Lead optimization of 7-benzyloxy 2-(4′-pyridylmethyl)thio isoflavone aromatase inhibitors

Article abstract of DOI:10.1016/j.bmc.2005.07.038

Aromatase, the enzyme responsible for estrogen biosynthesis, is a particularly attractive target in the treatment of hormone-dependent breast cancer. The synthesis and biological evaluation of a series of 2-(4′-pyridylmethyl)thio, 7-alkyl- or aryl-substit

Full text of DOI:10.1016/j.bmc.2005.07.038

Bioorganic & Medicinal Chemistry 13 (2005) 6571–6577
Lead optimization of 7-benzyloxy
2-(40-pyridylmethyl)thio isoflavone aromatase inhibitors
´
Bin Su, John C Hackett, Edgar S. Dıaz-Cruz,
Young-Woo Kim and Robert W. Brueggemeier*
Division of Medicinal Chemistry and Pharmacognosy, College of Pharmacy, The Ohio State University,
500 W. 12th Avenue, Columbus, OH 43210, USA
Received 6 May 2005; revised 12 July 2005; accepted 13 July 2005
Available online 24 August 2005
Abstract—Aromatase, the enzyme responsible for estrogen biosynthesis, is a particularly attractive target in the treatment of
hormone-dependent breast cancer. The synthesis and biological evaluation of a series of 2-(40-pyridylmethyl)thio, 7-alkyl- or
aryl-substituted isoflavones as potential aromatase inhibitors are described. The isoflavone derivatives demonstrate IC50 values from
79 to 553 nM and compete with the endogenous substrate, androstenedione. Data supporting the ability of these analogs to suppress
aromatase enzyme activity in the SK-BR-3 breast cancer cell line are also presented.
ꢀ 2005 Published by Elsevier Ltd.
1. Introduction
aromatase inhibition. The introduction of a 2-(40-pyr-
idylmethyl)thio functionality onto the isoflavone
nucleus (Fig. 1, compound A) afforded a 56-fold
enhancement in potency compared to the natural prod-
uct lead, Biochanin A (Fig. 1). In addition, changing the
7-hydroxy to 7-benzyloxy group generated a 3-fold more
active compound, the lead compound B.8 This observa-
tion was unexpected because the compound was as-
sumed to be too bulky to fit within the enzyme active
site. Herein, we hypothesize that the 7-benzyloxy group
Breast cancer is the most common cancer diagnosed in
women and represents the second leading cause of can-
cer-related deaths in women. Approximately one-third
of all breast cancer patients and two-thirds of postmen-
opausal breast cancer patients have estrogen-dependent
breast cancer, which contains estrogen receptors and
requires estrogen for tumor growth.1 Estrogens are
biosynthesized from androgens by aromatase, a member
of the cytochrome P450 superfamily.2 Aromatization of
androgen substrates is the terminal and rate-limiting
step in estrogen biosynthesis, and aromatase has been
a particularly attractive target in the treatment of
hormone-dependent breast cancer.3–6
OCH3
O
O
OH
O
O
HO
S
Many researchers have suggested that flavonoids are
responsible for the low incidence in breast cancer in
women from some regions of the world.7 Our research
group is pursuing drug discovery efforts exploiting flavo-
noids as core templates for novel aromatase inhibitors
(AIs). In our previous work, we demonstrated that
introducing a nitrogen-containing heterocycle at the
2-position of the isoflavone nucleus is beneficial for
HO
N
Compound A
IC50= 0.61
Biochanin A
µ
M
µ
IC50= 34
M
O
O
O
S
N
Compound B
IC50= 0.21
Keywords: Aromatase; Enzyme inhibitors; Isoflavones; SK-BR-3
breast cancer cells; p–p interaction.
