Cancer and Prostatic Diseases (2009) 12, 375–381 & 2009 Nature Publishing Group All rights reserved 1365-7852/09 $32.00 www.nature.com/pcan ORIGINAL ARTICLE

Effect of selective receptor modulators on cell proliferation and activities in normal human prostate stromal and epithelial cells

H Nomura, H Kawashima, S Masaki, TY Hosono, K Matsumura, S Tamada, T Tanaka and T Nakatani Department of Urology, Osaka City University Graduate School of Medicine, Asahimachi, Abenoku, Osaka, Japan

We examined the effect of E2 and selective estrogen receptor modulators (SERMs) on the proliferation and estrogen receptor (ER) activities in normal human prostate cells. SERMs such as toremifene, and suppressed the proliferation of prostate epithelial and stromal cells whereas anti- did not. In prostate stromal cells, the transactivation activities of ER were enhanced by adding E2 and reduced remarkably by toremifene. The results indicate that the ER-mediated pathway plays a central role in the growth of normal prostate cells. and Prostatic Diseases (2009) 12, 375–381; doi:10.1038/pcan.2009.20; published online 26 May 2009

Keywords: selective estrogen (SERM); estrogen receptors a and b; normal prostate stromal cells; normal prostate epithelial cells

Introduction investigated; the results indicate the possible involve- ment of estrogen receptors (ERs) in prostate cancer cell Although the most striking characteristic of the prostate proliferation.5,10 gland is its dependence, estrogen and the BPH is the most common prostatic disease that will estrogen receptors have been reported to be involved affect many men as they age. One of the hypothesized in the development of the prostate gland and the pathogeneses of BPH is that the stromal hyperplasia is pathogenesis of prostatic diseases.1 Maternal , related to an enhanced estrogenic status after middle age, for example, stimulates squamous metaplasia within the with a decreased androgen level and an increased developing prostatic epithelium in a male offspring.2 conversion of adrenal androgens to estrogen by aroma- With regard to localized prostate cancer, expression tase.11 Recently, key studies on the role of estrogen in the levels of the estrogen receptor b (ERb) decline with pathogenesis of BPH have been carried out using normal increasing grade from prostatic intraepithelial neoplasia human prostate stromal and epithelial cells. King et al.12 through low to high Gleason scores, suggesting that ERb reported that an increased estrogen/androgen ratio is a putative tumor suppressor.2 In contrast, stimulated the proliferation of prostate stromal cells seem to induce prostate stromal cell proliferation; it is and enhanced the growth of prostate epithelial cells co- well known that the administration of estrogens com- cultured with stromal cells. Zhong et al.13 focused on bined with androgens has been used to induce BPH ERK, which was shown to be activated by estradiol with experimentally in dogs.2 the proliferation of prostate stromal cells. Since the discovery of the ERb,3 expression of estrogen While those investigations support that ERs a and b receptors a and b has been extensively studied using may directly modulate the prostatic cell growth, it is RT-PCR or immunohistochemical technique in several necessary to investigate whether the ERs in prostatic cells prostate cancer cell lines as well as in normal and are actually activated by ligands with the corresponding malignant prostatic tissues, those authors associating the transcription of the downstream genes causing changes expression of each estrogen receptor (ER) in various cells in cell proliferation. with its possible function and role.4–9 Effect of ligands Primary cultures of normal human prostatic epithelial and selective estrogen receptor modulators (SERMs) on and stromal cells are commercially available and have the growth of several prostate cancer cells has also been been used as useful models for the normal prostate epithelium and normal stroma in many studies. They exhibit limited lifespan with cellular senescence in Correspondence: Dr H Kawashima, Department of Urology, Osaka culture.14,15 In cultured prostatic stromal cells, mRNA City University Graduate School of Medicine, 1-4-3 Asahimachi, of (AR) is expressed,16–18 whereas Abenoku, Osaka 545-8585, Japan. reduced expression of AR protein in late passages has E-mail: [email protected] 18 Received 21 January 2009; revised 7 April 2009; accepted 8 April 2009; been reported. The commercially available prostatic published online 26 May 2009 epithelial cells possess the features of basal epithelia that Effect of SERMs on normal human prostate cells H Nomura et al

