Inhibition of aromatase (P450Arom) by some 1-(benzofuran-2-ylmethyl)imidazoles.

Article abstract of DOI:10.1211/0022357991772637

Studies of a series of 1-(benzofuran-2-ylmethyl)imidazoles, 1-5, previously proposed as potential agents for prostatic cancer by their inhibition of 17beta-hydroxylase:17,20-lyase (P450 17), have been extended to their selectivity against placental microsomal aromatase (P450(Arom)) in man. The compounds were 3-7-fold more potent than aminoglutethimide and had some selectivity for P450 17 as expressed by the ratio (IC50 P450(Arom))/(IC50 P450) 17)/17.0 (2), 10.3 (3), 34.6 (4) and 42.0 (5), where IC50 is the concentration resulting in 50% inhibition. The lower potency of 1-5 towards P450(Arom) compared with the racemic alpha-phenyl-substituted compounds (6, 80-1000 x aminoglutethimide) and some racemic alpha-methyl (8.5 and 12.2 x aminoglutethimide) and alpha-ethyl (12.1 and 32.9 x aminoglutethimide) analogues has been rationalized. This work selectively extends studies of the P450 17 inhibitor 5, a potential prostatic cancer agent, towards other cytochrome P450 enzymes in the steroidogenic pathway and provides a general method for determining the relative influence of chemical manipulation of a parent inhibitor towards two enzymes in the pathway using additional literature data.

Full text of DOI:10.1211/0022357991772637

J. Pharm. Pharmacol. 1999, 51: 427± 433
Received September 21, 1998
Accepted November 25, 1998
# 1999 J. Pharm. Pharmacol.
Inhibition of Aromatase (P450Arom) by some
-(Benzofuran-2-ylmethyl)imidazoles
1
CAROLINE P. OWEN, PAUL J. NICHOLLS, H. JOHN SMITH AND RHYS WHOMSLEY
Welsh School of Pharmacy, Cardiff University, Cathays Park, Cardiff CF1 3XF, UK
Abstract
Studies of a series of 1-(benzofuran-2-ylmethyl)imidazoles, 1±5, previously proposed as
potential agents for prostatic cancer by their inhibition of 17b-hydroxylase:17,20-lyase
(
(
P450 17), have been extended to their selectivity against placental microsomal aromatase
P450Arom) in man.
The compounds were 3±7-fold more potent than aminoglutethimide and had some
selectivity for P450 17 as expressed by the ratio (IC50 P450Arom)=(IC50 P450) 17)=17Á0
2), 10Á3 (3), 34Á6 (4) and 42Á0 (5), where IC50 is the concentration resulting in 50%
inhibition. The lower potency of 1±5 towards P450 compared with the racemic a-
(
Arom
phenyl-substituted compounds (6, 80±1000 6 aminoglutethimide) and some racemic a-
methyl (8Á5 and 12Á2 6 aminoglutethimide) and a-ethyl (12Á1 and 32Á9 6 aminoglutethi-
mide) analogues has been rationalized.
This work selectively extends studies of the P450 17 inhibitor 5, a potential prostatic
cancer agent, towards other cytochrome P450 enzymes in the steroidogenic pathway and
provides a general method for determining the relative in¯uence of chemical manipulation
of a parent inhibitor towards two enzymes in the pathway using additional literature data.
Aromatase (P450Arom) catalyses the ®nal step in the
steroidogenesis pathway of oestrogens from cho-
lesterol. In postmenopausal women androstene-
dione synthesized in the adrenals is converted by
peripheral and breast P450Arom to oestrone and then
to the potent oestradiol by the action of 17b-
hydroxysteroid dehydrogenase. Aromatase has
been the target for the design of inhibitors as agents
in the treatment of breast cancer in postmenopausal
women (Brodie 1994). Inhibition of the enzyme
reduces plasma oestrogen levels and the stimulus to
growth of metastases.
Substituted 1-[(benzofuran-2-yl)phenylmethyl]-
imidazoles, 6, are very potent, in-vitro reversible
inhibitors of P450Arom (80±1000 6 aminoglutethi-
mide) (Whomsley et al 1993) but are non-selective
because of high potency towards other cytochrome
P450 enzymes in the steroidogenic pathway, i.e.
17a-hydroxylase:17,20-lyase (P450 17) and 11b-
hydroxylase (P450 11b) (Al Hamrouni 1996; Al
Hamrouni et al 1997). This lack of selectivity
would make them unsuitable as agents for the
treatment of breast cancer as cortisol and aldos-
terone production in-vivo would be affected.
Modi®cation of 6 by removal of the phenyl group
gives the 1-(benzofuran-2-ylmethyl)imidazoles 1±
5 which have been examined as selective inhibitors
of human testicular and bovine adrenal microsomal
P450 17 with a view to their use as potential agents
for prostatic cancer (Bahshwan et al 1998).
The ®rst-generation reversible inhibitor, still used
clinically, was aminoglutethimide (Shaw at al
1
988; Santen 1990); this has been replaced by other
more selective and potent non-steroidal inhibitors
Miller 1996). Irreversible steroidal inhibitors
(
which are mechanism-based inactivators of the
enzyme have also entered the clinic (Lombardi
1995; Miller 1996). Recently, we have described
some non-steroidal mechanism-based inactivators
Here we describe the synthesis of 1±5 and extend
the selectivity studies to inhibition of P450Arom to
determine the effect of the phenyl group on the
potency and selectivity of the compounds as inhi-
bitors of P450Arom and P450 17.
