New insights into the SAR and binding modes of bis(hydroxyphenyl)thiophenes and -benzenes: Influence of additional substituents on 17β-hydroxysteroid dehydrogenase type 1 (17β-HSD1) inhibitory activity and selectivity

6724 J. Med. Chem. 2009, 52, 6724–6743

DOI: 10.1021/jm901195w

New Insights into the SAR and Binding Modes of Bis(hydroxyphenyl)thiophenes and -benzenes:

Influence of Additional Substituents on 17β-Hydroxysteroid Dehydrogenase Type 1 (17β-HSD1)

Inhibitory Activity and Selectivity

Emmanuel Bey, Sandrine Marchais-Oberwinkler, Matthias Negri, Patricia Kruchten, Alexander Oster,

Tobias Klein, Alessandro Spadaro, Ruth Werth, Martin Frotscher, Barbara Birk, and Rolf W. Hartmann*

Pharmaceutical and Medicinal Chemistry, Saarland University, P.O. Box 15 11 50, D-66041 Saarbru€cken, Germany

Received August 11, 2009

17β-Hydroxysteroid dehydrogenase type 1 (17β-HSD1) is responsible for the catalytic reduction of

weakly active E1 to highly potent E2. E2 stimulates the proliferation of hormone-dependent diseases via

activation of the estrogen receptor R (ERR). Because of the overexpression of 17β-HSD1 in mammary

tumors, this enzyme should be an attractive target for the treatment of estrogen-dependent pathologies.

Recently, we have reported on a series of potent 17β-HSD1 inhibitors: bis(hydroxyphenyl) azoles,

thiophenes, and benzenes. In this paper, different substituents are introduced into the core structure and

the biological properties of the corresponding inhibitors are evaluated. Computational methods and

analysis of different X-rays of 17β-HSD1 lead to identification of two different binding modes for these

inhibitors. The fluorine compound 23 exhibits an IC₅₀of 8 nM and is the most potent nonsteroidal

inhibitor described so far. It also shows a high selectivity (17β-HSD2, ERR) and excellent pharmaco-

kinetic properties after peroral application to rats.

Introduction

estrone (E1). Until now, three subtypes (1, 7, and 12) are able

to catalyze thisreaction, the most importantbeing17β-HSD1.

The primary physiological role of 17β-HSD7 and 17β-HSD12

is supposed to be in the cholesterol synthesis^6,7and in the

regulation of the lipid biosynthesis,⁸respectively. In addition,

Day et al.⁹recently showed that 17β-HSD12, although highly

expressed in breast cancer cell lines, is inefficient in E2

formation.

Estrogens are involved in the regulation of the female

reproduction system. However, it is also well-known that

17β-estradiol (E2^a), the natural ligand of the estrogen recep-

tors (ERs) R and β, plays a critical role in the development of

several estrogen-dependent pathologies like breast cancer¹

and endometriosis.²

Until now, hormone-dependent breast cancers are treated

using three different endocrine therapies:^3,4aromatase inhibi-

tors and GnRH analogues disrupt the estrogen biosynthesis,

while selective estrogen receptor modulators (SERMs) or

pure antiestrogens⁵prevent E2 to unfold its action at the

receptor level. Besides specific disadvantages of each thera-

peutic approach, all of these strategies have in common a

rather radical reduction of estrogen levels in the whole body

leading to significant side effects.

17β-HSD1, which is responsible for the intracellular NAD-

(P)H-dependent conversion of the weak E1 into the highly

potent estrogen E2, is often overexpressed in breast cancer

cells^10-13and endometriosis.¹⁴Inhibition of this enzyme is

therefore regarded as a promising novel target for the treat-

ment of estrogen-dependent diseases.

Recently, two groups^9,15,16reported on the in vivo efficacy

of 17β-HSD1 inhibitors to reduce E1 induced tumor growth

using two different mouse models and indicating that the 17β-

HSD1 enzyme is a suitable target for the treatment of breast

cancer.

A softer approach could be the inhibition of an enzyme of

the 17β-hydroxysteroid dehydrogenase (17β-HSD) family,

especially one that is responsible for the E2 formation from

In order to not counteract the therapeutic efficacy of 17β-

HSD1 inhibitors, it is very important that the compounds are

selective toward 17β-HSD2, the enzyme that catalyzes the

deactivation of E2 into E1. Additionally, to avoid intrinsic

estrogenic effects, the inhibitors should not show affinity to

the estrogen receptors R and β.

During the past decade, several groups reported on

17β-HSD1 inhibitors, most of them having steroidal struc-

tures.^17-20Recently nonsteroidal cores have been pub-

lished too. Until today four classes of compounds were

described: thienopyrimidinones,^21,22biphenyl ethanones,²³

and from our group (hydroxyphenyl)naphthalenes^24-26

and bis(hydroxyphenyl)azoles, thiophenes, benzenes, and

azabenzenes.^27-32The most promising compounds of the

*To whom correspondence should be addressed. Phone: þ49 681 302

70300. Fax: þ49 681 302 70308. E-mail: rwh@mx.uni-saarland.de. Web

site: http://www.pharmmedchem.de.

^aAbbreviations: 17β-HSD1, 17β-hydroxysteroid dehydrogenase type

1; 17β-HSD2, 17β-hydroxysteroid dehydrogenase type 2; E1, estrone; E2,

17β-estradiol; ER, estrogen receptor; SERM, selective estrogen receptor

modulator; NADP(H), nicotinamide adenine dinucleotide phosphate;

NAD(H), nicotinamide adenine dinucleotide; SUB, substrate binding

site; COF, cofactor binding site; RBA, relative binding affinity; ESP,

electrostatic potential; P_app, apparent permeability coefficient; SAR,

structure-activity relationship; PDB, Protein Data Bank; CC, column

chromatography; MEP, molecular electrostatic potential; HYC, hybrid

inhibitor O5⁰-[9-(3,17β-dihydroxy-1,3,5(10)-estratrien-16β-yl)nonanoyl]-

adenosin; AUC, area under the curve; HOBt, 1-hydroxybenzotriazole;

EDCI, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride.

r

pubs.acs.org/jmc

Published on Web 10/15/2009

2009 American Chemical Society

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6725

Chart 1. Described Bis(hydroxyphenyl)azoles, Thiophenes and Benzenes

Chart 2. Title Compounds

that are important for high 17β-HSD1 inhibitory activity:

(1) one hydroxyphenyl moiety on the core structure is

not sufficient for a high potency; (2) only the meta-para

and meta-meta dihydroxy substitution pattern (O-O

distance in the same range as observed for the steroid,

˚

d = 11 A) are active; (3) the presence of the m-hydroxy

group is more important for inhibitory activity than the

p-hydroxy group; (4) only central aromatic rings without

hydrogen-bond-donor function like thiophene, thiazole,

benzene exhibit high inhibitory activity. It was also shown

that a correlation seems to exist between the activity of

the compounds and the electrostatic potential distribu-

tion of the molecules;²⁸to be active the ESP values of

the different regions of the inhibitor have to be in an

appropriate range.

In the present report, we will present the structure

optimization of this class of compounds leading to an

increase in activity and in selectivity of these inhibitors.

First, the influence of the bioisosteric exchange of one OH

group on the enzyme activity will be determined. Second,

the space availability around the inhibitors and the nature

of the most appropriate substituents will be investigated by

substitutions, either on the heterocycle or on the hydro-

xyphenyl moieties. The nature of the substituents will be

varied in order to investigate the possible interactions

between the inhibitor and the enzyme. Third, computa-

tional studies (docking studies and ESP calculations) will be

performed in order to identify the most plausible binding

mode for this class of compounds. Furthermore, the selec-

tivity toward 17β-HSD2 and the ERs R and β will be

determined as well as the potency of the compounds in

T-47D cells and inhibition of three hepatic CYP enzymes.

Finally, the pharmacokinetic profile of the two most

promising candidates will be evaluated in rats after oral

administration.

last series are thiophenes 1, 7, and 33, thiazole 36, and

phenylene 41, exhibiting IC₅₀values toward 17β-HSD1 in

the nanomolar range and high selectivity toward 17β-HSD2

and the ERs (Chart 1).

In the following, we will report on structural optimizations

that led to the discovery of new highly potent and selective

17β-HSD1 inhibitors (Chart 2).

Design

Up to now, several crystal structures of human 17β-HSD1

were resolved: as apoenzyme (i.e., PDB code 1BHS³³), as

binary complex (enzyme-E2, i.e., PDB code 1IOL³⁴), or as

ternary complex (enzyme-E2-NADP^þ, i.e., PDB codes

1FDT³⁵and 1A27³⁶and enzyme-HYC (hybride inhibitor),

PDB code 1I5R³⁷).

The analysis of the ternary complexes available from 17β-

HSD1 provides useful knowledge about the architecture of

the enzyme and important hints for structure based drug

design: a substrate binding site (SUB) and a cofactor binding

pocket (COF) are present as well as the most important amino

acids responsible for substrate and cofactor anchoring. The

SUB is a narrow hydrophobic tunnel containing two polar

regions at each end: His221/Glu282 on the one side and

Ser142/Tyr155 on the other side, corresponding tothe binding

oxygens in 3- and 17-hydroxy group of E2. Additionally a

flexible loop can be identified that is not well resolved in

almost all the structures.

Chemistry

From previous results obtained in the class of bis-

(hydroxyphenyl)azoles, thiophenes, benzenes, and azaben-

zenes,^27,28a SAR study highlighted four structural features

The synthesis of compounds 1-11, 21-25, and 32 is

presented in Scheme 1. Starting from the monobrominated

key intermediate 1b and the appropriate commercially

6726 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Scheme 1. Synthesis of Compounds 1-11, 21-25, and 32^a

Bey et al.

^aReagents and conditions: (a) method A, Cs₂CO₃, DME/EtOH/water (1:1:1), Pd(PPh₃)₄, microwave (150 W, 150 ꢀC, 15 bar), 15 min; (b) BBr₃,

CH₂Cl₂, -78 ꢀC to room temp, 18 h; (c) PhI(OAc)₂, I₂, AcOEt, 60 ꢀC, 5 h; (d) Ph-B(OH)₂; method B, Na₂CO₃, toluene/water (1:1), Pd(PPh₃)₄, reflux,

20 h; (e) (1) n-BuLi, dry THF, 5 min, -78 ꢀC, (2) B(OEt)₃, 2 h -78 ꢀC to room temp, (3) 1 N HCl, room temp; (f) method B, Na₂CO₃, toluene/water (1:1),

Pd(PPh₃)₄, reflux, 20 h.

available boronic acids, the preparation of compounds

1a-11a, 21a-25a, and 32a was accomplished via Suzuki

cross-coupling reaction³⁸under microwave assisted condi-

tions (method A: Cs₂CO₃, DME/EtOH/water (1:1:1), Pd-

(PPh₃)₄, microwave (150 W, 150 ꢀC, 15 bar), 15 min). The

resulting disubstituted thiophenes were subsequently sub-

mitted to ether cleavage with boron tribromide²⁸(method

C: BBr₃, CH₂Cl₂, -78 ꢀC to room temp, 18 h) leading to

compounds 1-11, 21-25, and 32 (Scheme 1). In the case of

intermediate 25a, the boronic acid 25b was prepared in a

three-step synthesis pathway. First, an iodine substituent

was selectively introduced in position 2 of the p-bromoani-

sole (compound 25d) using (diacetoxyiodo)benzene.³⁹Then

a selective Suzuki reaction on the iodo position of 25d under

method B (Na₂CO₃, toluene/water (1:1), Pd(PPh₃)₄, reflux,

20 h) led to the intermediate 25c, and the corresponding

boronic acid 25b was prepared using n-butyllithium and

triethyl borate followed by hydrolysis with diluted hydro-

chloric acid.

The preparation of compounds 31 and 33-42 is similar to

the synthetic pathway presented in Scheme 1 for compounds

1-11. The first Suzuki coupling was carried out according to

method B with the corresponding dibrominated heterocycle

and the methoxylated benzene boronic acid. The resulting

monosubstituted compounds 31b and 33b-42b were sub-

mitted to a second cross-coupling reaction under microwave

assisted conditionsfollowing method A. Thecompounds were

subsequently demethylated with boron tribromide to yield

compounds 31 and 33-42.

The synthesis of compounds 12, 14, and 15 is depicted

in Scheme 2. The key intermediate dimethoxylated bro-

mothiophene 12b was prepared following two successive

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6727

Scheme 2. Synthesis of Compounds 12, 14, and 15^a

^aReagents and conditions: (a) method B, Na₂CO₃, toluene/water (1:1), Pd(PPh₃)₄, reflux, 4 h; (h) boronic acid; method B, Na₂CO₃, toluene/water

(1:1), Pd(PPh₃)₄, reflux, 20 h; (d) BBr₃, CH₂Cl₂, -78 ꢀC to room temp, 18 h.

Scheme 3. Synthesis of Compounds 16-20^a

^aReagents and conditions: (a) method B, Na₂CO₃, toluene/water (1:1), Pd(PPh₃)₄, reflux, 4 h; (b) (1) n-BuLi, anhydrous THF, -78 ꢀC, 15 min,

(2) B(OEt)₃, THF, -78 ꢀC to room temp, 2 h, (3) 1 N HCl; (c) method A for 17a-20a (Cs₂CO₃, DME/EtOH/water (1:1:1), Pd(PPh₃)₄, microwave

(150 W, 150 ꢀC, 15 bar), 15 min); method B for 16a (Na₂CO₃, toluene/water (1:1), Pd(PPh₃)₄, reflux, 20 h); (d) BBr₃, CH₂Cl₂, -78 ꢀC to room temp, 18 h.

Suzuki coupling reactions according to method B (Na₂CO₃,

toluene/water (1:1), Pd(PPh₃)₄, reflux, 4 h) from 2,3,5-tribro-

mothiophene and methoxybenzene boronic acid. Thereaction

time of both cross-couplings was carefully controlled

(restricted to 4 h) in order to get a selective bromine replace-

ment each time. Intermediates 12a, 14a, and 15a were ob-

tained via a third Suzuki coupling using method B. The

methoxysubstituentswerecleaved in the last step, using boron

tribromide (method C: BBr₃, CH₂Cl₂, -78 ꢀC to room temp,

18 h).

Compound 13 was synthesized under microwave assisted

conditions in a one pot reaction using 2,5-dibromo-3-

methylthiophene and 3-hydroxyphenyl boronic acid follow-

ing method A (Cs₂CO₃, DME/EtOH/water (1:1:1), Pd-

(PPh₃)₄, microwave (150 W, 150 ꢀC, 15 bar)) for 15 min.

The synthesis of the molecules bearing an additional sub-

stituent on the m-hydroxyphenyl moiety of thiophene 1

(compounds 16-20) is shown in Scheme 3. Intermediate 16c

was prepared via Suzuki reaction from the p-methoxylated

benzene boronic acid and the 2,5-dibromothiophene follow-

ing method B, heating the reaction 4 h instead of 20 h in order

to avoid any dicoupling reaction. Treatment of 16c with

n-butyllithium and triethyl borate afforded after hydrolysis

with diluted hydrochloric acid the corresponding boronic acid

16b. The resulting compound was subjected to an additional

cross-coupling reaction which was carried out with the appro-

priate bromine derivative following method A for compounds

17a-20a and method B for compound 16a. The hydrolysis of

the methoxy groups with boron tribromide (method C) led to

compounds 16-20.

The synthesis of compounds 26-28 substituted in ortho-

position of the p-OH group is depicted in Scheme 4. The

preparation of the key intermediate 26b started from the

commercially available 3-formyl-4-methoxyphenylboronic

acid. Reduction of the aldehyde function using sodium boro-

hydride followed by a cross-coupling reaction with 1b under

microwave irradiation according to method A (Cs₂CO₃,

DME/EtOH/water (1:1:1), Pd(PPh₃)₄, microwave (150 W,

150 ꢀC, 15 bar), 15 min) afforded the disubstituted thiophene

26c. The alcohol function of 26c was subsequently oxidized

with pyridinium chlorochromate to yield to the key aldehyde

26b. It was subjected to the Horner-Wadworths-Emmons

6728 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Scheme 4. Synthesis of Compounds 26-28^a

Bey et al.

^aReagents and conditions: (a) NaBH₄, THF/EtOH (1:1), 0 ꢀC to room temp, 2 h; (b) method A, Cs₂CO₃, DME/EtOH/water (1:1:1), Pd(PPh₃)₄,

microwave (150 W, 150 ꢀC, 15 bar), 15 min; (c) PCC, CH₂Cl₂, room temp, 30 min; (d) NaH, THF dry, room temp, 4 h; (e) BBr₃, CH₂Cl₂, -78 ꢀC to room

temp, 18 h; (f) (1) LiOH, THF/H₂O (2:1), reflux, 20 h, (2) aniline, EDCI, HOBt, CH₂Cl₂, reflux, 20 h; (g) Pd(OH)₂, THF/EtOH (1:1), H₂, room temp,

20 h.

