Synthesis of N-substituted piperidine-4-(benzylidene-4-carboxylic acids) and evaluation as inhibitors of steroid-5α-reductase type 1 and 2

Article abstract of DOI:10.1016/S0968-0896(00)00070-5

The synthesis of N-substituted piperidine-4-(benzylidene-4-carboxylic acids) is described [benzoyl (1Scheme 2Synthetic strategy.), benzyl (2), adamantanoyl (3), cyclohexanoyl (4), cyclohexylacetyl (5), diphenylacetyl (6), dicyclohexylacetyl (7), 2-propylpentanoyl (8), diphenylcarbamoyl (9), trimethylacetyl (10), 3,3-dimethylacryloyl (11), dicyclohexylacetyl derivative of the benzyl compound (12)]. Compounds were tested for inhibitory activity toward 5α-reductase isozymes 1 and 2 in human and rat. The test compounds inhibited 5α-reductase, showing a broad range of inhibitory potencies. In rat, compounds 6 (IC₅₀=3.44 and 0.37 μM for type 1 and 2, respectively) and 9 (IC₅₀=0.54 and 0.69 μM for type 1 and 2, respectively) displayed the best inhibition toward both isozymes. Compound 7 showed a strong inhibition toward type 2 human and rat enzyme (IC₅₀=60 and 80 nM) but only a moderate activity versus type 1 enzyme (IC₅₀ approximately 10 μM for rat and human enzyme). In vivo, selected compounds reduced prostate weights in castrated testosterone treated rats. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(00)00070-5

Bioorganic & Medicinal Chemistry 8 (2000) 1479±1487
Synthesis of N-Substituted Piperidine-4-(benzylidene-4-carboxylic
acids) and Evaluation as Inhibitors of Steroid-5ꢀ-Reductase
Type 1 and 2
Franck Picard, Eckhard Baston,y Wolfgang Reichert and Rolf W. Hartmann*
FR 12.1 Pharmazeutische und Medizinische Chemie, Universitat des Saarlandes, PO Box 15 11 50, D-66041 Saarbrucken, Germany
È
È
Received 27 January 2000; accepted 29 February 2000
AbstractÐThe synthesis of N-substituted piperidine-4-(benzylidene-4-carboxylic acids) is described [benzoyl (1), benzyl (2), ada-
mantanoyl (3), cyclohexanoyl (4), cyclohexylacetyl (5), diphenylacetyl (6), dicyclohexylacetyl (7), 2-propylpentanoyl (8), diphe-
nylcarbamoyl (9), trimethylacetyl (10), 3,3-dimethylacryloyl (11), dicyclohexylacetyl derivative of the benzyl compound (12)].
Compounds were tested for inhibitory activity toward 5a-reductase isozymes 1 and 2 in human and rat. The test compounds
inhibited 5a-reductase, showing a broad range of inhibitory potencies. In rat, compounds 6 (IC50=3.44 and 0.37 mM for type 1 and
2, respectively) and 9 (IC50=0.54 and 0.69 mM for type 1 and 2, respectively) displayed the best inhibition toward both isozymes.
Compound 7 showed a strong inhibition toward type 2 human and rat enzyme (IC50=60 and 80 nM) but only a moderate activity
versus type 1 enzyme (IC50 approximately 10 mM for rat and human enzyme). In vivo, selected compounds reduced prostate weights
in castrated testosterone treated rats. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
5a-reductase inhibitor in BPH treatment is ®nasteride
(Proscar1), a steroidal compound inhibiting mainly
isozyme 2. The long term pharmacotherapy with ®nas-
teride has generally been well tolerated.12 Nevertheless,
its limited activity and its side e€ects which are related
with sexual function (libido, impotence, ejaculatory dis-
order)13 have caused us to look for a new class of 5a-
reductase inhibitors. The fact that ®nasteride decreases
circulating DHT levels by only 70%, whereas a combi-
nation of selective type 1 (MK 386) and type 2 (®nas-
teride) inhibitors shows a nearly complete suppression
of serum DHT,14 a strategy of dual inhibition, i.e. inhi-
bition of both isozymes by one compound, became
attractive. The steroidal compound dutasteride (GG
745) is up to now the most ecient dual inhibitor exhi-
biting Ki values towards isozymes 1 and 2 of 6 and 7 nM,
respectively.15 In vivo, it shows a nearly complete sup-
pression of serum DHT.15 18 It is generally accepted
that some of the side e€ects of steroidal compounds are
related to their steroidal structures.19 Therefore, it is our
goal to synthesize nonsteroidal dual inhibitors of 5a-
reductase isozymes 1 and 2. Potent nonsteroidal inhibi-
tors have already been described so far,9,10,19 some of
which show dual type of inhibition.10,19 Recently, we
found in the class of N-substituted 4-(5-indolyl) benzoic
acids (Scheme 1) dual inhibitors of the human iso-
zymes exhibiting mediocre potency (best IC50 values
5a-Reductase (EC 1.3.99.5) is an NADPH-dependent
enzyme that catalyzes the conversion of testosterone to
the more potent androgen dihydrotestosterone (DHT).1
High concentrations of DHT are associated with the
etiology and growth of diseases like benign prostatic
hyperplasia (BPH),2 prostatic carcinoma,3 male pattern
baldness4 and hirsutism in women.5 BPH is the most
common benign tumor a€ecting over 40% of men
above the age of 70.6 Surgical therapy (TURP) is cur-
rently used in the treatment of BPH,7 and until recently
no non invasive method was available. That is why 5a-
reductase has attracted a considerable interest as a new
therapeutic target and the search for potent inhibitors of
this enzyme has become the goal of several research
groups.8 10 However, the diculty with developing
inhibitors of 5a-reductase is related to the fact that there
are two isozymes (type 1 and 2) with a di€erent tissue
distribution pattern and with distinct biochemical and
pharmalogical properties.11 The most currently used
*Corresponding author. Tel.: +49-681-302-3424; fax: +49-681-302-
4386; e-mail: rwh@rz.uni-sb.de
yPresent address: Institut de Chimie organique, Universite de Lau-
sanne, CH-1015 Lausanne-Dorigny, Switzerland.
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00070-5

1480
F. Picard et al. / Bioorg. Med. Chem. 8 (2000) 1479±1487
using an appropriate solvent. In a last step (method C),
saponi®cation of the methyl esters 1a±11a was per-
formed with potassium carbonate and methanol/H2O to
give the corresponding carboxylic acids 1±11. Com-
pound 12 was prepared by hydrogenation of 7 using
palladium on charcoal as a catalyst (Scheme 2).
Scheme 1. General structures.
type 1: 2 mM, type 2: 6 mM).20 These compounds had
been designed as mimics of the steroidal substrate. In
contrast to steroidal inhibitors which bear a lipophilic
substituent in the 17b position, the substituent R of the
indole compounds is located in the plane of the mole-
cule. Aiming at the enhancement of inhibitory potency
we looked for compounds which are closer to the steroidal
structure. N-Substituted piperidine-4-(benzylidene-4-
carboxylic acids) (Scheme 1) are mimics having the
substituent R located outside the plane. In the following
we describe the synthesis of a series of these compounds
and the evaluation of their inhibitory activity toward rat
and human 5a-reductase isozymes 1 and 2 as well as
their in vivo activity in the rat.