(Saeed et al 1997).
The structures of compounds 1±6 are given in
Figure 1.
Correspondence: H. J. Smith, Welsh School of Pharmacy,
Cardiff University, Cardiff CF1 3XF, South Glamorgan, UK.

4
28
CAROLINE P. OWEN ET AL
Materials and Methods
Chemistry
The reagents used were either general purpose or
analytical grade and were obtained from either
Aldrich (Gillingham, Dorset, UK), BDH (Poole,
Dorset, UK) or Lancaster Synthesis (Morecombe,
Lancashire, UK).
Melting points were determined with an Electro-
thermal instrument and infrared spectra were
determined with a Perkin-Elmer 681 infrared
spectrophotometer. Thin-layer chromatography
was performed on aluminium-backed silica gel
1
plates with ¯uorescent indicator. H Nuclear mag-
netic resonance (NMR) spectra were recorded, with
TMS as an internal standard, on a Perkin-Elmer
R32 (90 MHz) spectrophotometer. Mass spectra
were acquired by the EPSRC National Mass
Spectrometry Service Centre, University of Wales,
Swansea. Elemental analyses were determined by
the School of Pharmacy, University of London.
Synthesis
The imidazoles 1±5 were synthesized by the gen-
eral method shown in Figure 2.
Figure 1. The structures of compounds 1±6.
Figure 2. The synthesis of 1-(benzofuran-2-ylmethyl)imidazoles.

INHIBITION OF AROMATASE BY 1-(BENZOFURAN-2-YLMETHYL)IMIDAZOLES
429
1
-(Benzofuran-2-y methyl)imidazole, 1
-Hydroxymethylbenzofuran 16. Methyl
The combined dichloromethane layers were
washed with water (2 6 50 mL), dried (MgSO4)
and ®ltered. Removal of the dichloromethane
gave 1-(benzofuran-2-ylmethyl)imidazole as a
pale buff solid. Crystallization (petroleum ether)
gave off-white crystals (yield 550 mg, 59%), mp
2
benzo-
furan-2-carboxylate (11) (2 g, 0Á011 mol) was dis-
solved in tetrahydrofuran (THF) (30 mL) and to this
was added dropwise, lithium borohydride (2Á0 M
solution in THF; 11Á5 mL, 0Á022 mol). The reaction
mixture was heated under re¯ux for 6 h, the THF
removed under vacuum and water (100 mL) added.
The mixture was carefully acidi®ed (HCl) and
extracted with dichloromethane (2 6 50 mL). The
combined organic layers were washed with water
ꢀ
77Á8±79Á8 C. Found: C, 72Á29; H, 5Á11; N, 14Á02%;
‡
M
198Á0793. C H ON requires C, 72Á71; H,
‡
max
12
10
2
5Á08; N, 14Á13%; M 198Á0793. n
(KBr): 3020
�
1
(Ar, C� H), 2970 (C� H), 1610 (Ar, C ˆ C) cm .
d (CDCl ): 7Á4 (7H, br. m, Ar, Im, C� H), 6Á7 (1H,
H
3
s, Ar, C� H), 5Á3 (2H, s, CH ).
2
(3 6 100 mL) and dried (MgSO ). The solvent was
4
removed under vacuum to give 2-hydroxymethyl-
benzofuran as a pale yellow oil (yield 1Á6 g, 95%).
1-(5-Chloro-benzofuran-2-ylmethyl)imidazole, 2
Ethyl 5-chlorobenzofuran-2-carboxylate, 12.
Chlorosalicylaldehyde (7) gave a pale yellow
solid of ethyl 5-chlorobenzofuran-2-carboxylate
which on crystallization (ethanol) gave pale-
‡
Found: C, 73Á08; H, 5Á39%; M 148Á052. C H O
9
8
2
‡
requires C, 72Á95; H, 5Á45%; M 148Á052. n
5-
max
(
(
(
neat): 3350 (br. OH), 3035 (Ar, C� H), 2940, 2885
�
1
C� H), 1605 (Ar, C ˆ C) cm . d (CDCl ): 7Á3
H
3
4H, m, Ar, CH), 6Á55 (H, s, Ar, CH), 4Á65 (2H, s,
ꢀ
CH � OH), 3Á17 (H, br. s, CH ±OH).
cream crystals (yield 66%); mp 49Á5±50Á7 C (Kur-
2
2
ꢀ
dukar & Subba Rao (1963); mp 65 C). Found: C,
H O Cl
9 3
‡
2
-Bromomethylbenzofuran, 21. 2-Hydroxymethyl-
58Á55; H, 4Á25%; M 224Á0240. C11
‡
benzofuran (16) (1Á5 g, 0Á01 mol) in anhydrous
requires C, 58Á8; H, 4Á14%; M 224Á0240. nmax
(KBr): 3010 (Ar, C� H), 2990, 2950 (C� H), 1730
ꢀ
diethyl ether (150 mL) was cooled to 0 C. Phos-
�
1
phorous tribromide (1Á5 mL, 0Á016 mol) was added
dropwise at a rate such that the temperature was
kept constant, and DMF (1 mL) was then added.
