Inhibitors of type II 17β-hydroxysteroid dehydrogenase

Molecular and Cellular Endocrinology 171 (2001) 119 – 128 www.elsevier.com/locate/mce

Inhibitors of type II 17b-hydroxysteroid dehydrogenase Donald Poirier *, Patrick Bydal, Martin R. Tremblay, Kay-Mane Sam, Van Luu-The Medicinal Chemistry Di6ision, Oncology and Molecular Endocrinology Research Center, La6al Uni6ersity Medical Centre (CHUL), 2705 Laurier Boule6ard, Quebec, Canada G1V 4G2

Abstract The 17b-hydroxysteroid dehydrogenases (17b-HSDs) are involved in the last step of the biosynthesis of sex steroids from cholesterol. This family of steroidogenic enzymes constitutes an interesting target in the control of the concentration of estrogens and androgens. Among the isoforms of 17b-HSD, type II preferentially catalyzes the oxidation of estradiol (E2), testosterone (T), dihydrotestosterone (DHT), and 20a-dihydroprogesterone (20a-DHP). Based on structure – activity relationship studies, we have developed steroidal spirolactones as inhibitors of type II 17b-HSD using different steroid nuclei: a C18-steroid (lactones 1 and 10), an antiestrogenic nucleus (lactone 2), and a C19-steroid (lactone 28). We know these inhibitors are selective for type II 17b-HSD as no or only weak inhibition was observed for types I and III. They also have no proliferative (androgenic) activity on androgen sensitive (AR+) Shionogi cells whereas their proliferative (estrogenic) activity on estrogen sensitive (ER+) ZR-75-1 cells depends on the nature of the steroid nucleus. Lactones 1 and 10 are weak estrogens, while lactones 2 and 28 do not exert estrogenic activity, in fact lactone 2 is an antiestrogen. Lactones 1, 2, 10 and 28 were also tested in an identical assay with a series of enzyme substrates, C19-steroid diols, and known inhibitors, for the oxidation of testosterone and estradiol into androstenedione and estrone, respectively. From this comparative study, the best inhibitors of type II 17b-HSD (oxidase activity) were identified, but none of them were clearly more potent than the hydroxylated (reduced) forms of enzyme substrates, E2, T, and DHT. Such inhibitors remain, however, useful tools to, (1) further elucidate the role of type II 17b-HSD, and (2) regulate the level of active estrogens, androgens and progesterone. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Steroids; Inhibitors; Enzyme; Dehydrogenases; 17b-HSD; Steroidogenesis; Hormones

1. Introduction The 17b-hydroxysteroid dehydrogenases (17b-HSDs) plays an important role in the regulation of steroid hormones, such as estrogens and androgens, by catalysing the reduction of 17-ketosteroids or the oxidation of 17b-hydroxysteroids using NAD(P)H or NAD(P)+ as cofactor (Labrie et al., 1997; Moghrabi and Andersson, 1998). Being involved in the last step of the biosynthesis of sex steroids from cholesterol, the family of 17b-HSD constitutes an interesting target to control the concentration of estrogens and androgens. The 17b-HSD activities are widespread in human tissues, not only in classic steroidogenic tissues, such as the testis, ovary, and placenta, but also in a large series of peripheral intracrine tissues (Martel et al., 1992). The * Corresponding author. Tel.: +1-418-6542296; fax: + 1-4186542761. E-mail address: [email protected] (D. Poirier).

cytosolic human placenta estradiol dehydrogenase (E.C.1.1.1.62) was the first 17b-HSD reported (Ryan and Engel, 1953) and also the most studied. With the nineties, however, new types of 17b-HSD were reported (Labrie et al., 1997; Peltoketo et al., 1999) indicating that a fine regulation is carried out. More importantly, each type of 17b-HSD has a selective substrate affinity, directional (reductase or oxidase) activity in intact cells, and a particular tissue distribution. Thus in human, 17b-HSD type I catalyzes the reduction of estrone (E1) and dehydroepiandrosterone (DHEA), type II catalyzes the oxidation of estradiol (E2), testosterone (T), dihydrotestosterone (DHT), and 20a-dihydroprogesterone (20a-DHP), type III and type V are responsible for the reduction of androstenedione (D4-dione), type IV catalyses the oxidation of E2, while type VIII is preferentially an oxidative enzyme transforming E2, T and DHT (Peltoketo et al., 1999). Fundamentally, these findings are important in understanding the mode of action of the 17b-HSD family. From a therapeutic

0303-7207/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 3 - 7 2 0 7 ( 0 0 ) 0 0 4 2 7 - 5

120

D. Poirier et al. / Molecular and Cellular Endocrinology 171 (2001) 119–128

point of view, this means that selectivity of drug action could be achieved by targetting a particular type of the 17b-HSD isozymes. Consequently, each study that leads to a better knowledge of the inhibition of 17bHSDs deserves attention from scientists working in this and related fields. Thus, inhibitors of 17b-HSDs are useful tools to elucidate the role of these enzymes, in particular, biological systems or for a therapeutic purpose. During the past few years, we have been developping inhibitors of reductive 17b-HSDs, type I (Poirier et al., 1998; Tremblay and Poirier, 1998), type III (Maltais and Poirier, 1998; Tchedam Ngatcha, 1999; Tchedam Ngatcha et al., 2000) and type V (Labrie et al., 1999), especially to block the formation of active hydroxysteroid that would interfere with estrogen-sensitive pathologies (breast, ovarian, and endometrium cancers) and androgen-sensitive pathologies (prostate cancer, benign prostatic hyperplasia, acne, hirsutism, etc). In this article, however, we will focus exclusively on the inhibitors of oxidative type II 17b-HSD. The presence of type II 17b-HSD or human placenta microsomal 17b-HSD was first revealed by Blomquist et al. (1985). After isolation and characterization (Wu et al., 1993; Andersson, 1995; Durocher et al., 1995), it was shown that type II 17b-HSD is a protein of 387 amino acids (42 782 Da) containing an amino-terminal signal-anchor motif and a carboxy-terminal endoplasmic reticulum retention motif. Type II 17b-HSD catalyzes the interconversion of T and D4-dione, DHT and