Biological Results
In vitro
The inhibitory activity of compounds 1±12 and ®nas-
teride in vitro was determined using rat prostate homo-
genates (pH 6.6, type 1; pH 5.5, type 2) and human
prostate homogenate (BPH tissue for type 2) according
to the method of Liang et al.,24 and the DU145 cell line
(for human type 1 enzyme) as described in the litera-
ture.25 As can be seen in Table 1, in rat type 1 isozyme,
most of the compounds only displayed a moderate
inhibitory activity. The most active compounds in this
study were 6 and 9, showing IC50 values of 3.44 and
0.54 mM, respectively. In rat type 2 isozyme, the test
compounds showed a broad range of inhibitory poten-
cies from no inhibition for compounds 1 and 2 to strong
inhibition for the methyldicyclohexyl compound 7
(IC50=80 nM). In human type 1 isozyme (DU145 cell
line), compounds were inactive or rather poor inhibi-
tors. The most active one was the methyldicyclohexyl
compound 7, showing a 46% inhibition at a concentra-
tion of 10 mM. Most of the compounds which were
potent inhibitors of the rat isozyme 2 were potent inhi-
bitors of the human type 2 isozyme as well. Compounds
3, 6, 8 and 12 showed IC50 values between 100 and
530 nM. The most active derivative was the methyl-
dicyclohexyl compound 7, exhibiting an IC50 value of 60
nM. Taking a closer look at the structure±activity rela-
tionships it becomes apparent that the compounds
should be amides. Reduction of the carbonyl group of
compound 1 resulting in the amine 2 leads to a complete
loss of human type 2 enzyme inhibitory activity. There
seems to be a correlation between the number of C-
atoms of the substituent R and the activity of the
corresponding compounds. The 2-methylpropylidene
compound 11 shows almost no inhibition and the t-
butyl compound 10 is only a mediocre inhibitor. A
much better inhibitor is the cyclohexyl compound 4.
The introduction of one more C-atom increases activity
further (compounds 5 and 8). The fact that the 4-heptyl
compound 8 is more active than the methylcyclohexyl
compound 5 is probably due to the higher ¯exibility of
the open ring substituent. Compound 3 containing an
adamantyl (10 C-atoms) substituent is also a highly
active inhibitor. Aromatic groups are less appropriate
than aliphatic residues. The phenyl compound 1 shows
a lower activity than the cyclohexyl compound 4 and
the diphenylamino and the methyldiphenyl compounds
(9 and 6) are less active than the methyldicyclohexyl
compound 7. The high activity of the latter compound
is probably caused by the bulkiness and the ¯exibility of
the substituent. A further increase of the ¯exibility of 7
by reducing the double bond of the benzylidene moiety,
however, decreases inhibitory activity (compound 12).
Chemistry
For the preparation of compounds 1±12 (Scheme 2) the
corresponding N-acyl piperidones 1b±11b were prepared
from the corresponding acids 3c±11c as described for N-
benzoyl piperidone.21 After reaction with thionyl chlo-
ride, the acid chlorides were reacted with piperidone
hydrochloride in a solution of triethylamine (method
A). The phosphonium bromide 13 necessary for Wittig
reaction with 1b±11b was obtained by bromination of
methyl toluene-4-carboxylate (13b) with NBS using
benzoylperoxide22 and subsequent reaction of 13a with
triphenylphosphine.23 The N-acylpiperidine-4-benzyl-
idene compounds 1a±11a were obtained by Wittig
reaction of the acylated piperidones 1b±11b and the
phosphonium bromide 13 using BuLi/THF (method B).
The crude compounds were puri®ed by chromatography
Scheme 2. Synthetic strategy.

F. Picard et al. / Bioorg. Med. Chem. 8 (2000) 1479±1487
1481
Table 1. Inhibition of rat and human 5a-reductase type 1 and 2 in vitro by compounds 1±12 and ®nasteride
Compound
X
R
RVPa: % inhibition (10 mM) [IC50 (mM)]
Human: % inhibition (10 mM) [IC50 (mM)]
type 1c
type 2c
DU145b,c,e
BPHc,d
1
2
3
4
5
6
7
8
9
10
11
12
O
H2
O
O
O
O
O
O
O
O
O
O
Phenyl
Phenyl
1-Adamantyl
Cyclohexyl
n.i.
8
61 [5.6]
20
44
71 [3.44]
51
28
[0.54]
25
15
n.i.
n.i.
17
n.t.
44
23
31
26
46
20
22
20
12
35
[0.039]
53 [9.3]
n.i.
94 [0.26]
[1.65]
[1.02]
[0.53]
[0.06]
[0.26]
[0.83]
66 [2.5]
[1.50]
[1.02]
[0.37]
[0.08]
[0.29]
[0.69]
62
Methylcyclohexyl
Methyldiphenyl
Methyldicyclohexyl
4-Heptyl
Diphenylamino
t-Butyl
75 [3.10]
16
[0.10]
2-Methylpropylidene
Methyldicyclohexyl
37
[0.75]
[0.011]
n.t.
[0.01]
Finasteride
[2±3 nM]
aEnzyme of rat ventral prostate, 200±250 mg protein, substrate [1b,2b-3H] testosterone, 0.21 mM.
bSubstrate: [3H] androstenedione, 5 nM.
cMean value; tests have been run in duplicate. The standard deviation for IC50 is 20%, for percent inhibition is Æ10%.
dEnzyme from BPH tissue (type 2), 200±300 mg protein, substrate [1b,2b-3H] testosterone, 0.21 mM.
eProstatic tumor cell line expressing type 1 enzyme. n.i.: no inhibition; n.t.: not tested.
In vivo
1 mg/kg). With the exception of compound 12, all other
compounds showed a signi®cant inhibition (33±47%) of
the prostate weight increase. However, they did not
reach the e€ect of ®nasteride (85% inhibition).
As the most active inhibitors of the human enzyme in
vitro were highly active towards the rat isozyme 2 as
well, they were tested for their in vivo activity using
immature castrated SD rats. According to the procedure
of Hausler et al., the animals were applied testosterone
propionate.26 This leads to a dramatic stimulation of
prostate weights caused by the conversion of testoster-
one to DHT. 5a-Reductase inhibitors can inhibit this
stimulation. Table 2 shows the e€ects of compounds 3, 6,
7, 8 and 12 administered subcutaneously at doses between
9.2 and 11.4 mg/kg (ten times the dose of ®nasteride,
Discussion and Conclusion
The structure±activity study described in this paper
demonstrates that it is possible to increase human type 2
isozyme inhibitory activity of the indolyl benzoic acids20
by a factor of 100 (IC50 6 mM and 60 nM, respectively).
However, the structure modi®cations also decreased
isozyme 1 inhibition by a factor of 4±5. The low activity
of the present compounds toward human type 1 iso-
zyme might be due to the fact that the assay (DU145
cell line) is based on intact cells. It cannot be excluded
that the cell permeability is reduced in this class of car-
boxylic acids compared to the indolyl benzoic acids.
Further experiments regarding this issue are in progress.
Comparing enzyme inhibitory activity of the title com-
pounds toward the target isozymes, it is striking that the
best correlation exists between the human and rat type 2
isozyme. The correlation between type 1 and 2 isozymes
in the human and in the rat is not as good. This obser-
vation is in accordance with the ®nding that the per-
centage of identical amino acids in human and rat type
2 enzyme is higher than in the other groups.11,27 The
results of this study, showing that very bulky and ¯ex-
ible substituents in a position which corresponds to the
17b position of the steroidal substrate are optimal for
high activity, support the hypothesis of the existence of a
hydrophobic pocket in the active site proposed by oth-
ers.28 The most potent compounds in vitro were active
in vivo as well, i. e., the compounds reduced the prostate
weights in castrated testosterone treated rats. However,
the in vivo activity was not very strong, which might to
Table 2. In vivo activity of compounds 3, 6, 7, 8, 12 and ®nasteride
on ventral prostate weights in juvenile castrated rats treated with tes-
tosterone propionate
Test group
In vivo
Doses
E€ect
% Inhibitionc
MVÆSDa (mg/kg)
Vehicle (not castrated)
Vehicle (castrated)
56.6Æ8.6
22.9Æ13.1
Testosterone propionate (tp) 78.6Æ25.4
1
tp+6b
60Æ10.5
54.9Æ15.7
56.7Æ12
11.0
11.3
9.2
11.4
1
33 (P<0.025)
42 (P<0.025)
39 (P<0.025)
25 (n.s.)
tp+7b
tp+8b
tp+12b
tp+®nasteride
64.6Æ15.5
31.1Æ9.5
85 (P<0.01)
Vehicle (castrated)
Testosterone propionate (tp) 61.4Æ9.6
17.6Æ3.7
1
10.1
1
tp+3b
tp+®nasteride
41Æ6.2
47 (P<0.01)
86 (P<0.01)
31Æ9
amg Prostate/100 g body weight (mean valuesÆstandard deviation).
bAll compounds were applied in doses 10 times equimolar to ®nasteride,
1 mg/kg.
cSigni®cance according to U-test (Wilcoxon, Mann and Whitney).