The resulting solution was stirred for 12 h at room
temperature and was then poured cautiously on to
ice. The ether layer was washed sequentially with
water (3 6 100 mL), sodium bicarbonate solution
(C ˆ O), 1590 (Ar, C ˆ C), 1110 (C� Cl) cm . d
H
(CDCl
J ˆ 7 Hz, CH
CH ).
� CH
): 7Á6 (4H, m, Ar, C� H), 4Á55 (2H, q,
3
� CH ), 1Á55 (3H, t, J ˆ 7 Hz,
2
3
2
3
5-Chloro-2-hydroxymethylbenzofuran, 17. Ethyl 5-
chlorobenzofuran-2-carboxylate (12) gave 5-
chloro-2-hydroxymethylbenzofuran as a pale yel-
low solid. Crystallization and recrystallization (pet-
roleum ether) gave pale yellow crystals (90%), mp
(1 6 100 mL), and water (3 6 100 mL). The ether
layer was dried (MgSO ) and the ether removed
under vacuum to give a yellow oil of 2-bromo-
4
‡
ꢀ
‡
methylbenzofuran (yield 1Á3 g, 61%). Found: M
80Á5±82 C. Found: C, 59Á72; H, 3Á92%; M
‡
2
09Á9680. C H OBr requires M 209Á9680. n
182Á0135. C
9
H
7
O
2
Cl requires C, 59Á34; H, 3Á88%;
9
7
max
‡
(
neat): 3020 (Ar C� H), 2880 (C� H), 1590 (Ar,
M
182Á0135. nmax (KBr): 3350 (br. OH), 2950
�
1
� 1
C ˆ C), 660 (C� Br) cm . d (CDCl ): 7Á35 (4H,
(C� H), 1610 (Ar, C ˆ C), 1115 (Ar, C� Cl) cm .
H
3
m, Ar, CH), 6Á75 (1H, s, Ar, C� H), 4Á55 (2H, s,
d
H
(CDCl
3
): 7Á4 (3H, m, Ar, C� H), 6Á62 (1H, s, Ar,
CH ±Br).
2
C� H), 4Á78 (2H, s, CH
2
� OH), 2Á05 (1H, br. s,
CH � OH).
2
1
-(Benzofuran-2-ylmethyl)imidazole, 1. Imidazole
(
0Á41 g, 0Á006 mol) was added to anhydrous potas-
2-Bromomethyl-5-chlorobenzofuran, 22. 5-Chloro-
2-hydroxymethylbenzofuran (17) gave 2-bromo-
methyl-5-chlorobenzofuran as a white solid. Crys-
tallization (methanol) furnished white crystals
sium carbonate (0Á35 g, 0Á0025 mol) and dry acet-
one (100 mL). The mixture was stirred at room
temperature for 10 min and then 2-bromomethyl-
benzofuran (21) (1 g, 0Á0047 mol) was added and
stirring continued at room temperature for a further
ꢀ
‡
(yield, 1Á1 g, 82%), mp 82Á9±84Á3 C. Found M
‡
245Á9270. C
H
9
OBrCl requires M 245Á9270. nmax
6
2
4 h. After ®ltration, the acetone was removed
(KBr): 3055 (Ar C� H), 2975 (C� H), 1590 (Ar,
�
H
1
under vacuum to leave a clear yellow oil which
was dissolved in dichloromethane (50 mL) and
extracted with hydrochloric acid (0Á1 M,
C ˆ C), 1060 (C� Cl), 695 (C� Br) cm . d
(CDCl
): 7Á35 (3H, m, Ar, C� H), 6Á71 (1H, s, Ar,
C� H), 4Á58 (2H, s, CH � Br).
3
2
3
6 50 mL). The combined acid layers were made
alkaline with saturated sodium bicarbonate and
then extracted with dichloromethane (2 6 50 mL).
1-(5-Chlorobenzofuran-2-ylmethyl)imidazole, 2. 2-
Bromomethyl-5-chlorobenzofuran (22) gave 1-(5-

4
30
CAROLINE P. OWEN ET AL
�
1
chloro-benzofuran-2-ylmethyl)imidazole
cream solid. Crystallization and recrystallization
as
a
(Ar, C ˆ C) cm . d (CDCl ): 7Á3 (6H, br. m, Ar,
H
3
Im, CH), 6Á61 (1H, s, Ar, C� H), 5Á25 (2H, s,
(
ethanol) gave clear crystals (yield 500 mg, 53%),
ꢀ
CH � Im).
2
mp 113Á6±114Á8 C. Found: C, 61Á68, H, 3Á97, N,
‡
1
6
(
2Á23%; M 232Á0403. C H OClN requires C,
1
2
9
2
‡
1
-(5,7-Dibromobenzofuran-2-ylmethyl)imidazole, 4
1Á93; H, 3Á90; N, 12Á04%; M 232Á0403. n
max
KBr): 3110 (Ar, C� H), 2980 (C� H), 1605 (Ar,
�
H 3
1
Ethyl 5,7-dichlorobenzofuran-2-carboxylate, 14.