376 lack the expression of AR protein;15,19,20 those cells are Luciferase assay not fully differentiated to be the secretory epithelial The luciferase assay was carried out using Dual- cells.15 luciferase reporter assay system (Promega) according to Currently, we show for the first time that endogenous the methods recommended by the manufacturer. In estrogen receptors were activated by estrogen in the short, cells were rinsed with PBS and harvested follow- primary cultures of normal human prostate cells, and ing the addition of 250 ml of the lysis buffer per well by that their activities were modulated by selective estrogen scraping with a rubber policeman. The cells were receptor modulators (SERMs) accompanied by changes subjected to two freeze/thaw cycles to accomplish in cell growth. complete lysis, and the lysate was cleaned by centrifuga- tion for 1 min. After preparation of the Luciferase Assay Reagent II and the Stop & Glo Reagent supplied in the kit, 20 ml of the cell lysate was added to 100 ml of the Materials and methods Luciferase Assay Reagent II in a luminometer tube and mixed. The firefly luciferase activity was measured using Materials a luminometer with the program of a 2-s premeasure- The normal human prostate stromal and epithelial cells, ment delay followed by a 10-s measurement period for PrSC and PrEC, were purchased from LonzaWalkersville each reporter assay. Then, 100 ml of Stop & Glo Reagent (Walkersville, MD, USA). The prostate cancer cell line, was further added and the second measurement of the LNCaP cells, was purchased from American Type renilla luciferase activity was performed. The results of Culture Collection (Rockville, MD, USA). Dual-luciferase the luciferase assay were normalized using the activity of reporter assay system was obtained from Promega the renilla expression vector pRL-TK co-transfected (Madison, WI, USA) and FuGENE 6 Transfecton Reagent together with the reporter plasmid. from Roche (Mannheim, Germany). A reporter gene for ER, pGL3-ERE-X3tK luciferase (pERE-luc), was a gener- ous gift from Dr. Shigeaki Kato, Tokyo University, Tokyo, Cell-proliferation assay Japan. An anti-androgen, flutamide and an anti-estrogen, The cell proliferation was measured using the water- toremifene, were generous gifts from Nippon Kayaku soluble tetrazolium salt assay system (WST-1 Cell Corp. (Tokyo, Japan). Raloxifene, tamoxifen, Proliferation System: TaKaRa Bio, Otsu, Shiga, Japan). and 5a- (DHT) were purchased from In short, PrSC or PrEC cells were seeded at 1 Â105 cells Sigma-Aldrich (Saint Louis, MO, USA) and a pure anti- per well onto 6-well plates and incubated for 24 h. Then, estrogen faslodex (ICI182,780) was from Tocris (Ellisville, ligands were added and the cells were further grown for MO, USA). b-Estradiol and (BPA) were from 48 h and subjected to WST-1 assay. WST-1 reagent was Wako Pure Chemical (Osaka, Japan). added to each well, cells were incubated at 37 1C for 2 h and the absorbance was measured at 450 nM.

