Calcitriol-induced prostate-derived factor ... - Wiley Online Library

7 downloads 18157 Views 280KB Size Report
Aug 18, 2004 - Grant sponsor: Center for International Mobility; Grant sponsor: the. Academy of ..... Asterisk, significantly different from control at p. 0.05.
Int. J. Cancer: 112, 951–958 (2004) © 2004 Wiley-Liss, Inc.

Publication of the International Union Against Cancer

CALCITRIOL-INDUCED PROSTATE-DERIVED FACTOR: AUTOCRINE CONTROL OF PROSTATE CANCER CELL GROWTH Nadja NAZAROVA1,2*, Shengjun QIAO1,3, Olga GOLOVKO1,3, Yan-Ru LOU1,3 and Pentti TUOHIMAA1,4 1 Department of Anatomy, Medical School, University of Tampere, Tampere, Finland 2 Drug Discovery Graduate School, University of Turku, Turku, Finland 3 Tampere Graduate School in Biomedicine and Biotechnology, University of Tampere, Tampere, Finland 4 Department of Clinical Chemistry, Tampere University Hospital, Tampere, Finland Calcitriol (1␣,25-dihydroxycholecalciferol) seems to play an important role in the complex control of prostate cell growth. It inhibits proliferation and induces differentiation and apoptosis in prostate cancer cells. However, the molecular mechanisms of the antiproliferative activity of calcitriol are not completely understood. The expression of prostatederived factor (PDF), a member of the transforming growth factor-␤ (TGF-␤) superfamily, has been shown to be associated with proapoptotic and antimitotic activities. We show that calcitriol induces PDF expression in LNCaP human prostate cancer cells in a concentration- and time-dependent manner. In LNCaP cells, the suppression of cell growth by calcitriol is accompanied by stimulation of PDF mRNA and protein synthesis. Human recombinant PDF inhibits LNCaP cell growth. We do not detect any effect of PDF-specific antibody on the basal growth of LNCaP cells, but this antibody partially reverses the suppression of LNCaP cell growth by calcitriol, suggesting that the effect of calcitriol on cell growth is at least partially mediated by PDF. In PC-3 cells, which are less responsive to the growth-inhibitory effect of calcitriol, it has no effect on PDF expression. We do not detect an effect of recombinant PDF on SMAD phosphorylation in LNCaP cells, but ERK1/2 kinases are transiently phosphorylated in response to PDF, which suggests that in LNCaP cells PDF may exert its action through pathway alternative to the classical TGF-␤ signaling pathway. The present study describes the regulation of PDF, the proapoptotic protein of the TGF-␤ superfamily, by calcitriol in androgen-responsive prostate cancer cells. This is a new link between calcitriol and growth factors of TGF-␤ superfamily in the control of prostate cell growth. © 2004 Wiley-Liss, Inc. Key words: vitamin D; prostate-derived factor; prostate cancer; transforming growth factor-␤; gene expression; cDNA microarray; quantitative RT-PCR

Epidemiologic studies suggest that vitamin D deficiency might be a risk factor in prostate cancer.1–3 For decades, vitamin D metabolites have been known as a potent modulators of cellular proliferation and differentiation, but the molecular mechanisms of their action are still not completely understood. Calcitriol (1␣,25dihydroxycholecalciferol), the potent hormonal form of vitamin D, has a complex and cell-specific effect on the growth and proliferation of a variety of tumor cells, including prostate cancer cells in culture, and has a protective effect against cancer development in vivo.4 The action of calcitriol is mediated by nuclear vitamin D receptor (VDR), ligand-dependent transcription factor, which regulates the expression of a number of genes involved in the control of cell cycle progression, apoptosis, differentiation, invasiveness and angiogenesis. Transforming growth factor-␤ (TGF-␤) superfamily of widely distributed multifunctional cytokines (TGF-␤s, activins, inhibins, bone morphogenic proteins) potently and differentially regulates cell proliferation, adhesion, differentiation and survival and is implicated in the initiation and progression of cancer of different origin.5 Often, growth factors of TGF-␤ superfamily have dual effect in cancer, their functioning being highly dependent on cellular context. Prostate-derived factor (PDF) is a recently isolated member of divergent group within the superfamily. Human

PDF cDNA has been cloned and characterized independently by several groups and it is known as PDF,6 placental bone morphogenic protein (PLAB),7 placental transforming growth factor␤ (PTGF-␤),8,9 macrophage inhibitory cytokine-1 (MIC-1),10 growth/differentiation factor-15 (GDF-15)11,12 and nonsteroidal anti-inflammatory drug-activated gene 1 (NAG-1).13 Similar to the other members of TGF-␤ superfamily, PDF is synthesized in a proform and is processed by furin-like proteases, but it is processed intracellularly and is secreted as an active disulfide-linked homodimer.14 PDF is widely distributed in human body, but compared to other members of the family, its expression is more tissue-specific, with the highest expression level in the placenta and the prostate and also in macrophages.6,8,11 Similar to the other TGF-␤ family growth factors, PDF exerts various biologic effects in an autocrine and paracrine manner and its effects are highly dependent on the cellular context. PDF was reported to have potent proapoptotic, prodifferentiative and antiproliferative properties toward different cell types,15–18 but some other studies demonstrated its stimulatory effect on cancer cell invasiveness and growth.19 PDF is expressed in human prostate with higher expression level in the epithelial cells.6,14,20 It is also expressed in various human prostate cancer cell lines, being most highly expressed in LNCaP and PC-3 cells, which were shown to secrete high amount of biologically active PDF.14,18,21 It was shown that PDF is expressed under the androgen control in vivo6 and its expression is modulated both by androgens and estrogens in LNCaP cells in vitro.18 Because of its marked antimitotic, prodifferentiative and proapoptotic properties, regulation by androgens and high expression in the prostate, it was suggested that PDF might be an important regulator of prostate cancer cell proliferation. Crosstalk between calcitriol and TGF-␤ and regulation of TGF-␤ signaling by calcitriol have been reported previously.22–24 Calcitriol enhances the expression of TGF-␤1 and its latent form binding protein in breast carcinoma cells22 and regulates SMADmediated signaling.23,24 Here we show that in human prostate cancer LNCaP cells, calcitriol induces expression of PDF, which plays a role in the complex control of prostate cell growth. We suggest that the induction of PDF may partially mediate various biologic responses of calcitriol in prostate cancer cells.