1482
F. Picard et al. / Bioorg. Med. Chem. 8 (2000) 1479±1487
some extent be due to the high lipophilicity of the com-
pounds leading to a low concentration of free com-
pound in the plasma. Thus, the fact that compound 7,
which in vitro is more active than 8 (by a factor of 4), is
not more potent than the latter compound in vivo might
be explained by its stronger lipophilicity.
N-(Diphenyl)acetyl-4-piperidone (6b). Synthesized from
1
diphenylacetyl chloride. H NMR (400 MHz, CDCl3):
d=1.61 (s, 1H, -COCH-), 2.0 and 2.14 (2t, 4H, broad,
3
3J=6 Hz, pip. H), 3.76 and 3.94 (2t, 4H, broad, J=6
Hz, pip. H), 7.25 (m, 10H, aromat. H); mp: 135 ꢀC,
yield: 46%, C19H19NO2 (293.36).
As a conclusion, we have presented in this paper an
interesting new class of nonsteroidal 5a-reductase inhi-
bitors which might lead to the development of highly
potent nonsteroidal dual inhibitors of 5a-reductase iso-
zymes 1 and 2.
N-(Dicyclohexyl)acetyl-4-piperidone (7b). Synthesized
1
from dicyclohexylacetyl chloride. H NMR (400 MHz,
CDCl3): d=0.94±1.16 (m, 11H, cyclohexane H), 1.56 (s,
1H, -COCH-), 1.69 (s, 11H, broad, cyclohexane H), 2.46
3
(t, 4H, J=6 Hz, pip. H), 3.86 and 3.93 (2t, 4H, broad,
3J=6 Hz, pip. H); mp: 168 ꢀC, yield: 68%, C19H31NO2
(305.45).
Experimental
N-(2-Propyl)pentanoyl-4-piperidone (8b). Synthesized
1
1H NMR spectra were measured on an AM 400 (400
MHz) in DMSO-d6 or CDCl3. Chemical shifts are
reported as d values (ppm) relative to internal tetra-
methylsilane (d=0 ppm). IR spectra were performed
with KBr ®lm on a Perkin±Elmer PE 398 spectrometer.
Melting points were determined on a Ko¯er melting
point apparatus Thermophan (Reichert). Column chro-
matography was performed on Merck silica gel 60 (40±
63 mm). All reactions were followed by thin layer chro-
matography using Alugram silica gel 60. Chemicals and
solvents used were commercially available and were
used without further puri®cation.
from 2-propylpentanoic acid chloride. H NMR (400
MHz, CDCl3): d=0.88±1.43 (m, 14H, propylpentane
H), 2.76 (s, 1H, -COCH-), 2.47 (s, 4H, pip. H), 3.85 and
3
3.93 (2t, 4H, broad, J=6 Hz, pip. H). Oil, yield: 75%,
C13H23NO2 (225.33).
N-(Diphenyl)carbamoyl-4-piperidone (9b). Synthesized
from diphenylcarbamoyl chloride. 1H NMR (400 MHz,
CDCl3): d=2.31 (t, 3J=6.2 Hz, 4H, pip. H), 3.64 (t, 4H,
3J=6.2 Hz, pip. H), 7.08 (d, 4H, broad, aromat. H),
7.16 (t, 2H, 3J=7.5 Hz, aromat. H), 7.33 (t, 4H, 3J=7.9
Hz, aromat. H); mp: 141±143 ꢀC, yield: 61%, C18H18
N2O2 (294.35).
Method A
N-(Trimethyl)acetyl-4-piperidone (10b). Synthesized from
1
N-Cyclohexanoyl-4-piperidone (4b). A mixture of cyclo-
hexane carboxylic acid 4c (13 g, 0.1 mol), one drop of
dimethyl formamide and thionyl chloride (50 mL) was
re¯uxed at 40 ꢀC for 2 h. After this, thionyl chloride
was removed by distillation. The crude acid chloride
was dissolved in 20 mL of CH2Cl2 and was added
dropwise to a suspension of 4-piperidone monohydrate
hydrochloride (15 g, 9.7 mmol) and dry triethylamine
(40 mL, 0.3 mol) in dry CH2Cl2 (170 mL). The solution
was stirred for 2 h. The organic phase was washed with
water (2Â20 mL) and dried over magnesium sulfate.
After ®ltration, the solvent was evaporated in vacuo
and the crude product was puri®ed by recrystallization
in n-hexane/ethyl acetate to yield 4b (10.45 g, 53%): mp
trimethylacetyl chloride. H NMR (400 MHz, CDCl3):
3
d=1.33 (s, 9H, -C(CH3)3), 2.46 (t, 4H, J=6.2 Hz, pip.
H), 3.89 (t, 4H, 3J=6.2 Hz, pip. H); mp: 87±89 ꢀC,
yield: 60%, C10H17NO2 (183.24).
N-(3,3-Dimethyl)acryloyl-4-piperidone (11b). Synthesized
1
from 3,3-dimethyl acrylic acid chloride. H NMR (400
MHz, CDCl3): d=1.84±1.90 (m, 6H, -C(CH3)2), 2.09 (s,
1H, -COCH-), 2.41 (t, 4H, pip. H), 3.61 (m, 4H, pip. H).
Oil, yield: 65%, C10H15NO2 (181.23).
4-(Carboxymethyl)benzyl bromide (13a). A mixture of
methyl p-toluate 13b (10 g, 0.066 mol), NBS (13 g, 0.073
mol) and benzoylperoxide (0.6 g, 2.4 mmol) in dry CCl4
(100 mL) was re¯uxed for 3 h at 100 ꢀC. After cooling
the suspension was ®ltered. The organic phase was
washed twice with water (50 mL) and dried over mag-
nesium sulfate. The solvent was evaporated in vacuo to
yield 11.64 g (77%) 13a. No further puri®cation was
necessary: mp 54 ꢀC. 1H NMR (400 MHz, CDCl3):
d=3.89 (s, 3H, -COOCH3), 4.49 (s, 2H, Ar-CH2-Br),
7.44 and 8.02 (AA0BB0, 4H, 3J=8 Hz, aromat. H).
C9H9O2Br (229.07).
ꢀ
1
57±58 C. H NMR (400 MHz, CDCl3): d=1.18±1.94
(m, 11H, cyclohexane H), 2.49±2.59 (m, 4H, pip. H),
3.80±3.88 (m, 4H, pip. H). C12H19NO2 (209.28). Com-
pounds 1b and 2b were obtained commercially.
N-(1-Adamantanoyl)-4-piperidone (3b). Synthesized from
1
1-adamantane carboxylic acid chloride. H NMR (80
MHz, CDCl3): d=1.70 (s, 6H, ada. H), 1.90 (s, 3H, ada.
H), 2.0 (s, 6H, ada. H), 2.42 (t, 3J=6 Hz, 4H, pip.
H), 3.90 (t, 3J=6 Hz, 4H, pip. H). Oil, yield: 64%,
C16H23NO2 (261.36).
4-(Carboxymethyl)benzyltriphenylphosphonium bromide
(13). A mixture of 13a (5 g, 22 mmol) and triphenyl-
phosphine (5.7 g, 22 mmol) in dry toluene (165 mL) was
re¯uxed for 5 h at 120 ꢀC. The clear solution became
quickly turbid due to precipitation of salt. The solution
was ®ltered to yield 8.1 g (75%) of compꢀound 13. No
further puri®cation was necessary: mp 238 C. 1H NMR
(400 MHz, CDCl3): d=3.86 (s, 3H, -COOCH3), 5.67 (d,
N-(Cyclohexyl)acetyl-4-piperidone (5b). Synthesized from
cyclohexylacetyl chloride. 1H NMR (400 MHz, CDCl3):
d=0.94±1.79 (m, 11H, cyclohexane H), 2.17 (s, 2H,
-COCH2-), 2.26 and 2.49 (2s, 4H, pip. H), 3.46±3.70 (t,
4H, broad, J=6.2 Hz, pip. H); mp: 50±52 C, yield:
31%, C13H21NO2 (223.31).