Fig. 1. Reactions catalysed by type II 17b-HSD and cofactors NAD(P)H or NAD(P)+. The Km values are from Wu et al. (1993).

androstanedione as well as E2 and El (Fig. 1). This isozyme also catalyzes the interconversion of 20a-DHP and progesterone to a lower extent. Experiments with cellular models revealed that the type II prefers an oxidative process using NAD+ as cofactor (Luu-The et al., 1995; Miettinen et al., 1996). The in vitro kinetic data showed that the substrate binding affinity was between 0.21 and 0.71 mM for the reduced substrates reported above (Wu et al., 1993). Type II 17b-HSD is expressed mainly in the placenta, liver, small intestine, endometrium, kidney, pancreas and colon (Casey et al., 1994; Moghrabi and Andersson, 1998). It has also been reported that benign and malignant prostate tissues as well as human meningioma tumors possess significant levels of type II 17b-HSD activity (Carsol et al., 1994, 1996; Elo et al., 1996). These data suggest that type II 17b-HSD inactivates the reduced steroidal hormone by oxidation in peripheral tissues. In classical endocrine tissues such as endometrium and placenta, it has been hypothesized that the type II isozyme together with the type I may control the intracellular ratio of active C18 and C19 steroid hormones during pregnancy (Beaudoin et al., 1995; Blomquist and D’Ascoli, 1995; Blomquist et al., 1994; Takeyama et al., 1998). Moreover, the 20a-HSD activity may provide high progesterone levels during pregnancy (Wu et al., 1993). The type II cDNA sequence showed only 20% homology with type I 17bHSD (Wu et al., 1993). Although the 3-D structures of cytosolic type I 17b-HSD complexed or not with steroid have been elucidated (Ghosh et al., 1995; Azzi et al., 1996; Mazza et al., 1998; Lin et al., 1999; Sawicki et al., 1999), the structure of type II remains unknown. Therefore, synthetic inhibitors should help to probe the active site and possibly clarify the role of type II 17b-HSD, which is not fully understood. To our knowledge, Blomquist and collaborators reported the first inhibitors of type II 17b-HSD in 1984, in a study investigating the ability of naturally occuring steroids, synthetic derivatives of steroids, and a series of non-steroidal agonists or antagonists of the estrogen receptor to inhibit the oxidation of testosterone by microsomal 17b-HSD (Blomquist, 1995; Blomquist et al., 1984). Among the Ki values of tested compounds, which range from 0.3 to 47 mM, three compounds emerged — ethynylestradiol (0.3 mM), estradiol (0.8 mM), and danazol (0.6 mM). The former two are, however, fully agonist on the estrogen receptor (von Angerer, 1995), while the latter is known for its weak androgenic property (Dmowski et al., 1971) and as a weak inhibitor of steroid sulfatase (Poirier et al., 1999). Specific inhibitors could be useful as tools to further investigate the fine regulation and to clarify the role of type II 17b-HSD in particular biological systems (LuuThe et al., 1995; Blomquist et al., 1997). Taking account of the lack of type II 17b-HSD inhibitors, we started the development of various kinds of steroidal inhibitors.

D. Poirier et al. / Molecular and Cellular Endocrinology 171 (2001) 119–128

2. Materials and methods

2.1. Inhibition of microsomal 17b-HSD or type II 17b-HSD from human placenta microsomes (reductase acti6ity) The partially purified 17b-HSD activity obtained from the microsomal fraction of human placenta was used to evaluate the ability of tested compounds to inhibit the transformation of [3H] 4-androstene-3,17dione ([3H] D4-dione) to [3H] testosterone ([3H] T). Briefly, the enzymatic assay was performed in a phosphate-based buffer at pH 7.4 with a low concentration of tritiated D4-dione (3 – 6 nM) to approximate physiological conditions. NADH (1 mM) was used as a cofactor, and the time of incubation and temperature were fixed at 1 h and 37°C, respectively. The residual aromatase activity of human placenta microsomes, which aromatizes the substrate [3H] D4-dione into [3H] estrone, was blocked with a selective aromatase inhibitor. Several concentrations of the tested compound (1 nM–10 mM) in EtOH were used to obtain the inhibition curve, which enabled us to determine the IC50 value, or the concentration of inhibitor that inhibited 50% of target enzyme activity. For screening experiments, only one or two concentrations of inhibitor was used to determine the percent of inhibition. The details of the enzymatic assay procedure were already reported (Auger et al., 1994; Sam et al., 1995).

2.2. Inhibition of transfected type II 17b-HSD (reductase acti6ity) Transfected human embryonal kidney (HEK)-293 cells with cDNA encoding for type II 17b-HSD were sonicated to liberate the crude enzyme that was used as the enzymatic pool without further purification (LuuThe et al., 1995). The enzymatic assay was performed as follows: a stock solution was first prepared containing the radiolabelled substrate [3H] D4-dione (5 nM), NADH (1 mM) in a phosphate buffer (pH 7.4, 50 mM KH2PO4, EDTA 1 mM, 20% glycerol). For the assay, 890 ml of the stock solution and 10 ml of a solution of inhibitor (EtOH or EtOH/DMSO, 95:5) were added in a tube. The reaction was started by adding 100 ml of a solution of crude enzyme (0.4 mg protein per tube) and the mixture was incubated for 1 h at 37°C. The percent of inhibition and the IC50 value were then calculated. The details of the enzymatic assay procedure were already reported (Tremblay et al., 1999).