Scheme 5. Synthesis of Compounds 29-30^a

^aReagents and conditions: (a) method B, Na₂CO₃, toluene/water (1:1), Pd(PPh₃)₄, reflux, 20 h; (b) method A, Cs₂CO₃, DME/EtOH/water (1:1:1),

Pd(PPh₃)₄, microwave (150 W, 150 ꢀC, 15 bar), 15 min; (c) BBr₃, CH₂Cl₂, -78 ꢀC to room temp, 18 h.

conditions⁴⁰to introduce the acrylic ester moiety

(intermediate 26a). Hydrolysis of the ester function using

lithium hydroxide,²⁵amide bond formation with aniline,

EDCI, and HOBt⁴¹afforded compound 27a. The catalytic

double bond hydrogenation of 27a was performed using

Perlman’s catalyst.⁴²The ether functions of 26a, 27a, and

28a were deprotected using boron tribromide (method C) to

give the desired compounds 26-28.

The synthesis of the difluorinated thiophenes 29 and 30 is

presented in Scheme 5. These compounds were obtained after

two successive cross-coupling reactions. In the first step 2,5-

dibromothiophene reacted with 3-fluoro-4-methoxyphenyl-

boronic acid following method B (Na₂CO₃, toluene/water

(1:1), Pd(PPh₃)₄, reflux, 20 h). In the second step, the result-

ing monosubstituted thiophene 29b was subsequently sub-

mitted to a second cross-coupling reaction under microwave

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6729

Table 1. Effect of the Exchange of One OH Substituent for Other Functional Groups on Human 17β-HSD1 and 17β-HSD2 Inhibitory Activities^e

^aMean values of three determinations, standard deviation less than 10%. ^bHuman placenta, cytosolic fraction, substrate [³H]E1 þ E1 [500 nM],

cofactor NADH [500 μM]. ^cHuman placenta, microsomal fraction, substrate [³H]E2 þ E2 [500 nM], cofactor NAD^þ[1500 μM]. ^dIC₅₀(17β-HSD2)/

IC₅₀(17β-HSD1). ^eni: no inhibition. nt: not tested.

Table 2. Inhibition of Human 17β-HSD1 and 17β-HSD2 by Compounds Bearing a Supplementary Substituent on the Thiophene Core Structure^f

^aMean values of three determinations, standard deviation less than 10%. ^bHuman placenta, cytosolic fraction, substrate [³H]E1 þ E1 [500 nM],

cofactor NADH [500 μM]. ^cHuman placenta, microsomal fraction, substrate [³H]E2 þ E2 [500 nM], cofactor NAD^þ[1500 μM]. ^dIC₅₀(17β-HSD2)/

IC₅₀(17β-HSD1). ^ePredicted value, obtained with the LOGIT transformed. ^fni: no inhibition. nt: not tested.

irradiation (method A: Cs₂CO₃, DME/EtOH/water (1:1:1),

Pd(PPh₃)₄, microwave (150 W, 150 ꢀC, 15 bar), 15 min) to

yield the intermediates 29a and 30a. Ether cleavage with

boron tribromide led to the final compounds 29 and 30.

the amount of labeled E2 formed was quantified. The inhibi-

tion values of the test compounds are shown in Tables 1-5.

Thiophenes 1, 2, 7, and 33, thiazole 36, and phenylenes 38 and

41, identified in our previous article,²⁸were used as reference

compounds.

It was first investigated whether one of the two hydro-

phenyl moieties could be exchanged by another functional

group having similar properties. Recent results²⁸showed

that the m-hydroxy group is highly important for activity

and was therefore maintained in the core structure.

Biological Results

Activity: Inhibition of Human 17β-HSD1. Placental enzyme

was partially purified following a described procedure.^27,28

Tritiated E1 was incubated with 17β-HSD1, cofactor, and

inhibitor. After HPLC separation of substrate and product,

6730 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Bey et al.

The exchange of the p-hydroxy group on the meta-para

disubstituted thiophene (1, IC₅₀= 69 nM) by a bioisosteric

function (F, NH₂, SH) resulted in moderate (3, IC₅₀= 717

nM) or weak inhibitors (4 and 5, IC₅₀> 5000 nM) of 17β-

HSD1 (Table 1). Moving the F atom from the para-

(compound 3) to the meta-position (compound 8) led to a

small increase in activity (8, IC₅₀= 535 nM vs 3, IC₅₀= 717

nM). Replacement of the m-fluorine for a methylsulfonamide

moiety (9) did not improve the activity (9, IC₅₀= 523 nM vs 8,

IC₅₀= 535 nM), while a compound bearing a bulky substi-

tuent like tolylsulfonamide (11, IC₅₀= 350 nM) showed

comparable activity to the monohydroxylated thiophene (2,

IC₅₀= 342 nM), indicating that there is some space in this

region of the enzyme for substitution but it is unlikely that

specific interactions between the tolylsulfonamide moiety and

amino acids of the active site take place. The insertion of a

C1-linker between the phenyl moiety and the methylsulfona-

mide group was detrimental for the activity (9, IC₅₀= 523 nM

vs 10, IC₅₀> 1000 nM). It can therefore be concluded that

the two hydroxy functions are necessary for high activity and

the p-hydroxy group cannot be replaced by a bioisoteric

group.

Table 3. Effect of a Supplementary Substituent on the Hydroxyphenyl

Moieties on the Inhibition of the Human 17β-HSD1 and 17β-HSD2^e

In order to improve the activity and the selectivity of

our inhibitors, substituents capable of establishing further

interactions with the enzyme were added either on

the heterocycle or on the hydroxyphenyl moieties. Addi-

tional functional groups were introduced in both of the

meta-para 2,5-bis(hydroxyphenyl)thiophene 1 (IC₅₀

=

69 nM) and the meta-meta 2,5-disubstituted derivative 7

(IC₅₀= 173 nM).

Concerning substitution on the heterocycle, two kinds of

hydrophobic substituents (Me, Ph) were introduced in

position 3 on the thiophene ring to investigate the space

availability around the core (Table 2). The meta-meta

thiophenes bearing a methyl (compound 13) or phenyl

(compound 14) showed a reduction of activity compared to

the reference compound 7 (IC₅₀= 173 nM vs IC₅₀, 567, and

493 nM for 13, and 14, respectively). It is striking that in only

the case of the meta-meta disubstituted series the insertion

of a polar m-hydroxyphenyl substituent leads to an increase

in activity (15, IC₅₀= 119 nM vs 12, IC₅₀> 1000 nM). This

exemplifies that there is space available for further substi-

tution around the heterocycle only in the case of the

meta-meta bis(hydroxyphenyl) substitution pattern and

that the third m-OH group is certainly at an appropriate

distance to establish supplementary hydrogen bond interac-

tions with the active site.

Concerning the substitution of the hydroxyphenyl rings,

different groups were introduced either on the m-hydroxy-

phenyl ring (compounds 16-20) or on the p-hydroxyphenyl

moiety (compounds 21-28, Table 3). The synthesis of

compounds bearing ortho (to the heterocycle) substituents

at the hydroxyphenyl moieties was not considered, as the

conformational constraints induced by the ortho effect will

not allow the compound to adopt a planar geometry. Pla-

narity of our structures was considered necessary to mimic

the steroidal substrate.^27,28

^aMean values of three determinations, standard deviation less than

15%. ^bHuman placenta, cytosolic fraction, substrate [³H]E1 þ E1 [500

nM], cofactor NADH [500 μM]. ^cHuman placenta, microsomal frac-

tion, substrate [³H]E2 þ E2 [500 nM], cofactor NAD^þ[1500 μM].

^dIC₅₀(17β-HSD2)/IC₅₀(17β-HSD1). ^eni: no inhibition. nt: not tested.

fluorinated compound 19 (IC₅₀= 113 nM) and a very weak

methylated inhibitor 18 (IC₅₀> 5000 nM). Substituents

have also been introduced in position 5 on the p-hydroxy-

phenyl ring: a polar group like hydroxy (compound 22) or a

bulky substituent like the phenyl (compound 25) in o- of the

p-OH induced a decrease in activity compared to thiophene

compound 1 (IC₅₀= 69 nM vs IC₅₀= 402 and >5000 nM

for 22 and 25, respectively). The introduction of a fluorine

substituent into the same position led to the highly potent

compound 23 (IC₅₀= 8 nM), while substituents like methyl

or trifluoromethyl showed similar or slightly better activities

compared to the reference compound 1 (IC₅₀= 69 nM vs

IC₅₀= 46 and 38 nM for 21 and 24, respectively). Other

functional groups showing a higher flexibility like ethyl

acrylate (compound 26), phenylacrylamide (compound 27),

or phenylpropaneamide (compound 28) were also synthe-

sized, and the resulting compounds 26, 27, and 28 turned out

to have weaker inhibitory activity compared to the unsubs-

tituted thiophene 1 (IC₅₀= 69 nM vs 130, 427, and 620 nM

Introduction of a substituent in position 5 on the m-

hydroxyphenyl moiety resulted in the case of a methyl group

in a reduction of activity (16, IC₅₀= 629 nM vs 1, IC₅₀= 69

nM). The introduction of a fluorine atom led to a slight

increase in activity in comparison to the unsubstituted

compound 1 (17, IC₅₀= 42 nM vs 1, IC₅₀= 69 nM).

Moving these functional groups to position 4 gave an active

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6731

Table 4. Effect of Two Additional Fluorine Atoms on the 17β-HSD1 and 17β-HSD2 Inhibitory Activity

IC₅₀(nM)^a

compd

R₁

17β-HSD1^b

17β-HSD2^c

selectivity factor^d

23

29

30

31

32

H

8

29

940

227

118

8

5-F

4-F

17

56

218

312

13

6

780

2640

3

^aMean values of three determinations, standard deviation less than 10%. ^bHuman placenta, cytosolic fraction, substrate [³H]E1 þ E1 [500 nM],

cofactor NADH [500 μM]. ^cHuman placenta, microsomal fraction, substrate [³H]E2 þ E2 [500 nM], cofactor NAD^þ[1500 μM]. ^dIC₅₀(17β-HSD2)/

IC₅₀(17β-HSD1).

for 26, 27, and 28, respectively). The low activity of the

unconjugated compound 28 indicates that an overall dis-

tributed electronic density is an important parameter for

activity. These results also indicate that there is space avail-

able in this area for substituents but the nature of the

substituents is probably not yet optimal (Table 3).

Table 5. Influence of the Core and a Supplementary Substituent on the

Inhibition of the Human 17β-HSD1 and 17β-HSD2

The influence of the introduction of a second fluorine on

the highly active thiophene 23 (IC₅₀= 8 nM), either one F on

each hydroxyphenyl ring or two F atoms on the same

hydroxyphenyl moiety, was also examined (Table 4). When

the two F atoms are located on each hydroxyphenyl

moiety, the 4-substituted fluoro derivative (compound 30)

is slightly more potent than the one with the fluorine in the

5-position (29 IC₅₀= 29 nM vs 30 IC₅₀= 17 nM). A small

decrease in activity was observed when the two fluorine

substituents were present at the same hydroxyphenyl ring

(compound 31, IC₅₀= 56 nM). The exchange of the p-OH

function of 23 by a fluorine atom (compound 32) confirmed

the essential role of this p-hydroxy moiety as previously

observed.

Methyl and fluorine substituents have been identified as

functional groups able to increase the inhibitory activity of

the 2,5-bis(hydroxyphenyl)thiophene 1. Previously²⁸we re-

ported that other central core structures like 2,4-thiophene,

2,5-thiazole, and 1,4-benzene lead to highly active com-

pounds. The influence of an additional methyl or fluorine

substituent at these structures was therefore also investigated

(Table 5). Introduction of CH₃or F into the p-hydroxyphe-

nyl ring of 33, 36, and 38 resulting in compounds 34, 35, 37,

39, and 40 led to similarly active derivatives in the cases of 34

and 35 (IC₅₀= 64 nM vs 21 IC₅₀= 46 nM). A decrease in

inhibitory activity in the thiazole and in the benzene classes

of compounds was observed compared to the thiophene

family (37 IC₅₀= 143 nM vs 21 IC₅₀= 46 nM; 40 and 42

IC₅₀= 123 and 51 nM, respectively, vs 23 IC₅₀= 8 nM).

Among the investigated molecules, introduction of a methyl

or fluorine substituent led to an increase in activity only in

the class of the bis(hydroxyphenyl)thiophenes.

Selectivity: Inhibition of 17β-HSD2 and Affinities to the

Estrogen Receptors r and β. In order to gain insight into the

selectivity of the most active compounds, inhibition of 17β-

HSD2 and the relative binding affinities to the estrogen

receptors R and β were determined.

^aMean values of three determinations, standard deviation less than

13%. ^bHuman placenta, cytosolic fraction, substrate [³H]E1 þ E1 [500

nM], cofactor NADH [500 μM]. ^cHuman placenta, microsomal frac-

tion, substrate [³H]E2 þ E2 [500 nM], cofactor NAD^þ[1500 μM].

^dIC₅₀(17β-HSD2)/IC₅₀(17β-HSD1).

Since 17β-HSD2 catalyzes the inactivation of E2 into E1,

inhibitory activity toward this enzyme must be avoided. The

17β-HSD2 inhibition was determined using an assay similar

to the 17β-HSD1 test. Placental microsomes were incubated

with tritiated E2 in the presence of NAD^þand inhibitor.

6732 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Bey et al.

Table 6. Binding Affinities for the Human Estrogen Receptors R and β

of Selected Compounds

Table 7. Pharmacokinetic Parameters of Compounds 21 and 23 in Male

Rats after Oral Application (10 mg/kg)

RBA^a(%)

compd

ERR^b

ERβ^b

parameters^a

21

23

1

0.1 < RBA < 1

<0.01

1.5

0.1 < RBA < 1

<0.01

C

_{max obs}(ng/mL)

905.0

43.3

4.0

1388.2

24.9

8.0

17

21

23

30

34

37

C_z(ng/mL)

_{max obs}(h)

t_z(h)

t

0.01 < RBA < 0.1

0.1

1

24.0

3.8

24.0

2.7

0.01 < RBA < 0.1

<0.01

t_1/2z(h)

AUC_0-¥(ng/mL)

0.01 < RBA < 0.1

12275

19407

^aC_{max obs}, maximal measured concentration; C_z, last analytical quan-

tifiable concentration; t_{max obs}, time to reach the maximum measured

concentration; t_z, time of the last sample that has an analytical quantifi-

able concentration; t_1/2z, half-life of the terminal slope of a concentra-

tion-time curve; AUC_0-¥, area under the concentration-time curve

extrapolated to infinity.

^aRBA (relative binding affinity). E2: 100%, mean values of three

determinations, standard deviations less than 10%. ^bHuman recombi-

nant protein, incubation with 10 nM [³H]E2 and inhibitor for 1 h.

Separation and quantification of labeled product (E1) was

performed by HPLC using radiodetection. A selection of the

most potent 17β-HSD1 inhibitors was tested for inhibition of

17β-HSD2. IC₅₀values and selectivity factors (IC₅₀(HSD2)/

IC₅₀(HSD1)) are presented in Tables 1-5.

Monohydroxylated compounds (Table 1) exhibited a poor

selectivity regarding 17β-HSD2, the most selective one being

compound 3 with a selectivity factor of 5. This finding

suggests that the p-OH is important for activity as well as

for selectivity (selectivity of the para-meta derivative 1 is

28). Introduction of further substituents (Tables 2-5) into

the highly active bis(hydroxyphenyl) scaffold induced a loss

of selectivity against 17β-HSD2 except in case of compounds

21 and 23, which exhibit excellent selectivity factors of 49 and

118, respectively.

process but should not have an impact on the proof of

concept in vitro.

The pharmacokinetic profiles of the two most active and

selective compounds of this study (21 and 23) were deter-

mined in rats after oral administration in a cassette dosing

approach. Each group consisted of four male rats, and the

compounds were administered in doses of 10 mg/kg. Plasma

samples were collected over 24 h, and plasma concentrations

were determined by HPLC-MS/MS. The pharmacokinetic

parameters are presented in Table 7. The maximal concen-

tration (C_{max obs}) as well as the AUC value is higher for

compound 23 (C_max= 1388.2 ng/mL, AUC = 19407 ng/

mL) than for compound 21 (C_max= 905.0 ng/mL, AUC =

12275 ng/mL). The maximal plasma concentration (t_{max obs}

)

for compounds 21 and 23 was reached after 4.0 and 8.0 h,

respectively. These data show that both compounds exhibit

excellent pharmacokinetic properties in the rat and might

therefore be good candidates for further experiments in

disease-oriented rat models.

A further prerequisite for 17β-HSD1 inhibitors to be

used as potential drugs is that they do not show affinity for

ERR and ERβ, since binding to these receptors could

counteract the therapeutic concept of selective 17β-HSD1

inhibition. The binding affinities of the most selective

compounds of this study were determined using recombi-

nant human protein in a competition assay applying [³H]E2

and hydroxyapatite (Table 6). All tested compounds show

very marginal to marginal affinity to the ERs except com-

pound 23, which binds weakly to ERβ (RBA = 1%).

Compound 21 was evaluated for estrogenic effects on the

ER-positive, mammary tumor T47D cell line. No agonistic,

i.e., stimulatory effect was observed after application of

compound 21 even at a concentration 1000-fold higher

compared to E2.

Further Biological Evaluations. Additionally, the intracel-

lular potency of compounds 21, 23, and 34 on E2 formation

was evaluated using a cell line that expresses both 17β-HSD1

and 17β-HSD2 (T47D cells). The compounds inhibited the

formation of E2 after incubation with labeled E1 showing

IC₅₀values of 426, 282, and 362 nM for 21, 23, and 34,

respectively. These results indicate that the compounds are

able to permeate the cell membrane and inhibit the transfor-

mation of E1 into E2.