ꢀ
3

F. Picard et al. / Bioorg. Med. Chem. 8 (2000) 1479±1487
1483
2H, 2J(H,P)=16 Hz, CH2-PPh+3 Br ), 7.59 and 7.76
(AA0BB0, 4H, 3J=8 Hz, aromat. H), 7.26±7.76 (m, 15H,
PPh+3 Br ). C27H24O2P Br (491.36).
broad, J=6.0 Hz, pip. H), 3.91 (s, 3H, -COOCH3),
6.41 (s, 1H, vinyl H), 7.24 and 7.98 (AA0BB0, 4H,
3J=8.12 Hz, aromat. H); mp: 65 ꢀC, yield: 17%,
C22H29NO3 (355.47).
3
Method B
N-(Diphenyl)acetyl-4-[4-(carboxymethyl)benzylidene]-
piperidine (6a). Synthesized from N-(diphenyl)acetyl-4-
piperidone (6b). Puri®ed by ¯ash chromatography using
n-hexane/ethyl acetate 7/3. 1H NMR (400 MHz,
CDCl3): d=1.58 (s, 1H, -COCH-), 2.04±2.48 (m, 4H, pip.
H), 3.44±3.77 (m 4H, pip. H), 3.93 (s, 3H, -COOCH3),
6.37 (s, 1H, vinyl H), 07.26±7.32 (m, 10H, aromat. H),
N-Cyclohexanoyl-4-[4-(carboxymethyl)benzylidene]piper-
idine (4a). To a suspension of 4-(carboxymethyl)benzyl-
triphenylphosphonium bromide (13, 4 g, 8.14 mmol) in
dry THF (40 mL) under nitrogen was added dropwise a
solution of butyllithium (1.6 M in n-hexane, 5.6 mL,
8.14 mmol). After 15 min the solution had changed to
orange and a solution of N-cyclohexanoyl-4-piperidone
(4b, 1.7 g, 8.14 mmol) in dry THF (30 mL) was added
dropwise at room temperature. The solution was stirred
overnight under nitrogen. The solvent was evaporated
in vacuo, and the residue was dissolved in CH2Cl2 (75
mL) and was washed twice with water (20 mL). The
solution was dried over magnesium sulfate. After ®ltra-
tion, the solvent was evaporated in vacuo. The crude
product was chromatographed using hexane/ethyl ace-
tate 1/1 as eluent to yield 388 mg of 4a (14%): mp 88±
0
3
7.30 and 7.96 (AA BB , 4H, J=7.96 Hz, aromat. H);
mp: 135 ꢀC, yield: 13%, C28H27NO3 (425.52).
N-(Dicyclohexyl)acetyl-4-[4-(carboxymethyl)benzylidene]-
piperidine (7a). Synthesized from N-(dicyclohexyl)-
acetyl-4-piperidone (7b). Puri®ed by ¯ash chromato-
1
graphy using n-hexane/ethyl acetate 8/2. H NMR (400
MHz, CDCl3): d=0.95±1.93 (m, 22H, cyclohexane H),
2.00 (s, 1H, -COCH-), 2.50 (m, 4H, pip. H), 3.44±3.75
(m, 4H, pip. H), 3.91 (s, 3H, -COOCH3), 6.40 (s, 1H,
vinyl H), 7.16±7.96 (m, 4H, 1,4 aromat. H): mp: 131 ꢀC,
yield: 38%, C28H39NO3 (437.62).
ꢀ
1
90 C. H NMR (400 MHz, CDCl3): d=1.25±1.80 (m,
11H, cyclohexane H), 2.39±2.51 (d, 4H, 3J=5.8 Hz, pip.
H), 3.34±3.59 (s, 4H, broad, pip. H), 3.91 (s, 3H,
-COOCH3), 6.42 (s, 1H, vinyl H), 7.26 and 8.0 (AA0BB0,
N-(2-Propyl)pentanoyl-4-[4-(carboxymethyl)benzylidene]-
piperidine (8a). Synthesized from N-(2-propyl)penta-
noyl-4-piperidone (8b). Puri®ed by ¯ash chromato-
3
4H, J=8.4 Hz, aromat. H). C21H27NO3 (341.22).
1
N-Benzoyl-4-[4-(carboxymethyl)benzylidene]piperidine (1a).
Synthesized from N-benzoyl-4-piperidone (1b). Chro-
matography was performed using benzin A/ethyl ace-
graphy using n-hexane/ethyl acetate 7/3. H NMR (400
MHz, CDCl3): d=0.87±1.43 (m, 14H, heptane H), 2.04
(s, 1H, -COCH-), 2.39±2.50 (m, 4H, pip. H), 3.72 (s, 4H,
broad, pip. H), 3.91 (s, 3H, -COOCH3), 6.41 (s, 1H,
1
tate 1/1. H NMR (400 MHz, CDCl3): d=2.50 (m, 4H,
0
0
3
pip. H), 3.41±3.80 (m, 4H, pip. H), 3.91 (s, 3H,
-COOCH3), 6.43 (s, 1H, vinyl H), 7.26 and 7.99 (2s, 4H,
1,4-disubst. aromat.), 7.42 (s, 5H, aromat. H); mp: 105±
106 ꢀC, yield: 41%, C21H21NO3 (335.40).
vinyl H), 7.24 and 8.0 (AA BB , 4H, J=8.4 Hz, aro-
mat. H). Oil, yield: 33%, C22H31NO3 (357.49).
N-(Diphenyl)carbamoyl-4-[4-(carboxymethyl)benzylidene]-
piperidine (9a). Synthesized from N-(diphenyl)carba-
moyl-4-piperidone (9b). Puri®ed by ¯ash chromato-
N-Benzyl-4-[4-(carboxymethyl)benzylidene]piperidine (2a).
Synthesized from N-benzyl-4-piperidone (2b). Chroma-
tography was performed using benzin A/ethyl acetate
1/1. 1H NMR (80 MHz, CDCl3): d=2.4 (s, 8H, pip. H),
3.5 (s, 2H, aromat-CH2-), 3.9 (s, 3H, -COOCH3), 6.15
(s, 1H, vinyl H), 7.1 (m, 7H, aromat. H, 2H of AA0BB0
systems and 5H aromat-CH2), 7.9 (d, 2H, 3J=8.2 Hz, of
AA0BB0 systems). Oil, yield: 31%, C21H23NO2 (321.41).
1
graphy with n-hexane/ethyl acetate 7/3. H NMR (400
3
MHz, CDCl3): d=2.26 and 2.35 (2t, 4H, J=6 Hz, pip.
H), 3.36 and 3.47 (2t, 4H, 3J=6 Hz, pip. H), 3.89 (s, 3H,
3
-COOCH3), 6.32 (s, 1H, vinyl H), 7.05 (d, 4H, J=6.7
3
Hz, aromat. H), 7.12 (t, 2H, J=7.5 Hz, aromat. H),
7.30 (t, 4H, 3J=7.5 Hz, aromat. H), 7.18 and 7.94
(AA0BB0, 4H, J=8.2 Hz, aromat. H). Whitish paste,
3
yield: 38%, C27H26N2O3 (426.51).
N-(1-Adamantanoyl)-4-[4-(carboxymethyl)benzylidene]-
piperidine (3a). Synthesized from N-(1-adamantanoyl)-
4-piperidone (3b). Chromatography was performed
N-(Trimethyl)acetyl-4-[4-(carboxymethyl)benzylidene]-
piperidine (10a). Synthesized from N-(trimethyl)acetyl-
4-piperidone (10b). Chromatography was performed
1
using n-hexane/ethyl acetate 1/1 . H NMR (400 MHz,
CDCl3): d=1.73 (s, 6H, ada. H), 2.03 (s, 6H, ada. H),
1
using n-hexane/ethyl acetate 1/1. H NMR (400 MHz,
CDCl3): d=1.30 (s, 9H, -C(CH3)3), 2.39 and 2.50 (2t,
4H, J=6 Hz, pip. H), 3.60 and 3.71 (2t, 4H, J=6 Hz,
2.05 (s, 3H, ada. H), 2.39 and 2.50 (2t, 4H, 3J=5.52 Hz,
3
3
3
pip. H), 3.65±3.74 (2t, 4H, broad, J=5.52 Hz, pip. H),
3.91 (s, 3H, -COOCH3), 6.39 (s, 1H, vinyl H), 7.25 and
7.99 (AA0BB0, 4H, 3J=8.12 Hz, aromat. H); mp:
147 ꢀC, yield: 60%, C25H31NO3 (393.52).
pip. H), 3.91 (s, 3H, -COOCH3), 6.39 (s, 1H, vinyl H),
7.24 and 7.97 (AA0BB0, 4H, J=8.4 Hz, aromat. H);
3
mp: 129±131 ꢀC, yield: 10%, C19H25NO3 (315.41).