121

37°C, pH 7.4) the oxidation of [14C] T (0.1 mM) to [14C] D4-dione or [14C] E2 (0.1 mM) to [14C] E1, using NAD+ (1 mM) as cofactor. The enzymatic assay was performed exactly as described in our earlier report (LuuThe et al., 1995).

2.4. Inhibitor sources The chemical synthesis of inhibitors 1–28 were already reported by us (Bydal, 1997; Sam et al., 1995, 2000; Tremblay et al., 1999). The other inhibitors tested in Fig. 8 were commercially available from Steraloids Inc. (Wilton, NH) or Aldrich-Sigma (Milwaukee, WI). Danazol (Win-17757) was supplied by Sterling– Winthrop Research Institute (Rensselaer, NY).

3. Results

3.1. Preliminary SAR study with C18 -steroid nucleus We started our study on type II 17b-HSD inhibitors in 1992. At that time, only three 17b-HSDs were reported in literature (Penning and Ricigliano, 1991) and few information were known about the human placenta microsomal 17b-HSD (later named type II 17b-HSD). Furthermore, it was known that the type II as well as other isoforms catalyze the reversible transformation (reduction and oxidation) of both substrate forms, ketosteroids and hydroxysteroids, but the preferential sense of the reaction catalyzed by each isozyme, particularly in intact cells, was not well established yet. It is why, the reductive transformation of tritiated D4-dione into tritiated T was selected for our preliminary structure–activity relationship (SAR) study. For enzymatic assay, we also selected a concentration of substrate and pH closely to physiological conditions. The study was performed using a series of 16a-derivatives of E2, and 17a-derivatives of E2, previously synthesized to inhibit type I 17b-HSD (Sam et al., 1998). As indicated in Table 1, only the spiro-g-lactone derivative of E2, inhibited the microsomal 17b-HSD (81%). Competition of labelled D4-dione by increased concentration of spiro-glactone 1 or unlabelled D4-dione gave the inhibition curves shown in Fig. 2A, which correspond to IC50 of 0.25 and 1.41 mM, respectively. No inactivation of enzyme was also observed according to time (Fig. 2B) suggesting that compound 1 acted as a reversible inhibitor (Sam et al., 1995).

3.2. Endocrine-profile modification of lactone 1 2.3. Inhibition of transfected type II 17b-HSD (oxidase acti6ity) The same enzymatic source as above, transfected HEK-293 cells sonicated, was used to catalyze (1–2 h,

To modify the residual estrogenic activity of C18steroid inhibitor 1, we decided to add the spiro-g-lactone functionality (the inhibiting group) to an antiestrogenic nucleus. It is well known that the long

122

D. Poirier et al. / Molecular and Cellular Endocrinology 171 (2001) 119–128

Table 1 Inhibition of microsomal 17b-HSD ([3H] D4-dione to [3H] T) by a series of E2 derivativesa

a

Error, 4%; inhibitor concentration, 1 mM; TBDMS, tert-butyldimethylsilyl.

methyl butyl alkanamide side chain at position 7a of E2 plays an essential role in the antiestrogenic activity of ICI-164384 (Wakeling and Bowler, 1987; Bowler et al., 1989) and EM-139 (Le´vesque et al., 1991). The target compound 2 was then synthesized and biological activity evaluated (Sam et al., 2000). Inhibition of microsomal 17b-HSD was demonstrated by an IC50 value of 0.35 mM for compound 2. In the absence of the 7aalkylamide chain, the corresponding lactone 1 had an IC50 value of 0.25 mM. Both lactones (1 and 2) fit the enzymatic pocket better than did the substrate itself, as indicated by the IC50 value of 1.27 mM obtained when unlabelled D4-dione was used to compete with labelled D4-dione (Fig. 3). Since we reported above that lactone 1 acts as a reversible inhibitor, we expect that lactone 2 with its spiro-g-lactone will share this same mechanism. Compound 2 also inhibits selectively the type 2 17bHSD (no inhibition was observed toward type 1 17bHSD) due mostly to the presence of the spiro-g-lactone group. In addition to its capacity to inhibit the microsomal 17b-HSD activity, lactone 2 exerts neither androgenic activity nor estrogenic activity and, furthermore, the alkylamide side chain at position 7a of the steroidal nucleus blocked the estrogenic activity associated with lactone 1, thus conferring antiestrogenic activity (Fig. 4).

study, (1) the substitution of OH; (2) the reduction of carbonyl into methylene; (3) the stereochemistry of lactone functionality (Cl7a or C17b orientation of oxygen); (4) the position of the lactone attachment (Cl7a/ C17b-O vs. C16b/C17b-O); and (5) the size of lactone ring. As shown in Table 2, a higher inhibition of microsomal 17b-HSD was obtained with the phenolic derivative 1 (81%), whereas we observed an important decrease of inhibitory activity with methoxy 3 (55%) and benzyloxy 4 (11%) derivatives. It appears that bulkier the chemical group in position 3, the lower the inhibition is. The presence of the carbonyl of lactone functionality is essential because the saturated analogue 5 exhibited only a weak inhibition (4%) of microsomal 17b-HSD. The crucial importance of 17b-orientation of oxygen atom was also established since the 17a-O analogue, compound 6, was 100-fold less potent inhibitor than lead lactone 1. For the same lactone ring size, the spiranic (C17/C17) form of lactone provoked a better inhibition than the other studied form (C16b/

3.3. Optimization of lactone 1 To optimize the lead compound (spiro-g-lactone 1), identified from our preliminary SAR study, a series of C18-steroid lactone derivatives 3 – 11 (Table 2) were synthesized and tested (Bydal, 1997). We introduced five different kinds of modifications in our second SAR

Fig. 2. Inhibition of microsomal 17b-HSD by spiro-g-lactone 1. (A) Curves of inhibition and (B) experiment of inactivation.

