Modification of estrone at the 6, 16, and 17 positions: Novel potent inhibitors of 17β-hydroxysteroid dehydrogenase type 1

J. Med. Chem. 2006, 49, 1325-1345

1325

Modification of Estrone at the 6, 16, and 17 Positions: Novel Potent Inhibitors of

17â-Hydroxysteroid Dehydrogenase Type 1

Gillian M. Allan,^†Harshani R. Lawrence,^†Josephine Cornet,^†Christian Bubert,^†Delphine S. Fischer,^†Nigel Vicker,^†

Andrew Smith,^†Helena J. Tutill,^‡Atul Purohit,^‡Joanna M. Day,^‡Mary F. Mahon,^§Michael J. Reed,^‡and Barry V. L. Potter^†,*

Medicinal Chemistry, Department of Pharmacy and Pharmacology and Sterix Ltd., UniVersity of Bath, ClaVerton Down, BA2 7AY, UK,

Endocrinology and Metabolic Medicine and Sterix Ltd., Faculty of Medicine, Imperial College London, St Mary’s Hospital,

London, W2 1NY, UK, and Department of Chemistry, UniVersity of Bath, BA2 7AY, UK

ReceiVed August 19, 2005

The 17â-hydroxysteroid dehydrogenases (17â-HSDs) catalyze the interconversion between the oxidized

and reduced forms of androgens and estrogens at the 17 position. The 17â-HSD type 1 enzyme (17â-

HSD1) catalyzes the reduction of estrone to estradiol and is expressed in malignant breast cells. Inhibitors

of this enzyme thus have potential as treatments for hormone dependent breast cancer. Here we report the

syntheses and biological evaluation of novel inhibitors based on the estrone or estradiol template. These

have been investigated by modification at the 6, 16 or 17 positions or combinations of these in order to gain

information about structure-activity relationships by probing different areas in the enzyme active site. Activity

data have been incorporated into a QSAR with predictive power, and the X-ray crystal structures of compounds

15 and 16c have been determined. Compound 15 has an IC₅₀of 320 nM for 17â-HSD1 and is selective for

17â-HSD1 over 17â-HSD2. Three libraries of amides are also reported that led to the identification of

inhibitors 19e and 20a, which have IC₅₀values of 510 and 380 nM respectively, and 20h which, having an

IC₅₀value of 37 nM, is the most potent inhibitor of 17â-HSD1 reported to date. These amides are also

selective for 17â-HSD1 over 17â-HSD2.

Introduction

androgens and estrogens at the 17 position. Although reversible,

their activity is mainly unidirectional and thus they can be

classified as oxidative or reductive. Thirteen members of this

enzyme family have been identified to date, eleven of which

exist in humans where they regulate the bioavailability of active

androgens and estrogens.⁵While all require NAD(P)H or NAD-

(P)⁺as cofactor, each type has a selective substrate affinity,

directional activity and a particular tissue distribution.

The 17â-HSD1 enzyme, which has a preferentially reductive

activity using NADPH as cofactor,^6,7is expressed in many

steroidogenic tissues, including breast tissue, and has been found

to be more highly expressed in malignant breast cells.^8,9E2 is

known to stimulate the growth and development of HDBC¹⁰

therefore inhibition of the final step in the synthesis of E2, by

the design of selective inhibitors of 17â-HSD1, is an attractive

option for the treatment of HDBC. Little progress has yet been

made in identifying inhibitors with proven in vivo activity using

this concept.

The 17â-HSD1 enzyme consists of 327 amino acid residues,

with a subunit mass of 35 kDa, and exists as a homodimer.¹¹

Much crystallographic information has been determined, includ-

ing that for the enzyme in its native form,¹²in complex with

estradiol and NADP,¹³with estradiol alone,¹⁴with equilin and

NADP¹⁵and with the inhibitor EM-1745.¹⁶This structural

information is an invaluable aid for identifying potential

inhibitors of 17â-HSD1 using molecular modeling.

Inhibitors of 17â-HSD1 have been reported by several groups,

and this field has recently been reviewed by Poirier.¹⁷Common

structural features can be identified which aid binding at the

active site. These include a phenol, which can undergo bifurcated

hydrogen bonding to His221 and Glu282 residues of the protein,

and a hydrophobic scaffold which inhabits the hydrophobic area

in the active site.

Breast cancer is one of the most common cancers in women

with an estimated global incidence in 2002 of approximately

1 150 000.¹Of these breast tumors the majority are initially

hormone-responsive with circulating estrogens playing a vital

role in their growth. Steroidogenic enzyme inhibitors can reduce

circulating and tissue levels of active estrogens by blocking their

biosynthetic pathways and thus can represent an effective

treatment for hormone dependent breast cancer (HDBC).

Aromatase inhibitors, which prevent the conversion of

androgens into estrogens, are currently used as an adjuvant

therapy to treat HDBC.²It has been proposed, however, that a

more important source of estrone (E1) in breast tumors is the

body’s reservoir of estrone 3-O-sulfate (E1S), that can be

converted to E1 by the action of estrone sulfatase: E1 production

in HDBC via the sulfatase pathway has been found to be

approximately 10-fold higher per gram of enzyme than via the

aromatase pathway.³Estrone sulfatase inhibitors have also been

investigated and are now progressing into clinical trials.⁴

Estrone itself, however, is not the most potent human estrogen

and, while inhibitors of E1 formation have reached an advanced

stage of use and development, another attractive target for the

treatment of HDBC is inhibition of 17â-hydroxysteroid dehy-

drogenase type 1 (17â-HSD1). This enzyme is responsible for

the reduction of the keto group of the weakly active E1 at the

17 position to give the most potent of the human estrogens,

17â estradiol (E2).

This enzyme is one of a class of enzymes known as the 17â-

hydroxysteroid dehydrogenases (17â-HSDs) that catalyze the

interconversion between the oxidized and reduced forms of

* Author to whom correspondence should be addressed. E-mail: B.V.L.

Potter@bath.ac.uk. Phone +44 1225 386639; Fax +44 1225 386114.

^†University of Bath, Department of Pharmacy & Pharmacology.

^‡Imperial College London.

The group of Poirier et al. has reported a number of inhibitors

of 17â-HSD1. These include E2 derivatives bearing a short side

^§University of Bath, Department of Chemistry.

Published on Web 01/26/2006

1326 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4

Allan et al.

Figure 1. 6-Oxo derivatives of E1.

chain at the C17R and C16R positions on the E2 skeleton¹⁸

and, more recently, novel hybrid inhibitors of 17â-HSD1.^16,19

The most potent of these hybrid inhibitors, EM-1745, has an

IC₅₀of 52 nM in an enzymatic assay and consists of an E2

template which is linked via an eight-methylene spacer to an

adenosine moiety. The crystal structure of the complex EM-

1745/17â-HSD1 was resolved to 1.6 Å and confirms that this

inhibitor interacts with both the substrate and the cofactor

enzyme-binding domains; it is, however, not drug-like. Two

patents from Solvay Pharmaceuticals have also recently been

published detailing nonsteroidal thiopyrimidone inhibitors²⁰and

steroidal inhibitors of 17â-HSD1 based on C15 substituted E1.²¹

The most active of these is a 3-O-methyl-15â-propyl-N-

cyclooctyl amide E1 derivative which shows 85% inhibition of

17â-HSD1 at 100 nM in a purified enzyme assay.

Figure 2. Oxime derivatives of E1 and 6-oxo-E1.

Figure 3. Pyrazolone derivative of E1 (9).

As part of our ongoing program to synthesize inhibitors of

17â-HSD1 we used E1 or E2 as a convenient scaffold and

investigated substitution at the 6, 16 and 17 positions of these

steroid templates, resulting in the filing of a patent by our

group.²²Herein is described the synthesis, molecular modeling

studies, and biological activities of some of these novel inhibitors

which have been investigated by expansion from the 6, 16 and

17 positions of the E1/E2 scaffold. The information gained about

the structure-activity relationships (SAR) for these inhibitors

of 17â-HSD1 will also be discussed in detail.

Results and Discussion

Our interest in the 6 position of the E1 template arose as a

result of docking studies, using the crystal structure of 17â-

HSD1 with E2 and NADP (PDB code 1FDT),^13,23that suggested

that a carbonyl at this position might display favorable interac-

tions in the active site. In particular, these studies indicated that

a small hydrogen bond accepting group at the 6 position might

capitalize on potentially beneficial interactions with the hydroxyl

of Ser222. Compounds 1-3 (Figure 1) were therefore synthe-

sized for comparison of the inhibitory activities of 6-oxo

compounds with their nonoxidized E1 counterparts. The non-

oxidized 16-hydroxymethylene and pyrazole derivatives of E1

have been previously reported by our group to be potent

inhibitors of 17â-HSD1.²⁴It was hypothesised that the combina-

tion of substituents at the 6 and the 16 or the 17 positions might

lead to additive effects.

This investigation was expanded by synthesis of oxime

derivatives 4-8, Figure 2. Compounds 4 and 5 were prepared

to create a pseudo E-ring system as discussed in our recent

publication;²⁴compounds 6-8 were synthesized with the aim

of further investigating the SAR around the 6 position in

combination with expansion from positions 16 and 17. It would

be desirable initially to convert the 6-oxo to another fairly small

substituent which would retain similar hydrogen bonding

properties. The oxime group was an attractive option for this

expansion as it can be easily introduced in one step without

3-O-protection of starting material and a large number of

hydroxylamines are commercially available for rapid lead

identification.

Figure 4. Derivatives of E1 and E2 with small substitutents at the 16

position.

novel E ring pyrazole and pyrazolone steroids. These have the

benefit of ease of further elaboration in different areas of space

and are achiral at the 16 and 17 positions. Some pyrazole

compounds have been reported previously by our group;²⁴here

we discuss the synthesis and activity of the novel pyrazolone

derivative of E1 9 (Figure 3).

Expansion from the 16 position alone is also an attractive

option, as the docking studies using 1FDT also suggested space

for significant extension from this position of the E1 scaffold,

possibly extending to interactions with the cofactor.

The scope for expansion at the 16 position was investigated

in a stepwise manner, starting with incorporation of small

substituents initially to give compounds 10-14 (Figure 4).

Compound 15 was obtained initially as a byproduct, as will be

discussed. Compounds 10-12 have been reported previously

to show promising inhibition of 17â-HSD1 at 10 µM of 77, 85

and 95% respectively.²⁴

Further to the high inhibitory activity of 12, the tolerance of

the active site for larger hydrophobic groups at the 16 position

was investigated by the incorporation of alkenyl substituents

to give a series of compounds 16 as shown in Figure 5. The

16-methylene derivative of E1 has been reported to be an active-

site directed alkylating agent of 17â-HSD1 which can be formed

As an extension of this work, the 16 position was further

explored in combination with the 17 position in the form of

Inhibitors of 17â-Hydroxysteroid Dehydrogenase Type 1

Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4 1327

Figure 6. Carboxyl derivative of E2 (19) and methyl carboxyl

derivatives of E1 (20) and E2 (21).

Figure 5. Alkenyl derivatives of E1 (16) and E2 (17) and 16â-alkyl

derivatives of E2 (18).

a

Table 2

a

Table 1

^ans ) not synthesized.

in situ by oxidation of the suicide inhibitor 16-methylene-E2.²⁵

Therefore 16-methylene-E1 and 16-methylene-E2 were synthe-

sized from 3-O-acetyl E1, using the method of Ringold and

Rosenkranz,²⁶for use as standards for comparison with 16-

alkenyl derivatives. Alkyl substituents were investigated by

synthesis of isopropyl and isobutyl compounds and aryl sub-

stituents containing heteroatoms in different areas in space were

also investigated. These were accessed via condensation reac-

^ans ) not synthesized.

tions of various aldehydes with E1 and are listed in Table 1.

The effect of reduction to the E2 derivatives 17 was also

investigated, and, for selected compounds, hydrogenation

1328 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4

Allan et al.

Chemistry

Oxidation at the 6 Position of E1 and Derivatives. The

syntheses of compounds 1-3 are shown in Scheme 1(a-c).

The oxidation of E1 at the 6 position was performed in the

presence of chromic acid by an adaptation of the procedure of

Schwenk and Montclair.²⁸To prevent oxidation at the 3 position

E1 was first protected as the 3-O-acetate derivative before

reaction with chromium trioxide in AcOH/water at ∼15 °C. The

yield of 23 after recrystallization was 20% which is in

accordance with literature values, as formation of byproducts

accounts for the low yield.²⁹Removal of the acetate using KOH/

MeOH gave compound 1 in 58% yield. Compound 2 was

prepared in a similar manner from the hydroxymethylene

derivative of E1, 24, the synthesis of which has been previously

described.³⁰This was protected as its bis-acetate derivative 25

and in this case the oxidation of 25 progressed in only 10%

yield. Removal of the acetate groups using K₂CO₃/MeOH gave

the compound 2 in 54% yield. In a similar manner the E-ring

pyrazole derivative of E1 was protected as the bis-acetate 28,

at the 3-O-position and on the 1′-N or 2′-N of the pyrazole ring,

and oxidation was attempted as before. In this case, however,

no 6-oxo product was observed (¹H NMR showed no C4-H

shift associated with deshielding by the new carbonyl group).

The use of chromium hexacarbonyl and tert-butyl hydroperoxide

in acetonitrile at reflux, as described by Pearson and Han,³¹did

give the oxidized product 29 in low yield after purification by

Figure 7. The 16â-diastereomer of compound 20h: the most potent

inhibitor of 17â-HSD1 reported to date.

of the alkene double bond was performed to give 16â-alkyl

compounds 18.

Since the activities of these compounds seemed promising it

was decided to incorporate a more flexible, functionalizable

substituent at the 16 position from which to probe this area

further in the form of syntheses of focused libraries. Two of

such linkers were investigated initially: a carboxylic acid in

19 and a methyl carboxylic acid in both 20 and 21, as shown

in Figure 6. These could then be reacted with a diverse set of

commercially available amines which was chosen to exhibit a

range of electron accepting, electron donating, charge transfer,

hydrophobic, hydrophilic and π-stacking properties, with the

aim of exploring the SAR in this area of the active site. The

amides synthesized for each of these templates are listed in Table

2.

Extension from the 16 position of E1 via the methyl

carboxylic linker resulted in the discovery of a highly potent

inhibitor 20h (Figure 7) that has an IC₅₀) 37 nM and shows

selectivity for 17â HSD1 over 17â HSD2, as recently reported

by our group in a preliminary report.²⁷

Scheme 1^a

^aReagents and conditions: (a) Ac₂O, pyr, 0 °C then ∆; (b) CrO₃, AcOH, 10-15 °C then rt; (c) KOH, MeOH, rt; (d) Ac₂O, pyr, 0 °C then ∆; (e) K₂CO₃

(aq), MeOH, rt; (f) t-BuO₂H, Cr(CO)₆, MeCN, ∆; (g) KOH(aq), EtOH, rt.

Inhibitors of 17â-Hydroxysteroid Dehydrogenase Type 1

Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4 1329

Scheme 2^a

^aReagents and conditions: (a) NH₂OH‚HCl, NaOAc, MeOH/H₂O, rt; (b) NH₂OMe‚HCl, NaOAc, MeOH/H₂O, rt.

Scheme 3^a

Figure 8. Possible tautomeric forms of compound 9.

Scheme 4^a

^aReagents and conditions: (a) BnBr, K₂CO₃, DMF, rt; (b) CH₃OCO₂CH₃,

NaH, THF, ∆; (c) H₂, Pd-C, THF, rt (d) NH₂NH₂‚H₂O, PhMe, ∆.

chromatography. Removal of both acetate groups using KOH/

EtOH yielded product 3.

Syntheses of Oxime Derivatives of E1. The syntheses of

compounds 4 and 5 have been previously described.^24,32

Compound 6 (Scheme 2) was prepared in the same manner by

adaptation of a literature procedure using hydroxylamine

hydrochloride and sodium acetate in ethanol/water.³³Reaction

of 1 with hydroxylamine hydrochloride or methoxylamine

hydrochloride under the same conditions gave compounds 7 or

8 respectively. Compound 7 was obtained in 41% yield after

purification by flash chromatography; the less polar compound

8 was obtained in a high 92% yield after recrystallization from

ethyl acetate/hexane. Compound 6 has been shown previously

to exist as the anti-isomer,³³and compounds 4-8 were all found

by NMR to consist of single geometrical isomers; however, as

yet these have not been assigned. It is likely that these are the

anti-isomers as these are the favored thermodynamic products.

^aReagents and conditions: (a) NaBH₄, THF/EtOH, rt.

pyrazolone to a six membered ring or an acyclic system.

Fortunately, the reaction does proceed under forcing conditions

at reflux in toluene in a sealed tube. Purification was problem-

atic, however, as in addition to the desired pyrazolone product

the hydrazone derivative of E1 is formed (with loss of the

carboxyl ester functionality) and is difficult to separate. The

product was precipitated selectively from AcOH/water to give

18% of pure compound 9. This compound can exist in different

tautomeric forms as shown in Figure 8. To investigate which

forms may predominate in solution and solid phases, ¹³C NMR

was performed in DMSO-d₆and IR spectroscopy in solid form

(KBr disk). The ¹³C NMR spectrum showed one major set of

peaks with a peak at 171 ppm which is indicative that in this

solution a keto-form predominates; the IR spectrum showed a

strong absorption at 1600 cm^-1which also suggests the presence

of a carbonyl functionality in the solid form.

Expansion from the 16 Position of E1. The syntheses of

compounds 10-12 have been described previously.²⁴Com-

pounds 13-15 were synthesized as shown in Schemes 4-6.

