Biochimica et Biophysica Acta, 1094 (1991) 103-112 © 1991 Elsevier Science Publishers B.V. 0167-4889/91/$03.50 ADONIS 016748899100219G

103

BBAMCR 12974

Effects of steroidal and non-steroidal antiandrogens on the androgen binding properties of the rat ventral prostate androgen receptor Jaime Steinsapir, Gloria Mora and Thomas G. Muldoon Department of Physiology and Endocrinology, Medical Collegeof Georgia, Augusta, GA (U.S.A.)

(Received 10 December 1990)

Key words: Antiandrogen; Androgen receptor; Prostate; Cycloheximide

Steroidal (cyproterone acetate) and non-steroidal (RU23908 and hydroxyflutamide) antiandrogens are able to block testosterone-induced increases in nuclear androgen receptor (All) in the prostate of l-day orchidectomized rats, but when given alone, RU23908 and hydroxyflutamide increase nuclear AR (RU23908 > hydroxyflutamide) in the same animal model. The increases in nuclear AIR induced by antiandrogen alone or with testosterone alone are blocked by cycloheximide ! h after administration, suggesting that androgen or antiandrogens induce de novo AR synthesis. Concomitant to nuclear AIR accumulation, testosterone is able to induce depletion of cytosol and microsomal AIR. Blockade of testosterone-induced depletion of microsomal AR, but not of cytosoi AR, occurs in the presence of antiandrogens. Cyproterone acetate has a higher relative binding affinity (RBA) for microsomal AR and cytosol AIR than RU23908 or hydroxyflutamide. This phenomenon is in good agreement with the degree of inhibition by these compounds of the association rate of androgen for the microsomal AR. This correlation between RBA and inhibition of the initial rate of hormone binding to the receptor is not found for cytosol AIR. The results show that antiandrogens are not 'pure' antagonists of androgen action and they are potent agonists in the absence of testosterone. Furthermore, testosterone alone or antiandrogens per se regulate AIR levels acutely by protein-synthesis dependent mechanisms of action, in rat ventral prostate.

Introduction

Co-existence of three distinct androgen receptor (AR) forms have been described in ventral prostate [1]. These AR forms are associated with the microsomal, eytosolic and nuclear compartments of target cells. AR is necessary for androgen action in responsive tissues and cells [2]. Each AR form may mediate different responses in prostate [3].

Abbreviations: AR, androgen receptor; DMNT, 17~-hydroxy7a,17u-dimethyl-4-estren-3-one, mibolerone; PMSF, phenylmethylsulfonylfluoridc; RU23908, 5,5-dimethyl-3-(4-nitro-3-(triflu~romethyl)phenyl)-2,4-imidazolidinedione, tAnandron®); cyproterone acetate, 6a-chloro- 17a-hydroxy-1 ~,2a-methylene-4,6-pregna-diene3,20-dione-17-acetate; hydroxyflutamide, 2.hydroxy-2-methyI-N-(4nitro-3-(trifluoromethyl)phenyl)-propanamide; (SCH-16423), RBA, relative binding affinity. Correspondence: J. Steinsapir, Department of Physi:logy and Endocrinology, Medical College of Georgia, Augusta, Georgia 30912, U.S.A.

Antiandrogens are of widespread use in the treatment of prostate cancer [4]. They are described as antiandrogens mainly by the criteria of antagonism with biological responses to androgens. Current understanding of the mechanisms of interference of androgen action by these compounds involve AR [5]. However, little is known on the relative binding properties of each AR form, their relative responsiveness to androgen and antiandrogen as well as the mechanisms of antiandrogen action. The aims of this study were: (a) to study the effects of RU23908, hydroxyflutamide and cyproterone acetate on the microsomal, eytosol and nuclear AR levels, in the presence or absence of testosterone, in rat ventral prostate; (b) to assess the effects of these antiandrogens on the relative binding affinities, association rate and dissociation rate kinetics of androgen from microsomal and cytosol AR; and (c) to determine whether the mechanisms by which antiandrogens control AR levels are protein-synthesis dependent by the analysis of the effects of eycloheximide on the relative sensitivity of nuclear, cytosol and microsoma! AR to androgen a n d / o r antiandrogens.

