Ii

JLLQ

a

ARR2AA

KISE1HAP

A4tDac1lsot

(C

0h

Journal of Steroid Biochemistry & Molecular Biology 141 (2014) 44–51

Contents lists available at ScienceDirect

Journal of Steroid Biochemistry and Molecular Biology

j ourna l h o mepa ge: www.elsev ier .com/ locate / j sbmb

n vitro and in vivo evaluation of a 3�-androsterone derivative asnhibitor of 17�-hydroxysteroid dehydrogenase type 3

enny Roy, Michelle-Audrey Fournier, René Maltais,ucie Carolle Kenmogne, Donald Poirier ∗

aboratory of Medicinal Chemistry, CHU de Québec (CHUL, T4) – Research Center and Laval University (Faculty of Medicine), Québec,uébec G1V 4G2, Canada

r t i c l e i n f o

rticle history:eceived 23 August 2013eceived in revised form9 December 2013ccepted 31 December 2013vailable online 13 January 2014

eywords:nhibitorteroidnzyme7�-Hydroxysteroid dehydrogenaseormonendrogenrostate cancer

a b s t r a c t

17�-Hydroxysteroid dehydrogenase type 3 (17�-HSD3 or HSD17B3) catalyzes the last step in thebiosynthesis of the potent androgen testosterone (T), by stereoselectively reducing the C17 ketoneof 4-androstene-3,17-dione (4-dione), with NADPH as cofactor. Since T plays an important role inandrogen-sensitive diseases, this enzyme is thus an interesting therapeutic target. In an attempt to designcompounds to lower the level of T, we synthesized androsterone derivatives substituted at position 3 asinhibitors of 17�-HSD3, and selected one of the most potent compounds for additional studies. In anenzymatic assay in homogenized and whole HEK-293 cells overexpressing 17�-HSD3, the inhibitor RM-532-105 efficiently inhibited the conversion of natural substrate 4-dione (50 nM) into T with an IC50 of26 nM and 5 nM, respectively. Moreover, the inhibitor RM-532-105 (10 mg/kg) reached a plasma concen-tration of 250 ng/mL at 7 h (AUC 24 h: 3485 ng h/mL) after subcutaneous (s.c.) injection in the rat. In orderto mimic the human situation in which 4-dione is converted to T in the testis, we used intact rats. Treat-ment for 7 days with 17�-HSD3 inhibitor RM-532-105 by s.c. injection or oral gavage exerted no effect onthe testis, prostate and seminal vesicle weight and no modification in the levels of plasma steroids. How-ever, after this treatment, the concentration of inhibitor in plasma increased depending on the dose. We

thereafter determined the concentration of inhibitor in the testis and we discovered that the compoundwas slightly present. In fact, at 10 mg/kg, the inhibitor RM-532-105 seems to have difficulty penetrat-ing inside the testis and was found to be concentrated in the testicular capsule, and therefore unable toinhibit the 17�-HSD3 located inside the testis. However, with a higher dose of 50 mg/kg injected s.c. inrats, RM-532-105 significantly decreased the level of T and dihydrotestosterone measured in plasma at2 h.

Abbreviations: AR, androgen receptor; A-diol, 5�-androstane-3�,17�-diol;-dione, 5�-androstane-3,17-dione; 5-diol, 5-androstene-3�,17�-diol; 4-dione,-androstene-3,17-dione; ADT, androsterone; AUC, area under curve; CTX, cas-rated; CTL, control; DHEA, dehydroepiandrosterone; DHT, dihydrotestosterone;MSO, dimethyl sulfoxide; E2, estradiol; E1, estrone; EDTA, ethylenedi-minetetraacetic acid; GC, gas chromatography; IC50, half maximal inhibitoryoncentration; P450c17, 17�-hydroxylase/17,20-lyase; 17�-HSD3 or HSD17B3,7�-hydroxysteroid dehydrogenase type 3; INT, intact; i.p., intraperitoneal; LH,

uteinizing hormone; LHRH, luteinizing hormone-releasing hormone; MS, masspectrometry; NADPH, nicotinamide adenine dinucleotide phosphate; p.o., per osr gavage; RIA, radioimmunoassay; 5�-Rs, 5�-reductases; s.c., subcutaneous; T,estosterone.∗ Corresponding author at: Laboratory of Medicinal Chemistry, CHU de Québec

CHUL) – Research Center, 2705 Laurier Boulevard, Québec, Québec G1V 4G2,anada. Tel.: +1 418 654 2296; fax: +1 418 654 2298.

E-mail address: donald.poirier@crchul.ulaval.ca (D. Poirier).

960-0760/$ – see front matter © 2014 Elsevier Ltd. All rights reserved.ttp://dx.doi.org/10.1016/j.jsbmb.2013.12.019

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Prostate cancer accounts for about 29% of cancer cases in menand is the second cause of death in men after lung cancer [1].This cancer is generally an androgen-sensitive disease and theandrogens testosterone (T) and dihydrotestosterone (DHT) playan important role in the development, growth and progressionof prostate cancer [2–4]. The optimal androgen blockade therapypresently used consists in the administration of an LHRH agonist(medical castration) in combination with a pure antiandrogen (flu-tamide or bicalutamide) [5,6]. However, studies have shown thatthe widely used chemical castration with depot LHRH agonist failsto achieve castrate levels of T in men in nearly 20% of cases [7,8].Moreover, the relative weak affinity of the antiandrogens flutamide

and bicalutamide for the androgen receptor (AR) [9] leaves theresidual androgens free to interact with the AR and potentially acti-vate the growth of prostate cancer cells. To find a better cure againsthormone-dependent prostate cancer, some different enzymes that