,5-Dichloro-2-hydroxybenzaldehyde (9) gave
ethyl 5,7-dichlorobenzofuran-2-carboxylate as
C ˆ C) cm . d (CDCl ): 7Á35 (6H, br. m, Ar, Im,
3
C� H), 6Á58 (1H, s, Ar, C� H), 5Á22 (2H, s,
CH � Im).
2
pale buff crystals (yield, 48%) from ethanol, mp
ꢀ
7
8Á9±80Á7 C (Kurdukar & Subba Rao (1963); mp
ꢀ
‡
1
-(5-Bromobenzofuran-2-ylmethyl)imidazole, 3
123±124 C). Found: C, 50Á86; H, 3Á12%; M
57Á9850. C H O Cl requires C, 50Á97; H,
2
11
8
3
2
‡
Ethyl 5-bromobenzofuran-2-carboxylate, 13. 5-
Bromosalicylaldehyde (8) gave ethyl 5-bromoben-
zofuran-2-carboxylate as pale brown crystals (yield
3Á11%; M 257Á9851. n
(KBr): 3090, 3010
(Ar, C� H), 2980, 2920 (C� H), 1725 (C ˆ O),
max
�
1
1590 (Ar, C ˆ C) cm . d (CDCl ): 7Á55 (1H, d,
H
3
ꢀ
6
0%) from ethanol, mp 53Á5±54Á8 C (Kurdukar &
J ˆ 2 Hz, Ar, C� H), 7Á45 (1H, d, J ˆ 2 Hz, Ar,
C� H), 7Á28 (1H, s, Ar, C� H), 4Á46 (2H, q,
J ˆ 7 Hz, � CH CH ), 1Á42 (3H, t, J ˆ 7 Hz,
ꢀ
Subba Rao (1963); mp 90 C). Found: C, 49Á56; H,
‡
3
4
3
(
(
Á44%; M 267Á9735. C H O Br requires C,
11
9
3
2
3
‡
9Á26; H, 3Á38%; M 267Á9735. n
(KBr):
100, 3070 (Ar, C� H), 2990 (C� H), 1740
� CH CH ).
max
2
3
�
1
C ˆ O), 1610 (Ar, C ˆ C) cm . d (CDCl ): 7Á5
5,7-Dichloro-2-hydroxymethylbenzofuran, 19. Ethyl
5,7-dichlorobenzofuran-2-carboxylate (14) gave 2-
hydroxymethyl-5,7-dichlorobenzofuran as a white
solid. Crystallization (petroleum ether) furnished
H
3
4H, m, Ar, C� H), 4Á5 (2H, q, J ˆ 7 Hz, CH -
2
CH ), 1Á48 (3H, t, J ˆ 7 Hz, CH � CH ).
3
2
3
ꢀ
5-Bromo-2-hydroxymethylbenzofuran, 18. Ethyl 5-
bromobenzofuran-2-carboxylate (13) gave 5-
white crystals (yield 95%), mp 113Á5±114Á5 C.
‡
Found: C, 50Á21; H, 2Á77%; M 215Á9745.
‡
bromo-2-hydroxymethylbenzofuran as a white
ꢀ
C H Cl O requires C, 50Á00; H, 2Á80%; M
9
6
2 2
solid (yield, 94%), mp 95Á1±97Á0 C (Dann et al
215Á9745. n
(KBr): 3250 (br. OH), 3100 (Ar,
C� H), 2970, 2880 (C� H), 1610 (Ar, C ˆ C) cm .
d (CDCl ): 7Á36 (1H, d, J ˆ 2 Hz, Ar, C� H), 7Á29
max
ꢀ
‡
� 1
(
1982); mp 99±103 C). Found M 225Á9629.
‡
C H O Br requires M 225Á9629. n
(Nujol):
9
7
2
max
H
3
�
1
3
(
300 (br. OH), 1610 (Ar, C ˆ C) cm . d
(1H, d, J ˆ 2 Hz, Ar, C� H), 6Á66 (1H, s, Ar,
H
CDCl ): 7Á38 (3H, m, Ar, C� H), 6Á84 (1H, s, Ar,
C� H), 4Á82 (2H, s, CH OH), 2Á2 (1H, br. s,
3
2
C� H), 4Á78 (2H, s, CH � OH), 2Á18 (1H, br. s,
CH OH).
2
2
CH � OH).
2
2
-Bromomethyl-5,7-dichlorobenzofuran, 24. 5,7-
5
-Bromo-2-bromomethylbenzofuran, 23. 5-Bromo-
Dichloro-2-hydroxymethylbenzofuran (19) gave
2-bromomethyl-5,7-dichlorobenzofuran as a white
solid. Crystallization (ethanol) furnished white
2-hydroxymethylbenzofuran (18) gave 5-bromo-2-
bromomethylbenzofuran as a pale cream solid
ꢀ
ꢀ
(
(
yield 78%), mp 166Á7±167Á4 C (Dann et al
crystals (yield 71%), mp 67Á1±68Á5 C. Found: C,
ꢀ
‡
‡
1982); mp 84±85 C) Found: M 287Á8786.