Corresponding author. Tel.: +1 614 292 5231; fax: +1 614 292
3113; e-mail: Brueggemeier.1@osu.edu
µ
M
*
Figure 1. Chemical structures and aromatase inhibitory activities of
natural product biochanin A and lead compounds.8
0968-0896/$ - see front matter ꢀ 2005 Published by Elsevier Ltd.
doi:10.1016/j.bmc.2005.07.038

6572
B. Su et al. / Bioorg. Med. Chem. 13 (2005) 6571–6577
plays a very important role in the binding process and
that by introducing similar functional groups into the
7-position of compound A may lead to better AIs.
yl)pyridine hydrobromide were used as alkyl halides,
even after refluxing for 72 h. All the synthesized com-
pounds were confirmed by HNMR and HRMS, and
1
compounds were also confirmed by elemental analyses.
2. Results and discussion
2.1. Synthesis approach
2.2. Biological evaluation
2.2.1. Aromatase activity assay in human placental
microsomes. Aromatase inhibition was performed with
human placental microsomes using methods previously
reported by our laboratory.9 The IC50 values of the
compounds were determined by nonlinear regression
analysis of the dose–response curves (Fig. 2) and the val-
ues are given in Table 1. The IC50 value of ( )-amino-
glutethimide (AG) is also listed for comparison.8 To
examine the mode of aromatase inhibition and to more
accurately distinguish the relative potencies, enzyme
kinetic studies for selected compounds (5d, 5e, and
5m) were also performed, and their apparent Ki values
are given in Table 2, along with apparent Km values
and Ki/Km ratios. Lineweaver–Burk plot of compound
5m, the most potent analog in this series, is shown in
Figure 3.
The synthesis of the target compounds was carried out,
as outlined in Scheme 1. Compound 1 was commercially
available. The 4-hydroxyl group of compound 1 was
selectively protected with a benzyl group under Mitsun-
obu reaction conditions to give a monoalkyl ether 2.
Compound 2 was treated with carbon disulfide and
4-(bromomethyl)pyridine hydrobromide in a THF-
aqueous NaOH solution in the presence of 10 mol% of
tetrabutylammonium hydrogensulfate (n-Bu4NÆHSO4)
to give the corresponding 2-(alkylthio)isoflavone 3 in
good yield. Dealkylation reaction of compound 3 was
performed with BF3ÆOEt2–Me2S in dichloromethane,
yielding hydroxy compound 4.8 Compound 4 was treat-
ed with K2CO3 and alkyl halide in DMF at room tem-
perature or refluxed to obtain the desired compounds.
Enzyme inhibition studies have demonstrated that
extending the hydrocarbon chain from benzyloxy (Com-
pound B) to phenylethoxy (5f) decreased the activity
almost 2-fold. Expansion of the aromatic system of the
benzyloxy analog to naphthyl (5d, 5e) enhanced activity
approximately 2-fold. However, dose–response studies
alone could not discern whether attachment at the
a- or b-position produces more desirable inhibitory
activity. Enzyme kinetic studies corroborated the
dose–response results, revealing the b-naphthyl analog
The nitrogen atom of the 2-(4-pyridylmethyl)thio group
and the 7-hydroxy group of intermediate 4 are reactive
toward alkyl halides and result in dialkylated quaterna-
ry amine side products, respectively. For some alkyl
halides, such as methyl, dimethyl, and 3-methoxy benzyl
chloride, we obtained the quaternary amines as the ma-
jor products and the yields for the desired compounds
were below 3%. No product was found when 3-(bromo-
methyl)pyridine hydrobromide and 4-(bromometh-
O
O
BnOH
CS2, aq. NaOH, THF
Ph3P, DIAD, THF
Bu4NHSO4,
(bromomethyl)pyridine
hydrobromide
BnO
OH
HO
OH
O
4-
1
2 (87%)
O
O
Me2S, BF3OEt2, CH2Cl2
HO
S
BnO
O
S
N
N
4
(88.6%)
3 (86%)
O
5a: R= 2'-nitrophenylmethoxy (71%)
5b: R= 3'-nitrophenylmethoxy (66%)
5c: R= 4'-nitrophenylmethoxy (60%)
K2CO3, DMF
RX
RO
O
S
5d: R= naphthylmethoxy (58%)
α-
5e: R= ß-naphthylmethoxy (78%)
N
5
5f: R= phenylethoxy (
88%)
5g: R= 2'-methoxyphenylmethoxy (60%)
5h: R= 4'-methoxyphenylmethoxy (39%)
5i: R= cyclohexylmethoxy (81%)
5j: R= 4'-fluorophenylmethoxy (87%)
5k: R= 4'-bromophenylmethoxy (97%)
5l: R= 4'-chlorophenylmethoxy (83%)
5m:R= 4'-biphenylmethoxy (71%)
5n: R= 2'-pyridylmethoxy (54%)
Scheme 1. Synthesis of isoflavones 5a–n.