D. Poirier et al. / Molecular and Cellular Endocrinology 171 (2001) 119–128

Fig. 3. Inhibition of microsomal 17b-HSD (comparison of lactones 1 and 2).

C17b-O). Thus, no inhibition was observed for compound 7 at 1 mM compared with 85% for spiro-g-lactone 1. Similarly, weak inhibition was obtained for the six-member ring analogue 8 (7 and 44%) compared with corresponding spiro-d-lactone 10 (99 and 100%), respectively, at 1 and 10 mM. The inhibitory activity increases, however, with the size of the C16b/Cl7b-lactone (compare 7, 8 and 9). Among the three spirolactones (g, d, and o) synthesized, compounds 1, 10 and 11, the six-member d-lactone 10 emerged clearly as the most potent inhibitor with IC50 values of 6 nM (Fig. 5). Thus, the spiro-d-lactone 10 was 150-fold more potent than D4-dione used itself as inhibitor regarding the reduction of labelled D4-dione by transfected 17b-HSD.

3.4. SAR study with C19 -steroid lactones 12 – 19 The type II 17b-HSD is the only one of known human 17b-HSD isoforms that catalyses the transformation of both estrogenic (C18-steroids) and androgenic (C19-steroids) substrates. A spirolactone functionality is furthermore an important requirement to inhibit the type II 17b-HSD. It is thus likely that a

Fig. 4. Proliferative and antiproliferative activities of compound 2 on AR+ Shionogi cells (A) and ER+ ZR-75-1 cells (B).

123

C19-steroid nucleus bearing a spirolactone functionality might be a good inhibitor. A series of spiro-g-lactones was thus synthesized with modification mainly on A and/or B ring and tested for their inhibitory effects on microsomal 17b-HSD (D4-dione to T). Surprisingly, only three lactones showed a significant but weak inhibitory effect (25–28% of inhibition) on microsomal 17b-HSD activity (Table 3). In contrast, the C18 analogue spiro-g-lactone 1 showed 81% inhibition. Among those C19 steroid derivatives, the well-known spironolactone (16) showed the highest inhibitory effect. The backbone of spironolactone resembles spiro-g-lactone 14, but the presence of an acetylthio residue at position7a seemed to improve the inhibition of microsomal 17b-HSD (compare 14 with 16). The results suggest also that introducing one or two double bonds in the A-ring decreased the inhibitory effect of the spiro-g-lactone (compare 13, 14 and 15 with 12; compare 17 with 18). We also synthesized the bis spiro-g-lactone 19 and observed that the presence of an additional lactone group at position 3 of the A-ring lowered the inhibitory effect (compare 19 with 12). The spiro-g-lactone derivatives of C19-steroid were weak inhibitors instead of their structural analogy with C19-substrates of type II 17b-HSD, D4-dione and T. An inhibitory effect of a 7a-acetylthio (SCOCH3) residue was, however, observed suggesting a possible optimization.

3.5. Optimization of spironolactone 16 To further characterize the importance of a 7aacetylthio group to enhance the inhibitory potency of C19-steroid derivatives bearing a spiro-g-lactone, the 7a-acetylthio of spironolactone 16 was substituted with various thioalkyl and thioaryl side chains (Table 4). A 2-(1-piperidinyl)-ethoxybenzyl group was also introduced at the 7a position to assess the ability of type II 17b-HSD to tolerate this large pharmacophore and possibly modify the endocrine profile of the C19-steroid nucleus. A screening test was first performed with compounds 20–26 using the reductase activity of transfected type II 17b-HSD. The reductive activity was initially tested to make the connection with our previously published work that was reported before the preferred oxidase activity of the enzyme was established. It can be seen that neither compound 20 without substitution at position 7a nor compounds bearing a thioester group (16 and 24) are very potent against type II 17b-HSD. On the other hand, thioalkyl (21–23) and particularly thioaryl compounds (25 and 26) displayed a more potent inhibitory effect. The 7a-thiobenzyl analogue 26 exhibited the highest inhibitory effect with an IC50 value of 0.42 mM, while other compounds showed IC50 values close to micromolar. The thiobenzyl compound 26, phenolic derivative 27 and its O-substituted analogue 28 were next tested using the oxidase activity

124

D. Poirier et al. / Molecular and Cellular Endocrinology 171 (2001) 119–128

Table 2 Inhibition of transfected type II 17b-HSD ([3H] D4-dione to [3H] T) by a series of C18-steroid lactones

of transfected type II 17b-HSD transforming T into D4-dione (Table 4). Thus, the phenolic derivative 27 (IC50 = 0.53 mM) is slighly better than its substituted analogue 28 (IC50 =0.7 mM), which is more potent than the lead compound 16 (IC50 =1.1 mM) and 26 (IC50 = 1.0 mM). It is interesting to note that compounds 27 and 28 displayed inhibitory potency very similar to our previously reported C18-steroid inhibitor 1 (IC50 = 0.7 mM). Compound 28 also inhibits the oxidative transformation of E2 into E1 (Fig. 6). Contrary to the firstly reported inhibitor 1 that efficiently bound to the estrogen receptor, none of the spironolactone analogs 16 – 28 had any affinity for this receptor. However, several 7a-derivatives of spironolactone bound to progestin or glucocorticoid receptors and, to a lesser extent, to androgen receptor. Among the tested compounds, only 28 was devoid of any affinity for all of the tested steroid receptors (Tremblay et al., 1999). This compound did not display significant proliferative/antiproliferative activity on estrogen-sensitive (ER+) ZR-75-1 cells and showed weak antiproliferative effect on androgen-sensitive (AR+) Shionogi cells (Fig. 7). Moreover, compound 28 did not show any inhibitory activity against reductive isozymes I, III, and V of 17b-HSD as well as P450 aromatase when tested at 3 mM (Tremblay et al., 1999).