104 Materials and Methods

Materials [17a.methyl.3H]Mibolerone (7a,17a-dimethyl-19[17a.methyl.3H]nortestosterone, DMNT; 85 Ci/mmol) and radioinert DMNT were purchased from Amersham, Searle. Cyproterone acetate (6a-chloro-17a-hydroxy-I a,2 a-methylene-4,6-pregnadiene-3,20 dione17-acetate) was from Schering AG, Berlin, F.R.G. RU23908 (5,5.dimethyl-3-(4-nitro-3-(trifluoro-methyl) phenyl)-2,4-imidazolidinedione, Anandron ®) was from Roussel-UCLAF, Paris, France. Hydroxyflutamide (2hydroxT.2.methyl.N-[4-nitro-3-(trifluoro-methyl)phenyl]propanamide, SCH-16423) was from Schering Corp., Bloomfield, N J, U.S.A.). Dextran-coated charcoal was prepared with Norit A (0.5%, Amend Drug and Chem. Co., Irvington, NJ, U.S.A.) and Dextran (0.05%, Nutritional Biochem. Co., Cleveland, OH, U.S.A,). Biogel-HTP was from Bio.Rad (Richmond, CA, U.S.A.). Triton X-100 was from Sigma. All other reagents were of analytical grade.

Animals and preparation of sources of binding activity Male rats were obtained at 120 days of age (400-440 g body wt.) from Holtzman (Madison, WI, U.S.A.) and were a!iowed a 5 day period of acclimatization to a 12 h light-dark cycle. Bilateral orchidectomy was performed through the scrotal route under diethyl ether anesthesia 24 h prior to each experiment. Animals were killed by decapitation, and individual ventral prostates were collected in ice.cold homogenization buffer (10 mM Tris, 1.5 mM Na2EDTA, 0.5 mM dithiothreitol, 0.25 M sucrose, 10 mM Na2MoO 4, 1 mM PMSF, pH 7.4 at 22 ° C), rinsed and minced with scissors prior to homogenization by three 15-s bursts and 30-s cooling periods, with a Polytron PT-10 homogenizer, at a (tissue/buffer) ratio of 500 mg/ml (1 ml/prostate). The homogenate was centrifuged at 800 × g for 20 rain. Nuclei was prepared by three different methods from the 800 × g pellet. In Method I, modified from Blobel and Potter [6], nuclei were resuspended in 'nuclear' buffer (10 mM Tris, 0.5 mM dithiothreitol, 0.25 M sucrose, 1 mM PMSF, 2.5 mM MgCi2-61-120, pH 7.4 at 220C) at 0.5 ml/prostate. The nuclei suspension was adjusted to 0.88 M and layered on top of a 1.7 M sucrose pad prepared in the same buffer. The volume ratio between top and bottom layers was 0.75/1.0 to allow optimal nuclei migration [6] during centrifugation. The samples were spun down for 60 rain at 60000 × g (50.2 Ti rotor, L8-80 Beckman), and the white nuclear pellet is resuspended in 'nuclear' buffer [6]. In Method II [7], nuclei were washed at 4500 × g for 10 rain in 'nuclear" buffer with 1% Triton X-100, at 1 ml/prostate. A second wash in the same buffer without detergent was performed at 27 000 × g