J. Roy et al. / Journal of Steroid Biochemistry & Molecular Biology 141 (2014) 44–51 45

F A) Las( -105 a

clbnadd1treumiatmt1

ima3wwutptPchFtmm

ig. 1. The role of 17�-HSD3 inhibitor in reducing the production of androgens. (DHT) from cholesterol. (B) The chemical structure of 17�-HSD3 inhibitors RM-532

atalyze the biosynthesis of androgens have been targeted [10]. Theast step in the formation of androgens and estrogens is controlledy the key steroidogenic enzymes 17�-hydroxysteroid dehydroge-ases (17�-HSDs or HSD17Bs) and a variety of 17�-HSD isoformsre responsible for the interconversion of 17-ketosteroids (e.g.,ehydroepiandrosterone (DHEA), 4-androstene-3,17-dione (4-ione) and estrone (E1)) as well as their corresponding more active7�-hydroxysteroids (e.g., androst-5-ene-3�,17�-diol (5-diol),estosterone (T) and 17�-estradiol (E2)) [11–15]. Fundamentaloles played by the 17�-HSD family in steroid biology probablyxplain the existence of a large series of isoforms having individ-al cell-specific expression, substrate specificity, and regulatoryechanisms. Among all 17�-HSD isoforms, we are particularly

nterested in type 3 (17�-HSD3 or HSD17B3), which is expressedlmost exclusively in the testis, principally in the microsomal frac-ion of Leydig cells [16]. These cells produce T, the precursor of the

ost active androgen DHT, by a process that requires the reduc-ion of C17 carbonyl of non-androgenic 4-dione [17] catalyzed by7�-HSD3 and its cofactor NADPH (Fig. 1A) [18–21].

While 17�-HSD3 is expressed almost exclusively in the testis,t is sometimes expressed in other tissues. As examples, 17�-HSD3

RNA increased over 30-fold in cancerous prostate biopsies [22]nd the enzyme is overexpressed 8-fold in LuCaP-23 and LuCAP-5 cell lines, both obtained from metastatic tissues of a patientho was resistant to castration therapy [23–25]. The level of Tithin these metastatic tumors was thus sufficiently high to stim-late cancer cell proliferation despite the castrated level of T inhe bloodstream. 17�-HSD3 was also up-regulated in AR-positiverostate cell line LNCaP after it was treated for 48 h with dutas-eride, an inhibitor of 5�-reductases (5�-Rs) types 1 and 2 [26].roduced from T by the action of 5�-Rs, DHT is the main intra-ellular androgen in the prostate and stimulates the growth oformone-dependent prostate cancer via its interaction with the AR.

urthermore, it was reported that 17�-HSD3 catalyzes the biosyn-hesis of approximately 50% of the total amount of androgen in

en [16]. The remaining 50% would result from the same enzy-atic reaction catalyzed by 17�-HSD5 [27,28] or 17�-HSD15 [29]

t steps in the biosynthesis of androgens testosterone (T) and dihydrotestosteronend D-5-2.

in peripheral tissues. In fact, the recently identified 17�-HSD15is expressed in the prostate and is able to catalyze the transfor-mation of 5�-androstane-3,17-dione (5�-dione) into DHT [29,30].An alternative pathway, independent of DHEA, 4-dione and T asintermediates, was also recently reported for the synthesis of DHTfrom progesterone and androsterone (ADT) precursors [31,32].In addition to 17�-hydroxylase/17,20-lyase (P450c17) and 5�-Rs,this backdoor pathway involves 17�-HSD3, for the conversion ofADT to 5�-androstane-3�,17�-diol (A-diol), and 17�-HSD6, for theconversion of A-diol to DHT. Inhibitors of these enzymes can be con-sequently used as a therapeutic approach for prostate cancer [10].However, 17�-HSD3 is a very important enzyme in the productionof T since a deficiency of this enzyme has been associated with pseu-dohermaphroditism [20]. These observations show the importantrole of 17�-HSD3 and suggest that this enzyme is an interestingtherapeutic target. Thus, 17�-HSD3 inhibitors are attractive toolsto lower the level of T (Fig. 1A) and to become a potential agent forandrogen-dependent prostate cancer.

Several groups have so far developed inhibitors of 17�-HSD3as reported in recently published review articles [19,33,34].However, there are few reports of efficacy of these inhibitors inin vivo models and this weakness prompted us to study the in vivoefficacy of a new family of 17�-HSD3 inhibitors developed by ourgroup. These new inhibitors were generated from a preliminarystudy on the inhibition of 17�-HSD3 that identified ADT as alead compound with an IC50 value of 330 nM in microsomes oftransfected HEK-293 cells [35]. Several libraries of ADT derivativeswere next synthesized by adding a hydrophobic group at position3� of ADT nucleus and measuring their inhibitory potencies andandrogenicity [36,37]. Among all these compounds, some ADTderivatives were found to be potent inhibitors of 17�-HSD3 inhomogenized and whole HEK-293 cells overexpressing 17�-HSD3activity while devoid of androgenic activity [38]. Based on these

results, we further investigated the potential of one of our best17�-HSD3 inhibitors (RM-532-105; Fig. 1B) by carrying out in vitroand in vivo studies on rat models and measuring different biologicalparameters.

4 istry &

2

2

2

HcCvs1tffpoaPdideteoRgG([r(tlaowt(

2

psnpafpri4iemmta

2

2

(w

6 J. Roy et al. / Journal of Steroid Biochem

. Experimental

.1. Inhibition of 17ˇ-HSD3

.1.1. In vitro activity using homogenized cellsAn expression vector encoding for human HSD17B3 (17�-

SD3) was transfected into human embryonic kidney (HEK)-293ells using the Exgen 500 procedure (Fermentas, Burlington, ON,anada) as previously reported [39]. Transfected cells were har-ested by trypsinization and heat shock was performed in 50 mModium phosphate buffer (pH 7.5), containing 20% glycerol and

mM EDTA, to obtain cellular fragmentation (−80 ◦C to 37 ◦C, threeimes, 5 min). This crude enzyme preparation was used withouturther purification. The enzymatic assay was performed at 37 ◦Cor 2 h in 1 mL of a solution containing 860 �L of 50 mM sodiumhosphate buffer (pH 7.5, 20% glycerol and 1 mM EDTA), 100 �Lf 5 mM NADPH in phosphate buffer, 10 �L of 5 �M [4-14C]-4-ndrostene-3,17-dione ([14C]-4-dione) in ethanol (53.6 mCi/mmol,erkin Elmer Life Sciences Inc., Boston, MA, USA), 10 �L of inhibitorissolved in ethanol and 20 �L of diluted enzymatic source

n phosphate buffer. The final concentrations for androstene-ione and NADPH were 60 nM and 500 �M, respectively, whereasach inhibitor was assessed in triplicate at several concentra-ions (0.1 nM to 100 �M). Afterwards, radiolabeled steroids werextracted from the reaction mixture with 1 mL of diethyl ether. Therganic phases were evaporated to dryness with nitrogen stream.esidues were dissolved in 50 �L of CH2Cl2 and dropped on silicael 60 F254 thin layer chromatography plates (EMD Chemicals Inc.,ibbstown, NJ, USA) and eluted with a mixture of toluene/acetone

