-transfected MCF-7 cells quantitative analysis of Shh expression in breast carcinoma was low compared to that of control cells in the presence of E2 tissue specimens. There was no significant correlation (Figure 4E), suggesting that the E2-induced Shh production between the ERα status and the major clinicopathological was a consequence of ERα activation. These results were parameters in these specimens (Table I). The level of Shh supported by the following data using small interfering RNA mRNA expression in ERα-positive tissues was significantly against ERα (ERα-siRNA). Knockdown of ERα was higher than that in ERα-negative tissues (Figure 6). confirmed at both mRNA and protein level (Figure 4F, left and inset). In the presence of E2, the Shh protein level of the ERα- Time-course analysis of pS2 and Shh mRNA expression. The siRNA-transfected MCF-7 cells decreased, in contrast to that MCF-7 cells were treated with E2 and pS2 and Shh mRNA of control the siRNA-transfected cells (Figure 4F, right). expression were analyzed by real-time RT-PCR. Shh mRNA expression significantly increased from 16 h onwards after Effect of Shh induced by estrogen on cell proliferation in E2 treatment (Figure 7A), whereas pS2 mRNA significantly ERα-positive cells. Supplement ation of E2 in 5% DCC- increased from 4 h after E2 treatment (Figure 7B). FBS-MEM increased the proliferation activity of MCF-7 cells in a dose-dependent manner, which was inhibited by Discussion ICI (Figure 5A, left and middle, respectively) and Cyc (Figure 5A, right). The αShh-Ab also inhibited proliferation Here for the first time ERα regulation of Hh pathway of MCF-7 cells in the presence of E2 (Figure 5B, right), activation in an autocrine manner through up-regulation of while consistent with the above data, rhShh increased Shh expression in ERα-positive breast cancer, is reported. In

736 Koga et al : Cross-talk OF ERα and Hh Pathway IN Breast Cancer

Table I. Clinicopathological features in the 14 breast specimens.

No. of specimens p-value

ER α + ve group ER α - ve group

T classification* n.s. † pT1 43 pT2 43 Hist ology n.s. † Intraductal cacinoma 74 Invasive ductal carcinoma 12 PgR status n.s. † Positive 50 Negative 36 LN involvement n.s. † Figure 6. Shh mRNA expression in resected breast cancer tissues. A Negative 31 scatter plot of relative Shh mRNA expression in ERα-negative or ERα- Positive 55 positive breast cancer groups are shown after normalization to the Stage n.s. † corresponding β-actin mRNA expression. Filled circles, mean of I22experiments; bars, SD; open circles, relative Shh mRNA expression of IIA 54 each sample. * p<0.05. IIB 10

Total 8614

Major clinicopathological parameters were comparable in both groups. There were no significant differences in other parameter s, such as age, HER2/neu, nuclear atypia, and vessel invasion (data not shown). *According to the TNM classification system. †Mann-Whitney U-test; n.s., not significant.

the present study, E2 increased ERα activation, Shh expression, and Hh pathway activation only in the ERα- positive MCF-7 cells. Blockade of ERα activation by ICI or transfection with DNER or ERα-siRNA decreased the increase in Shh expression and Hh pathway activation induced by E2 in the ERα-positive MCF-7 cells. The data from surgically resected specimens indicated that this pathway consisting of ERα activation, Shh expression, and Hh pathway activation may be operating even in breast cancer tissue. The data obtained from the two cell lines indicated that ERα is one of the regulators of ligand-dependent Hh pathway activation in breast cancer. Namely, ERα may be contributing to Shh production in ERα-positive breast cancer Figure 7. Time-course analysis of Shh and pS2 mRNA expression in cells. The time-course analysis indicated that activation of MCF-7 cells treated with E2. MCF-7 cells were treated with E2 for indicated number of hours. Relative Shh or pS2 mRNA expression is ERα preceded Shh mRNA expression (Figure 7). shown after normalization to the corresponding β-actin mRNA Furthermore, incubation of MCF-7 cells with cycloheximide, expression as fold change relative to control. Columns, mean of a protein synthesis inhibitor, attenuated the ability of ERα to triplicate experiments; bars, SD. *p<0.05. induce Shh mRNA expression (data not shown). These data suggested that ERα indirectly regulated Shh expression. Our previous data showed that the Hh pathway was Several mechanisms may be operating dependently or constitutively activated not only in MCF-7 cells but also in independently in Hh pathway activation in different cells. For MDA-MB-231 cells (15), thus Hh pathway activation is also example, we have recently reported that nuclear factor κB found even in ERα-negative cells. It is still unclear how the activation increased Shh expression, which activated the Hh Hh pathway is activated in ERα-negative breast cancer cells. pathway in an autocrine manner and enhanced cell