N-(Cyclohexyl)acetyl-4-[4-(carboxymethyl)benzylidene]-
piperidine (5a). Synthesized from N-(cyclohexyl)acetyl-
4-piperidone (5b). Chromatography was performed
using n-hexane/ethyl acetate 1/1. H NMR (400 MHz,
CDCl3): d=0.93±1.78 (m, 11H, cyclohexane H), 2.17 (s,
2H, -COCH2-), 2.24±2.49 (m, 4H, pip. H), 3.58 (t, 4H,
N-(3,3-Dimethyl)acryloyl-4-[4-(carboxymethyl)benzyl-
idene]piperidine (11a). Synthesized from N-(3,3-dimeth-
yl)acryloyl-4-piperidone (11b). Puri®ed by ¯ash chro-
matography with n-hexane/ethyl acetate 7/3. H NMR
(400 MHz, CDCl3): d=1.86 (m, 6H, -C(CH3)2), 2.09 (s,
1H, -COCH-), 2.39 (m, 4H, pip. H), 3.59 (m, 4H, pip.
1
1

1484
F. Picard et al. / Bioorg. Med. Chem. 8 (2000) 1479±1487
H), 3.91 (s, 3H, -COOCH3), 6.42 (s, 1H, vinyl H), 7.24
and 7.98 (AA0 BB0, 4H, J=7.9 Hz, aromat. H); mp:
1H, -COCH-), 2.39±2.50 (m, 4H, pip. H), 3.33±3.60 (m,
4H, pip. H), 6.41 (s, 1H, vinyl H), 07.22±7.31 (m, 10H,
3
110±112 ꢀC, yield: 11%, C19H23NO3 (313.39).
aromat. H), 7.21 and 7.89 (AA BB , 4H, J=7.96 Hz,
aromat. H), 12.85 (s, 1H, COOH). IR (KBr): 2960,
2860, 1695, 1645, 1610, 1430, 1410, 1300, 1285, 1210,
1185, 1070, 995, 880, 790, 750, 740, 700; mp: 228±
229 ꢀC, yield: 62%, Anal. C27H25NO3 (411.49).
0
3
Method C
N-Cyclohexanoyl-piperidine-4-(benzylidene-4-carboxylic
acid) (4). A mixture of N-cyclohexanoyl-4-[4-(carboxy-
methyl)benzylidene]piperidine (4a, 0.4 g, 1.16 mmol) and
potassium carbonate (0.7 g) in methanol/water 9/1 was
re¯uxed for 3 h at 90 ꢀC. The solution was stirred over-
night at room temperature. After this, the reaction was
acidi®ed with 1 N hydrochloric acid. The compound
was extracted with CH2Cl2 (3Â30 mL), washed with
water, and dried over magnesium sulfate. The solvent
was evaporated in vacuo to yield 322 mg (85%) of 4: mp
N-(Dicyclohexyl)acetyl-piperidine-4-(benzylidene-4-carb-
oxylic acid) (7). 1H NMR (400 MHz, DMSO-d6):
d=0.95±1.93 (m, 22H, cyclohexane H), 2.0 (s, 1H,
-COCH-), 2.50 (m, 4H, pip. H), 3.34±3.71 (m, 4H, pip.
H), 6.44 (s, 1H, vinyl H), 7.21 and 7.91 (AA0BB0, 4H,
3J=7.96 Hz, aromat. H), 12.86 (s, 1H, COOH). IR
(KBr): 2920, 2850, 1690, 1630, 1610, 1430, 1360, 1295,
1180, 995, 880, 790, 750, 710; mp: 167±168 ꢀC, yield:
37%, Anal. C27H37NO3 (423.59).
166 ꢀC. H NMR (400 MHz, DMSO-d6): d=1.16±1.70
1
(m, 11H, cyclohexane H), 2.29±2.50 (m, 4H, pip. H),
3.32±3.50 (m, 4H, pip. H), 6.44 (s, 1H, vinyl H), 7.34
N-(2-Propyl)pentanoyl-piperidine-4-(benzylidene-4-carb-
oxylic) acid (8). 1H NMR (400 MHz, DMSO-d6):
d=0.81±1.22 (m, 14H, heptane H), 2.29 (s, 1H, -COCH-),
2.39±2.50 (m, 4H, pip. H), 3.51±3.63 (m, 4H, pip. H),
6.45 (s, 1H, vinyl H), 7.34 and 7.91 (AA0BB0, 4H,
3J=7.96 Hz, aromat. H), 12.89 (s, 1H, COOH). IR
(KBr): 2960, 2940, 1680, 1640, 1470, 1420, 1360,
1310, 1290, 1260, 1200, 1185, 1110, 990, 880, 790, 755,
705; mp: 124±125 ꢀC, yield 48%, Anal. C21H29NO3
(343.46).
and 7.91 (AA0BB0, 4H, J=7.96 Hz, aromat. H), 12.86
3
(s, 1H, COOH). IR (KBr): 2940, 2860, 1680, 1645, 1610,
1430, 1355, 1315, 1290, 1260, 1205, 1180, 995, 795, 755,
710. Anal. C20H25NO3 (327.42).
N-Benzoyl-piperidine-4-(benzylidene-4-carboxylic acid) (1).
1H NMR (400 MHz, DMSO-d6): d=2.5 (m, 4H, pip.
H), 3.31±3.72 (m, 4H, pip. H), 6.47 (s, 1H, vinyl H), 7.36
and 7.90 (2s, 4H, 1,4-disubst. aromat), 7.46 (m, 5H,
aromat. H), 12.89 (s, 1H, COOH). IR (KBr): 3000,
2850, 1690, 1640, 1605, 1430, 1365, 1310, 1270, 1240,
1175, 1070, 990, 885, 790, 750, 730, 710; mp: 166 ꢀC,
yield: 74%. C20H19NO3 (321.37), MS (CI) 321.
N-(Diphenyl)carbamoyl-piperidine-4-(benzyl-4-carboxylic
1
acid) (9). H NMR (400 MHz, DMSO-d6): d=2.31 (s,
4H, pip. H), 3.37 (s, 4H, pip. H), 6.37 (s, 1H, vinyl H),
7.02 (d, 4H, 3J=7.5 Hz, aromat. H), 7.14 (t, 2H, 3J=7.5
3
N-Benzyl-piperidine-4-(benzylidene-4-carboxylic acid) (2).
1H NMR (400 MHz, DMSO-d6): d=2.5±3.2 (m, 8H,
pip. H), 4.30 (s, 2H, Ar-CH2-), 6.51 (s, 1H, vinyl H),
Hz, aromat. H), 7.34 (t, 4H, J=7.5 Hz, aromat. H),
7.30 and 7.88 (AA0BB0, 4H, 3J=8.1 Hz, aromat.), 12.89
(s, 1H, COOH). IR (KBr): 2980, 2810, 1690, 1650, 1610,
1500, 1410, 1300, 1280, 1240, 1180, 1070, 1020, 990,
900, 760, 700; mp: 189 ꢀC, yield: 66%, Anal.
C26H24N2O3 (412.48).
7.35 and 7.91 (AA0BB0, 4H, J=8.4 Hz), 7.45 (m, 3H,
3
aromat. H), 7.63 (m, 2H, aromat H), 11.36 (s, 1H,
COOH). IR (KBr): 3000, 2950, 1680, 1610, 1430, 1325,
1220, 1180, 1105, 950, 890, 755, 705: mp: 285 ꢀC, yield:
68%. C20H21NO2 (307.39); MS (CI) 307.4.