B. Su et al. / Bioorg. Med. Chem. 13 (2005) 6571–6577
6573
Table 2. Enzyme kinetic parameters for selected isoflavones and
reference compounds
100
80
60
40
20
0
Compound
Apparent
Ki (nM) (SE)a
Apparent Km
(nM) (SE)a
Ki/Km
Compound B
5d
5e
220 20
114 15.7
74 8.0
130 10
100 8.8
89 7.2
77 9.0
90 10
1.69b
1.14
0.83
0.51
15.7b
5m
AG
39 5.0
1410 10
a Apparent Km, apparent Ki, and SE values were calculated by
weighted regression analysis.13
b Ref. 8
1
2
3
4
5
log [nM]
O
15000
Figure 2. Aromatase inhibitory activities of compounds 5f (j), 5e (m),
5d (.), 5m (Ç), and 5i (d). Aminoglutethimide (s) was used as
reference. Error bars represent standard error (n = 3), and the data
were statistically analyzed by a nonlinear regression analysis method.
O
O
S
N
10000
5000
0
5m
(5e) to be slightly more active. Introduction of an addi-
tional phenyl ring generated the biphenyl analog (5m),
which was the most potent one in this series of com-
pounds. The cyclohexyl analog (5i) demonstrated a
greater than 2-fold loss of activity, thus illustrating the
importance of unsaturation in the 7-substituent to
enhance aromatase inhibitory activity. Introduction of
a nitro group to the 7-benzyloxy functionality leads to
an almost 2-fold enhancement in activity (5a, 5b, and
5c). This enhancement appears to be independent of
the position of the electron-withdrawing substituent.
These observations, taken together with the apparent
necessity for an aromatic moiety at the 7-position, sug-
gest that these molecules exploit a beneficial p–p interac-
tion and/or a planar structural feature in the aromatase
active site.
0.000
0.005
0.010
0.015
0.020
1/[Androstenedione (nM)]
Figure 3. Lineweaver–Burk plot of aromatase inhibition by compound
5m. Various concentrations of androstenedione (50, 100, 200, 400, and
800 nM) were incubated with microsomal enzyme preparations at
inhibitor concentrations of 0 nM (j), 25 nM (m), 100 nM (.), or
200 nM (Ç). Each point represents the average of three determinations
( )standard error.
