Vol 44#2

The Prostate 44:111–117 (2000)
Modulating Effect of Estrogen and Testosterone
on Prostatic Stromal Cell Phenotype
Differentiation Induced by Noradrenaline
and Doxazosin
Paul Smith,1* Nicholas P. Rhodes,2 Youqiang Ke,1 and Christopher S. Foster1 1Department of Cellular and Molecular Pathology, University of Liverpool, 2Department of Clinical Engineering, University of Liverpool, Liverpool, United Kingdom BACKGROUND. Noradrenaline (NA) has been shown to enhance expression of the contrac-
tile phenotype of human prostatic stromal cells in tissue culture. This study examined the
possibility that changing levels of sex hormones in elderly men with BPH may modulate the
differentiating effect of NA and hence the efficacy of ␣ -adrenoceptor-blocking drugs.
METHODS. Confluent, quiescent stromal cell cultures from 6 different patients were treated
with combinations of 20 ␮M NA, 1 ␮M doxazosin, 0.1 ␮M ␤-estradiol, and 0.1 ␮M testoster-
one, over a period of 10 days. Harvested cells were labelled with fluorescein-conjugated
antisera to ␣-smooth muscle actin and myosin to identify cells of contractile phenotype which
were thereafter analyzed flow-cytometrically.
RESULTS. NA increased mean immunoexpression of both actin and myosin. Enhancement of
myosin expression was highly significant (P Յ 0.02). This effect was incompletely opposed by
doxazosin. Neither estradiol nor testosterone influenced mean expression of contractile fila-
ments and did not significantly enhance or inhibit the effects of NA or doxazosin. However,
both sex hormones exhibited a differentially powerful effect on cell lines from individual
patients. The expression of myosin increased by NA was further elevated by addition of
estradiol in four of the cell lines and by testosterone in three.
CONCLUSIONS. The data suggest that levels of estrogens and androgens, either alone or in
combination, are unlikely to predict the development of obstructive symptoms in patients
with BPH or their response to doxazosin. Nevertheless, prostatic stromal cells from individual
patients may be exceptionally sensitive to both sex hormones, with enhanced modulation
towards a contractile phenotype. Since ␣- and ␤-subtypes of the estrogen receptor are differ-
entially expressed between the stroma and epithelium of the early fetal prostate, it is likely
that interaction between sex hormones and noradrenaline is an important factor in determin-
ing the phenotypic composition of prostatic stroma at this early stage of development, and
possibly predisposition to BPH during later adult life. Prostate 44:111–117, 2000.
2000 Wiley-Liss, Inc.
KEY WORDS:
human prostate; smooth muscle cells; flow cytometry; doxazosin; estro-
gen; androgen; noradrenaline; differentiation
INTRODUCTION
Grant sponsor: Pfizer Research International; Grant sponsor: Engi- Benign prostatic hyperplasia (BPH) is frequently as- neering and Physical Science Research Council.
sociated with symptomatic urinary outflow obstruc- *Correspondence to: Dr. Paul Smith, Department of Cellular and tion accompanied by urinary retention. Although Molecular Pathology, University of Liverpool, Liverpool L69 3GA, much of this obstruction may be due to physical oc- clusion of the urethra by the enlarged prostate gland, Received 12 May 1999; Accepted 10 March 2000 2000 Wiley-Liss, Inc.
Smith et al.
it is also exacerbated by contraction of stromal smooth is required for successful functional development of muscle cells surrounding the urethra. This contraction prostatic epithelium [23]. However, the role of estro- is mediated, in part, by noradrenaline which binds genic stimulation is complex with respect to prolifera- ␣1A-adrenoceptors expressed on plasma membranes tion and to expression of other differentiation pro- of stromal smooth muscle cells and myofibroblasts teins, including the androgen receptor. Depending on [1–6]. In recent years, ␣1-adrenoceptor antagonists the stage of intrauterine growth and relative levels of have been employed clinically to successfully oppose hormone achieved, estrogen exhibits a biphasic re- the influence of noradrenaline (NA) and hence im- sponse on prostatic enlargement in mice. While low concentrations modulate the actions of androgen to Recently, we showed by in vitro studies that nor- stimulate a 40% growth in epithelium through a two- adrenaline not only induces contraction of prostatic fold induction of androgen receptors, high concentra- stromal cells but also induces expression of a more tions of estrogen cause a reduction in prostatic size contractile cell-type [10]. Such phenotypic modulation is most pronounced when cells are plated out sparsely Since estrogen and noradrenaline, independently, and are in a logarithmic phase of active proliferation.