Cell culture and transient transfection TUNEL assay The normal human prostate stromal and epithelial cells, The TUNEL (terminal deoxynucleotidyl transferase- PrSC and PrEC, were maintained in Prostate Stromal mediated dUTP nick end labeling) assay was carried Cell Growth Medium (PrSCGM) supplemented with out using APO-DIRECT Kit (BD Biosciences, San Jose, PrSCGM SingleQuots (Lonza) and Prostate Epithelial CA, USA) according to the manufacturer’s instruction. Cell Growth Medium (PrEGM) supplemented with After labeling DNA breaks with FITC-dUTP, samples PrEGM SingleQuots (Lonza), respectively according were subjected to flow cytometric analysis using a FACS to the manufacturer’s instruction. LNCaP cells were Calibur flow cytometer (Becton Dickinson, San Jose, CA, maintained in RPMI-1640 supplemented with 10% USA). FCS, 100 U mlÀ1 of penicillin and 100 mgmlÀ1 of streptomycin. Cells (2 Â 105) for LNCaP or 1 Â105 cells for PrEC and Statistical analysis PrSC per each well were plated onto six-well culture Differences between treatment groups were assessed by plates, incubated for 24 h and used for assay. The PrEC Student’s paired t-test. Po0.05 was considered as and PrSC plated onto six-well culture plates were statistically significant. transiently transfected using FuGENE transfection re- agent (Roche) with 1.05 mg of DNA mixture containing pERE-luc reporter (1 mg) and pRL-TK (50 ng) in accor- dance with the manufacturer’s instructions and further Results grown for 24 h in PrECGM for PrEC or in PrSCGM for Effect of and an anti-estrogen toremifene on PrSC. LNCaP cells plated onto six-well culture plates were transiently transfected using FuGENE transfection proliferation of normal human prostate epithelial and reagent with pERE-luc (1 mg) and pRL-TK (50 ng) stromal cells according to the manufacturer’s instruction and further To access the importance of the estrogenic signaling in grown for 24 h in RPMI-1640 with 10% fetal bovine normal human prostate cells, DHT, estradiol (E2), serum dialyzed (10 000 MW cutoff, Sigma), penicillin and estrogenic chemicals and an anti-estrogen were admini- streptomycin. Then, b-estradiol or SERM was added to strated to prostate epithelial and stromal cells. As a the final concentration at 10 nM for b-estradiol or at 10 mM control, -dependent prostate cancer cells, for SERM, and cells were further cultured for another LNCaP, were used. In prostate epithelial cells, the 24 h followed by luciferase assay. proliferation was enhanced by adding E2, genistein (a

Prostate Cancer and Prostatic Diseases Effect of SERMs on normal human prostate cells H Nomura et al phyto-estrogen) or bisphenol A (BPA) compared with the To validate whether or not the negative growth effect 377 control cells. An anti-estrogen toremifene suppressed the of toremifene is due to apoptosis or blockade of the cell proliferation of epithelial cells (Figure 1a). In stromal proliferation, we carried out TUNEL (terminal deoxynu- cells, only DHT enhanced the proliferation. Suppression cleotidyl transferase-mediated dUTP nick end labeling) of the ER by toremifene reduced the growth of the cells assay with FITC-dUTP followed by flow cytometric remarkably in both prostate epithelial and stromal cells analysis. After incubation of prostatic stromal cells with indicating that the ER activation is necessary in the 10 mM of toremifene for 48 h, no peak of apoptotic nuclei growth of normal prostate cells (Figures 1a and b). In was detected by flow cytometric analysis: a single peak LNCaP cells, on the other hand, estrogenic stimulation of the normal DNA which was the same pattern as the seems to reduce cell proliferation: suppression of ER by control was observed (data not shown). toremifene did not reduce cell growth strongly (Figure 1c). Effect of SERMs on proliferation of normal human prostate cells Effect of varying toremifene concentration on growth As shown in Figures 3a and b, SERMs such as suppression of prostatic stromal cells toremifene, tamoxifen and raloxifene suppressed the As toremifene reduced the growth of prostatic stromal proliferation of prostate epithelial and stromal cells. cells remarkably, we varied its concentration and carried Faslodex did not reduce cell proliferation albeit tended to out cell-proliferation assays to define the optimum increase the epithelial cell growth; the possible explana- concentration. Toremifene suppressed the growth of tion is the agonistic property of faslodex in epithelial stromal cells dose-dependently. The suppressive growth cells. Anti-androgens, however, did not suppress the cell effect of toremifene was reduced to some extent by the growth in normal prostate cells. The observation that the longer incubation time (Figure 2). suppression of ER by SERMs, rather than that of AR by

