Novel imidazolyl and triazolyl substituted biphenyl compounds: Synthesis and evaluation as nonsteroidal inhibitors of human 17α-Hydroxylase-C17, 20-lyase (P450 17)

Article abstract of DOI:10.1016/S0968-0896(00)00076-6

The synthesis of a new series of P450 17 inhibitors is described. The imidazol-1-yl compounds 5 showed strong inhibition of P450 17 rat and especially human enzyme, the most active compounds being 5ax, 5ay and 5bx with IC₅₀ values of 0.17, 0.24 and 0.25 μM, respectively (ketoconazole: 0.74 μM). The 1,2,4-triazol-1-yl compounds 6 were less active, while the 1,2,4-triazol-4-yl compounds 7 were inactive. The title compounds showed little inhibition of P450 arom. The most active P450 17 inhibitors 5ax and 5ay markedly decreased the testosterone plasma concentration of SD rats 2 h after application of 0.019 mmol/kg. After 6 h, 5ay still exhibited a strong effect. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(00)00076-6

Bioorganic & Medicinal Chemistry 8 (2000) 1245±1252
Novel Imidazolyl and Triazolyl Substituted Biphenyl Compounds:
Synthesis and Evaluation as Nonsteroidal Inhibitors of Human
17ꢀ-Hydroxylase-C17, 20-Lyase (P450 17)
Yan Zhuang, Bertil G. Wachall and Rolf W. Hartmann*
Fachrichtung 12.1 Pharmazeutische und Medizinische Chemie, Universitat des Saarlandes, PO Box 15 11 50,
È
D-66041 Saarbrucken, Germany
È
Received 17 March 1999; accepted 29 December 1999
AbstractÐThe synthesis of a new series of P450 17 inhibitors is described. The imidazol-1-yl compounds 5 showed strong inhibition
of P450 17 rat and especially human enzyme, the most active compounds being 5ax, 5ay and 5bx with IC50 values of 0.17, 0.24 and
0.25 mM, respectively (ketoconazole: 0.74 mM). The 1,2,4-triazol-1-yl compounds 6 were less active, while the 1,2,4-triazol-4-yl
compounds 7 were inactive. The title compounds showed little inhibition of P450 arom. The most active P450 17 inhibitors 5ax and
5ay markedly decreased the testosterone plasma concentration of SD rats 2 h after application of 0.019 mmol/kg. After 6 h, 5ay still
exhibited a strong e€ect. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
5ax±5az, 5bx±5bz, 6ax±6az, 6bx±6bz, 7ax±7az and 7bx±
7bz. Inhibition of P450 17 rat and human enzyme and
human P450 arom by the test compounds and the
in¯uence of selected compounds on testosterone plasma
concentration in rats are presented.
Inhibitors of enzymes involved in the key steps of
androgen synthesis are important for the treatment of
prostatic diseases, such as benign prostatic hyperplasia
(BPH) and prostatic cancer.1 4 17a-Hydroxylase-
C17,20-lyase (P450 17) produces androstenedione and
dehydroepiandrosterone (DHEA), the immediate pre-
cursors of testosterone, from progesterone and pregne-
nolone, in both testes and adrenals.5,6 We have
shown2,7,8 that a dihydro- or tetrahydronaphthalene is
appropriate to give Ð after linkage with an N contain-
ing heterocycle, like imidazole, pyrazine or pyridine Ð
highly potent inhibitors of P450 17 (e.g., I and II,
Chart 18). While the heterocyclic N is complexing the
heme iron of P450 17, the rest of the molecule interacts
with the apoprotein moiety. In the case of steroidal
compounds, it has been shown recently,9,10 that an imi-
dazole ring at the C-17 position leads to potent inhibi-
tors as well (e.g., III, Chart 110). Very recently we found
in the class of N-imidazolylmethyl-substituted biphenyls
highly potent inhibitors of P450 17 (like IV, Chart 111).
To further increase in vitro activity and in vivo ecacy
we describe in the present paper the synthesis of the
imidazolyl and triazolyl substituted biphenyl compounds
Chemistry
As starting materials for the synthesis of the imidazolyl
and triazolyl substituted biphenyl compounds 5, 6 and
7, 4-bromo-acetophenone 2a or 5-bromo-1-indanone 2b
and di€erent para-substituted benzeneboronic acids (1x,
1y, 1z) were used (Scheme 1). The two series of impor-
tant intermediates, the 4-phenyl-acetophenones 3ax±3az
and the 5-phenyl-1-indanones 3bx±3bz, were obtained
by ¯uoride-mediated boronic acid coupling reaction
using the palladium complex Pd(PPh3)4 as catalyst
(Method A).12 The title compounds, the imidazole sub-
stituted racemic biphenyls 5 were prepared by reaction
of 4 with CDI (1,10-carbonyldiimidazole) (Method C)13
and the triazole substituted racemic biphenyls 6 and 7
were produced by similar reaction of 4 with 1,10-carbo-
nyldi(1,2,4-triazole) (CDT; Method D). The alcohols 4
were obtained as racemates by reduction of 3 with
NaBH4 (Method B).13 The isomers 6 and 7 were
separated by silica gel column chromatography using
CH2Cl2:CH3OH (19:1) as eluent.
*Corresponding author. Tel.: +49-681-302-3424; fax: +49-681-302-
4386; e-mail: rwh@rz.uni-sb.de
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00076-6

1246
Y. Zhuang et al. / Bioorg. Med. Chem. 8 (2000) 1245±1252
Chart 1. Selected inhibitors of P450 17 (I±IV) and the title compounds.
Results and Discussion
indane and the open ring class of compounds. The
energy di€erence from the low energy conformation of
the open ring compounds (a series) to the conformation
which corresponds to the indane structure is approxi-
mately 3 kcal/mol. From the fact that the indane com-
pounds are less active than the corresponding open ring
compounds, it can be concluded that the cyclic con-
formation of the indane compounds (b series) is less
appropriate for optimal interaction with the enzyme.