4:1) solvent system. Substrate [14C]-4-dione and metabolite14C]-testosterone ([14C]-T) were identified by comparison witheference steroids and quantified using the Storm 860 SystemMolecular Dynamics, Sunnyvale, CA, USA). The percentage ofransformation of [14C]-4-dione into [14C]-T was calculated as fol-ows: % transformation = 100 × ([14C]-T/([14C]-T + [14C]-4-dione)),nd subsequently, % inhibition = 100 × ((% transformation with-ut inhibitor − % transformation with inhibitor)/% transformationithout inhibitor). The IC50 value, the concentration of inhibitor

hat gives 50% of enzymatic inhibition, was calculated by computerDE50 program, CHUL Research Center, Québec, QC, Canada).

.1.2. In vitro activity using whole cellsHEK-293 cells were seeded at 200,000 cells/well in a 12-well

late (BD Falcon) at 37 ◦C under 95% air 5% CO2 humidified atmo-phere in minimum essential medium (MEM) supplemented withon-essential amino acids (0.1 mM), glutamine (2 mM), sodiumyruvate (1 mM), 10% fetal bovine serum, penicillin (100 IU/mL)nd streptomycin (100 �g/mL). The expression vector encodingor human HSD17B3 [39] was transfected using the Exgen 500rocedure (Fermentas, Burlington, ON, Canada) with 2.5 �g ofecombinant plasmid per well. After a 3-h transfection, cells werencubated for 2 h with fresh culture medium containing [4-14C]--androstene-3,17-dione (50 nM) and an ethanolic solution of

nhibitor (0.5% v/v). Each inhibitor was assessed in triplicate at sev-ral concentrations (0.01 nM–100 �M). After incubation, cultureedium was removed, radiolabeled steroids were extracted fromedium and quantified, the percentage of transformation and then

he IC50 value was calculated as described above for the enzymaticssay in homogenized cells.

.2. In vivo treatments with 17ˇ-HSD3 inhibitor RM-532-105

.2.1. AnimalsSix to seven week-old male Sprague-Dawley rats

Crl:CD®(SD)Br VAF/PlusTM) weighing approximately 220 gere obtained from Charles-River, Inc. (St-Constant, QC, Canada).

Molecular Biology 141 (2014) 44–51

The animals were acclimatized to environmental conditions(temperature: 22 ± 3 ◦C; humidity: 50 ± 20%; 12-h light/12-h darkcycles, lights on at 07:15 h) for at least 1 week before startingthe experiment. The animals were housed three per cage, wereallowed free access to water and a certified commercial rodentfood (Lab Diet 5002, Ralston Purina, St. Louis, MO, USA) and wererandomized according to their body weight. The experiments withanimals were conducted in an animal facility approved by theCanadian Council on Animal Care (CCAC) and the Association forAssessment and Accreditation of Laboratory Animal Care (AALAC).The study was performed in accordance with the CCAC Guide forCare and Use of Experimental Animals. Institutional approval wasobtained.

2.2.2. First assay (1-day treatment)A pharmacokinetic study was carried out following one subcu-

taneous (s.c.) injection of 17�-HSD3 inhibitor RM-532-105 at twodifferent doses (0.2 and 10 mg/kg of body weight in 1 mL of vehiclefluid) in rats. The inhibitor was first dissolved in DMSO to which weadded an aqueous methylcellulose solution (0.4%, w/v) to obtain afinal concentration of 10% DMSO. During this experiment, the ratswere housed individually and fasted for 6 h before the inhibitorinjection. Blood samples for determination of inhibitor plasma con-centration were collected at the jugular vein (0.4 mL by animal)after 1, 2, 3.5, 7 and 24 h post-dose from six rats per time point.After collection at 7 h, a replacement fluid (3 mL of 0.9% sodiumchloride injection USP) was injected s.c. in rats. Blood sampleswere collected into Microvette potassium-EDTA (ethylenediaminetetra-acetic acid)-coated tube (Sarstedt, Montréal, QC, Canada) andcentrifuged at 3200 rpm for 10 min at 4 ◦C. The plasma was collectedand stored at −80 ◦C until analyzed by liquid chromatography/massspectrometry/mass spectrometry (LC/MS/MS) analysis.

2.2.3. Second assay (7-day treatment)Forty-eight rats were assigned to 6 groups comprised of 8 ani-

mals each. Six days before the study, the animals from group 1 werecastrated (CTX) under isoflurane anesthesia. Rats from groups 2–6were kept intact (INT). Rats in the CTX and INT control groups (1 and2 respectively) received a daily s.c. injection (1 mL) of the vehiclealone (10% DMSO–90% aqueous methylcellulose (0.4%, w/v)) duringthe 7-day period. The 17�-HSD3 inhibitor RM-532-105 was admin-istered each day early morning as a suspension in the vehicle at adaily oral dose of 10 mg/kg of body weight in 1 mL (group 3) ora daily s.c. injection of 10, 1 or 0.2 mg/kg of body weight in 1 mL(groups 4–6). Seven hours after the seventh day of treatment (at2 h PM), the rats under isoflurane anesthesia were sacrificed byexsanguination by cardiac venipuncture followed by cervical dis-location. Blood from each rat was then collected into heparinizedmicrotainer tubes (Becton Dickinson, Franklin Lakes, NJ, USA) andcentrifuged at 3200 rpm for 10 min at 4 ◦C. The plasma was col-lected, pooled for each group, and stored at −80 ◦C until use forthe determination of inhibitor concentration by LC/MS/MS analysis,steroid concentration by gas chromatography/mass spectrome-try (GC/MS) analysis and luteinizing hormone (LH) measurementby RIA analysis. Whole testes from rats were rapidly trimmed,weighed, and kept in liquid nitrogen and stored at −80 ◦C untiluse for the determination of inhibitor concentration by LC/MS/MSanalysis or enzymatic assay.