4. Discussion Based on the observation that a spiro-g-lactone functionality is an important requirement of type II 17bHSD inhibition, we have developed inhibitors using different steroid nucleus, viz. a C18-steroid (lactones 1 and 10), an antiestrogenic nucleus (lactone 2) and a

Fig. 5. Inhibition of transfected type II 17b-HSD by spirolactones g (1), d (10) and o (11) and unlabelled D4-dione.

D. Poirier et al. / Molecular and Cellular Endocrinology 171 (2001) 119–128

125

Table 3 Inhibition of microsomal 17b-HSD ([3H] D4-dione to [3H] T) by a series of C19-steroid lactones

C19-steroid (lactone 28). These inhibitors are also selective for type II 17b-HSD as no or only weak inhibition was observed for types I and III. They also have no proliferative (androgenic) activity on androgen sensitive (AR+) Shionogi cells while their proliferative (estrogenic) activity on estrogen sensitive (ER+) ZR-75-1 cells depends on the nature of the steroid nucleus. Lactones 1 and 10 are weak estrogen while lactones 2 and 28 do not exerts estrogenic activity. In addition to its inhibitory activity on type II 17b-HSD, lactone 2 is also an antiestrogen. Few compounds were reported to inhibit type II 17b-HSD activity (Penning, 1996). Estradiol, ethynylestradiol and danazol (Ki =0.8, 0.3 and 0.6 mM, respectively) emerged from the pionneer work of Blomquist et al. (1985). They also reported that naturally occurring steroids (5-androsten-3b,17b-diol, 20aDHP, 5a-DHT and 5b-DHT) inhibited testosterone oxidation by microsomal 17b-HSD with Ki values ranging from 1.4 to 1.9 mM. In 1994, some retinoids were tested as inhibitors of microsomal 17b-HSD oxidoreductive activities in rat liver (Murray et al., 1994). 13-cis-Retinoic acid and 9-cis-retinoic acid inhibited testosterone oxidation by type II 17b-HSD with a Ki of 2.4 and 4.1 mM, but the reduction of D4-dione was less sensitive to inhibition (Ki =27 and 35 mM, respectively). More recently, Puranen et al. (1999) characterized the substrate specificity of human type II 17b-HSD, using a series of C18, C19 and C21 steroids that were tested for their ability to compete with 0.2 mM of estradiol. The best inhibition percentages at 10 mM were obtained with 17b-dihydroequilin (DHE; 88%), 5a-androstan-3a,17b-diol (87%), 5-androsten3b,17b-diol (83%), 5b-androstan-3a,17b-diol (83%), 2hydroxyestradiol (67%), 5a-androstan-3b,17b-diol (55%), ethynylestradiol (54%), and enzyme substrates 20a-DHP (71%) and testosterone (70%). We then decided to compare most of the inhibitors reported above in an identical protocol (Fig. 8). Be-

cause it is now well established that oxidation is the preferential reaction catalyzed by type II 17b-HSD, we studied the transformations of [14C] T (0.1 mM) into [14C] D4-dione and [14C] E2 (0.1 mM) into [14C] E1 using NAD+ (1 mM) as cofactor. The reaction time, temperature and pH were fixed to 2 h, 37°C, and 7.4, respectively, and radioactivity associated with the steroids (remaining substrate and product formed) was measured to calculate the percent of inhibition at two concentrations of inhibitor (0.1 and 1.0 mM). As expected, the reduced form (17b-OH) of substrates (T, DHT, and E2) were better competitors of labelled T and labelled E2 than the oxidized forms (D4-dione and E1). On the other hand, C21 steroid substrates, 20aDHP and Prog, were weak competitors for the type II 17b-HSD transformation of C18 and C19 substrates. The importance of a 17b-OH was also observed in the series of C19-steroid diols (37–69% of inhibition at 1 mM). DHE, which differs from C18 substrate E2 by only an additional double bond in B-ring (D7,8), strongly inhibited the enzyme (53 and 64%). DHE and E2 both have a 17b-OH. Contrary to a first report and despite the presence of a 17b-OH, EE2 is not a good inhibitor (0–24%). In this case, the 17b-ethynyl group had a negative effect on the inhibitory activity of 17b-OH, probably through an attracting effect or a steric hindrance. Danazol, a C19 derivative with an additional heterocycle in positions 2–4, moderately inhibited the enzyme (23–34%). Interestingly, danazol has a 17b-OH and a 17b-ethynyl as EE2. The only nonsteroidal inhibitor tested, the 9-cis-retinoic acid, weakly inhibited the enzyme (16–20%). It was not, however, possible to test the 13-cis retinoic acid that was previously reported as a better inhibitor than the 9-cis analogue. For the spirolactone derivatives represented by inhibitors 1, 2, 10, and 28, the percent of inhibition ranged from 29 to 69%, reflecting the results previously obtained for the reductive transformation of labelled D4-dione. Thus, spiro-g-lactone 1 and the ana-