The fact that the imidazolyl compounds (series 5) are
more active than the triazolyl compounds (series 6 and
7) might be explained by the di€erent atomic charges of
the nitrogens (Table 1) being responsible for the inter-
action with the heme iron.
Table 1 shows the IC50 values and the inhibitory
potencies of the compounds towards P450 17 relative to
ketoconazole (KTZ; rp values). P450 17 from rat and
human origin was used as it is known that there is no
correlation between inhibition data obtained from these
species.11 Testicular enzyme and progesterone as sub-
strate were employed using the procedure recently
described by us.14 The imidazolyl compounds showed
strong inhibitory activity toward the human enzyme
exhibiting IC50 values of 0.17 mM (5ax), 0.24 mM (5ay),
and 0.25 mM (5bx), respectively. They were clearly more
potent than the reference ketoconazole (0.74 mM). It is
striking that the indane compounds were less active
than the corresponding open ring compounds. Similar
results were observed in the rat enzyme test (5ax, 5ay
and 5bx with IC50 values of 1.2, 0.54 and 2.1 mM,
respectively, ketoconazole: 67 mM). The 1,2,4-triazol-1-
yl compounds 6 were less active, while the 1,2,4-triazol-
4-yl compounds 7 were inactive. Methoxy substitution
of the phenyl ring decreases activity in both series 5a
and 5b, while ¯uorine substitution leads to an increase
in the rat and to a decrease in the human enzyme.
As P450 arom plays an important role in androgen
metabolism it should not be inhibited by the title com-
pounds to avoid accumulation of testosterone. There-
fore the compounds were tested for inhibition of P450
arom (Table 1). Most imidazolyl compounds 5 inhibited
P450 arom only in high concentrations of approxi-
mately 20 mM. As observed with P450 17, the triazole
compounds 6 and 7 were even less active. The only
compounds clearly more potent than the reference
compound aminoglutethimide were the methoxy sub-
stituted phenylindanes 5bz and 7bz.
In order to ®nd an explanation for the results, mole-
cular modeling studies were performed using energy-
minimized conformations of the title compounds.
Superimposition of the unsubstituted compounds (x
series) is shown in Figure 1. It becomes apparent that
the orientation of the heterocycle is di€erent in the
The most active inhibitiors of P450 17 rat and human
enzyme compounds 5ax, 5ay were tested for reduction
of the plasma testosterone concentration in male Spra-
gue±Dawley rats (Table 2).

Y. Zhuang et al. / Bioorg. Med. Chem. 8 (2000) 1245±1252
1247
Scheme 1. Synthesis of the title compounds 5, 6 and 7. Method A: CsF, Pd(PPh3)4, DME, re¯ux 12 h; B: NaBH4, THF, CH3OH, rt 1 h; C: 1,10-
carbonyldiimidazole (CDI), THF, 80±90ꢀC, 3±4 days; D: 1,10-carbonyldi(1,2,4-triazole) (CDT), THF, 80±90 ꢀC, 1 day followed by 140±150ꢀC, 7 days.
The two compounds were clearly superior to the refer-
ence compound ketoconazole. Compound 5ax strongly
inhibited testosterone concentration 2 h after appli-
cation. As observed after ketoconazole, 5ax led to a
stimulation of the testosterone concentration above
control after 6 h. This phenomenon has already been
observed11,15 and is probably due to hypothalamic
pituitary feedback. In contrast to 5ax, 5ay reduced
testosterone concentration to castration level after 2 h
and still exhibits strong activity after 6 h.
has to be concluded that 5ay might be a new lead for the
development of a drug candidate for the treatment of
prostate cancer.
Experimental
Melting points were determined on a Ko¯er melting
point Thermopan apparatus and are uncorrected. IR
Spectra were recorded on a Perkin±Elmer 398 Infrared
spectrometer as KBr disks or ®lms as indicated. 1H
NMR spectra were measured on a Bruker AM-400
instrument (400 MHz). Chemical shifts are given in
parts per million, and TMS was used as an internal
standard for spectra obtained in CDCl3. All J values are
given in Hz. Purity was checked by GC±MS on HP
G1800A GCD system. Reagents and solvents were used
as obtained from commercial suppliers without further
puri®cation. Column chromatography was performed
using silica-gel 60 (50±200 mm), and reaction progress
was determined by TLC analysis on ALUGRAM1
SILG/UV254 (Macherey and Nagel). Boronic acids and
bromoaryls used as starting materials were commer-
cially obtained (Lancaster, Fluka, Acros).
Comparing the results of the present study with the data
obtained for the correspondingly H, F or OCH3 sub-
stituted imidazolylmethyl substituted biphenyls (Type
IV compounds, Chart 1),11 it becomes apparent that the
introduction of a methyl group into the methylene
spacer between imidazole and biphenyl moiety increases
P450 17 inhibitory activity in both species. Ring closure
leading to the corresponding phenylindanes, however,
reduces inhibition of the target enzymes. The most
active compound toward P450 17 rat enzyme (5ay)
showed a marked in vivo activity. It reduced plasma
testosterone concentration to a stronger extent than the
corresponding imidazolylmethyl compound.11 Thus, it

1248
Y. Zhuang et al. / Bioorg. Med. Chem. 8 (2000) 1245±1252
Table 1. Structures, atomic charges and inhibition of steroidogenic P450 enzymes (P450 17, P450 arom) by heterocyclic substituted biphenyls and
phenylindanes
Compound
no.
R
a.c.b
N1/N3/N4
P450 17 (rat)
P450 17 (human)
P450 arom (human)
% Inhibitionc IC50 [mM]d r.p.e % Inhibitiong IC50 [mM]d r.p.e % Inhibitionh IC50 [mM]d r.p.i
5ax
5ay
5az
5bx
5by
5bz
H
F
OCH3
H
F
OCH3
0.204
0.205
0.206
0.205
0.204
0.205
1.2
56
124
27
32
39
0.17
0.24
1.8
0.25
1.11
2.1
4.4
3.1
0.41
3.0
0.67
0.35
22
21
25
20
21
0.84
0.88
0.74
0.93
0.88
3.6
0.54
2.51
2.1
1.7
10
6.7
5.1
6ax
6ay
6az
6bx
6by
6bz
H
F
OCH3
H
F
OCH3
0.221
0.219
0.220
0.220
0.221
0.222
n.i.f
n.i.