increase in activity. Whether this minor enhancement
in activity arises from the hydrophobic nature of this
substituent or from its hydrogen-bonding potential is
unclear. Introduction of a 2-pyridyl substituent at the
7-position (5n) leads to an almost 2-fold decrease in
activity. Introduction of chlorine or bromine (5l, 5k)
at the para position apparently has no effect on the
Based on the IC50 values alone, it appears that introduc-
tion of an o-methoxy group to the 7-benzyloxy substitu-
ent (5g) does not impart a significant change in
inhibitory activity. However, introduction of this func-
tional group in the para position (5h) leads to a slight
Table 1. Aromatase inhibitory activities of isoflavones 5a–n
Compound
R
IC50 (nM)
Log IC50 (nM) SEa
Compound B
Benzyloxy
210
138
113
132
112
90
2.33 0.03b
2.14 0.03
2.06 0.05
2.12 0.08
2.05 0.15
1.96 0.05
2.55 0.04
2.38 0.04
2.21 0.02
2.74 0.08
2.53 0.05
2.33 0.05
2.37 0.05
1.90 0.16
2.58 0.06
3.45 0.05
5a
5b
5c
5d
5e
5f
20-Nitrophenylmethoxy
30-Nitrophenylmethoxy
40-Nitrophenylmethoxy
a-Naphthylmethoxy
b-Naphthylmethoxy
Phenylethoxy
359
243
161
553
337
213
233
79
5g
5h
5i
20-Methoxy-phenylmethoxy
40-Methoxy-phenylmethoxy
Cyclohexylmethoxy
40-Fluorophenylmethoxy
40-Bromophenylmethoxy
40-Chlorophenylmethoxy
40-Biphenylmethoxy
20-Pyridylmethoxy
5j
5k
5l
5m
5n
AGb
378
2800
a IC50 values were calculated by a nonlinear regression analysis (GraphPad Prism). Each dose–response curve contained 10 concentrations, each in
triplicate.
b Ref. 8

6574
B. Su et al. / Bioorg. Med. Chem. 13 (2005) 6571–6577
aromatase inhibitory activity compared to the lead com-
pound. Interestingly, the p-fluoro analog (5j) is approx-
imately 50% less active. Additional analogs of the
7-benzyloxy lead will need to be examined in order to
determine the contributions from hydrophobicity and
hydrogen-bonding potential of substituents to aroma-
tase inhibitory activity.
ously been considered an inappropriate scaffold for the
development of AIs. This study optimized further the
7-substituent of the lead compound and generated ana-
logs with greater than 2-fold activity improvements. In
addition, these structure–activity studies suggest a
potential role for p–p stacking interactions within the
aromatase active site. Although no direct structure
information for the aromatase active site exists, the
potent inhibitory activities of some analogs described
here indicate that the substrate binding pocket may be
able to accept larger ligands than previously thought
possible. These observations on potential p–p stacking
interactions in this series of nonsteroidal agents are
congruent with structure–activity studies of 7a-aryl-
substituted C19 steroidal inhibitors.9–11 Therefore, we
are currently investigating the aromatase inhibitory
activity of other bulky pyridine-containing flavonoids
to better understand the binding site of aromatase.
2.2.2. Aromatase activity assay in SK-BR-3 breast cancer
cell lines. The synthesized compounds were evaluated at
1 lM concentrations for their abilities to inhibit aroma-
tase in SK-BR-3 breast cancer cell cultures (Fig. 4). Our
results showed that halogen-substituted compounds 5j,
5k, 5l, 2-nitro and 3-nitro compounds 5a and 5b, and
a-naphthyl compound 5d had no effect on aromatase
activity. This may be due to their weak ability to pene-
trate the cell membrane and/or their potential binding
to other proteins in our cell culture system, on the cell
surface, or within the cell membrane. The remaining
compounds in this series display a statistically significant
suppression of aromatase enzyme activity. These results
proved that some of these bulky compounds can pass
through the cell membrane and inhibit aromatase activ-
ity. In fact, large lipophilic compounds, such as 5e and
5m, apparently penetrate cells and exhibit promising
aromatase inhibitory activity. The significant suppres-
sion of aromatase by compounds 5e and 5m may be
related to hydrophobicity of the 7-substituent of these
analogs. However, the disparity in aromatase suppres-
sion achieved in microsomes and cell culture by the
more hydrophilic analog, 5n, suggests that other factors,
such as intracellular transport or decreased extracellular
sequestration by the culture system, may be involved.