Grant sponsor: Center for International Mobility; Grant sponsor: the Academy of Finland; Grant sponsor: the University of Tampere. *Correspondence to: Department of Anatomy, Medical School, FIN33014 University of Tampere, 33014 Tampere, Finland. Fax: ⫹358-3-2156170. E-mail: nadya.nazarova@uta.fi Received 23 February 2004; Accepted 11 June 2004 DOI 10.1002/ijc.20510 Published online 18 August 2004 in Wiley InterScience (www. interscience.wiley.com).

952

NAZAROVA ET AL.

MATERIAL AND METHODS

Cells Human prostate cancer LNCaP and PC-3 cell lines were obtained from American Type Culture Collection (Rockville, MD). Chemicals RPMI-1640 and DMEM/F12 media, 5␣-dihydrotestosterone, pifithrin-␤, actinomycin D and sodium orthovanadate were obtained from Sigma-Aldrich Laboratories (St. Quentin Fallavier, France). Fetal bovine serum was obtained from Gibco-BRL (Life Technology, Paisley, Scotland) and COT-1 DNA, PolyA and yeast tRNA were purchased from Gibco-BRL (Grand Island, NY). Calcitriol was a gift from Leo Pharmaceuticals (Ballerup, Denmark) and Casodex was donated from AstraZeneca (London, U.K.). Recombinant human PDF (GDF-15/MIC-1) and human TGF-␤1 were purchased from R&D Systems (Minneapolis, MN). Cy3dUTP, Cy5-dUTP, dNTP and Oligo(dT)(12–18) primers were from Amersham Parmacia Biotech (Piscataway, NJ). Oligonucleotide primers for RT-PCR were ordered from TAG Copenhagen (Copenhagen, Denmark). TRIzol reagent was purchased from Invitrogen (Carlsbad, CA). High-capacity cDNA archive kit and SYBR Green PCR Master Mix were purchased from Applied Biosystems (Forster City, CA). M-PER mammalian protein extraction reagent and BCA protein assay kit were purchased from Pierce (Rockford, IL). Complete Mini Protease inhibitor was obtained from Roche (Mannheim, Germany). Protran nitrocellulose transfer membrane was obtained from Schleider and Schuell (Dassel, Germany). PDF-specific (anti-NAG-1/PTGF-␤) rabbit polyclonal antibody was purchased from Upstate Biotechnology (Lake Placid, NY). Antiphospho-SMAD1/5/8 and antiphospho-SMAD2 rabbit polyclonal antibody were purchased from Cell Signaling Technology (Beverly, MA). Anti-ERK1/2 and antiphosphoERK1/2 rabbit polyclonal antibody were obtained from New England BioLabs (Beverly, MA). Mouse monoclonal anti-␤-actin antibody was obtained from Sigma (St. Louis, MO). Anti-p53 antibody (DO-7) mouse monoclonal was purchased from Novocastra Laboratories (Newcastle upon Tyne, U.K.). Protein molecular weight standards were obtained from Bio-Rad (Richmond, CA). Bovine serum albumin was from Boehringer Mannheim (Mannheim, Germany). Cell culture LNCaP cells were routinely cultured in phenol-red-free RPMI1640 medium supplemented with 10% fetal bovine serum (FBS), 100 ␮g/ml streptomycin and 100 U/ml penicillin. PC-3 cells were cultured in phenol-red DMEM/F12 medium supplemented with 5% FBS. Cells were cultured at 37°C in a humidified atmosphere of 5% CO2. Cell treatment and RNA isolation Cells were plated in 25 cm2 flasks and grown to 70% confluency. Twenty-four hours before treatment, the medium was replaced with one containing 5% dextran-coated charcoal-treated fetal bovine serum or 1% bovine serum albumin (BSA). Cells were treated with calcitriol and other reagents at indicated concentrations for 4 – 48 hr. In all the samples, the concentration of vehicle (ethanol) was adjusted to 0.1% except for RNA stability assay, where it was adjusted to 0.2%. For RNA stability, assay cells were pretreated with 10 nM calcitriol or vehicle before addition of 5 ␮g/ml actinomycin D for 8 hr. Total RNA was isolated using TRIzol reagent (Life Technologies) according to the manufacturer’s instructions. In brief, cells were lysed in 1 ml TRIzol reagent, phases were separated by centrifugation and total RNA was precipitated from the RNAcontaining phase with isopropanol. The pellets were washed once with 75% ethanol, dried in air and diluted in DEPC-treated water. RNA concentration and purity were assessed by A260/A280 absorption using GeneQuant II RNA/DNA Calculator (Pharmacia Biotech, Uppsala, Sweden) and RNA quality was assessed by agarose