126

D. Poirier et al. / Molecular and Cellular Endocrinology 171 (2001) 119–128

Table 4 Inhibition of transfected type II 17b-HSD by a series of spironolactone analogues

logue 2 with a bulky alkylamide side chain inhibited the oxidase activity of type II 17b-HSD roughly at a same level (30–46%). The C19 steroid 28, bearing a spiro-glactone and a bulky piperidinylethoxybenzyl group also moderately inhibited the enzyme (29 – 37%). As previously observed for the reductase activity of type II 17b-HSD, the spiro-d-lactone 10 was the best inhibitor for oxidase activity. With 62 – 66% of inhibition, the potency of lactone 10 was similar to the best of 17bOH-substrates (43– 65%) or naturally occurring C19steroid diols (55–69%). In general, however, the spirolactone inhibitors were more potent when inhibiting the reductase than oxidase transformation. Since the presence of a spiro-d-lactone increased significantly the ability of spirolactone to inhibit the type II 17bHSD, the potency of dual inhibitors 2 and 28 could be increased by changing the spiro-g-lactone for a spiro-dlactone. In summary, we report the development of steroidal spirolactones 1, 2, 10, and 28, which inhibit type II 17b-HSD and have different biological characteristics. These inhibitors were also tested in a same assay with a series of enzyme substrates, C19-steroid diols and inhibitors reported in the literature. From this comparative study, the best inhibitors of oxidase activity were identified, but none of them were clearly more potent than the hydroxylated (reduced) forms of enzyme substrates, E2, T, and DHT, independently of the labelled substrate used (T or E2). Since it is now established that type II 17b-HSD prefers the oxidase activity in intact cells, and is involved in degradation of active estrogens and androgens, the inhibition of type II is not desirable for use in a therapy of hormono-sensitive cancers. Such inhibitors remain, however, useful tools to, (1) help elucidate the role of type II 17b-HSD; and (2) better

regulate the level of active estrogens, androgens and progesterone. In this light, additional work will be necessary to improve the potency of known inhibitors or to develop new ones.

Acknowledgements We thank the Medical Research Council of Canada (MRC) and Le Fonds de la Recherche en Sante´ du Que´bec (FRSQ) for operating grants. We gratefully acknowledge Serge Auger and Mei Wang for their participation to the enzymatic assays, Fernand Labrie, Jacques Simard and Diane Michaud for providing biological evaluation on Shionogi and ZR-75-1 cells, and Caroline Mercier for the chemical synthesis of compound 2 and dihydroequilin.

Fig. 6. Inhibition (%) of the oxidase activity of transfected type II 17b-HSD according to substrates (T and E2).

D. Poirier et al. / Molecular and Cellular Endocrinology 171 (2001) 119–128

Fig. 7. Proliferative and antiproliferative activities of C19-steroid spiro-g-lactone 28 on AR+ Shionogi cells (A) and ER+ ZR-75-1 cells (B).

Fig. 8. Comparison of various compounds reported to inhibit type II 17b-HSD. Each compound was dissolved in EtOH and tested in duplicate on the transformation of 0.1 mM of labelled substrate (T and E2) by transfected type II 17b-HSD in HEK-293 cells. E2, estradiol; El, estrone; T, testosterone; D4-dione, 4-androsten-3,17dione; DHT, dihydrostesterone; 20a-DHP, 20a-dihydroprogesterone; Prog, progesterone; D5-diol, 5-androsten-3b,17b-diol; (5a)-3b-diol, 5a-androstan-3b,17b-diol; (5b)-3a-diol, 5a-androstan-3a,17b-diol; EE2, ethynylestradiol; DHE, dihydroequilin; danazol, 17a-ethynyl17b-hydroxy-4-androsteno-[2,3-d]isoxazole; 9-cis-RA, 9-cis-retinoic acid.

References Andersson, S., 1995. Molecular genetics of androgenic 17b-hydroxysteroid dehydrogenases. J. Steroid Biochem. Mol. Biol. 55, 533 – 534. Auger, S., Luu-The, V., Sam, K.M., Poirier, D., 1994. 3-Hydroxy-19-