The 16 cyano substituted E2 derivative 13 was obtained by

reduction of 10 using sodium borohydride in THF/EtOH in 37%

yield after recrystallization from IPA/water. The stereochemistry

of this compound at the 16 position was assigned as â by

comparison of NMR coupling constants with those reported for

the R isomer.^{37 1}H NMR of the crude product prior to

recrystallization showed the presence of the 16R isomer as only

5% of the reduced products obtained.

Substitution at the 16/17 Positions. The route to the

pyrazolone compound 9 is shown in Scheme 3. Estrone was

protected as its 3-O-benzyl ether by a literature procedure using

benzyl bromide and K₂CO₃in DMF.³⁴The protected compound

30 was then reacted using the procedure of Paquette et al. with

sodium hydride and dimethyl carbonate in THF at reflux

followed by acidification at 0 °C then recrystallization from

MeOH to give 31 in 81% yield.³⁵Hydrogenation using 10%

Pd-C (catalytic) gave 32 in 93% yield after recrystallization

from methanol. The pyrazolone compound 9 was obtained from

condensation of 32 with hydrazine monohydrate, although this

condensation reaction was found not to be successful using

general literature described methods at reflux in EtOH.³⁶This

is probably due to the fact that the new E-ring formed will be

strained due to being attached to the five-membered D-ring and

thus the reaction will be more difficult than if attaching the

Compound 14 was obtained in two steps from 31. Reduction

of the latter using sodium borohydride at 0 °C in THF/MeOH

1330 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4

Allan et al.

Scheme 5^a

^aReagents and conditions: (a) NaBH₄, THF/MeOH, 0-20 °C; (b) H₂, Pd-C, THF, rt.

Scheme 6^a

Scheme 7^a

^aReagents and conditions: (a) RCHO, LDA, THF, -78 °C or ArCHO,

NaOH, EtOH, rt; (b) NaBH₄, THF/EtOH, 0 °C then rt; (c) H₂, Pd-C, THF/

EtOH, rt.

^aReagents and conditions: (a) (CH₂O)_n, i-AmOH, ∆; (b) KOH(aq),

EtOH, 0 °C.

Figure 9. ORTEX⁴⁰plot of the X-ray crystal structure of 15. Ellipsoids

are shown at the 30% probability level.

followed by acidification gave 33 in 13% yield as one of the

products isolated by flash chromatography (the other products

being the separated diastereomers of 3-O-benzyl-16-carboxy-

methyl-E2 31). Subsequent hydrogenation of 33 using catalytic

Pd-C (10wt %) in THF gave compound 14 in 93% yield.

Compounds 33 and 14 have been described in the literature

therefore assignment of the 16-substituent as â was based on

comparison with reported NMR data.³⁸

The 16-ethoxymethyl derivative 15 was initially isolated as

a byproduct, in this case from deprotection of 3-O-acetyl-16-

methylene E1 using KOH in EtOH as shown in Scheme 6.

Compound 15 was obtained from this reaction in 13% yield

and was shown by X-ray crystallography to possess the ethyl

ether substituent in the â orientation (Figure 9). This compound

was subsequently prepared in higher yield in two steps from

the 3-O-benzyl protected version of the hydroxymethylene

compound 24. This could be alkylated according to literature

procedure³⁹using potassium carbonate and ethyl iodide before

hydrogenation of the alkene double bond. This hydrogenation

did yield a mixture of products; however, the desired product

15 was obtained as the major product in 46% yield.

Figure 10. (a) ORTEX⁴⁰plot of the X-ray crystal structure of a

polymorph of 16c in which pyridyl ring is in plane with E1 skeleton

(blocklike crystals) and (b) ORTEX⁴⁰plot of the X-ray crystal structure

of a polymorph of 16c in which the pyridyl ring is out of plane with

E1 skeleton (needlelike crystals). Ellipsoids are shown at the 30%

probability level.

condensation of aromatic aldehydes with E1 using NaOH in

EtOH at room temperature. The syntheses of some of these

compounds at high temperature have been previously reported

to yield the Z-isomers.⁴²However compound 16c was crystal-

lized from EtOH and shown by X-ray crystallography to possess

the alkene double bond in the E-orientation. Two polymorphs

were obtained from this recrystallization: colorless blocks and

needle-like crystals (Figures 10a and 10b, respectively). These

exhibited differing orientations of the phenolic OH and of the

pyridyl ring, and this impacted heavily on the hydrogen bonds

that direct the crystal packing. The remaining aromatic analogues

of the series were assigned the geometry E by analogy since

they were prepared in the same manner and showed similar

chemical shifts for their alkene protons. Subsequent reduction

using NaBH₄gave compounds 17 in high yield (average 85%)

Compounds 16 were formed by condensation of E1 with the

appropriate aldehyde (Scheme 7). For 16e-f reactions were

started at -78 °C using LDA as base then allowed to warm to

ambient temperature to give products in 64 and 21% yield

respectively after work up and purification. These were assigned

as the more thermodynamically stable E isomer based on

comparison of NMR with published data.⁴¹The remaining

compounds 16 were obtained in high yields (average 87%) by

Inhibitors of 17â-Hydroxysteroid Dehydrogenase Type 1

Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4 1331

Scheme 8^a

^aReagents and conditions: (a) NaBH₄, THF/MeOH, 0 °C; (b) H₂, Pd-C, THF, rt; (c) NaOH(aq), MeOH, rt; (d) Oxime resin, DIC, HOBt, DMF, rt; (e)

furfurylamine, DCM, 40 °C; (f) NaOH (aq), THF, rt; (g) (p-NO₂C₆H₄O)₂CO, NEt₃, DMF, rt; (h) R₁R₂NH, MeCN, rt; (i) H₂, Pd-C, THF, rt.

Scheme 9^a

Scheme 10^a

^aReagents and conditions: (a) R₁R₂NH, EDC, DMAP, DCM, rt.

amides; however, decarboxylation of the â-keto acid proved to

be a significant problem with this approach. The carboxymethyl

amides were therefore investigated using the reduced (E2)

scaffold which was obtained by reduction with sodium boro-

hydride at 0 °C to give 36a and 36b as shown in Scheme 8.

This reduction was kept at 0 °C for 4 h since allowing the

reaction to warm to 20 °C over a longer time (18 h) resulted in

further reduction to form 33 (Scheme 5). In this case the

diastereomers were separated by chromatography in 48% and

18% yields and the major product 36a was used for further

reaction. This was assigned as the 16â-carboxyl ester by

comparison of ¹H NMR coupling constants for the two isomers,

^aReagents and conditions: (a) NaOH (aq), THF/MeOH, ∆; (b) H₂, Pd-

C, EtOH, rt; (c) NaBH₄, MeOH, 0 °C to room temperature.

1

and then hydrogenation using Pd-C catalyst yielded compounds

18 (average yield 71% after purification by chromatography).

Compounds 18 were assigned as having the 16-substituent in

the expected â orientation based on comparison of NMR spectra

with those of known structures.⁴³

A similar approach was used to attach the carboxymethyl

linker using 3-O-benzyl-E1 with NaH and dimethyl carbonate

to give 31 (Scheme 3) which was obtained in a 1:1.9 ratio of

diastereomers after recrystallization from methanol. The initial

intention was then to saponify the ester and form a library of

and comparison of the H NMR of 37 with published NMR

data for the 16R diastereomer.³⁹The carbon center at the 16

position was however found to epimerize during subsequent

saponification. Initially, saponification followed by amide

coupling was attempted; however, amide coupling reactions

were found to be problematic using standard coupling condi-

tions. Two routes were thus investigated for amide synthesis,

both involving activation of the starting carboxylic acid: (Route

A) removal the benzyl protecting group first to give 37, followed

by saponification and loading of 19 onto Oxime resin for solid-

1332 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4

Allan et al.

Scheme 11^a

^aReagents and conditions: (a) Oxime resin, DIC, HOBt, DMF, rt; (b) R₁R₂NH, DCM, 40 °C; (c) H₂, Pd-C, MeOH, rt; (d) NaBH₄, THF/EtOH, 0 °C to

room temperature.

Scheme 12^a

for these compounds was 40% based on a loading of 1.06

mmol/g on Oxime resin. Amides 20j-m were also formed via

a solid-phase route but in these cases the unprotected precursor

20 was loaded onto Oxime resin to give intermediate 48

(Scheme 12). The amides were therefore obtained directly by

reaction of 48 with the specific amines at 40 °C. Once again,

an average yield of 40% was obtained.

Of inhibitors 21a-f, amide 21a was obtained by solution

phase amide coupling of 21 with a yield of 33%; 21b-f were

obtained by sodium borohydride reduction of 20b-f. This

reduction progressed cleanly in quantitative yield.

Biological Activity: Inhibition of 17â-HSD Type 1 and

Type 2

^aReagents and conditions: (a) Oxime resin, DIC, HOBt, DMF, rt; (b)

R₁R₂NH, DCM, 40 °C.

The compounds synthesized were tested for their ability to

inhibit 17â-HSD1 activity in T47-D cells. As a measure of

selectivity, their ability to inhibit 17â-HSD2 activity was also

measured in MDA-MB-231 cells. Since 17â-HSD2 catalyses

the oxidative (inactivating) process, ability to inhibit this enzyme

is not a desirable property for potential treatments for HDBC.

These evaluations of inhibition were performed by measuring

the amount of labeled E1 or E2 formed from the labeled natural

substrate in the presence of the required cofactor. The percentage

of inhibition is then calculated by comparison of conversion in

the absence and presence of inhibitor. The percentages of

inhibition achieved for a 10 µM concentration of the inhibitors,

as well as the IC₅₀values for some of the most potent

compounds, are shown in Tables 3-8. Compounds which gave

less than 10% inhibition at 10 µM in the 17â-HSD2 assay were

considered to be inactive against this enzyme.

Oxo, Oxime, and Pyrazolone Derivatives of E1. 6-Oxo

Derivatives of E1. The inhibition data for the 6-oxo derivatives

of E1 are shown in Table 3. The activities of the corresponding

nonoxidized E1 derivatives are also shown for comparison. From

these results it can be concluded that the presence of the 6-oxo

substituent does not significantly enhance activity. The 6-oxo

derivative of E1, 1, has comparable activity to the natural

substrate itself which indicates that although the keto group is

well tolerated it does not enhance binding in the active site.

For compounds 2 and 3 a slightly detrimental effect was

observed which suggests that these combinations of substitution

at the 6 and 17 positions are not favored.

phase synthesis, or (Route B) saponification first to give 39

followed by formation of the active ester 40 for amide coupling

before deprotection. Route A was used to synthesize the furfuryl

amide 19d but in only 4% yield over the final two steps (loading

on resin followed by amide coupling). Route B was found to

be superior since the active ester 40 was obtained in 52% yield,

and the average yields for amide coupling and debenzylation

were 80% and 36% respectively. Amides 19a-c and 19e-j

were thus synthesized using Route B.

The methylcarboxyl linker was attached to 3-O-benzyl-E1

in 70% yield using LDA and bromoethyl acetate to give 42 as

previously described (Scheme 9).³⁹In this case decarboxylation

of the resulting γ-keto acid was no longer a problem, enabling

the synthesis of libraries of both E1 and E2 derivatives. The

syntheses of methyl carboxylic acid intermediates are also shown

in Scheme 9. Compound 42 could be debenzylated first to give

44, which was then saponified to give 20, or could be reduced

to the E2 derivative 45 followed by saponification to give 21.

Acids 20 and 21 were then used for solution phase amide

couplings using standard conditions (EDC, DMAP, DCM)

(Scheme 10). Other amides were synthesized via solid-phase

reactions using Oxime resin (Schemes 11 and 12). Scheme 11

shows the solid-phase route using the 3-O-benzyl intermediate

43; Scheme 12 shows this route using acid 20. All of these

amide couplings progressed in moderate yields. In the case of

the E1 derivatives, amides 20a, 20g, 20h and 20i were prepared

by solution phase chemistry as shown in Scheme 10 with yields

of 40, 33, 50, and 40% respectively. Amides 20b-f were

prepared by solid-phase amide formation followed by deben-

zylation as illustrated in Scheme 11. The average overall yield

Oxime Derivatives of E1. By comparing the activities of

compounds 4-8 information can be gained about the combina-

tion of effects of the oxime functionality at different positions

on the E1 skeleton. These results are shown in Table 3.

Comparison of 4 with E1 shows that oxime substitution at the

Inhibitors of 17â-Hydroxysteroid Dehydrogenase Type 1

Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4 1333

Table 3. Inhibition of 17â-HSD Type 1 and Type 2 by 6-oxo, 6-, 16- and 17-Oxime and Pyrazolone Derivatives of E1^a,b

^aMean of at least two measurements with typically a SD or spread of (5%. ^bnd ) not determined.

Table 4. Inhibition of 17â-HSD Type 1 and Type 2 by Derivatives of

of 6 with E1 shows that replacing the 17-carbonyl with an oxime

moiety is not tolerated. It is interesting therefore that a slightly

better inhibition is exhibited by compound 5 in which there is

an oxime moiety at the both 16 and 17 positions. Compound 7

also shows improved inhibition compared with 6, indicating that

having the oxime moiety at the 6 position must have a beneficial

effect on binding in the active site, sufficient to substantially

compensate for the detrimental effect of the 17-oxime. This

illustrates how a combination of substituents may lead to a

different more favored orientation. The methyl-oxime compound

8 does not show good inhibition of 17â-HSD1, which could be

due to steric effects of the methyl group clashing with residues

in the active site or perhaps due to removal of hydrogen bonding

interactions by replacement of the oxime hydrogen with a methyl

group. It can be concluded from these results that substitution

at the 6 position in combination with the 17 position could in

some cases lead to improved synergistic effects. One can also

deduce that disfavored interactions can to some degree be

alleviated by substitution at a different position on the E1

skeleton.

E1 and E2 Possessing a Small Substituent at the 16 Position^a,b

^aMean of at least two measurements with typically a SD or spread of

(5%. ^bnd ) not determined.

16 position is slightly detrimental to potency although 4 does

still show very high inhibition of 17â-HSD1 at 10 µM and low

inhibition of 17â-HSD2 at the same concentration. Comparison

Pyrazolone Derivative of E1. The pyrazolone compound 9

(Table 3) showed 86% inhibition of 17â-HSD1 at 10 µM and

1334 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4

Allan et al.

Table 5. Inhibition of 17â-HSD Type 1 and Type 2 by 16 Alkenyl

Derivatives of E1 and E2 and 16 â-alkyl Derivatives of E2^a-c

Table 6. Inhibition of 17â-HSD Type 1 and Type 2 by Carboxyl

Derivatives of E2^a,b

^aMean of at least two measurements with typically a SD or spread of

(5%. ^bnd ) not determined.

16-Substituted Derivatives of E1 and E2. Small Substit-

uents. The results for compounds 13-15 were encouraging

(Table 4), as all inhibited 17â-HSD1 to a moderate or higher

degree and all showed reasonable selectivity for Type 1 over

Type 2. Compound 15, in particular, showed a promising

inhibition of 17â-HSD1 of 92% at 10 µM with an IC₅₀value

of 0.32 µM. This emphasizes that there is potential for expansion

at the 16 position, with the caveat of substituent optimization,

and also suggests that â-substitution is well tolerated. This has

significance for compounds 18 since, due to steric factors, the

â-substituted products would be those expected from the

hydrogenation of compounds 17.

^aMean of at least two measurements with typically a SD or spread of

(5%. ^bnd ) not determined. ^cns ) not synthesized.

Alkenyl Substituents. Compounds 16 showed a wide range

of inhibitory activity as shown in Table 5. The methylene

derivative of E1,²⁵35 (16, R ) H, Table 5) showed 70%

inhibition of 17â-HSD1 at 10 µM, and 10% inhibition of Type

2 in our assay. The isopropyl analogue 16e showed similar

inhibition and the isobutyl version 16f gave an inhibition of

82% for Type 1 but was less selective, with 31% inhibition of

Type 2. Of the aryl substituents the only promising compounds

were 16d and 16k which showed 72 and 67% inhibition of Type

1 at 10 µM, respectively. Reduction of the 17-carbonyl to give

the E2 derivatives 17 yielded interesting results. The methylene-

E2 derivative 17a²⁵and the two other alkyl derivatives (17e

was selective for Type 1 over Type 2 (inhibition of 17â-HSD2

was 7% at 10 µM). However, these results were less promising

than for the corresponding E-ring pyrazole 27 that showed 97%

inhibition of 17â-HSD1 at 10 µM. Compound 9 can exist as

the different tautomeric forms shown in Figure 8, four of which

can be thought of as a hydroxyl-substituted pyrazole. As

mentioned previously, however, ¹³C NMR of 9 in deuterated

DMSO suggests that in polar solution a keto form predominates.

The inferior activity of 9 compared to 27 is probably due to

differences in the electronic properties of the E-ring system.

Inhibitors of 17â-Hydroxysteroid Dehydrogenase Type 1

Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4 1335

Table 7. Inhibition of 17â-HSD Type 1 and Type 2 by Methyl

Table 8. Inhibition of 17â-HSD Type 1 and Type 2 by Methyl

Carboxyl Derivatives of E1^a,b

Carboxyl Derivatives of E2^a

^aMean of at least two measurements with typically a SD or spread of

(5%.

position. Here the results were more promising, with inhibitions

of 84% and over for compounds 18a-c at 10 µM. These

findings suggest that some flexibility of the substituent may be

important for binding in the active site. The results also show

that bulky aryl substituents are tolerated although optimization

of these systems would be needed to obtain a potent inhibitor.