737 ANTICANCER RESEARCH 28 : 731-740 (2008) proliferation in pancreatic cancer (12). The present data 9 Sanchez P, Hernandez AM, Stecca B, Kahler AJ, DeGueme AM, support a causal link between ERα activation and Shh Barrett A, Beyna M, Datta MW, Datta S and Ruiz i Altaba A: expression, even in primary breast cancer tissue. Thus, our Inhibition of prostate cancer proliferation by interference with SONIC HEDGEHOG-GLI1 signaling. Proc Natl Acad Sci U S A findings that ERα links with the Hh pathway appear to be 101 : 12561-12566, 2004. important in the clinical setting, because both ERα and the 10 Sheng T, Li C, Zhang X, Chi S, He N, Chen K, McCormick F, Hh pathway play important roles in the development of Gatalica Z and Xie J: Activation of the hedgehog pathway in breast cancer, and are also useful therapeutic targets. advanced prostate cancer. Mol Cancer 3: 29, 2004. 11 Watkins DN, Berman DM, Burkholder SG, Wang B, Beachy PA and Acknowledgements Baylin SB: Hedgehog signalling within airway epithelial progenitors and in small-cell cancer. Nature 422 : 313-317, 2003. We thank Dr. Kiyoko Kato (Kyushu University, Fukouka, Japan), 12 Nakashima H, Nakamura M, Yamaguchi H, Yamanaka N, Kaori Nomiyama, Takaaki Kanemaru, Nobuhiro Torata, Miyuki Akiyoshi T, Koga K, Yamaguchi K, Tsuneyoshi M, Tanaka M Manabe, and Yoshiko Mikami for their skillful technical assistance. and Katano M: Nuclear factor-kappaB contributes to hedgehog General Scientific Research Grants 18390350, 18659372 and signaling pathway activation through sonic hedgehog induction 18591439, the Ministry of Education, Culture, Sports, and in pancreatic cancer. Cancer Res 66 : 7041-7049, 2006. Technology of Japan supported this work. 13 Lewis MT: Hedgehog signaling in mouse mammary gland development and neoplasia. J Mammary Gland Biol Neoplasia 6: 53-66, 2001. References 14 Lewis MT, Ross S, Strickland PA, Sugnet CW, Jimenez E, Scott MP and Daniel CW: Defects in mouse mammary gland 1 Ingham PW and McMahon AP: Hedgehog signaling in animal development caused by conditional haploinsufficiency of development: paradigms and principles. Genes Dev 15 : 3059- Patched-1. Development 126 : 5181-5193, 1999. 3087, 2001. 15 Kubo M, Nakamura M, Tasaki A, Yamanaka N, Nakashima H, 2 Kusano KF, Pola R, Murayama T, Curry C, Kawamoto A, Nomura M, Kuroki S and Katano M: Hedgehog signaling Iwakura A, Shintani S, Ii M, Asai J, Tkebuchava T, Thorne T, pathway is a new therapeutic target for patients with breast Takenaka H, Aikawa R, Goukassian D, von Samson P, Hamada cancer. Cancer Res 64 : 6071-6074, 2004. H, Yoon YS, Silver M, Eaton E, Ma H, Heyd L, Kearney M, 16 Cho HS, Ng PA and Katzenellenbogen BS: Differential Munger W, Porter JA, Kishore R and Losordo DW: Sonic regulation of by estrogen in estrogen growth- hedgehog myocardial gene therapy: tissue repair through independent and -dependent MCF-7 human breast cancer cell transient reconstitution of embryonic signaling. Nat Med 11 : sublines. Mol Endocrinol 5: 1323-1330, 1991. 1197-1204, 2005. 17 Platet N, Cathiard AM, Gleizes M and Garcia M: Estrogens and 3 Adolphe C, Narang M, Ellis T, Wicking C, Kaur P and their receptors in breast cancer progression: a dual role in cancer Wainwright B: An in vivo comparative study of sonic, desert and proliferation and invasion. Crit Rev Oncol Hematol 51 : 55-67, 2004. Indian hedgehog reveals that hedgehog pathway activity 18 Dowsett M and Ashworth A: New biology of the oestrogen regulates epidermal stem cell homeostasis. Development 131 : receptor. Lancet 362 : 260-262, 2003. 5009-5019, 2004. 19 Oro AE, Higgins KM, Hu Z, Bonifas JM, Epstein EH Jr and 4 Thayer SP, di Magliano MP, Heiser PW, Nielsen CM, Roberts Scott MP: Basal cell carcinomas in mice overexpressing sonic DJ, Lauwers GY, Qi YP, Gysin S, Fernandez-del Castillo C, hedgehog. Science 276 : 817-821, 1997. Yajnik V, Antoniu B, McMahon M, Warshaw AL and Hebrok M: 20 Hatsell S and Frost AR: Hedgehog signaling in mammary gland Hedgehog is an early and late mediator of pancreatic cancer development and breast cancer. J Mammary Gland Biol tumorigenesis. Nature 425 : 851-856, 2003. Neoplasia 12 : 163-173, 2007. 5 Karhadkar SS, Bova GS, Abdallah N, Dhara S, Gardner D, 21 Vorechovsky I, Benediktsson KP and Toftgard R: The Maitra A, Isaacs JT, Berman DM and Beachy PA: Hedgehog patched/hedgehog/ signalling pathway in human signalling in prostate regeneration, neoplasia and metastasis. breast cancer: no evidence for H133Y SHH, PTCH and SMO Nature 431 : 707-712, 2004. mutations. Eur J Cancer 35 : 711-713, 1999. 6 Johnson RL, Rothman AL, Xie J, Goodrich LV, Bare JW, 22 Wicking C, Evans T, Henk B, Hayward N, Simms LA, Bonifas JM, Quinn AG, Myers RM, Cox DR, Epstein EH Jr Chenevix-Trench G, Pietsch T and Wainwright B: No evidence and Scott MP: Human homolog of patched, a candidate gene for the H133Y in SONIC HEDGEHOG in a collection for the basal cell nevus syndrome. Science 272 : 1668-1671, of common tumour types. Oncogene 16 : 1091-1093, 1998. 1996. 23 Lee J, Platt KA, Censullo P and Ruiz i Altaba A: Gli1 is a target 7 Berman DM, Karhadkar SS, Hallahan AR, Pritchard JI, Eberhart of Sonic hedgehog that induces ventral development. CG, Watkins DN, Chen JK, Cooper MK, Taipale J, Olson JM Development 124 : 2537-2552, 1997. and Beachy PA: Medulloblastoma growth inhibition by 24 Bai CB, Stephen D and Joyner AL: All mouse ventral spinal hedgehog pathway blockade. Science 297 : 1559-1561, 2002. cord patterning by hedgehog is Gli dependent and involves an 8 Berman DM, Karhadkar SS, Maitra A, Montes De Oca R, function of Gli3. Dev Cell 6: 103-115, 2004. Gerstenblith MR, Briggs K, Parker AR, Shimada Y, Eshleman 25 Ali S, Metzger D, Bornert JM and Chambon P: Modulation of JR, Watkins DN and Beachy PA: Widespread requirement for transcriptional activation by ligand-dependent phosphorylation Hedgehog ligand stimulation in growth of digestive tract of the human oestrogen receptor A/B region. Embo J 12 : 1153- tumours. Nature 425 : 846-851, 2003. 1160, 1993.