Figure 1 Effect of hormones and an anti-estrogen, toremifene, on proliferation of normal human prostatic epithelial (a) and stromal cells (b) and prostate cancer cells LNCaP (c). 2 Â 105 cells for LNCaP or 1 Â105 cells for PrEC and PrSC per each well were seeded onto 6-well plates and incubated for 24 h. Then, DHT, b-estradiol, toremifene, genistein or BPA was added to the concentration of 10 nM for DHT and b-estradiol or 10 mM for toremifene, genistein and BPA and cells were further cultured for another 48 h, and subjected to WST-1 assay. The values represent the mean±s.e. of five independent transfections for PrEC and three for PrSC and LNCaP. *Po0.05, as compared with the control.

Prostate Cancer and Prostatic Diseases Effect of SERMs on normal human prostate cells H Nomura et al

378 esis of BPH,2,11 the ERb, through activation by its putative 5a-androstane-3b,17b-diol, was reported to suppress the growth of the ventral prostate and to decrease the AR content in rodents.21,22 Accordingly, there have been several studies describing that a loss of ERb expression is seen in high-grade prostate intrae- pithelial neoplasia and high-grade dysplasia compared with normal prostate epithelium expressing ERb.23–25 Recent experimental studies have demonstrated that ERb behaves as a tumor suppressor gene. Hurtado et al.26 for example showed that overexpression of ERb1 (ERb) induced a cell cycle arrest in the early G1 phase in LNCaP cells. Walton et al.27 showed that re-expression of ERb using a DNA demethylating agent and a histone deacetylase inhibitor caused increased apoptosis in prostate cancer cells. Thus, carcinogenesis in prostate seems to be characterized by a loss of ERb expression. However, conflicting results have been reported regard- ing the frequency of the ERb expression and its correlation to the grade in prostate cancer.24,25,28,29 With regard to ERa, recent investigations have emphasized the 30,31 Figure 2 Dose-suppression relationship of toremifene in prostatic important role of ERa in prostatic carcinogenesis. stromal cells. Cells were plated at 7 Â 103 cells per well onto 24-well Although those studies seem to illustrate the role of plates and incubated for 24 h. Then, toremifene was added to each ERs in the prostatic gland, an issue still remains whether concentration, and cells were further grown for 24, 48 and 72 h and the endogenous ERs will actually be activated by their subjected to WST-1 proliferation assay. The values represent the ligands followed by the ER-mediated induction of the ± mean s.e. (n ¼ 3). relevant genes. In this study, we demonstrated for the first time that anti-androgens, reduced the growth of normal prostate endogenous ERs in normal human prostate stromal cells stromal cells emphasizes the importance of estrogenic were activated by estrogen and suppressed by SERMs. signaling in normal prostate. In LNCaP cells, SERMs did We also showed that the growth of normal prostate cells not reduce cell proliferation strongly. Because LNCaP was suppressed by SERMs and not by anti-androgens, cells have a mutated AR at the codon 877 that is activated the weight of evidences indicating that the ER-mediated by flutamide and suppressed by , the result signaling is more relevant to the growth of normal that flutamide enhanced the growth of LNCaP cells and prostate cells than the AR-mediated one in the primary bicalutamide seemed to suppress its growth indicates the culture model. relevance of AR-mediated signaling in LNCaP cells. Concerning the effect of hormones on proliferation of human normal prostate cells, there have been several reports showing variable results. Some reported that DHT Effect of E2 and SERMs on ER activities in human normal or estradiol stimulated the growth of stromal cells 18,32–36 prostate cells and LNCaP cells whereas others published that they did not. In our To demonstrate the functionality of the endogenous ERs study, the proliferation of prostate epithelial cells was in normal prostate cells, the transactivation activities of significantly enhanced by adding E2, and that of stromal ERs were measured using pERE-luc as a reporter gene. cells by DHT although the effect was marginal. This may As pERE-luc is a reporter plasmid for both ERs a and b, be attributed to the serum and other additives in the the activities measured with each cell line represent the culture medium of these slow-growing cells. The suppres- sum of those of ERsa and b. The activities of ERs seemed sion study using SERMs and anti-androgens seemed to yield a good understanding of the relevance of the ER- to be enhanced by adding E2 and suppressed by toremifene and tamoxifen in prostate epithelial cells mediated signaling in normal human prostate cells. (Figure 4a). In prostate stromal cells, the activities of ERs The downstream genes of which ERs activate the transcription have not been studied well in normal were enhanced by adding E2 and reduced remarkably by 13 adding toremifene, and seemed to be suppressed by prostate cells. Zhang et al. showed that ERK was raloxifene whereas they were not suppressed by tamox- activated by estradiol with the proliferation of prostate ifen nor faslodex (Figure 4b). In LNCaP cells, the ER stromal cells and that a small interfering RNA of ERa blocked the estrogen-mediated activation of ERK and the activities were increased by E2 and tended to be 37 suppressed by all of the SERMs used (Figure 4c). proliferation of stromal cells. Chen et al. showed that ERa activity was enhanced by overexpressing ERAP75, a co-activator of ERa, with upregulation of the ERa target gene, stromal-derived factor-1, in normal prostate stro- Discussion mal cells transfected with ERa. Stromal-derived factor-1 has been reported to be associated with the Recently, much attention has been paid to the ERs a and and progression of prostate cancer.38 b in terms of their involvement in prostatic diseases. The ERb, originally isolated from rat prostate, has a Contrary to the ERa that is expressed at a higher level in variety of its isoforms, that is, splicing variants. Those stromal cells and seems to be involved in the pathogen- isoforms are expressed in an organ-specific manner, but