48
37
30
35
30
11
18
21
22
25
26
11
6.1
42
24
1.6
2.8
41
26
27
0.71
7ax
7ay
7az
7bx
7by
7bz
H
F
OCH3
H
F
OCH3
0.144
0.144
0.144
0.142
0.141
0.142
n.i.
n.i.
n.i.
n.i.
n.i.
n.i.
n.i.
12
n.i.
n.i.
n.i.
n.i.
20
7.1
17
35
0.68
5.0
3.7
III
IV
KTZa
AGa
0.18
2.6
67
372
26
1
0.04
0.31
0.74
19
2.4
1
18.5
1
aKTZ: ketoconazole; AG: aminoglutethimide; III, IV: see Chart 1.
ba.c.: atomic charges; N3: for imidazolyl compound (series 5); N4: for 1,2,4-triazol-1-yl compounds (series 6); N1: for 1,2,4-triazol-4-yl compounds
(series 7).
cConcentration of inhibitors for rat enzyme: 125 mM, concentration of progesterone (substrate): 25 mM.
dConcentration of inhibitors required to give 50% inhibition. The given values are mean values of at least two experiments. The deviations were
within Æ10%.
er.p.: relative potency (against ketoconazole (KTZ), rp KTZ=1).
fn.i.: no inhibition: values ꢁ10%.
gConcentration of inhibitors for human enzyme: 2.5 mM, concentration of progesterone (substrate): 25 mM.
hHuman placental microsomes, concentration of inhibitors: 25 mM, concentration of testosterone (substrate): 2.5 mM, (1b,2b-3H)-testosterone: 0.225
mCi.
ir.p.: relative potency (against aminoglutethimide (AG): r.p. AG=1).
Method A. 5-Phenyl-1-indanone (3bx). To a stirred mixture
of benzeneboronic acid (1x, 0.25 g, 2.01 mmol), 5-bromo-
1-indanone (2b, 0.4 g, 1.90 mmol), and powdered CsF
(0.61 g, 4.0 mmol) in 20 mL of DME (dimethoxyethane)
was added 68 mg (58.8 mmol) of Pd(PPh3)4. Under
nitrogen the reaction mixture was re¯uxed until no more
starting material was detected by TLC. The reaction
mixture was worked up with 1 N HCl and extracted three
times with CH2Cl2. The organic layer was washed with
brine, dried over Na2SO4 and the solvent was removed
in vacuo. The crude product was puri®ed by silica gel
column chromatography using 4% EtOAc/petroleum
ether as eluent to give 0.38 g (91%) of 3bx as a white
solid: mp 103±105 ꢀC; IR 3080, 2950, 1710, 1610, 1580,
Hz, 2H, H2), 3.19±3.22 (t, J=6.2 Hz, 2H, H3), 7.40±7.43
(m, 1H, Ar0-H, H40), 7.47±7.50 (dd, J=8.4 Hz, 2H, Ar0-
H, H30, H50), 7.60±7.62 (d, J=7.96 Hz, 1H, Ar-H, H6),
7.63±7.65 (dd, J=8.4 Hz, 2H, Ar0-H, H20, H60), 7.68 (s,
1H, H4), 7.82±7.84 (d, J=7.96 Hz, 1H, Ar-H, H7).
5-(4-Fluorophenyl)-1-indanone (3by). Compound 1y
(0.56 g, 4.0 mmol) gave 0.72 g (80%) of 3by as white
crystals: mp 91±93 ꢀC; IR 3030, 2960, 1710, 1605, 1500
1
(vs)cm 1; H NMR (CDCl3) 2.74±2.77 (t, J=6.2 Hz,
2H, H2), 3.19±3.22 (t, J=6.2 Hz, 2H, H3), 7.15±7.19
(dd, J=8.84 Hz, 2H, Ar0-H, H20, H60), 7.54±7.57 (d,
J=7.96 Hz, 1H, Ar-H, H7), 7.58±7.62 (dd, J=8.84 Hz,
2H, Ar0-H, H30,H50), 7.63 (s, 1H, H4), 7.81±7.83 (d,
J=7.96 Hz, 1H, Ar-H, H6).
1
1510 (vs)cm 1; H NMR (CDCl3) 2.74±2.77 (t, J=6.2

Y. Zhuang et al. / Bioorg. Med. Chem. 8 (2000) 1245±1252
1249
mp 100±102 ꢀC; IR 3400, 3020, 2960, 2880, 1610, 1510,
1
1490 (vs)cm
;
1H NMR (CDCl3) 1.76 (s, 1H, OH),
1.96±2.05 (m, 1H, H2a), 2.50±2.59 (m, 1H, H2b), 2.84±
2.92 (m, 1H, H3a), 3.08±3.15 (m, 1H, H3b), 5.28±5.31
(t, J=5.96 Hz, 1H, H1), 7.09±7.13 (dd, J=8.84 Hz, 1H,
Ar0-H, H20, H60), 7.41±7.42 (m, 2H, Ar-H, H4, H6),
7.47±7.49 (d, J=8.4 Hz, 1H, Ar-H, H7), 7.51±7.54 (dd,
J=8.84 Hz, 2H, Ar0-H, H30, H50).
Compounds 4ax,20 4ay,17 4az17 and 4bz,19 prepared by
literature methods, provided spectral and analytical
data as described.17,19,20
Method C. 4-[1-(1H-Imidazolyl)ethyl]biphenyl (5ax).