gel electrophoresis. The total RNA was immediately reversed to cDNA or stored at ⫺70°C for not longer than 1 month. cDNA microarray and data analysis cDNA microarray was performed according to the manufacturer’s instructions. In brief, 20 ␮g of the total RNA were labeled with Cy5-dUTP by reverse transcription under Oligo(dT)(12–18) primer direction. The RNA labeling was performed at 42°C for 80 min. After the labeling reaction, cDNA was separated from RNA by the addition of a small amount of NaOH solution (1 M) followed by neutralization with Tris-HCl (1 M, pH 7.5). Cy3cDNA and Cy5-cDNA were combined in a Microcon Column (Millipor, Bedford, MA) and washed 4 times in TE buffer (pH 7.4) by centrifugation. In the final washing step, COT-1 DNA, PolyA and yeast tRNA were added to the washing buffer and centrifuged to make the final volume of the labeled cDNA mixture less than 10 ␮l. Glass chip containing 12,000 cDNA probes (Biomedicum Biochip Center, Helsinki, Finland) was pretreated with 0.1% SDS, sterile H2O and 95% ethanol and air-dried. The labeled cDNA mixture was hybridized with the chip in a humid chamber at 65°C overnight. After hybridization, the chip was washed 4 times in sterile water with slight shaking and spun-dry by centrifugation. The chip was scanned in ScanArray 4000 Series (Packard BioScience, Billerica, MA), the hybridization images were analyzed using QuantArray Microarray Analysis Software (Packard BioScience) and the data were finally normalized to median using Excel Data Normalization Macro. Microarray analysis was performed in duplicates and the data are presented as the average of 2 independent experiments. Quantitative RT-PCR Reverse transcription was performed in GeneAmp PCR System 2400 (Perkin Elmer, Oak Brook, IL) using high-capacity cDNA archive kit (Applied Biosystems); 8 ␮g of total RNA were taken in one RT reaction. Reaction conditions were as follows: MultiScribe reverse transcriptase activation at 25°C for 10 min, reverse transcription at 37°C for 2 hr and enzyme inactivation at 95°C for 5 min; cDNA was stored at ⫺20°C. Specific primers for real-time RT-PCR were designed for each gene of interest using PrimerExpress software (Applied Biosystems). Primers were as follows: PDF (NM 004864) forward 5⬘-CCCGGGACCCTAGAGTT-3⬘, reverse 5⬘-CAGGTCCTCGTAGCGTTTCC-3⬘; RPLPO (NM 053275) forward 5⬘-AATCTCCAGGGGCACCATT-3⬘, reverse 5⬘-CGCTGGCTCCCACTTTGT-3⬘. Quantitative RT-PCR was performed in ABI Prism 2000 Sequence Detection System (Applied Biosystems) using SYBR Green PCR Master Mix reagents (Applied Biosystems). Polymerase chain reaction conditions were as follows: AmpliTaq gold polymerase activation at 95°C for 10 min, followed by 40 cycles of denaturation at 95°C for 15 sec and annealing/extension at 60°C for 1 min. The relative expression ratio of a target gene was calculated based on the reaction efficiency value and the cycle at threshold (CT) deviation of an unknown sample versus a control and expressed in comparison to a reference gene. Acidic ribosomal phosphoprotein large P0 subunit (RPLPO) was used as a reference gene. SDS-PAGE and Western blot LNCaP cells were seeded in 25 cm2 flasks and grown to 70% confluency in RPMI-1640 medium containing 10% FBS. Cells were treated with 10 nM calcitriol for 6 – 48 hr. If cells were treated in serum-free conditions, 24 hr before treatment the conditioning medium was changed to one containing 1% BSA and cells were treated with calcitriol in 1% BSA-containing medium. In every sample, concentration of the vehicle (ethanol) was adjusted to 0.1%. Cells were washed in ice-cold PBS and lysed according to the manufacturer’s instructions in M-PER mammalian protein extraction reagent modified with protease inhibitors (Complete Mini Protease inhibitor cocktail, 1 tablet per 10 ml buffer) and sodium orthovanadate (1 mM) activated according to the manufacturer’s instructions. In brief, 400 ␮l of modified M-PER reagent

953

CALCITRIOL-INDUCED PROSTATE-DERIVED FACTOR

were added to each 25 cm2 flask and left at RT for 5 min with mild agitation. Lysates were collected into Eppendorf tubes, left on ice for 5–10 min and microcentrifuged at 12,000g for 10 min to remove cell debris. Supernatant was collected and immediately analyzed in Western blot or stored at ⫺70°C for not longer than 1 month. Total protein concentration in the lysate was measured using BCA protein assay kit. Total cell lysate samples containing the same amount of proteins (25 ␮g) were mixed with 2% SDS Laemmli buffer, heated at 95°C for 5 min and analyzed by electrophoresis in polyacrylamide gel (SDS-PAGE). Proteins separated by SDS-PAGE were electrophoretically transferred to the nitrocellulose membrane during 1 hr at RT using transfer buffer containing 25 mM Tris, 192 mM glycine and 20% methanol, pH 8.3. Membranes were incubated at 37°C for 1 hr in TBST buffer (50 mM Tris-HCl, 150 mM NaCl, 0.05% Tween 20, pH 8.0) containing 5% nonfat dry milk powder to saturate the nonspecific protein binding sites and incubated with specific primary antibody diluted in 5% nonfat milk-TBST (anti-NAG-1/PTGF-␤ rabbit polyclonal antibody, anti-␤-actin antibody) or 5% BSA-TBST (antiphospho-SMAD, antiphospho-ERK1/2, anti-ERK1/2) at 4°C overnight with mild agitation. The membranes were washed 3 times for 10 min with TBST and incubated for 1 hr with secondary antirabbit IgG HRP-conjugated antibody in 1% nonfat milk-TBST with gentle agitation. The membranes were washed 3 times for 10 min with TBST and subjected to ECL detection according to the manufacturer’s instructions. The intensities of scanned bands were measured by ImageJ software.

TABLE I – EXPRESSION OF GENES ENCODING TGF-␤ FAMILY AND BINDING PROTEINS IN LNCAP CELLS TREATED WITH CALCITRIOL Gene product name

Transforming growth factor-␤1 Latent TGF-␤ binding protein 1 Latent TGF-␤ binding protein 3 Prostate-derived factor Bone morphogenic protein 1 Bone morphogenic protein 4 Bone morphogenic protein 7 Inhibin-␣ Inhibin-␤A

Genbank accession number1

BC004248 BC006391

Fold

0.8 1.0 1.1 2.5 0.8 0.7 1.0 0.6 1.0

1 Only Genbank accession numbers indicated in the cDNA chip documentation are listed.

Cell growth assay Cells were plated in 96-well plates 1,500 cells/well and left to attach for 48 hr followed by treatment with calcitriol, hrPDF, TGF-␤1 and anti-NAG-1/PTGF-␤ antibody for 4 – 6 days. For the assay with PDF-specific antibody (anti-NAG-1/PTGF-␤), cells were pretreated with 10 nM calcitriol or vehicle for 12 hr before the addition of 2 ␮g/ml anti-NAG-1/PTGF-␤ antibody. In all the samples, the concentration of the vehicle was adjusted to 0.1% ethanol. The medium was changed and the treatment was renewed every second day. The relative cell number was quantified using crystal violet assay.25 In brief, cells were fixed with 11% glutaraldehyde, stained with crystal violet and counted in the microplate reader (Wallac Victor 1420 Multilabel counter, Wallac Oy, Turku, Finland). Absorbance at 590 nm represents relative cell number. Values are presented as absorbance at 590 nm multiplied to 1,000 and normalized to control sample values. Statistics All the experiments were performed at least in duplicates and repeated independently 3 to 5 times unless otherwise stated. All numerical data are expressed as mean ⫾ SD. Significance was analyzed using Student’s t-test. RESULTS