Computational Chemistry

Molecular Modeling. From the biological results it became

apparent that introduction of a fluorine atom in the position

ortho to the p-OH phenylthiophene (compound 23) led to a

significant increase in the 17β-HSD1 inhibitory activity. To

get an insight into the binding mode of this compound and to

better understand the favorable interactions achieved by this

inhibitor in the active site, computational studies were

performed by means of the docking software GOLD, version

3.2, and Autodock 4.1.

The choice of the 3D-structure of the enzyme, i.e., crystal

structure, and the oxidation state of the cofactor, i.e.,

NADPH or NADP^þ, used for the docking studies are crucial

for obtaining reliable results.

We focused on X-ray structures of 17β-HSD1 having a

high resolution and showing a ternary complex (to get closer

to the in vivo conditions). Three structures fulfilled these

criteria: 1FDT and 1A27 both describing the ternary com-

plex; enzyme-E2-NADP^þand 1I5R, describing the binary

complex; enzyme-steroidal hybride inhibitor (HYC), which

is an adenosine moiety linked to an E2 core via a C9-linker.

These three crystal structures differ mainly in the location of

the amino acids belonging to the flexible loop RG⁰βF

(Pro187-Pro200). Since this loop borders both the SUB

and the COF, its conformational variations strongly influ-

ence the size of both binding cavities. It is therefore impor-

tant to take care of the position of this loop in the structures

used for the docking studies.

The same compounds (21, 23, and 34) were further in-

vestigated for inhibition of three human hepatic enzymes:

CYP3A4, CYP2D6, and CYP2C19. The three compounds

showed very little to weak inhibition of the hepatic CYP

enzymes (10 μM > IC₅₀> 1.50 μM) except for CYP3A4

(IC₅₀= 0.50 and 0.82 μM for 21 and 23, respectively) and

CYP2C19 (IC₅₀= 0.94 μM for 21) for which an enhanced

inhibition was observed.

The relatively high inhibition of CYP3A4 by 21 and 23 has

to be taken into consideration for the further optimization

Article

In the X-ray structure 1FDT, the residues 187-200 are not

well resolved but two plausible conformations for the loop

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6733

(noted 1FDT-A and 1FDT-B) have been described.³⁵The

˚

backbones of these two loops are similar (rmsd of ∼1 A),

while the main difference is given by the orientation of the

side chains, mainly concerning the four amino acids Phe192,

Met193, Glu194, and Lys195. In 1FDT-A, Phe192 and

Met193 are turned toward the outer part of the enzyme while

Glu194 and Lys195 are oriented toward the substrate and the

cofactor (extending the substrate binding site = open con-

formation). On the other hand, in 1FDT-B these two pairs of

residues show a reversed orientation limiting length and

volume of the steroid binding site compared to 1FDT-A

(=closed conformation). Although others²²have only con-

sidered 1FDT-B, we decided to investigate both conforma-

tions of this loop.

Interestingly, the flexible loop in 1A27 shows a compar-

able geometry as observed in 1FDT-B, with Phe192 and

Met193 oriented toward the nicotinamide moiety, also re-

stricting the space in the substrate binding site. In the case of

1I5R, the loop is shifted in the direction of the cofactor,

resulting in a different conformation compared to both

1FDT-A and 1FDT-B, although, like for 1FDT-A, it ex-

tends the SUB.

Figure 1. Docking complex between 17β-HSD1 (X-ray 1FDT-B)

and compound 23 (blue, SUB binding mode). NADPH, interacting

residues, and ribbon rendered tertiary structure of the active site are

shown. Residues of the flexible loop are rendered in sticks and

colored in yellow. Hydrogen bonds and π-π stackings (and hydro-

phobic interactions) are drawn in yellow and blue dashed lines,

respectively. For comparison, E2 is depicted in magenta lines.

Figures were generated with Pymol (http://www.pymol.org).

The docking studies were performed with NADPH as

cofactor, as it is described that NADPH is present in a

500-fold higher concentration than its oxidized form

NADP^þin living cells.^43,44

Compound 23 was docked with NADPH into four differ-

ent X-ray structures: 1FDT-A, 1FDT-B, 1A27, and 1I5R.

Two different binding modes were observed for compound

23. In the cases of 1FDT-B and 1A27, the inhibitor is located

exclusively in the steroid binding site (Figure 1), adopting a

similar orientation as previously described for the bis-

(hydroxyphenyl)oxazole B²⁷(Chart 1), while for 1FDT-A

and 1I5R, the inhibitor is located between the steroid and the

cofactor binding sites, interacting with the nicotinamide

moiety. In the following, this binding mode will be named

as alternative binding mode (Figure 2).

In the case of the steroidal binding mode (1FDT-B and

1A27, Figure 1) the following specific interactions can be

observed: hydrogen bond interactions between the m-hydro-

xy group of 23 and Ser142/Tyr155 (d_O-O= 2.6 A for both

amino acids) and between the p-OH group and His221/

˚

Figure 2. Docking complex between 17β-HSD1 (X-ray 1FDT-A)

and compound 23 (green, alternative binding mode). NADPH,

interacting residues, and ribbon rendered tertiary structure of the

active site are shown. Residues of the flexible loop are rendered in

sticks and colored in blue. Hydrogen bonds and π-π stackings (and

hydrophobic interactions) are drawn in yellow and blue dashed

lines, respectively. For comparison, E2 is depicted in magenta lines.

˚

Glu282 (d_O-N= 2.8 A and d_O-O= 3.8 A, Figure 1).

Additionally, hydrophobic interactions and π-π stackings

(Phe226, Phe259) are also involved.

In the alternative binding mode obtained using 1FDT-A

and 1I5R (Figure 2), compound 23 is also stabilized by

hydrogen bond interactions: the m-OH group forms a strong

˚

(T-shape conformation; closest C-C contact distance, 3.7 A).

Moreover, electrostatic interactions between the sulfur atom

of the heterocycle with the surrounding amino acids like

Tyr155 and Ser142 might also play a role as described.⁴⁵

The results presented so far suggest that both binding

modes have to be considered as possible for this class of

inhibitors. They depend mainly on the orientation of the

flexible loop. There is only one conformation of the loop

leading to a steroidal binding mode (1FDT-B/1A27). In the

case of 1FDT-A/1I5R the pose showing the alternative

binding mode is obtained using two X-ray structures having

two different conformations of the loop. Unfortunately,

because of the almost identical scoring function values

observed for both poses with the docking programs (Gold

and Autodock), it was not possible to determine which

H-bond with the phosphate group of the cofactor (d_O-O

˚

=

2.9 A). The fluorine atom could establish halogen bonds with

the backbone -NH- of Val143 and Gly144 (d_F-N= 3.2 and

3.8 A, respectively), in addition to a halogen bond with the

˚

OH-group of Ser142 (d_F-O= 3.5 A) which is involved in the

catalytic process. Further, the p-OH points perpendicular

˚

toward Phe259 (d_O-centroid= 4.5 A), indicating a possible

OH-π interaction. This could explain the importance of this

group observed in the SARs. Moreover, strong π-π stacking

interactions seem to stabilize the inhibitor in this binding

mode: between the m-OH-phenylthiophene moiety and the

nicotinamide part of the cofactor (parallel-displaced config-

˚

uration; distance between the two ring centers, 4.3 A) and

between the p-OH-phenylthiophene moiety and Phe226

6734 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Bey et al.

Figure 3. Structures and MEP maps of both ventral (steroidal R-side) and dorsal (steroidal β-side) views of truncated NADPH (A), thiophene

23 (B), E2 (C), and E1 (D). MEP surfaces were plotted with GaussView, version 3.0.

model (1FDT-A/1I5R or 1FDT-B/1A27) is the most appro-

priate to describe the interactions between the inhibitor and

the enzyme and therefore which is the most plausible binding

mode.

Comparing both poses obtained by docking of 23 in

1FDT-A and 1FDT-B shows that there is a common area

in the neighborhood of the catalytic tetrade that corresponds

to the D-ring of the enzyme-substrate complex (figure in

Supporting Information).

Molecular Electrostatic Potential (MEP). Recently we

reported on the influence of the electronic density (MEP

maps, “semi-QMAR”) on the potency of the inhibitors in

this class of compounds.²⁸The 3D-structures of the inhibi-

tors were virtually divided into three areas, and a given

optimal range of ESP values (in hartree) was determined

for each region (-1.7 ꢀ 10^-2to -1.2 ꢀ 10^-2for I, -1.6 ꢀ

10^-2to -0.9 ꢀ 10^-2for II, and -1.2 ꢀ 10^-2to -0.5 ꢀ 10^-2

for III). The MEPs of compound 23 were calculated as

shown in Figure 3. The molecular ESP distribution observed

(-1.8 ꢀ 10^-2to -1.2 ꢀ 10^-2for I, -1.6 ꢀ 10^-2to -0.8 ꢀ

10^-2for II, and -1.1 ꢀ 10^-2to -0.4 ꢀ 10^-2for III) fitted

well to the optimal ranges identified previously, confirming

the correlation between the ESP range and the potency of the

compounds. The MEP maps of the natural substrate E1 and

of E2 were also calculated (Figure 3) and compared to the

one of 23. The finding that the ESP distribution of 23 and E2/

E1 is very different might be an indication that compound 23

does not bind in the same way as the steroid.

by ab initio methods. As can be seen in Figure 3, a certain

complementarity was observed. The NADPH MEP maps

give the explanation for the observation²⁸that a strong

polarization between the vertex and the base of the central

ring of the inhibitors is negative for binding. Positive

ESP values on the vertex side lead to repulsion effects

with the nicotinamide and therefore reduce the inhibitory

activity. This finding indicates that this class of com-

pound might bind according to the alternative binding mode

(Figure 2). However, this hypothesis needs to be further

investigated.

Discussion and Conclusion

Structural optimizations of compound 1 led to the discov-

ery of new substituted 2,5-bis(hydroxyphenyl)thiophene deri-

vatives, the fluorinated 23 and the methylated 21 being the

most active and selective inhibitors identified.

From a previous work in this class of compounds,²⁸it

was demonstrated that removal of one of the two hydro-

xyphenyl moieties is detrimental for the activity. In this

paper it was shown that replacement of the p-OH function

by a bioisoteric group like F, NH₂, SH, CN leads to a

reduction of activity. The lack of hydrogen donating

properties of the fluoro and cyano substituents might not

be the only reason for this decrease in activity, as the amino

and the thiol derivatives are also less active than the parent

compound 1. Interestingly, the omission of a C1-linker

between the methylsulfonamide moiety and the phenyl ring

(compound 9) resulted in an increase of potency. Depriva-

tion of electrons from the phenyl ring obviously is neces-

sary for good inhibition. The relatively high activity

observed for compound 11 (IC₅₀= 350 nM) especially

compared to compound 9 (IC₅₀= 523 nM) demonstrates

that in the protein there is some space available in this

position for a bulky substituent. Furthermore, the tolyl

According to the alternative binding mode, the m-hydro-

xyphenylthiophene part of 23 overlaps with the nicotina-

mide part of the cofactor and forms stabilizing π-π inter-

actions. The ESP distribution of these two entities should

therefore show complementarity. To get an insight into

this, the MEP map of a truncated NADPH, the counter-

part of the m-OH-phenylthiophene moiety, was calculated

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6735

group might also be involved in the stabilization of the inhi-

bitor in the binding site, establishing π-π stacking interactions

with appropriate amino acid residues present in this region.

With the aim to increase the activity and the selectivity in

this class of compounds, substituents were introduced on the

2,5-bis(hydroxyphenyl)thiophenes 1 and 7. This was success-

ful for compound 15 (IC₅₀= 119 nM vs 7, IC₅₀= 173 nM).

Apparently, the formation of an additional hydrogen bond is

responsible for this increase in inhibitory activity, while a pure

π-π stacking interaction as supposed for compound 14 is not

sufficient. The 2,5-disubstituted thiophenes 12 and 15 differ

only in the position of one hydroxy group (para, compound

12; meta, compound 15). The fact that compound 15 shows a

much higher activity (IC₅₀= 119 nM vs 12, IC₅₀> 1000 nM)

indicates that only in case of 15 is the geometry of the OH

groups is acceptable for a reasonable interaction. It demon-

strates, as observed already,^24,27a sharp SAR and a reduced

flexibility in this region of the active site.

The trisubstituted compound 14 bearing a phenyl substi-

tuent at the thiophene differs from the triazole A³²(Chart 1)

only in the nature of the heterocycle. The following compar-

isons highlight the importance of the heterocycle for the

potency of the molecules: inactive compound A vs thiophene

14 (IC₅₀= 493 nM), thiophene 21 (IC₅₀= 46 nM) vs thiazole

37 (IC₅₀= 143 nM), and thiophene 23 (IC₅₀= 8 nM) vs

benzene 40 (IC₅₀= 123 nM). It becomes apparent that the

thiophene ring is the most appropriate heterocycle for high

inhibitory activity. Provided that all compounds bind accord-

ing to the same binding mode, there are different explanations

for these results: (1) The presence and the position of one or

several nitrogens in the heterocycle modulate the MEP dis-

tribution of the compound and in particular the electronic

distribution of the m-hydroxyphenyl moiety²⁸(in the case of

the compounds under consideration, resulting in inadequate

electronic properties for binding). (2) Similarly the replace-

ment of the sulfur leads to an inadequate repartition of the

electron density in the heterocycle. (3) A reduced flexibility in

the binding site is responsible for the enzyme not being able to

adjust its geometry to the different hydroxyphenyl moieties

(depending on the heterocycle, the angles between the phenyl-

OHs are different).

Introduction of a second fluorine atom into this fluorohy-

droxyphenyl ring (compound 31) does not enhance the acti-

vity, suggesting that the effects of the fluorine are not additive.

A second fluorine was also added to the other hydroxyphenyl

moiety (in ortho and meta of the m-OH group), leading to

compounds 29 and 30. However, no enhancement of the

activity compared to the monofluorinated 23 was observed,

indicating that there are no specific interactions of the second

fluoro substituent.

A close look at the X-ray structures of 17β-HSD1 crystal-

lized in the presence of different steroidal ligands showed that

the flexible loop (amino acids 187-200) can adopt different

geometries depending on the nature of the ligand and on the

absence or presence of the cofactor in the catalytic region. It

indicates that some parts of the enzyme can adapt their

geometry to the molecule present in the active site in order

to stabilize it. However, other parts are rigid, explaining the

sharp SAR observed in this paper and previously.^27,28

Two plausible conformations of the loop in the ternary

complex enzyme-E2-NADP^þ, PDB code 1FDT, have been

described (1FDT-A, 1FDT-B).We have shown that both can

be used for docking studies. In the case of 1FDT-A the

substrate binding pocket is extended, enhancing the volume

of the active site. It is therefore a good model to evaluate an

alternativebinding mode for inhibitors thatare largerthan the

steroid. A binding mode as observed for steroids in the X-ray

structures was found when the loop closes the SUB (1FDT-B).

Surprisingly, when the inhibitors were docked to the protein

with the loop in the open conformation, they interact with the

nicotinamide part of the cofactor. MEP calculations showed a

certain complementarity between the electronic density of 23

and of the nicotinamide moiety of the cofactor, indicating that

this alternative binding mode is not only plausible but also

might be the one that is more likely.

Up to now, in the design of compounds as potential

17β-HSD1 inhibitors, one group^37,49-51tried to mimic the

cofactor. Our finding of the above-mentioned alternative

binding mode makes another strategy very promising: the

cofactor, which is likely to be present in the active site when

the inhibitor is entering, could be used as partner to achieve

additional interactions rather than trying to displace it.

The most potent 17β-HSD1 inhibitors 21 and 23 exhibit a

higher selectivity toward 17β-HSD2 compared to parent

compound (selectivity factors 49 and 118, respectively, vs 28

for1). Thisindicates thatthe aminoacidsclose to theCH₃or F

substituents must have different properties in the two 17β-

HSD enzymes, which could be further exploited to increase

selectivity.

A high increase in activity and selectivity could be reached

by introduction of substituents into the hydroxyphenyl moi-

ety, especially when the substituent is located ortho of the

p-hydroxyphenyl group (compounds 21-24). Not all substitu-

ents are equally well tolerated: there is no space available for

a phenyl group (compound 25). An additional OH group

(compound 22) is obviously not able to establish specific

interactions, while small lipophilic substituents (methyl, 21;

fluorine, 23; trifluoromethyl, 24) are enhancing the activity.

There is enough space in this region of the enzyme to introduce

a flexible chain (26), but conjugation seems to be necessary to

achieve a higher activity as already observed with the tolyl-

sulfonamide substituent (compound 11).

The most potent inhibitors show only marginal to very little

affinity to the ERR, and no stimulation of cell proliferation

(agonistic effect) in the ER-positive T47D cell line could be

observed. The weak affinity of compound23 for ERβ may not

be critical, as it is reported that ERβ exhibits antiproliferative

effects in breast cancer cells.⁵²

The positive influence of the fluorine atom has often been

demonstrated in medicinal chemistry^46-48and was also pro-

ven in this study with compound 23 (IC₅₀= 8 nM). The

position of the fluorine is decisive for an increase in activity: it

has tobe inthe meta-position (17and 23). Highest activity was

achieved at the position ortho of the hydroxy group

(compound 23). This indicates that either direct interactions

of the fluorine with amino acid residues in this region of the

active site or the increase of acidity of the neighboring OH

groups might be responsible for this effect.