N-(Trimethyl)acetyl-4-piperidine-4-(benzyl-4-carboxylic
1
acid) (10). H NMR (400 MHz, DMSO-d6): d=1.21 (s,
N-Adamantanoyl-piperidine-4-(benzylidene-4-carboxylic
1
9H, -(CH3)3), 2.33 (s, 4H, pip. H), 3.52 (s, 4H, pip. H),
6.43 (s, 1H, vinyl H), 7.33 and 7.88 (AA0BB0, 4H,
3J=8.4 Hz, aromat. H), 12.92 (s, 1H, -COOH). IR
(KBr): 2980, 2800, 1670, 1600, 1410, 1290, 1160, 980,
880, 750, 700, 650; mp: 220±222 ꢀC, yield: 66%, Anal.
C18H23NO3 (301.38).
acid) (3). H NMR (400 MHz, DMSO-d6): d=1.69 (s,
6H, ada. H), 1.93 (s, 6H, ada. H), 1.99 (s, 3H, ada. H),
2.33 and 2.45 (2t, 4H, broad, pip. H), 3.59 and 3.65 (2t,
4H, broad, pip. H), 6.51 (s, 1H, vinyl H), 7.35 and 7.89
(AA0BB0, 4H, J=7.96 Hz, aromat. H), 12.85 (s, 1H,
3
COOH). IR (KBr): 3000, 2900, 1705, 1590, 1430, 1380,
1260, 1230, 1170, 1100, 1060, 990, 880, 770, 750; mp:
256 ꢀC, yield: 42%. Anal. C24H29NO3 (379.49).
N-(3,3-Dimethyl)acryloyl-4-piperidine-4-(benzyl-4-carb-
oxylic acid) (11). 1H NMR (400 MHz, DMSO-d6):
d=1.81 (s, 6H, -(CH3)2), 2.33 (s, 1H, -COCH-), 2.5 (m,
8H, pip. H), 6.45 (s, 1H, vinyl H), 7.34 and 7.89
(AA0BB0, 4H, 3J=7.9 Hz, aromat. H), 12.91 (s, 1H,
-COOH). IR (KBr): 2980, 2960, 2850, 1680, 1620, 1600,
1420, 1360, 1320, 1280, 1220, 1170, 1060, 990, 880, 750,
700; mp: 155±157 ꢀC, yield: 57%, Anal. C18H21NO3
(299.36).
N-(Cyclohexyl)acetyl-piperidine-4-(benzylidene-4-carb-
oxylic acid) (5). 1H NMR (400 MHz, DMSO-d6):
d=0.91±1.66 (m, 11H, cyclohexane H), 2.19 (s, 2H,
-COCH2-), 2.50±3.32 (m, 8H, pip. H), 6.44 (s, 1H, vinyl
H), 7.33 and 7.90 (AA0BB0, 4H, 3J=7.96 Hz, aromat. H),
12.89 (s, 1H, COOH). IR (KBr): 2950, 2870, 1720, 1610,
1460, 1360, 1310, 1230, 1180, 1080, 995, 885, 775, 765,
750; mp: 173 ꢀC, yield: 44%. Anal. C21H27NO3 (341.45).
N-(Dicyclohexyl)acetyl-piperidine-4-(benzyl-4-carboxylic
acid) (12). N-(Dicyclohexyl)-acetyl-piperidine-4-(benzyl-
idene-4-carboxylic acid) (7) (300 mg) dissolved in
methanol was subjected to a hydrogenation under
N-(Diphenyl)acetyl-piperidine-4-(benzylidene-4-carboxylic
1
acid) (6). H NMR (400 MHz, DMSO-d6): d=2.29 (s,

F. Picard et al. / Bioorg. Med. Chem. 8 (2000) 1479±1487
1485
atmospheric pressure for 2 h using Pd/C as a catalyst.
The solvent was extracted in vacuo, and the compound
was recrystallized from n-hexane/ethyl acetate to yield
Human type 1 inhibition: DU145-assay. Intact human
prostatic carcinoma DU145 cells were used as the
source of type 1 5a-reductase.25 The inhibitory potency
of the compounds was determined by monitoring the
conversion of the tritiated substrate androstenedione
(5 nM) to androstanedione during an incubation period
of 6 h. A day before the experiment DU145 cells were
seeded in a 24-multiwell-plate at a density of 200,000
cells/well and allowed to become adherent overnight.
Compounds to be tested were dissolved in DMSO and
5 mL of each were added to the cells in a ®nal volume of
0.5 mL complete medium. Inhibitors were screened at
a concentration of 10 mM. As control of conversion
(typically about 35% under these conditions) served
a triplicate of wells without inhibitors and as positive
control for inhibition ®nasteride (80, 60, 40, 20 nM)
was used. After the 6 h incubation period in 5% CO2
at 37 ꢀC the medium samples were extracted twice
with 1 mL diethylether and the steroids were separated
by HPLC. Results are expressed as amount of formed
androstanedione as percentage of control values.
271 mg (90%) of compound 12: mp: 190±192 ꢀC. H
1
NMR (400 MHz, DMSO-d6): d=0.89±1.59 (m, 22H,
cyclohexane H), 2.00 (s, 1H, -COCH), 2.34±2.57 (m,
8H, pip. H), 2.58 (s, 2H, -CH2-arom.), 7.14 and 7.86
(AA0BB0, 4H, J=7.96 Hz, aromat. H), 12.74 (s, 1H,
3
COOH). IR (KBr): 2920, 2850, 1690, 1630, 1610, 1450,
1430, 1310, 1290, 1230, 1175, 1110, 1070, 1020, 960,
860, 760, 700. Anal. C27H39NO3 (425.61).
Enzyme inhibition test
Reagents. [1,2-3H]Androstenedione (4-androstene-3,17-
dione, AD) and [1,2-3H]testosterone (17b-hydroxy-4-
androstene-3-one, T) were purchased from DuPont,
Bad Homburg, Germany.
Preparation of tissue. Rat prostatic enzyme was pre-
pared according to the method of Liang et al.24 with
slight modi®cations.29 Male rats were killed and pros-
tates were taken within 5 min and put in ice cold 0.9%
NaCl solution. All the following operations were per-
formed at 0±4 ꢀC. The prostates were dissected free from
fat and connective tissue, cut into pieces and weighed.
Per 1 g of tissue, 3 mL of 20 mM phosphate bu€er, pH
6.5, containing 0.32 mM sucrose and 1 mM DTT were
added. The tissue was homogenized by ten 10-s strokes
at 20,500 rpm of an ultraturax (IKA) in 60-s intervals,
®ltered through cheesecloth and centrifuged for 60 min
at 105,000 g. The pellet obtained was resuspended in
phosphate bu€er. The centrifugation was repeated, the
®nal pellet resuspended in a minimum volume of phos-
phate bu€er and stored in 300 mL portions at 70 ꢀC.
The 105,000 g pellet contains nuclei, mitochondria and
microsomes and is referred to as the enzyme prepara-
tion. The protein content was determined and was in the
range of 15±25 mg/mL. Human prostatic tissue from
BPH patients was processed in the same way using
citrate bu€er, pH 5.5.
HPLC procedure. The procedure was carried out29
similar to the method of Cook et al.30 The steroids were
dissolved in 50 mL methanol and 25 mL injected into the
computer-controlled HPLC system, which was checked
before using labelled reference controls. Radioactivity
was measured using a Berthold LB 506C monitor.
Using methanol/water (55/45, w/w) for T and DHT,
with a ¯ow of 0.4 mL/min and an additive ¯ow of
1.0 mL for scintillator, base-line-separation of T and
DHT was achieved within 20 min. For the steroids
androstenedione and dihydroandrostenedione we used
methanol/water (50/50, w/w).
Calculation procedure. The amount of DHT formed was
calculated (% DHT). The zero value was subtracted
from the control (cv) and inhibition (iv) values (cvcorr
and ivcorr). Inhibition (I) was calculated using the fol-
lowing equation: %I ˆ ꢁ1 ivcorr=cvcorr†100.