Compound 4ax (178 mg, 0.9 mmol) and 300 mg (1.85
mmol) of CDI (1,10-carbonyldiimidazole) in dry THF
(20 mL) were stirred at 70 ꢀC for 4 days. Water was
added and the reaction mixture was extracted three
times with CH2Cl2 (50 mL). The organic layer was
washed with brine, dried over Na2SO4 and the solvent
was removed in vacuo. The crude product was puri®ed
by silica gel column chromatography (CH2Cl2:CH3OH:
19:1) and the resulting solid was recrystallized from n-
hexane/ethyl acetate to give 130 mg (58%) of 5ax as a
Figure 1. Superimposition of the title compounds (x series): indane
compounds (b series, pink) show di€erent orientation of their hetero-
cycles compared to open ring compounds (a series, blue).
Table 2. Reduction of the plasma testosterone concentrations in
Sprague±Dawley rats by select compoundsa
Compound
Plasma testosterone level (ng/mL)
light white solid: mp 102±103 ꢀC. IR 3110, 3080, 3020,
2 h
6 h
1
2960, 2940, 1580, 1500, 1485 (vs)cm
;
1H NMR
Control
KTZ
5ax
4.54Æ1.58
1.69Æ0.61b
0.24Æ0.15b
0.028Æ0.046b
0.94Æ0.77
3.13Æ1.73b
4.11Æ1.94b
0.34Æ0.27c
(CDCl3) 1.90±1.91 (d, J=7.04 Hz, 3H, CH3), 5.38±5.43
(q, J=7.08 Hz, 1H, CH), 6.97 (s, 1H, imidazole-H5),
7.11 (s, 1H, imidazole-H4), 7.21±7.23 (d, J=8.4 Hz, 2H,
Ar-H, H3, H5), 7.34±7.38 (m, 1H, Ar0-H, H40), 7.42±
7.46 (m, 2H, Ar0-H, H20, H60), 7.55±7.58 (m, 4H, Ar-H,
H2, H6, Ar0-H, H30, H50), 7.64 (s, 1H, imidazole-H2);
GC±MS m/z (M+): 248.3 (calcd 248.3).
5ay
aAll compounds were applied equimolar to 10 mg/kg KTZ (0.019
mmol/kg) intraperitoneally. Blood was taken by cardiac puncture.
Testosterone plasma concentrations were determined by RIA. Each
group consisted of 7±8 animals. Signi®cant di€erence from control
group (Wilcoxon); bP<0.01; cP<0.05.
40-Fluoro-4-[1-(1H-imidazolyl)ethyl]biphenyl (5ay). Com-
pound 4ay (200 mg, 0.93 mmol) gave 142 mg (61%) of
5ay as a light white solid: mp 81±83 ꢀC. IR 3110, 3050,
2950, 1605, 1500 (vs)cm 1; 1H NMR (CDCl3) 1.89±1.91
(d, J=7.08 Hz, 3H, CH3), 5.37±5.42 (q, J=7.08 Hz, 1H,
CH), 6.97 (s, 1H, imidazole-H5), 7.10 (s, 1H, imidazole-
H4), 7.13±7.15 (d, J=8.84 Hz, 2H, Ar-H, H3, H5),
7.20±7.22 (d, J=8.4 Hz, 2H, Ar0-H, H20, H60), 7.50±
7.53 (m, 4H, Ar-H, H2, H6, Ar0-H, H30, H50), 7.63 (s,
1H, imidazole-H2); GC±MS m/z (M+): 266.3 (calcd
266.3).
Compounds 3ax,16 3ay,17 3az18 and 3bz,19 prepared by
literature methods, provided spectral and analytical
data as described.16 19
Method B. 5-Phenyl-1-indanol (4bx). A solution of
NaBH4 (120 mg, 3.17 mmol) in methanol (10 mL) was
added dropwise to a solution of 3bx (110 mg, 0.53
mmol) in THF (10 mL). The reaction mixture was
stirred at room temperature for 1 h. Water was added
and the solution was extracted three times with CH2Cl2.
The organic layer was washed with brine, dried over
Na2SO4 and the solvent was removed in vacuo. The
crude product was puri®ed by recrystallization from
diethyl ether and gave 92 mg (83%) of 4bx as light
40-Methoxy-4-[1-(1H-imidazolyl)ethyl]biphenyl (5az). Com-
pound 4az (175 mg, 0.77 mmol) gave 170 mg (80%) of
5az as a light white solid: mp 130±131 ꢀC. IR 3100,
1
3040, 2960, 1610, 1580, 1500 (vs)cm
;
1H NMR
(CDCl3) 1.89±1.90 (d, J=6.64 Hz, 3H, CH3), 3.85 (s,
3H, OCH3), 5.36±5.41 (q, J=6.64 Hz, 1H, CH), 6.96±
6.99 (m, 3H, Ar0-H, H30, H50, imidazole-H5), 7.10 (s,
1H, imidazole-H4), 7.18±7.20 (d, J=7.96 Hz, 2H, Ar-H,
H3, H5), 7.49±7.53 (dd, J=8.84 Hz, 4H, Ar-H, H2, H6,
Ar0-H, H20, H60), 7.63 (s, 1H, imidazole-H2); GC±MS
m/z (M+): 278.4 (calcd 278.4).
brown crystals: mp 105±107 ꢀC. IR 3400, 3020, 2960,
1
2870, 1615, 1600, 1570, 1480 (vs)cm
;
1H NMR
(CDCl3) 1.57 (s, 1H, OH), 1.97±2.05 (m, 1H, H2a),
2.50±2.59 (m, 1H, H2b), 2.84±2.92 (m, 1H, H3a), 3.09±
3.16 (m, 1H, H3b), 5.28±5.31 (t, J=5.32 Hz, 1H, H1),
7.32±7.36 (m, 1H, Ar0-H, H40), 7.41±7.43 (d, J=7.96
Hz, 1H, H6), 7.43±7.45 (d, J=8.4 Hz, 1H, Ar0-H, H20),
7.47±7.56 (m, 3H, Ar-H, H4, H7, Ar0-H, H60), 7.56±7.58
(m, 2H, Ar0-H, H30, H50).