4. Experimental section
4.1. Chemistry
Chemicals were commercially available and used as
received without further purification unless otherwise
noted. Moisture sensitive reactions were carried out
under a dry argon atmosphere in flame-dried glassware.
Solvents were distilled before use under argon. Tetrahy-
drofuran was distilled from sodium metal in the pres-
ence of benzophenone; dichloromethane was distilled
from calcium hydride. Thin-layer chromatography was
performed on precoated silica gel F254 plates (What-
man). Silica gel column chromatography was performed
using silica gel 60A (Merck, 230–400 Mesh). High-
resolution electrospray ionization mass spectra were
obtained on the Micromass QTOF Electrospray mass
spectrometer at The Ohio State Chemical Instrumenta-
tion Center. All the NMR spectra were recorded on a
Bruker DPX 250 in either DMSO-d6 or CDCl3. Chemi-
3. Conclusion
Introduction of functional groups with the potential to
coordinate the heme iron of aromatase resulted in new
lead AIs with the isoflavone nucleus, which has previ-
1
cal shifts (d) for H NMR spectra are reported in parts
per million to residual solvent protons.
125
100
Compounds 2, 3, and 4 were prepared, as described by
Kim et al.8
*
*
*
*
75
50
25
0
*
*
*
General procedure for the preparation of 7-substituted
2-alkylthioisoflavones (5a–n) from 7-hydroxy-2-(40-pyr-
idylmethylthio)isoflavones. An alkyl halide (0.3 mmol)
was added to a stirred mixture of 7-hydroxy-3-phenyl-
2-[(40-pyridylmethyl)thio]-4H-1-benzopyran-4-one (72 mg,
0.2 mmol) and K2CO3 (83 mg, 0.6 mmol) in DMF
(3 mL) at room temperature. The resulting mixture
was vigorously stirred at room temperature or refluxed
for several hours and monitored by TLC. When the
starting material was exhausted, 15 mL of water was
added and the product was extracted with EtOAc (2·
10 mL). The combined organic layer was washed with
water (10 mL) and then with brine (10 mL), dried over
MgSO4, and filtered. The filtrate was concentrated un-
der reduced pressure, and the residue was purified by sil-
ica gel column chromatography (eluting with MeOH/
CHCl3 or EtOAc/hexane) to yield the desired product.
*
*
Figure 4. Suppression of aromatase activity in SK-BR-3 breast cancer
cells. SK-BR-3 cells were treated with aminoglutethimide (AG, 10 lM)
and compounds 5a–n (1 lM). Aromatase activity was measured, as
described in Section 4. Values are expressed as picomoles 3H2O formed
per hour incubation time per million cells. The results were normalized
against a control treatment with vehicle. The value of 100% is equal to
0.003 pmol/h/106 cells. Each data bar represents the mean results of
three independent determinations. *P < 0.01 versus control by
unpaired t test, n = 3.

B. Su et al. / Bioorg. Med. Chem. 13 (2005) 6571–6577
6575
3-Phenyl-7-(20-nitrophenylmethoxy)-2-[(40-pyridylmeth-
yl)thio]-4H-1-benzopyran-4-one (5a). With 2-nitroben-
zyl bromide as an alkyl halide, 71 mg (71%) of the
title compound was obtained as a pale yellow solid:
CDCl3): d 8.65 (m, 2H), 8.13 (d, J = 8.9 Hz, 1H), 7.29–
7.57 (m, 12H), 6.99 (d, J = 9.1 Hz, 1H), 6.73 (s, 1H),
4.28–4.33 (m, 4H), 3.18 (dd, J = 6.4, 6.4 Hz, 2H); HRMS
Calcd for C29H23NNaO3S (M+Na)+ 488.1296. Found:
488.1310. Anal. Calcd for C29H23NO3S: C, 74.81; H,
4.98; N, 3.01. Found: C, 74.51; H, 5.06; N, 2.89.