for 10 min and the resuspended nuclear material was used as source of binding activity. In method III, modified from Allfrey et al. [8], the crude 800 × g nuclear pellet was resuspended in 'nuclear' buffer at 1 ml/prostate and unbroken cells and cell debris were allowed to settle for 10 min at 4°C. The nuclear suspension was then decanted through a double layer of gauze and centrifuged at 400 × g for 7 rain. An additional wash was performed in the same buffer at 700 × g for 20 min and the final resuspended material was used for nuclear AR assay. The 800 x g supernatant obtained from the centrifugation of the homogenate was then centrifuged at 11000 × g for 15 min, and the post-mitochondrial supernatant was centrifuged for 60 rain at 223000 × g (Beckman L8.80 centrifuge), to yield the cytosol and the microsomal pellet. Microsomal pellets were rinsed with 1 ml of homogenization buffer to remove cytoplasmic contaminants and resuspended in this buffer. Microsomal suspensions were gently homogenized by 3 or 4 strokes of a Dounce homogenizer (pestle A) to yield the source of the microsomal binding activity, at a (buffer/tissue) ratio of I ml/prostate. Sodium molybdate has been reported before [9] to exert a stabilizing effect on cytosol AR [9]. However, we did not know its effects on microsomal AR. At the same time, this newly described AR form, associated with endoplasmic reticulum [10], could contribute significantly to the total binding capacity of crude nuclear preparations. Therefore, we compared the effects of the presence (25 mM) or absence of sodium molybdate in the homogenization buffer in nuclear AR after nuclei purification by the three methods previously described. After selection of an appropriate method of nuclei purification, we examined the effects of the absence or presence of 10, 25 or 40 mM sodium molybdate in the homogenization buffer on nuclear, cytosol and microsomal AR. Subcellular fractions were isolated as described before. AR was measured in each subcellular compartment with equilibrium binding assays to be described below.

Equilibrium binding studies Effect of antiandrogens on the testosterone-induced nuclear AR accumulation and cymsol or microsomal AR depletion of ventral prostate. 1-day castrates were injected with 400/zg testosterone/100 g body wt., 5 mg RU23908/100 g body wt., 2 mg cyproterone acetate/100 g body wt. or 2 mg hydroxy[lutamide/100 g body wt. intraperitoneally. Testosterone was dissolved in 50% ethanol/saline. Antiandrogens were dissolved in 1,2-propanediol. 500 /zl of testosterone a n d / o r each antiandrogen (given immediately before testosterone) was administered per animal. Groups in which testosterone or antiandrogen were administered alone received also 1,2-propanediol or 50% ethanol/

105 saline, respectively. Groups treated with vehicle alone received both ethanol/saline and 1,2-propanediol. Animals were killed 1 h after treatment. Nuclear, cytosol and microsomal A R concentration and affinity for mibolerone were measured as described below, under equilibrium conditions.

Cytosol and microsomal AR assays 200 /zl aliquots of diluted cytosol (1-4 mg protein/ml) or microsomal suspensions (1-4 mg protein/ml) were incubated f o r 2 0 h at 4 ° C with 0.1-3 nM [3H]DMNT (50 # l / t u b e , final volume, 0.5 ml) in the presence or absence of a 100-fold molar excess of radioinert DMNT. The synthetic androgen, DMNT, was selected for its high affinity for AR, low progesterone receptor binding activity and high biological activity [11]. Bound and unbound DMNT were separated by dextran-coated charcoal adsorption of free and loosely bound steroid (0.5% charcoal/0.05% dextran in homogenization buffer). 1 ml dextran-coated charcoal was added, and the tubes were centrifuged at 800 × g for 10 rain. The supernatants were collected in scintillation vials. 10 ml of cocktail was added to each vial. Samples were shaken for 2 h and counted. In some experiments, a hydroxylapatite 'batch' technique to absorb cytosol or microsomal AR complexes was used, exactly as previously described [12]. We have found no differences in receptor concentration or affinity of mibolerone for cytosol or microsomal AR when hydroxylapatite and dextran-coated charcoal adsorption techniques have been compared, in 1-day castrates (data not shown). Nuclear AR assay 200 #1 aliquots of the nuclear suspension (240-580 /~g D N A / m l ) were dispensed into 12 × 75 mm glass tubes containing increasing concentrations of either [3H]DMNT (0.1-3 nM) or [aH]DMNT plus a 100-fold molar excess of radioinert DMNT (50 /.d/tube, final volume, 0.5 mi). Incubations were carried out for 20 h at 4 ° C. These conditions allow measurement of total nuclear AR. We have not found differences in nuclear AR levels after incubations with iigand during 20 h at 15°C or 4 ° C (data not shown). At the end of each incubation, 1 ml of 'nuclear' buffer was added, followed by centrifugation at 800 x g for 10 min. The 800 × g nuclear pellets were subsequently washed three times by resuspension in 1 ml of ice-cold 'nuclear' buffer, mixing and centrifugation as above. After di;carding the final supernatant, 1 ml of 100% ethanol was added to each pellet and mixed. They were then placed in a 30 ° C water bath for 30 min, with vortexing every 10 min, and finally centrifuged at 800 × g for 10 min. The ethanolic extracts were poured into vials containing 10 ml of scintillation mixture. Samples were shaken for 1 h and counted.