738 Koga et al : Cross-talk OF ERα and Hh Pathway IN Breast Cancer

26 Nunez AM, Berry M, Imler JL and Chambon P: The 5’ flanking 30 Sasaki H, Nishizaki Y, Hui C, Nakafuku M and Kondoh H: region of the pS2 gene contains a complex enhancer region Regulation of Gli2 and Gli3 activities by an amino-terminal responsive to oestrogens, epidermal growth factor, a tumour repression domain: implication of Gli2 and Gli3 as primary promoter (TPA), the c-Ha-ras oncoprotein and the c-jun protein. mediators of Shh signaling. Development 126 : 3915-3924, 1999. Embo J 8: 823-829, 1989. 31 Bumcrot DA, Takada R and McMahon AP: Proteolytic 27 Kato K, Horiuchi S, Takahashi A, Ueoka Y, Arima T, Matsuda processing yields two secreted forms of sonic hedgehog. Mol T, Kato H, Nishida Ji J, Nakabeppu Y and Wake N: Contribution Cell Biol 15 : 2294-2303, 1995. of to oncogenic K-Ras-mediated NIH3T3 cell transformation and its implication for escape from senescence by modulating the pathway. J Biol Chem 277 : 11217-11224, 2002. 28 Green S, Issemann I and Sheer E: A versatile in vivo and in vitro eukaryotic expression vector for protein engineering. Nucleic Acids Res 16 : 369, 1988. 29 Fan L, Pepicelli CV, Dibble CC, Catbagan W, Zarycki JL, Laciak R, Gipp J, Shaw A, Lamm ML, Munoz A, Lipinski R, Thrasher JB and Bushman W: Hedgehog signaling promotes Received October 23, 2007 prostate xenograft tumor growth. Endocrinology 145 : 3961- Revised December 14, 2007 3970, 2004. Accepted February 4, 2008

739