Compound 23 might be used in an appropriate animal

model to prove the concept of 17β-HSD1 inhibition with

nonsteroidal inhibitors. This compound shows a good phar-

macokinetic profile in rats.

In this paper, we described the synthesis of substituted

bis(hydroxyphenyl)thiophenes, thiazoles, and benzenes as

inhibitors of 17β-HSD1 and the evaluation of their biological

properties. The most promising compounds of this study, 21

and 23, exhibit high selectivity toward 17β-HSD2, marginal

binding to ERR, and excellent pharmacokinetic profiles in

6736 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Bey et al.

rats after peroral application. These new compounds provide

useful tools to validate 17β-HSD1 as a target for the treatment

of estrogen-dependent diseases.

nitrogen atmosphere. The reaction mixture was cooled to room

temperature. The aqueous layer was extracted with ethyl acet-

ate. The combined organic layers were washed with brine, dried

over magnesium sulfate, filtered, and concentrated to dryness.

The product was purified by CC.

Experimental Section

General Procedure for Ether Cleavage. Method C. To a

solution of methoxyphenyl derivative (1 equiv) in dry dichloro-

methane at -78 ꢀC (dry ice/acetone bath), boron tribromide in

dichloromethane (1 M, 3 equiv per methoxy function) was

added dropwise. The reaction mixture was stirred for 20 h at

room temperature under nitrogen atmosphere. Water was

added to quench the reaction, and the aqueous layer was

extracted with ethyl acetate. The combined organic layers were

washed with brine, dried over sodium sulfate, filtered, and

concentrated to dryness. The product was purified by prepara-

tive thin layer chromatography (TLC).

2-(4-Fluorophenyl)-5-(3-methoxyphenyl)thiophene (3a). The

title compound was prepared by reaction of 2-bromo-5-(3-

methoxyphenyl)thiophene (1b) (150 mg, 0.56 mmol), 4-fluoro-

phenylboronic acid (94 mg, 0.67 mmol), cesium carbonate (383

mg, 1.24 mmol), and tetrakis(triphenylphosphine)palladium

(6.4 mg, 5.6 μmol) according to method A. The product was

purified by CC (hexane/ethyl acetate 9:1). Yield: 75% (120 mg).

3-[5-(4-Fluorophenyl)-2-thienyl]phenol (3). The title com-

pound was prepared by reaction of 2-(4-fluorophenyl)-5-(3-

methoxyphenyl)thiophene (3a) (80 mg, 0.28 mmol) and boron

tribromide (0.84 mmol) according to method C. The product

was purified by preparative TLC (hexane/ethyl acetate 1:1).

Yield: 68% (52 mg). MS (ESI): 270 (M þ H)^þ. Anal.

(C₁₆H₁₁FOS) C, H, N.

Chemical Methods. Chemical names follow IUPAC nomen-

clature. Starting materials were purchased from Aldrich, Acros,

Lancaster, Roth, Merck, or Fluka and were used without

purification.

Column chromatography (CC) was performed on silica gel

(70-200 μm) coated with silica. Preparative thin layer chroma-

tography (TLC) was performed on 1 mm SIL G-100 UV₂₅₄glass

plates (Macherey-Nagel), and reaction progress was monitored

by TLC on Alugram SIL G UV₂₅₄(Macherey-Nagel).

IR spectra were recorded on a Bruker Vector 33 spectrometer

(neat sample).

¹H NMR and ¹³C NMR spectra were measured on a Bruker

AM500 spectrometer (500 MHz) at 300 K. Chemical shifts are

reported in δ (parts per million, ppm) by reference to the

hydrogenated residues of deuteriated solvent as internal stan-

dard (CDCl₃, δ = 7.24 ppm (¹H NMR) and δ = 77 ppm (¹³C

NMR); CD₃OD, δ = 3.35 ppm (¹H NMR) and δ = 49.3 ppm

(¹³C NMR); CD₃COCD₃, δ = 2.05 ppm (¹H NMR) and δ =

29.9 ppm (¹³C NMR); CD₃SOCD₃, δ = 2.50 ppm (¹H NMR)

and δ = 39.5 ppm (¹³C NMR)). Signals are described as s, d, t, q,

dd, m, dt for singlet, doublet, triplet, quadruplet, doublet of

doublets, multiplet, and doublet of triplets, respectively. All

coupling constants (J) are given in hertz (Hz).

Mass spectra (ESI) were recorded on a TSQ Quantum

(Thermofischer) instrument. Elemental analyses (C, H, N) were

performed at the Department of Instrumental Analysis and

Bioanalysis, Saarland University, Germany. The purity of the

compounds, which in all cases was higher than 95%, was

determined using two different methods: elemental analysis

for compounds 3-6, 8-32, 34, and 37 and LC/MS for com-

pounds 35, 39, 41, and 42.

4-[5-(3-Methoxyphenyl)-2-thienyl]aniline (4a). The title com-

pound was prepared by reaction of 2-bromo-5-(3-methoxyphe-

nyl)thiophene (1b) (150 mg, 0.56 mmol), 4-aminophenylboronic

acid (92 mg, 0.67 mmol), cesium carbonate (383 mg, 1.24 mmol),

and tetrakis(triphenylphosphine)palladium (6.4 mg, 5.6 μmol)

according to method A. The product was purified by CC

(hexane/ethyl acetate 7:3). Yield: 63% (100 mg).

Compounds 2-bromo-5-(3-methoxyphenyl)thiophene (1b),²⁸

2-(3-methoxyphenyl)-5-(4-methoxyphenyl)thiophene (1a),²⁸3-[5-

(4-hydroxyphenyl)-2-thienyl]phenol (1),²⁸2-(3-methoxyphenyl)-5-

phenylthiophene (2a),²⁸3-(5-phenyl-2-thienyl)phenol (2),²⁸2,5-bis-

(3-methoxyphenyl)thiophene (7a),²⁸3,3⁰-thiene-2,5-diyldiphenol

(7),²⁸2-bromo-5-(4-methoxyphenyl)thiophene (16c),²⁸4-bromo-

2-iodo-1-methoxybenzene (25d),⁵³5-bromo-2-methoxybiphenyl (25c),⁵⁴

[6-methoxy-1,1⁰-biphenyl-3-yl]boronic acid (25b),⁵⁴[3-(hydroxy-

methyl)-4-methoxyphenyl]boronic acid (26d),⁵⁵4-bromo-2-(3-

methoxyphenyl)thiophene (33b),²⁸2-(3-methoxyphenyl)-4-(4-

methoxyphenyl)thiophene (33a),²⁸3-[4-(4-hydroxyphenyl)-

2-thienyl]phenol (29),²⁸5-bromo-2-(3-methoxyphenyl)-1,3-thia-

zole (36b),²⁸2-(3-methoxyphenyl)-5-(4-methoxyphenyl)-1,3-thiazole

(36a),²⁸3-[5-(4-hydroxyphenyl)-1,3-thiazol-2-yl]phenol (36),²⁸4⁰-

bromo-3-methoxybiphenyl (38b),²⁸3,4⁰⁰-dimethoxy-1,1⁰:4⁰,1⁰⁰-ter-

phenyl (38a),²⁸1,1⁰:4⁰,1⁰⁰-terphenyl-3,4⁰⁰-diol (38),²⁸3,3⁰⁰-dimethoxy-

1,1⁰:4⁰,1⁰⁰-terphenyl (40a),²⁸and 1,1⁰:4⁰,1⁰⁰-terphenyl-3,3⁰⁰-diol (40)²⁸

were prepared following described procedures.

General Procedure for Suzuki Coupling. Method A. A mix-

ture of aryl bromide (1 equiv), arylboronic acid (1.2 equiv),

cesium carbonate (2.2 equiv), and tetrakis(triphenylphos-

phine)palladium (0.01 equiv) was suspended in an oxygen-free

DME/EtOH/water (1:1:1) solution. The reaction mixture was

exposed to microwave irradiation (15 min, 150 W, 150 ꢀC,

15 bar). After cooling to room temperature, water was added

and the aqueous layer was extracted with ethyl acetate. The

combined organic layers were washed with brine, dried over

magnesium sulfate, filtered, and concentrated to dryness. The

product was purified by column chromatography (CC).

Method B. A mixture of aryl bromide (1 equiv), arylboronic

acid (1 equiv), sodium carbonate (2 equiv), and tetrakis-

(triphenylphosphine)palladium (0.05 equiv) in an oxygen-free

toluene/water (1:1) solution was stirred at 100 ꢀC for 20 h under

3-[5-(4-Aminophenyl)-2-thienyl]phenol (4). The title com-

pound was prepared by reaction of 4-[5-(3-methoxyphenyl)-2-

thienyl]aniline (4a) (100 mg, 0.37 mmol) and boron tribromide

(1.11 mmol) according to method C. The product was purified

by preparative TLC (hexane/ethyl acetate 1:1). Yield: 82% (82

mg). MS (ESI): 268 (M þ H)^þ. Anal. (C₁₆H₁₃NOS) C, H, N.

4-[5-(3-Methoxyphenyl)-2-thienyl]benzenethiol (5a). The title

compound was prepared by reaction of 2-bromo-5-(3-methoxy-

phenyl)thiophene (1b) (250 mg, 0.93 mmol), 4-mercaptophenyl-

boronic acid (172 mg, 1.12 mmol), cesium carbonate (636 mg,

2.05 mmol), and tetrakis(triphenylphosphine)palladium (10.8

mg, 9.3 μmol) according to method A. The product was purified

by CC (hexane/ethyl acetate 7:3). Yield: 61% (160 mg).

3-[5-(4-Sulfanylphenyl)-2-thienyl]phenol (5). The title com-

pound was prepared by reaction of 4-[5-(3-methoxyphenyl)-2-

thienyl]benzenethiol (5a) (150 mg, 0.50 mmol) and boron tribro-

mide (1.50 mmol) according to method C. The product was puri-

fied by preparative TLC (hexane/ethyl acetate 1:1). Yield: 81%

(115 mg). MS (ESI): 285 (M þ H)^þ. Anal. (C₁₆H₁₂OS₂) C, H, N.

4-[5-(3-Methoxyphenyl)-2-thienyl]benzonitrile (6a). The title

compound was prepared by reaction of 2-bromo-5-(3-methoxy-

phenyl)thiophene (1b) (200 mg, 0.74 mmol), 4-cyanophenyl-

boronic acid (131 mg, 0.89 mmol), cesium carbonate (508 mg,

1.64 mmol), and tetrakis(triphenylphosphine)palladium (8.5

mg, 7.4 μmol) according to method A. The product was purified

by CC (hexane/ethyl acetate 7:3). Yield: 27% (60 mg).

4-[5-(3-Hydroxyphenyl)-2-thienyl]benzonitrile (6). The title

compound was prepared by reaction of 4-[5-(3-metho-

xyphenyl)-2-thienyl]benzonitrile (6a) (42 mg, 0.14 mmol) and

boron tribromide (0.42 mmol) according to method C. The

product was purified by preparative TLC (hexane/ethyl acetate

6:4). Yield: 62% (25 mg). MS (APCI): 277 (M)^þ. Anal.

(C₁₇H₁₁NOS) C, H, N.

Article

2-(3-Fluorophenyl)-5-(3-methoxyphenyl)thiophene (8a). The

title compound was prepared by reaction of 2-bromo-5-(3-

methoxyphenyl)thiophene (1b) (150 mg, 0.56 mmol), 3-fluoro-

phenylboronic acid (94 mg, 0.67 mmol), cesium carbonate

(381 mg, 1.22 mmol), and tetrakis(triphenylphosphine)-

palladium (6.5 mg, 5.6 μmol) according to method A. The

product was purified by CC (hexane/ethyl acetate 9:1). Yield:

82% (130 mg).

3-[5-(3-Fluorophenyl)-2-thienyl]phenol (8). The title com-

pound was prepared by reaction of 2-(3-fluorophenyl)-5-

(3-methoxyphenyl)thiophene (8a) (130 mg, 0.45 mmol) and

boron tribromide (1.35 mmol) according to method C. The

product was purified by preparative TLC (dichloromethane/

methanol 99:1). Yield: 66% (82 mg). MS (ESI): 271 (M þ H)^þ.

Anal. (C₁₆H₁₁FOS) C, H, N.

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6737

3,5-dibromo-2-(3-methoxyphenyl)thiophene (12c) (500 mg,

1.43 mmol), 4-methoxybenzeneboronic acid (268 mg, 1.72

mmol), sodium carbonate (333 mg, 3.15 mmol), and tetrakis-

(triphenylphosphine)palladium (82.6 mg, 71.5 μmol) according

to method B. The product was purified by CC (petroleum ether/

ethyl acetate 9:1). Yield: 52% (278 mg).

2,3-Bis(3-methoxyphenyl)-5-(4-methoxyphenyl)thiophene (12a).

The title compound was prepared by reaction of 3-bromo-2-(3-

methoxyphenyl)-5-(4-methoxyphenyl)thiophene (12b) (250 mg,

0.67 mmol), 3-methoxybenzeneboronic acid (124 mg, 0.80

mmol), sodium carbonate (142 mg, 1.34 mmol), and tetrakis-

(triphenylphosphine)palladium (38.7 mg, 33.5 μmol) according

to method B. The product was purified by CC (petroleum ether/

ethyl acetate 9:1). Yield: 72% (194 mg).

3,3⁰-[5-(4-Hydroxyphenyl)thiene-2,3-diyl]diphenol (12). The

title compound was prepared by reaction of 2,3-bis(3-

methoxyphenyl)-5-(4-methoxyphenyl)thiophene (12a) (100 mg,

0.24 mmol) and boron tribromide (2.16 mmol) according to

method C. The product was purified by preparative TLC

(hexane/ethyl acetate 1:1). Yield: 92% (79 mg). MS (ESI): 359

(M - H)^þ. Anal. (C₂₂H₁₆O₃S) C, H, N.

N-(3-[5-(3-Methoxyphenyl)-2-thienyl]phenyl)methanesulfona-

mide (9a). The title compound was prepared by reaction

of 2-bromo-5-(3-methoxyphenyl)thiophene (1b) (160 mg,

0.60 mmol), 3-methylsulfonylaminophenylboronic acid (155

mg, 0.72 mmol), cesium carbonate (410 mg, 1.32 mmol), and

tetrakis(triphenylphosphine)palladium (6.9 mg, 6.0 μmol) ac-

cording to method A. The product was purified by CC

(petroleum ether/ethyl acetate 6:4). Yield: 75% (150 mg).

N-(3-[5-(3-Hydroxyphenyl)-2-thienyl]phenyl)methanesulfona-

mide (9). The title compound was prepared by reaction of

N-(3-[5-(3-methoxyphenyl)-2-thienyl]phenyl)methanesulfona-

mide (9a) (150 mg, 0.44mmol) andborontribromide (1.32mmol)

according to method C. The product was purified by preparative

TLC (hexane/ethyl acetate 1:1). Yield: 61% (92 mg). MS (ESI):

346 (M þ H)^þ. Anal. (C₁₇H₁₅NO₃S₂) C, H, N

3,3⁰-(3-Methylthiene-2,5-diyl]diphenol (13). The title com-

pound was prepared by reaction of 2,5-dibromo-3-methylthio-

phene (150 mg, 0.58 mmol), 3-hydroxybenzeneboronic acid (179

mg, 1.27 mmol), cesium carbonate (868 mg, 2.79 mmol), and

tetrakis(triphenylphosphine)palladium (6.7 mg, 5.8 μmol) ac-

cording to method A. The product was purified by CC (hexane/

ethyl acetate 4:6). Yield: 45% (73 mg). MS (ESI): 281 (M - H)^þ.

Anal. (C₁₇H₁₄O₂S) C, H, N.

3-Bromo-2,5-bis(3-methoxyphenyl)thiophene (14b). The title

compound was prepared by reaction of 3,5-dibromo-2-(3-meth-

oxyphenyl)thiophene (12c) (250 mg, 0.72 mmol), 3-methoxy-

benzeneboronic acid (134 mg, 0.86 mmol), sodium carbonate

(148 mg, 1.44 mmol), and tetrakis(triphenylphosphine)palla-

dium (41.6 mg, 36.0 μmol) according to method B. The product

was purified by CC (petroleum ether/ethyl acetate 9:1). Yield:

72% (194 mg).

2,5-Bis(3-methoxyphenyl)-3-phenylthiophene (14a). The title

compound was prepared by reaction of 3-bromo-2,5-bis(3-

methoxyphenyl)thiophene (14b) (102 mg, 0.27 mmol), benzene-

boronic acid (38 mg, 0.27 mmol), sodium carbonate (58 mg, 0.54

mmol), and tetrakis(triphenylphosphine)palladium (15.6 mg,

13.5 μmol) according to method B. The product was purified

by CC (petroleum ether/ethyl acetate 9:1). Yield: 54% (51 mg).