Relatit'e binding affittity of antiandrogens for microsomal and cytosol AR The relative binding affinities (RBA) of cyproterone acetate, RU23908 and hydroxyflutamide for microsomal and cytosol AR of ventral prostate were examined using single point assays, in duplicate, in the presence of 5 nM [3H]DMNT and different amounts of radioinerr DMNT or antiandrogen (0.01-10000-fold molar excess; 250/zl/tube) for 20 h at 4°C. 250/zl of cytosol or microsomal suspension were used in a final volume of 1 ml, adjusted with homogenization buffer. Unbound steroid was removed with dextran-coated charcoal. The relative binding affinity (RBA) of each antiandrogen for microsomal and cytosol AR is assessed as RBA ~ ICso(DMNT)/IC~0(antiandrogen), where IC5c~ = concentration of antiandrogen needed to inhibit 50% of the initial binding activity. Appropriate controls were made in the absence of antiandrogen. Association rate kh~etics studies The effects of RU23908, cyproterone acetate and hydroxyflutamide on the association rate of radiolabeled DMNT to microsomal or cytosol A R were examined by the use of 2 h preincubations at 4 ° C with each antiandrogen before to initiate the association reaction between the receptor source and tritiated DMNT. The concentrations used of RU23908, cyproterone acetate and hydroxyflutamide were 5000-, 500- and 5000-fold molar excess over radiolabeled DMNT, respectively. Fhese concentrations were selected according to the IC50 for each antiandrogen obtained from the relative binding affinity studies previously described. The association rate constants of the interaction between mibolerone and microsomal or cytosol AR were determined as described before [1]. Dissociation rate kinetics studies Microsomal or cytosol fractions were labeled for 20 h at 4 ° C with [3H]DMNT (5.5 nM) to allow the formation of androgen-receptor complexes at equilibrium. After subsequent removal of unbound steroid with dextran-coated charcoal, a 100-fold excess molar of radioinert DMNT (0.55 /zM) or 5000-, 500- and 5000-fold molar excess, respectively, of RU23908, cyproterone acetate or hydroxyflutamide was added to the medium. Degradation rate kinetics of complexes were controlled, after equilibrium, in the absence of an ~xcess zadioinert androgen. Dissociation and degradation rate constants of AR complexes were determined as described before [1]. Effects of cycloheximide on antiandrogen and~or androgen-induced changes #7 nuclear, cytosol and microsoreal AR from rat ventral prostate 1-day orchidectomized rats were given 400/.~g cydoheximide/100 g body wt., 5 mg RU23908/100 g body

106 wt., 2 mg cyproterone acetate/100 g body wt., 2 mg hydroxyflutamide/100 g body wt. or 400 /zg testosterone/100 g body wt. Cycloheximide and testosterone were dissolved in 50% ethanol/saline. The cycloheximide dose used in these studies was selected keeping in mind that the same dose was able to inhibit microsomai and cytosol AR replenishment in ventral prostate of 1-day castrates [12]. Antiandrogens were dissolved in 1,2-propanediol. 500 #1 of each antiandrogen, testosterone or cycloheximide (given immediately before each antiandrogen or before testosterone) was administered per animal. Groups where cycloheximide, antiandrogen or testosterone were administered alone received also 1,2-propanediol or 50% ethanol/saline, according to each treatment condition. Vehicle-alone treated groups received both ethanol/saline and 1,2-propanediol. Animals were killed 1 h after treatment, and nuclear, cytosol and microsomal AR concentration and affinity for mibolerone were measured as described before, under equilibrium conditions.