appear to influence the phenotype of prostatic stromal Furthermore, addition of the ␣1-adrenoceptor antago- cells in a similar manner, this current study was per- nist, doxazosin, to the medium largely reverses this formed to test the hypothesis that a synergistic rela- phenotypic modulation [11], particularly when the tionship between these two humoral agents may re- cells are confluent and proliferatively quiescent, as in sult in enhanced differentiation of a population of the intact hyperplastic gland. These findings suggest stromal cells. Following our previous observation that that ␣1-adrenoceptor antagonists may not only inhibit the selective ␣1 adrenoceptor antagonist, doxazosin, contraction of prostatic stroma, but may also reduce inhibits, or partially reverses, the differentiating effect its initial capacity to contract. However, despite giving of noradrenaline, we predicted that any synergistic valuable insight into one important aspect of the in- estrogen effect should be diminished in its presence.
teraction between noradrenaline and prostatic stromal Although androgen receptors are found in the stroma cells, these particular experiments did not address the as well as the glandular epithelium [25,26], little is possible modulating role of sex hormones on the re- currently known about the effect of testosterone on the sponse of hyperplastic prostatic tissues to NA. It is phenotype of stromal cells. Therefore, the final aim of generally acknowledged that an imbalance between this study was to analyze the interactive effect of tes- the relative levels of estrogens and androgens, conse- tosterone with noradrenaline and doxazosin on the quent upon a declining secretion of testosterone from phenotypic differentiation of cultured human prostat- middle age onwards, is an important factor in the eti- ic stromal cells, as assessed by flow cytometry. The data from this study confirm the initial hypothesis that Studies of rat early fetal prostate have shown that a synergistic effect between estrogen and noradrena- estrogen receptor (ER) subtypes are differentially ex- line modulates the differentiated phenotype of stro- pressed between the epithelium and stroma [16], such mal cells derived from adult human prostate. How- that ER␣ is expressed by stromal mesenchymal cells, ever, this interaction is not one of simple synergy but while ER␤ is differentially expressed by epithelial cells is complex and likely to be influenced by other factors.
within different lobes of the gland. In the nonneoplas-tic mature adult prostate, estrogen receptors are lo- MATERIALS AND METHODS
cated almost exclusively in stromal cells which are,therefore, able to respond to any relative estrogen ex- Prostatic tissues were obtained from 6 patients who cess [13,17–20]. One such response in vitro is an in- had undergone transurethral prostatic resection for creased density of intracellular desmin and myosin BPH. Stromal cells were isolated from the tissues by filaments [21]. Stromal estrogen receptors also modu- collagenase digestion, as previously described [10].
late the mitogenic effects of estrogen on prostatic epi- Cells were grown in RPMI-1640 medium supple- thelium [22]. Neonatal exposure to low-dose estrogen mented with 10% fetal calf serum (FCS), 2 mM L- regulates estrogen receptor expression in the develop- glutamine, 100 units/ml penicillin, and 0.1 mg/ml ing and adult rat prostate stroma and epithelium [15], streptomycin (Gibco, Life Technologies, Paisley, Scot- thus providing mechanisms for genetic imprinting land, UK) in a humidified atmosphere of 5% CO2 and or for possible development of hyperplastic/prolifera- air at 37°C. Cells were passaged 4–5 times to provide tive diseases in later life. In fibroblasts known to ex- six different cell lines for study. Cells from each line press ER, exposure to estrogen increases levels of the were seeded into 16 tissue culture dishes (9-cm diam- androgen receptor. Furthermore, functional interac- eter) at a density of approximately 6 × 103 cells/cm2 tion with androgen receptor-expressing stromal cells and maintained in the above growth medium until Sex Hormones and Stromal Differentiation
TABLE I. Mean Fluorescence Intensities of Alpha-Smooth Muscle Actin and Myosin Expressed by Six Prostatic
Stromal Cell Lines†
†C, untreated control; D, doxazosin at 1 ␮M; T, testosterone at 0.1 ␮M; N, noradrenaline at 20 ␮M; E, estradiol at 0.1 ␮M. Numbersin parentheses are ± one standard deviation from the mean.