127

nor-17b-pregna-1,3,5(10)-triene-21,17b-carbolactone as inhibitor of 17b-hydroxysteroid dehydrogenase type 2. Bioorg. Med. Chem. Lett. 4, 2045 – 2048. Azzi, A., Rehse, P., Zhu, D.W., Campbell, R.L., Labrie, F., Lin, S.X, 1996. Crystal structure of human estrogenic 17b-hydroxysteroid dehydrogenase complexed with 17b-estradiol. Nat. Struct. Biol. 3, 665 – 668. Beaudoin, C., Blomquist, C.H., Tremblay, Y., 1995. Gene expression of 17b-hydroxysteroid dehydrogenase type 2 isozyme in primary cultures of human trophoblasts predicts different mechanisms regulating type 1 and type 2 enzymes. Endocrinology 136, 3807– 3814. Blomquist, C.H., 1995. Kinetic analysis of enzymic activities: prediction of multiple forms of 17b-hydroxysteroid dehydrogenase. J. Steroid Biochem. Mol. Biol. 55, 515 – 524. Blomquist, C.H., D’Ascoli, P.T., 1995. Gestational development of human placental 17b-hydroxysteroid oxidoreductase types 1 and 2. Hum. Reprod. 10, 2685 – 2689. Blomquist, C.H., Lindemann, N.J., Hakanson, E.Y., 1984. Inhibition of 17b-hydroxysteroid dehydrogenase (17b-HSD) activities of human placenta by steroids and non-steroidal hormone agonists and antagonists. Steroids 43, 571 – 586. Blomquist, C.H., Lindemann, N.J., Hakanson, E.Y., 1985. 17b-Hydroxysteroid and 20a-hydroxysteroid dehydrogenase activities of human placental microsomes: kinetic evidence for two enzymes differing in substrate specificity. Arch. Biochem. Biophys. 239, 206 – 215. Blomquist, C.H., Bealka, D.G., Hensleigh, H.C., Tagatz, G.E., 1994. A comparison of 17b-hydroxysteroid oxidoreductase type 1 and type 2 activity of cytosol and microsomes from human term placenta, ovarian stroma and granulosa-luteal cells. J. Steroid Biochem. Mol. Biol. 49, 183 – 189. Blomquist, C.H., Leung, B.S., Beaudoin, C., Poirier, D., Tremblay, Y., 1997. Intracellular regulation of 17b-hydroxysteroid dehydrogenase type 2 catalytic activity in A431 cells. J. Endocrinol. 153, 453 – 464. Bowler, J., Lilley, T.J., Pittam, J.D., Wakeling, A.E., 1989. Novel steroidal pure antiestrogens. Steroids 54, 71 – 99. Bydal, P., 1997. Synthe`se de de´rive´s de l’oestradiol, comportant une lactone sur le cycle D, comme inhibiteurs des isoformes 2 et 5 de la 17b-hydroxyste´roide de´shydroge´nase. M.Sc. thesis. Faculte´ de me´decine, Universite´ Laval, Que´bec, Canada. Carsol, J.L., Martin, P.M., de Launoit, Y., 1994. Characterization of 17b-hydroxysteroid dehydrogenase activity and mRNA abundance in human meningioma tumors. Neuroendocrinology 60, 445 – 451. Carsol, J.L., Adamski, J., Guirou, O., Gerard, H., Martin, P.M., de Launoit, Y., 1996. 17b-Hydroxysteroid dehydrogenase activity correlates with the type-2 17 beta-hydroxysteroid dehydrogenase mRNA abundance in human meningioma tumors. Neuroendocrinology 64, 70 – 78. Casey, M.L., MacDonald, P.C., Andersson, S., 1994. 17b-Hydroxysteroid dehydrogenase type 2: chromosomal assignment and progestin regulation of gene expression in human endometrium. J. Clin. Invest. 94, 2135 – 2141. Dmowski, W.P., Scholer, H.F.L., Mahesh, V.B., Greenblatt, R.B., 1971. Danazol — a synthetic steroid derivative with interesting physiologic properties. Fertil. Steril. 22, 9 – 18. Durocher, F., Morissette, J., Labrie, Y., Labrie, F., Simard, J., 1995. Mapping of the HSD17B2 gene encoding type II 17b-hydroxysteroid dehydrogenase close to D16S422 on chromosome 16q 24.1 – q 24.2. Genomics 25, 724 – 726. Elo, J.P., Akinola, L.A., Poutanen, M., Vihko, P., Kyllonen, A.P., Lukkarinen, O., Vihko, R., 1996. Characterization of 17b-hydroxysteroid dehydrogenase isoenzyme expression in benign and malignant human prostate. Int. J. Cancer 66, 37 – 41.

128

D. Poirier et al. / Molecular and Cellular Endocrinology 171 (2001) 119–128

Ghosh, D., Pletnev, V.Z., Zhu, D.W., Wawrzak, Z., Duax, W.L., Pangborn, W., Labrie, F., Lin, S.X., 1995. Structure of human estrogenic 17b-hydroxysteroid dehydrogenase at 2.20A resolution. Structure 3, 503 – 513. Labrie, F., Luu-The, V., Lin, S.X., Labrie, C., Simard, J., Breton, R., Be´langer, A., 1997. The key role of 17b-hydroxysteroid dehydrogenases in sex steroid biology. Steroids 62, 148–158. Labrie, F., Be´langer, A., Gauthier, S., Luu-The, V., Me´rand, Y., Poirier, D., Provencher, L., Singh, S.M., 1999. Inhibitors of type 5 and type 3 17b-hydroxysteroid dehydrogenase and methods for their use. WO9946279. Le´vesque, C., Me´rand, Y., Dufour, J.M., Labrie, C., Labrie, F., 1991. Synthesis and biological activity of new halo-steroidal antiestrogens. J. Med. Chem. 34, 1624–1630. Lin, S.X., Han, Q., Azzi, A., Zhu, D.W., Gangloff, A., Campbell, R.L., 1999. 3D-structure of human estrogenic 17b-HSD1: binding with various steroids. J. Steroid Biochem. Mol. Biol. 69, 425 – 429. Luu-The, V., Zhang, Y., Poirier, D., Labrie, F., 1995. Characteristics of human types 1, 2 and 3 17b-hydroxysteroid dehydrogenase activities: oxidation – reduction and inhibition. J. Steroid Biochem. Mol. Biol. 55, 581 – 587. Maltais, R., Poirier, D., 1998. A solution-phase combinatorial parallel synthesis of 3b-amido-3a-hydroxy-5a-androstane-17-ones. Tetrahedron Lett. 39, 4151–4154. Martel, C., Rhe´aume, E., Takahashi, M., Trudel, C., Couet, J., Luu-The, V., Simard, J., Labrie, F., 1992. Distribution of 17b-hydroxysteroid dehydrogenase gene expression and activity in rat and and human tissues. J. Steroid Biochem. Mol. Biol. 41, 597 – 603. Mazza, C., Breton, R., Housset, D., Fontecilla-Camps, J.C., 1998. Unusual charge stabilization of NADP+ in 17b-hydroxysteroid dehydrogenase. J. Biol. Chem. 273, 8145–8152. Miettinen, M.M., Mustonen, M.V., Poutanen, M.H., Isomaa, V.V., Vihko, R.K., 1996. Human 17b-hydroxysteroid dehydrogenase type 1 and type 2 isoenzymes have opposite activities in cultured cells and characteristic cell- and tissue-specific expression. Biochem. J. 314, 839 –845. Moghrabi, N., Andersson, S., 1998. 17b-Hydroxysteroid dehydrogenases: physiological roles in health and disease. TEM 9, 265 – 270. Murray, M., Butler, A.M., Martini, R., 1994. Inhibition of microsomal 17b-hydroxysteroid oxidoreduction activities in rat liver by all-trans-, 9-cis- and 13-cisretinoic acid. Biochim. Biophys. Acta 1222, 227 – 233. Peltoketo, H., Luu-The, V., Simard, J., Adamski, J., 1999. 17b-Hydroxysteroid dehydrogenase (HSD)/17-ketosteroid reductase (KSR) family; nomenclature and main characteristics of the 17HSD/KSR enzymes. J. Mol. Endocrinol. 23, 1–11. Penning, T.M., Ricigliano, J.W., 1991. Mechanism based inhibition of hydroxysteroid dehydrogenases. J. Enzyme Inhib. 5, 165 – 198. Penning, T.M., 1996. 17b-hydroxysteroid dehydrogenase: inhibitors and inhibitor design. Endocrine-Related Cancer 3, 41–56. Poirier, D., Dionne, P., Auger, S., 1998. A 6b-(thiaheptanamide) derivative of estradiol as inhibitor of 17b-hydroxysteroid dehydrogenase type 1. J. Steroid Biochem. Mol. Biol. 64, 83–90. Poirier, D., Ciobanu, L.C., Maltais, R., 1999. Steroid sulfatase inhibitors. Exp. Opin. Ther. Patents 9, 1083–1099.