Analytical procedures Radioactivity was quantified with a Beckman LS7500 liquid scintillation spectrometer, having 56% efficiency for tritium. Microsomal and cytosolic samples were counted in a mixture composed of 5 g of Permablend I1 (Packard) dissolved in 1 liter of toluene. Nuclear samples were counted in a mixture composed of 15 g of butyl.PBD (2-(4'-tert-butylphenyl)-5-(4'-biphenylyl)-1, 3,4-oxadiazole; Sigma), 1770 ml toluene, 620 ml S-570 (ethoxylated octylphenols, Fisher) and 110 ml distilled water. Counting was done at a level permitting less than 2% error. The method of Lowry et al. [13] was used for determination of protein and the method of Burton [14] for determination of DNA concentration.

Statistical analysis The results were analyzed using the Student's t-test. A difference of P < 0.05 was considered significant. Results

Effects of antiandrogen on equilibrium binding properties of rat ~ntral prostate AR Nuclear and cytosoi AR assays are well established [12,15] in ventral prostate. However, the recent identification of a microsomal AR [16] in prostate prompted us to purify the nuclei fractions used for assay to avoid significant membrane presence in nuclei samples. We first examined three different methods of prostate nuclei purification. We compared the nuclear AR levels recovered with each method. Method I showed the highest nuclear AR levels, when results are expressed as fmol/100 # g DNA or as total specific binding cal~city per tissue equivalent (fmol/prostate, data not shown). We subsequently studied the effects of the

presence or absence of molybdate on nuclear AR after isolation and purification of nuclei with these methods, keeping in mind that 25 mM molybdate had been shown to be beneficial for nuclear AR recovery in rat anterior pituitary [17] and that there are well known stabilizing effects of molybdate on cytosol AR from rat ventral prostate [9,18]. The presence of 25 mM molybdate in the homogenization buffer as well as during nuclei purification was deleterious for prostate nuclear AR, regardless of the method used to prepare the source of nuclear AR activity (data not shown). Further analyses of the effects of different concentrations of sodium molybdate on nuclear, cytosol and microsoreal AR showed that optimum binding has specific requirements in each subcellular compartment, i.e., absence of molybdate for nuclear AR vs. 10 mM molybdate for microsomal AR and cytosol AR (data not shown). Although the concentration of binding sites in cytosol and microsomes reached a peak at 10 mM molybdate and it decreased at 25 and 40 mM molybdate, the affinity of mibolerone for cytosol and microsomal AR was higher in the presence of any molybdate concentration than in the absence of molybdate (Ka(× 109 M-m): cytosol AR, 3.3, 14.5, 28.5, 29.9 and microsomal AR, 3.8, 17.1, 10.0 and 19.9 in the absence or presence of 10, 25 or 40 mM sodium molybdate, respectively). These results are in agreement with results reported in the glucocorticoid receptor system [19]. A basic step required for androgen action in target cells is the accumulation of its receptor in the nuclear compartment and the achievement of appropriate levels of nuclear-receptor complex needed to induce a response. The previous development of an optimum binding assay in purified prostate nuclear preparations allowed us to study the effects of antiandrogens on testosterone-induced nuclear AR accumulation in ventral prostate. Fig. 1 (top panel) shows that all antiandrogens used were able to block testosterone-induced increase in nuclear AR with a similar degree of potency. An unexpected result was the effect of these compounds alone: Fig. 1 shows that they have an androgenic effect, increasing nuclear, cytosol and microsomal AR in the rat ventral prostate, in the absence of testosterone, with the following order of potency: RU23908 > hydroxyflutamide > cyproterone acetate. The experiments represented in Fig. 1 (top panel) were performed with purified nuclei and the results are consistent with preliminary experiments made with crude nuclei (washed three times with isotonic sucrose) under the same conditions. Cytosolic depletion of steroid receptors occurs associated with its nuclear accumulation in target cells [20]. Testosterone-induced depletion of cytosol AR was not blocked by RU23908 or cyproterone acetate (Fig. 1, middle panel). Furthermore, hydroxyflutamide potenti-

107 u.J o6

1000"] A 800-1

~

i

Iml Vel~icle(VEH) m Testosterone(T) ~ RU23908(RU) ~irrrltCyproterone Acetate Hydroxyflutami d(OHF) e {CPA)

600-1

=,

a

b

,001

A o 0 ._1

o

I.Z

'°°lli j

0

0 u.

300] C

0 uJ

0

,00]

uJ

o i.u o,.