In vivo assay.26 One day after their arrival, 21 day old
male rats were castrated by scrotal incision under ether
anesthesia. All animals were fed commercially available
chow and housed in temperature controlled rooms with
lights on between 08:00 and 18:00. Rats were divided in
7 groups with 8 rats each. Oil solution of compounds
(or vehicle) and testosterone propionate were applied by
separate subcutaneous injections to the rats once daily
for 4 days at doses of 9.2±11.4 mg/kg for title compounds,
1 mg/kg for testosterone propionate and ®nasteride.
Twenty four h after the last application, the rats were
killed by CO2 inhalation, and the ventral prostates were
removed. The prostates were dissected free from fat and
connective tissue and weighed. The mean percentage of
inhibition of the T-induced hypertrophic response was
calculated according to the following equation:
% Inhibition=100  ꢁCt D†=ꢁCt Cc†, where Ct, Cc,
and D are the mean prostate weights of T-treated
control, castrated control and drug treated group,
respectively. Mean values and standard deviation were
calculated. For determination of signi®cance the U-test
according to Wilcoxon, Mann and Whitney was used.
Incubation procedure. The assay was performed as
described24 with modi®cations.29 All values were run in
duplicate. The incubation was carried out for 30 min at
37 ꢀC in a total volume of 250 mL. In the case of rat
enzyme preparation phosphate bu€er (40 mM, pH 6.6
for type 1 and pH 5.5 for type 2) and in the case of
human enzyme preparation citrate bu€er (40 mM, pH
5.5) was used. The incubation mixture contained
approximately 250 mg rat protein (125 mg human pro-
tein), 200 mM NADPH (human enzyme: 100 mM
NADPH), 0.21 mM T including 45 nCi [1,2-3H]T, and
2% DMSO with or without test compound (10 mM). In
case of exceeding 60% inhibition, three concentrations
were chosen for measurement of IC50 values. The reaction
was started by adding the prostatic enzyme preparation
and stopped by addition of 50 mL NaOH (10 M). The
steroids were extracted using 500 mL of diethylether.
The mixture was shaken for 10 min and centrifuged
10 min at 4000 rpm. The water layer was frozen and the
ether layer was decanted in fresh tubes and evaporated to
dryness.

1486
F. Picard et al. / Bioorg. Med. Chem. 8 (2000) 1479±1487
McNulty, A. M.; Neubauer; B. L.; Rocco, V. P.; Audia, J. E.
Acknowledgements
Synthesis and 5a-reductase inhibitory activity of 8-substituted
benzo[f]quinolinones derived from palladium mediated coup-
ling reactions. Bioorg. Med. Chem. Lett. 1998, 8, 395.
10. Ishibashi, K.; Nakajima, K.; Sugioka,. Y.; Sugiyama, M.;
Hamada, T.; Horikoshi, H.; Nishi, T. Synthesis of 2-phe-
nylbenzofuran derivatives as testosterone 5a-reductase inhi-
bitor. Chem. Pharm. Bull. 1999, 47, 226. Yoshida, K.;
Horikoshi, Y.; Eta, M.; Chikazawa, J.; Ogishima, M.;
Fukuda, Y.; Sato, H. Synthesis of benzanilide derivatives as
dual acting agents with a1-adrenoceptor antagonistic action
and steroid 5a-reductase inhibitory activity. Bioorg. Med.
Chem. Lett. 1998, 8, 2967.
Thanks are due to the Deutsche Forschungsgemein-
schaft (DFG) and to the Fonds der Chemischen Indus-
trie, who supported this work by grants. Franck Picard
is grateful to the Hermann-Schlosser-Stiftung for ®nan-
cial support. We thank Mrs Anja Palusczak for per-
forming the biological tests.
References and Notes
1. Andersson, S.; Russell, D. W. Structural and biochemical
properties of cloned and expressed human and rat steroid 5a-
reductase. Proc. Natl. Acad. Sci. USA 1990, 87, 3640.
2. Imperato-McGinley, J.; Guerrero, L.; Gautier, T.; Peterson,
R. E. Steroid 5a-reductase de®ciency in man: an inherited
form of male pseudohermaphroditism. Science 1974, 186, 1213.
3. Gormley, G. J. 5a-Reductase inhibitors in prostate cancer.
Endocrine-Related Cancer 1996, 3, 57.
4. Dallob, A. L.; Sadick, N. S.; Unger, W.; Lipert, S.; Geissler,
L. A.; Gregoire, S. L.; Nguyen, H. H.; Moore, E. C.; Tanaka,
W. K. The e€ect of ®nasteride, a 5 alpha-reductase inhibitor,
on scalp skin testosterone and dihydrotestosterone concentra-
tions in patients with male pattern baldness. J. Clin. Endocri-
nol. Metab. 1994, 79, 703.
11. Russell, D. W.; Berman, D. M.; Bryant, J. T.; Cala, K. M.;
Davis, D. L.; Landrum, C. P.; Prihoda, J. S.; Silver, R. I.;
Thigpen, A. E.; Wigley, W. C. The molecular genetics of ster-
oid 5a-reductases. Recent Prog. Horm. Res. 1994, 49, 275.
Russell, D. W.; Wilson, J. D. Steroid 5a-reductase: two genes/
two enzymes. Annu. Rev. Biochem. 1994, 63, 26.
12. Marberger, M. J. Long-term e€ects of ®nasteride in
patients with benign prostatic hyperplasia: a double-bind,
placebo-controlled, multicenter study. Urol. 1998, 51, 677.
13. Uygur, M. C.; Gur, E.; Arik, A. I.; Altud, U.; Erol, D.
Erectile dysfunction following treatments of benign prostatic
hyperplasia: a prospective study. Androl. 1998, 30, 5.
14. Schwartz, J. I.; Van Hecken, A.; De Schepper, P. J.; De
Lepeleire, I.; Lasseter, K. C.; Cooper Shamblen, E.; Winchell,
G. A.; Constanzer, M. L.; Chavez, C. M.; Wang, D. Z.; Ebel,
D. L.; Justice, S. J.; Gertz, B. J. E€ect of MK-386, a novel
inhibitor of type 1 5a-reductase, alone and in combination
with ®nasteride on serum dihydrotestosterone concentrations
in men. J. Clin. Endocrinol. Metab. 1996, 81, 2942.
15. Dutasteride. Drugs Fut. 1999, 24, 246.
5. Sera®ni, P.; Lobo, R. A. 5a-Reductase activity in idiopathic
hirsutism. Fertil. Steril. 1985, 43, 74.
6. Keetch, D. W.; Andriole, G. L. Medicinal therapy for
benign prostatic hyperplasia. Am. J. Roentgenol. 1995, 164, 11.
7. Barry, M. J.; Fowler, F. J. Jr.; Bin, L.; Oesterling, J. E. A
nationwide survey of practising urologists: current manage-
ment of benign prostatic hyperplasia and clinically localized
prostate cancer. J. Urol. 1997, 158, 488. Barry, M.; Roehrborn,
C. Management of benign prostatic hyperplasia. Ann. Rev.
Med. 1997, 158, 177.
16. Bramson, H. N.; Hermann, D.; Batchelor, K. W.; Lee, F.
W.; James, M. K.; Frye, S. V. Unique preclinical character-
istics of GG745, a potent dual inhibitor of 5AR. J. Pharmacol.
Exp. Ther. 1997, 282, 1496.
8. Sharp, M. J.; Fang, F. G. Ecient construction of 6-azas-
teroids: dual inhibitors of steroidal 5a-reductase. Bioorg. Med.
Chem. Lett. 1998, 8, 3291. Ling, Y. Z.; Li, J. S.; Kato, K.; Liu,
Y.; Wang, X.; Klus, G. T.; Marat, K.; Nnane, I. P.; Brodie, A.
M. Synthesis and in vitro activity of some epimeric 20 alpha-
hydroxy, 20-oxime and aziridine pregnene derivatives as inhi-
bitors of human 17 alpha-hydroxylase/C17,20-lyase and 5
alpha-reductase. Bioorg. Med. Chem. 1998, 6, 1683. Guarna,
A.; Occhiato, E. G.; Danza, G.; Conti, A.; Serio, M. 5a-
Reductase inhibitors, chemical and clinical models. Steroids
1998, 63, 355. Kenny, B.; Ballard, S.; Blagg, J.; Fox, D.