2.2.4. Third assay (2- and 6-h treatment)Eighteen rats were assigned to three groups of six animals

each. Group 1 was treated with one s.c. injection of RM-532-105

(50 mg/kg in 0.3 mL of 8% DMSO–92% aqueous methylcellulose(0.4%, w/v)), group 2 was treated with one intraperitoneal (i.p.)injection of abiraterone (10 mg/kg in 0.3 mL of a vehicle constitutedof DMSO (8%), Tween 80 (2%) and phosphate buffer pH 7.4 (90%))

istry &

aoB2biaagt

2

2

dtp6tt(wT1a

Fdcla

J. Roy et al. / Journal of Steroid Biochem

nd group 3 (control) received only the vehicle (0.3 mL, i.p. injectionf DMSO (8%), Tween 80 (2%) and phosphate buffer pH 7.4 (90%)).lood samples were collected at the jugular vein (0.5 mL by animal)

h after the injection, whereas the rats were sacrificed after 6 h andlood collected by cardiac puncture. Blood samples were collected

nto Microvette potassium-EDTA (ethylenediamine tetraaceticcid)-coated tube (Sarstedt, Montréal, QC, Canada) and centrifugedt 3200 rpm for 10 min at 4 ◦C. The plasma samples of the sameroup at a given time point were pooled and stored at −80 ◦C untilhe concentrations of T and DHT were measured by LC/MS/MS.

.3. Determination of inhibitor, steroid and LH concentrations

.3.1. Inhibitor measurementThe concentration of 17�-HSD3 inhibitor RM-532-105 was

etermined by LC/MS/MS analysis using a procedure developed athe CHUQ (CHUL) – Research Center. Briefly, for extraction fromlasma, a 100 �L sample is transferred to individual tubes and00 �L of ammonium acetate (1 mM) are added. A methanolic solu-ion (50 �L) containing a steroid internal standard is then addedo each tube. Samples are transferred on Strata-X SPE columnsPhenomenex, Torrance, CA, USA) and each column is washed

ith water (1× 2 mL) and methanol:water (10:90, v/v) (1× 2 mL).

he inhibitor is then eluted with 5 mL of methanol containing mM ammonium acetate. Methanol is evaporated at 45 ◦C under

stream of nitrogen and the dried residue dissolved in 100 �L of

ig. 2. Inhibitory potency of RM-532-105, D-5-2 and unlabeled natural substrate 4-ione on HEK-293 cells overexpressing 17�-HSD3. Homogenized cells (A) or wholeells (B) were incubated with various concentrations of inhibitor in the presence ofabeled 4-dione (50 nM). IC50 represents the concentration that inhibited 17�-HSD3ctivity by 50%. Results are the means (±SEM) of a triplicate.

Molecular Biology 141 (2014) 44–51 47

methanol:water (85:15, v/v). For the inhibitor analysis, the HPLCsystem uses a 75-mm × 4.6-mm reversed-phase ACE3 C18-HL col-umn (Canadian Life Science, Montréal, QC, Canada) and a mixtureof methanol:water (90:10, v/v) containing 1 mM of ammonium for-mate at a flow rate of 0.8 mL/min. The inhibitor is detected usingan API 5000 mass spectrometer equipped with a TurboIonSpray(Applied Biosystems, Montréal, QC, Canada). Electrospray ioniza-tion in positive ion mode was used.

2.3.2. Steroid measurementThe concentration of C19-steroids T, DHT, 4-dione and 5-diol

were determined by GC/MS analysis according to a procedure wellestablished at the CHUQ (CHUL) – Research Center (BioanalyticalService) and previously published [40].

2.3.3. LH measurementLH plasma concentration was determined in 100 �L of plasma

by a RIA method using the Biotrak rat LH hormone 125I assay sys-tem (Amersham Biosciences, England). The minimum detectableLH concentration in this assay was approximately 0.011 ng/100 �L(0.11 ng/mL). Plasma samples from in vivo experiments werepooled and assayed in triplicate.

2.4. Homogenization of testes for enzymatic assay (inhibition ofrat 17ˇ-HSD3)

Whole rat testes were homogenized on ice with a Polytron incold phosphate buffer (20 mM KH2PO4, 0.25 M sucrose, 1 mM EDTA,

pH 7.5) containing protease inhibitors mini-complete (Roche Diag-nostics, Laval, QC, Canada) and centrifuged at 12,500 × g for 15 minto remove the mitochondria, plasma membranes, and cell frag-ments. The supernatant was further centrifuged at 100,000 × g for

Fig. 3. Plasma concentration of 17�-HSD3 inhibitor RM-532-105 as a function oftime. The inhibitor was injected subcutaneous (s.c.) at a dose of 10 mg/kg (A) or0.2 mg/kg (B) in Sprague-Dawley rats. Results are the means (±SEM) of a triplicate.

4 istry &

1macsau

2

ttcDc

3

Hlgsifco

F(

8 J. Roy et al. / Journal of Steroid Biochem

5 min using an ultracentrifuge equipped with a 70.1 Ti rotor. Theicrosomal pellet was washed three times with phosphate buffer

nd centrifuged at 100,000 × g for 15 min. All these operations wereonducted at 4 ◦C. The protein concentration of the washed micro-omes was determined by the Bradford method using bovine serumlbumin as standard [41]. The enzymatic assay was performednder the same conditions mentioned above for homogenized cells.

.5. Statistical analysis

The area under the curve of the inhibitor plasma concentra-ion from time 0 to 24 h (AUC0–24h) was calculated by a linearrapezoidal method. Data are presented as means ± SEM. Statisti-al significance was determined according to the multiple-rangeuncan–Kramer test [42]. P values that were less than 0.05 wereonsidered as statistically significant.