1-(1H-Imidazolyl)-5-phenylindane (5bx). Compound 4bx
(80 mg, 0.38 mmol) gave 60 mg (60%) of 5bx as a light
green oil. IR 3110, 3040, 2970, 2850, 1610, 1570, 1490
1
5-(4-Fluorophenyl)-1-indanol (4by). Compound 3by (0.4 g,
1.17 mmol) gave 0.37 g (92%) of 4by as white crystals:
(vs)cm 1; H NMR (CDCl3) 2.22±2.31 (m, 1H, H2a),
2.72±2.81 (m, 1H, H2b), 3.0±3.08 (m, 1H, H3a), 3.16±3.23

1250
Y. Zhuang et al. / Bioorg. Med. Chem. 8 (2000) 1245±1252
(m, 1H, H3b), 5.69±5.73 (t, J=7.08 Hz, 1H, H1), 6.88
(s, 1H, imidazole-H5), 7.09 (s, 1H, imidazole-H4), 7.17±
7.19 (d, J=7.96 Hz, 1H, H7), 7.35±7.38 (m, 1H, Ar0-H,
H40), 7.43±7.47 (d, J=7.96 Hz, 2H, Ar0-H, H20, H60),
7.47 (s, 1H, Ar-H, H4), 7.55 (s, 1H, imidazole-H2),
7.57±7.60 (m, 3H, Ar-H, H6, Ar0-H, H30, H50); GC±MS
m/z (M+): 260.3 (calcd 260.3).
Ar-H, H2, H6), 7.56±7.58 (m, 2H, Ar0-H, H20, H60),
7.60±7.62 (d, J=8.4 Hz, 2H, Ar-H, H3, H5), 8.23 (s,
2H, triazole-H2, triazole-H5); GC±MS m/z (M+): 249.3
(calcd 249.3).
40-Fluoro-4-[1-(1H-tri-1,2,4-azolyl)ethyl]biphenyl (6ay) and
40-¯uoro-4-[1-(4H-tri-1,2,4-azolyl)ethyl]biphenyl (7ay).
Compound 4ay (360 mg, 1.67 mmol) gave 177 mg
(40%) of 6ay as a light white solid: mp 60±62 ꢀC. IR
1-(1H-Imidazolyl)-5-(4-¯uorophenyl)indane (5by). Com-
pound 4by (0.19 g, 0.81 mmol) gave 0.15 g (67%) of 5by
as a light brown oil. IR 3110, 3040, 2970, 2850, 1610,
1
3100, 3060, 2950, 2940, 1600, 1500 (vs)cm 1; H NMR
(CDCl3) 1.96±1.98 (d, J=7.08 Hz, 3H, CH3), 5.57±5.63
(q, J=7.08 Hz, 1H, CH), 7.10±7.12 (d, J=8.84 Hz, 2H,
Ar-H, H3, H5), 7.32±7.34 (d, J=8.4 Hz, 2H, Ar0-H,
H20, H60), 7.50±7.52 (d, J=8.84 Hz, 2H, Ar-H, H2, H6),
7.53±7.55 (d, J=8.4 Hz, 2H, Ar0-H, H30, H50), 7.99 (s,
1H, triazole-H5), 8.10 (s, 1H, triazole-H3); GC±MS m/z
(M+): 267.3 (calcd 267.3) and 101 mg (23%) of 7ay as
white crystals: mp 122±124 ꢀC. IR 3100, 3030, 2980,
1600, 1515, 1500 (vs)cm 1; 1H NMR (CDCl3) 1.96±1.98
(d, J=7.08 Hz, 3H, CH3), 5.49±5.54 (q, J=7.08 Hz, 1H,
CH), 7.12±7.16 (d, J=8.84 Hz, 2H, Ar-H, H3, H5),
7.26±7.28 (d, J=7.96 Hz, 2H, Ar0-H, H20, H60), 7.51±
7.54 (d, J=8.84 Hz, 2H, Ar-H, H2, H6), 7.56±7.58 (d,
J=7.96 Hz, 2H, Ar0-H, H30, H50), 8.34 (s, 2H, triazole-
H2, triazole-H5); GC±MS m/z (M+): 267.3 (calcd
267.3).
1
1520, 1490 (vs)cm 1; H NMR (CDCl3) 2.22±2.30 (m,
1H, H2a), 2.72±2.80 (m, 1H, H2b), 2.99±3.07 (m, 1H,
H3a), 3.15±3.22 (m, 1H, H3b), 5.69±5.72 (t, J=7.06 Hz,
1H, H1), 6.87 (s, 1H, imidazole-H5), 7.10 (s, 1H, Ar-H,
H4), 7.11±7.16 (dd, J=8.84 Hz, 2H, Ar0-H, H20, H60),
7.16±7.18 (d, J=7.96 Hz, 1H, Ar-H, H7), 7.39±7.41 (d,
J=7.96 Hz, 1H, Ar-H, H6), 7.50 (s, 1H, imidazole-H4),
7.52±7.55 (dd, J=8.84 Hz, 2H, Ar0-H, H30, H50), 7.57
(s, 1H, imidazole-H2);GC±MS m/z (M+): 278.3 (calcd
278.3).
1-(1H-Imidazolyl)-5-(4-methoxyphenyl)indane (5bz). Com-
pound 4bz (0.04 g, 0.17 mmol) gave 37 mg (77%) of 5bz
as white crystals: mp 130±131 ꢀC; IR 3120, 3040, 2970,
1
2850, 1610, 1590, 1525, 1490 (vs)cm
;
1H NMR
(CDCl3) 2.21±2.30 (m, 1H, H2a), 2.71±2.80 (m, 1H,
H2b), 3.0-3.08 (m, 1H, H3a), 3.14±3.22 (m, 1H, H3b),
3.86 (s, 3H, OCH3), 5.69±5.72 (t, J=7.08 Hz, 1H, H1),
6.88 (s, 1H, imidazole-H5), 6.98±7.00 (d, J=8.84 Hz,
2H, Ar0-H, H30, H50), 7.10 (s, 1H, imidazole-H4), 7.15±
7.17 (d, J=7.96 Hz, 1H, Ar-H, H6), 7.41±7.43 (d,
J=7.96 Hz, 1H, Ar-H, H7), 7.51 (s, 1H, Ar-H, H4),
7.55±7.54 (d, J=8.84 Hz, 2H, Ar0-H, H20, H60), 7.59 (s,
1H, imidazole-H2); GC±MS m/z (M+): 290.4 (calcd
290.4).