VEH

T

RU CPA OHF T+ T+ T+ RU CPA OHF

Fig. 1. Effects of testosterone, RU23908, cyproterone acetate and hydroxFflutamide on nuclear (A), cytosol (B) and microsomal (C) AR of ventral prostate from l-day castrates. Male rats (6 animals/group) were treated with an intraperitoneal injection of 400 /.tg testosterone/100 g body wt., 5 mg RU23908/100 g body wt.; 2 mg cyproterone acetate/100 g body wt.; 2 mg hydroxFflutamide/100 g body wt.; or vehicle (representative symbols are shown in the figure). Other groups received the same previous doses of each antiandrogen immediately prior to the administration of testosterone. Animals were killed 1 h after administration. Control levels (in fmol of binding sites/rag protein) were 182.0 (cytosol) and 70.0 (microsomes). 5.2 fmol of binding sites/100/.~g DNA was the control level observed in the nuclei. Each bar value is derived from a Scatchard analysis of a seven-point saturation binding curve. Results are expressed as percentages of control values__ S.E. Each percentage represents the mean of three different experiments. " P < 0.05, as compared to the respective vehicle group: b p < 0.05, as compared to the respective group treated with testosterone alone.

ated the decrease in cytosol AR observed with testosterone alone (Fig. 1, middle panel). Microsomal AR levels decreased after testosterone stimuli in the ventral prostate of 1-day castrated rats [12]. All antiandrogens examined were able to block

testosterone-induced depletion of microsomal AR with the following order of potency: RU23908 > hydroxyflutamide > cyproterone acetate (Fig. 1, bottom panel). An effort was made to correlate the biological potency of each antiandrogen to block testosteroneinduced microsomal or cytosol AR depletion with the relative binding affinity of each compound for each AR form. Table I shows that the relative binding affinities followed the same order of magnitude for both microsomal and cytosol AR: cyproterone acetate > RU23908 = hydroxyflutamide. Therefore, there is no correlation between the relative binding affinity of an antiandrogen for AR and its biological potency to block cytosol or microsomal AR depletion (Fig. 1) in ventral prostate. Cyproterone acetate showed higher relative binding affinity for cytosol AR than for microsomal AR (Table I). A similar phenomena has been shown with the estrogen analog, tamoxifen, which displayed higher relative binding affinity for cytosol than microsomal estrogen receptor of calf uterus, as assessed by competition analysis [21].

Effects of antiandrogen on kinetic binding properties of rat ventral prostate A R The affinity of the interaction between hormone and receptor can be independently determined as the ratio of the association rate and dissociation rate constants of the reaction. The equilibrium association constants usually measured to describe steroid-receptor interactions can be modified by changes of the association or dissociation rate constants of the hormone-receptor complex. These kinetic constants can be modulated independently and represent different phenomena [22]. The IC50 values for each antiandrogen obtained from the previously described competition studies (Table I) were used as criteria for the selection of the antiandrogen concentrations to be examined, in vitro, in order to quantitate these kinetic constants. The association rate kinetics of mibolerone binding to cytosol AR and that of microsomal AR were almost identical (Fig. 2). Both

TABLE I

Relatire binding affinities of RU23908, cyproterone acetate and hydroxvflutamide for cytoso! and microsomal AR Cytosol or microsomal aliquots (250/zl) were incubated for 20 h at 4 ° C with 5 nM [3H]DMNT and increasing concentrations (0.01-10.000-fold molar excess) of each competitor. Protein: 697.5 /~g/tube (microsomes) and 650 #,g/tube (cytosol). Relative binding affinity (RBA) ICso (DMNT)/ICs0 (antiandrogens). The ICs0 is the concentration of competitor needed to inhibit 50% of the initial binding activity (fold molar excess of compelitol ~. Results are presented as averages_+ range of two experimental measurements (duplicates of n = 1). Competitor DMNT RU23908 Cyproteroneacetate Hydroxyflutamide

IC50

RBA

cytosol

microsomes

cytosol

microsomes

6.5 + 0.1 2100 + 300 68 + 3 3100 + 100

8.4+ 0.2 3900 + 100 375 + 15 3250 + 50

1.00 0.0032 +0.0004 0.097 +0.006 0.0021 +0.00002

1.00 0,002 +0.00005 0.023 +0.002 0.0026+0.00002

108 CYTOSOL i

0"12"1 • Control O.1O-I o RU2SS08 / aCygretemneAeetale .