38Á54; H, 1Á78%; M 279Á8880. C H OBrCl
9
5
2
‡
‡
C H Br O requires M 287Á8786. n
(Nujol):
requires C, 38Á59; H, 1Á80%; M 279Á8881. n
9
6
2
max
max
�
1
1
610 (Ar, C ˆ C), 670 (C� Br) cm . d (CDCl ):
(KBr): 3040 (Ar, C� H), 2980 (C� H), 1595 (Ar,
�
H 3
H
3
1
7
Á4 (3H, m, Ar, C� H), 6Á85 (1H, s, Ar, C� H), 6Á30
C ˆ C) cm . d (CDCl ): 7Á43 (1H, d, J ˆ 2 Hz,
(2H, s, CH � Br).
Ar, C� H), 7Á34 (1H, d, J ˆ 2 Hz, Ar, C� H), 6Á78
1H, s, Ar, C� H), 4Á57 (2H, s, CH � Br).
2
(
2
1
-(5-Bromobenzofuran-2-ylmethyl)imidazole, 3. 5-
Bromo-2-bromomethylbenzofuran (23) gave 1-(5-
bromobenzofuran-2-ylmethyl)imidazole as a pale
cream solid. Crystallization (petroleum ether)
1-(5,7-Dichlorobenzofuran-2-ylmethyl)imidazole,
4. 2-Bromomethyl-5,7-dichlorobenzofuran
(24)
gave 1-(5,7-dichlorobenzofuran-2-ylmethyl)imida-
zole as a white solid. Crystallization (diethyl
ether) gave white needles (yield 42%), mp 82Á8±
gave off-white crystals, (yield 56%), mp 124±
ꢀ
1
M
25Á2 C. Found: C, 48Á69; H, 3Á68; N, 9Á53%;
‡
ꢀ
‡
276Á9898. C H OBrN .H O requires C,
83Á7 C. Found: C, 52Á89; H, 3Á22; N, 9Á97%; M
12
9
2
2
‡
4
8Á99; H, 3Á76; N, 9Á52%, M 276Á9898). n
266Á0014. C H OCl N .5H O requires C, 52Á38;
‡
max
max
12
8
2
2
2
(KBr): 3100 (Ar, C� H), 2990, 2930 (C� H), 1605
H, 3Á29; N, 10Á15%; M 266Á0014. n
(KBr):

INHIBITION OF AROMATASE BY 1-(BENZOFURAN-2-YLMETHYL)IMIDAZOLES
431
�
1
3
(
100 (Ar, C� H), 1610 (Ar, C ˆ C) cm . d
Biochemistry
H
CDCl ): 7Á65 (1H, br. s, Im, C� H), 7Á42 (1H, d,
All non-radioactive steroids, NADPH (mono
sodium salt), D-glucose-6-phosphate (mono sodium
salt) and protein standards were from Sigma
(Poole, Dorset). All unlabelled laboratory reagents
were Analar grade and from BDH. D-Glucose-6-
phosphate dehydrogenase (suspension in ammo-
3
J ˆ 2 Hz, Ar, C� H), 7Á32 (1H, d, J ˆ 2 Hz, Ar,
C� H), 7Á1 (2H, m, Im, C� H), 6Á59 (1H, s, Ar,
C� H), 5Á28 (2H, s, CH -Im).
2
1
-(5,7-Dibromobenzofuran-2-ylmethyl)imidazole, 5
nium sulphate) was from Boehringer Mannheim
(
dione (40 Ci mmol , in ethanol) was from NEN,
Dupont (UK), Stevenage, Hertfordshire, UK.
Aminoglutethimide was a gift from Ciba-Geigy,
Horsham, Sussex, UK.
Radioactivity was measured with a LKB Wallac
1217 Rackbeta liquid scintillation counter. Scintil-
lation ¯uid was Optiphase III (Hisafe) from Phar-
macia LKB.
Preliminary homogenization of placental tissue
was performed with a Kenwood Chef blender.
Low-speed centrifugation was performed with a
MSE Europa 65M Ultracentrifuge.
3
Mannheim, Germany). [1b,2b- H]Androstene-
�
1
Ethyl 5,7-dibromobenzofuran-2-carboxylate, 15.
,5-Dibromosalicylaldehyde (10) gave ethyl 5,7-
dibromobenzofuran-2-carboxylate as pale cream
3
crystals (yield 2Á4 g, 66%) from ethanol, mp
ꢀ
102±103 C).
106Á9±108 C (Kurdukar & Subba Rao (1963);
mp
ꢀ
‡
Found
M
347Á8820.
(Nujol):
max
‡
C H O Br requires M 347Á8820. n
11
8
3
2
3
040 (Ar, C� H), 1720 (C ˆ O), 1590 (Ar, C ˆ C)
�
H 3
1
cm . d (CDCl ): 7Á75 (2H, m, Ar, C� H), 7Á51
(
1H, s, Ar, C� H), 4Á5 (2H, q, J ˆ 7 Hz,
CH � CH ), 1Á46 (3H, t, J ˆ 7 Hz, CH � CH ).
2
3
2
3
5
,7-Dibromo-2-hydroxymethylbenzofuran, 20.
The NADPH-generating system consisted of
D-glucose-6-phosphate (28Á2 mg), NADP (8Á6 mg),
D-glucose-6-phosphate dehydrogenase (4 interna-
tional units mL , 15 mL) and phosphate buffer pH
7Á4 (50 mM, 1 mL).