Calcitriol induces PDF mRNA in human prostate cancer LNCaP cells We performed cDNA microarray screening of calcitriol-regulated genes in LNCaP cells. Microarray analysis revealed 2.5-fold increase in mRNA level of PDF, the distantly related member of TGF-␤ superfamily, in cells treated with 10 nM calcitriol for 24 hr. Genes encoding TGF-␤ superfamily members and binding proteins present on the cDNA chip, which was taken for the screening, and the effect of calcitriol on their mRNA levels are listed in Table I. Induction of PDF mRNA by calcitriol was reevaluated and quantified by real-time RT-PCR. To characterize the time-course of mRNA, induction cells were treated with 10 nM calcitriol for 4 – 48 hr followed by RNA isolation and quantitative RT-PCR analysis with PDF-specific oligonucleotide primers. Time-course for PDF mRNA induction by calcitriol is shown in Figure 1a. After treatment with calcitriol for 4 hr, PDF mRNA was increased

FIGURE 1 – Induction of PDF mRNA by calcitriol in LNCaP cells. (a) Cells were treated with 10 nM calcitriol or vehicle (0.1% ethanol) for the periods of time shown. (b) Cells were treated with calcitriol in the indicated concentrations or vehicle (0.1% ethanol) for 24 hr. Total RNA was isolated and analyzed by quantitative RT-PCR. The relative expression ratio of the target gene was calculated based on reaction efficiency value and the CT deviation of the unknown sample versus the control and expressed in comparison to the reference gene (RPLPO). Asterisk, significantly different from control at p ⬍ 0.05.

1.5-fold and reached maximum, 2.5- to 3-fold, after 24-hr treatment. To characterize the dependence of PDF mRNA induction on the concentration of calcitriol, cells were treated for 24 hr with

954

NAZAROVA ET AL.

FIGURE 2 – Calcitriol does not affect the stability of PDF mRNA. LNCaP cells were pretreated with 10 nM calcitriol or vehicle (0.2% ethanol) for 8 hr followed by addition of 5 ␮g/ml actinomycin D. RNA was isolated after treatment with actinomycin D for periods of time shown and analyzed with quantitative RT-PCR. The relative expression ratio of the target gene was calculated as in Figure 1 and normalized to the control (cells pretreated with vehicle).

calcitriol in concentrations increasing from 1 to 100 nM. Induction of PDF mRNA was highly dependent on calcitriol dose and was maximal (about 5-fold) at maximal calcitriol concentration used (100 nM; Fig. 1b). Time- and concentration-course were analyzed both in serum and serum-free conditions and were largely similar independently of presence of serum. Calcitriol does not affect PDF mRNA stability in LNCaP cells To find out if calcitriol induces PDF mRNA by transcriptional or posttranscriptional mechanism, we used actinomycin D to block the transcription in LNCaP cells. Cells pretreated with 10 nM calcitriol or vehicle (0.1% ethanol) for 8 hr were subsequently treated with 5 ␮g/ml actinomycin D in the presence or absence of calcitriol for 1–5 hr. RNA was isolated and analyzed by quantitative RT-PCR with PDF-specific oligonucleotide primers. The halflife of PDF mRNA was 1.5–2 hr and was not affected by calcitriol (Fig. 2). Thus, calcitriol does not affect the stability of PDF mRNA and the induction is due to the activation of PDF gene transcription. To see if PDF mRNA induction is mediated by protein synthesis, we used cycloheximide to block protein synthesis in LNCaP cells. Cells were treated with 10 nM calcitriol for 24 hr in the presence or absence of 10 ␮g/ml cycloheximide followed by RNA isolation and quantitative RT-PCR analysis. Cycloheximide did not prevent PDF mRNA induction mediated by calcitriol but itself potently induced PDF mRNA. Also, cycloheximide attenuated the effect of calcitriol on PDF mRNA level, which raises the possibility that the induction of PDF mRNA by calcitriol does not depend on protein synthesis (data not shown). The time-course for PDF mRNA induction, which is significant already after 4-hr treatment with calcitriol, supports this proposal. Calcitriol does not affect PDF expression in human prostate cancer PC-3 cells To characterize PDF expression and its regulation by calcitriol in prostate cancer cells, we analyzed PDF mRNA level in another calcitriol-responsive though less sensitive to growth-inhibitory effect of calcitriol human prostate cancer cell line PC-3. PC-3 cells were treated with 10 nM calcitriol or vehicle for 10 and 24 hr and the total RNA was isolated and analyzed by quantitative RT-PCR with PDF-specific primers. PDF mRNA level in PC-3 cells was 3–10 times lower than in LNCaP (Fig. 3a) and calcitriol had no significant effect on PDF mRNA level (Fig. 3b).

FIGURE 3 – PDF mRNA expression and its regulation by calcitriol in PC-3 cells. (a) Comparison of PDF mRNA level in LNCaP and PC-3 cells. (b) Calcitriol does not affect PDF mRNA level in PC-3 cells. Cells were treated with 10 nM calcitriol or vehicle (0.1% ethanol) for the periods of time shown. Total RNA was isolated and analyzed by quantitative RT-PCR. The relative expression ratio of the target gene was calculated as in Figure 2.

Casodex and pifithrin-␣ do not affect calcitriol-mediated induction of PDF mRNA The earlier reports demonstrate that the effect of calcitriol on the growth of LNCaP cells is androgen-dependent.26,27 It was also reported that the expression of PDF is regulated by androgens.6,18 To see if the effect of calcitriol on the induction of PDF mRNA in LNCaP cells is mediated by androgens, we treated cells with 10 nM calcitriol in the presence or absence of antiandrogen Casodex for 24 hr. Casodex in concentration range of 1–50 ␮M did not prevent calcitriol-mediated induction of PDF mRNA (data not shown). PDF gene is known to be a primary p53 target gene.15,16,28 To see if the effect of calcitriol is mediated by p53, we treated cells with the specific p53 inhibitor pifithrin-␣, which is a low-molecular-weight inhibitor shown to prevent p53 effects on gene expression.29 LNCaP cells were treated with 10 nM calcitriol in the presence or absence of 1–30 ␮M pifithrin-␣; RNA was isolated and analyzed by quantitative

CALCITRIOL-INDUCED PROSTATE-DERIVED FACTOR

FIGURE 4 – Induction of PDF protein by calcitriol in LNCaP cells. Cells were treated with 10 nM calcitriol (⫹) or vehicle (⫺) for the indicated periods of time. Cell lysate was analyzed by Western blot with PDF-specific antibody. The same blots were rehybridized with antibodies specific for ␤-actin. Blots were analyzed and quantified by ImageJ software. Relative PDF protein level was calculated and normalized to sample treated with vehicle for 24 hr. Asterisk, significantly different from control at p ⬍ 0.05.