1H NMR (250 MHz, DMSO-d6):
d
8.47 (d,
J = 5.5 Hz, 2H), 8.21 (d, J = 8.5 Hz, 1H), 7.95 (d,
J = 8.8 Hz, 1H), 7.84 (d, J = 3 Hz, 2H), 7.70–7.72 (m,
1H), 7.37–7.44 (m, 6H), 7.18–7.23 (m, 3H), 5.68 (s,
2H), 4.45 (s, 2H); HRMS Calcd for C28H21N2O5S
(M+H)+ 497.1171. Found: 497.1153. Anal. Calcd for
C28H20N2O5S: C, 67.73; H, 4.06; N, 5.64. Found: C,
67.30; H, 4.04; N, 5.44.
3-Phenyl-7-(20-methoxyphenylmethoxy)-2-[(40-pyridylmeth-
yl)thio]-4H-1-benzopyran-4-one (5g). With 2-methoxyben-
zyl chloride as an alkyl halide, 57.5 mg (60%) of the title
1
compound was obtained as a pale yellow powder: H
NMR (250 MHz, DMSO-d6): d 8.73 (d, J = 6.7 Hz, 2H),
7.92 (m, 3H), 6.92–7.44 (m, 11H), 5.25 (s, 2H), 4.64 (s,
2H); HRMS Calcd for C29H24NO4S (M+H)+ 482.1426.
Found: 482.1403.
3-Phenyl-7-(30-nitrophenylmethoxy)-2-[(40-pyridylmeth-
yl)thio]-4H-1-benzopyran-4-one (5b). With 3-nitroben-
zyl bromide as an alkyl halide, 66 mg (66%) of the title
1
compound was obtained as a yellow solid: H NMR
3-Phenyl-7-(40-methoxyphenylmethoxy)-2-[(40-pyridylmeth-
yl)thio]-4H-1-benzopyran-4-one (5h). With 4-methoxyben-
zyl chloride as an alkyl halide, 37.3 mg (39%) of the title
compound was obtained as a pale yellow crystal: 1H
NMR (250 MHz, CDCl3): d 8.60 (d, J = 4.2 Hz, 2H),
8.14 (d, J = 9.1 Hz, 1H), 7.39–7.46 (m, 7H), 7.28 (d,
J = 6.8 Hz, 2H), 7.06 (d, J = 8.5 Hz, 1H), 6.97 (d,
J = 8.4 Hz, 2H), 6.80 (s, 1H), 5.13 (s, 2H), 4.29 (s, 2H);
HRMS Calcd for C29H24NO4S (M+H)+ 482.1426.
Found: 482.1439. Anal. Calcd for C29H23NO4S: C,
72.33; H, 4.81; N, 2.91. Found: C, 71.87; H, 4.92; N, 2.91.
(250 MHz, DMSO-d6): d 8.69 (d, J = 5 Hz, 2H), 8.41
(s, 1H), 8.26 (d, J=7.5 Hz, 1H), 7.95–8.02 (dd, J = 12,
7.7 Hz, 2H), 7.75–7.84 (m, 3H), 7.39–7.45 (m, 4H),
7.20–7.25 (m, 3H), 5.48 (s, 2H), 4.59 (s, 2H); HRMS
Calcd for C28H21N2O5S (M+H)+ 497.1171. Found:
497.1196.
3-Phenyl-7-(40-nitrophenylmethoxy)-2-[(40-pyridylmeth-
yl)thio]-4H-1-benzopyran-4-one (5c). With 4-nitroben-
zyl bromide as an alkyl halide, 60 mg (60%) of the title
compound was obtained as a pale yellow solid: 1H
NMR (250 MHz, DMSO-d6): d 8.48 (d, J = 3.6 Hz,
2H), 8.30 (d, J = 8.3 Hz, 2H), 7.94 (d, J = 8.8 Hz, 1H),
7.79 (d, J = 8.7 Hz, 2H), 7.38–7.43 (m, 6H), 7.20–7.22
(m, 3H), 5.50 (s, 2H), 4.44 (s, 2H); HRMS Calcd for
C28H21N2O5S (M+H)+ 497.1171, Found: 497.1156.