40-Methoxy-4-[1-(1H-tri-1,2,4-azolyl)ethyl]biphenyl (6az)
and 40-methoxy-4-[1-(4H-tri-1,2,4-azolyl)ethyl]biphenyl
(7az). Compound 4az (140 mg, 0.61 mmol) gave 77 mg
(45%) of 6az as a light white solid: mp 128±130 ꢀC. IR
1
3100, 3030, 2960, 2940, 1610, 1500 (vs)cm 1; H NMR
(CDCl3) 1.96±1.98 (d, J=7.08 Hz, 3H, CH3), 3.85 (s,
3H, OCH3), 5.57±5.62 (q, J=7.08 Hz, 1H, CH), 6.97±
6.99 (d, J=8.84 Hz, 2H, Ar0-H, H30, H50), 7.31±7.33 (d,
J=8.4 Hz, 2H, Ar-H, H2, H6), 7.50±7.52 (d, J=8.84
Hz, 2H, Ar0-H, H20, H60), 7.54±7.56 (d, J=8.4 Hz, 2H,
H3, H5), 7.99 (s, 1H, triazole-H5), 8.10 (s, 1H, triazole-
H3); GC±MS m/z (M+): 279.3 (calcd 279.3) and 38 mg
Method D. 4-[1-(1H-Tri-1,2,4-azolyl)ethyl]biphenyl (6ax)
and 4-[1-(4H-tri-1,2,4-azolyl)ethyl]biphenyl (7ax). Com-
pound 4ax (120 mg, 0.61 mmol) and 600 mg (3.7 mmol)
of 1,10-carbonyldi(1,2,4-triazole)(CDT) in dry THF (20
mL) were re¯uxed for 1 day. The solvent was removed
by distillation and the residue heated at 140±150 ꢀC for
7 days. Water was added and the reaction mixture was
extracted three times with CH2Cl2. The organic layer
was washed with brine, dried over Na2SO4 and the sol-
vent was removed in vacuo. The crude product was
puri®ed by silica gel column chromatography using
CH2Cl2:CH3OH (19:1) as eluent to give 75 mg (50%) of
(22%) of 7az as white crystals: mp 208±210 ꢀC. IR 3100,
1
3020, 2960, 1650, 1600, 1500 (vs)cm
;
1H NMR
(CDCl3) 1.93±1.95 (d, J=7.08 Hz, 3H, CH3), 3.85 (s,
3H, OCH3), 5.41±5.46 (q, J=7.08 Hz, 1H, CH), 6.97±
6.99 (d, J=8.84 Hz, 2H, Ar0-H, H30, H50), 7.21±7.23 (d,
J=8.4 Hz, 2H, Ar-H, H2, H6), 7.49±7.51 (d, J=8.84
Hz, 2H, Ar0-H, H20, H60), 7.55±7.57 (d, J=8.4 Hz, 2H,
H3, H5), 8.20 (s, 2H, triazole-H2, triazole-H5); GC±MS
m/z (M+): 279.3 (calcd 279.3).
6ax as a light white solid: mp 68±70 ꢀC. IR 3105, 3030,
5-Phenyl-1-(1H-tri-1,2,4-azolyl)indane (6bx) and 5-phen-
yl-1-(4H-tri-1,2,4-azolyl)indane (7bx). Compound 4bx
(400 mg, 1.9 mmol) gave 227 mg (46%) of 6bx as green
solid: mp 109±111 ꢀC. IR 3110, 3030, 2990, 2940, 2860,
1610, 1540, 1500 (vs)cm 1; 1H NMR (CDCl3) 2.45±2.53
(m, 1H, H2a), 2.73±2.82 (m, 1H, H2b), 3.04±3.11 (m,
1H, H3a), 3.21±3.28 (m, 1H, H3b), 5.95±5.98 (t, J=6.2
Hz, 1H, H1), 7.25±7.27 (d, J=7.96 Hz, 1H, H7), 7.35±
7.39 (m, 1H, Ar0-H, H40), 7.43±7.49 (m, 3H, Ar-H, H6,
Ar0-H, H30, H50), 7.57±7.60 (m, 3H, Ar-H, H4, Ar0-H,
H20, H60), 8.01 (s, 1H, triazole-H5), 8.05 (s, 1H, triazole-
H3); GC±MS m/z (M+): 261.3 (calcd 261.3) and 57 mg
1
2990, 2940, 1500 (versus) cm
;
1H NMR (CDCl3)
1.97±1.99 (d, J=7.08 Hz, 3H, CH3), 5.58±5.64 (q,
J=7.08 Hz, 1H, CH), 7.33±7.35 (d=7.96 Hz, 2H, Ar0-
H, H30, H50), 7.43±7.46 (m, 2H, Ar0-H, H40, triazole-
H5), 7.56±7.58 (d, J=8.4 Hz, 2H, Ar-H, H2, H6), 7.59±
7.61 (d, J=8.4 Hz, 2H, Ar0-H, H20, H60), 8.0±8.02 (d,
J=8.4 Hz, 2H, Ar-H, H3, H5), 8.11 (s, 1H, triazole-
H3); GC±MS m/z (M+): 249.3 (calcd 249.3) and 60 mg
(40%) of 7ax as white crystals: mp 66±68 ꢀC. IR 3100,
1
3015, 2990, 2940, 1530, 1485 (vs)cm
;
1H NMR
(CDCl3) 1.95±1.97 (d, J=7.08 Hz, 3H, CH3), 5.44±5.49
(q, J=7.08 Hz, 1H, CH), 7.25±7.27 (m, 2H, Ar0-H, H30,
H50), 7.36±7.40 (m, 1H, Ar0-H, H40), 7.44±7.48 (m, 2H,
(11%) of 7bx as a light green oil. IR 3120, 3040, 2980,
1
2960, 1630, 1610, 1560, 1520 (vs)cm
;
1H NMR

Y. Zhuang et al. / Bioorg. Med. Chem. 8 (2000) 1245±1252
1251
(CDCl3) 2.20±2.30 (m, 1H, H2a), 2.78±2.87 (m, 1H,
H2b), 3.05±3.12 (m, 1H, H3a), 3.18±3.26 (m, 1H, H3b),
5.78±5.81 (t, J=6.2 Hz, 1H, H1), 7.20±7.22 (d, J=7.96
Hz, 1H, H7), 7.36±7.41 (m, 1H, Ar0-H, H40), 7.44±7.51
(m, 3H, Ar-H, H6, Ar0-H, H30, H50), 7.58±7.60 (m, 3H,
Ar-H, H4, Ar0-H, H20, H60), 8.17 (s, 2H, triazole-H2,
triazole-H5); GC±MS m/z (M+): 261.3 (calcd 261.3).