*Significantly different from control (C).
**Significantly different from noradrenaline (N); P < 0.05.
they had proliferated to confluence. The medium was Cells from each treatment were then divided equally then replaced with 10 ml RPMI-1640 containing 2 mM into two microcentrifuge tubes and rehydrated with L-glutamine, antibiotics, and 1% (v/v) dialyzed and PBS for 30 min, and the supernatants were replaced carbon-stripped FCS (to remove exogenous steroid with 200 ␮l of blocking buffer, comprising PBS con- hormones) to render cells quiescent. The dishes were taining 1% (w/v) bovine serum albumin, and left then divided into four groups of four dishes each, to overnight. After centrifugation, supernatants were re- which noradrenaline bitartrate (Arterenol, Sigma placed with 100 ␮l of monoclonal antibodies to either Chemical Co., Poole, Dorset, UK), ␤-estradiol, testos- ␣-smooth muscle actin (Sigma clone 1A4, dilution 1: terone (Sigma), and doxazosin mesylate (Cardura௡, 640) or to smooth muscle myosin (Sigma clone hSM-V, Pfizer, Ltd., Sandwich, Kent, UK) were added with a dilution 1:200). Cells were incubated at 37°C for 40 micropipette, according to the scheme summarized in min with periodic agitation to maintain cells in a dis- Table I. The final concentrations of each substance in persed state. The antisera were then withdrawn and the medium in each case were as follows: noradrena- the cells were washed in 1 ml of blocking buffer for 10 line, 20 ␮M; ␤-estradiol, 0.1 ␮M; testosterone, 0.1 ␮M; min, followed by the addition of 100 ␮l of sheep anti- and doxazosin, 1.0 ␮M. This is higher than the plasma (mouse IgG) conjugated to fluorescein isothiocyanate, concentration of approximately 0.075 ␮M of doxazosin diluted 1:250 in blocking buffer. Incubation was for a in a man taking 4 mg of the drug daily [27].
further 30 min at 37°C, after which supernatants were An extra dish of untreated cells was reserved as a replaced with 200 ␮l PBS. Negative control cells from negative control for subsequent immunolabeling. The each cell line were treated with an antibody of irrel- medium, hormones, and doxazosin were replaced on evant specificity but of the same isotype as the pri- alternate days for a total of 10 days.
mary antibody, followed by the fluorescent secondary Cells were released from each of the 102 dishes by exposure to trypsin/EDTA for 5 min, followed by agi- The fluorescence intensity of each sample of cells tation until a monodisperse suspension of cells was was measured in a Becton Dickinson FACsort flow obtained. Trypsin was inactivated by addition of 1 ml cytometer (San Jose, CA) in which the sensitivity was FCS, and the cells washed with phosphate-buffered adjusted so that 95% of the negative controls fell saline (PBS) and centrifuged, and the pelleted cells within the first log decade. A total of 20,000 events resuspended in 1 ml of 70% (v/v) chilled methanol.
was counted, with electronic noise and subcellular de- Smith et al.
TABLE II. Percentage Increases in Fluorescence Above Baseline for Actin and Myosin in Each Stromal Cell Line
Treated With Noradrenaline, Estradiol, and Testosterone
Percent increase in filament expression for cell lines: bris excluded by setting a threshold on forward light hormones (TE) to the medium induced negligible scatter. Data were recorded as frequency distribution changes to mean fluorescence of either actin or myo- histograms and as mean fluorescence intensity units sin, as shown by ratios close to unity. Similarly, both derived from cells with a fluorescence greater than hormones exhibited little effect on the ability of nor- that of 95% of the negative control cells. Data were adrenaline (N) to alter the expression of myofilaments.
analyzed statistically by ANOVA and paired t-tests, Thus, although EN, TN, and TEN were all higher than and were considered to be statistically significantly the control (significantly so for myosin), in no instance were they significantly different from N or from oneanother.