RT-PCR with PDF-specific primers. The concentration of pifithrin-␣ as high as 30 ␮M did not prevent but stimulated the induction of PDF by calcitriol (data not shown). Also, we studied the effect of calcitriol on p53 expression in LNCaP cells by Western blot with p53-specific monoclonal antibody. No difference in p53 protein level in LNCaP cells untreated or treated with 10 nM calcitriol for 10 –24 hr could be seen (data not shown). Calcitriol induces PDF protein in LNCaP cells We investigated PDF expression in LNCaP cells by Western blot analysis using PDF-specific (anti-NAG-1/PTGF-␤) polyclonal antibody. To characterize the time-course of PDF protein induction, LNCaP cells were treated with 10 nM calcitriol for 24 –72 hr. Cellular proteins were separated by SDS-PAGE and analyzed by Western blot. Membranes were scanned and bands were analyzed and quantified using ImageJ software. PDF protein induction mediated by calcitriol was clearly time-dependent, being about 1.5fold after 24 hr of treatment with calcitriol and about 3-fold maximal after 72 hr (Fig. 4). To analyze the dependence of PDF protein induction of concentration of calcitriol, we treated LNCaP cells for 48 hr with the calcitriol in concentrations increasing from 0.1 to 100 nM. The induction was highly dependent on calcitriol concentration; it could be already seen at 0.1 nM concentration of calcitriol and was maximal (over 5-fold) at the maximal calcitriol concentration used (100 nM; Fig. 5). Time- and concentrationcourse were analyzed both in serum and serum-free conditions and were largely similar independent of the presence of serum.

955

FIGURE 5 – Induction of PDF protein by calcitriol in LNCaP cells. Cells were treated for 48 hr with the concentrations of calcitriol shown or vehicle (0.1% ethanol). Cell lysate was analyzed by Western blot with PDF-specific antibodies. The same blots were rehybridized with antibodies specific for ␤-actin. Blots were analyzed as in Figure 4 and the relative PDF protein level was normalized to vehicle-treated sample. Asterisk, significantly different from control at p ⬍ 0.05.

Human recombinant PDF has inhibitory effect on LNCaP cell growth To study the role of PDF in a complex control of LNCaP cell growth, we treated cells with human recombinant PDF (hrPDF). Cells were grown in the presence of 10, 50 and 250 ng/ml hrPDF for 6 days and the growth curves were made; 10 ng/ml hrPDF had no significant effect on cell growth, while 50 and 250 ng/ml hrPDF inhibited LNCaP cell growth for about 15–20% at day 6. The maximum growth inhibition was reached already with 50 ng/ml hrPDF (Fig. 6). It was suggested in earlier reports that PDF might act through the same signaling pathway, as does TGF-␤1.16 But whether TGF-␤1 signaling pathway is functional in LNCaP cells is not clear. To compare the effect of PDF on the growth of LNCaP cells with TGF-␤1 response, we treated cells with 10 ng/ml human TGF-␤1 and the growth curve was made the same way as with hrPDF. We could not detect any effect of TGF-␤1 on the basal growth of LNCaP cells (Fig. 6). PDF-specific antibody does not affect basal growth of LNCaP cells but partially reverses growth-inhibitory effect of calcitriol To study if the autocrine PDF significantly contributes to the control of LNCaP cell growth, we used PDF-specific antibodies to neutralize the factor secreted into the culture medium. Cells were grown in the presence or absence of the pretested dilution of PDF-specific antibodies and the growth curves were made. AntiPDF antibody had no significant effect on LNCaP cells growth (Fig. 7). To study if the induction of PDF mediated by calcitriol contributes to the inhibitory effect of calcitriol on LNCaP cell growth, we

956

NAZAROVA ET AL.

FIGURE 6 – Inhibition of LNCaP cell growth by calcitriol and hrPDF. Cells were treated with 10 nM calcitriol or 50 ng/ml hrPDF or vehicle (0.1% ethanol) for 6 days. Cells were fixed and the treatment renewed every second day. Relative cell number was quantified using crystal violet assay. Relative cell number counted as absorbance at 590 nm multiplied to 1,000 and normalized to vehicle-treated sample. Asterisk, significantly different from calcitriol-treated sample at p ⬍ 0.05.

treated cells with PDF-specific antibody in the absence or presence of calcitriol. Cells were pretreated with 10 nM calcitriol for 24 hr and pretested dilution of PDF-specific antibodies was added to the culture medium. The growth curves were made as previously. PDF-specific antibody partially reversed inhibition of LNCaP cell growth caused by calcitriol (Fig. 7). Human recombinant PDF does not affect SMAD phosphorylation in LNCaP cells To see if PDF acts through the classical TGF-␤1 or BMP signaling pathway in LNCaP cells, we studied the effect of PDF on SMAD phosphorylation. Cells were treated with 50 ng/ml hrPDF for 15–90 min in serum-free medium. Cellular proteins were separated by SDS-PAGE and analyzed by Western blot with either phospho-SMAD2- or phospho-SMAD1/5/8-specific antibody. PDF had no significant effect on either SMAD2 or SMAD1/5/8 phosphorylation (data not shown). Human recombinant PDF stimulates ERK1/2 phosphorylation in LNCaP cells A resent report demonstrated that PDF acts through MAPK signaling pathway independently of TGF-␤ receptor activation in gastric cancer cells.19 In these cells, PDF induced phosphorylation of ERK1/2, which resulted in the activation of the system of cellular proteases. To see if PDF affects ERK1/2 phosphorylation in LNCaP cells, we treated cells with 50 ng/ml hrPDF for 2– 15 min in serum-free medium. Cellular proteins were separated by SDS-PAGE and analyzed by Western blot with antibody specific for phospho-ERK1 and phospho-ERK2. The same blots were stripped and rehybridized with antibody specific for total ERK1 and ERK2. hrPDF induced transient phosphorylation of ERK1/2 after 2 min of treatment, which was reversed by 10 min of treatment (Fig. 8). DISCUSSION

Hormonal forms of vitamin D play an important role in the complex control of prostate cell growth and might be involved in the prevention of prostate cancer. Calcitriol inhibits prostate cancer

FIGURE 7 – Effect of PDF-specific antibody on growth of LNCaP cells. Cells were pretreated with 10 nM calcitriol or vehicle (0.1% ethanol) for 12 hr followed by addition of pretested dilution of PDFspecific antibody. Cells were treated for 4 days, fixed on every second day and relative cell number was quantified using crystal violet assay. Asterisk, significantly different from calcitriol-treated sample at p ⬍ 0.05.