Prostate Cancer and Prostatic Diseases Effect of SERMs on normal human prostate cells H Nomura et al 379

Figure 3 Effect of SERMs and anti-androgens on proliferation of normal human prostate epithelial (a) and stromal cells (b) and LNCaP cells (c). Cells were plated onto 6-well plates as described in the Materials and methods section and incubated for 24 h. Then, SERMs and anti- androgens were added to the concentration of 10 mM and the cells were further grown for 48 h and subjected to WST-1 assay. The values represent the mean±s.e. (n ¼ 4). *Po0.05, as compared with the control. their function has not been well investigated, which may Of particular interest is a case report that a 56-year-old account for the complexity of the ERb signaling.39 male-to-female transsexual who had undergone sex- Accordingly, in our experiments, some SERMs func- changing surgery 25 years ago followed by continuous tioned as an antagonist whereas others tended to act as estrogen substitution presented with obstructive voiding an in normal human prostate stromal and symptoms. The patient was diagnosed as BPH and epithelial cells. treated with TUR-P with the histopathology showing an A few studies demonstrated the association of ERb ER expression and a strong AR expression.42 Probably, and the downstream genes regulated by ERb in prostate the gene amplification of AR has been caused by the low cancer cells. Leng et al.40 reported that ICI182,780 androgen level. One can imagine that continuous induced several key genes by activating ERb that bound stimulation of ER in stromal cells played an important to the upstream NFk consensus sequences. Matsumura role in the pathogenesis of BPH in the case. et al.41 showed that genistein, a phyto-estrogen from From a chemopreventive point of view, Price et al.43 soybean, induced p21 and ERb expression in PC-3 performed a using toremifene for the prostate cancer cells. Silencing ERb by siRNA sup- prevention of developing prostate cancer in men with pressed the p21 expression and the p21 promoter high-grade prostatic intraepithelial neoplasia, a precan- activity. Further study is necessary to elucidate the roles cerous lesion. They concluded that toremifene decreased of the ERs and the genes of which transcriptions are the incidence of developing prostate cancer in men with regulated by ERs in normal and malignant prostate cells. high-grade prostatic intraepithelial neoplasia. In our

Prostate Cancer and Prostatic Diseases Effect of SERMs on normal human prostate cells H Nomura et al 380