3,3⁰-(3-Phenylthiene-2,5-diyl)diphenol (14). The title com-

pound was prepared by reaction of 2,5-bis(3-methoxyphenyl)-

3-phenylthiophene (14a) (50 mg, 0.13 mmol) and boron tribro-

mide (0.78 mmol) according to method C. The product was

purified by preparative TLC (hexane/ethyl acetate 1:1). Yield:

53% (49 mg). MS (ESI): 345 (M þ H)^þ. Anal. (C₂₂H₁₆O₂S)

C, H, N.

2,3,5-Tris(3-methoxyphenyl)thiophene (15a). The title com-

pound was prepared by reaction of 3-bromo-2,5-bis(3-methox-

yphenyl)thiophene (14b) (102 mg, 0.27 mmol), 3-methoxy-

benzene boronic acid (42 mg, 0.27 mmol), sodium carbonate

(58 mg, 0.54 mmol), and tetrakis(triphenylphosphine)palladium

(15.6 mg, 13.5 μmol) according to method B. The product was

purified by CC (petroleum ether/ethyl acetate 9:1). Yield: 34%

(37 mg).

3,3⁰,3⁰⁰-Thiene-2,3,5-triyltriphenol (15). The title compound

was prepared by reaction of 2,3,5-tris(3-methoxyphenyl)-

thiophene (15a) (37 mg, 0.09 mmol) and boron tribromide

(0.81 mmol) according to method C. The product was purified

by preparative TLC (hexane/ethyl acetate 1:1). Yield: 67%

(21 mg). MS (ESI): 361 (M þ H)^þ. Anal. (C₂₂H₁₆O₃S) C, H, N.

5-(4-Methoxyphenyl)-2-(boronic acid)thiophene (16b). To a

solution of 2-bromo-5-(4-methoxyphenyl)thiophene (16c) (100

mg, 0.37 mmol, 1 equiv) in anhydrous THF cooled to -78 ꢀC for

5 min, n-BuLi (1.6 M in hexane, 0.28 mL, 0.44 mmol, 1.2 equiv)

was added dropwise and stirred at -78 ꢀC. After 15 min, triethyl

N-(3-[5-(3-Methoxyphenyl)-2-thienyl]benzyl)methanesulfona-

mide (10a). The title compound was prepared by reaction of 2-

bromo-5-(3-methoxyphenyl)thiophene (1b) (150 mg, 0.56 mmol),

3-[(methylsulfonylamino)methyl]benzeneboronic acid (153 mg,

0.67 mmol), cesium carbonate (382 mg, 1.23 mmol), and tetrakis-

(triphenylphosphine)palladium (6.5 mg, 5.6 μmol) according to

method A. The product was purified by CC (petroleum ether/

ethyl acetate 8:2). Yield: 58% (122 mg).

N-(3-[5-(3-Hydroxyphenyl)-2-thienyl]benzyl)methanesulfona-

mide (10). The title compound was prepared by reaction of

N-(3-[5-(3-methoxyphenyl)-2-thienyl]benzyl)methanesulfona-

mide (10a) (122 mg, 0.37 mmol) and boron tribromide (1.11

mmol) according to method C. The product was purified by

preparative TLC (hexane/ethyl acetate 1:1). Yield: 33% (44 mg).

MS (ESI): 360 (M þ H)^þ. Anal. (C₁₈H₁₇NO₃S₂) C, H, N.

N-(3-[5-(3-Methoxyphenyl)-2-thienyl]phenyl)-4-methylbenze-

nesulfonamide (11a). The title compound was prepared by reac-

tion of 2-bromo-5-(3-methoxyphenyl)thiophene (1b) (150 mg,

0.56 mmol), [3-[[(4-methylphenyl)sulfonyl]amino]phenyl]boronic

acid (195 mg, 0.67 mmol), cesium carbonate (383 mg, 1.23 mmol),

and tetrakis(triphenylphosphine)palladium (6.5 mg, 5.6 μmol)

according to method A. The product was purified by CC

(petroleum ether/ethyl acetate 9:1). Yield: 88% (214 mg).

N-(3-[5-(3-Hydroxyphenyl)-2-thienyl]phenyl)-4-methylbenzene-

sulfonamide (11). Thetitle compoundwas preparedbyreactionof

N-(3-[5-(3-methoxyphenyl)-2-thienyl]phenyl)-4-methylbenzene-

sulfonamide (11a) (214 mg, 0.49 mmol) and boron tribromide

(1.47 mmol) according to method C. The product was purified

by preparative TLC (hexane/ethyl acetate 1:1). Yield: 75%

(156 mg). MS (APCI): 421 (M)^þ. Anal. (C₂₃H₁₉NO₃S₂) C, H, N.

3,5-Dibromo-2-(3-methoxyphenyl)thiophene (12c). The title

compound was prepared by reaction of 2,3,5-tribromo-

thiophene (100 mg, 0.31 mmol), 3-methoxybenzeneboronic acid

(46 mg, 0.31 mmol), sodium carbonate (67 mg, 0.62 mmol), and

tetrakis(triphenylphosphine)palladium (17.9 mg, 15.5 μmol)

according to method B, heating the mixture 4 h instead of

20 h. The product was purified by CC (petroleum ether/ethyl

acetate 9:1). Yield: 23% (25 mg).

3-Bromo-2-(3-methoxyphenyl)-5-(4-methoxyphenyl)thio-

phene (12b). The title compound was prepared by reaction of

6738 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Bey et al.

borate (0.37 mL, 2.22 mmol, 6 equiv) was added at -78 ꢀC and

the mixture was stirred for 2 h. After warming to room tem-

perature, the crude material was acidified with 20 mL of a 1 N

hydrochloric acid solution. The aqueous layer was washed with

ethyl acetate. The combined organic layers were dried over

sodium sulfate, filtered, and evaporated under reduced pressure.

The title compound was not characterized and used without

further purification.

2-(3-Methoxyphenyl-5-methylphenyl)-5-(4-methoxyphenyl)-

thiophene (16a). The title compound was prepared by reaction of

1-bromo-3-methoxy-5-methylbenzene (150 mg, 0.74 mmol),

[5-(4-methoxyphenyl)-2-thienyl]boronic acid (16b) (206 mg,

0.88 mmol), sodium carbonate (181 mg, 1.76 mmol), and

tetrakis(triphenylphosphine)palladium (42.7 mg, 37.0 μmol)

according to method B. The product was purified by CC

(hexane/ethyl acetate 7:3). Yield: 22% (50 mg).

3-[5-(4-Hydroxyphenyl)-2-thienyl]-5-methylphenol (16). The

title compound was prepared by reaction of 2-(3-methoxyphe-

nyl-5-methylphenyl)-5-(4-methoxyphenyl)thiophene (16a) (50

mg, 0.16 mmol) and boron tribromide (0.96 mmol) according

to method C. The product was purified by preparative TLC

(hexane/ethyl acetate 1:1). Yield: 90% (41 mg). MS (ESI): 281

(M - H)^þ. Anal. (C₁₇H₁₄O₂S) C, H, N.

2-(3-Fluoro-5-methoxyphenyl)-5-(4-methoxyphenyl)thiophene

(17a). The title compound was prepared by reaction of 2-bromo-

5-(4-methoxyphenyl)thiophene (16c) (200 mg, 0.75 mmol),

3-fluoro-5-methoxybenzeneboronic acid (152 mg, 0.89 mmol),

cesium carbonate (513 mg, 1.65 mmol), and tetrakis(triphenyl-

phosphine)palladium (8.7 mg, 7.5 μmol) according to method

A. The product was purified by CC (hexane/ethyl acetate 7:3).

Yield: 43% (122 mg).

3-Fluoro-5-[5-(4-hydroxyphenyl)-2-thienyl]phenol (17). The ti-

tle compound was prepared by reaction of 2-(3-fluoro-5-meth-

oxyphenyl)-5-(4-methoxyphenyl)thiophene (17a) (100 mg,

0.32 mmol) and boron tribromide (1.92 mmol) according to

method C. The product was purified by preparative TLC

(hexane/ethyl acetate 7:3). Yield: 88% (80 mg). MS (APCI):

286 M^þ. Anal. (C₁₆H₁₁FO₂S) C, H, N.

5-[5-(4-Methoxyphenyl)-2-thienyl]-2-methylphenol (18a). The

title compound was prepared by reaction of 5-bromo-2-methyl-

phenol (250 mg, 1.34 mmol), [5-(4-methoxyphenyl)-2-

thienyl]boronic acid (16b) (690 mg, 2.95 mmol), cesium car-

bonate (914 mg, 2.94 mmol), and tetrakis(triphenylphos-

phine)palladium (15.5 mg, 13.4 μmol) according to method A.

The product was purified by CC (hexane/ethyl acetate 7:3).

Yield: 47% (193 mg).

5-[5-(4-Hydroxyphenyl)-2-thienyl]-2-methylphenol (18). The

title compound was prepared by reaction of 5-[5-(4-meth-

oxyphenyl)-2-thienyl]-2-methylphenol (18a) (161 mg, 0.54 mmol)

and boron tribromide (3.24 mmol) according to method C. The

product was purified by preparative TLC (hexane/ethyl acetate

7:3). Yield: 27% (42 mg). MS (ESI): 283 (M þ H)^þ. Anal.

(C₁₇H₁₄O₂S) C, H, N.

2-(4-Fluoro-3-methoxyphenyl)-5-(4-methoxyphenyl)thiophene

(19a). The title compound was prepared by reaction of 2-bromo-

5-(4-methoxyphenyl)thiophene (16c) (200 mg, 0.75 mmol),

4-fluoro-3-methoxybenzeneboronic acid (152 mg, 0.89 mmol),

cesium carbonate (553 mg, 1.78 mmol), and tetrakis(triphenyl-

phosphine)palladium (8.7 mg, 7.5 μmol) according to method

A. The product was purified by CC (hexane/ethyl acetate 7:3).

Yield: 49% (149 mg).

2-bromo-5-(4-methoxyphenyl)thiophene (16c) (195 mg, 0.73

mmol), 3,4-dimethoxybenzeneboronic acid (160 mg, 0.88mmol),

cesium carbonate (500 mg, 1.61 mmol), and tetrakis(triphenyl-

phosphine)palladium (8.4 mg, 7.3 μmol) according to method A.

The product was purified by CC (dichloromethane/methanol

99:1). Yield: 46% (119 mg).

4-[5-(4-Hydroxyphenyl)-2-thienyl]benzene-1,2-diol (20). The

title compound was prepared by reaction of 2-(3,4-

dimethoxyphenyl)-5-(4-methoxyphenyl)thiophene (20a) (100

mg, 0.31 mmol) and boron tribromide (2.79 mmol) according

to method C. The product was purified by preparative TLC

(hexane/ethyl acetate 1:1). Yield: 17% (49 mg). MS (ESI): 285

(M þ H)^þ. Anal. (C₁₆H₁₂O₃S) C, H, N.

2-(4-Methoxy-3-methylphenyl)-5-(3-methoxyphenyl)thiophene

(21a). The title compound was prepared by reaction of 2-bromo-

5-(3-methoxyphenyl)thiophene (1b) (250 mg, 0.92 mmol), 3-

methyl-4-methoxybenzeneboronic acid (152.8 mg, 0.92 mmol),

sodium carbonate (243 mg, 2.36 mmol), and tetrakis-

(triphenylphosphine)palladium (53.1 mg, 46.0 μmol) according

to method B. The product was purified by CC (hexane/ethyl

acetate 7:3). Yield: 54% (154 mg).

4-[5-(3-Hydroxyphenyl)-2-thienyl)-2-methyl]phenol (21). The

title compound was prepared by reaction of 2-(4-methoxy-3-

methylphenyl)-5-(3-methoxyphenyl)thiophene (21a) (100 mg,

0.32 mmol) and boron tribromide (1.92 mmol) according to

method C. The product was purified by preparative TLC

(hexane/ethyl acetate 6:4). Yield: 79% (72 mg). MS (ESI): 281

(M - H)^þ. Anal. (C₁₇H₁₄O₂S) C, H, N.

2-(3,4-Dimethoxyphenyl)-5-(3-methoxyphenyl)thiophene (22a).

The title compound was prepared by reaction of 2-bromo-5-(3-

methoxyphenyl)thiophene (1b) (206 mg, 1.14 mmol), 3,4-di-

methoxybenzeneboronic acid (247 mg, 1.36 mmol), cesium car-

bonate (779 mg, 2.51 mmol), and tetrakis(triphenylphos-

phine)palladium (13.2 mg, 11.4 μmol) according to method A.

The product was purified by CC (dichloromethane/methanol

99:1). Yield: 34% (126 mg).

4-[5-(3-Hydroxyphenyl)-2-thienyl]benzene-1,2-diol (22). The

title compound was prepared by reaction of 2-(3,4-

dimethoxyphenyl)-5-(3-methoxyphenyl)thiophene (22a) (100

mg, 0.32 mmol) and boron tribromide (2.88 mmol) according

to method C. The product was purified by preparative TLC

(hexane/ethyl acetate 6:4). Yield: 61% (56 mg). MS (ESI): 283

(M - H)^þ. Anal. (C₁₆H₁₂O₃S) C, H, N.

2-(3-Fluoro-4-methoxyphenyl)-5-(3-methoxyphenyl)thiophene

(23a). The title compound was prepared by reaction of 2-bromo-

5-(3-methoxyphenyl)thiophene (1b) (370 mg, 1.37 mmol), 3-

fluoro-4-methoxybenzeneboronic acid (255 mg, 1.50 mmol),

cesium carbonate (717 mg, 3.01 mmol), and tetrakis-

(triphenylphosphine)palladium (15.8 mg, 13.7 μmol) according

to method A. The product was purified by CC (hexane/ethyl

acetate 7:3). Yield: 98% (421 mg).

2-Fluoro-4-[5-(3-hydroxyphenyl)-2-thienyl]phenol (23). The ti-

tle compound was prepared by reaction of 2-(3-fluoro-4-meth-

oxyphenyl)-5-(3-methoxyphenyl)thiophene (23a) (240 mg, 0.76

mmol) and boron tribromide (4.56 mmol) according to method

C. The product was purified by preparative TLC (hexane/ethyl

acetate 1:1). Yield: 90% (195 mg). MS (ESI): 285 (M - H)^þ.

Anal. (C₁₆H₁₁FO₂S) C, H, N.

2-(3-Methoxyphenyl)-5-[4-methoxy-3-(trifluoromethyl)-

phenyl]thiophene (24a). The title compound was prepared

by reaction of 2-bromo-5-(3-methoxyphenyl)thiophene (1b) (408

mg, 1.51 mmol), 3-trifluoromethyl-4-methoxybenzeneboronic

acid (398 mg, 1.81 mmol), cesium carbonate (1033 mg, 3.32 mmol),

and tetrakis(triphenylphosphine)palladium (17.5 mg, 15.1 μmol)

according to method A. The product was purified by CC (hexane/

ethyl acetate 7:3). Yield: 75% (412 mg).

2-Fluoro-5-[5-(4-hydroxyphenyl)-2-thienyl]phenol (19). The ti-

tle compound was prepared by reaction of 2-(4-fluoro-3-meth-

oxyphenyl)-5-(4-methoxyphenyl)thiophene (19a) (100 mg,

0.32 mmol) and boron tribromide (1.92 mmol) according to

method C. The product was purified by preparative TLC

(hexane/ethyl acetate 7:3). Yield: 88% (80 mg). MS (ESI): 287

(M þ H)^þ. Anal. (C₁₆H₁₁FO₂S) C, H, N.

4-[5-(3-Hydroxyphenyl)-2-thienyl]-2-(trifluoromethyl)phenol

(24). The title compound was prepared by reaction of

2-(3-methoxyphenyl)-5-[4-methoxy-3-(trifluoromethyl)phenyl]-

thiophene (24a) (300 mg, 0.82 mmol) and boron tribromide

2-(3,4-Dimethoxyphenyl)-5-(4-methoxyphenyl)thiophene

(20a). The title compound was prepared by reaction of

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6739

(4.95 mmol) according to method C. The product was purified

by preparative TLC (hexane/ethyl acetate 1:1). Yield: 98% (272

mg). MS (ESI): 285 (M - H)^þ. Anal. (C₁₇H₁₁F₃O₂S) C, H, N.

2-(6-Methoxybiphenyl-3-yl)-5-(3-methoxyphenyl)thiophene

(25a). The title compound was prepared by reaction of 2-bromo-

5-(3-methoxyphenyl)thiophene (1b) (287 mg, 1.07 mmol),

[6-methoxy-1,1⁰-biphenyl-3-yl]boronic acid (25b) (338 mg, 1.29

mmol), sodium carbonate (250 mg, 2.35 mmol), and tetrakis-

(triphenylphosphine)palladium (61.8 mg, 53.5 μmol) according

to method B. The product was purified by CC (hexane/ethyl

acetate 7:3). Yield: 35% (135 mg).

brine, dried over sodium sulfate, evaporated under reduced

pressure, and purified by CC (hexane/ethyl acetate 7:3). Yield:

51% (120 mg). MS (ESI): 442 (M þ H)^þ.

(2E)-3-[2-Hydroxy-5-[5-(3-hydroxyphenyl)-2-thienyl]phenyl]-

N-phenylacrylamide (27). The title compound was prepared by

reaction of (2E)-3-[2-methoxy-5-[5-(3-methoxyphenyl)-2-thienyl]-

phenyl]-N-phenylacrylamide (27a) (55 mg, 0.13 mmol) and

boron tribromide (0.78 mmol) according to method C. The

product was purified by preparative TLC (hexane/ethyl acetate

4:6). Yield: 31% (17 mg). MS (ESI): 414 (M þ H)^þ. Anal.