Carboxyl Derivatives of E2. The inhibitory activities of the

carboxylic acid derivatives of E2 are shown in Table 6. These

were tested as mixtures of diastereomers in ratios which in most

cases were undetermined as for the majority of the amides only

1

one set of peaks was observed by H NMR. Apart from the

ester and acid intermediates 37 and 19 which showed 63 and

39% inhibition of 17â-HSD1 at 10 µM respectively, the

carboxyl derivatives of E2 show good inhibition of 17â-HSD1

(70-91%). It can also be seen from the results that aromatic

amide moieties show better inhibition than the alkyl amides.

This indicates the potential for enhancing activity through

beneficial π-π interactions. The best compound in the series

proved to be the m-pyridyl amide 19e which has an IC₅₀of

510 nM. All of the compounds in this series showed good

selectivity for Type 1 over Type 2.

Methyl Carboxyl Derivatives of E1. The results for the

methyl carboxylic acid derivatives of E1, shown in Table 7,

were the most promising of all the series investigated to date.

For this series the ester intermediate 44 itself, tested as a 3:1

mixture of diastereomers, showed 97% inhibition of 17â-HSD1

at 10 µM, with an IC₅₀value of 300 nM. This activity was lost

upon hydrolysis to the acid 20, which was also tested as an

approximately 3:1 ratio of diastereomers. The majority of the

amides 20a-m, however, showed excellent inhibition of 17â-

HSD1 and, importantly, selectivity for Type 1 over Type 2. It

can be deduced from the comparison of results for 20a (which

shows good inhibition) with 20j and 20k that very flexible

^aMean of at least two measurements with typically a SD or spread of

(5%. ^bnd ) not determined.

and 17f) did not show much difference in activity from their

E1 analogues. For the aryl compounds, activities were improved

in all cases except for the two most active compounds from

series 16: compound 17d showed no improvement in inhibition

of Type 1 compared with 16d, but interestingly the selectivity

for Type 1 over Type 2 was lost; in fact 17d showed a high

81% inhibition of Type 2 at 10 µM, and compound 17k showed

decreased inhibition when compared with 16k and also showed

less selectivity. Other examples (17a, 17h and 17i) also exhibited

some drop in selectivity compared with their 16 series coun-

terparts. Generally, however, inhibition of Type 1 was improved

by reduction of the E1 derivatives 16 to the E2 derivatives 17.

As yet the irreversibility of the inhibition shown by compounds

16a-l and 17a-k has not been fully investigated.

16â-Alkyl Substituents. Compounds 18 were synthesized

to further investigate the potential of â-substitution at the 16

1336 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4

Allan et al.

moieties are less well tolerated in the active site. Also, from

comparison of the inhibition by 20m with 20i, one can deduce

that there is an optimum distance at which the pyridyl moiety

interacts (20m appears to have exceeded this). By far the best

result was for compound 20h which, as mentioned previously,

is the most potent inhibitor of 17â-HSD1 reported to date.²⁷

This compound was initially tested as a mixture of diastereo-

mers, along with the other members of the library, before the

major diastereomer (the 16â-substituted compound) was sepa-

rated and its biological activity evaluated. This single 16â-

substituted diastereomer was found to have the same potency

as the 2:1 mixture; as yet the 16R-substituted compound has

not been isolated for testing. The diastereomers of the other

less active amides were not separated.

Methyl Carboxyl Derivatives of E2. The percentage inhibi-

tion of 17â-HSD1 and 17â-HSD2 at 10 µM for the methyl

carboxylic acid derivatives of E2, compounds 21, are listed in

Table 8. These analogues showed reduced inhibitory activities

compared with compounds 20 and also showed less activity

than the carboxyl derivatives 19. The selectivity for inhibition

of 17â-HSD1 over 17â-HSD2 was however conserved.

Figure 11. Compound 1 (orange) docked into 1FDT in the presence

of NADP with substrate removed. E2 docked (not shown) overlays

with the steroid template of 1.

Comparison of results for series 19 and 21 may suggest that

the carboxyl linker in series 19 might be preferable to the methyl

carboxyl, since overall the activities of compounds 19 are higher

than those of 21. However, this conclusion can only be tentative

since most of series 21 contain only simple alkyl amides (ideally

a more comprehensive library 21 including greater variety of

amides might confirm the preferential binding of one series over

another). It can be concluded, however, that the linker/amide

combination is more successful for 19a than for 21a and that

compounds 19 appear to be more selective for 17â-HSD1 over

17â-HSD2. Comparison of activities for series 20 and 21 suggest

that E1 is preferable as a template for this particular methyl

carboxyl system than E2.

Figure 12. Compound 20h (orange) docked into 1FDT in the presence

of NADP with substrate removed.

Leu149, Val255, Phe226, and Phe259, the pyridyl nitrogen is

only 3.36 Å from the nearest phosphate oxygen of the cofactor.

In addition, the hydrogen atoms in the methyl linker of the side

chain are 1.71 and 3.33 Å from the Pro(S) hydrogen of the

cofactor, which could interfere with the transfer of the Pro(S)

hydrogen to the 17-carbonyl of 20h.

However, since inhibitor 20h, N-(pyridin-3-ylmethyl)estrone-

16-methylcarboxamide, showed vastly superior potency over

the other compounds it seems reasonable to pursue further

investigation into compounds which to a large degree mimic

the attributes of this compound in similar areas of space.

The novel 17â-HSD1 inhibitors described herein, along with

compounds previously published by our group, for which an

IC₅₀had been determined and which possess only defined

stereocenters were used to develop a quantitative structure-

activity relationship (QSAR). This was achieved using the three-

dimensional method of comparative molecular similarity indices

analysis (CoMSIA), part of the SYBYL suite, where a set of

aligned molecules is used in QSAR generation.⁴⁵These

molecules include the highly potent compound 20h (the

separated diastereomer which is â-substituted at the 16 position),

the equipotent 2-ethyl analogue,²⁷which has also been con-

firmed by X-ray crystallography as â-substituted at the 16

position, and a variety of pyrazole derivatives. The set of

molecules was aligned using FlexS^46,47and CoMSIA was then

used to assess steric, electrostatic, hydrophobic, and hydrogen

bond donor/acceptor fields. The two parameters which were

given the highest relative contributions by CoMSIA were

electrostatic and hydrogen bond acceptor fields: the aligned

molecules with these fields are shown in Figure 13. Areas of

favored and disfavored hydrogen bond acceptor interactions are

shown by green and yellow contours, respectively. The elec-

trostatic contours can be interpreted as the blue contour defining

a region where increased positive charge will result in increased

activity (or a more negative charge will result in decreased

activity); the red contour defines a region of space where

electron density is favorable.

Molecular Modeling. To investigate the binding of our

inhibitors in the active site of 17â-HSD1, the crystal structure

of 17â-HSD1 in complex with E2 and NADP (PDB code 1FDT)

was used.^13,23Compounds were docked into 1FDT with E2

removed using the docking program GOLD version 2.2.⁴⁴For

all compounds docked, the steroid backbone overlaid closely

with that of E2 as it is positioned in the crystal structure 1FDT.

When E2 is docked back into the active site after being removed

it docks back into the same position as it is in the crystal

structure. Shown in Figure 11 is inhibitor 1 docked into 1FDT

with a docking score of 50; the docked E2 is left out for clarity

as the steroid backbones of E2 and 1 overlay. The distance

between the 6-oxo carbonyl oxygen of 1 and the hydroxyl on

serine 222 is 2.21 Å; from the carbonyl oxygen to the tyrosine

phenolic oxygen is 2.86 Å. These distances suggest that

hydrogen bonding in this area is possible for 6-oxo compounds,

and it was hoped that the presence of a 6-oxo substituent might

enhance the binding of inhibitors. SAR results for the examples

synthesized, however, showed no improvement in inhibitory

activity of the 6-oxo over nonoxidized compounds.

Also shown (Figure 12) is the potent inhibitor 20h docked

into 1FDT. This docking gives an indication of what interactions

in the active site might confer superiority to 20h over the other

compounds synthesized. For example, in addition to the common

hydrophobic interactions of the steroid backbone with residues

Inhibitors of 17â-Hydroxysteroid Dehydrogenase Type 1

Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4 1337

tions in an important substrate binding region. Although the

presence of a 6-oxo functionality does not significantly enhance

inhibitory activity, and in some cases can be detrimental, the

presence of a carbonyl group at C6 can block a known⁴⁹point

for metabolism of the substrate, and these compounds are novel

and interesting inhibitors in their own right. A combination of

substitutions at different positions on the E1 scaffold can lead

to improved activity, as exemplified by the activity of com-

pounds 6 and 7 where an oxime moiety at the 6 position can

alleviate the highly detrimental effect of 17-oxime group.

Comparison of the inhibitory activities of compounds 5 and 6

show that the combination of an oxime moiety at the 16 and

17 positions is also more favorable than having the oxime at

the 17 position alone. Our group further investigated this 16/

17 combined substitution in the form of E-ring pyrazole steroids,

which have been reported previously, and has extended the SAR

in this area.²⁴Herein is reported the novel pyrazolone compound

9 which has been found not to show such high inhibition as the

E-ring pyrazole compound. This highlights the importance of

electronic effects in this area on binding in the active site.

In addition, a diverse array of new E1 and E2 derivatives

with substitution at the 16 position has been synthesized. From

their biological activities it can be concluded that compounds

containing a flexible linker to the 16 position give better

inhibition than those with a rigid alkene linker and, in the

majority of cases, substitution at the 16 position of E2 confers

selectivity for 17â-HSD1 over 17â-HSD2. This work has led

to the discovery of novel potent inhibitors of 17â-HSD1

including compounds 15 and 20a which show IC₅₀values of

320 and 380 nM, respectively. Also discovered was compound

20h which, with an IC₅₀of 37 nM, is the most potent inhibitor

of 17â-HSD1 reported to date.²⁷In addition a predictive QSAR

has been developed which will be modified to identify novel

nonsteroidal 17â-HSD1 inhibitors by in silico screening.

Figure 13. CoMSIA alignment showing hydrogen bond acceptor and

electrostatic fields: regions where a hydrogen bond acceptor is favorable

or unfavorable are shown in green and yellow respectively; blue

contours indicate regions where increased positive charge is favorable

and red contours show regions where electron density is favorable.

Figure 14. QSAR showing activity against predicted activity (units

are -log IC₅₀values). Triangles are the training set; circles are the

validation set.

Experimental Section

Chemistry. All chemicals were purchased from Aldrich Chemi-

cal Co. (Gillingham, UK) or Lancaster Synthesis (Morecambe,

U.K.). All organic solvents of A. R. grade were supplied by Fisher

Scientific (Loughborough, U.K.). E1 was purchased from Sequoia

Research Products (Oxford, UK). Reactions using anhydrous

solvents were carried out under nitrogen.

Using the IC₅₀values as continuous response data, the

statistical tool in QSAR for CoMSIA utilizes partial least squares

(PLS) for regression analysis.⁴⁸The results of the QSAR are

shown in Figure 14, where the -log of the observed IC₅₀is

plotted against the -log of the predicted IC₅₀. Compound 1

was included in the validation set; compounds 3, 15, 17c, 20h,

and 27 were included in the training set to predict activity. The

full list of compounds used with their IC₅₀values can be found

in Supporting Information. The graph in Figure 14 shows good

correlation of observed versus predicted activity with an r²value

of 0.94 and q²value of 0.86, indicating that the QSAR can be

used in a predictive fashion to calculate activities in silico. This

QSAR will be further refined as more data become available

from other series.

Thin-layer chromatography (TLC) was performed on precoated

plates (Merck TLC aluminum sheets silica gel 60 F₂₅₄). Product(s)

and starting material(s) were detected by either viewing under UV

light and/or treating with an ethanolic solution of phosphomolybdic

acid followed by heating. Flash column chromatography was

performed on silica gel (Sorbsil/Matrex C60) or using Argonaut

prepacked columns with a Flashmaster II. IR spectra were recorded

on a Perkin-Elmer Spectrum RXI FT-IR as KBr disks and peak

1

positions are expressed in cm^-1. H NMR (270 or 400 MHz) and

DEPT-edited ¹³C NMR (100.4 MHz) spectra were recorded with a

JEOL Delta 270 or a Varian Mercury VX 400 NMR spectrometer,

and chemical shifts are reported in parts per million (ppm, δ). HPLC

analyses were performed on a Waters Millenium 32 instrument

equipped with a Waters 996 PDA detector using either a Waters

Radialpack C₁₈reversed phase column (8 × 100 mm) eluting with

MeOH/H₂O at 2 mL/min, or a Symmetry C18 reverse phase column

(4.6 × 150 mm) eluting with MeCN/H₂O at 0.3 mL/min. FAB low

and high-resolution mass spectra were recorded at the Mass

Spectrometry Service Centre, University of Bath, using m-nitroben-

zyl alcohol (NBA) as the matrix. ES and APCI low resolution mass

spectra were obtained on a Waters Micromass ZQ. Elemental

analyses were performed by the Microanalysis Service, University

of Bath. Melting points were determined using a Reichert-Jung

Thermo Galen Kofler block and are uncorrected. X-ray crystal-

lographic studies of compounds 15 and 16c were carried out on a

kappa CCD diffractometer with area detector.

Conclusions

Novel potent inhibitors of 17â-HSD1 have been identified

by the modification of E1 at the 6, 16 and 17 positions. This

work has provided valuable information about the SAR around

the substituted D ring of the E1/E2 template for this emerging

and topical area of research. A number of conclusions can be

drawn from these results about the effects of substitution at

positions 6, 16 and 17 and combinations of these. Docking

studies indicate there is limited space for expansion at the 6

and 17 positions and results for the oxime compounds agree

with this hypothesis (the methyl-oxime compound 8 being a

lot less active than compound 7). This diminished activity could

also be a consequence of disrupted hydrogen bonding interac-

1338 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4

Allan et al.

Molecular Modeling. For the docking studies of compound 1

in 1FDT the starting conformations used for receptor docking were

generated from an energy minimization performed using the

MMFF94s force field, with MMFF94 charges applied, as imple-

mented in Sybyl 7.0. The resulting lowest energy conformer was

then used for docking studies using Gold version 2.2 with default

parameters. The active site was defined as a 12 Å radius around

the C alpha atom of Serine 142 and 30 attempts were computed

and scored using Gold Score.

To develop a QSAR using CoMSIA, the ligands were initially

minimized using the MMFF94s force field as implemented within

the Sybyl 7.0 package. The molecules were aligned using FlexS,

with a common core elucidated by DISTILL which corresponded

to the steroid backbone. Gastegier-Hu¨ckel charges were used for

the charge descriptors in FlexS and CoMSIA was performed using

the aligned compounds and the standard Sybyl 7.0 CoMSIA fields.

Biology. Radiolabeled E1 and E2 (³H and ¹⁴C) were purchased

from New England Nuclear (Boston, MA) or Amersham Bio-

sciences UK Limited (Amersham, U.K.).

mmol) in aq. AcOH (0.75 mL, 10%). The resulting dark brown

solution was stirred for 40 h at room temperature. The solvent was

then removed under reduced pressure and ice/water (50 mL) was

added. The organics were extracted with EtOAc (2 × 50 mL),

washed with H₂O (2 × 30 mL), then brine (2 × 30 mL), dried

(Na₂SO₄), filtered, and concentrated under reduced pressure to give

a brown crude product. This was purified by flash chromatography

with CHCl₃/EtOAc (8:2) as eluent to give the title compound as a

pale yellow solid (13 mg, 10%): mp 96-101 °C; TLC (CHCl₃/

1

EtOAc, 4:1) R_f0.40 cf. R_f0.63; H NMR (400 MHz, CDCl₃) δ

0.95 (s, 3H, H-18), 1.24-2.92 (m, 11H), 2.26 (s, 3H, OAc), 2.32

(s, 3H, OAc), 7.29 (dd, J ) 8.4, 2.6 Hz, 1H, H-2), 7.46 (d, J ) 8.4

Hz, 1H, H-1), 7.77 (d, J ) 2.6 Hz, 1H, H-4), 8.17 (dd, J ) 2.9, 1.7

Hz, 1H, dCH); LRMS (FAB+) m/z 663.5 [48], 397.2 [38, (M +

H)⁺], 73.0 [100]; HRMS (FAB+) m/z calcd. for C₂₃H₂₅O₆(M +

H)⁺397.1651, found 397.1670.

6-Oxo-16-formyl-estrone (2). To a stirred solution of 6-oxo-3-

O-acetyl-16-acetoxymethylene-estrone 26 (30 mg, 0.08 mmol) in

MeOH (10 mL) was added dropwise aq. K₂CO₃(63 mg, 0.46 mmol

in 1 mL). The resulting pale yellow solution was stirred at room

temperature for 45 min before the final pale brown mixture was

acidified with 3 drops of 5 M HCl and concentrated in vacuo. H₂O

was added (30 mL) and the organics were extracted with EtOAc

(2 × 50 mL). The combined organic layers were washed with H₂O

(2 × 30 mL) then brine (2 × 30 mL), dried (Na₂SO₄), and

concentrated in vacuo. The resulting beige crude product was

recrystallized from EtOAc/hexane to give the title compound as

white powder (13 mg, 54%): mp 224-227 °C; ¹H NMR (400 MHz,

DMSO-d₆) 0.80 (s, 3H, H-18), 1.41-2.65 (m, 11H), 7.01 (dd, J )

8.4, 2.7 Hz, 1H, H-2), 7.28 (d, J ) 2.7 Hz, 1H, H-4), 7.31(d, J )

8.4 Hz, 1H, H-1), 7.41 (1H, s, 1H, C-1′-H), 9.64 (1H, s, exchanged

with D₂O, C-3-OH) and 10.77 (1H, br s, exchanged with D₂O, C-1′-

OH); LRMS (FAB-) m/z 311.2 [100, (M - H)^-]; HRMS (FAB+)

m/z calcd. for C₁₉H₂₁O₄(M + H)⁺313.1440, found 313.1450.