Figure 4 Effect of E2 and SERMs on endogenous ER activities in human normal prostate epithelial (a) and stromal cells (b) and LNCaP cells (c). PrEC and PrSC (1 Â105 cells per well) and LNCaP cells (2 Â 105 cells per well) plated onto six-well culture plates and incubated for 24 h were transiently transfected with the DNA mixture containing pERE-luc reporter (1 mg) and pRL-TK (50 ng) and further grown for 24 h. Then, b-estradiol or SERM was added to the final concentration at 10 nM for b-estradiol or at 10 mM for SERM, and the cells were further cultured for another 24 h followed by luciferase assay. The values represent the mean±s.e. (n ¼ 5). *Po0.05, as compared with the control (Student’s t-test).

study, toremifene showed a strong property of suppres- Conflict of interest sing ER activity and cell proliferation in prostatic stromal cells. The ability of toremifene to decrease the incidence The authors declare no conflict of interest. of prostate cancer may be attributed to the suppression of the stromal paracrine that stimulates the precancerous lesion. The possible use of SERMs for treating BPH may References be of further study. In conclusion, endogenous estrogen receptors were 1 Prins GS, Huang L, Birch L, Pu Y. The role of estrogens in normal functional in normal human prostate stromal cells, and and abnormal development of the prostate gland. Ann NY Acad their activities were reduced by toremifene accompanied Sci 2006; 1089: 1–13. by suppression in cell growth. 2 Prins GS, Korach KS. The role of estrogens and estrogen receptors in normal prostate growth and disease. 2008; 73: 233–244. 3 Kuiper GG, Enmark E, Pelt-Huikko M, Nilsson S, Gustafsson J-A. Acknowledgements Cloning of a novel receptor expressed in rat prostate and ovary. We thank Dr Shigeaki Kato, Tokyo University, for Proc Natl Acad Sci USA 1996; 93: 5925–5930. 4 Chang WY, Prins GS. Estrogen receptor b: implication for the providing us with pGL3-ERE-X3tK luc (pERE-luc). We prostate gland. Prostate 1999; 40: 115–124. also thank Nippon Kayaku Corp. (Tokyo, Japan) for 5 Lau K-M, LaSpina M, Long J, Ho S-M. Expression of estrogen flutamide and toremifene. This work was supported in receptor (ER)-a and ER-b in normal and malignant prosratic part by grants from Osaka City University Medical epithelial cells: regulation by methylation and involvement in Research Foundation. growth regulation. Cancer Res 2000; 60: 3175–3182.