(C₂₅H₁₉NO₃S) C, H, N.

5-[5-(3-Hydroxyphenyl)-2-thienyl]biphenyl-2-ol (25). The title

compound was prepared by reaction of 2-(6-methoxybiphenyl-

3-yl)-5-(3-methoxyphenyl)thiophene (25a) (100 mg, 0.26 mmol)

and boron tribromide (1.56 mmol) according to method C. The

product was purified by preparative TLC (hexane/ethyl acetate

1:1). Yield: 88% (81 mg). MS (ESI): 343 (M - H)^þ. Anal.

(C₂₂H₁₆O₂S) C, H, N.

[2-Methoxy-5-[5-(3-methoxyphenyl)-2-thienyl]phenyl]methanol

(26c). The title compound was prepared by reaction of 2-bromo-

5-(3-methoxyphenyl)thiophene (1b) (408 mg, 1.51 mmol),

[3-(hydroxymethyl)-4-methoxyphenyl]boronic acid (26d) (329

mg, 1.81 mmol), cesium carbonate (1032 mg, 3.32 mmol), and

tetrakis(triphenylphosphine)palladium (17.5 mg, 15.1 μmol)

according to method A. The product was purified by CC

(hexane/ethyl acetate 8:2). Yield: 12% (59 mg).

2-Methoxy-5-[5-(3-methoxyphenyl)-2-thienyl]benzaldehyde

(26b). To a solution of [2-methoxy-5-[5-(3-methoxyphenyl)-2-

thienyl]phenyl]methanol (26c) (100 mg, 0.31 mmol, 1 equiv) in

dichloromethane, pyridium chlorochromate (66 mg, 0.31 mmol,

1 equiv) was added in small portions over 5 min and stirred at

room temperature. After 30 min, the reaction was quenched

with water. The resulting organic layer was dried over sodium

sulfate, filtered, and concentrated to dryness. The title com-

pound was not characterized and used in the next step without

purification.

Ethyl (2E)-3-[2-Methoxy-5-[5-(3-methoxyphenyl)-2-thienyl]-

phenyl]acrylate (26a). To a solution of sodium hydride (10.4

mg, 0.43 mmol, 1 equiv) in anhydrous THF, triethyl phospho-

noacetate (93 μL, 0.46 mmol, 1.1 equiv) was added dropwise and

stirred at room temperature. After 15 min, 2-methoxy-5-[5-(3-

methoxyphenyl)-2-thienyl]benzaldehyde (26b) (100 mg, 0.31

mmol, 0.6 equiv) was added and stirred for 4 h at room

temperature. To quench the reaction water was added and

the resulting organic layer was washed with brine, dried over

sodium sulfate, filtered, evaporated, and purified by CC

(hexane/ethyl acetate 7:3). Yield: 98% (120 mg).

Ethyl (2E)-3-[2-Hydroxy-5-[5-(3-hydroxyphenyl)-2-thienyl]-

phenyl]acrylate (26). The title compound was prepared by reac-

tion of ethyl (2E)-3-[2-methoxy-5-[5-(3-methoxyphenyl)-2-

thienyl]phenyl]acrylate (26a) (60 mg, 0.15 mmol) and boron

tribromide (0.90 mmol) according to method C. The product

was purified by preparative TLC (hexane/ethyl acetate 1:1).

Yield: 17% (10 mg). MS (APCI): 366 (M)^þ. Anal. (C₂₁H₁₈O₄S)

C, H, N.

(2E)-3-[2-Methoxy-5-[5-(3-methoxyphenyl)-2-thienyl]phenyl]-

N-phenylacrylamide (27a). Ethyl (2E)-3-[2-methoxy-5-[5-(3-

methoxyphenyl)-2-thienyl]phenyl]acrylate (26a) (720 mg, 2.22

mmol, 1 equiv) in a solution of THF/water (2:1) was refluxed for

20 h together with lithium hydroxide (320 mg, 13.33 mmol, 6

equiv). After the mixture was cooled to room temperature, ether

was added, the aqueous layer was acidified with 1 N hydro-

chloric acid and washed with dichloromethane. The combined

organic layers were dried over sodium sulfate, filtered, and

evaporated under reduced pressure. The resulting carboxylic

acid was solubilized in dichloromethane (180 mg, 0.53 mmol,

1 equiv) and refluxed for 20 h with EDCI (102 mg, 0.53 mmol,

1 equiv) and HOBt (72 mg, 0.53 mmol, 1 equiv). After the

mixture was cooled to room temperature, the organic layer was

washed with a 1.5 M sodium hydrogenocarbonate solution,

3-[2-Methoxy-5-[5-(3-methoxyphenyl)-2-thienyl]phenyl]-N-

phenylpropanamide (28a). (2E)-3-[2-Methoxy-5-[5-(3-meth-

oxyphenyl)-2-thienyl]phenyl]-N-phenylacrylamide (27a) (50

mg, 0.11 mmol, 1 equiv) was solubilized in a mixture of THF/

EtOH (1:1). After addition of palladium hydroxide (1.7 mg,

0.01 mmol, 0.1 equiv) the mixture was stirred at room tempera-

ture under nitrogen atmosphere for 20 h. The crude mixture was

filtered, and the organic layer was evaporated under reduced

pressure. Yield: quantitative.

3-[2-Hydroxy-5-[5-(3-hydroxyphenyl)-2-thienyl]phenyl]-N-

phenylpropanamide (28). The title compound was prepared by

reaction of 3-[2-methoxy-5-[5-(3-methoxyphenyl)-2-thienyl]-

phenyl]-N-phenylpropanamide (28a) (55 mg, 0.13 mmol) and

boron tribromide (0.78 mmol) according to method C. The

product was purified by preparative TLC (hexane/ethyl acetate

1:1). Yield: 20% (10 mg). MS (ESI): 416 (M þ H)^þ. Anal.

(C₂₅H₂₁NO₃S) C, H, N.

2-Bromo-5-(3-fluoro-4-methoxyphenyl)thiophene (29b). The

title compound was prepared by reaction of 2,5-dibromothio-

phene (500 mg, 2.10 mmol), 3-fluoro-4-methoxybenzeneboronic

acid (357 mg, 2.10 mmol), sodium carbonate (432 mg, 4.20

mmol), and tetrakis(triphenylphosphine)palladium (121 mg,

1.05 mmol) according to method B. The product was purified

by CC (hexane/ethyl acetate 95:5). Yield: 85% (427 mg).

2-(3-Fluoro-4-methoxyphenyl)-5-(3-fluoro-5-methoxyphenyl)thio-

phene (29a). The title compound was prepared by reaction of

2-bromo-5-(3-fluoro-4-methoxyphenyl)thiophene (29b) (100 mg,

0.41 mmol), 3-fluoro-5-methoxybenzeneboronic acid (85 mg, 0.50

mmol), cesium carbonate (280 mg, 0.90 mmol), and tetrakis-

(triphenylphosphine)palladium (4.7 mg, 4.1 μmol) according to

method A. The product was purified by CC (hexane/ethyl acetate

9:1). Yield: 82% (113 mg).

2-Fluoro-4-[5-(3-fluoro-5-hydroxyphenyl)thien-2-yl]phenol (29).

The title compound was prepared by reaction of 2-(3-fluoro-4-

methoxyphenyl)-5-(3-fluoro-5-methoxyphenyl)thiophene (30a)

(100 mg, 0.30 mmol) and boron tribromide (1.80 mmol) according

to method C. The product was purified by preparative TLC

(hexane/ethyl acetate 6:4). Yield: 38% (35 mg). MS (APCI): 304

(M)^þ. Anal. (C₁₆H₁₀F₂O₂S) C, H, N.

2-(3-Fluoro-4-methoxyphenyl)-5-(4-fluoro-3-methoxyphenyl)-

thiophene (30a). The title compound was prepared by reac-

tion of 2-bromo-5-(3-fluoro-4-methoxyphenyl)thiophene (29b)

(100 mg, 0.41 mmol), 4-fluoro-3-methoxybenzeneboronic acid

(85 mg, 0.50 mmol), cesium carbonate (279 mg, 0.90 mmol), and

tetrakis(triphenylphosphine)palladium (4.7 mg, 4.1 μmol) ac-

cording to method A. The product was purified by CC (hexane/

ethyl acetate 9:1). Yield: 72% (100 mg).

2-Fluoro-4-[5-(4-fluoro-3-hydroxyphenyl)thien-2-yl]phenol

(30). The title compound was prepared by reaction of 2-(3-

fluoro-4-methoxyphenyl)-5-(3-fluoro-5-methoxyphenyl)thio-

phene (30a) (100 mg, 0.30 mmol) and boron tribromide (1.80

mmol) according to method C. The product was purified by

preparative TLC (hexane/ethyl acetate 6:4). Yield: 75% (69

mg). MS (APCI): 304 (M)^þ. Anal. (C₁₆H₁₀F₂O₂S) C, H, N.

2-(3,5-Difluoro-4-methoxyphenyl)-5-(3-methoxyphenyl)thio-

phene (31a). The title compound was prepared by reaction

of 2-bromo-5-(3-methoxyphenyl)thiophene (1b) (430 mg, 1.60

mmol), 3,5-difluoro-4-methoxybenzeneboronic acid (357 mg,

1.92 mmol), cesium carbonate (1094 mg, 3.52 mmol), and

6740 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Bey et al.

tetrakis(triphenylphosphine)palladium (18.5 mg, 16.0 μmol)

according to method A. The product was purified by CC

(petroleum ether/ethyl acetate 9:1). Yield: 42% (223 mg).

2,6-Difluoro-4-[5-(3-hydroxyphenyl)-2-thienyl]phenol (31). The

title compound was prepared by reaction of 2-(3,5-difluoro-4-

methoxyphenyl)-5-(3-methoxyphenyl)thiophene (31a) (220 mg,

0.62 mmol) and boron tribromide (3.72 mmol) according to

method C. The product was purified by preparative TLC

(hexane/ethyl acetate 5:5). Yield: 10% (18 mg). MS (ESI): 305

(M þ H)^þ. Anal. (C₁₆H₁₀F₂O₂S) C, H, N.

3,4⁰⁰-Dimethoxy-4-methyl-1,1⁰:4⁰,1⁰⁰-terphenyl (39a). The title

compound was prepared by reaction of 4⁰-bromo-3-methoxybi-

phenyl (38b) (230 mg, 0.87 mmol), 4-methoxy-3-methylbenze-

neboronic acid (172 mg, 1.04 mmol), cesium carbonate (595 mg,

1.91 mmol), and tetrakis(triphenylphosphine)palladium (10.1

mg, 8.7 μmol) according to method A. The product was purified

by CC (hexane/ethyl acetate 98:2). Yield: 53% (140 mg).

4-Methyl-1,1⁰:4⁰,1⁰⁰-terphenyl-3,4⁰⁰-diol (39). The title com-

pound was prepared by reaction of 3,3⁰⁰-dimethoxy-4-methyl-

1,1⁰:4⁰,1⁰⁰-terphenyl (41a) (120 mg, 0.39 mmol) and boron

tribromide (2.34 mmol) according to method C. The product

was purified by preparative TLC (dichloromethane/methanol

97:3). Yield: 38% (42 mg). MS (ESI): 277 (M þ H)^þ.

2-(3,4-Difluorophenyl)-5-(3-methoxyphenyl)thiophene (32a).

The title compound was prepared by reaction of 2-bromo-

5-(3-methoxyphenyl)thiophene (1b) (150 mg, 0.56 mmol), 3,4-

difluorobenzeneboronic acid (105 mg, 0.67 mmol), cesium

carbonate (383 mg, 1.23 mmol), and tetrakis(triphenylphos-

phine)palladium (6.5 mg, 5.6 μmol) according to method A.

The product was purified by CC (hexane/ethyl acetate 9:1).

Yield: 90% (152 mg).

3-Fluoro-3⁰⁰,4-dimethoxy-1,1⁰:4⁰,1⁰⁰-terphenyl (40a). The title

compound was prepared by reaction of 4⁰-bromo-3-methoxybi-

phenyl (38b) (175 mg, 0.67 mmol), 3-fluoro-4-methoxybenzene-

boronic acid (136.7 mg, 0.88 mmol), cesium carbonate (457 mg,

1.47 mmol), and tetrakis(triphenylphosphine)palladium (7.7

mg, 6.7 μmol) according to method A. The product was purified

by CC (hexane/ethyl acetate 98:2). Yield: 58% (117 mg).

3⁰⁰-Fluoro-1,1⁰:4⁰,1⁰⁰-terphenyl-3,4⁰⁰-diol (40). The title com-

pound was prepared by reaction of 3-fluoro-3⁰⁰,4-dimethoxy-

1,1⁰:4⁰,1⁰⁰-terphenyl (40a) (115 mg, 0.37 mmol) and boron

tribromide (2.22 mmol) according to method C. The product

was purified by preparative TLC (dichloromethane/methanol

99:1). Yield: 62% (65 mg). MS (ESI): 281 (M þ H)^þ.

3-[5-(3,4-Difluorophenyl)-2-thienyl]phenol (32). The title com-

pound was prepared by reaction of 2-(3,4-difluorophenyl)-5-(3-

methoxyphenyl)thiophene (32a) (120 mg, 0.40 mmol) and boron

tribromide (1.20 mmol) according to method C. The product

was purified by CC (dichloromethane/methanol 99:1). Yield:

85% (98 mg). MS (ESI): 289 (M þ H)^þ. Anal. (C₁₆H₉F₂OS) C,

H, N.

4-(4-Methoxy-3-methylphenyl)-2-(3-methoxyphenyl)thiophene

(34a). The title compound was prepared by reaction of 4-bromo-

2-(3-methoxyphenyl)thiophene (33b) (400 mg, 1.49 mmol),

3-methyl-4-methoxybenzeneboronic acid (296 mg, 1.79 mmol),

cesium carbonate (1019 mg, 3.27 mmol), and tetrakis-

(triphenylphosphine)palladium (17.2 mg, 14.9 μmol) according

to method A. The product was purified by CC (hexane/ethyl

acetate 7:3). Yield: 69% (320 mg).

4-[5-(3-Hydroxyphenyl)-3-thienyl]-2-methylphenol (34). The

title compound was prepared by reaction of 4-(4-methoxy-3-

methylphenyl)-2-(3-methoxyphenyl)thiophene (34a) (180 mg,

0.58 mmol) and boron tribromide (3.48 mmol) according to

method C. The product was purified by preparative TLC

(hexane/ethyl acetate 1:1). Yield: 54% (88 mg). MS (ESI): 281

(M - H)^þ. Anal. (C₁₇H₁₄O₂S) C, H, N.

4-(3-Fluoro-4-methoxyphenyl)-2-(3-methoxyphenyl)thiophene

(35a). The title compound was prepared by reaction of 4-bromo-

2-(3-methoxyphenyl)thiophene (33b) (400 mg, 1.49 mmol), 3-

fluoro-4-methoxybenzeneboronic acid (303 mg, 1.78 mmol),

cesium carbonate (1019 mg, 3.30 mmol), and tetrakis-

(triphenylphosphine)palladium (17.2 mg, 14.9 μmol) according

to method A. The product was purified by CC (petroleum ether/

ethyl acetate 9:1). Yield: 80% (403 mg).

2-Fluoro-4-[5-(3-hydroxyphenyl)-3-thienyl]phenol (35). The

title compound was prepared by reaction of 4-(3-fluoro-

4-methoxyphenyl)-2-(3-methoxyphenyl)thiophene (35a) (400

mg, 1.27 mmol) and boron tribromide (7.63 mmol) according

to method C. The product was purified by CC (dichloro-

methane/methanol 98:2). Yield: 22% (88 mg). MS (ESI): 287

(M - H)^-

4-(4-Methoxy-3-methylphenyl)-2-(3-methoxyphenyl)-1,3-thia-

zole (37a). The title compound was prepared by reaction of

5-bromo-2-(3-methoxyphenyl)-1,3-thiazole (36b) (402 mg, 1.49

mmol), 3-methyl-4-methoxybenzeneboronic acid (247 mg, 1.79

mmol), cesium carbonate (1019 mg, 3.27 mmol), and tetrakis-

(triphenylphosphine)palladium (17.2 mg, 14.9 μmol) according

to method A. The product was purified by CC (hexane/ethyl

acetate 7:3). Yield: 69% (320 mg).

4-[2-(3-Hydroxyphenyl)-1,3-thiazol-5-yl]-2-methylphenol (37).

The title compound was prepared by reaction of 4-(4-methoxy-

3-methylphenyl)-2-(3-methoxyphenyl)-1,3-thiazole (37a) (80 mg,

0.26 mmol) and boron tribromide (1.56 mmol) according to

method C. The product was purified by preparative TLC

(hexane/ethyl acetate 3:7). Yield: 16% (11 mg), MS (ESI): 274

(M þ H)^þ. Anal. (C₁₆H₁₃NO₂S) C, H, N.