Acetic Acid 8-Acetyl-6a-methyl-4b,5,6,6a,8,10,10a,10b,11,12-

decahydro-7,8-diaza-pentaleno[2,1-a]phenanthene-2-yl Ester (28).

To a solution of 27 (0.598 g, 2.0 mmol) in dry pyridine (8 mL)

was added acetic anhydride (2 mL), and the mixture was refluxed

under nitrogen for 16 h. The mixture was then cooled and poured

onto ice, and the resulting cream colored precipitate was collected

by filtration and purified by column chromatography using EtOAc/

hexane (1/1) to give the title compound as white powder (0.303 g,

T47-D and MDA-MB-231 cells have previously been shown to

possess predominantly reductive or oxidative 17B-HSD activity,

respectively.⁵⁰

Measurement of Inhibition of 17â-HSD Type 1. T47-D human

3

breast cancer cells were incubated with H-E1 at a concentration

of 2 nM per well, in a 24-well tissue culture plate, in the absence

or presence of the inhibitor (0.1 nM to 10 µM). After incubation

of the substrate ( inhibitor for 30 min at 37 °C, the products were

isolated from the mixture by extraction with Et₂O (4 mL), using

¹⁴C-E2 (5000 dpm) to monitor procedural losses. Separation of ³H-

E2 from the mixture was achieved using TLC (DCM/EtOAc, 4:1

3

v/v) and the mass of H-E2 produced was calculated from the H

counts detected and recovery of ¹⁴C-E2.

Measurement of Inhibition of 17â-HSD Type 2. MDA-MB-

3

231 human breast cancer cells were incubated with H-E2 at a

concentration of 2 nM per T25 flask, in the absence or presence of

the inhibitor (0.1 nM to 10 µM). After incubation of the substrate

( inhibitor for 3 h at 37 °C, the products were isolated from the

mixture by extraction with Et₂O (4 mL), using ¹⁴C-E1 (5000 dpm)

to monitor procedural losses. Separation of ³H-E1 from the mixture

was achieved using TLC (DCM/EtOAc, 4:1 v/v) and the mass of

3

³H-E1 produced was calculated from the H counts detected and

recovery of ¹⁴C-E1.

1

Compounds 4,³²5,²⁴6,^33,5110,²⁴21,³⁹22,⁵²24,³⁰27,⁵³30,³⁴42,³⁹

and 45³⁹have been described in the literature. Experimental details

and data for compounds 23 and 1 are reported in Supporting

Information as are analytical and spectroscopic data for compounds

16b-d, 16f-l, 17b-d, 17h, 17i, 18b-d, 19b, 19c, 19e-j, 20b-

m, and 21b-f.

40%): TLC (EtOAc/hexane, 1/1) R_f0.48; H NMR (270 MHz,

CDCl₃) δ 1.07 (s, 3H, H-18), 1.40-1.60 (m, 1H), 1.62-2.10 (m,

5H), 2.28 (s, 3H, OCOCH₃), 2.28-2.48 (m, 4H), 2.64 (s, 3H,

NCOCH₃), 2.64-2.72 (m, 1H), 2.87-2.96 (m, 2H), 6.82 (d, J )

2.2 Hz, 1H, H-4), 6.86 (dd, J ) 8.4, 2.5 Hz, 1H, H-2), 7.30 (d, J

) 8.4 Hz, 1H, H-1), 7.86 (s, 1H); LCMS (ES+) 380.23 (M +

2H)⁺.

3-O-Acetyl-16-acetoxymethylene-estrone (25). To a stirred

solution of 16-formyl-estrone 24 (230 mg, 0.77 mmol) in anhydrous

pyridine (15 mL) at 0 °C was added dropwise over 10 min acetic

anhydride (8.34 mL, 88.4 mmol). The resulting yellow mixture was

then heated to reflux for 1 h. The final brown solution was cooled

then poured into ice/water (50 mL) and acidified with 5 M HCl.

The organics were extracted with EtOAc (2 × 50 mL), washed

with H₂O (30 mL), Na₂CO₃10% (30 mL), then brine (30 mL),

dried (Na₂SO₄), filtered, and concentrated in vacuo to give orange

foam (250 mg). This was purified by flash chromatography using

CHCl₃/EtOAc (95:5) as eluent to give the title compound as a pale

yellow oil that crystallized on standing (140 mg, 47%): mp 115-

Acetic Acid 8-Acetyl-6a-methyl-12-oxo-4b,5,6,6a,8,10,10a,-

10b,11,12-decahydro-7,8-diaza-pentaleno-[2,1-a]phenanthren-

2-yl Ester (29). To a solution of 28 (0.121 g, 0.32 mmol) in MeCN

(2.5 mL) was added tert-butyl hydroperoxide (0.14 mL of a 70wt

% solution in water, 1 mmol) and chromium hexacarbonyl (0.021

g, 0.09 mmol). The solution was heated to reflux for 24 h before

being cooled to room temperature. Water (10 mL) was added, and

the products were extracted with diethyl ether (3 × 10 mL). The

ether extracts were combined, washed with H₂O, saturated aqueous

NaHCO₃and brine before being dried (MgSO₄) and concentrated

in vacuo. The title compound was isolated by flash column

chromatography using ethyl acetate/hexane (1/1) as eluent (29 mg,

1

117 °C; TLC (CHCl₃/EtOAc, 9:1) R_f0.70 cf. R_f0.30; H NMR

1

23%): TLC (EtOAc/hexane 1/1) R_f) 0.4; H NMR (400 MHz,

(400 MHz, CDCl₃) δ 0.93 (s, 3H, H-18), 1.42-2.84 (m, 11H), 2.24

(s, 3H, OAc), 2.29 (s, 3H, OAc), 2.89-2.96 (m, 2H, H-6), 6.81 (d,

J ) 2.6 Hz, 1H, H-4), 6.86 (dd, J ) 8.3, 2.6 Hz, 1H, H-2), 7.29 (d,

J ) 8.3 Hz, 1H, H-1), 8.15 (dd, J ) 2.9, 1.7 Hz, 1H, dCH); LRMS

(FAB+) m/z 383.0 [100, (M + H)⁺]; HRMS (FAB+) m/z calcd.

for C₂₃H₂₇O₅(M + H)⁺383.1858, found 383.1852.

6-Oxo-3-O-acetyl-16-acetoxymethylene-estrone (26). To a

stirred solution of 3-O-acetyl-16-acetoxymethylene-estrone 25 (120

mg, 0.31 mmol) in AcOH at 10-15 °C in an ice/water bath was

added dropwise over 30 min a solution of CrO₃(132 mg, 1.32

CDCl₃) δ 1.09 (s, 3H, 18-H), 1.8-2.0 (m, 2H), 2.2-2.5 (m, 4H),

2.33 (s, 3H), 2.5-2.9 (m, 5H), 2.65 (s, 3H), 7.34 (dd, J ) 8.4, 2.5

Hz, 1H, H-2), 7.65 (d, J ) 8.6 Hz, 1H, H-1), 7.79 (d, J ) 2.3 Hz,

1H, H-4), 7.88 (s, 1H).

2-Hydroxy-6a-methyl-5,6,6a,8,10,10a,10b,11-octahydro-4bH-

7,8-diaza-pentaleno[2,1-a]phenanthren-12-one (3). To a solution

of 29 (0.014 g, 0.04 mmol) in EtOH (1 mL) was added a solution

of KOH (0.004 g) in ethanol/water (2 mL of 1/1), and the mixture

was shaken then left at room temperature for 30 min. TLC at this

Inhibitors of 17â-Hydroxysteroid Dehydrogenase Type 1

Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4 1339

stage showed no starting material remaining; therefore, the mixture

was acidified with glacial AcOH, concentrated in vacuo and H₂O

was added. The solution was left standing until a pale orange

precipitate formed. This powder was collected by filtration and

washed with water to yield the title compound (6 mg, 46%): ¹H

NMR (400 MHz, DMSO-d₆) δ 0.91 (s, 3H, 18-H), 1.60-1.82 (m,

2H), 2.1-2.35 (m, 5H), 2.4-2.7 (m, 5H), 7.02 (dd, J ) 8.6, 2.7

Hz, 1H, H-2), 7.29 (d, J ) 2.7 Hz, 1H, H-4), 7.34 (d, J ) 8.6 Hz,

1H, H-1), 9.65 (s, 1H), 12.04 (bs, 1H); LCMS (ES+) 309.99 (M

+ 2H)⁺; HRMS (FAB+) calcd. for C₁₉H₂₁N₂O₂(M + H)⁺

309.16030, found 309.16132.

¹H NMR (400 MHz, CDCl₃) δ 0.92 (s, 3H, H-18), 1.24-2.58 (m,

12H), 3.12 (dd, J ) 18.1, 4.5 Hz, 1H, H-7), 3.84 (s, 3H, OMe),

3.99 (s, 3H, OMe), 4.85 (br s, exchanged with D₂O, 1H, OH), 6.84

(dd, J ) 8.6, 2.7 Hz, 1H, H-2), 7.20 (d, J ) 8.6 Hz, 1H, H-1), 7.29

(d, J ) 2.7 Hz, 1H, H-4); LRMS (FAB+) m/z 343.2 [100, (M +

H)⁺]; HRMS (FAB+) m/z calcd. for C₂₀H₂₇N₂O₃(M + H)⁺

343.2022, found 343.2033. Anal. (C₂₀H₂₆N₂O₃) C, H, N.

3-O-Benzyl-estrone-16-carboxylic Acid Methyl Ester (31). To

a stirred suspension of NaH (3.48 g of a 60% dispersion, 87 mmol)

in anhydrous THF (50 mL) was added dimethyl carbonate (6.11

mL, 72 mmol), and the mixture was heated to reflux. To this was

added dropwise a solution of 3-O-benzylestrone 30 (10.45 g, 29

mmol) in THF (50 mL), and the reaction was refluxed for 8 h before

being cooled to 0 °C. The mixture was acidified with 3 M AcOH

(50 mL) before being poured into brine. The product was extracted

with CHCl₃and the solution dried over Na₂SO₄before concentration

in vacuo. Recrystallization of the solid residue from MeOH gave

the title compound as beige solid (9.8 g, 81%): ¹H NMR (400

MHz, CDCl₃) δ 0.97 (s, 1.05H, CH₃), 0.99 (s, 1.95H, CH₃), 1.34-

2.50 (m, 11H), 2.91-2.93 (m, 2H, H-6), 3.22 (dd, J ) 10.1, 8.6

Hz, 0.65H, H-16), 3.58 (dd, J ) 9.4, 1.7 Hz, 0.35H, H-16), 3.77

(s, 3H, OCH₃), 5.03 (s, 2H, CH₂Ph), 6.73 (d, J ) 2.5 Hz, 1H, H-4),

6.80 (dd, J ) 8.6, 2.5 Hz, 1H, H-2), 7.20 (d, 1H, J ) 8.6 Hz, 1H,

H-1), 7.30-7.50 (m, 5H); LRMS (FAB+) m/z 418.20 (M⁺).

Estrone-16-carboxylic Acid Methyl Ester (32). To a solution

of 31 (2.5 g, 5.97 mmol) in THF (30 mL) was added Pd/C (10wt

%, catalytic), and the reaction mixture was stirred at room

temperature under H₂(balloon) for 8 h. The mixture was filtered

through Celite, and the filtrate was evaporated in vacuo. Recrys-

tallization of the solid residue from MeOH gave the title compound

as white solid (1.82 g, 93%): mp 215-217 °C; ¹H NMR (400 MHz,

DMSO-d₆) δ 0.87 (s, 1.95H, CH₃), 0.91 (s, 1.05H, CH₃), 1.20-

2.40 (m, 11H), 2.73-2.76 (m, 2H, H-6), 3.39 (dd, J ) 9.8, 8.6 Hz,

0.65H, H-16), 3.65 (s, 1.05H, OCH₃), 3.65 (s, 1.95H, OCH₃), 3.80

(dd, J ) 9.8, 8.6 Hz, 0.35H, H-16), 6.45 (d, J ) 2.3 Hz, 1H, H-4),

6.05 (dd, J ) 8.6, 2.7, 1H, H-2), 7.04 (d, J ) 8.2 Hz, 1H, H-1),

9.04 (s, 1H, OH); HRMS (FAB+) m/z calcd. for C₂₀H₂₄O₄(M⁺)

328.1675, found 328.1684; HPLC > 97% (t_R) 2.75, 10:90 H₂O:

MeOH). Anal. (C₂₀H₂₄O₄) C, H.

3-Hydroxy-13-methyl-6,7,8,9,11,12,13,14,15,16-decahydro-cy-

clopenta[a]phenanthren-17-one Oxime (6). To a suspension of

E1 (500 mg, 1.85 mmol) in a mixture of MeOH/H₂O (5:1, 90 mL)

was added NaOAc (1.50 g, 18.80 mmol) followed by hydroxy-

lamine hydrochloride (1.40 g, 20.72 mmol). The resulting suspen-

sion was stirred at room temperature overnight. The solvent was

then removed under reduced pressure and H₂O (100 mL) added.

The organics were extracted with EtOAc (100 mL + 50 mL),

washed with H₂O (50 mL), then brine (50 mL), dried (Na₂SO₄),

filtered, and concentrated in vacuo to give a white crystalline crude

product (627 mg). This was recrystallized from MeOH to give the

title compound as white crystals (426 mg, 81%): mp 251-253 °C

[lit. 248-250 °C];⁵¹TLC (CHCl₃/EtOAc, 4:1) R_f0.13 cf. R_f0.55

(E1); IR (KBr) 3415 (NOH), 3270 (OH), 2930 (aliph CH), 1620

1

(CdN or arom CdC), 1585-1460 (arom CdC) cm^-1; H NMR

(400 MHz, DMSO-d₆) δ 0.85 (s, 3H, H-18), 1.32-2.41 (m, 13H),

2.65-2.80 (m, 2H, H-6), 6.44 (d, J ) 2.6 Hz, 1H, H-4), 6.50 (dd,

J ) 8.5, 2.6 Hz, 1H, H-2), 7.05 (d, J ) 8.5 Hz, 1H, H-1), 9.01 (s,

exchanged with D₂O, 1H, OH) and 10.10 (s, exchanged with D₂O,

1H, NOH); LRMS (FAB+) m/z 286.1 (100, [M + H]⁺); HRMS

(FAB+) m/z calcd. for C₁₈H₂₄NO₂(M + H)⁺286.1807, found

286.1809.

3-Hydroxy-13-methyl-8,9,11,12,13,14,15,16-octahydro-7H-cy-

clopenta[a]phenanthrene-6,17-dione Dioxime (7). To a solution

of 1 (100 mg, 0.35 mmol) in a mixture of MeOH/H₂O (5:1, 18

mL) was added NaOAc (293 mg, 3.58 mmol) followed by

hydroxylamine hydrochloride (274 mg, 3.94 mmol). The resulting

solution was stirred at room temperature overnight before the

solvent was removed under reduced pressure and H₂O (50 mL)

was added. The organics were extracted with EtOAc (50 mL + 20

mL) and the combined organic layers washed with H₂O (2 × 20

mL), then brine (2 × 20 mL), dried (Na₂SO₄), filtered, and

concentrated in vacuo to give a light brown crude product (120

mg). This was purified by flash chromatography using a gradient

of CHCl₃/EtOAc (7:3 to 1:1) then CHCl₃/EtOAc/acetone (2:2:1 to

1:1:2) as eluent to give the title compound as cream colored powder

(46 mg, 41%): mp 341-344 °C; TLC (CHCl₃/EtOAc, 7:3) R_f0.11

cf. R_f0.44; IR (KBr) 3410, 3265-3050 (br, NOH, OH), 2930-

2-Hydroxy-6a-methyl-5,6,6a,7,8,10,10a,10b,11,12-decahydro-

4bH-7,8-diaza-pentaleno[2,1-a]phenanthren-9-one (9). To a sus-

pension of 32 (0.657 g, 2 mmol) in toluene (10 mL) was added

hydrazine monohydrate (0.11 mL, 2.2 mmol), and the mixture was

heated at 150 °C in a sealed tube for 4 h. During this time the

starting material dissolved then another white precipitate formed.

The reaction was then cooled to room temperature and acidified

with glacial AcOH, and H₂O was added to ensure that all product

precipitated. The white powder was collected by filtration and

washed with H₂O, EtOH (minimum), diethyl ether, and hexane.

LCMS showed two compounds with masses corresponding to the

desired product and the hydrazone derivative of E1. This mixture

was suspended in EtOH, and aq. NaOH was added slowly until

the solid dissolved. The solution was then acidified to pH 3 with

glacial AcOH, and a small amount of H₂O was added until a fine

white powder precipitated (0.112 g, 18%): mp > 275 °C (dec); IR

2850 (aliph CH), 1705 (CdN), 1580-1495 (arom CdC) cm^-1

;

¹H NMR (400 MHz, DMSO-d₆) δ 0.83 (s, 3H, H-18), 1.14-2.42

(m, 12H), 3.06 (m, 1H, H-7), 6.73 (dd, J ) 8.3, 2.5 Hz, 1H, H-2),

7.14 (d, J ) 8.3 Hz, 1H, H-1), 7.29 (d, J ) 2.8 Hz, 1H, H-4) and

9.28, 10.15, 11.08 (each 1H, s, 2×NOH, OH); LRMS (FAB+) m/z

315.1 [82, (M + H)⁺], 73.0 [100]; HRMS (FAB+) m/z calcd. for

C₁₈H₂₃N₂O₃(M + H)⁺315.1709, found 315.1715.