. Results and discussion

During the past few years we have designed a new family of 17�-SD3 inhibitors [35–38]. Following the screening results from our

ast optimization study [38], we have decided to further investi-ate one of the best candidates, the 3�-dimethylpiperazino-ADTulfonamide derivative RM-532-105 (Fig. 1B). In addition to its

nteresting inhibitory activity for 17�-HSD3, this compound wasound to be nonandrogenic on androgen-sensitive (AR+) Shionogiells [38], a crucial characteristic for a potential use in the treatmentf androgen-sensitive diseases, and especially prostate cancer.

ig. 4. Effect of 17�-HSD3 inhibitor RM-532-105 on the weight of four androgen-sensitivB), dorsal prostate (C) and seminal vesicles (D) were measured. CTL, control group; CTX,

Molecular Biology 141 (2014) 44–51

Using HEK-293 cells overexpressing 17�-HSD3 activity, RM-532-105 was a potent inhibitor with IC50 of 26 nM in homogenized cells(Fig. 2A) and 5 nM in whole cells (Fig. 2B). Furthermore, it was amore efficient inhibitor than steroid analog D-5-2 (IC50 = 37 and51 nM), a first generation inhibitor previously synthesized in ourlaboratory (Fig. 1B) [37]. RM-532-105 was also a better competi-tor of labeled enzyme substrate [14C]-4-dione than the unlabelednatural substrate 4-dione (IC50 = 341 and 337 nM) used itself as aninhibitor. These interesting results prompted us to further investi-gate the potential of this second generation 17�-HSD3 inhibitor incarrying out in vivo studies.

As the first step of our in vivo studies, we measured at dif-ferent times (1, 2, 3.5, 7 and 24 h) the concentration of inhibitorpresent in blood following a single s.c. injection of 10 or 0.2 mg/kgin rat (Fig. 3). At the highest dose, the inhibitor achieved a maxi-mum plasma concentration of 250 ng/mL at 7 h and the AUC0–24hwas 3485 ng h/mL. For the lower dose of 0.2 mg/kg, the maximumplasma concentration of the inhibitor reached 14 ng/mL at 1 h anddecreased quickly in the bloodstream almost completely disap-pearing (2.7 ng/mL) 7 h after the injection. This observation andthe weak AUC0–24h value of 87 ng h/mL suggest that the dose of0.2 mg/kg is too low to be administered only once a day. However,the highest dose of 10 mg/kg provided a 40-time higher AUC0–24hvalue that roughly correlated with the 50-time higher dose of

inhibitor injected.

After we established that our 17�-HSD3 inhibitor was found inplasma after a one-day single s.c. injection, we assessed its capacityto block the production of T in the testis using an intact rat model.

e tissues in rats. After 7 days of treatment, the weights of testes (A), ventral prostatecastrated rats; INT, intact rats.

J. Roy et al. / Journal of Steroid Biochemistry & Molecular Biology 141 (2014) 44–51 49

Table 1Plasma concentration of 17�-HSD3 inhibitor RM-532-105 (INH) in rats after 7 daysof treatment.

Group Treatment ConcentrationRM-532-105 (ng/mL)

1 CTX BLQ2 INT BLQ3 INT + INH (10 mg/kg/d) (p.o.) 1.34 INT + INH (10 mg/kg/d) (s.c.) 80.45 INT + INH (1 mg/kg/d) (s.c.) 5.86 INT + INH (0.2 mg/kg/d) (s.c.) 1.1

Co

IhtfidAvpo1pitattsihop

rl(aidCt4ttwoCioi

TP1

Co

lower concentrations (0.2 and 1 mg/kg) did not produce a signif-icant inhibition effect on 17�-HSD3 whereas the 10 mg/kg dosesproduced only a weak effect. This last result allowed us to con-clude that our inhibitor was not present in the testis of treated

TX, castrated rats; INT, intact rats; BLQ, below limit of quantification. The coefficientf variation for inhibitor measurement is 8%.

n fact, contrary to humans [16], the adrenal glands in rats do notave the capacity to produce C19-steroids (such as T and DHEA) andhe testosterone measured in rat plasma comes from the testis. Atrst, inhibitor RM-532-105 was administered once a day during 7ays to four different groups by s.c. injection or oral gavage (p.o.).s shown in Fig. 4, the inhibitor did not reduce the weight of testes,entral prostate, dorsal prostate and seminal vesicles when com-ared with control. These results suggest a problem with the modef action of this inhibitor regardless of the dose used (0.2, 1 and0 mg/kg) and the two modes of administration tested (s.c. and.o.). To understand this result, we decided to explore some phys-

ological parameters more thoroughly. After 7 days of treatment,he plasma concentration of inhibitor, the levels of key steroidsnd the level of LH were determined for each group. The result ofhese analyses at first confirmed that the plasma concentration ofhe inhibitor is dose dependent (Table 1). In fact, after 7 days of.c. injection of 0.2, 1 and 10 mg/kg, the plasma concentration ofnhibitor reaches 1.1, 5.8 and 80.4 ng/mL, respectively. This dosageas also indicated that the s.c. administration of inhibitor is rec-mmended since 7 days of oral treatment with 10 mg/kg leads to alasma concentration of only 1.3 ng/mL.

Following the measurement of RM-532-105 concentration inats, we evaluated the level of three steroids (T, DHT and 4-dione)ikely to be influenced by the action of this 17�-HSD3 inhibitorTable 2). The plasma levels of these steroids remained unchangedfter 7 days of treatment except for the 0.2 mg/kg dose whichncreased the levels of T (61%), DHT (121%) and 4-dione (76%). Weo not know what caused this unexpectedly increase of these three-19 steroids at the lower inhibitor dose used. Following adminis-ration of the inhibitor, we should normally have a higher level of-dione and lower levels of T and DHT. That was not the case andhe levels of measured steroids were roughly the same in groupsreated or not by the inhibitor. Although 80.4 ng/mL of inhibitoras present in the plasma of the rats in group 4, the concentration

f key androgenic steroids T and DHT never dropped or reached the

TX level. The measure of LH in plasma was next assessed because

t is known that LH level rise rapidly following castration. Indeed,rchidectomy produces a reduction of T and alteration in pitu-tary function, which is characterized by a sharp increase in plasma

able 2lasma concentration of three C-19 steroids in rats after 7 days of treatment by7�-HSD3 inhibitor RM-532-105 (INH).

Group Treatment T(pg/mL)

DHT(pg/mL)

4-Dione(pg/mL)

1 CTX BLQ BLQ BLQ2 INT 1900 67 0.213 INT + INH (10 mg/kg/d) (p.o.) 2310 80 0.244 INT + INH (10 mg/kg/d) (s.c.) 1690 95 0.175 INT + INH (1 mg/kg/d) (s.c.) 2130 87 0.236 INT + INH (0.2 mg/kg/d) (s.c.) 3060 148 0.37

TX, castrated rats; INT, intact rats; BLQ, below limit of quantification. The coefficientf variation for steroid measurement range from 3% to 13%.