Ethyl 5,7-dibromobenzofuran-2-carboxylate (15)
gave 5,7-dibromo-2-hydroxymethylbenzofuran as
a white solid. Crystallization (petroleum ether)
furnished white crystals (yield 94%), mp 130Á2±
�
1
ꢀ
‡
1
M
31Á5 C. Found M 305Á8714. C H O Br requires
9
6
2
2
‡
305Á8714. n
(KBr): 3210 (br. OH), 3080
max
�
1
(
Ar, C� H), 2945 (C� H), 1600 (Ar, C ˆ C) cm .
P450Arom assay
d (CDCl ): 7Á65 (2H, m, Ar, C� H), 6Á72 (1H, s,
Placental microsomes were prepared by the method
of Thompson & Siiteri (1974). The assay followed
was that described by Khodarahmi et al (1998).
Incubation mixtures (0Á5 mL) in triplicate contain-
H
3
Ar, C� H), 4Á85 (2H, s, CH � OH), 2Á0 (1H, br. s,
2
CH � OH).
2
3
2
-Bromomethyl-5,7-dibromobenzofuran, 25. 5,7-
Dibromo-2-hydroxymethylbenzofuran (20) gave
-bromomethyl-5,7-dibromobenzofuran as a white
ing [1b,2b- H]androstenedione, unlabelled sub-
strate (10 mL, 0Á5 mM ®nal concentration) and
2
NADPH-generating system (50 mL) in phosphate
ꢀ
shaking water-bath. The placental microsomes
solid. Crystallization (ethanol) produced white
ꢀ
buffer (50 mM, pH 7Á4) were warmed to 37 C in a
crystals (yield 500 mg, 83%), mp 91Á7±93 C.
‡
‡
ꢀ
Found M 367Á7870. C H OBr requires M
were thawed and warmed to 37 C before addition
9
5
3
�
1
3
67Á7870. n
(Nujol): 3030 (Ar, C� H), 1595
(20 mL, 0Á452 mg mL ®nal concentration) to the
incubation mixtures. After incubation for 6 min, a
sample (300 mL) from each assay tube was added to
1% activated charcoal (900 mL) and mercuric
chloride (1 mM, 300 mL) and mixed thoroughly.
After standing on ice for 15 min the tubes were
centrifuged for 20 min at 3000 rev min (2000 g).
Samples (500 mL) of the supernatant from each tube
were then removed and dispersed in 2 mL scintil-
max
�
1
(
(
Ar, C ˆ C), 670 (C� Br) cm . d (CDCl ): 7Á62
H
3
2H, m, Ar, C� H), 6Á78 (H, s, Ar, C� H), 4Á6 (2H,
s, CH � Br).
2
1
5
-(5,7-Dibromobenzofuran-2-ylmethyl)imidazole,
. 2-Bromomethyl-5,7-dibromobenzofuran (25)
�
1
gave 1-(5,7-dibromobenzofuran-2-ylmethyl)imida-
zole as a pale brown solid. Crystallization (petro-
leum ether) gave cream crystals (yield 260 mg,
3
lation ¯uid and counted ( H) for 1 min.
ꢀ
6
7%), mp 121Á7±123Á1 C. Found: C, 40Á14; H,
‡
2
Á29; N, 7Á76%; M 355Á8983. C H OBr N
12
8
2
2
‡
requires C, 40Á45; H, 2Á25; N, 7Á86%; M
Inhibition studies
3
55Á8983. n
(KBr): 3070 (Ar, C� H), 2990
Inhibitors 1±5 in ethanol (10 mL) were included in
the assay procedure which was conducted in tri-
plicate. Controls (10 mL, ethanol) were also incu-
bated and aminoglutethimide was used as a
max
�
1
(
(
6
C� H), 1605 (Ar, C ˆ C) cm . d (CDCl ): 7Á64
H
3
3H, m, Ar, Im, C� H), 7Á15 (2H, b. m, Im, C� H),
Á61 (H, s, Ar, C� H), 5Á25 (2H, s, CH � Im).
2

4
32
CAROLINE P. OWEN ET AL
Table 1. Some substituted 1-(benzofuran-2-ylmethyl)imida-
zoles as inhibitors of placenta microsomal P450Arom and
nylmethyl]imidazoles 6 were also potent inhibitors of
P450Arom (IC50 ˆ 7Á3±91Á8 nM), 80±100-fold more
potent than aminoglutethimide (Whomsley et al
1993). Here the imidazoles 1±5 have been examined
for their additionalselectivitytowards thetarget P450
a
testicular microsomal P450 17 in man.
b
a
Compound
P450Arom IC50 (mM) P450 17 IC50 (mM)
17 enzyme by evaluation of their potency towards
P450Arom
The 1-(benzofuran-2-ylmethyl)imidazoles 1±5
were 3±7-fold (approx.) more potent than ami-
noglutethimide (Table 1) towards aromatase, the 5-
chloro- and 5-bromo-substituted compounds, 2 and
Aminoglutethimide
Compound 1
Compound 2
Compound 3
Compound 4
Compound 5
Ketoconazole
28Á69Æ 1Á65
7Á32Æ 0Á12
3Á9Æ 0Á02
3Á91Æ 0Á08
6Á25Æ 0Á3
8Á06Æ 0Á1
±
±
ND
.