Determination of plasma testosterone concentration.
Tests were performed with adult male Sprague±Dawley
rats (each group consisted of 7±8 animals). All com-
pounds were dissolved in 0.1 N HCl and administered
once i.p. equimolar to 10 mg/kg ketoconazole (0.019
mmol/kg). Blood samples were taken by cardiac punc-
ture under diethyl ether anesthesia after 2 and 6 h.
Plasma testosterone values were determined by com-
mercially available RIA and are given in ng/mL plas-
maÆstandard deviation.
5-(4-Fluorophenyl)-1-(1H-tri-1,2,4-azolyl)indane (6by) and
5-(4-¯uorophenyl)-1-(4H-tri-1,2,4-azolyl)indane (7by).
Compound 4by (235 mg, 1.0 mmol) gave 143 mg
(50%) of 6by as light white solid: mp 80±82 ꢀC; IR 3110,
Molecular modeling
1
3040, 2960, 1600, 1500 (vs)cm
;
1H NMR (CDCl3)
2.46±2.54 (m, 1H, H2a), 2.74±2.82 (m, 1H, H2b), 3.04±
3.11 (m, 1H, H3a), 3.21±3.29(m, 1H, H3b), 5.96±5.99 (t,
J=7.48 Hz, 1H, H1), 7.11±7.16 (m, 2H, H20, H60), 7.25±
7.27 (d, J=7.96 Hz, 1H, Ar-H, H7), 7.42±7.44 (d,
J=7.96 Hz, 1H, Ar-H, H6), 7.51±7.57 (m, 3H, Ar-H,
H4, Ar0-H, H30, H50), 8.04 (s, 1H, triazole-H5), 8.13 (s,
1H, triazole-H3); GC±MS m/z (M+): 279.3 (calcd
279.3) and 32 mg (11%) of 7by as white solid: mp 153±
155 ꢀC; IR 3120, 2980, 2960, 1600, 1580, 1530, 1490
HyperChem molecular modeling software22 was used to
build and minimize the structures of the title com-
pounds. Molecular mechanics calculations employed
the MM+ force ®eld23 and the structures were re®ned
using a conjugate gradient (Polak-Ribiere) method.
Conformations were optimized by molecular dynamics
simulations using the heating-cooling-heating proce-
dure.24 Each molecule was heated several times to 1000
K and then cooled down to 0 K (5, 10, 5 ps). Between
each cycle, the molecule was reminimized using the
MM+ force ®eld and the procedure was iterated until it
provided the minimum energy conformation. Molecular
Similarity ®tting operation was performed between two
molecules, matching the pairs of atoms, using energy-
minimized 5ax as a template and C1, C3, CH2 or C5,
C7, C1 as ®t atoms for biphenyl and indane compounds,
respectively (Fig. 1). The charges of compounds were
obtained, after minimization, by the semiempirical
CNDO approximation.25
1
(vs)cm 1; H NMR (CDCl3) 2.16±2.30 (m, 1H, H2a),
2.79±2.87 (m, 1H, H2b), 3.04±3.12 (m, 1H, H3a), 3.18±
3.26 (m, 1H, H3b), 5.79±5.80 (t, J=7.12 Hz, 1H, H1),
7.12±7.17 (m, 2H, H20, H60), 7.19±7.27 (d, J=7.96 Hz,
1H, Ar-H, H7), 7.44±7.46 (d, J=7.96 Hz, 1H, Ar-H,
H6), 7.53±7.56 (m, 3H, Ar-H, H4, Ar0-H, H30, H50),
8.18 (s, 2H, triazole-H2, triazole-H5); GC±MS m/z
(M+): 279.3 (calcd 279.3).
5-(4-Methoxyphenyl)-1-(1H-tri-1,2,4-azolyl)indane (6bz)
and 5-(4-methoxyphenyl)-1-(4H-tri-1,2,4-azolyl)indane
(7bz). Compound 4bz (200 mg, 0.83 mmol) gave 142
mg (58%) of 6bz as a light white solid: mp 134±136 ꢀC.
IR 3110, 3020, 2960, 2940, 1610, 1590, 1510, 1500
Acknowledgements
1
(vs)cm 1; H NMR (CDCl3) 2.43±2.51 (m, 1H, H2a),
Thanks are due to the Fonds der Chemischen Industrie,
who supported this work by a grant. We thank Dr.
Markus Hector for discussion, Dr. Joachim Schroder
for providing the GC±MS spectra, Mrs. Anja Palusczak
for performing the P450 17 and P450 arom inhibitory
assays, Mrs. Alice Kleber and Mr. Wolfgang Reichert
for their help regarding the in vivo study and Professor
Vincent Njar for providing compound III.