Pharmacological options in the treatment of benign prostatic
hyperplasia. J. Med. Chem. 1997, 40, 1293. Fei, X. S.; Tian,
W. S.; Chen, Q. Y. Synthesis of 4-tri¯uoromethylsteroids: a
novel class of steroid 5a-reductase inhibitors. Bioorg. Med.
Chem. Lett. 1997, 7, 3113. Frye, S. V. Inhibitors of 5a-
reductase. Curr. Pharm. Des. 1996, 2, 59. Hartmann, R. W.;
Reichert, M.; Gohring, S. Novel 5a-reductase inhibitors.
Synthesis and structure±activity studies of 5-substituted 1-
methyl-2-pyridones and 1-methyl-2-piperidones. Eur. J. Med.
Chem. 1994, 29, 807.
9. Nakakoshi, M.; Kimura, K. I.; Nakajima, N.; Yoshihama,
M. SNA-4606-1, a new member of elaiophylins with enzyme
inhibition activity against testosterone 5a-reductase. J. Anti-
biot. 1999, 52, 175. Guarna, A.; Occhiato, E. G.; Scarpi, D.;
Tsai, R.; Comerci, A.; Mancina, R.; Serio, M. Synthesis of
benzo[c]quinolizin-3-ones: selective non-steroidal inhibitor of
steroid 5a-reductase 1. Bioorg. Med. Chem. Lett. 1998, 8,
2871. Takami, H.; Kishibayashi, N.; Ishi, A.; Kumazawa, T.
Indole and benzimidazole derivatives as steroid 5a-reductase
inhibitors in the rat prostate. Bioorg. Med. Chem. 1998, 6,
2441. Smith, E. C. R.; McQuaid, L. A.; Goode, R. L.;
17. Hermann, D. J.; Davis, I. M.; Wilson, T. H. E€ects of
GI198745 (GG745), a novel 5a-reductase (5 AR) inhibitor on
dihydrotestosterone (DHT). Am. Soc. Clin. Pharm. Ther.
1996, 59, 162.
18. Hobbs, S.; Hermann, D. J.; Gabriel, T.; Wilson, B.; Morrill,
R.; Clark, R. V. Marked suppression of dihydrotestosterone
in men by a novel 5 alpha reductase inhibitor. Fertil. Steril.
1998, 70 (suppl. 3), 4555.
19. Sawada, K.; Okada, S.; Golden, P.; Kayakiri, N.; Sawada,
Y.; Hashimoto, M.; Tanaka, H. 4-(1-Benzoylindol-3-yl)bu-
tyric acids and FK143: novel nonsteroidal inhibitors of 5a-
reductase (II). Chem. Pharm. Bull. 1999, 47, 481.
20. Baston, E.; Hartmann, R. W. N-Substituted 4-(5-indo-
lyl)benzoic acids. Synthesis and evaluation of steroid 5a-
reductase type I and II inhibitory activity. Bioorg. Med. Chem.
Lett. 1999, 9, 1601.
21. Richter, P.; Wagner, G. Synthese antiproteolytisch wirk-
samer 3,5-bis(4-amidinobenzyl)- und 3,5-bis(4-amidinobenzy-
liden)piperidon-(4)-derivatives. Pharmazie 1980, 35, 75.
22. Gorins, G.; Kuhnert, L.; Johnson, C. R.; Marnett, L. J.
(Carboxyalkyl)benzyl propargyl ethers as selective inhibitors
of leukocyte-type 12-lipoxygenases. J. Med. Chem. 1996, 61,
4871.
23. Gust, D.; Moore, T. A.; Bensasson, R. V.; Mathis, P.;
Land, E. J.; Chachaty, C.; Moore, A. L.; Liddell, P. A.;
Nemeth, G. A. Stereodynamics of intramolecular triplet
energy transfer in carotenoporphyrins. J. Am. Chem. Soc.
1985, 107, 3631.
24. Liang, T.; Cascieri, M. A.; Cheung, A. H.; Reynolds, G.
F.; Rasmusson, G. H. Species di€erences in prostatic steroid 5
alpha-reductases of rat, dog, and human. Endocrinology 1985,
117, 571.

F. Picard et al. / Bioorg. Med. Chem. 8 (2000) 1479±1487
1487
25. Delos, S.; Iehle, C.; Martin, P. M.; Raynaud, J. P. Inhibi-
tion of the activity of `basic' 5 alpha-reductase (type 1) detec-
ted in DU 145 cells and expressed in insect cells. J. Steroid
Biochem. Molec. Biol. 1994, 48, 347. Kaefer, M.; Audia, J. E.;
Bruchoversusky, N.; Goode, R. L.; Hsiao, K. C.; Leibovitch,
I. Y.; Krushinski, J. H.; Lee, C.; Steidle, C. P.; Sutkowski, D.
M.; Neubauer, B. L. Characterization of type I 5 alpha-
reductase activity in DU145 human prostatic adenocarcinoma
cells. J. Steroid Biochem. Molec. Biol. 1996, 58, 195.
26. Hausler, A.; Allegrini, P. R.; Biollaz, M.; Batzl, C.;
Scheidegger, E.; Bhatnagar, A. S. CGP 53153: a new potent
inhibitor of 5 alpha-reductase. J. Steroid Biochem. Molec. Biol.
1996, 57, 187. Hirosumi, J.; Nakayama, O.; Chida, N.; Inami,
M.; Fagan, T.; Sawada, K.; Shigematsu, S.; Kojo, H.; Notsu,
Y.; Okuhara, M. FK143, a novel nonsteroidal inhibitor of
steroid 5 alpha-reductase: (2) in vivo e€ects on rat and dog
prostates. J. Steroid Biochem. Molec. Biol. 1995, 52, 365.
27. Levy, M. A.; Brandt, M.; Sheedy, K. M.; Holt, D. A.;
Heaslip, J. I.; Trill, J. J.; Ryan, P. J.; Morris, R. A.; Garrison,
L. M.; Bergsma, D. J. Cloning, expression and functional
characterization of type 1 and type 2 steroid 5a-reductases
from cynomolgus monkey: comparison with human and
rat isoforms. J. Steroid Biochem. Molec. Biol. 1995, 52, 307.
Normington, K.; Russell, D. W. Tissue distribution and
kinetic characteristics of rat steroid 5a-reductase isozymes.
Evidence for distinct physiological functions. J. Biol. Chem.
1992, 267, 19548.
28. Rasmusson, G. H.; Reynolds, G. F.; Steinberg, N. G.;
Walton, E.; Patel, G. F.; Liang, T.; Cascieri, M. A.; Cheung,
A. H.; Brooks, J. R.; Berman, C. Azasteroids: structure±
activity relationships for inhibition of 5 alpha-reductase and of
androgen receptor binding. J. Med. Chem. 1986, 29, 2298.
Frye, S. V.; Ha€ner, C. D.; Maloney, P. R.; Mook, R. A. Jr.;
Dorsey, G. F. Jr.; Hiner, R. N.; Batchelor, K. W.; Bramson, H.
N.; Stuart, J. D.; Schweiker, S. L.; Van Arnold, J.; Bickett, D.
M.; Moss, M. L.; Tiang, G.; Unwalla, R. J.; Lee, F. W.; Tippin,
T. K.; James, M. K.; Grizzle, M. K.; Long, J. E.; Schuster, S.
V. 6-Azasteroids: potent dual inhibitors of human type 1 and 2
steroid 5 alpha-reductase. J. Med. Chem. 1993, 36, 4313.
29. Hartmann, R. W.; Reichert, M.; Gohring, S. Novel 5a-
reductase inhibitors. Synthesis and structure±activity studies
of 5-substituted 1-methyl-2-pyridones and 1-methyl-2-piper-
idones. Eur. J. Med. Chem. 1994, 29, 807.
30. Cook, S. J.; Rawlings, N. C.; Kennedy, R. I. Quantitation
of six androgens by combined high performance liquid chro-
matography and radioimmunoassay. Steroids 1982, 40, 369.