FIGURE 8 – Induction of ERK phosphorylation by hrPDF in LNCaP cells. Cells were treated with 50 ng/ml hrPDF for the periods of time shown. Cell lysate was analyzed by Western blot with antibody specific for phosphorylated ERK1/2. The same blots were reprobed with antibodies for total ERK1/2. Blots were analyzed as in Figure 4.

cell growth by the complex mechanism that includes cell cycle arrest, inhibition of proliferation, induction of differentiation and apoptosis. LNCaP, androgen-responsive human prostate cancer cell line, is highly responsive to growth-inhibitory action of calcitriol.30 –32 Using cDNA microarray technique, we screened for calcitriol-regulated genes in LNCaP cells. A set of genes was differentially regulated in LNCaP cells treated with 10 nM calcitriol for 24 hr. cDNA microarray analysis revealed increased mRNA level of PDF, a member of divergent group within TGF-␤ superfamily. A list of genes encoding TGF-␤ family and binding proteins present in the chip is shown in Table I. Induction of PDF mRNA detected by cDNA microarray was reevaluated and quantified by real-time RT-PCR with PDF-specific primers (Fig. 1). Calcitriol induces PDF mRNA time- and dose-dependently and does not affect the half-life of PDF mRNA, which means that it affects the transcription of PDF gene (Fig. 2). Because in the various cell types PDF is characterized by the proapoprotic and prodifferentiative properties, we suggested that PDF could mediate the growth-inhibitory effect of calcitriol in prostate cancer cells. We could not detect any effect of calcitriol on the expression of PDF in PC-3 human prostate cancer cells (Fig. 3b), which are responsive to the growth-inhibitory effect of calcitriol, though to a lesser extent than LNCaP.30 –32 It was previously reported that in PC-3 cells, the level of PDF expression is lower compared to LNCaP.18,21,30 –32 Our results are in agreement with the literature data (Fig. 3a). The difference in ratio is most likely due to the

CALCITRIOL-INDUCED PROSTATE-DERIVED FACTOR

different techniques used for the quantification. The fact that calcitriol does not affect PDF expression in PC-3, p53 null androgen-unresponsive cells, and that androgens and p53 are potent regulators of PDF expression in human prostate, leads us to the suggestion that the induction of PDF in LNCaP cells may be mediated by p53 and/or androgens. The differential regulation of PDF expression by the various factors including different retinoids,33 genistein,34 NSAIDs,17,28 hHG35 and anoxia36 has been reported recently. It was shown that the expression of PDF is efficiently regulated both in p53-dependent and p53-independent manner. In our study, neither antiandrogen Casodex nor specific p53 inhibitor pifithrin-␣ prevented induction of PDF mRNA mediated by calcitriol, which suggest a possibility of p53- and androgen-independent induction by calcitriol, though the present data are not sufficient to reach a conclusion. Calcitriol dramatically induces PDF protein synthesis in LNCaP cells, as we have shown by Western blot analysis. The effect is clearly time- and dose-dependent. The protein is induced stably, significantly after 24 hr and increasing toward 72 hr of the treatment with calcitriol (Figs. 4 and 5). The major function of PDF and also the role of PDF in tumorigenesis are not known. We demonstrated the inhibitory effect of human recombinant PDF on LNCaP cell growth (Fig. 6), though PDF-specific antibodies did not affect basal growth of LNCaP cells (Fig. 7). We suppose that the level of basal autocrine PDF production is not sufficient to inhibit LNCaP cell growth, as it was previously reported that high levels of PDF are necessary for its growth-inhibitory effect.16,17,36 A recent report on PDF expression and functioning in human prostate cancer cells demonstrated a lack of PDF effect on the proliferation of human prostate cancer cells (LNCaP, PC-3 and DU145), as was shown by BrdU incorporation, but suggested a role for PDF in decreasing cell adhesion, which may subsequently lead to apoptosis, and also its role in prostate cell differentiation.18 Our results do not exclude the possibility that the effect on cell growth observed is due to the decrease in cell adhesion and secondary apoptosis and not the inhibition of mitotic activity. The role of PDF in the processes of cell adhesion and invasiveness has also been suggested by other studies,19 but this function of PDF is not completely understood. It was demonstrated that the apoptosis induced by the various anticancer agents such as NSAIDs and resveratrol is due to PDF induction in human colorectal cells.13,28,37,38 The ability of PDF to suppress tumor cell growth in vitro and in vivo was demonstrated on colorectal carcinoma and glioblastoma cell lines.16,17,36 PDF is often overexpressed in cancer39 – 42 and is upregulated during the transition of human prostate cancer LNCaP cells to the androgen-independent state.21,43 Studies on gastric cancer cells also propose the involvement of PDF in tumor progression.19 In situ hybridization studies

957

demonstrated the high level of PDF in normal prostate with considerable decrease during the progression of cancer at the primary site and reexpression in an osseous metastatic lesions.20 In this sense, PDF shares the main characteristics of growth factors belonging to the TGF-␤ superfamily, which act both positively and negatively on tumorigenesis depending on the molecular and cellular context. The molecular mechanisms of PDF action are not fully investigated. It has been reported that PDF, like TGF-␤, requires signaling pathway mediated by TGF-␤ receptors as well as receptoractivated SMAD4 to suppress tumor cell growth.16 We could detect neither SMAD2 nor SMAD 1/5/8 activation in LNCaP cells treated with hrPDF (data not shown). LNCaP cells are TGF-␤unresponsive and the nature of this resistance is poorly investigated. It was reported that at certain concentration of androgen, the response to growth-inhibitory effect of TGF-␤1 is restored in LNCaP cells.44 PDF can act as well through alternative signaling pathways. It was demonstrated that in gastric cancer cells, ERK1/2 is strongly activated in response to PDF and independent of TGF-␤ receptors and leads to activating of the urokinase-type plasminogen activator system and the potentiation of gastric cells invasion.19 In our study, PDF caused transient activation of ERK1/2 (Fig. 8), which raises the possibility that its effect in LNCaP cells may be mediated through a pathway different from the classical TGF-␤ signaling pathway, such as MAP kinase pathway, which promotes processes of differentiation and apoptosis. PDF-specific antibody partially reverses growth-inhibitory effect of calcitriol in LNCaP cells, which supports the proposal that PDF might partly mediate the effect of calcitriol on LNCaP cell growth. The crosstalk between calcitriol and various peptide growth factors such as insulin-like growth factor in prostate cancer LNCaP cells and TGF-␤ in breast cancer MCF-7 cells has been reported, which mediates inhibition of cellular growth by calcitriol. The present data make a new link between calcitriol and TGF-␤ superfamily in the control of prostate cancer cell growth and suggest a new mechanism of the inhibition of cell growth by calcitriol. ACKNOWLEDGEMENTS

The authors thank Dr. Heimo Syva¨la¨, Department of Anatomy, University of Tampere Medical School, for valuable advices; Mrs. Hilkka Ma¨kinen and Mrs. Mirja Hippo¨nen for excellent technical assistance; Virginia Mattila, Translation Services, Language Center, University of Tampere, for revising the language; and Leo Pharmaceuticals (Ballerup, Denmark) for the generous gift of calcitriol and AstraZeneca (London, UK) for donating Casodex.