3-Phenyl-7-(cyclohexylmethoxy)-2-[(40-pyridylmethyl)thio]-
4H-1-benzopyran-4-one (5i). With cyclohexylmethyl bro-
mideas an alkylhalide, 73.8 mg (81%)ofthe title compound
was obtained as a yellow solid: 1H NMR (250 MHz,
DMSO-d6) d 8.50 (dd, J = 2.8, 4.5 Hz, 2H), 7.89 (d,
J = 8.9 Hz, 1H), 7.38–7.47 (m, 5H), 7.19–7.26 (m, 3H),
7.05 (dd, J = 2.0, 8.8 Hz, 1H), 4.46 (s, 2H), 3.96 (d,
J = 5.9 Hz, 2H), 1.73–1.88 (m, 6H), 1.07–1.28 (m, 5H);
HRMS Calcd for C28H27NNaO3S (M+Na)+ 480.1609.
Found: 480.1615. Anal. Calcd for C28H27NO3S: C, 73.49;
H, 5.95; N, 3.06. Found: C, 73.36; H, 6.00; N, 2.97.
3-Phenyl-7-(b-naphthylmethoxy)-2-[(40-pyridylmethyl)thio]-
4H-1-benzopyran-4-one (5d). With 1-(chloromethyl)naph-
thalene as an alkyl halide, 59 mg (58%) of the title com-
1
pound was obtained as a pale yellow solid: H NMR
(250 MHz, CDCl3): d 8.53 (dd, J = 1.6, 4.6 Hz, 2H),
8.17 (d, J = 8.9 Hz, 1H), 8.06–8.09 (m, 1H), 7.94–8.00
(m, 2H), 7.29–7.64 (m, 11H), 7.13–7.17 (dd, J = 2.4,
8.9 Hz, 1H), 6.89 (d, J = 2.3, 1H), 5.65 (s, 2H), 4.26 (s,
2H); HRMS Calcd for C32H23N NaO3S (M+Na)+
524.1296. Found: 524.1274. Anal. Calcd for C32H23NO3S:
C, 76.62; H, 4.62; N, 2.79. Found: C, 76.23; H, 4.64; N,
2.61.
3-Phenyl-7-(40-fluorophenylmethoxy)-2-[(40-pyridylmeth-
yl)thio]-4H-1-benzopyran-4-one (5j). With 4-fluoroben-
zyl chloride as an alkyl halide, 82 mg (87%) of the
title compound was obtained as a pale green powder:
1H NMR (250 MHz, DMSO-d6) d 8.49 (d, J = 5.9 Hz,
2H), 7.92 (d, J = 8.8 Hz, 1H), 7.57–7.60 (m, 2H),
7.18–7.46 (m, 11H), 5.29 (s, 2H), 4.45 (s, 2H); HRMS
Calcd for C28H20FNNaO3S (M+Na)+ 492.1046.
Found: 492.1047.
3-Phenyl-7-(a-naphthylmethoxy)-2-[(40-pyridylmethyl)thio]-
4H-1-benzopyran-4-one (5e). With 2-(bromomethyl)naph-
thalene as an alkyl halide, 78 mg (78%) of the title com-
1
pound was obtained as a pale yellow solid: H NMR
3-Phenyl-7-(40-bromophenylmethoxy)-2-[(40-pyridylmeth-
yl)thio]-4H-1-benzopyran-4-one (5k). With 4-bromoben-
zyl bromide as an alkyl halide, 103 mg (97%) of the
(250 MHz, DMSO-d6) d 8.42 (dd, J = 1.6, 4.4 Hz, 2H),
8.09 (s, 1H), 7.93–8.03 (m, 4H), 7.64 (dd, J = 1.6, 8.4 Hz,
1H), 7.55–7.59 (dd, J = 3.3, 6.3 Hz, 2H), 7.38–7.43 (m,
6H), 7.19–7.24 (m, 3H), 5.50 (s, 2H), 4.44 (s, 2H); HRMS
Calcd for C32H23N NaO3S (M+Na)+ 524.1296. Found:
524.1315. Anal. Calcd for C32H23NO3S: C, 76.62; H, 4.62;
N, 2.79. Found: C, 76.32; H, 4.60; N, 2.65.