1

(KBr) 3164 (OH), 2910 (NH), 1599 (CdO) cm^-1; H NMR (400

MHz, DMSO-d₆) δ 0.89 (s, 3H, H-18), 1.30-1.70 (m, 4H), 1.80-

1.87 (m, 1H), 1.93-2.11 (m, 3H), 2.17-2.40 (m, 3H), 2.71-2.78

(m, 2H), 6.43 (d, J ) 2.2 Hz, 1H), 6.50 (dd, J ) 8.1, 2.6 Hz, 1H),

7.03 (d, J ) 8.4 Hz, 1H), 9.01 (s, 1H), 10.3 (bs, 1H); ¹³C NMR

(100 MHz, DMSO-d₆) δ 18.2 (CH₃), 23.3 (CH₂), 25.9 (CH₂), 27.0-

(CH₂), 29.0(CH₂), 33.7(CH₂), 37.2(CH), 41.1(C), 43.9(CH), 61.0-

(CH), 102.4(C), 112.8(CH), 115.0(CH), 125.9(CH), 130.4(C),

137.1(C), 155.0(C), 158.7(C), 170.8(C); LCMS (APCI+) m/z

311.29 (M + H)⁺; HRMS (FAB+) m/z calcd. for C₁₉H₂₃N₂O₂(M

+ H)⁺311.1759, found 311.1761.

16â-Cyano-estradiol (13). To a stirred solution of 16-cyano-

estrone 10 (150 mg, 0.51 mmol) in a mixture of THF/EtOH (3:2,

5 mL) at room temperature was added portionwise NaBH₄(50 mg,

1.32 mmol). The resulting pale yellow solution was stirred for 30

min before the solvent was removed under reduced pressure and

3-Hydroxy-13-methyl-8,9,11,12,13,14,15,16-octahydro-7H-cy-

clopenta[a]phenanthrene-6,17-dione Bis-(O-methyl-oxime) (8).

To a solution of 1 (50 mg, 0.18 mmol) in a mixture of MeOH/H₂O

(5:1, 9 mL) was added NaOAc (146 mg, 1.79 mmol) followed by

O-methyl-hydroxylamine hydrochloride (164 mg, 1.97 mmol). The

resulting solution was stirred at room temperature overnight before

the solvent was removed under reduced pressure and H₂O (30 mL)

was added. The organics were extracted with EtOAc (20 mL + 10

mL), and the combined organic layers were washed with H₂O (2

× 10 mL) then brine (2 × 10 mL), dried (Na₂SO₄), filtered, and

concentrated in vacuo. The white crystalline crude product was

recrystallized from EtOAc/hexane to give the title compound as

white crystals (55 mg, 92%): mp 206-208 °C; TLC (CHCl₃/

EtOAc, 8:2) R_f0.70 cf. R_f0.28; IR (KBr) 3134 (br, OH), 2995

(arom CH), 2935-2890 (aliph CH), 1570-1490 (arom CdC) cm^-1

;

1340 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4

Allan et al.

H₂O (20 mL) was added. The organics were extracted with EtOAc

(20 mL + 10 mL), and the combined organic layers washed with

H₂O (20 mL) then brine (20 mL), dried (Na₂SO₄), filtered, and

concentrated in vacuo to give a light yellow crude product (159

mg). This was recrystallized from isopropyl alcohol/H₂O to give

white crystals (16 mg) and a further crop of the product (40 mg)

was obtained from the residue of the mother liquor upon recrys-

tallization from EtOH (overall yield 37%): mp 252-254 °C; IR

(KBr) 3430-3235 (OH), 2925-2850 (aliph CH), 2255 (CN),

1610-1500 (arom CdC) cm^-1; ¹H NMR (400 MHz, DMSO-d₆) δ

0.79 (s, 3H, H-18), 1.10-2.27 (m, 11H), 2.66-2.76 (m, 2H, H-6),

∼3.33 (m, ∼1H, H-16, under solvent peaks), 3.68 (dd, J ) 9.4,

4.7 Hz, 1H, H-17), 5.50 (d, J ) 4.7 Hz, 1H, C-17-OH), 6.43 (d, J

) 2.5 Hz, 1H, H-4), 6.50 (dd, J ) 8.5, 2.5 Hz, 1H, H-2), 7.03 (d,

J ) 8.5 Hz, 1H, H-1) and 9.01 (s, exchanged with D₂O, 1H, C-3-

OH); LRMS (FAB+) m/z 297.1 [100, (M⁺)]; HRMS (FAB+) m/z

calcd. for C₁₉H₂₃NO₂(M⁺) 297.1729, found 297.1724.

3-O-Benzyl-16â-hydroxymethyl-estradiol (33). To a cold (0

°C), stirred solution of 31 (5 g, 12 mmol) in THF/MeOH (90:10)

was added NaBH₄(452 mg, 12 mmol), and the reaction mixture

was stirred for 18 h at 0-20 °C, acidified with 2 N HCl, and

extracted with ether (2×). The combined organic layers were

washed with brine, dried (Na₂SO₄), and concentrated in vacuo. The

solid was purified by column chromatography using CHCl₃/EtOAc

(90:10) as eluent to separate 36 (2.50 g, 57%) and the title

compound 33, 0.63 g, 13%: ¹H NMR (270 MHz, DMSO-d₆) δ

0.68 (s, 3H, H-18), 0.95-1.45 (m, 4H), 1.70-1.90 (m, 4H), 2.00-

2.35 (m, 4H), 2.74-2.76 (m, 2H, H-6), 3.29 (m, 1H), 3.67 (m,

2H), 4.13 (dd, J ) 5.9, 4.4 Hz, 1H, H-17), 4.60 (d, J ) 4.4 Hz,

1H, OH), 5.03 (s, 2H, CH₂Ph), 6.69 (d, J ) 2.7 Hz, 1H, H-4), 6.74

(dd, J ) 8.4, 2.7 Hz, 1H, H-2), 7.14 (d, J ) 8.6 Hz, 1H, H-1),

7.25-7.46 (m, 5H, Ph); HRMS (FAB+) m/z calcd. for C₂₆H₃₂0₃

(M⁺) 392.2351, found 392.2342.

16â-Hydroxymethyl-estradiol (14). Prepared by the same

method as 32. Yield ) 78%, white solid: mp > 275 °C (dec); ¹H

NMR (400 MHz, DMSO-d₆) δ 0.72 (s, 3H, H-18), 1.02-1.13 (m,

2H), 1.19-1.42 (m, 4H), 1.78-1.89 (m, 3H), 2.10 (td, J ) 10.0,

3.3 Hz, 1H), 2.23-2.27 (m, 2H), 2.70-2.77 (m, 2H, H-6), 3.30-

3.36 (m, 1H), 3.65-3.72 (m, 2H), 4.18 (dd, J ) 6.2, 4.8 Hz, 1H,

H-17), 4.64 (d, J ) 4.8 Hz, 1H, OH), 6.45 (d, J ) 2.6 Hz, 1H,

H-4), 6.52 (dd, J ) 8.4, 2.6 Hz, 1H, H-2), 7.06 (d, J ) 8.4 Hz, 1H,

H-1), 9.02 (s, 1H, phenol); LRMS (FAB+) m/z 302.1 [100, (M⁺)];

HPLC > 99% (t_R) 1.74, 100% MeOH); Anal. (C₁₉H₂₆O₃) C, H.

3-O-Acetyl-16-methylene-estrone (34). To a stirred solution of

22 (1.00 g, 3.21 mmol) in anhydrous isoamyl alcohol (8 mL) were

added paraformaldehyde (480 mg, 16.0 mmol for one unit) and

dimethylamine hydrochloride (1.6 g, 19.6 mmol). The resulting

mixture was heated to reflux for 24 h. The resulting light yellow

solution was poured into H₂O (30 mL) and acidified with 5 M HCl.

The organics were extracted with EtOAc (2 × 50 mL), and the

combined organic layers washed with saturated aqueous NaHCO₃

(20 mL), H₂O (20 mL), then brine (20 mL), dried (Na₂SO₄), and

concentrated in vacuo to give a light yellow oil. The isoamyl alcohol

was distilled off and the oily residue crystallized overnight.

Purification by flash chromatography using CHCl₃/EtOAc (95:5)

as eluent gave the title compound as white solid (527 mg, 51%).

For analysis some crystals were obtained by adding hexane to the

oily crude residue and allowing to stand for 1 h: ¹H NMR (400

MHz, CDCl₃) δ 0.93 (s, 3H, H-18), 1.41-2.72 (m, 11H), 2.29 (s,

3H, OAc), 2.90-2.95 (m, 2H, H-6), 5.41-5.43 (m, 1H, H-1′trans),

6.09-6.12 (m, 1H, H-1′cis), 6.81 (d, J ) 2.5 Hz, 1H, H-4), 6.86

(dd, J ) 8.3, 2.5 Hz, 1H, H-2), 7.29 (d, J ) 8.3 Hz, 1H, H-1);

LRMS (FAB+) m/z 325.2 [100, (M + H)⁺]; HRMS (FAB+) m/z

calcd. for C₂₁H₂₅O₃(M + H)⁺325.1804, found 325.1822.

dried under high vacuum. This was purified by flash chromatog-

raphy with CHCl₃/EtOAc (95:5) as eluent to give the title compound

as pale yellow powder (26 mg, 13%). This was recrystallized from

EtOH/H₂O to give pale yellow crystals (13 mg, 6%): mp 208-

1

210 °C; TLC (CHCl₃/EtOAc, 8:2) R_f0.40 cf. R_f0.86; H NMR

(400 MHz, CDCl₃) δ 0.89 (3H, s, H-18), 1.16 (t, J ) 7.0 Hz, 3H,

H-3′), 1.36-2.42 (m, 12H), 2.81-2.85 (m, 2H, H-6), 3.42-3.50

(m, 2H, H-2′), 3.59 (m, 2H, H-1′), 4.66 (s, 1H, OH), 6.58 (d, J )

2.8 Hz, 1H, H-4), 6.64 (dd, J ) 8.1, 2.8 Hz, 1H, H-2) and 7.15

(1H, d, J ) 8.1 Hz, H-1); LRMS (FAB+) m/z 329.1 [100, (M +

H)⁺]; HRMS (FAB+) m/z calcd. for C₂₁H₂₉O₃(M + H)⁺329.2117,

found 329.2124. Anal. (C₂₁H₂₈O₃) C, H.

16-(Pyridin-3-yl)methylene-estrone (16a). To a stirred suspen-

sion of E1 (1.35 g, 5.0 mmol) and pyridine-3-carbaldehyde (595

mg, 5.0 mmol) in EtOH (40 mL) at room temperature was added

NaOH (1.0 g, 25 mmol). The resulting dark orange solution was

stirred at room temperature for 4 h before glacial AcOH (ca. 10

mL) was added with stirring. The color changed to light yellow,

and a light yellow solid precipitated. The solid was collected by

filtration and washed with H₂O (50 mL), EtOH (20 mL), diethyl

ether (50 mL), and hexane (50 mL) and dried under high vacuum

to give the title compound as yellow powder (1.63 g, 90%): mp >

280 °C (dec); ¹H NMR (400 MHz, DMSO-d₆) δ 0.90 (s, 3H, H-18),

1.26-1.66 (m, 5H), 1.82-1.90 (m, 1H), 1.95-2.02 (m, 1H), 2.14-

2.22 (m, 1H), 2.30-2.38 (m, 1H), 2.57-2.66 (m, 1H), 2.70-2.81

(m, 2H), 2.83-2.92 (m, 1H), 6.45 (d, J ) 2.3 Hz, 1H, H-4), 6.51

(dd, J ) 8.2, 2.3 Hz, 1H, H-2), 7.04 (d, J ) 8.2 Hz, 1H, H-1), 7.34

(s, 1H), 7.47 (dd, J ) 8.2, 5.1 Hz, 1H), 8.06 (d, J ) 8.2 Hz, 1H),

8.56 (dd, J ) 4.7, 1.2 Hz, 1H), 8.82 (d, J ) 1.6 Hz, 1H), 9.03 (s,

1H, -OH); LRMS (FAB+) m/z 360.2 [100, (M + H)⁺]; LRMS

(FAB-) m/z 276.1 [100], 358.2 [90, (M - H)^-]; HRMS (FAB+)

m/z calcd. for C₂₄H₂₆NO₂(M + H)⁺360.1963, found 360.1964.

16-Isobutylidene-estrone (16e). A solution of E1 (420 mg, 1.55

mmol) in dry THF (5 mL) was added dropwise to a stirred solution

of LDA (2.47 mL of a 1.8 M solution in heptane/THF/ethyl

benzene, 4.44 mmol) in dry THF (2 mL) at -78 °C, under an

atmosphere of N₂. After stirring for 2 h at -78 °C, isobutyraldehyde

(185 µL, 2.03 mmol, freshly distilled from Na₂SO₄) was added.

The resulting mixture was allowed to warm to room temperature

with stirring overnight. The solvent was then removed under

reduced pressure and H₂O (50 mL) added. The organics were

extracted with EtOAc (50 mL + 20 mL), and the combined organic

layers washed with H₂O (20 mL) then brine (20 mL), dried (Na₂-

SO₄), and concentrated in vacuo to give white foam. This was

purified by flash chromatography using CHCl₃/EtOAc (9:1) as

eluent to give the title compound as white foam (401 mg, 79%).

This was recrystallized from EtOAc/hexane to give white crystals

(325 mg, 64%): mp 188-190 °C; TLC (chloroform/EtOAc, 8:2)

R_f0.65 cf. R_f0.74 (E1); IR (KBr) 3370 (OH), 2930-2890 (aliph

CH), 1710 (CdO), 1645-1445 (arom CdC and exocyclic CdC)

1

cm^-1; H NMR (400 MHz, CDCl₃) δ 0.92 (s, 3H, H-18), 1.06 (d,

J ) 6.6 Hz, 3H, H-3′), 1.04 (d, J ) 6.6 Hz, 3H, H-4′), 1.40-2.70

(m, 12H), 2.84-2.90 (m, 2H, H-6), 4.67 (s, exchanged with D₂O,

1H, OH), 6.46 (dd, J ) 9.8, 1.9 Hz, 1H, H-1′), 6.59 (d, J ) 2.7

Hz, 1H, H-4), 6.64 (dd, J ) 8.3, 2.7 Hz, 1H, H-2), 7.16 (d, J ) 8.3

Hz, 1H, H-1); ¹³C NMR (100 MHz, CDCl₃) 14.99 (C-18), 22.34

(2×CH₃), 26.42 (CH₂), 26.44 (CH₂), 27.18 (CH₂), 29.69 (CH),

29.94 (CH₂), 31.98 (CH₂), 38.32 (CH), 44.38 (CH), 48.34 (CH),

48.79 (C-13), 113.12 (CH), 115.54 (CH), 126.66 (CH), 132.20 (C),

134.73 (C), 138.12 (C), 144.25 (CH, C-1′), 153.78 (C-3), 210.04

(CdO); LRMS (FAB+) m/z 325.1 [100, (M + H)⁺]; HRMS

(FAB+) m/z calcd. for C₂₂H₂₉O₂(M + H)⁺325.2167, found

325.2166. Anal. (C₂₂H₂₈O₂) C, H.

16-(Pyridin-3-yl)methylene-estradiol (17a). To a solution of

16-(pyridin-3-yl)methylene-estrone 16a (360 mg, 1.0 mmol) in

EtOH (20 mL) and THF (20 mL), cooled to 0 °C (ice bath), was

added NaBH₄(0.100 g, 2.6 mmol). The reaction mixture was

allowed to warm to room temperature with stirring overnight. The

clear and colorless solution was concentrated in vacuo to ca. 20

mL volume, and H₂O (50 mL) was added to precipitate the product.

This was collected by filtration and washed with H₂O (50 mL),

16â-Ethoxymethyl-estrone (15). To a stirred solution of 3-O-

acetyl-16-methylene-estrone 34 (200 mg, 0.62 mmol) in EtOH (20

mL) at 0 °C was added dropwise aq. KOH (41 mg, 0.74 mmol in

2 mL). The resulting light yellow solution was stirred for 30 min

at 0 °C. The solvent was then removed under reduced pressure,

and H₂O was added (40 mL), followed by a few drops of 5 M

HCl. The resulting white precipitate was collected by filtration and

Inhibitors of 17â-Hydroxysteroid Dehydrogenase Type 1

Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4 1341

MeOH (20 mL), and diethyl ether (50 mL) and dried under high

vacuum to give the title compound as white powder (333 mg,

92%): mp > 150 °C (dec); ¹H NMR (400 MHz, DMSO-d₆) δ 0.58

(s, 3H, H-18), 1.15-1.43 (m, 5H), 1.83-1.96 (m, 2H), 2.12-2.31

(m, 3H), 2.62-2.80 (m, 3H), 3.99 (bs, 1H, H-17), 5.25 (bs, 1H,

-OH), 6.41 (d, J ) 2.0 Hz, 1H), 6.43 (d, J ) 2.3 Hz, 1H, H-4),

6.49 (dd, J ) 8.6, 2.3 Hz, 1H, H-2), 7.04 (d, J ) 8.6 Hz, 1H, H-1),

7.34 (dd, J ) 8.2, 5.1 Hz, 1H), 7.80 (d, J ) 8.2 Hz, 1H), 8.36 (dd,

J ) 4.7, 1.2 Hz, 1H), 8.56 (d, J ) 2.0 Hz, 1H), 9.00 (bs, 1H, -OH);

LRMS (FAB+) m/z 362.2 [100, (M + H)⁺]; LRMS (FAB-) m/z

360.2 [100, (M - H)^-]; HRMS (FAB+) m/z calcd. for C₂₄H₂₈NO₂

(M + H)⁺362.2120, found 362.2117.