Prostate Cancer and Prostatic Diseases Effect of SERMs on normal human prostate cells H Nomura et al 381 6 Latil A, Bieche I, Vidaud D, Lidereau R, Berthon P, Cussenot O cycle-dependent expression and regulates the G1 phase through et al. Evaluation of androgen, estrogen (ERa and ERb), and a non-genomic mechanism in prostate carcinoma cells. Cell progesterone receptor expression in human prostate cancer by Oncol 2008; 30: 349–365. real-time quantitative reverse transcription-polymerase chain 27 Walton TJ, Li G, Seth R, McArdle ES, Bishop MC, Rees RC. DNA reaction assays. Cancer Res 2001; 61: 1919–1926. demethylation and histone deacetylation inhibition co-operate to 7 Royuela M, de Miguel MP, Bethencourt FR, Sanchez-Chapado re-express and induce apoptosis in M, Fraile B, Arenas MI et al. Estrogen receptors a and b in prostate cancer cell-lines. Prostate 2008; 68: 210–222. the normal, hyperplastic and carcinomatous human prostate. 28 Torlakovic E, Lilleby W, Torlakovic G, Fossa SD, Chibber R. J Endocrinol 2001; 168: 447–454. Prostate carcinoma expression of estrogen receptor-b as 8 Tsurusaki T, Aoki D, Kanetake H, Inoue S, Muramatsu M, detected by PPG5/10 antibody has positive association with Nishikawa Y et al. Zone-dependent expression of estrogen primary Gleason grade and Gleason score. Hum Pathol 2002; 33: receptors a and b in human benign prostatic hyperplasia. J Clin 646–651. Endocrinol Metab 2003; 88: 1333–1340. 29 Walton TJ, Li G, McCulloch TA, Seth R, Powe DG, Bishop MC 9 Linja MJ, Savinainen KJ, Tammela TL, Isola JJ, Visakorpi T. et al. Quantitative RT-PCR analysis of estrogen receptor gene Expression of ERa and ERb in prostate cancer. Prostate 2003; 55: expression in Laser microdissected prostate cancer tissue. 180–186. Prostate 2009; 69: 810–819. Published online 2 Feb. 10 Kim IY, Kim B-C, Seong DH, Lee DK, Seo J-M, Hong YJ et al. 30 Singh PB, Matanhelia SS, Martin FL. A potential paradox in Raloxifene, a mixed estrogen agonist/antagonist, induces prostate adenocarcinoma progression: oestrogen as the initiating apoptosis in androgen-independent human prostate cancer cell driver. Eur J Cancer 2008; 44: 928–936. lines. Cancer Res 2002; 62: 5365–5369. 31 Ricke WA, McPherson SJ, Bianco JJ, Cunha GR, Wang Y, 11 Griffiths K, Denis LJ, Behre HE, Bracke M, Krieg M, Kyprianou Risbridger GP. Prostatic hormonal carcinogenesis is mediated N et al. Estrogens and prostatic disease. Prostate 2000; 45: 87–100. by in situ estrogen production and 12 King KJ, Nicholson HD, Assinder SJ. Effect of increasing ratio of signaling. FASEB J 2008; 22: 1512–1520. estrogen:androgen on proliferation of normal human prostate 32 Collins AT, Zhiming B, Gilmore K, Neal DE. Androgen and stromal and epithelial cells, and the malignant cell line LNCaP. oestrogen responsiveness of stromal cells derived from the Prostate 2006; 66: 105–114. human hyperplastic prostate: oestrogen regulation of the 13 Zhang Z, Duan L, Du X, Ma H, Park I, Lee C. The proliferative androgen receptor. J Endocrinol 1994; 143: 269–277. effect of estradiol on human prostate stromal cells is mediated 33 Zhang J, Hess MW, Thurnher M, Hobisch A, Radmayr C, through activation of ERK. Prostate 2008; 68: 508–516. Cronauer MV. Human prostatic smooth muscle cells in culture: 14 Krill D, Shuman M, Thompson MT, Becich MJ, Strom SC. A estradiol enhances expression of smooth muscle cell-specific simple method for the isolation and culture of epithelial and markers. Prostate 1997; 30: 117–129. stromal cells from benign and neoplastic . Urology 1997; 34 Planz B, Wang Q, Kirley SD, Lin C-W, McDougal WS. Androgen 49: 981–988. responsiveness of stromal cells of the human prostate: regulation 15 Peehl DM. Primary cell cultures as models of prostate cancer of cell proliferation and keratinocyte growth factor by androgen. development. Endocr Relat Cancer 2005; 12: 19–47. JUrol1998; 160: 1850–1855. 