REFERENCES

1. 2.

3. 4. 5. 6.

7.

Schwartz GG, Hulka BS. Is vitamin D deficiency a risk factor for prostate cancer? Anticancer Res 1990;10:1307–11. Corder EH, Guess HA, Hulka BS, Friedman GD, Sadler M, Vollmer RT, Lobaugh B, Drezner MK, Vogelman JH, Orentreich N. Vitamin D and prostate cancer: a prediagnostic study with stored sera. Cancer Epidemiol Biomarkers Prev 1993;2:467–72. Ahonen MH, Tenkanen L, Teppo L, Hakama M, Tuohimaa P. Prostate cancer risk and prediagnostic serum 25-hydroxyvitamin D levels (Finland). Cancer Causes Control 2000;11:847–52. Ylikomi T, Laaksi I, Lou YR, Martikainen P, Miettinen S, Pennanen P, Purmonen S, Syvala H, Vienonen A, Tuohimaa P. Antiproliferative action of vitamin D. Vitam Horm 2002;64:357– 406. Gumienny TL, Padgett RW. The other side of TGF-beta superfamily signal regulation: thinking outside the cell. Trends Endocrinol Metab 2002;13:295–9. Paralkar VM, Vail AL, Grasser WA, Brown TA, Xu H, Vukicevic S, Ke HZ, Qi H, Owen TA, Thompson DD. Cloning and characterization of a novel member of the transforming growth factor-beta/bone morphogenetic protein family. J Biol Chem 1998;273:13760 –7. Hromas R, Hufford M, Sutton J, Xu D, Li Y, Lu L. PLAB, a novel placental bone morphogenetic protein. Biochim Biophys Acta 1997; 1354:40 – 4.

8.

9. 10.

11.

12.

13.

Lawton LN, Bonaldo MF, Jelenc PC, Qiu L, Baumes SA, Marcelino RA, de Jesus GM, Wellington S, Knowles JA, Warbutron D, Brown S, Soares MB. Identification of a novel member of the TGF-beta superfamily highly expressed in human placenta. Gene 1997;203:17– 26. Yokoyama-Kobayashi M, Saeki M, Sekine S, Kato S. Human cDNA encoding a novel TGF-beta superfamily protein highly expressed in placenta. J Biochem (Tokyo) 1997;122:622– 6. Bootcov MR, Bauskin AR, Valenzuela SM, Moore AG, Bansal M, He XY, Zhang HP, Donnellan M, Mahler S, Pryor K, Walsh BJ, Nicholson RC. MIC-1, a novel macrophage inhibitory cytokine, is a divergent member of the TGF-beta superfamily. Proc Natl Acad Sci USA 1997;94:11514 –9. Bottner M, Laaff M, Schechinger B, Rappold G, Unsicker K, SuterCrazzolara C. Characterization of the rat, mouse, and human genes of growth/differentiation factor-15/macrophage inhibiting cytokine-1 (GDF-15/MIC-1). Gene 1999;237:105–11. Strelau J, Bottner M, Lingor P, Suter-Crazzolara C, Galter D, Jaszai J, Sullivan A, Schober A, Krieglstein K, Unsicker K. GDF-15/MIC-1 a novel member of the TGF-beta superfamily. J Neural Transm 2000;60(Suppl):273– 6. Baek SJ, Horowitz JM, Eling TE. Molecular cloning and character-

958

14.

15.

16.

17.

18.

19.

20.

21. 22. 23.

24.

25. 26. 27. 28.

29.

NAZAROVA ET AL.

ization of human nonsteroidal anti-inflammatory drug-activated gene promoter: basal transcription is mediated by Sp1 and Sp3. J Biol Chem 2001;276:33384 –92. Uchida K, Chaudhary LR, Sugimura Y, Adkisson HD, Hruska KA. Proprotein convertases regulate activity of prostate epithelial cell differentiation markers and are modulated in human prostate cancer cells. J Cell Biochem 2003;88:394 –9. Li PX, Wong J, Ayed A, Ngo D, Brade AM, Arrowsmith C, Austin RC, Klamut HJ. Placental transforming growth factor-beta is a downstream mediator of the growth arrest and apoptotic response of tumor cells to DNA damage and p53 overexpression. J Biol Chem 2000; 275:20127–35. Tan M, Wang Y, Guan K, Sun Y. PTGF-beta, a type beta transforming growth factor (TGF-beta) superfamily member, is a p53 target gene that inhibits tumor cell growth via TGF-beta signaling pathway. Proc Natl Acad Sci USA 2000;97:109 –14. Baek SJ, Kim KS, Nixon JB, Wilson LC, Eling TE. Cyclooxygenase inhibitors regulate the expression of a TGF-beta superfamily member that has proapoptotic and antitumorigenic activities. Mol Pharmacol 2001;59:901– 8. Liu T, Bauskin AR, Zaunders J, Brown DA, Pankurst S, Russell PJ, Breit SN. Macrophage inhibitory cytokine 1 reduces cell adhesion and induces apoptosis in prostate cancer cells. Cancer Res 2003;63: 5034 – 40. Lee DH, Yang Y, Lee SJ, Kim KY, Koo TH, Shin SM, Song KS, Lee YH, Kim YJ, Lee JJ, Choi I, Lee JH. Macrophage inhibitory cytokine-1 induces the invasiveness of gastric cancer cells by up-regulating the urokinase-type plasminogen activator system. Cancer Res 2003;63:4648 –55. Thomas R, True LD, Lange PH, Vessella RL. Placental bone morphogenetic protein (PLAB) gene expression in normal, premalignant and malignant human prostate: relation to tumor development and progression. Int J Cancer 2001;93:47–52. Karan D, Chen SJ, Johansson SL, Singh AP, Paralkar VM, Lin MF, Batra SK. Dysregulated expression of MIC-1/PDF in human prostate tumor cells. Biochem Biophys Res Commun 2003;305:598 – 604. Koli K, Keski-Oja J. 1,25-dihydroxyvitamin D3 enhances the expression of transforming growth factor beta 1 and its latent form binding protein in cultured breast carcinoma cells. Cancer Res 1995;55:1540– 6. Yanagisawa J, Yanagi Y, Masuhiro Y, Suzawa M, Watanabe M, Kashiwagi K, Toriyabe T, Kawabata M, Miyazono K, Kato S. Convergence of transforming growth factor-beta and vitamin D signaling pathways on SMAD transcriptional coactivators. Science 1999;283: 1317–21. Subramaniam N, Leong GM, Cock TA, Flanagan JL, Fong C, Eisman JA, Kouzmenko AP. Cross-talk between 1,25-dihydroxyvitamin D3 and transforming growth factor-beta signaling requires binding of VDR and Smad3 proteins to their cognate DNA recognition elements. J Biol Chem 2001;276:15741– 6. Kueng W, Silber E, Eppenberger U. Quantification of cells cultured on 96-well plates. Anal Biochem 1989;182:16 –9. Zhao XY, Ly LH, Peehl DM, Feldman D. 1alpha,25-dihydroxyvitamin D3 actions in LNCaP human prostate cancer cells are androgendependent. Endocrinology 1997;138:3290 – 8. Bao BY, Hu YC, Ting HJ, Lee YF. Androgen signaling is required for the vitamin D-mediated growth inhibition in human prostate cancer cells. Oncogene 2004;23:3350 – 60. Bottone FG Jr, Baek SJ, Nixon JB, Eling TE. Diallyl disulfide (DADS) induces the antitumorigenic NSAID-activated gene (NAG-1) by a p53-dependent mechanism in human colorectal HCT 116 cells. J Nutr 2002;132:773– 8. Komarov PG, Komarova EA, Kondratov RV, Christov-Tselkov K, Coon JS, Chernov MV, Gudkov AV. A chemical inhibitor of p53 that