Fig. 5. Plasma concentration of LH after 7 days of treatment by 17�-HSD3 inhibitorRM-532-105 in intact rats. CTL, control group; CTX, castrated rats; INT, intact rats.

levels of the LH and FHS hormones. However, in our in vivo exper-iment, none of the inhibitor doses in long term 7-day treatmenthas allowed to increase the LH concentration (Fig. 5), suggestingno retro feedback produced by the expected low level of T. All theresults reported above suggest that our inhibitor did not block theformation of T in the rat testis.

To confirm that the low potency of our inhibitor in the rat testiswas not due to the species, since our inhibitor was developed usinghuman 17�-HSD3, we verified its activity directly in homogenizedtestes of untreated rats. In our preparation of whole testes, we elim-inated the cytosolic fraction containing 17�-HSD5, another enzymeable to transform 4-dione into T, and we selected the micro-somal fraction containing 17�-HSD3 [43]. At 0.1, 1 and 10 �M,the inhibitor RM-532-105 inhibited 45, 60 and 73%, respectively,of the transformation of [14C]-4-dione into [14C]-T by 17�-HSDactivity found in homogenized rat testes (Fig. 6). Despite an esti-mated IC50 of 100 nM obtained in homogenized rat testes, whichvalue is higher than the IC50 value (26 nM) obtained in homoge-nized HEK-293 cells overexpressing human 17�-HSD3, inhibitorRM-532-105 is clearly efficient in both species (rat and human).We next measured the 17�-HSD3 activity in homogenized testesof rat treated during 7 days with inhibitor RM-532-105 at differ-ent concentrations and two modes of administration (Fig. 7). The

Fig. 6. 17�-HSD3 activity in homogenized testes (25 �g of protein) from untreatedrats. **Indicates a significantly different result (P < 0.01) from control group (CTL).

50 J. Roy et al. / Journal of Steroid Biochemistry &

Fig. 7. 17�-HSD3 activity in homogenized testes (25 �g of protein) of rats treated7 days by inhibitor RM-532-105. *Indicates a significantly different result (P < 0.05)from control group (CTL). **Indicates a significantly different result (P < 0.01) fromCTL.

Table 3Concentration of 17�-HSD3 inhibitor RM-532-105 in the testis after 7 days oftreatment.

Treatment (s.c) Animal # Testis Inhibitor (ng/g of tissue)

0.2 mg/kg/d 1 Capsule 6.51 Interior BLQ2 Capsule 6.62 Interior BLQ

1 mg/kg/d 3 Capsule 14.93 Interior BLQ4 Capsule 9.94 Interior 5.7

10 mg/kg/d 5 Capsule 45.95 Interior 9.16 Capsule 68.56 Interior 7.2

Bm

reotd

RM-532-105 (10 mg/kg) did not inhibit the concentration of T in

Frfi

LQ, below limit of quantification. The coefficient of variation for inhibitor measure-ent is 8%.

ats in sufficient concentration to produce a significant inhibitoryffect on 17�-HSD3. We thereafter determined the concentration

f inhibitor within the testis of treated rats and we discovered thathe inhibitor concentration is low (Table 3). In fact, the inhibitor isose-dependent concentrated in the testicular capsule, but is not

ig. 8. Effect of 17�-HSD3 inhibitor RM-532-105 and P450c17 inhibitor abiraterone on thesults are expressed as the concentration of testosterone (T) (A) and the concentration

xed at 100%). *Indicates a significantly different result (P < 0.05) from control group. **In

Molecular Biology 141 (2014) 44–51

sufficiently concentrated to inhibit the 17�-HSD3 located insidethe testis. There are two possible hypotheses that could explainthis result. The first is the nature of the capsule. Since the capsule isa very compact and fibrous tissue, it can accumulate the inhibitorwhich will not reach the Leydig cells inside the testis in sufficientconcentration to inhibit the enzyme. The second hypothesis is thatthe inhibitor penetrates inside the testis, but is quickly eliminatedbefore reaching the concentration to exert its action.

In order to shed new light to the previous results, we proceededwith an additional in vivo assay (short-term study) with a higherdose of 17�-HSD3 inhibitor RM-532-105. We thus measured theplasma concentration of both T and DHT in rats, 2 h and 6 h fol-lowing a single dose (50 mg/kg) s.c. injection of RM-532-105. Inthis experiment, the untreated (control) group received a differentvehicle by i.p. injection than the group treated with RM-532-105,but this should not affect our results, since the administration ofa vehicle had no effect on the levels of T and DHT [44]. As results,we were pleased to see that RM-532-105 was able to decrease thelevel of T by 31% and the level of DHT by 45%, 2 h post-injection(Fig. 8). Furthermore, this blocking action of the inhibitor was inthe same order as observed with abiraterone (31 and 27% for T andDHT, respectively), a known P450c17 inhibitor [45] injected i.p. asa positive control in the experiment. At 6 h post-injection, how-ever, the level of T recovered was close to the level of the untreated(control) group. This fact by itself could explain why we did notobserve a decrease in the T level in our long-term (7-day) in vivostudy (Table 2) considering that the blood sampling was taken 7 hpost-injection. Supplemental pharmacokinetic studies will thus benecessary to determine the optimal dosage, as well as the frequencyof injection required to exert a durable effect on the T blockadeaction of inhibitor RM-532-105.