4-Fluoro-3,3⁰⁰-dimethoxy-1,1⁰:4⁰,1⁰⁰-terphenyl (42a). The title

compound was prepared by reaction of 4⁰-bromo-3-methoxybi-

phenyl (38b) (200 mg, 0.76 mmol), 4-fluoro-3-methoxybenzene-

boronic acid (154 mg, 0.91 mmol), cesium carbonate (520 mg,

1.67 mmol), and tetrakis(triphenylphosphine)palladium (8.8

mg, 7.6 μmol) according to method A. The product was purified

by CC (hexane/ethyl acetate 9:1). Yield: 75% (175 mg).

4-Fluoro-1,1⁰:4⁰,1⁰⁰-terphenyl-3,3⁰⁰-diol (42). The title com-

pound was prepared by reaction of 4-fluoro-3,3⁰⁰-dimethoxy-

1,1⁰:4⁰,1⁰⁰-terphenyl (42a) (175 mg, 0.57 mmol) and boron

tribromide solution (3.42 mmol) according to method C. The

product was purified by preparative TLC (hexane/ethyl acetate

1:1). Yield: 63% (100 mg). MS (ESI): 281 (M þ H)^þ.

Biological Methods. [2,4,6,7-³H]E2 and [2,4,6,7-³H]E1 were

bought from Perkin-Elmer, Boston, MA. Quickszint Flow 302

scintillator fluid was bought from Zinsser Analytic, Frankfurt,

Germany.

17β-HSD1 and 17β-HSD2 were obtained from human pla-

centa according to previously described procedures.^31,37,56

Fresh human placenta was homogenized and centrifuged. The

pellet fraction contains the microsomal 17β-HSD2, while 17β-

HSD1 was obtained after precipitation with ammonium sulfate

from the cytosolic fraction.

1. Inhibition of 17β-HSD1. Inhibitory activities were evaluated

by a well established method with minor modifications.^57-59

Briefly, the enzyme preparation was incubated with NADH [500

μM] in the presence of potential inhibitors at 37 ꢀC in a

phosphate buffer (50 mM) supplemented with 20% of glycerol

and EDTA (1 mM). Inhibitor stock solutions were prepared

in DMSO. Final concentration of DMSO was adjusted to

1% in all samples. The enzymatic reaction was started by

addition of a mixture of unlabeled E1 and [2,4,6,7-³H]E1

(final concentration: 500 nM, 0.15 μCi). After 10 min, the

incubation was stopped with HgCl₂and the mixture was

extracted with ether. After evaporation, the steroids were dis-

solved in acetonitrile. E1 and E2 were separated using acetoni-

trile/water (45:55) as mobile phase in a C18 rp chromatography

column (Nucleodur C18 Gravity, 3 μm, Macherey-Nagel,

€

Duren, Germany) connected to a HPLC-system (Agilent 1100

Series, Agilent Technologies, Waldbronn, Germany). Detection

and quantification of the steroids were performed using a

radioflow detector (Berthold Technologies, Bad Wildbad,

Germany). The conversion rate was calculated according to

% E2

the following equation:% conversion ¼

ꢀ 100. Each

% E2þ% E1

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6741

value was calculated from at least three independent experi-

ments.

Compounds 21 and 23 were applied orally in a cassette dosing

in four rats at a dose of 10 mg/kg body weight by using a feeding

needle. The compounds were dissolved in a mixture labrasol/

water (1:1) and given at a volume of 5 mL/kg. Blood samples

(0.2 mL) were taken at 0, 1, 2, 3, 4, 6, 8, 10, and 24 h postdose and

collected in heparinised tubes. They were centrifuged at 3000g

for 10 min, and plasma was harvested and kept at -20 ꢀC until

analyzed.

HPLC-MS/MS analysis and quantification of the samples

was carried out on a Surveyor HPLC system coupled with a TSQ

Quantum (Thermo/Fisher) triple quadrupole mass spectro-

meter equipped with an electrospray interface (ESI).

Computational Chemistry. Molecular Modeling. All molecu-

lar modeling studies were performed on Intel(R) P4 CPU 3.00

GHz running Linux CentOS 5.2. The X-ray structures of 17β-

HSD1 (PDB codes 1A27, 1FDT, and 1I5R) were obtained from

the Protein Data Bank⁶²and further prepared using the BIO-

POLYMER module of SYBYL version 8.0 (Sybyl, Tripos Inc.,

St. Louis, MO). Water molecules, E2 (or HYC for 1I5R), and

sulfate ions were stripped from the PDB files, and missing

protein atoms were added and correct atom types set. Finally

hydrogen atoms and neutral end groups were added. All basic

and acidic residues were considered protonated and deproto-

nated, respectively. Since almost all histidines are oriented

toward the outer part of the enzyme, accessible for the surface,

they were considered as protonated (HIP) after a prediction run

made by MolProbity.⁶³For 1I5R the cofactor NADPH was

merged into the enzyme after an accurate overlay with the

hybrid inhibitor HYC and the X-rays 1A27 and 1FDT. Further,

every crystal structure was minimized for 500 steps with the

steepest descent minimizer as implemented in SYBYL with the

backbone atoms kept at fixed positions in order to fix close

contacts, followed by 2000 steps conjugate gradient minimiza-

tion requested for an overall better starting structure.

2. Inhibition of 17β-HSD2. The 17β-HSD2 inhibition assay

was performed similarly to the 17β-HSD1 procedure. The

microsomal fraction was incubated with NAD^þ[1500 μM], test

compound, and a mixture of unlabeled E2 and [2,4,6,7-³H]E2

(final concentration: 500 nM, 0.11 μCi) for 20 min at 37 ꢀC.

Further treatment of the samples and HPLC separation were

carried out as mentioned above.

3. ER Affinity. The binding affinity of selected compounds to

the ERR and ERβ was determined according to Zimmermann

et al.⁶⁰Briefly, 0.25 pmol of ERR or ERβ, respectively, was

incubated with [2,4,6,7-³H]E2 (10 nM) and test compound for

1 h at room temperature. The potential inhibitors were dissolved

in DMSO (5% final concentration). Nonspecific-binding was

performed with diethylstilbestrol (10 μM). After incubation,

ligand-receptor complexes were selectively bound to hydro-

xyapatite (5 g/ 60 mL TE-buffer). The formed complex was

separated, washed, and resuspended in ethanol. For radiodetec-

tion, scintillator cocktail (Quickszint 212, Zinsser Analytic,

Frankfurt, Germany) was added and samples were measured

in a liquid scintillation counter (Rack Beta Primo 1209, Wallac,

Turku, Finland). For determination of the relative binding

affinity (RBA), inhibitor and E2 concentrations required to

displace 50% of the receptor bound labeled E2 were determined.

RBA values were calculated according to the following equa-

IC₅₀ðE2Þ

tion:RBA ½%ꢁ ¼

_ðcompoundÞꢀ 100. The RBA value for E2 was

IC₅₀

arbitrarily set at 100%.

4. Evaluation of the Estrogenic Activity Using T-47D Cells.

Phenol red free medium was supplemented with sodium bicar-

bonate (2 g/L), streptomycin (100 μg/mL), insuline zinc salt (10

μg/mL), sodium pyruvate (1 mM), ^L-glutamine (2 mM), peni-

cillin (100 U/mL), and DCC-FCS 5% (v/v). RPMI 1640

(without phenol red) was used for the experiments. Cells were

grown for 48 h in phenol red free medium. Compound 21 was

added at a final concentration of 100 nM. Inhibitors and E2

were diluted in ethanol (final ethanol concentration was ad-

justed to 1%). As a positive control, E2 was added at a final

concentration of 0.1 nM. Ethanol was used as negative control.

Medium was changed every 2-3 days and supplemented with

the respective additive. After 8 days of incubation, the cell

viability was evaluated measuring the reduction of 3-(4,5-

dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide (MTT).

The cleavage of MTT to a blue formazan by mitochondrial

succinate dehydrogenase was quantified spectrophoto-

metrically at 590 nm as described by Denizot and Lang⁶¹with

minor modifications. The control proliferation was arbi-

trarily set at 1 and the stimulation induced by the inhi-

bitor was calculated according to following equation:

Inhibitor 23 was built with SYBYL and energy-minimized in

a MMFF94s force field as implemented in Sybyl. Subsequently

an ab initio geometry optimizations was performed in the gas

phase at the B3LYP/6-311þþG (d,p) level of density functional

theory (DFT) by means of the Gaussian 03 software,^64,65in

order to obtain the RESP charges of compound 23.

Two different softwares were used for docking studies:

GOLD, version 3.2,⁶⁶and Autodock4,^67,68using the graphical

user interface AutoDockTools (ADT 1.5.2). Since both allow

flexible docking of ligands, no conformational search was

employed to the ligand structure. For both programs the

compound 23 was docked in 50 independent genetic algorithm

(GA) runs.

GOLD, Version 3.2. Active-site origin was set at the center of

˚

the steroid binding site, while the radius was set equal to 13 A.

The automatic active-site detection was switched on. Further, a

slightly modified GOLDSCORE fitness function (increased

scaling for hydrophobic contacts) was used and genetic algo-

rithm default parameters were set as suggested by the GOLD

authors.

% stimulation ¼ ^{½proliferationðcompound-inducedÞ -1ꢁ}ꢀ 100. Eachvalue

is calculated as the mean value of at least three independent

experiments

½proliferationðE2-inducedÞ -1ꢁ

5. Inhibition of Human Hepatic CYPs. The commercially

available P450 inhibition kits from BD Gentest (Heidelberg,

Germany) were used according to the instructions of the man-

ufacturer. Compounds 21, 23, and 34 were tested for inhibition

of the following enzymes: CYP3A4, 2D6, and 2C19. Inhibitory

potencies were determined as IC₅₀values.

6. In Vivo Pharmacokinetics. Male Wistar rats weighing

300-330 g (Janvier France) were housed in a temperature-

controlled room (20-22 ꢀC) and maintained in a 12 h light/

12 h dark cycle. Food and water were available ad libitum. They

were anesthetized with a ketamine (135 mg/kg)/xylyzine (10 mg/

kg) mixture and cannulated with silicone tubing via the right

jugular vein and attached to the skull with dental cement. Prior

to the first blood sampling, animals were connected to a counter-

balanced system and tubing to perform blood sampling in the

freely moving rat.

Autodock4. The docking area has been defined by a box,

centered on the mass center of the CD-rings of the cocrystallized

˚

E2. Grids points of 60 ꢀ 70 ꢀ 74 with 0.375 A spacing were

calculated around the docking area for all the ligand atom types

using AutoGrid4. The Lamarckian genetic algorithm local

search (GALS) method was used. Each docking run was per-

formed with a population size of 200. A mutation rate of 0.02

and a crossover rate of 0.8 were used to generate new docking

trials for subsequent generations. The GALS method evaluates

a population of possible docking solutions and propagates the

most successful individuals from each generation into the next

one.

Both programs performed in a similar way, supporting the

herein suggested binding modes. The quality of the docked poses

was evaluated based mainly on visual inspection of the putative

binding modes of the ligand and on the scoring functions, which

6742 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21

Bey et al.

give a good measure of the discrimination between the found

binding modes for one single X-ray conformation but do not

help us to compare the poses of different X-rays.

dehydrogenase type 1 in maintenance of intratumoral high estra-

diol levels in postmenopausal breast cancers. Int. J. Cancer 2001,

94, 685–689.

(13) Suzuki, T.; Moriya, T.; Ariga, N.; Kaneko, C.; Kanazawa, M.;

Sasano, H. 17Beta-hydroxysteroid dehydrogenase type 1 and type

2 in human breast carcinoma: a correlation to clinicopathological

parameters. Br. J. Cancer 2000, 82, 518–523.

MEP. For selected compounds ab initio geometry optimiza-

tions were performed for the gas phase at the B3LYP/

6-311þþG (d,p) level of density functional theory (DFT) by

means of the Gaussian 03 software, and the molecular electro-

statics potential map (MEP) was plotted using GaussView,

version 3.0, the 3D molecular graphics package of Gaussian.⁶⁹

These electrostatic potential surfaces were generated by map-

ping 6-311þþG electrostatic potentials onto surfaces of molecu-

ꢀ

(14) Smuc, T.; Ribic Pucelj, M.; Sinkovec, J.; Husen, B.; Thole, H.;

ꢀ

Lanisnik Rizner, T. Expression analysis of the genes involved in

estradiol and progesterone action in human ovarian endometriosis.

Gynecol. Endocrinol. 2007, 23, 105–111.

(15) Husen, B.; Huhtinen, K.; Poutanen, M.; Kangas, L.; Messinger, J.;

Thole, H. Evaluation of inhibitors for 17beta-hydroxysteroid

dehydrogenase type 1 in vivo in immunodeficient mice inoculated

with MCF-7 cells stably expressing the recombinant human en-

zyme. Mol. Cell. Endocrinol. 2006, 248, 109–113.

(16) Husen, B.; Huhtinen, K.; Saloniemi, T.; Messinger, J.; Thole, H.;

Poutanen, M. Human hydroxysteroid (17-beta) dehydrogenase

1 expression enhances estrogen sensitivity of MCF-7 breast cancer

cell xenografts. Endocrinology 2006, 147, 5333–5339.

(17) Penning, T. M. 17Beta-hydroxysteroid dehydrogenase: inhibitors

and inhibitor design. Endocr. Relat. Cancer 1996, 3, 41–56.

(18) Poirier, D. Inhibitors of 17 beta-hydroxysteroid dehydrogenases.

Curr. Med. Chem. 2003, 10, 453–477.

˚

lar electron density (isovalue of 0.0002 e/A). The MEP maps are

color-coded, where red stands for negative values (3.1 ꢀ 10^-2

hartree) and blue for positive ones (4.5 ꢀ 10^-2hartree).

Acknowledgment. We are grateful to the Deutsche For-

schungsgemeinschaft (Grant HA1315/8-1) for financial sup-

port. We thank Anja Palusczak, Beate Geiger, and Jannine

Ludwig for their help in performing the enzyme inhibition

tests (17β-HSD1, 17β-HSD2, ERR, and ERβ) and Dr. Ursula

ꢀ

(19) Brozic, P.; Lanisnik Rizner, T.; Gobec, S. Inhibitors of 17beta-

hydroxysteroid dehydrogenase type 1. Curr. Med. Chem. 2008, 15,

137–150 and references cited therein .

€

Muller-Viera, Pharmacelsus CRO Saarbrucken, Germany,

for the hepatic CYP inhibition tests. Patricia Kruchten is

grateful to the European Postgraduate School 532 (DFG) for

a scholarship.

(20) Poirier, D. Advances in development of inhibitors of 17beta

hydroxysteroid dehydrogenases. Anti-Cancer Agents Med. Chem.

2009, 9, 642–660.

€

(21) Messinger, J.; Hirvela, L.; Husen, B.; Kangas, L.; Koskimies, P.;

Supporting Information Available: NMR spectroscopic data

for all compounds, elemental analysis results of compounds 3-6,

8-32, 34, and 37, purity determination of compounds 35, 39, 41,

and 42 using LC/MS, and overlay of the binding poses of

compound 23 obtained with the two 1FDT X-rays. This material

is available free of charge via the Internet at http://pubs.acs.org.

€

Pentikainen, O.; Saarenketo, P.; Thole, H. New inhibitors of

17beta-hydroxysteroid dehydrogenase type 1. Mol. Cell. Endocri-

nol. 2006, 248, 192–198.

(22) Karkola, S.; Lilienkampf, A.; Wahala, K. A 3D QSAR model of

€ € €

17beta-HSD1 inhibitors based on a thieno[2,3-d]pyrimidin-4(3H)-

one core applying molecular dynamics simulations and ligand-

protein docking. ChemMedChem 2008, 3, 461–472.

(23) Allan, G. M.; Vicker, N.; Lawrence, H. R.; Tutill, H. J.; Day, J. M.;

Huchet, M.; Ferrandis, E.; Reed, M. J.; Purohit, A.; Potter, B. V.

Novel inhibitors of 17beta-hydroxysteroid dehydrogenase type 1:

templates for design. Bioorg. Med. Chem. 2008, 16, 4438–4456.

(24) Frotscher, M.; Ziegler, E.; Marchais-Oberwinkler, S.; Kruchten,

References

(1) Travis, R. C.; Key, T. J. Oestrogen exposure and breast cancer risk.

Breast Cancer Res. 2003, 5, 239–247.

(2) Dizerega, G. S.; Barber, D. L.; Hodgen, G. D. Endometriosis: role

of ovarian steroids in initiation, maintenance and suppression.

Fertil. Steril. 1980, 33, 649–653.

(3) Miller, W. R.; Bartlett, J. M.; Canney, P.; Verrill, M. Hormonal

therapy for postmenopausal breast cancer: the science of sequen-

cing. Breast Cancer Res. Treat. 2007, 103, 149–160.

€

P.; Neugebauer, A.; Fetzer, L.; Scherer, C.; Muller-Vieira, U.;

Messinger, J.; Thole, H.; Hartmann, R. W. Design, synthesis and

biological evaluation of (hydroxyphenyl)-naphthalene and quino-

line derivatives: potent and selective non steroidal inhibitors of

17beta-hydroxysteroid dehydrogenase type 1 (17beta-HSD1) for

the treatment of estrogen-dependent diseases. J. Med. Chem. 2008,

51, 2158–2169.

(4) Bush, N. J. Advances in hormonal therapy for breast cancer.

Semin. Oncol. Nurs. 2007, 23, 46–54.