0Á23Æ 0Á043
0Á38Æ 0Á04
0Á18Æ 0Á033
0Á185Æ 0Á021
0Á029Æ 0Á003
3
, respectively, being the most potent with IC50
Values are meansÆ s.d. of results from three determinations,
a
b
values of 3Á9 mM (aminoglutethimide, 28Á7 mM).
The in-vitro selectivity of 2±5, as inhibitors of
testicular P450 17 in man (Table 1; Bahshwan et al
each in triplicate. From Bahshwan et al (1998). Androstene-
dione, 0Á5 mM. IC50 is the concentration resulting in 50%
inhibition. ND ˆ not determined.
(
1998)), towards P450Arom can be expressed by the
standard for comparative purposes. The percentage
conversion of androstenedione to oestrone and
oestradiol was calculated and the percentage inhi-
bition determined by comparing the conversion in
the presence of inhibitors with that of the controls.
IC50 values (amounts resulting in 50% inhibition)
were determined by use of a range of inhibitor
concentrations, from a plot of percentage activity
remaining against log [inhibitor], using Cricket
Graph (Table 1). All points were means of three
ratio (IC50 P450Arom) = (IC50 P450 17) although
this is not an absolute value because IC50 values
are substrate-dependent. Values of 17Á0 (2), 10Á3
(3), 34Á6 (4) and 42Á0 (5) were obtained, implying
reasonable selectivity for 4 and 5. In the context of
the role of 5 as a potential prostatic cancer agent
targetting P450 17, the level of selectivity towards
P450Arom is considered unimportant. This is
because inhibition of the low level of P450Arom
activity present in males is unlikely to enhance
conversion of androstenedione (breaking through
because of incomplete inhibition of P450 17) by
alternative pathways to its main androgen meta-
bolites, testosterone and dihydrotestosterone, with
an associated stimulus to prostatic growth.
2
determinations, each conducted in triplicate and r
was in the range 0Á987±0Á995.
Results and Discussion
Within the two series of related compounds, 6
and 1±5, as inhibitors of P450Arom and P450 17,
the overall effect of removal of the a-phenyl group
was determined by normalizing the potency in each
series to aminoglutethimide and ketoconazole,
respectively. Comparisons of the ratio (relative
potency to aminoglutethimide)=(relative potency to
ketoconazole) for 1±5 gave values of 22Á8±96Á2
whereas for 6 (4-Cl-, 4-F-, 2-CH -, 4-CH - and 2-
We have found that a series of 16 substituted
1
(
-[(benzofuran-2-yl)phenylmethyl]imidazoles
Whomsley et al 1993) were potent inhibitors of
6
microsomal testicular and bovine adrenal P450 17
fromman(AlHamrounietal1997).Wehavemodi®ed
the general structure (6) by removal of the phenyl
group to give the 1-(benzofuran-2-ylmethyl)imida-
zoles, 2±5, which as inhibitors of P450 17 were
evaluated as potential anti-prostatic agents. Their
speci®city towards other steroidogenic and liver
enzymes was compared with that of ketoconazole
3
3
OCH phenyl derivatives; (Whomsley et al 1993;
3
Al Hamrouni et al 1997)), the values were 566±
9150. These different ratios indicate that removal
(
Bahshwan et al 1998). All four compounds were
of the a-phenyl group in 6, to give 2±5, leads to a
reduction in potency towards P450Arom relative to
P450 17. Direct comparison of IC50 values for
either series with either enzyme was not possible
because different workers used different substrate
concentrations, and this affects measurements of
IC50 values. The considerable reduction in potency
towards P450Arom observed on removal of the
phenyl group in the imidazoles 1±5 can be
explained on the basis of the mode of binding of 6
to the active site (Khodarahmi 1996) of a model of
aromatase (Laughton et al 1993). The imidazole
inhibitors of the testicular enzyme (4,
IC50 ˆ 0Á18 mM; 5, 0Á185 mM) but less potent than
ketoconazole (0Á03 mM). Towards bovine adrenal
enzyme, 4 and 5 were 35- and 31-fold, respectively,
more potent than ketoconazole (IC50 ˆ 39Á8 mM).
Compound5wasconsideredausefulleadcompound;
although less potent than ketoconazole towards
P450scc and P45011b, but not P450c21, at the
enhanced dose required for equivalent effects in-vivo
on P450 17 it is likely that cortisol and aldosterone
productionwillbeaffectedtoagreaterextentthanwith
3
‡
ketoconazole.
The
1-[(benzofuran-2-yl)phe-
ring is coordinated to the haem-Fe through N3,

INHIBITION OF AROMATASE BY 1-(BENZOFURAN-2-YLMETHYL)IMIDAZOLES
433
Al Hamrouni, A., Ahmadi, M., Nicholls, P. J., Smith, H. J.,
Lombardi, P., Pestellini, V. (1997) 1-[(Benzofuran-2-yl)phe-
nylmethyl]imidazoles as inhibitors of 17a-hydroxy-
lase:17,20-lyase (P450 17): species and tissues differences.