Doxazosin alone (D) was associated with reduced expression of both myofilaments, which was signifi- Flow cytometric fluorescence intensities expressed cant for mysoin and also for the ratio of actin:control.
as the mean of the six stromal cell lines are shown in Similar reductions occurred in groups 2 and 3 (ED and Table I. To aid comparison, the data are also normal- TD) and were significant in several instances (Table I).
ized by expressing each fluorescence as a ratio of the In group 4 (TED), there was little change from control untreated control (C). Analysis of variance (ANOVA) values. Doxazosin in conjunction with noradrenaline was performed on the data derived from the four (ND) induced a significantly higher immunofluores- treatments collectively in each experimental group, as cence than control values (C) for both actin and myo- defined in Table I. For cells labelled for ␣-smooth sin. Similar increases were obtained from groups 2 muscle actin, differences in the variance of fluores- and 3 (NED and NTD). These values, particularly cence intensity were not significant and, when ex- those for myosin, were lower than the corresponding pressed as a ratio of control, were significant only in values obtained from noradrenaline in the absence of group 1. Similarly, ANOVA of myosin fluorescence doxazosin (EN and TN), indicating that the drug had intensity was not significant but, when expressed as a incompletely reversed the influence of noradrenaline ratio of control, all four of the experimental groups on myofilament expression. However, only in two in- showed a significant difference (group 1, P < 0.001; stances (ND, actin ratio; and TEND, myosin fluores- group 2, P < 0.05; group 3, P < 0.025; group 4, P < 0.01).
cence) were these values significantly lower than those The data were also analyzed by paired t-tests which for noradrenaline alone (N). Sex hormones did not compare the arithmetic differences between related significantly enhance or inhibit the influence of dox- pairs of data. The results of this analysis are indicated azosin plus noradrenaline, despite the fact that testos- in Table I. Addition of noradrenaline to the medium terone (NTD and TEND) appeared to prevent doxazo- increased the mean fluorescence of ␣-smooth muscle sin from reducing the effect of noradrenaline on actin above control values in each instance, although this was statistically significant only for group 1. For Although sex hormones had little effect on mean myosin the increases were more pronounced and sig- fluorescence values from the six cell lines, they had a considerable effect on myofilament expression by Addition of estradiol (E), testosterone (T), or both some individual cell lines. Table II shows the percent- Sex Hormones and Stromal Differentiation
age increases above baseline (C, E, T, and TE) for each minimal dose dependency over the concentration stromal cell line treated with noradrenaline in con- range of 0.001–10.0 ␮M [11], suggesting that the drug junction with estradiol, testosterone, and both hor- may have an action other than that of a simple adre- mones, respectively (N, EN, TN, and TEN). In cell line noceptor antagonist. This possibility is supported by C there was a 28% increase in fluorescence of actin the observation that doxazosin, when administered to induced by noradrenaline alone, which was raised to patients with BPH, enhances apoptosis of prostate 34% in the presence of estradiol, to 45% with testos- cells, particularly those of the stroma [28,29]. Also, terone, and further to 50% with both hormones. The mitogenesis and migration of vascular smooth muscle increased expression of myosin was further raised by in culture are inhibited by doxazosin, even when there addition of estradiol in four cell lines (A, C, D, and F) is prior ␣1-adrenoceptor blockade by phenoxybenza- and by testosterone in three cell lines (A, B, and C).