16 Peehl DM, Sellers RG. Cultured stromal cells: an in vitro model 35 Minamiguchi K, Kawada M, Someno T, Ishizuka M. Androgen- of prostatic mesenchymal biology. Prostate 2000; 45: 115–123. independent prostate cancer DU145 cells suppress androgen- 17 Planz B, Kirley SD, Wang Q, Tabatabaei S, Aretz HT, McDougal dependent growth of prostate stromal cells through production WS. Characterization of a stromal cell model of the human of inhibitory factors for androgen responsiveness. Biochem benign and malignant prostate from explant culture. JUrol1999; Biophys Res Commun 2003; 306: 629–636. 161: 1329–1336. 36 Nakano K, Fukabori Y, Itoh N, Lu W, Kan M, McKeehan WL 18 Janen M, Albrecht M, Mo¨schler O, Renneberg H, Fritz B, et al. Androgen-stimulated human prostate epithelial growth Aumu¨ller G et al. Cell lineage characteristics of human prostatic mediated by stromal-derived fibroblast growth factor-10. Endocr stromal cells cultured in vitro. Prostate 2000; 43: 20–30. J 1999; 46: 405–413. 19 Tran CP, Lin C, Yamashiro J, Reiter RE. Prostate stem cell antigen 37 Chen M, Ni J, Zhang Y, Muyan M, Yeh S. ERAP75 functions as a is a marker of late intermediate prostate epithelial cells. Mol coactivator to enhance estrogen receptor a transactivation in Cancer Res 2002; 1: 113–121. prostate stromal cells. Prostate 2008; 68: 1273–1282. 20 Garraway LA, Lin D, Signoretti S, Waltregny D, Dilks J, 38 Wang J, Shiozawa Y, Wang J, Wang Y, Jung Y, Pienta KJ et al. The Bhattacharya N et al. Intermediate basal cells of the prostate: role of CXCR7/RDC1 as a chemokine receptor for CXCL12/ in vitro and in vivo characterization. Prostate 2003; 55: 206–218. SDF-1 in prostate cancer. J Biol Chem 2008; 283: 4283–4294. 21 Weihua Z, Ma¨kela¨ S, Andersson LC, Salmi S, Saji S, Webster JI 39 Lewandowski S, Kalita K, Kaczmarek L. Estrogen receptor b: et al. A role for estrogen receptor bin the regulation of growth of potential functional significance of a variety of RNA isoforms. the ventral prostate. Proc Natl Acad Sci USA 2001; 98: 6330–6335. FEBS Lett 2002; 524: 1–5. 22 Weihua Z, Lathe R, Warner M, Gustafsson J. An endocrine 40 Leung Y-K, Gao Y, Lau K-M, Zhang X, Ho S-M. ICI 182,780- pathway in the prostate, ERb, AR, 5a-androstane-3b,17b-diol, regulated gene expression in DU145 prostate cancer cells is and CYP7B1, regulates prostate growth. Proc Natl Acad Sci USA mediated by estrogen receptor-b/NFkB crosstalk. Neoplasia 2006; 2002; 99: 13589–13594. 8: 242–249. 23 Horvath LG, Henshall SM, Lee C-S, Head DR, Quinn DI, 41 Matsumura K, Tanaka T, Kawashima H, Nakatani T. Involve- Makela S et al. Frequent loss of estrogen receptor-b expression in ment of the estrogen receptor b in genistein-induced expression prostate cancer. Cancer Res 2001; 61: 5331–5335. of p21 waf1/cip1 in PC-3 prostate cancer cells. Anticancer Res 24 Leav I, Lau K-M, Adams JY, McNeal JE, Taplin M-E, Wang J et al. 2008; 28: 709–714. Comparative studies of the estrogen receptors b and a and the 42 Casella R, Bubendorf L, Schaefer DJ, Bachmann A, Gasser TC, androgen receptor in normal human prostate glands, dysplasia, Sulser T. Does the prostate really need androgens to grow? and in primary and metastatic carcinoma. Am J Pathol 2001; 159: Transurethral resection of the prostate in a male-to-female 79–92. transsexual 25 years after sex-changing operation. Urol Int 25 Fixemer T, Remberger K, Bonkhoff H. Differential expression of 2005; 75: 288–290. the estrogen receptor b (ERb) in human prostate tissue, 43 Price D, Stein B, Sieber P, Tutrone R, Bailen J, Goluboff E et al. premalignant changes, and in primary, metastatic and recurrent Toremifene for the prevention of prostate cancer in men with prostatic adenocarcinoma. Prostate 2003; 54: 79–87. high grade prostatic intraepithelial neoplasia: results of a 26 Hurtado A, Pinos T, Barbosa-Desongles A, Lopez-Aviles S, double-blind, placebo controlled phase IIB clinical trial. JUrol Barquinero J, Petriz J et al. Estrogen receptor beta displays cell 2006; 176: 965–970.

Prostate Cancer and Prostatic Diseases