16-(3,4,5-Trimethoxy-benzylidene)-estradiol (17j). To a solu-

tion of 16-(3,4,5-trimethoxy-benzylidene)-estrone 16j (448 mg, 1.0

mmol) in THF/MeOH (1/1, 40 mL) cooled to 0 °C (ice bath) was

added NaBH₄(100 mg, 2.64 mmol). The reaction was stirred for

2 h at this temperature, while the color changed from slightly yellow

to colorless. The solution was diluted with EtOAc (100 mL) and

washed with water (100 mL + 50 mL) and brine (50 mL), dried

(Na₂SO₄), and concentrated in vacuo to give the title compound as

white solid (445 mg, 99%): mp 135-140 °C; ¹H NMR (400 MHz,

CDCl₃) δ 0.74 (s, 3H, H-18), 1.20-1.60 (m, 6H), 1.90-1.98 (m,

1H), 2.01-2.08 (m, 1H), 2.21-2.41 (m, 3H), 2.74-2.90 (m, 3H),

3.86 (s, 3H), 3.89 (s, 6H), 4.15 (d, J ) 9.4 Hz, 1H, H-17), 4.62 (s,

1H, -OH), 6.49-6.51 (m, 1H), 6.59 (d, J ) 2.3 Hz, 1H, H-4),

6.63-6.68 (m, 3H), 7.18 (d, J ) 8.2 Hz, 1H); LRMS (FAB+) m/z

451.3 [100, (M + H)⁺].

16-(Thiophen-2-yl)methylene-estradiol (17k). To a stirred

solution of 16-(thiophenen-2-yl)methylene-estrone 16k (365 mg,

1.0 mmol) in THF (20 mL) and EtOH (40 mL) at 0 °C was added

NaBH₄(100 mg, 2.64 mmol), and the reaction was allowed to warm

to room temperature overnight before being concentrated in vacuo.

The residue was dissolved in EtOAc (40 mL) and the solution

washed with water (40 mL) and brine (20 mL), dried (Na₂SO₄),

and concentrated in vacuo. The white crude product was crystallized

from EtOAc/hexane to give the title compound as colorless crystals

(208 mg, 57%): mp > 180 °C (dec); ¹H NMR (400 MHz, CDCl₃)

δ 0.71 (s, 3H, H-18), 1.38-1.58 (m, 6H), 1.98-2.06 (m, 2H), 2.12-

2.22 (m, 1H), 2.26-2.40 (m, 2H), 2.68 (dd, J ) 16.9, 6.6 Hz, 1H),

2.82-2.92 (m, 2H), 4.17 (d, J ) 9.4 Hz, 1H, H-17), 4.56 (s, 1H,

-OH), 6.59 (d, J ) 2.3 Hz, 1H, H-4), 6.64 (dd, J ) 8.6, 2.3 Hz,

1H, H-2), 6.81 (d, J ) 2.3 Hz, 1H), 7.00-7.06 (m, 2H), 7.18 (d,

J ) 8.6 Hz, 1H, H-1), 7.27 (d, J ) 4.7 Hz, 1H); LRMS (FAB+)

m/z 366.2 [100, (M⁺)].

16-Isobutylidene-estradiol (17e). To a stirred solution of 16-

isobutylidene-estrone 16e (100 mg, 0.31 mmol) in a mixture of

MeOH/THF (3:1, 9 mL) at 0 °C was added dropwise a solution of

NaBH₄(57 mg, 1.51 mmol) in H₂O (3 mL). The resulting solution

was stirred at 0 °C for 20 min before glacial AcOH (5 drops) was

added, followed by aq. NaCl (10% solution, 20 mL), and the white

precipitate that formed was collected by filtration and dried under

high vacuum (102 mg). This was recrystallized from acetone/hexane

to give the title compounds as white crystals (60 mg, 60%): TLC

(CHCl₃/EtOAc, 8:2) R_f0.72 cf. R_f0.82; mp 143-145 °C; IR (KBr)

3435-3325 (OH), 2955-2865 (aliph CH), 1695-1500 (arom

1

CdC and exocyclic CdC) cm^-1; H NMR (400 MHz, CDCl₃) δ

0.67 (s, 3H, H-18), 0.96 (d, J ) 6.6 Hz, 3H, H-3′), 0.99 (d, J )

6.6 Hz, 3H, H-4′), 1.22-2.02 (m,13H), 2.81-2.87 (m, 2H, H-6),

3.91-3.98 (m, 1H, H-17), 4.66 (s, exchanged with D₂O, 1H, OH),

5.32 (dd, J ) 9.4, 2.3 Hz, 1H, H-1′), 6.56 (d, J ) 2.7 Hz, 1H,

H-4), 6.63 (dd, J ) 8.6, 2.7 Hz, 1H, H-2), 7.16 (d, J ) 8.6 Hz, 1H,

H-1); LRMS (FAB+) m/z 326.2 [95, (M⁺)], 309.2 [100, (M -

OH)⁺]; HRMS (FAB+) m/z calcd. for C₂₂H₃₀O₂(M⁺) 326.2246,

found 326.2252.

General Procedure for Hydrogenation: 18a-d. The starting

16-methylene estradiol was dissolved in a volume of THF and the

same volume of EtOH added to the solution. The solution was then

degassed by bubbling nitrogen through for 40 min before Pd/C (5wt

%, catalytic) was added and hydrogen gas (balloon) was passed

over the reaction. The reaction was stirred under hydrogen at room-

temperature overnight before being filtered through Celite and

purified by flash chromatography.

16-(2′,2′-Dimethyl)-propylidene-estradiol (17f). To a stirred

solution of 16-(2′,2′-dimethylpropylidene)-estrone 16f (100 mg, 0.31

mmol) in a mixture of MeOH/THF (2:1, 15 mL) at 0 °C was added

dropwise a solution of NaBH₄(44 mg, 1.16 mmol) in H₂O (2.5

mL). The resulting solution was stirred at 0 °C for 1 h before glacial

AcOH (4 drops) was added, followed by aq. NaCl (10% solution,

15 mL), and the white precipitate that formed was collected by

filtration and dried (79 mg). This was recrystallized from EtOAc/

hexane (1:8) to give the title compound as white crystals (63 mg,

79%): mp 219-222 °C; IR (KBr) 3550, 3140 (OH), 2960-28765

(aliph CH), 1610, 1505 (aliph CdC and arom CdC) cm^-1; ¹H NMR

(400 MHz, CDCl₃) δ 0.65 (s, 3H, H-18), 1.10 (s, 9H, C(CH₃)₃),

1.18-2.59 (m, 12H), 2.82-2.88 (m, 2H, H-6), 3.85-3.90 (m, 1H,

H-17), 4.60 (s, exchanged with D₂O, 1H, OH), 5.46 (dd, J ) 4.9,

2.5 Hz, 1H, H-1′), 6.55 (d, J ) 2.8 Hz, 1H, H-4), 6.61 (dd, J )

8.5, 2.8 Hz, 1H, H-2), 7.14 (d, J ) 8.5 Hz, 1H, H-1); LRMS

(FAB+) m/z 340.2 [49, (M⁺)], 323.2 [100, (M - OH)⁺], 283.2

[92, (M - C(CH₃)₃)⁺]; HRMS (FAB+) m/z calcd. for C₂₃H₃₂O₂

(M⁺) 340.2402, found 340.2390.

Representative example:

16-(Pyridin-3-yl)methyl-estradiol (18a). Purification by column

chromatography using an elution gradient of 100% hexane to 100%

EtOAc gave the title compound in 88% yield: mp 224-227 °C;

TLC (EtOAc/hexane, 1:3) R_f0.1; ¹H NMR (270 MHz, DMSO-d₆)

δ 0.76 (s, 3H, H-18), 1.00-1.04 (m, 2H), 1.20-1.35 (m, 5H), 1.54-

1.56 (m, 1H), 1.73-1.90 (m, 2H), 2.13-2.32 (m, 3H), 2.64 (m,

2H), 3.01 (d, J ) 9.4 Hz, 1H), 4.77 (d, J ) 4.5 Hz, 1H), 6.41 (d,

J ) 2.5 Hz, 1H, H-4), 6.49 (dd, J ) 8.4, 2.5 Hz, 1H, H-2), 7.03 (d,

J ) 8.4 Hz, 1H, H-1), 7.25-7.31 (m, 1H), 7.60-7.64 (m, 1H),

8.34-8.39 (m, 1H), 8.41 (d, J ) 1.7 Hz, 1H), 9.02 (s, 1H); LRMS

(FAB+) m/z 364.1 [100, (M + H)⁺]; HRMS (FAB+) m/z calcd.

for C₂₄H₃₀NO₂(M + H)⁺364.2276, found 364.2287.

3-O-Benzyl-estradiol-16-carboxylic Acid Methyl Ester (36).

To a cooled (0 °C), stirred solution of 31 (5 g, 12 mmol) in THF/

MeOH (90:10) was added NaBH₄(456 mg, 12 mmol), the reaction

mixture was stirred for 4 h at 0 °C, acidified with 2 N HCl, and

the products were extracted with ether (2×). The combined organic

layers were washed with brine, dried (Na₂SO₄), and concentrated

in vacuo. The solid was purified by column chromatography using

CHCl₃/EtOAc (90:10) as eluent to give 2.48 g of the 16â-carboxyl

diastereomer 36a (48%) and 0.91 g of the 16R 36b (18%) as white

solids.

16-Benzylidene-estradiol (17g). To a solution of 16-benz-

ylidene-estrone 16g (179 mg, 0.50 mmol) in EtOH (10 mL) and

THF (10 mL), cooled to 0 °C (ice bath), was added NaBH₄(100

mg, 2.64 mmol). The reaction was stirred overnight at room

temperature before being concentrated in vacuo and the residue

dissolved in EtOAc (40 mL). The solution was washed with H₂O

(2 × 25 mL) and brine (25 mL), dried (Na₂SO₄), and concentrated

in vacuo. The residue was dissolved in a small amount of EtOAc

and precipitated by addition of hexane. The white solid was

collected by filtration and dried under high vacuum to give the

1

title compound (175 mg, 97%): mp 213-216 °C; H NMR (270

36a: ¹H NMR (400 MHz, CDCl₃) δ 0.85 (s, 3H, H-18), 1.19

(ddd, J ) 12.9, 10.9, 6.2 Hz, 1H), 1.29-1.43 (m, 2H), 1.46-1.56

(m, 2H), 1.76 (dt, J ) 12.9, 8.6 Hz, 1H), 1.87-1.93 (m, 1H), 2.00-

2.13 (m, 2H), 2.20-2.27 (m, 1H), 2.30-2.35 (m, 1H), 2.83-2.89

(m, 2H, H-6), 3.15 (q, J ) 9.8 Hz, 1H, H-16R), 3.34 (d, J ) 8.2

Hz, 1H, OH), 3.72 (s, 3H, OCH₃), 3.89 (dd, J ) 9.8, 8.2 Hz, 1H,

H-17R), 5.04 (s, 2H, CH₂Ph), 6.72 (d, J ) 2.7 Hz, 1H, H-4), 6.79

(dd, J ) 8.4, 2.7 Hz, 1H, H-2), 7.21 (d, J ) 8.6 Hz, 1H, H-1),

MHz, DMSO-d₆) δ 0.61 (s, 3H, H-18), 1.20-1.50 (m, 5H), 1.82-

1.98 (m, 2H), 2.20-2.35 (m, 3H), 2.58-2.82 (m, 3H), 3.95-4.05

(m, 1H, H-17), 5.15 (d, J ) 6.2 Hz, 1H), 6.43 (s, 1H), 6.46 (d, J

) 2.3 Hz, 1H, H-4), 6.52 (dd, J ) 8.2, 2.3 Hz, 1H, H-2), 7.06 (d,

J ) 8.2 Hz, 1H, H-1), 7.15-7.20 (m, 1H), 7.31-7.41 (m, 4H);

LRMS (FAB+) m/z 360.2 [100, (M⁺)]; LRMS (FAB-) m/z 359.3

[100, (M - H)^-]; HRMS (FAB+) m/z calcd. for C₂₅H₂₈O₂(M⁺)

360.2089, found 360.2097.

1342 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4

Allan et al.

7.30-7.46 (m, 5H, Ph); LRMS (FAB+) m/z 91.0 [100], 420.1 [42,

(M⁺)]; HRMS (FAB+) m/z calcd. for C₂₇H₃₂O₄(M⁺) 420.2301,

found 420.2298. Anal. (C₂₇H₃₂O₄) C, H.

(dd, J ) 8.4, 2.7 Hz, 1H, H-2), 7.04 (d, J ) 8.1 Hz, 1H, H-1), 7.28

(dd, J ) 2.0; 1.0 Hz, 1H, furan); LCMS (APCI+) m/z 396.18 (M

+ H)⁺.

36b: ¹H NMR (400 MHz, CDCl₃) δ 0.84 (s, 3H, H-18), 1.20-

1.60 (m, 5H), 1.72 (app. q, J ) 12.4 Hz, 1H), 1.85-1.88 (m, 1H),

1.97 (dt, J ) 12.1, 3.3 Hz, 1H), 2.02-2.10 (m, 2H), 2.21-2.26

(m, 1H), 2.31-2.35 (m, 1H), 2.77 (ddd, J ) 12.1, 8.4, 4.0 Hz, 1H,

H-16â), 2.86 (m, 2H, H-6), 3.75 (s, 3H, OCH₃), 3.89 (dd, J ) 8.1,

4.4 Hz, 1H, H-17R), 5.04 (s, 2H, CH₂Ph), 6.73 (d, J ) 2.9 Hz, 1H,

H-4), 6.79 (dd, J ) 8.4, 2.6 Hz, 1H, H-2), 7.21 (d, J ) 8.1 Hz, 1H,

3-O-Benzyl-estradiol-16-carboxylic Acid (39). To a solution

of 36 (2.00 g, 4.75 mmol) in THF (20 mL) was added aq. NaOH

(285 mg, 7.13 mmol in 5 mL), and the reaction was stirred at room

temperature for 24 h. The solvents were removed in vacuo, and

6N HCl was added to the residue. The organics were extracted

with diethyl ether (5×), and the product was extracted into aq. K₂-

CO₃(10%, 5×). This basic solution was acidified with 6 N HCl

and the product extracted into EtOAc (5×). The combined organic

layers were washed with brine, dried (Na₂SO₄), and concentrated

in vacuo. The white solid was suspended in hexane and collected

by filtration to give the title compound (720 mg, 37%): ¹H NMR

(270 MHz, DMSO-d₆) δ 0.68 (s, 1.05H, H-18), 0.72 (s, 1.95H,

H-18), 1.00-1.50 (m, 6H), 1.60-1.90 (m, 3H), 2.08-2.15 (m, 1H),

2.23-2.29 (m, 1H), 2.72-2.76 (m, 2H, H-6), 2.92 (q, J ) 8.9 Hz,

0.65H, H-16), 3.68 (dd, J ) 7.2, 5.3 Hz, 0.35H, H-16), 3.80 (d, J

) 10.5 Hz, 0.65H, H-17), 4.91 (d, J ) 5.3, 0.35H, H-17), 5.03 (s,

2H, CH₂Ph), 6.68 (s, 1H, H-4), 6.74 (dd, J ) 8.6, 2.6 Hz, 1H,

H-2), 7.14 (d, J ) 8.6 Hz, 1H, H-1), 7.25-7.50 (m, 5H, Ph); LRMS

(FAB+) m/z 91.0 [100], 406.1 [40, (M⁺)].

(4′-Nitro-phenyl) 3-O-Benzyl-estradiol-16-carboxylate (40). To

a stirred solution of 39 (0.400 g, 0.98 mmol) in dry DMF (4 mL)

was added NEt₃(0.14 mL, 0.98 mmol) followed by bis(4-

nitrophenyl) carbonate (299 mg, 0.98 mmol). The solution was

stirred at room temperature for 2 h under N₂atmosphere before

being cooled to 0 °C and acidified with 2 N HCl. The mixture was

extracted with DCM (3×), and the combined organic layers were

washed with sat. NaHCO₃(5×) and brine, dried over Na₂SO₄, and

concentrated in vacuo. The residue was purified by flash chroma-

tography using an elution gradient of hexane to EtOAc to give the

title compound as yellow solid (270 mg, 52%). ¹H NMR (270 MHz

CDCl₃,) δ 0.87 (s, 3H, H-18), 1.27-1.50 (m, 4H), 1.80-2.05 (m,

4H), 2.10-2.40 (m, 3H), 2.82-2.88 (m, 2H, H-6), 3.05 (ddd, J )

11.7, 7.7, 3.7 Hz, 1H, H-16), 4.06 (dd, J ) 8.1, 5.1 Hz, 1H, H-17),

5.01 (s, 2H, CH₂Ph), 6.70 (d, J ) 2.6 Hz, 1H, H-4), 6.76 (dd, J )

8.8, 2.6 Hz, 1H, H-2), 7.19 (d, J ) 8.8 Hz, 1H, H-1), 7.25-7.45

(m, 7H), 8.27 (d, J ) 12.1 Hz, 2H); LRMS (FAB+) m/z 527.4

[20, (M⁺)]; 389.4 [6, (M - pNO₂PhO)⁺]; 91.1 [100, (PhCH₂⁺)].

General Procedure for Amide Coupling: 41a-c and 41e-j.

To a solution of the active ester 40 (0.090 g, 0.17 mmol) in dry

MeCN (2 mL) was added amine (2 equiv), and the solution was

stirred overnight. The solvents were removed in vacuo, and the

residue was semipurified by flash column chromatography using

an elution gradient of hexane to EtOAc to give 3-O-benzyl protected

E1-16-carboxyl amides with an average yield of 80%. These were

then debenzylated without further purification.