2.71±2.80 (m, 1H, H2b), 3.02±3.09 (m, 1H, H3a), 3.19±
3.27 (m, 1H, H3b), 3.86 (s, 3H, OCH3), 5.93±5.96 (m,
1H, H1), 6.97±7.00 (d, J=8.84 Hz, 2H, Ar0-H, H30,
H50), 7.23±7.25 (d, J=7.96 Hz, 1H, Ar-H, H6), 7.43±
7.45 (d, J=7.96 Hz, 1H, Ar-H, H7), 7.51±7.53 (m, 3H,
Ar-H, H4, Ar0-H, H20, H60), 8.00 (s, 1H, triazole-H3),
8.02 (s, 1H, triazole-H5); GC±MS m/z (M+): 291.4
(calcd 291.4) and 51 mg (21%) of 7bz as a light white
solid: mp 139±140.5 ꢀC. IR 3105, 3020, 2970, 2940,
1
1610, 1590, 1520, 1490 (vs)cm
;
1H NMR (CDCl3)
References and Notes
2.21±2.29 (m, 1H, H2a), 2.78±2.87 (m, 1H, H2b), 3.04±
3.11 (m, 1H, H3a), 3.17±3.25 (m, 1H, H3b), 3.86 (s, 3H,
OCH3), 5.80±5.83 (t, J=6.64 Hz, 1H, H1), 6.99±7.01 (d,
J=8.84 Hz, 2H, Ar0-H, H30, H50), 7.18±7.20 (d, J=7.96
Hz, 1H, Ar-H, H6), 7.45±7.47 (d, J=7.96 Hz, 1H, Ar-
H, H7), 7.52±7.54 (m, 3H, Ar-H, H4, Ar0-H, H20, H60),
8.22 (s, 2H, triazole-H2, triazole-H5); GC±MS m/z
(M+): 291.4 (calcd 291.4).
1. For review see: Njar, V. C. O.; Brodie, A. M. H. Curr.
Pharm. Design 1999, 5, 165.
2. Hartmann, R. W.; Wachter, G. A.; Sergejew, T. F.; Wurtz,
R.; Duerkop, J. Arch. Pharm. Pharm. Med. Chem. 1995, 328,
573.
3. De Coster, R.; Wouters, W.; Bruynseeles, J. J. Steroid Bio-
chem. Mol. Biol. 1999, 56, 133.
4. Jarman, M.; Smith, H. J.; Nicholls, P. J.; Simons, C. Nat.
Prod. Rep. 1998, 495.
Biological methods
5. Nakajin, S.; Hall, P. F. Biochemistry 1981, 20, 4037.
6. Hall, P. F. J. Steroid Biochem. Mol. Biol. 1991, 40, 527.
7. Wacher, G. A.; Hartmann, R. W.; Sergejew, T. F.; Grun, G.
L.; Ledergerber, D. J. Med. Chem. 1996, 39, 834.
8. Hartmann, R. W.; Frotscher, M.; Ledergerber, D.; Wach-
ter, G. A.; Grun, G. L.; Sergejew, T. F. Arch. Pharm. Pharm.
Med. Chem. 1996, 329, 251.
Enzyme preparations and assays. The enzymes were pre-
pared according to described methods: human and rat
testicular P450 1714 and human placental P450 arom.21
Enzyme assays were performed as described: human
P450 17,11 rat P450 1714 and P450 arom.21

1252
Y. Zhuang et al. / Bioorg. Med. Chem. 8 (2000) 1245±1252
9. Ling, Y.-Z.; Li, J.-S.; Liu, Y.; Kato, K.; Llus, G. T.; Brodie,
A. J. Med. Chem. 1997, 40, 3297.
10. Njar, V. C. O.; Kato, K.; Nnane, I. P.; Grigoryev, D. N.;
Long, B. J.; Brodie, A. M. H. J. Med. Chem. 1998, 41, 902.
11. Wachall, B. G.; Hector, M.; Zhuang, Y.; Hartmann, R. W.
Bioorg. Med. Chem. 1999, 7, 1913.
16. Olah, G. A.; Vankar, Y. D.; Fung, A. P. Synthesis 1979, 1, 59.
17. Dovgosheya, M. I.; Mil0ner, R. S. J. Gen. Chem. USSR
(Engl. Transl.) 1964, 34, 969.
18. Shunk, W. J. Am. Chem. Soc. 1950, 72, 2388.
19. Minssen-Guette, M.; Dvolaitzky, M.; Jacques, J. Bull. Soc.
Chim. Fr. 1968, 5, 2111.
12. Wright, S. W.; Hageman, D. L.; McClue, L. D. J. Org.
Chem. 1994, 59, 6095.
20. Hutchins, R. O.; Kandasamy, D. J. Am. Chem. Soc. 1973,
95, 6131.
13. Borgne, M. L.; Marchand, P.; Du¯os, M.; Delevoye-Seil-
ler, B.; Piessard-Kobert, S.; Le Baut, G. L.; Hartmann, R. W.;
Palzer, M. Arch. Pharm. Pharm. Med. Chem. 1997, 330, 141.
14. Sergejew, T. F.; Hartmann, R. W. J. Enz. Inhib. 1994, 8,
113.
15. Barrie, S. E.; Potter, G. A.; Goddard, P. M.; Haynes, B.
P.; Dowsett, M.; Jarman, M. J. Steroid Biochem. Mol. Biol.
1994, 50, 267.
21. Hartmann, R. W.; Batzl, C. J. Med. Chem. 1986, 29, 1362.
22. Molecular-modeling software HyperChemTM Version 5.02
(Hypercube Inc., Ontario, Canada).
23. Allinger, N. L. J. Am. Chem. Soc. 1997, 99, 8127.
24. van Gunsteren, W. F.; Berendsen, H. J. C. Angew. Chem.
(Engl. Transl.) 1990, 29, 992.
25. Pople, J. A.; Beveridge, D. L. Approximate Molecular
Orbital Theory; McGraw-Hill: New York, 1970.