• .*al -.i~==~".~.mtdo

°"1

o °"1

/

~,,,/"A

/

.

•

..,*"¢"

-- 0.02-1 ,.g,~¢~= A

_

.

°'1=t 0~1o MICROSOMES

0"0°1

/ *

/

_

o ,'

-

.j."

1

3

S

r

tl

Time (minutes) Fig. 2, Association rate kinetics of the interaction between ['~H]DMNT and cytosol (top panel) or microsomes (bottom panel) of ventral prostate homogenates in the presence or absence of RU23908, cyproterone acetate or hydroxyflutamide, Control samples were examined in the presence of labelled Ilgand alone. The initial receptor concentrations (R o) were: cytosol, 0.95 nM, microsomes 0.19 nM, The ordinate denotes the incremental changes over the specified time intervals in the funetion=2,3/(Ao-Ro) Iog(At/Rt)xlO 9 M " t Notation for initial concentration of androgen: A o. The initial values of this function were: cytosoh 1.363; microsomes: 2.952. The reactions were started by the addition of 0.55 nM [3H]DMNT (A0) at time zero after a 2 h preincubation at 4 ° C with a 5000-, 500- or 5000-fold molar excess over the labeled ligand of RU23908, cyproterone acetate or hydroxyflutamide, respectively. The amount of [SH]DMNT-receptor complex was measured at each denoted interval, after removal of aliquots in duplicate, in each time point. Control samples: r = 0.94 for cytosol and r = 0.98 for microsomes. Bottom panel: + hydroxyflutamide: r - 0.80; + RU23908: r = 0.83. The association rate constants, k . t (×10 s M -~ s - t ) were the following: eytosoh control (e), 1.32; mierosomes: control (e), 1,38; + RU23908 (o), 0.18; + hydroxyflutamide ( Lx), 0.78; +cyproterone acetate (t:l), not detectable, k . t values for cytosol samples in the presence of all antiandrogens tested were also not detectable,

AR forms displayed second order rate behavior in their interaction with mibolerone. All antiandrogens tested suppressed completely the binding of mibolerone to cytosol AR (Fig. 2, top panel) at short periods of time (between 0 and 11 min) after the initiation of the binding reaction between the ligand and the source of receptor activity, regardless of the steroidal (cyproterone acetate) or non-steroidal (RU23908 and hydroxyflutamide) nature of these compounds. On the other hand, there is correlation between the degree of inhibition by each antiandrogen of the association reaction between mibolerone and microsomal AR with the relative binding affinity of each compound for microsomal AR: cyproterone acetate > RU23908 > hydroxyflutamide (compare Fig. 2, bottom panel and Table I). This phenomenon does not occur with cytosol AR (compare Fig. 2, top panel and Table I). The half-life of dihydrotestosterone dissociation from rat ventral prostate homogenates of 1-day cas-

trated rats has been reported to be 52-65 h at 40C [23]. We, therefore, studied the kinetics of dissociation of mibolerone from cytosol and microsomal AR of ventral prostate over a 15-day period, at 4 ° C under conditions that minimize receptor breakdown [24]. The half-life for dissociation of the cytosol AR-mibolerone complex was 78 h. The same long tl/2 and kinetic behavior was observed for microsomal AR which had a tt/2 of 86 h. Both cytosol AR and microsomal AR were very stable in the presence of hormone and the t~/2 of degradation values were 365 h (cytosol AR) and 707 h (microsomal AR). Similar long ill 2 values of dissociation for microsomal AR and cytosol AR have been reported before, in the presence of androgen alone [1]. There were no differences in dissociation rate constants measured in the presence of RU23908, cyproterone acetate or hydroxyflutamide, as compared to control groups, for cytosol AR or microsomal AR (data not shown), in spite of the fact that cyproterone acetate had a higher relative binding affinity for cytosol

:[ U.J O'J 10001 A +1