Cell line A in particular showed a 122% increase in Recently, Boesch et al. [31] demonstrated a change myosin expression with estradiol and a 139% increase in the relative proportions of mRNA encoding the two with testosterone, as contrasted with a 42% increase isoforms of smooth muscle type myosin heavy chain with noradrenaline alone. In other cell lines, sex hor- (SM1 and SM2) in cultured prostatic stromal cells ex- mones failed to augment, or even reduced, the influ- posed to doxazosin for 6 days. Thus, the ratio of SM2:SM1 was increased 10-fold, suggesting a changefrom a secretory to a more contractile phenotype. Thisfinding is at variance with our own observations and DISCUSSION
also with the authors’ earlier paper [32], both of which The data from this study confirm our previous re- imply that doxazosin reduces the contractility of stro- ports that noradrenaline promotes expression of con- mal smooth muscle. The anti-smooth muscle myosin tractile filament proteins in prostatic stromal smooth antibody which we employed does not distinguish be- muscle cells [10,11]. The study further supports our tween SM1 and SM2, but demonstrates instead a re- observation that the ␣1-adrenoceptor antagonist, dox- duction in total myosin immunoexpression induced azosin, reduces expression of myofilaments actin and by doxazosin. Furthermore, Boesch et al. [31] demon- myosin within these cells [11]. Doxazosin, in the ab- strated that stimulation of stromal adrenoceptors by sence of noradrenaline, induced a fall in the mean phenylephrine for 6 days had a negligible effect on the expression of actin and myosin, an effect which was myosin heavy-chain ratio, though we have shown re- greater and significant for the latter filament protein in peatedly in this and previous papers [10,11] that nor- experimental groups 1–3 (see Table I). This reduction adrenaline consistently increases the quantity of myo- in contractile filaments was greater than that recorded sin in stromal cells. Clearly, the SM2:SM1 ratio and the in our earlier study [11], in which neither actin nor total myosin content are independently modulated by myosin showed a significant fall. Doxazosin also par- tially reversed the increase in myofilament expression In the present study, additional modulating effects induced by noradrenaline in most of the experimental of estrogen and androgen were tested concurrently groups (the exceptions being NTD and TEND labeled with noradrenaline and doxazosin. Since the concen- for actin). However, this response was incomplete and trations of sex hormones change during the aging pro- varied greatly within the six cell lines examined, so cess [12–14], we reasoned that they may modify the that statistical significance was reached only in two influences of noradrenaline and doxazosin on the instances (Table I). In a previous study of 10 stromal prostatic stromal cell phenotype. This effect appeared cell lines, we found a greater and more consistent re- likely following the report by Zhang et al. that estra- sponse to doxazosin, such that the level of actin fluo- diol (and to a lesser extent testosterone) increases the rescence and the ratios of actin and myosin to control density of myosin and desmin filaments in cultured values were significantly lower than their correspond- human prostatic stromal smooth muscle cells [21]. We ing levels with noradrenaline alone [11]. These find- found no such enhanced fluorescence as assessed by ings suggest that noradrenaline may enhance the con- flow cytometry in the mean values from six stromal tractile phenotype of prostatic stromal cells in BPH, cell lines, despite the use of identical concentrations with the potential effect of increasing their contractile (0.1 ␮M) of estradiol and testosterone. However, an force and thus exacerbating urethral obstruction.
important methodological difference may be respon- Moreover, doxazosin may not only inhibit noradren- sible for this apparent discrepancy. Zhang et al. [21] ergic contraction of stromal smooth muscle cells but, added hormones while their cultures were in a prolif- by reducing their quantity of myofilaments, may re- erative phase and then fixed them at confluence. In duce their original capacity to contract. The effect of our experimental model, cells were allowed to reach doxazosin on the expression of myofilaments shows a confluence before the addition of hormones, and were Smith et al.
then prevented from any further growth by replacing program grant on biocompatibility. We thank Mrs. Jill the serum in the medium with 1% dialyzed and car- bon-stripped FCS. The purpose of this treatment wasto more closely simulate the normal physiological REFERENCES
state of stromal smooth muscle cells in the intact hy-perplastic prostate in which the stromal cells are com- 1. Andersson KE, Lepor H, Wyllie H, Wyllie M. Prostatic alpha1- pact and largely nonproliferating. Under these condi- adrenoceptors and uroselectivity. Prostate 1997;30:202–215.
tions, neither estradiol nor testosterone exhibited any 2. Chapple CR, Burt RP, Andersson PO, Greengrass P, Wyllie M, significant influence on the phenotype of prostatic Marshall I. Alpha-1-adrenoceptor subtypes in the human pros- stromal cells, either with or without the addition of 3. Lepor H, Tang R, Kobayashi S, Shapiro E, Forray C, Wetzel JM, noradrenaline. Furthermore, the two hormones failed Gluchowski C. Localization of the alpha 1A-adrenoceptor in the to affect the efficacy of doxazosin to modulate the phe- human prostate. J Urol 1995;154:2096–2099.
notype of stromal cells to a less contractile form. With 4. Marshall I, Burt RP, Chapple CR. Noradrenaline contractions of respect to an individual patient with BPH, our results human prostate mediated by alpha 1A-(alpha 1c-) adrenoceptor suggest that relative levels of androgens and estrogens subtype. Br J Pharmacol 1995;115:781–786.