4. Conclusion

The 3�-(o-trifluoromethylphenyl-sulfonyl-dimethylpipera-zino)-androsterone (RM-532-105) inhibited human 17�-HSD3,the enzyme responsible for the formation of T from 4-dione inlow nM concentrations and in two preparations (homogenizedand whole transfected HEK-293 cells). This ADT derivative alsoinhibited the formation of T in a preparation of rat testicular17�-HSD3. In a long-term (7-day) study performed with rats,

plasma. However, in a short-term (2–6 h) study using a higherdose of inhibitor (50 mg/kg), it reduced the formation of T and DHTin plasma. Considering this positive result, the next step will be

e plasma concentration of testosterone and dihydrotestosterone in intact rats. Theof dihydrotestosterone (DHT) (B) reported versus the control group (concentrationdicates a significantly different result (P < 0.01) from control group.

istry &

tott

A

RQnwtcM

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

J. Roy et al. / Journal of Steroid Biochem

o further investigate the pharmacokinetic aspect in order to findptimal conditions that will provide a sufficient concentration ofhe inhibitor to block the formation of T in androgen-sensitiveissues or in tumors expressing 17�-HSD3.

cknowledgments

This work was supported by the Canadian Institutes of Healthesearch (CIHR). We are grateful to Dr. Van Luu-The (CHU deuébec (CHUL) – Research Center) for providing the plasmideeded for overexpressing the 17�-HSD3 in HEK-293 cells. Weould also like to thank Patrick Bélanger and René Bérubé, from

he CHUQ (CHUL)-Bioanalytical Service, for inhibitor and steroidoncentration determination. Careful reading of the manuscript byicheline Harvey is also greatly appreciated.

eferences

[1] R. Siegel, D. Naishadham, A. Jamal, Cancer Statistics 2012, CA Cancer J. Clin. 62(2012) 10–29.

[2] J.D. McConnell, Physiologic basis of endocrine therapy for prostatic cancer, Urol.Clin. North Am. 18 (1991) 1–13.

[3] F. Labrie, A. Belanger, V. Luu-The, C. Labrie, J. Simard, L. Cusan, J. Gomez, B.Candas, Gonadotropin-releasing hormone agonists in the treatment of prostatecancer, Endocr. Rev. 26 (2005) 361–379.

[4] F. Labrie, Screening and early hormonal treatment of prostate cancer are accu-mulating strong evidence and support, Prostate 40 (2000) 215–222.

[5] F. Labrie, L. Cusan, C. Labrie, J. Simard, V. Luu-The, P. Diamond, F. Gomez, B.Candas, History of LHRH agonist and combination therapy in prostate cancer,Endocr. Relat. Cancer 3 (1996) 243–278.

[6] E.L. Gheiler, R. Tiguert, Current concepts in androgen deprivation ther-apy – is there a best endocrine treatment? World J. Urol. 18 (2000)190–193.

[7] M.G. Oefelein, R. Cornum, Failure to achieve castrate levels of testosteroneduring luteinizing hormone releasing hormone agonist therapy: the case formonitoring serum testosterone and a treatment decision algorithm, J. Urol.164 (2000) 726–729.

[8] M.G. Oefelein, Serum testosterone-based luteinizing hormone-releasing hor-mone agonist redosing schedule for chronic androgen ablation: a phase Iassessment, Urology 54 (1999) 694–699.

[9] J. Simard, I. Luthy, J. Guay, A. Bélanger, F. Labrie, Characteristics of interactionof the antiandrogen flutamide with the androgen receptor in various targettissues, Mol. Cell. Endocrinol. 44 (1986) 261–270.

10] D. Poirier, New cancer drugs targeting the biosynthesis of estrogens and andro-gens, Drug Dev. Res. 69 (2008) 304–318.

11] G. Moeller, J. Adamski, Integrated view on 17beta-hydroxysteroid dehydroge-nases, Mol. Cell. Endocrinol. 301 (2009) 7–19.

12] S. Marchais-Oberwinkler, C. Henn, G. Möller, T. Klein, M. Negri, A. Oster, A.Spadaro, R. Werth, M. Wetzel, K. Xu, M. Frotscher, R.W. Hartmann, J. Adamski,17�-Hydroxysteroid dehydrogenases (17�-HSDs) as therapeutic targets: pro-tein structures, functions, and recent progress in inhibitor development, J.Steroid Biochem. Mol. Biol. 125 (2011) 66–82.

13] V. Luu-The, F. Labrie, The intracrine sex steroid biosynthesis pathways, in: L.Martini (Ed.), Progress in Brain Research 181 (2010) 177–192 (Chap. 10).

14] F. Labrie, V. Luu-The, S.X. Lin, C. Labrie, J. Simard, R. Breton, A. Bélanger, The keyrole of 17�-hydroxysteroid dehydrogenases in sex steroid biology, Steroids 62(1997) 148–158.

15] M. Rotinen, J. Villar, I. Encio, Regulation of 17�-hydroxysteroid dehydrogenasesin cancer: regulating steroid receptor at pre-receptor stage, J. Physiol. Biochem.68 (2012) 461–473.

16] F. Labrie, V. Luu-The, C. Labrie, A. Bélanger, J. Simard, S.X. Lin, G. Pelletier,Endocrine and intracrine sources of androgens in women: inhibition of breastcancer and other roles of androgens and their precursor dehydroepiandros-terone, Endrocr. Rev. 24 (2003) 152–182.

17] Y. Laplante, D. Poirier, Proliferative effect of androst-4-ene-3,17-dione andits metabolites in the androgen-sensitive LNCaP cell line, Steroids 73 (2008)266–271.

18] T.M. Penning, Molecular endocrinology of hydroxysteroid dehydrogenases,Endocr. Rev. 18 (1997) 281–305.

19] M.L. Mohler, R. Narayanan, Y. He, D.D. Miller, J.T. Dalton, Hydroxysteroiddehydrogenase (17�-HSD3, 17�-HSD5, and 3�-HSD3) inhibitors: extragonadalregulation of intracellular sex steroid hormone levels, Recent Patents Endocr.

Metab. Immune Drug Discov. 1 (2007) 103–118.

20] W.M. Geissler, D.L. Davis, L. Wu, K.D. Bradshaw, S. Patel, B.B. Mendoca, K.O.Elliston, J.D. Wilson, D.W. Russell, S. Andersson, Male pseudohermaphroditismcaused by mutations of testicular 17�-hydroxysteroid dehydrogenase 3, Nat.Genet. 7 (1994) 34–39.

[

Molecular Biology 141 (2014) 44–51 51

21] S. Andersson, W.M. Geissler, S. Patel, L. Wu, The molecular biology of androgenic17�-hydroxysteroid dehydrogenases, J. Steroid Biochem. Mol. Biol. 53 (1995)37–39.

22] E. Koh, T. Noda, J. Kanaya, M. Namiki, Differential expression of 17�-hydroxysteroid dehydrogenase isozyme genes in prostate cancer andnoncancer tissues, Prostate 53 (2002) 154–159.