1

H-1), 7.30-7.46 (m, 5H, Ph); H NMR (400 MHz, CDCl₃, D₂O)

δ 3.80 (d, J ) 7.8 Hz, H-17R); LRMS (FAB+) m/z 91.0 [100],

420.1 [37, (M⁺)]; HRMS (FAB+) m/z calcd. for C₂₇H₃₂O₄(M⁺)

420.2301, found 420.2301; HPLC > 99% (t_R) 3.03, 4:96 H₂O:

MeOH). Anal. (C₂₇H₃₂O₄) C, H.

Methyl Estradiol-16â-carboxylate (37). Synthesized from 36a,

procedure as for 32. Recrystallization of the solid residue from

MeOH/H₂O gave the title compound as white solid (92%): mp

222-224 °C; ¹H NMR (400 MHz, DMSO-d₆) δ 0.75 (s, 3H, H-18),

1.04-1.46 (m, 5H), 1.64-1.86 (m, 4H), 2.08 (td, J ) 10.5, 3.3

Hz, 1H), 2.21-2.25 (m, 1H), 2.66-2.79 (m, 2H, H-6), 3.05 (q, J

) 9.0 Hz, 1H), 3.56 (s, 3H, OCH₃), 3.83 (dd, J ) 10.5, 5.5 Hz,

1H, H-17R), 5.02 (d, J ) 5.0 Hz, 1H, OH), 6.42 (d, J ) 2.7 Hz,

1H, H-4), 6.48 (dd, J ) 8.2, 2.7 Hz, 1H, H-2), 7.03 (d, J ) 8.6 Hz,

1H, H-1), 8.99 (s, 1H, OH); LRMS (FAB+) m/z 331.1 [100, (M

+ H)⁺]; HRMS (FAB+) calcd. for C₂₀H₂₆O₄(M⁺) 330.1831, found

330.1831; HPLC > 95% (t_R) 2.35, 8:92 H₂O:MeOH). Anal.

(C₂₀H₂₆O₄) C, H.

Estradiol-16-carboxylic Acid (19). To a suspension of 37 (1.1

g, 3.20 mmol) in MeOH (10 mL) was added aq. NaOH (0.58 g, 14

mmol in 3 mL). The solution was stirred at room temperature for

5 h before solvents were removed in vacuo and 6 N HCl was added

to the residue. The products were extracted with EtOAc (5×), and

the combined organic layers were washed with brine, dried (Na₂-

SO₄), and concentrated in vacuo. Recrystallization of the solid

residue from MeCN gave the title compound as white solid (470

1

mg, 46%): mp > 235 °C (dec); H NMR (400 MHz, DMSO-d₆)

δ 0.69 (s, 1.86H, H-18), 0.73 (s, 1.14H, H-18), 1.20-1.40 (m, 5H),

1.45-1.85 (m, 4H), 2.08-2.12 (m, 1H), 2.22-2.33 (m, 1H), 2.45

(m, 0.62H, H-16), 2.70-2.75 (m, 2H, H-6), 2.95 (q, J ) 9.0 Hz,

0.38H, H-16), 3.69 (d, J ) 8.0 Hz, 0.62H, H-17), 3.60 (d, J )

10.8 Hz, 0.38H, H-17), 4.92 (bs, 1H, OH), 6.42 (d, J ) 2.7 Hz,

1H, H-4), 6.48 (dd, J ) 8.2, 2.7 Hz, 1H, H-2), 7.03 (d, J ) 8.2 Hz,

1H, H-1), 8.99 (s, 1H); LRMS (FAB+) m/z 316.1 [100, (M⁺)];

HRMS (FAB+) calcd. for C₁₉H₂₄O₄(M⁺) 316.1675, found

316.1677.

Loading of 19 onto Oxime Resin (38). Oxime resin (0.500 g,

loading of 1.06 mmol/g resin) was swollen in dry DMF (7 mL) for

15 min after which 19 (0.251 g, 0.79 mmol) was added, followed

by DIC (41 µL) and HOBt (0.358 g, 2.61 mmol). The mixture was

shaken for 48 h at room temperature under inert atmosphere, and

the resin was filtered, washed with DCM, DMF, MeOH (three

cycles), and MeOH (5 times) and dried in vacuo. IR (KBr) 1750

General Procedure for Debenzylation: 19a-c and 19e-j. To

a solution of 3-O-benzyl amide 41 in THF (3 mL) was added Pd/C

(5wt %, catalytic), and the reaction mixture was stirred under H₂

(balloon) for 24 h. The Pd/C was removed by filtration and the

filtrate was concentrated in vacuo. The final product was precipi-

tated in a mixture of EtOAc/hexane. Compounds 19a-c and 19e-j

were obtained with an average yield of 36%.

(CdO, ester), 1600 (CdN, oxime) cm^-1

.

Representative example:

N-(Furan-2-ylmethyl)estradiol-16-carboxamide (19d). To a

suspension of loaded resin 38 (200 mg, maximum loading of 1.06

mmol/g resin) swollen in dry DCM (4 mL) was added furfurylamine

(47 µL, 0.53 mmol). The mixture was shaken at 40 °C under

nitrogen atmosphere for 24 h. After filtration, the resin was washed

with DCM (five times), and the filtrated was evaporated in vacuo.

The resulting brown oil was purified by column chromatography

using EtOAc/hexane (50:50) as eluent to give the title compound

as a yellow solid (3 mg, 3.6% over two steps): ¹H NMR (400

MHz, CDCl₃) δ 0.82 (s, 2.49H, H-18), 0.84 (s, 0.51H, H-18), 1.08-

1.60 (m, 6H), 1.84-1.89 (m, 2H), 2.16-2.24 (m, 2H), 2.28-2.36

(m, 1H), 2.56-2.62 (m, 1H, H-16), 2.80-2.83 (m, 2H, H-6), 3.74

(d, J ) 8.4 Hz, 0.83H, H-17), 3.80 (d, J ) 8.9 Hz, 0.17H, H-17),

4.41 (d, J ) 5.7 Hz, 2H, CH₂N), 6.15 (dd, J ) 3.2, 1.7 Hz, 1H,

furan), 6.25 (m, 1H, furan), 6.50 (d, J ) 2.7 Hz, 1H, H-4), 6.55

N-(Tetrahydro-furan-2-ylmethyl)estradiol-16-carboxamide

(19a). Yield ) 12%. H NMR (270 MHz, MeOH-d₃) δ 0.81 (s,

1

3H, H-18), 1.26-1.45 (m, 6H),1.50-1.70 (m, 3H), 1.80-2.00 (m,

6H), 2.10-2.18 (m, 1H), 2.27-2.32 (m, 1H), 2.60 (ddd, J ) 11.4,

8.1, 3.7 Hz, 1H, H-16), 2.74-2.79 (m, 2H, H-6), 3.70-3.95 (m,

3H), 3.93-4.00 (m, 1H), 6.45 (d, J ) 2.6 Hz, 1H, H-4), 6.52 (dd,

J ) 8.4, 2.6 Hz, 1H, H-2), 7.06 (d, J ) 8.4 Hz, 1H, H-1); LCMS

(APCI+) m/z 401.34 (M + 2H)⁺.

Ethyl Estrone-16-methylcarboxylate (44). Procedure as for

18a-d using EtOH as solvent. The progress of the reaction was

monitored by TLC (EtOAc/hexane, 3:7) and was found to be

complete after 48 h. The crude product was purified by column

chromatography (5-20% EtOAc in hexane) to obtain the deben-

zylated product 44 (0.828 g, 70%, mixture of two diastereomers

3:1) as a white solid: mp 140-143 °C; ¹H NMR (400 MHz, CDCl₃)

Inhibitors of 17â-Hydroxysteroid Dehydrogenase Type 1

Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4 1343

[chemical shifts of major to minor isomer ratio approximately 3:1

was determined using Me signals at 13-position], δ 0.916 and 0.988

(major isomer) [2 × s, 3H, H-18], 1.28 (t, J ) 7.0 Hz, 3H, CH₃),

1.40-1.60 (m, 5H), 1.75-1.80 (m, 1H), 1.92-2.54 (m, 6H), 2.74-

2.79 (dd, J ) 3.9, 4.2 Hz, 1H), 2.85-2.88 (m, 2H), 2.99-3.02 (m,

1H), 4.15-4.17 (q, 2H, CH₂), 4.63 (1H, OH), 6.57-6.58 (appd, J

) 2.7 Hz, 1H), 6.62-6.65 (dd, J ) 8.5, 2.7 Hz, 1H), 7.13-7.16

(dd, J ) 8.5, 3.9 Hz, 1H); LRMS (FAB+) m/z 357 [100, (M +

H)⁺]; HRMS (FAB+) m/z calcd. for C₂₂H₂₉O₄(M + H)⁺357.2065,

found 357.2054.

heated at 40 °C for 72 h. The reaction mixtures were filtered and

the filtrates concentrated under reduced pressure to obtain pale

yellow solids. The crude compounds were purified by flash

chromatography using DCM/MeOH (gradient elution, starting with

2% MeOH in DCM) to obtain pure benzylated intermediates

(average recovery 15 mg/100 mg of resin bound intermediate,

analyzed by ¹H NMR). These intermediates were then debenzylated

under H₂in the GreenHouse Synthesizer to obtain final compounds

in average 90% purity and with an average overall yield of 40%.

Synthesis of Resin Bound Intermediate (48). To a suspension

of Oxime resin (0.500 g, loading 1.06 mmol/g) swollen in DMF (7

mL) was added 20 (0.521 g, 5.3 mmol) followed by DIC (0.667 g,

5.3 mmol) and HOBt (0.715 g, 5.3 mmol). The reaction mixture

was shaken using a flask shaker for approximately 72 h under N₂.

The yellow reaction mixture obtained was filtered and the resin

washed with DCM, DMF, MeOH (three cycles with each solvent),

then MeOH (5×) and dried under high vacuum to obtain the resin

bound intermediate as yellow beads: IR (cm^-1) 1662 (CdO), 1739

(CO-O-Nd), 3502 (OH).

Parallel Synthesis of E1-Methylcarboxyl-amides. To a suspen-

sion of resin bound 48 (0.100 g, 0.106 mmol) in anhydrous DCM

(3.0 mL) under N₂in a glass tube in Radleys GreenHouse

Synthesizer was added the corresponding amine (5 molar equiva-

lents), and the reaction was heated at 40 °C under N₂for 72 h. The

reaction mixtures were filtered, and the filtrates were evaporated

using Genevac to obtain yellow solids. The crude compounds were

purified by flash chromatography using DCM/MeOH (gradient

elution, starting with 2% MeOH in DCM) to obtain compounds

20j-m with an average recovery of 14 mg/100 mg of resin bound

intermediate.

Estrone-16-methylcarboxylic Acid (20). To a solution of E1-

16-methyl carboxylic acid ethyl ester 44 (1.00 g, 2.8 mmol) in THF:

MeOH:H₂O (1:1:0.5, 10 mL) was added NaOH (0.56 g, 1.4 mmol),

and the mixture was heated to 80 °C in a sealed tube overnight.

The reaction was monitored by TLC (EtOAc: hexane, 1:1). The

crude mixture was acidified to pH ) 3-4, and the organics were

extracted with EtOAc and DCM. The combined organic layers were

dried (MgSO₄) and concentrated to obtain white solid. Recrystal-

lization from MeOH/hexane gave the title compound as a white

1

solid (0.920 g, quantitative yield): mp > 250 °C (dec); H NMR

(400 MHz, DMSO-d₆) [chemical shifts of major to minor isomer

ratio approximately 3:1 was determined using Me signals at 13-

position] δ 0.786 [major isomer] and 0.896 (2 × s, 3H, H-18),

1.31-1.52 (m, 6H), 1.74-1.77 (m, 1H), 1.88-1.91 (m, 1H), 2.16-

2.33 (m, 4H), 2.40-2.45 (m, 1H), 2.55-2.62 (m, 1H), 2.72-2.74

(m, 2H), 6.43-6.44 (appd, J ) 2.5 Hz, 1H), 6.48-6.51 (d, J )

8.5, 2.4 Hz, 1H, H-2), 7.02-7.04 (d, J ) 8.5 Hz, 1H, H-1), 9.0

(bs, 1H); LRMS (FAB+) m/z 329 [100, (M + H)⁺]; HRMS

(FAB+) m/z calcd. for C₂₀H₂₆O₄(M + 2H)⁺330.1831, found

330.1751.

3-O-Benzyl-estrone-16-methylcarboxylic Acid (43). The reac-

tion wascarried out using the procedure outlined for compound 20

with 3-O-benzyl-estrone-16-methylcarboxylic acid ethyl ester 42

(1.714 g, 3.8 mmol), NaOH (1.537 g, 38.4 mmol) and THF:MeOH:

H₂O (1:1:0.5, 10 mL). The title compound (1.114 g, 90%) was

isolated as white solid after recrystallization from EtOAc/hexane;

¹H NMR (400 MHz,DMSO-d₆) δ 0.81, 0.84 (2 × s, 3H, H-18),

1.36-2.64 (m, 14H), 2.81-2.83 (m, 2H), 5.06 (s, 2H, CH₂Ph),

6.73 (appd, J ) 2.3 Hz, 1H), 6.77 (dd, J ) 8.5 and 2.5 Hz, 1H),

7.18 (d, J ) 8.5 Hz, 1H), 7.31-7.44 (5H); LRMS (FAB+) m/z

90.9 [100], 418.0 [21, (M⁺)]; HRMS (FAB+) m/z calcd. for

Representative example:

N-(Tetrahydrofuran-2-ylmethyl)estrone-16-methylcarbox-

amide (20a). ¹H NMR (400 MHz, CDCl₃) δ 0.69, 0.73, 0.83, 0.87

(4 × s, 3H, H-18), 1.2-3.1 (m, 21H), 3.50-3.95 (m, 4H), 6.0 (bs)

and 6.15-6.17 (m, 1H), 6.50 (d, J ) 2.3 Hz, 0.8H, H-4, major),

6.55-6.58 (m, 0.8H, H-2, major), 6.76-6.81 (m, H-4 and H-2

minor, 0.4H) 7.05 (d, J ) 8.6 Hz, H-1 major + minor underneath,

1H); LCMS (ES-) m/z 410.2 (M - H)^-; HRMS (FAB+) m/z calcd.

for C₂₅H₃₄NO₄(M + H)⁺412.2487, found 408.2491.

Representative example:

N-(Tetrahydro-furan-2-ylmethyl)estradiol-16-methylcarbox-

C₂₇H₃₀O₄(M⁺) 418.2099, found 418.2148; HPLC > 99% (t_R

)

1

amide (21a). H NMR (400 MHz, CDCl₃) δ 0.78, 0.79, (2 × s,

1.89, 100% MeOH).

3H, H-18), 0.86-0.88 (m, 1H), 1.19-1.55 (m, 16H), 1.67-1.97

(m, 2H), 2.12-2.30 (m, 3H), 2.74-2.85 (m, 3H), 3.73-3.77 (m,

1H), 4.48 (d, J ) 9.8 Hz, 1H), 4.63 (bs, 1H), 6.57 (s, 1H, H-4),

6.64 (dd, J ) 8.6, 2.8 Hz, 1H, H-2), 7.16 (d, J ) 8.5 Hz, 1H,

H-1); LCMS (ES-) m/z 311 (M - C₅H₉NO)^-.

General Method for Solution Phase Parallel Synthesis of E1-

16-Methylcarboxyl-amides. The carboxylic acid 20 (0.050 g, 0.15

mmol) or 21 (0.050 g, 0.15 mmol) was suspended in DCM (1.0

mL) in a glass tube in a Radleys GreenHouse Parallel Synthesizer.

To this was added a solution of EDC (0.043 g, 0.227 mmol), DMAP

(catalytic), and NEt₃(0.100 mL) in DCM (1.0 mL). The reaction

mixture was stirred for 30 min before the corresponding amine (5

molar equivalents) in DCM (1.0 mL) was added and stirring was

continued overnight. The completions of the reactions were

monitored by TLC (EtOAc:hexane, 2:7). The crude mixtures were

concentrated under reduced pressure and purified using flash

chromatography with DCM:MeOH (gradient elution, starting with

2% MeOH in DCM). Average yield 39%.

X-ray Crystallography. Crystal Data for 15, C₂₁H₂₈O₃, M )

328.43, λ ) 0.71073 Å, orthorhombic, space group P2₁2₁2₁, a )

7.1090(1), b ) 11.9830(2), c ) 21.4370(4) Å, U ) 1826.16(5)

Å³, Z ) 4, D_c)1.195 mg/m³, µ ) 0.078 mm^-1, F(000) ) 712,

crystal size 0.40 × 0.13 × 0.04 mm, unique reflections ) 4176

[R(int) ) 0.0742], observed I > 2σ(I) ) 3169, data/restraints/

paramaters ) 4176/0/221, R1 ) 0.0412 wR2 ) 0.0857 (obsd data),

R1 ) 0.0663 wR2 ) 0.0952 (all data), max peak/hole 0.259 and

-0.283 eÅ^-3, software used, SHELXS,⁵⁴SHELXL,⁵⁵and OR-

TEX.⁴⁰

Crystal Data for 16c (block crystals), C₂₄H₂₄NO₂, M ) 358.44,

λ ) 0.71073 Å, tetragonal, space group P4₃, a ) 6.8940(1), b )

6.8940(1), c ) 38.7530(6)Å, U ) 1841.82(5) Å³, Z ) 4, D_c)

1.293 mg/m³, µ ) 0.082 mm^-1, F(000) ) 764, crystal size 0.25 ×

0.25 × 0.12 mm, unique reflections ) 4064 [R(int) ) 0.0383],

observed I > 2σ(I) ) 3861, data/restraints/paramaters ) 4064/1/

247, R1 ) 0.0338 wR2 ) 0.0813 (obs. data), R1 ) 0.0376 wR2 )

0.0842 (all data), max peak/hole 0.201 and -0.228 eÅ^-3, software

used, SHELXS,⁵⁴SHELXL,⁵⁵and ORTEX.⁴⁰

Synthesis of Resin Bound Intermediate (46). To a suspension

of Oxime resin (2.0 g, loading 1.06 mmol/g) swollen in DMF (7

mL) was added 43 (1.428 g, 3.18 mmol) followed by DIC (1.33 g,

10.6 mmol) and HOBt (1.431 g, 10.6 mmol). The reaction mixture

was shaken using a flask shaker for approximately 72 h under N₂.