5. Price DT, Schwinn DA, Lomasney JW, Allen LF, Caron MG, are unlikely to be useful predictors of the develop- Lefkowitz RJ. Identification, quantification, and localization of ment of obstructive symptoms, or for the response to mRNA for three distinct alpha 1 adrenergic receptor subtypes in human prostate. J Urol 1993;150:546–551.
Although our data and conclusions are based on 6. Tseng-Crank J, Kost T, Goetz A, Hazum S, Robertson KM, Haiz- mean values obtained from six cell lines, there was lip J, Godinot N, Robertson CN, Saussy D. The alpha 1c-adrenoceptor in human prostate: cloning, functional expression, considerable variation in response between individual and localization to specific prostatic cell types. Br J Pharmacol cell lines, with some showing a dramatic increase in myofilament expression in response to estradiol and/ 7. Chapple CR. Selective alpha 1-adrenoceptor antagonists in be- or testosterone together with noradrenaline, while in nign prostatic hyperplasia: rationale and clinical experience. Eur others there was little change or even a negative re- 8. Hieble JP, Ruffolo RR. The use of alpha-adrenoceptor antago- sponse (Table II). Hence, although patients with BPH nists in the pharmacological management of benign prostatic may not show a consistent relationship between levels hypertrophy: an overview: Pharmacol Res 1996;33:145–160.
of sex hormones and expression of prostatic stromal 9. Lepor H, Kaplan SA, Klimberg I, Mobley DF, Fawzy A, Gaffney cell myofilaments, there may be some individuals in M, Ice K, Dias N. Doxazosin for benign prostatic hyperplasia: whom sex hormones are of profound importance.
long-term efficacy and safety in hypertensive and normotensivepatients. The Multicenter Study Group. J Urol 1997;157:525–530.
However, a similar or greater response to testosterone 10. Smith P, Rhodes NP, Beesley C, Ke Y, Foster CS. Prostatic stro- was obtained in those stromal cell cultures in which mal cell phenotype is directly modulated by norepinephrine.
estrogen stimulated an enhanced response to nor- adrenaline. In such individuals, predisposition to cer- 11. Smith P, Rhodes NP, Ke Y, Foster CS. Influence of the alpha1- tain responses may reflect cellular and molecular adrenergic antagonist, doxazosin, upon noradrenaline-induced changes occurring during fetal development, includ- modulation of cytoskeletal proteins in cultured hyperplasticprostatic stromal cells. Prostate 1999;38:216–227.
ing exposure to different levels of estrogen, resulting 12. Coffey DS, Walsh PC. Clinical and experimental studies of be- in differential genetic imprinting. Thus, prostatic stro- nign prostatic hyperplasia. Urol Clin N Am 1990;17:461–475.
mal cells from some patients with BPH may be highly 13. Habenicht UF, el-Etreby MF. Rationale for using aromatase in- sensitive to sex hormones and may show enhanced hibitors to manage benign prostatic hyperplasia. Experimental modulation towards a contractile phenotype under studies. J Androl 1991;12:395–402.
14. Krieg M, Nass R, Tunn S. Effect of aging on endogenous level of noradrenergic stimulation. Our results suggest that a 5 alpha-dihydrotestosterone, testosterone, estradiol and estrone shift in hormone balance towards estrogen excess is in epithelium and stroma of normal and hyperplastic human unlikely to be a significant contribution to this pheno- prostate. J Clin Endocrinol Metab 1993;77:375–381.
typic modulation, and that other humoral factors are 15. Prins GS, Birch L. Neonatal estrogen exposure up-regulates es- probably of greater importance, such as those derived trogen receptor expression in the developing and adult rat pros- tate lobes. Endocrinology 1997;138:1801–1809.
16. Prins GS, Marmer M, Woodham C, Chang W, Kuiper G, Gus- tafsson JA, Birch L. Estrogen receptor-beta messenger ribo-nucleic acid ontogeny in the prostate of normal and neonatally ACKNOWLEDGMENTS
estrogenized rats. Endocrinology 1998;139:874–883.