23] R.B. Montgomery, E.A. Mostaghel, R. Vessella, D. Hess, T.F. Kalhorn, C.S. Higano,L.D. True, P.S. Nelson, Maintenance of intratumoral androgens in metastaticprostate cancer: mechanism for castration-resistant tumor growth, Cancer Res.68 (2008) 4447–4454.

24] J.A. Locke, E. Guns, A.A. Lubik, H.H. Adomat, S.C. Hendy, C.A. Wood, S.L. Ettinger,M.E. Gleave, C.C. Nelson, Androgen levels increase by intratumoral de novosteroidogenesis during progression of castration-resistant prostate cancer,Cancer Res. 68 (2008) 6407–6415.

25] E.A. Mostaghel, S.T. Page, D.W. Lin, L. Fazli, I.M. Coleman, L.D. True, B. Knud-sen, D.L. Hess, C.C. Nelson, A.M. Matsumoto, W.J. Bremner, M.E. Gleave, P.Nelson, Intraprostatic androgens and androgen-regulated gene expressionpersist after testosterone suppression: therapeutic implications for castration-resistant prostate cancer, Cancer Res. 67 (2007) 5033–5041.

26] M. Biancolella, A. Valentini, D. Minella, L. Vecchione, F. D’Amico, G. Chillemi,P. Gravina, S. Bueno, G. Prosperini, A. Desideri, G. Federici, S. Bernardini, G.Novelli, Effects of dutasteride on the expression of genes related to androgenmetabolism and related pathway in human prostate cancer cell lines, Invest.New Drugs 25 (2007) 491–497.

27] I. Dufort, P. Rheault, X.F. Huang, P. Soucy, Characteristics of a highly labilehuman type 5 17�-hydroxysteroid dehydrogenase, Endocrinology 140 (1999)568–574.

28] M.C. Byrns, Y.T. Jin, T.M. Penning, Inhibitors of type 5 17�-hydroxysteroid dehy-drogenase (AKR1C3): overview and structural insights, J. Steroid Biochem. Mol.Biol. 125 (2011) 95–114.

29] V. Luu-The, A. Bélanger, F. Labrie, Androgen biosynthetic pathways in thehuman prostate, Best Pract. Res. Clin. Endocrinol. Metab. 22 (2008) 207–221.

30] V. Luu-The, Assessment of steroidogenesis and steroidogenic enzyme func-tions, J. Steroid Biochem. Mol. Biol. 137 (2013) 176–182.

31] J.L. Mohler, M.A. Titus, E.M. Wilson, Potential prostate cancer drug target: bioac-tivation of androstanediol by conversion to dihydrotestosterone, Clin. CancerRes. 17 (2011) 5844–5849.

32] N. Sharifi, R.J. Auchus, Steroid biosynthesis and prostate cancer, Steroids 77(2012) 719–726.

33] D. Poirier, 17�-Hydroxysteroid dehydrogenase inhibitors: a patent review, Exp.Opin. Ther. Patents 20 (2010) 1123–1145.

34] J.M. Day, H.J. Tutill, A. Purohit, M.J. Reed, Design and validation of specificinhibitors of 17�-hydroxysteroid dehydrogenases for therapeutic applicationin breast and prostate cancer, and in endometriosis, Endocr. Relat. Cancer 15(2008) 665–692.

35] B. Tchédam-Ngatcha, Y. Laplante, F. Labrie, V. Luu-The, D. Poirier, 3�-Alkyl-androsterones as inhibitors of type 3 17�-hydroxysteroid dehydrogenase:Inhibitory potency in intact cells, selectivity towards isoforms 1, 2, 5 and7, binding affinity for steroid receptors, and proliferative/antiproliferativeactivities on AR+ and ER+ cell lines, Mol. Cell. Endocrinol. 248 (2006)225–232.

36] B. Tchédam-Ngatcha, V. Luu-The, F. Labrie, D. Poirier, Androsterone 3�-ether-3�-substituted and androsterone 3�-substituted derivatives as inhibitorsof type 3 17�-hydroxysteroid dehydrogenase: chemical synthesis andstructure–activity relationship, J. Med. Chem. 48 (2005) 5257–5268.

37] R. Maltais, V. Luu-The, D. Poirier, Synthesis and optimization of a new family oftype 3 17 �-hydroxysteroid dehydrogenase inhibitors by parallel liquid-phasechemistry, J. Med. Chem. 45 (2002) 640–653.

38] R. Maltais, M.A. Fournier, D. Poirier, Development of 3�-substituted andros-terone derivatives as potent inhibitors of 17�-hydroxysteroid dehydrogenasetype 3, Bioorg. Med. Chem. 19 (2011) 4652–4668.

39] V. Luu-The, Y. Zhang, D. Poirier, F. Labrie, Characteristics of human types 1, 2and 3 17�-hydroxysteroid dehydrogenase activities: oxidation/reduction andinhibition, J. Steroid Biochem. Mol. Biol. 55 (1995) 581–587.

40] F. Labrie, A. Bélanger, P. Bélanger, R. Bérubé, C. Martel, L. cusan, J. Gomez, B.Candas, V. Chaussade, I. Castiel, C. Deloche, J. Leclaire, Metabolism of DHEAin postmenopausal women following percutaneous administration, J. SteroidBiochem. Mol. Biol. 103 (2007) 178–188.

41] M.M. Bradford, A rapid and sensitive method for the quantification of micro-gram quantities of protein utilizing the principle of protein–dye binding, Anal.Biochem. 72 (1976) 248–254.

42] C.Y. Kramer, Extension of multiple range tests to group with unique numbersof replications, Biometrics 12 (1956) 307–310.

43] H. Peltoketo, V. Luu-The, J. Simard, J. Adamski, 17�-Hydroxysteroid dehydro-genase (HSD)/17-ketosteroid reductase (KSR) family; nomenclature and maincharacteristics of the 17HSD/KSR enzymes, J. Mol. Endocrinol. 23 (1999) 1–11.

44] P.E. Juniewicz, B.H. Johnson, D.J. Bolt, Effect of adrenal steroid on testos-

terone and luteinizing hormone secretion in the ram, J. Androl. 30 (1987)190–196.

45] G. Attard, A.S. Belldegrun, J.S. De Bono, Selective blockade of androgenicsteroid synthesis by novel lyase inhibitors as a therapeutic strategy for treatingmetastatic prostate cancer, BJU Int. 96 (2005) 1241–1246.