.

Puranen, T.J., Kurkela, R.M., Lakkakorpi, J.T., Poutanen, M.H., Itaranta, P.V., Melis, J.P.J., Ghosh, D., Vihko, R.K., Vihko, P.T., 1999. Characterization of molecular and catalytic properties of intact and truncated human 17b-hydroxysteroid dehydrogenase type 2 enzymes: intracellular localization of the wild-type enzyme in the endoplasmic reticulum. Endocrinology 140, 3334 –3341. Ryan, K.J., Engel, L.L., 1953. The interconversion of estrone and estradiol by human tissue slices. Endocrinology 52, 287–291. Sam, K.M., Auger, S., Luu-The, V., Poirier, D., 1995. Steroidal spiro-gamma-lactones that inhibit 17b-hydroxysteroid dehydrogenase activity in human placental microsomes. J. Med. Chem. 38, 4518 – 4528. Sam, K.M., Boivin, R.P., Tremblay, M.R., Auger, S., Poirier, D., 1998. C16 and C17 derivatives of estradiol as inhibitors of 17b-hydroxysteroid dehydrogenase type 1: chemical synthesis and structure – activity relationships. Drug Des. Discov. 15, 157 –180. Sam, K.M., Labrie, F., Poirier, D., 2000. N-butyl, N-methyl-11-(3%hydroxy-21%,17%-carbolactone-19%-nor-17%a-pregna-l%,3%,5% (10%)trien-7%a-yl) undecanamide: as inhibitor of type 2 17b-hydroxysteroid dehydrogenase that does not have estrogenic or androgenic activity. Eur. J. Med. Chem. 35, 217 – 225. Sawicki, M.W., Erman, M., Puranen, T., Vihko, P., Ghosh, D., 1999. Structure of the ternary complex of human 17b-hydroxysteroid dehydrogenase type 1 with 3-hydroxyestra-1,3,5,7-tetraen-17-one (equilin) and NADP+. Proc. Natl. Acad. Sci. USA 96, 840–845. Takeyama, J., Sasano, H., Suzuki, T., Iinuma, K., Nagura, H., Andersson, S., 1998. 17b-Hydroxysteroid dehydrogenase types 1 and 2 in human placenta: an immunohistochemical study with correlation to placental development. J. Clin. Endocrinol. Metab. 83, 3710 – 3715. Tchedam Ngatcha, B., 1999. Les de´rive´s de l’androste´rone substitue´s en position 16, 3 beta et/ou 3 alpha-O et les ste´roides delta-lactones: synthe`se chimique et inhibition des isoformes 3 et 5 de la 17b-hydroxyste´roide de´shydroge´nase. Ph.D. thesis. Faculte´ de me´decine, Universite´ Laval, Que´bec, Canada. Tchedam Ngatcha, B., Luu-The, V., Poirier, D., 2000. Androsterone 3b-substituted derivatives as inhibitors of type 3 17b-hydrosysteroid dehydrogenase. Bioorg. Med. Chem. Lett. 10, 2533–2536. Tremblay, M.R., Poirier, D., 1998. Overview of a rational approach to design type I 17b-hydroxysteroid dehydrogenase inhibitors without estrogenic activity: chemical synthesis and biological evaluation. J. Steroid Biochem. Mol. Biol. Vol. 66, 179 – 191. Tremblay, M.R., Luu-The, V., Leblanc, G., Noel, P., Breton, E., Labrie, F., Poirier, D., 1999. Spironolactone-related inhibitors of type II 17b-hydroxysteroid dehydrogenase: chemical synthesis, receptor binding affinities, and proliferative/antiproliferative activities. Bioorg. Med. Chem. 7, 1013 – 1023. von Angerer, E., 1995. In: Molecular Biology Intelligence Unit, The Estrogen Receptor as a Target for Rational Drug Design. R.G. Landes Company, Austin, TX, pp. 31 – 48. Wakeling, A.E., Bowler, J., 1987. Steroidal pure antiestrogens. J. Endocrinol. 112, R7 – R10. Wu, L., Einstein, M., Geissler, W.M., Chan, K.H., Elliston, K.O., Andersson, S., 1993. Expression cloning and characterization of human 17b-hydroxysteroid dehydrogenase type 2, a microsomal enzyme possessing 20a-hydroxysteroid dehydrogenase activity. J. Biol. Chem. 268, 12964 – 12969.