30. 31. 32. 33.

34.

35.

36.

37.

38. 39.

40.

41.

42.

43.

44.

protects mice from the side effects of cancer therapy. Science 1999; 285:1733–7. Skowronski RJ, Peehl DM, Feldman D. Vitamin D and prostate cancer: 1,25 dihydroxyvitamin D3 receptors and actions in human prostate cancer cell lines. Endocrinology 1993;132:1952– 60. Schwartz GG, Oeler TA, Uskokovic MR, Bahnson RR. Human prostate cancer cells: inhibition of proliferation by vitamin D analogs. Anticancer Res 1994;14:1077– 81. Skowronski RJ, Peehl DM, Feldman D. Actions of vitamin D3, analogs on human prostate cancer cell lines: comparison with 1,25dihydroxyvitamin D3. Endocrinology 1995;136:20 – 6. Newman D, Sakaue M, Koo JS, Kim KS, Baek SJ, Eling T, Jetten AM. Differential regulation of nonsteroidal anti-inflammatory drugactivated gene in normal human tracheobronchial epithelial and lung carcinoma cells by retinoids. Mol Pharmacol 2003;63:557– 64. Wilson LC, Baek SJ, Call A, Eling TE. Nonsteroidal anti-inflammatory drug-activated gene (NAG-1) is induced by genistein through the expression of p53 in colorectal cancer cells. Int J Cancer 2003;105: 747–53. Graichen R, Liu D, Sun Y, Lee KO, Lobie PE. Autocrine human growth hormone inhibits placental transforming growth factor-beta gene transcription to prevent apoptosis and allow cell cycle progression of human mammary carcinoma cells. J Biol Chem 2002;277: 26662–72. Albertoni M, Shaw PH, Nozaki M, Godard S, Tenan M, Hamou MF, Fairlie DW, Breit SN, Paralkar VM, de Tribolet N, Van Meir EG, Hegl ME. Anoxia induces macrophage inhibitory cytokine-1 (MIC-1) in glioblastoma cells independently of p53 and HIF-1. Oncogene 2002;21:4212–9. Baek SJ, Wilson LC, Lee CH, Eling TE. Dual function of nonsteroidal anti-inflammatory drugs (NSAIDs): inhibition of cyclooxygenase and induction of NSAID-activated gene. J Pharmacol Exp Ther 2002;301: 1126 –31. Baek SJ, Wilson LC, Eling TE. Resveratrol enhances the expression of non-steroidal anti-inflammatory drug-activated gene (NAG-1) by increasing the expression of p53. Carcinogenesis 2002;23:425–34. Welsh JB, Sapinoso LM, Su AI, Kern SG, Wang-Rodriguez J, Moskaluk CA, Frierson HF, Jr. Hampton GM. Analysis of gene expression identifies candidate markers and pharmacological targets in prostate cancer. Cancer Res 2001;61:5974 – 8. Nakamura T, Scorilas A, Stephan C, Yousef GM, Kristiansen G, Jung K, Diamandis EP. Quantitative analysis of macrophage inhibitory cytokine-1 (MIC-1) gene expression in human prostatic tissues. Br J Cancer 2003;88:1101– 4. Welsh JB, Sapinoso LM, Kern SG, Brown DA, Liu T, Bauskin AR, Ward RL, Hawkins NJ, Quinn DI, Russell PJ, Sutherland RL, Breit SN. Large-scale delineation of secreted protein biomarkers overexpressed in cancer tissue and serum. Proc Natl Acad Sci USA 2003; 100:3410 –5. Koopmann J, Buckhaults P, Brown DA, Zahurak ML, Sato N, Fukushima N, Sokoll LJ, Chan DW, Yeo CJ, Hruban RH, Breit SN, Kinzler KW. Serum macrophage inhibitory cytokine 1 as a marker of pancreatic and other periampullary cancers. Clin Cancer Res 2004; 10:2386 –92. Karan D, Kelly DL, Rizzino A, Lin MF, Batra SK. Expression profile of differentially-regulated genes during progression of androgen-independent growth in human prostate cancer cells. Carcinogenesis 2002;23:967–75. Kim IY, Zelner DJ, Sensibar JA, Ahn HJ, Park L, Kim JH, Lee C. Modulation of sensitivity to transforming growth factor-beta 1 (TGFbeta 1) and the level of type II TGF-beta receptor in LNCaP cells by dihydrotestosterone. Exp Cell Res 1996;222:103–10.