17. Ehara H, Koji T, Deguchi T, Yoshii A, Nakano M, Nakane PK, C.S.F. thanks Pfizer Research International for con- Kawada Y. Expression of estrogen receptor in diseased humanprostate assessed by nonradioactive in situ hybridization and tinued generous support of this project. N.P.R. ac- immunohistochemistry. Prostate 1995;27:304–313.
knowledges the support of the Engineering and Physi- 18. Kirschenbaum A, Ren M, Erenburg I, Schachter B, Levine AC.
cal Science Research Council through a rolling Estrogen receptor messenger RNA in human benign prostatic Sex Hormones and Stromal Differentiation
hyperplasia: detection, localization and modulation with a long- 26. Suzuki K, Ito K, Suzuki T, Honma S, Yamanaka H. Synergistic acting gonadotrophin-releasing hormone agonist. J Androl effects of estrogen and androgen on the prostate: effects of es- trogen on androgen–and estrogen-receptors, Brd U uptake, im- 19. Schulze H, Claus S. Histological localization of estrogen recep- munohistochemical study of AR, and responses to antiandro- tors in normal and diseased human prostates by immunocyto- chemistry. Prostate 1990;16:331–343.
27. Frick MH, Halttunen P, Himanen P, Huttunen M, Po¨rsti P, Pit- 20. Zhao GQ, Holterhus PM, Dammshauser I, Hoffbauer G, Au- tajarvi T, Poyhonen L, Pyykonen M-L, Reinikainen P, Salmela P, muller G. Estrogen-induced morphological and immunohisto- Saraste M. A long-term double-blind comparison of doxazosin chemical changes in stroma and epithelium of rat ventral pros- and atenolol in patients with mild to moderate essential hyper- tension. Br J Clin Pharmacol 1986;21:55S–62S.
21. Zhang J, Hess MW, Thurnher M, Hobisch A, Radmayr C, Cro- nauer MV, Hittmair A, Culig Z, Bartsch G, Klocker H. Human 28. Kyprianou N, Litvak JP, Borkowski A, Alexander R, Jacobs SC.
prostatic smooth muscle cells in culture: estradiol enhances ex- Induction of prostate apoptosis by doxazosin in benign prostatic pression of smooth muscle cell-specific markers. Prostate 1997; hyperplasia. J Urol 1998;159:1810–1815.
29. Chon JK, Borkowski A, Partin AW, Isaacs JT, Jacobs SC, Kypri- 22. Cooke PS, Buchanan DL, Young P, Setiawan T, Brody J, Korach anou N. Alpha 1-adrenoceptor antagonists terazosin and dox- KS, Taylor J, Lubahn DB, Cunha GR. Stromal estrogen receptors azosin induce prostate apoptosis without affecting cell prolif- mediate mitogenic effects of estradiol on uterine epithelium.
eration in patients with benign prostatic hyperplasia. J Urol Proc Nat Acad Sci USA 1997;94:6535–6540.
23. Blanchere M, Berthaut I, Portois MC Mestayer C, Mowszowicz 30. Hu ZW, Shi XY, Hoffman BB. Doxazosin inhibits proliferation I. Hormonal regulation of the androgen receptor expression in and migration of human vascular smooth–muscle cells indepen- human prostatic cells in culture. J Steroid Biochem Mol Biol dent of alpha 1-adrenergic receptor antagonism. J Cardiovasc 24. vom Saal FS, Timms BG, Montano MM, Palanza P, Thayer KA, Nagel SC, Dhar MD, Ganjam VK, Parmigiani S, Welshons WV.
31. Boesch ST, Corvin S, Zhang J, Rogatsch H, Bartsch G, Klocker H.
Prostate enlargement in mice due to fetal exposure to low doses Modulation of the differentiation status of cultured prostatic of estradiol or diethylstilbestrol and opposite effects at high smooth muscle cells by an alpha 1-adrenergic receptor antago- doses. Proc Nat Acad Sci USA 1997;94:2056–2061.
25. Sar M, Lubahn DB, French FS, Wilson EM. Immunohistochem- 32. Corvin S, Boesch ST, Eder I, Thurnher M, Bartsch G, Klocker H.
ical localization of the androgen receptor in rat and human tis- Videoimaging of prostatic stromal-cell contraction: an in vitro sues. Endocrinology 1990;127:3180–3186.
model for studying drug effects. Prostate 1998;37:209–214.

Source: http://www.nickrhodes.org/nick/Smith%202000.pdf