The yellow reaction mixture obtained was filtered and the resin

washed with DCM, DMF, MeOH (three cycles with each solvent),

then MeOH (5×) and dried in vacuo to obtain the resin bound

intermediate (2.391 g) as yellow beads: IR (KBr) 3502 (OH), 1739

(CO-O-Nd), 1662 (CdO) cm^-1

.

Parallel Synthesis of E1-Methylcarboxyl-amides Followed by

Debenzylation. To a suspension of resin bound intermediate 46

(0.100 g, 0.106 mmol) in anhydrous DCM (3.0 mL) under N₂in a

glass tube in a Radleys GreenHouse Synthesizer was added the

corresponding amine (5 molar equivalents), and the reactions were

Crystal Data for 16c (needle crystals), C₂₄H₂₅NO₂, M ) 359.45,

λ ) 0.71073 Å, monoclinic, space group C2, a ) 31.4610(8), b )

6.4880(2), c ) 9.2690(3) Å, â ) 90.923(1)° U ) 1891.73(10) Å³,

Z ) 4, D_c) 1.262 Mg/m³, µ ) 0.080 mm^-1, F(000) ) 768, crystal

size 0.38 × 0.13 × 0.05 mm, unique reflections ) 4183 [R(int) )

1344 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4

Allan et al.

(15) Sawicki, M. W.; Erman, M.; Puranen, T.; Vihko, P.; Ghosh, D.

Structure of the ternary complex of human 17â-hydroxysteroid

dehydrogenase type 1 with 3-hydroxyestra-1,3,5,7-tetraen-17-one

(equilin) and NADP+. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 840-

845.

0.0627], observed I > 2σ(I) ) 3680, data/restraints/paramaters )

4183/61/293, R1 ) 0.0602 wR2 ) 0.1044 (obsd data), R1 ) 0.0741

wR2 ) 0.1083 (all data), max peak/hole 0.193 and -0.272 eÅ^-3

,

software used, SHELXS,⁵⁴SHELXL,⁵⁵and ORTEX.⁴⁰

(16) 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 type1 17â hydroxysteroid dehydrogenase inhibitors: estradiol-

adenosine hybrids with high affinity. FASEB J. 2002, 16, 1829-

1830.

Acknowledgment. This research was supported by Sterix

Ltd., a member of the Ipsen group.

Supporting Information Available: Experimental details and

data for compounds 23 and 1 and analytical and spectroscopic data

for compounds 16b-d, 16f-l, 17b-d, 17h, 17i, 18b-d, 19b, 19c,

19e-j, 20b-m, and 21b-f. Also available are microanalysis data

on selected compounds, HPLC data for biologically tested com-

pounds for which microanalysis data are not available, structures

and IC₅₀data for compounds used to develop the QSAR, figure

showing the crystal packing of polymorphs of 16c, and crystal-

lographic data for compounds 15 and the two polymorphs of 16c.

Crystallographic data have been deposited with the Cambridge

Crystallographic Data Centre as CCDC 278452, 278453, and

278454, respectively. Copies of these data can be obtained free of

charge on application to CCDC, 12 Union Rd., Cambridge CB2

1EZ, UK [fax (+44) 1223 336033, e-mail: deposit@ccdc.cam.ac.uk].

This material is available free of charge via the Internet at http://

pubs.acs.org.

(17) Poirier, D. Inhibitors of 17â-hydroxysteroid dehydrogenase. Curr.

Med. Chem. 2003, 10, 453-477.

(18) Sam, K.-M.; Boivin, R. P.; Tremblay, M. R.; Auger, S.; Poirier, D.

C16 and C17 derivatives of estradiol as inhibitors of 17â-hydroys-

teroid dehydrogenase type1: chemical synthesis and structure activity

relationships. Drug Des. DiscoVery 1998, 15, 157-180.

(19) Poirier, D.; Boivin, R. P.; Be´rube´, M.; Lin, S.-X. Synthesis of a first

estradiol-adenosine hybrid compound. Synth. Commun. 2003, 33,

3183-3192.

(20) Hirvela¨, L.; Johansson, N.; Koskimies, P.; Pentika¨inen, O. T.;

Nyro¨nen, T.; Salminen, T. A.; Johnson, M. S. Novel compounds and

their use in therapy. PCT WO 2005032527, 2005.

(21) Messinger, J.; Thole, H.-H.; Husen, B.; Van Steen, B. J.; Schneider,

G.; Hulshof, J. B. E.; Koskimies, P.; Johansson, N.; Adamski, J.

Novel 17beta hydroxysteroid dehydrogenase type 1 inhibitors. PCT

WO 2005047303, 2005.

(22) Vicker, N.; Lawrence, H. R. R.; Allan, G. M.; Bubert, C.; Fischer,

D. S. M.; Purohit, A.; Reed, M. J.; Potter, B. V. L. Preparation of

estranes as 17 â-hydroxysteroid dehydrogenase inhibitors. PCT WO

2004085457, 2004.

(23) 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.

References

(1) Ferlay, J.; Bray, F.; Pisani, P.; Parkin, D. M. IARC Cancerbase No.

5 version 2.0, IARC Press: Lyon, 2004.

(24) Fischer, D. S.; Allan, G. M.; Bubert, C.; Vicker, N.; Smith, A.; Tutill,

H. J.; Purohit, A.; Wood, L.; Packham, G.; Mahon, M. F.; Reed, M.

J.; Potter, B. V. L. E-Ring modified steroids as novel potent inhibitors

of 17â-hydroxysteroid dehydrogenase type 1. J. Med. Chem. 2005,

48, 5749-5770.

(2) Miller, W. R. Biological rationale for endocrine therapy in breast

cancer. Best Pract. Res. Clin. Endoc. Metab. 2004, 18, 1-32.

(3) Santner, S. J.; Feil, P. D.; Santen, R. J. In situ estrogen production

via the estrone sulfatase pathway in breast tumors - relative

importance versus the aromatase pathway. J. Clin. Endocrinol. Metab.

1984, 59, 29-33.

(4) Reed, M. J.; Purohit, A.; Woo, L. W. L.; Newman, S. P.; Potter, B.

V. L. Steroid sulfatase: molecular biology, regulation, and inhibition.

Endocr. ReV. 2005, 26, 171-202.

(5) Mindnich, R.; Mo¨ller, G.; Adamski, J. The role of the 17 beta-

hydroxysteroid dehydrogenases. J. Mol. Cell. Endocrinol. 2004, 218,

7-20. The 17â-HSD type 13 enzyme was disclosed at a workshop

on 11â- and 17â-hydroxysteroid dehydrogenases: role in human

disease, Germany 2005. Here 17â-HSD type 13 was identified as

human retinal short chain dehydrogenase/reductase 3, PDB code

1YDE.

(6) Puranen, T.; Poutanen, M.; Ghosh, D.; Vihko, R.; Vihko, P. Origin

of substrate specificity of human and rat 17â-hydroxsteroid dehy-

drogenase type 1 using chimeric enzymes and site-directed substitu-

tions. Endocrinology 1997, 138, 3532-3539.

(7) Jin, J. Z.; Lin, S. X. Human estrogenic 17 beta-hydroxysteroid

dehydrogenase: predominance of estrone reduction and its induction

by NADPH. Biochem. Biophys. Res. Commun. 1999, 259, 489-493.

(8) Spiers, V.; Green, A. R.; Atkin, S. L. Activity and gene expression

of 17â-hydroxysteroid dehydrogenase type 1 in primary cultures of

epithelial and stromal cells derived from normal and tumorous breast

tissue: the role of IL8. J. Steroid Biochem. Mol. Biol. 1998, 67, 267-

274.

(9) Martel, C.; Rhe´aume, E.; Takahashi, M.; Trudel, C.; Coue¨t, J.; Luu-

The, V.; Simard, J.; Labrie, F. Distribution of 17â-hydroxysteroid

dehydrogenase gene expression and activity in rat and human tissues.

J. Steroid Biochem. Mol. Biol. 1992, 41, 597-603.

(10) Penning, T. M. 17â-hydroxysteroid dehydrogenase: Inhibitors and

inhibitor design. Endocrine-Related Cancer 1996, 3, 41-56.

(11) Peltoketo, H.; Isomaa, V.; Maentausta, O.; Vihko, R. Complete amino

acid sequence of human placental 17 beta-hydroxysteroid dehydro-

genase deduced from cDNA. FEBS Lett. 1988, 239, 73-77.

(12) Ghosh, D.; Pletnev, V. Z.; Zhu, D. W.; Wawrzak, Z.; Duax, W. L.;

Pangborn, W.; Labrie, F.; Lin, S. X. Structure of human estrogenic

17 beta-hydroxysteroid dehydrogenase at 2.20 Å resolution. Structure

1995, 3, 503-513.

(13) Breton, R.; Housset, D.; Mazza, C.; Fontecilla-Camps, J. C. The

structure of a complex human 17â-hydroxysteroid dehydrogenase

with estradiol and NADP+ identifies two principal targets for the

design of inhibitors. Structure 1996, 4, 905-915.

(25) Thomas, J. L.; LaRochelle, M. C.; Covey, D. F.; Strickler, R. C.

Inactivation of human placental 17â,20R-hydroxysteroid dehydro-

genase by 16-methylene estrone, an affinity alkylator enzymatically

generated from 16-methylene estradiol-17â. J. Biol. Chem. 1983, 258,

19, 11500-11504.

(26) Ringold, H. J.; Rosenkranz, G. 16-Methyl derivatives of estrone and

estradiol. US 3,103,521, 1963.

(27) Lawrence, H.; Vicker, N.; Allan, G. M.; Smith, A.; Mahon, M.; Tutill,

H.; Purohit, A.; Reed, M. J.; Potter, B. V. L. Novel and potent 17â-

hydroxysteroid dehydrogenase type 1 inhibitors. J. Med. Chem. 2005,

48, 2759-2762.

(28) Schwenk, E.; Montclair, N. J. Oxygenated estrogenic hormones and

method of preparing same. US 2,294,938, 1942.

(29) Cambie, R. C.; Manning, T. D. R. Oxidation products of 3-meth-

oxyoestra1,3,5(10)-trien-17â-yl acetate. J. Chem Soc. C. 1968, 2603-

2608.

(30) Katzenellenbogen, J. A.; Myers, H. N.; Johnson, H. J. Reagents for

photoaffinity labeling of estrogen binding proteins. Synthesis of some

azide and diazo derivatives of estradiol, estrone and hexestrol. J. Org.

Chem. 1973, 38, 3525-3533.

(31) Pearson, A. J.; Han, G. R. Benzylic oxidation using tert-butyl

hydroperoxide in the presence of chromium hexacarbonyl. J. Org.

Chem. 1985, 50, 2791-2792.

(32) Fischer, D. S.; Woo, L. W. L.; Mahon, M. F.; Purohit, A.; Reed, M.

J.; Potter, B. V. L. D-ring modified estrone derivatives as novel potent

inhibitors of steroid sulfatase. Bioorg. Med. Chem. 2003, 11, 1685-

1700.

(33) Hejaz, H. A. M.; Purohit, A.; Mahon, M. F.; Reed, M. J.; Potter, B.

V. L. Synthesis and biological activity of the superestrogen (E)-17-

oximino-3-O-sulfamoyl-1,3,5(10)-estratriene: X-ray crystal structure

of (E)-17-oximino-3-hydroxy-1,3,5(10)-estratriene. J. Med. Chem.

1999, 42, 3188-3192.

(34) Tanabe, M.; Peters, R. H.; Chao, W.-R.; Shigeno, K. Estrone

sulfamate inhibitors of estrone sulfatase, and associated pharmaceuti-

cal compositions and methods of use. WO 9933858, 1999.

(35) Paquette, L. A.; Dahnke, K.; Doyon, J.; He, W.; Wyant, K.; Friedrich,

D. Regioselective conversion of cycloalkanones to vinyl bromides

with 1,2-functionality transportation. A general stratagem. J. Org.

Chem. 1991, 56, 21, 6199-6205.

(36) Nicolai, E.; Cure, G.; Goyard, J.; Kirchner, M.; Teulon, J. M.;

Versigny, A.; Cazes, M.; Vironeoddos, A.; Caussade, F.; Cloarec,

A. Synthesis and angiotensin-II receptor agonist activity of C-linked

pyrazole derivatives. Chem. Pharm. Bull. 1994, 42, 8, 1617-1630.

(14) Azzi, A.; Reshe, P. H.; Zhu, D. W.; Campbell, R. L.; Labrie, F.;

Lin, S. X. Crystal structure of human estrogenic 17â-hydroxysteroid

dehydrogenase-1 complexed with 17â-estradiol. Nat. Struct. Biol.

1996, 3, 665-668.

Inhibitors of 17â-Hydroxysteroid Dehydrogenase Type 1

Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4 1345

(37) Fevig, T. L.; Mao, M. K.; Katzenellenbogen, J. A. Estrogen receptor

binding tolerance of 16R-substituted estradiol derivatives. Steroids

1988, 51, 5-6, 471-497.

(38) Tapolcsa´nyi, P.; Wo¨lfling, J.; Falkay, G.; Ma´rki, A.; Minorics, R.;

Schneider, G. Synthesis and receptor-binding examination of 16-

hydroxymethyl-3-17-estradiol stereoisomers. Steroids 2002, 67,

371-377.

(39) Labaree, D. C.; Reynolds, T. Y.; Hochberg, R. B. Estradiol-16R-

carboxylic acid esters as locally active estrogens. J. Med. Chem. 2001,

44, 1802-1814.

(40) McArdle, P. ORTEX2.1 - a 1677-atom veriosn of ORTEP with

automatic cell outline and cell packing for use on a PC. J. Appl.

Crystallogr. 1994, 27, 438-439.

(41) Goto, G.; Yoshikoda, K.; Hiraga, K. The Serini reaction of 16,17-

dihydroxysteroid monoacetates. Tetrahedron 1974, 30, 2107-2114.

(42) Minu, M.; Jindal, D. P. Synthesis and biological activity of

16-arylidene derivatives of estrone and estrone methyl ether. Indian.

J. Chem. Sect. B. 2003, 42, 166-172.

(43) Boivin, R. P.; Luu-The, V.; Lachance, R.; Labrie, F.; Poirier, D.

Structure-activity relationships of 17R-derivatives of estradiol as

inhibitors of steroid sulfatase. J. Med. Chem. 2000, 43, 4465-4478.

(44) 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.

(45) Klebe, G. Comparative Molecular Similarity Indices: CoMSIA.

Perspect. Drug Disc. Des. 1998, 12-14, 87-104.

(46) Lemmen, C.; Lengauer, T. Time-efficient flexible superposition of

medium-sized molecules. J. Comput.-Aided Mol. Des. 1997, 11, 357-

368.

(47) Lemmen, C.; Lengauer, T.; Klebe, G. FlexS: A method for fast

flexible ligand superposition. J. Med. Chem. 1998, 41, 4502-4520.

(48) Cramer, R. D., III; Bunce, J. D.; Patterson, D. E.; Frank, I. E. Cross-

validation, bootstrapping, and partial least squares compared with

multiple regression in conventional QSAR Studies. Quant. Struct-

Act. Relat. 1988, 7, 18-25.

(49) Zhu, B. T.; Lee, A. J. NADPH-dependent metabolism of 17â-estradiol

and estrone to polar and nonpolar metabolites by human tissues and

cytochrome P450 isoforms. Steroids 2005, 70, 225-244.

(50) Duncan, L. J.; Reed, M. J. The role and proposed mechanism by

which oestradiol 17â-hydroxysteroid dehydrogenase regulates breast

tumour oestrogen concentrations. J. Steroid Biochem. Mol. Biol. 1995,

55, 565-572.

(51) Regan, B. M.; Hayes, F. N. 17-and 17a-Aza-D-homosteroids. J. Am.

Chem. Soc. 1956, 78, 639-643.

(52) Matkovics, B.; Tarodi, B.; Balaspiri, L. Rearrangement of steroids,

VII. Schmidt reaction and Beckmann rearrangement of estrone and

its derivatives. Acta C. Acad. Sci. Hung. 1974, 80, 79-87.

(53) Sweet, F.; Boyd, J.; Medina, O.; Konderski, L.; Murdock, G. L.

Hydrogen bonding in steroidogenesis - studies on new heterocyclic

analogues of estrone that inhibit human estradiol 17 beta-dehydro-

genase. Biochem. Biophys. Res. Commun. 1991, 180, 1057-1063.

(54) Sheldrick, G. M. Phase annealing in SHELX-90: Direct methods

for larger structures. Acta Crystallogr. 1990, A46 (6), 467-473.

(55) Sheldrick, G. M. SHELXL-97, a computer program for crystal

structure refinement, University of Go¨ttingen, 1997.

JM050830T

Article Doi

Modification of estrone at the 6, 16, and 17 positions: Novel potent inhibitors of 17β-hydroxysteroid dehydrogenase type 1

DOI: 10.1021/jm050830t

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm050830t

Full text of DOI:10.1021/jm050830t

Products guided by the article

R&D Labs maybe for 879880-13-4

Relevant to this article

Hot Product