(25) Marchais-Oberwinkler, S.; Kruchten, P.; Frotscher, M.; Ziegler,

(5) Adamo, V.; Iorfida, M.; Montalto, E.; Festa, V.; Garipoli, C.;

Scimone, A.; Zanghi, M.; Caristi, N. Overview and new strategies

in metastatic breast cancer (MBC) for treatment of tamoxifen-

resistant patients. Ann. Oncol. 2007, 18 (Suppl. 6), vi53–vi57.

(6) Ohnesorg, T.; Keller, B.; Hrabe de Angelis, M.; Adamski, J.

Transcriptional regulation of human and murine 17beta-hydro-

xysteroid dehydrogenase type-7 confers its participation in choles-

terol biosynthesis. J. Mol. Endocrinol. 2006, 37, 185–197.

(7) Shehu, A.; Mao, J.; Gibori, G. B.; Halperin, J.; Le, J.; Devi, Y. S.;

Merrill, B.; Kiyokawa, H.; Gibori, G. Prolactin receptor-asso-

ciated protein/17beta-hydroxysteroid dehydrogenase type 7 gene

(Hsd17b7) plays a crucial role in embryonic development and fetal

survival. Mol. Endocrinol. 2008, 22, 2268–2277.

(8) Sakurai, N.; Miki, Y.; Suzuki, T.; Watanabe, K.; Narita, T.; Ando,

K.; Yung, T. M.; Aoki, D.; Sasano, H.; Handa, H. Systemic

distribution and tissue localizations of human 17beta-hydroxyster-

oid dehydrogenase type 12. J. Steroid Biochem. Mol. Biol. 2006, 99,

174–181.

(9) Day, J. M.; Foster, P. A.; Tutill, H. J.; Parsons, M. F.; Newman,

S. P.; Chander, S. K.; Allan, G. M.; Lawrence, H. R.; Vicker, N.;

Potter, B. V.; Reed, M. J.; Purohit, A. 17Beta-hydroxysteroid

dehydrogenase type 1, and not type 12, is a target for endocrine

therapy of hormone-dependent breast cancer. Int. J. Cancer 2008,

122, 1931–1940.

€

E.; Neugebauer, A.; Bhoga, U. D.; Bey, E.; Muller-Vieira, U.;

Messinger, J.; Thole, H.; Hartmann, R. W. Substituted 6-phenyl-2-

naphthols. Potent and selective non-steroidal inhibitors of 17β-

hydroxysteroid dehydrogenase type 1 (17beta-HSD1): design,

synthesis, biological evaluation and pharmacokinetics. J. Med.

Chem. 2008, 51, 4685–4698.

(26) Marchais-Oberwinkler, S.; Frotscher, M.; Ziegler, E.; Werth, R.;

Kruchten, P.; Messinger, J.; Thole, H.; Hartmann, R. W.

Structure-activity study in the class of 6-(3⁰-hydroxyphenyl)-

naphthalenes leading to an optimization of a pharmacophore

model for 17beta-hydroxysteroid dehydrogenase type 1 (17beta-

HSD1) inhibitors. Mol. Cell. Endocrinol. 2009, 301, 205–211.

(27) Bey, E.; Marchais-Oberwinkler, S.; Kruchten, P.; Frotscher, M.;

€

Werth, R.; Oster, A.; Algul, O.; Neugebauer, A.; Hartmann, R. W.

Design, synthesis and biological evaluation of bis(hydroxyphenyl)

azoles as potent and selective non-steroidal inhibitors of 17beta-

hydroxysteroid dehydrogenase type 1 (17beta-HSD1) for the treat-

ment of estrogen-dependent diseases. Bioorg. Med. Chem. 2008, 16,

6423–6435.

(28) Bey, E.; Marchais-Oberwinkler, S.; Werth, R.; Negri, M.; Al-Soud,

Y. A.; Kruchten, P.; Oster, A.; Frotscher, M.; Birk, B.; Hartmann,

R. W. Design, synthesis, biological evaluation and pharmacoki-

netics of bis(hydroxyphenyl) substituted azoles, thiophenes,

benzenes and aza-benzenes as potent and selective non-steroidal

inhibitors of 17beta-hydroxysteroid dehydrogenase type 1 (17beta-

HSD1). J. Med. Chem. 2008, 51, 6725–6739.

(29) Kruchten, P.; Werth, R.; Bey, E.; Oster, A.; Marchais-Oberwinkler,

S.; Frotscher, M.; Hartmann, R. W. Selective inhibition of 17beta-

hydroxysteroid dehydrogenase type 1 (17beta-HSD1) reduces

estrogen responsive cell growth of T-47D breast cancer cells.

J. Steroid Biochem. Mol. Biol. 2009, 114, 200–206.

(10) Gunnarsson, C.; Hellqvist, E.; Stal, O. 17Beta-hydroxysteroid

˚

dehydrogenases involved in local oestrogen synthesis have prog-

nostic significance in breast cancer. Br. J. Cancer 2005, 92, 547–552.

(11) Gunnarsson, C.; Olsson, B. M.; Stal, O. Abnormal expression of

˚

17beta-hydroxysteroid dehydrogenases in breast cancer predicts

late recurrence. Cancer Res. 2001, 61, 8448–8451.

(12) Miyoshi, Y.; Ando, A.; Shiba, E.; Taguchi, T.; Tamaki, Y.

Noguchi, S. Involvement of up-regulation of 17beta-hydroxysteroid

Article

Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6743

(30) Kruchten, P.; Werth, R.; Marchais-Oberwinkler, S.; Bey, E.;

Ziegler, E.; Oster, A.; Frotscher, M.; Hartmann, R. W. Develop-

ment of biological assays for the identification of selective inhibi-

tors of estradiol formation from estrone in rat liver preparations.

C. R. Chim. 2009, DOI: 10.1016/j.crci.2009.05.005.

(31) Kruchten, P.; Werth, R.; Marchais-Oberwinkler, S.; Frotscher, M.;

Hartmann, R. W. Development of a biological screening system for the

evaluation of highly active and selective 17beta-HSD1-inhibitors as

potential therapeutic agents. Mol. Cell. Endocrinol. 2009, 301, 154–157.

(32) Al-Soud, Y. A.; Bey, E.; Oster, A.; Marchais-Oberwinkler, S.;

Werth, R.; Kruchten, P.; Frotscher, M.; Hartmann, R. W. The role

of the heterocycle in bis(hydroxyphenyl) triazoles for inhibition of

17beta-hydroxysteroid dehydrogenase (17beta-HSD) type 1 and

type 2. Mol. Cell. Endocrinol. 2009, 301, 212–215.

growth of T47D breast cancer xenografts. Cancer Res. 2006, 66,

11207–11213.

(53) Smith, G.; Mikkelsen, G.; Eskildsen, J.; Bundgaard, C. The synthe-

sis and SAR of 2-arylsulfanylphenyl-1-oxyalkylamino acids as

GlyT-1 inhibitors. Bioorg. Med. Chem. 2006, 16, 3981–3884.

(54) Pfahl, M.; Al-Shamma, H. A.; Fanjul, A.; Pleynet, D. P. M.;

Spruce, L. W.; Wiemann, T. R.; Ibarra, J. B.; Tachdjian, C.

Preparation of benzylidenethiazolidenediones and analogs as anti-

tumor agents. WO 2002072009 A2 20020919, 2002.

(55) Duggan, H. M. E.; Leroux, F. G. M.; Malagu, K.; Martin, N. M. B.;

Menear, K. A.; Smith, G. C. M. 2-Methylmorpholine-substituted

pyrido-, pyrazo- and pyrimidopyrimidine derivatives as mTOR

inhibitors and their preparation, pharmaceutical compositions and

use in the treatment of cancer. WO 2008023161 A1 20080228, 2008.

(56) Zhu, D. W.; Lee, X.; Breton, R.; Ghosh, D.; Pangborn, W.; Duax,

W. L.; Lin, S.-X. Crystallization and preliminary X-ray diffraction

analysis of the complex of human placental 17beta-hydroxysteroid

dehydrogenase with NADP^þ. J. Mol. Biol. 1993, 234, 242–244.

(57) Lin, S.-X.; Yang, F.; Jin, J. Z.; Breton, R.; Zhu, D. W.; Luu-The,

V.; Labrie, F. Subunit identity of the dimeric 17beta-hydroxyster-

oid dehydrogenase from human placenta. J. Biol. Chem. 1992, 267,

16182–16187.

(33) Ghosh, D.; Pletnev, V. Z.; Zhu, D. W.; Wawrzak, Z.; Duax, W. L.;

Pangborn, W.; Labrie, F.; Lin, S.-X. Structure of human estrogenic

˚

17beta- hydroxysteroid dehydrogenase at 2.20 A resolution. Struc-

ture 1995, 3, 503–513.

(34) Azzi, A.; Rehse, P. H.; Zhu, D. W.; Campbell, R. L.; Labrie, F.;

Lin, S.-X. Crystal structure of human estrogenic 17beta-hydro-

xysteroid dehydrogenase complexed with 17 beta-estradiol. Nat.

Struct. Biol. 1996, 3, 665–668.

(35) Breton, R.; Housset, D.; Mazza, C.; Fontecilla-Camps, J. C. The

structure of a complex of human _þ17beta-hydroxysteroid dehydro-

genase with estradiol and NADP identifies two principal targets

for the design of inhibitors. Structure 1996, 4, 905–915.

(36) Mazza, C. Human Type I 17Beta-hydroxysteroid Dehydrogenase:

Site Directed Mutagenesis and X-ray Crystallography Structure-

(58) Sam, K. M.; Auger, S.; Luu-The, V.; Poirier, D. Steroidal spiro-

gamma-lactones that inhibit 17beta-hydroxysteroid dehydrogen-

ase activity in human placental microsomes. J. Med. Chem. 1995,

38, 4518–4528.

(59) Sam, K. M.; Boivin, R. P.; Tremblay, M. R.; Auger, S.; Poirier, D.

C16 and C17 derivatives of estradiol as inhibitors of 17beta-

hydroxysteroid dehydrogenase type 1: chemical synthesis and

structure-activity relationships. Drug. Des. Discovery 1998, 15,

157–180.

(60) Zimmermann, J.; Liebl, R.; von Angerer, E. 2,5-Diphenylfuran-

based pure antiestrogens with selectivity for the estrogen receptor

alpha. J. Steroid Biochem. Mol. Biol. 2005, 94, 57–66.

(61) Denizot, F.; Lang, R. Rapid colorimetric assay for cell growth and

survival. Modifications to the tetrazolium dye procedure giving

improved sensitivity and reliability. J. Immunol. Methods 1986, 89,

271–277.

ꢁ

-Function Analysis. Ph.D. Thesis, Universite Joseph Fourier,

Grenoble, France, 1997.

(37) Qiu, W.; Campbell, R. L.; Gangloff, A.; Dupuis, P.; Boivin, R. P.;

Tremblay, M. R.; Poirier, D.; Lin, S.-X. A concerted, rational

design of type 1 17beta-hydroxysteroid dehydrogenase inhibitors:

estradiol-adenosine hybrids with high affinity. FASEB J. 2002, 16,

1829–1831.

(38) Miyaura, N.; Suzuki, A. The palladium catalysed cross-coupling

reaction of phenylboronic acid with haloarenes in the presence of

bases. Synth. Commun. 1995, 11, 513–519.

(39) Togo, H.; Nogami, G.; Yokoyama, M. Synthetic application of

poly[styrene(iodoso diacetate)]. Synlett 1998, 5, 534–536.

(40) Emmons, W. D. The utility of phosphonate carbanions in olefin

synthesis. J. Am. Chem. Soc. 1961, 83, 1733–1738.

(41) Pascal, C.; Dubois, J.; Guenard, D.; Gueritte, F. Synthesis of

biphenyls mimicking the structure of the antimitotic rhazinilam.

J. Org. Chem. 1998, 63, 6414–6420.

(42) Hwang, K.-J.; O’Neil, J. P.; Katzenellenbogen, J. A. 5,6,11,12-

Tetrahydrochrysenes: synthesis of rigid stilbene systems designed

to be fluorescent ligands for the estrogen receptor. J. Org. Chem.

1992, 57, 1262–1271.

(43) Sherbet, D. P.; Guryev, O. L.; Papari-Zareei, M.; Mizrachi, D.;

Rambally, S.; Akbar, S.; Auchus, R. J. Biochemical factors govern-

ing the steady-state estrone/estradiol ratios catalyzed by human

17beta-hydroxysteroid dehydrogenases types 1 and 2 in HEK-293

cells. Endocrinology 2009, 150, 4154–4162.

(44) Sherbet, D. P.; Papari-Zareei, M.; Khan, N.; Sharma, K. K.;

Brandmaier, A.; Rambally, S.; Chattopadhyay, A.; Andersson,

S.; Agarwal, A. K.; Auchus, R. J. Cofactors, redox state, and

directional preferences of hydroxysteroid dehydrogenases. Mol.

Cell. Endocrinol. 2007, 265-266, 83–88.

(45) Bruno, I. J.; Cole, J. C.; Lommerse, J. P. M.; Rowland, R. S.;

Taylor, T.; Verdonk, M. L. Isostar: a library of information about

non-bonded interactions. J. Comput. Chem. 1997, 11, 523–537.

(46) Hagmann, W. K. The many roles for fluorine in medicinal chem-

istry. J. Med. Chem. 2008, 51, 4359–4369.

(62) Berman, H. M.; Westbrook, J.; Feng, Z.; Gilliland, G.; Bhat, T. N.;

Weissig, H.; Shindyalov, I. N.; Bourne, P. E. The Protein Data

Bank. Nucleic Acids Res. 2000, 28, 235–242.

(63) Davis, I. W.; Leaver-Fay, A.; Chen, V. B.; Block, J. N.; Kapral,

G. J.; Wang, X.; Murray, L. W.; Arendall, W. B.; Snoeyink, J.;

Richardson, J. S.; Richardson, D. C. MolProbity: all-atom con-

tacts and structure validation for proteins and nucleic acids.

Nucleic Acids Res. 2007, 35, W375–W383.

(64) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,

M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;

Kudin, K. N.; Burant, J. C.; Millam, J. M.; Lyengar, S. S.; Tomasi,

J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;

Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.;

Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.;

Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian,

H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.;

Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi,

R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;

Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.;

Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D.

K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.;

Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.;

Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.;

Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara,

A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.;

Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03; Gaussian,

Inc.: Pittsburgh, PA, 2004.

(65) Bayly, C. I.; Cieplak, P.; Cornell, W. D.; Kollman, P. A. A well

behaved electrostatic potential based method using charge re-

straints for determining atom-centered charges: the RESP model.

J. Phys. Chem. 1993, 97, 10269–10280.

(66) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R.

Development and validation of a genetic algorithm for flexible

docking. J. Mol. Biol. 1997, 267, 727–748.

(67) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart,

W. E.; Belew, R. K.; Olson, A. J. Automated docking using a

Lamarckian genetic algorithm and empirical binding free energy

function. J. Comput. Chem. 1998, 19, 1639–1662.

(68) Huey, R.; Morris, G. M.; Olson, A. J.; Goodsell, D. S. A semi-

empirical free energy force field with charge-based desolvation.

J. Comput. Chem. 2007, 28, 1145–1152.

ꢁ

(47) Bohm, H. J.; Banner, D.; Bendels, S.; Kansy, M.; Kuhn, B.; Muller,

K.; Obst-Sander, U.; Stahl, M. Fluorine in medicinal chemistry.

ChemBioChem 2004, 5, 637–643.

(48) Muller, K.; Faeh, C.; Diederich, F. Fluorine in pharmaceuticals:

looking beyond intuition. Science 2007, 317, 1881–1886.

ꢁ

(49) Poirier, D.; Boivin, R. P.; Tremblay, M. R.; Berube, M.; Qiu, W.;

Lin, S.-X. Estradiol-adenosine hybrid compounds designed to

inhibit type 1 17beta-hydroxysteroid dehydrogenase. J. Med.

Chem. 2005, 48, 8134–8147.

ꢁ

(50) Berube, M.; Poirier, D. Synthesis of simplified hybrid inhibitors of

type 1 17beta-hydroxysteroid dehydrogenase via cross-metathesis

and Sonogashira coupling reactions. Org. Lett. 2004, 6, 3127–3130.

(51) Fournier, D.; Poirier, D.; Mazumdar, M.; Lin, S.-X. Design and

synthesis of bisubstrate inhibitors of type 1 17beta-hydroxysteroid

dehydrogenase: overview and perspectives. Eur. J. Med. Chem.

2008, 43, 2298–2306.

(69) Dennington, I.; Roy, K. T.; Millam, J.; Eppinnett, K.; Howell, W.

L.; Gilliland, R. GaussView, version 3.0; Semichem Inc.: Shawnee

Mission, KS, 2003.

(52) Hartman, J.; Lindberg, K.; Morani, A.; Inzunza, J.; Strom, A.;

Gustafsson, J. A. Estrogen receptor beta inhibits angiogenesis and

Article Doi

New insights into the SAR and binding modes of bis(hydroxyphenyl)thiophenes and -benzenes: Influence of additional substituents on 17β-hydroxysteroid dehydrogenase type 1 (17β-HSD1) inhibitory activity and selectivity

DOI: 10.1021/jm901195w

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm901195w

Full text of DOI:10.1021/jm901195w

Products guided by the article

R&D Labs maybe for 1193525-64-2

Relevant to this article

Hot Product