Pharm. Sci. 3: 259±263
and the phenyl and benzofuran rings bind in the
hydrophobic steroidal substrate backbone binding
cavity and the hydrophobic cavity below the A ring
of the steroid (Khodarahmi 1996). The positions
taken by the two ring structures are interchangeable
as suggested by the high potency for both enan-
Bahshwan, S. A., Owen, C. P., Nicholls, P. J., Smith, H. J.,
Ahmadi, M. (1998) Some 1-[(benzofuran-2yl)methyl]imida-
zoles as inhibitors of 17a-hydroxylase:17,20-lyase (P450
0
0
tiomers of the 4 -chloro- and 4 -¯uoro (phenyl ring)
1
7) and their speci®city patterns. J. Pharm. Pharmacol. 50:
compounds with IC50 values of 8Á4 nM ((‡) form),
1
109±1116
8
Á4 nM ((� ) form) and 5Á3 nM ((‡) form), 65Á0 nM
Brodie, A. (1994) Aromatase inhibitors. In: Sandler, M.,
Smith, H. J. (eds) Design of Enzyme Inhibitors as Drugs.
Vol. 2. Oxford University Press, Oxford, pp 424±438
Dann, O., Char, H., Greissmeir, H. (1982) Synthesis of
biscationic trypanocidal 1-benzofuran compounds. Liebigs
Ann. Chem. Part 10. 1836±1869
Khodarahmi, G. A. (1996) Synthesis, Chirality and Molecular
Modelling Studies on Some Aromatase Inhibitors as Poten-
tial Agents for the Treatment of Breast Cancer. PhD. Thesis,
University of Wales
((� ) form), respectively (Khodarahmi 1996). The
1
-(benzofuran-2-ylmethyl)imidazoles 1±5 lacking
the phenyl substituent of 6 could adopt either of the
positions occupied by the benzofuran group in the
enantiomers of 6 but with considerably reduced
binding and potency.
The racemic dichloro (4) and dibromo (5) com-
pounds were 4Á8- and 3Á5-fold more potent than
aminoglutethimide but less active than the corre-
sponding racemic a-methyl analogues (12Á2 and
Khodarahmi, G. A., Smith, H. J., Nicholls, P. J., Ahmadi, M.
(
1998) Enantioselectivity of some 1-(benzofuran-2-yl)-1-(1-
H-imidazol-1-yl) alkanes as inhibitors of P450Arom. J.
Pharm. Pharmacol. 50: 1321±1330
8
Á5 6 aminoglutethimide, respectively) and a-ethyl
Kurdukar, R., Subba Rao, N. V. (1963) Physiologically active
compounds. VII. Synthesis of halo and nitro coumarones.
Proc. Indian Acad. Sci. Sect. A 58: 336±342
Laughton, A. C., Zvelebil, M., Neidle, S. (1993) A detailed
molecular model for human aromatase. J. Steroid Biochem.
Mol. Biol. 44: 399±407
analogues (32Á9 and 12Á1 6 aminoglutethimide,
respectively) (Khodarahmi et al 1998). These dif-
ferences in potency re¯ect the additional binding as
a result of the hydrophobic methyl and ethyl resi-
dues.
Lombardi, P. (1995) The irreversible inhibition of aromatase
Overall this work extends selectively studies of
the P450 17 inhibitor 5, a potential prostatic cancer
agent, toward other cytochrome P450 enzymes in
the steroidogenic pathway. It also provides a gen-
eral method for determining the relative in¯uence
of chemical manipulation of a parent inhibitor
towards two enzymes in the pathway using addi-
tional literature data.
(oestrogen sythetase) by steroidal compounds. Curr. Pharm.
Des. 1: 23±50
Miller, W. R. (1996) Aromatase inhibitors. Endocr. Rel.
Cancers 3: 65±79
Saeed, G., Khodarahmi, A., Fradgley, S. J., Smith, H. J.,
Nicholls, P. J., King, M., Ahmadi, M. (1997) Non-steroidal
irreversible inhibitors of aromatase based on N-pentyl-3-
phenylpyrrolidine-2,5-dione. Pharm. Sci. 3: 265±277
Santen, R. J. (1990) Clinical use of aromatase inhibitors:
current data and future perspectives. J. Enzyme Inhib. 4:
7
9±99
Acknowledgements
Shaw, M. A., Nicholls, P. J., Smith, H. J. (1988) Aminoglu-
tethimide and ketoconazole: historical perspectives and
future prospects. J. Steroid Biochem. 31: 137±146
Thompson, E. A., Siiteri, P. K. (1974) The involvement of
human placental microsomal cytochrome P-450 in aromati-
zation. J. Biol. Chem. 249: 5373±5378
We wish to thank the Cancer Research Campaign
for ®nancial support and the EPSRC National Mass
Spectrometry Service Centre, University of Wales,
Swansea, for the mass spectra determinations.
Whomsley, R., Fernandez, E., Nicholls, P. J., Smith. H. J.,
Lombardi, P., Pestellini, V. (1993) Substituted 1-[(benzo-
furan-2-yl)phenylmethyl]imidazoles as potent inhibitors of
aromatase in vitro and in female rats in vivo. J. Steroid
Biochem. Mol. Biol. 44: 675±676
References
Al Hamrouni, A. M. (1996) Studies on Tissue Selectivity of
inhibitors of P450 17. MPhil. Thesis, University of Wales