1
title compound was obtained as a yellow powder: H
NMR (250 MHz, DMSO-d6) d 8.49 (dd, J = 1.6,
4.4 Hz, 2H), 7.92 (d, J = 8.9 Hz, 1H), 7.64(d,
J = 8.4 Hz, 2H), 7.35–7.52 (m, 8H), 7.19–7.23 (m, 3H),
5.30 (s, 2H), 4.45 (s, 2H); HRMS Calcd for C28H20
BrNNa O3S (M+Na)+ 552.0245. Found: 552.0264.
3-Phenyl-7-(phenylethoxy)-2-[(40-pyridylmethyl)thio]-4H-
1-benzopyran-4-one (5f). With (2-bromoethyl) benzene
as an alkyl halide, 82 mg (88%) of the title compound
was obtained as a white solid: 1H NMR (250 MHz,
3-Phenyl-7-(40-chlorophenylmethoxy)-2-[(40-pyridylmeth-
yl)thio]-4H-1-benzopyran-4-one (5l). With 4-chloroben-
zyl chloride as an alkyl halide, 81 mg (83%) of the title

6576
B. Su et al. / Bioorg. Med. Chem. 13 (2005) 6571–6577
1
compound was obtained as a green powder: H NMR
(250 MHz, DMSO-d6) d 8.49 (d, J = 5.6 Hz, 2H), 7.92
(d, J = 9.0 Hz, 1H), 7.34–7.59 (m, 11H), 7.15–7.22 (m,
3H), 5.32 (s, 2H), 4.45 (s, 2H); HRMS Calcd for
C28H20ClNNaO3S (M+Na)+ 508.0750. Found: 508.0773.
ple was run in triplicate and background values were
determined with microsomal protein inactivated by boil-
ing. Samples containing 50 lM ( ) aminoglutethimide
(Sigma, St. Louis, MO) were used a positive control.
IC50 sigmoidal dose-response data were analyzed with
the Graphpad Prism (Version 3.0) program.
3-Phenyl-7-(40-biphenylmethoxy)-2-[(40-pyridylmethyl)thio]-
4H-1-benzopyran-4-one (5m). With 4-phenylbenzyl chlo-
ride as an alkyl halide, 75 mg (71%) of the title com-
pound was obtained as a yellow powder: 1H NMR
(250 MHz, DMSO-d6): d 8.48 (d, J = 4.8 Hz, 2H), 7.92
(d, J = 7.5 Hz, 1H), 7.39–7.77 (m, 15H), 7.20–7.23 (m,
3H), 5.37 (s, 2H), 4.45 (s, 2H); HRMS Calcd for
C34H25NNaO3S (M+Na)+ 550.1453. Found: 550.1436.
Anal. Calcd for C34H25NO3S: C, 77.40; H, 4.78; N,
2.65. Found: C, 76.95; H, 4.98; N, 2.27.
4.2.3. Kinetic study. Enzyme kinetic studies of com-
pounds 5d, 5e, and 5m were conducted to investigate the
nature of aromatase inhibition. Michaelis–Menten en-
zyme kinetic parameters were determined by varying the
concentration of androst-4-ene-3,17-dione from 50 to
800 nM in the presence of fixed concentrations of 0, 25,
100, and 200 nM for all three compounds. Assay condi-
tions were the same as those described in the IC50 studies,
except that reactions were initiated by the addition of
15 lg microsomal protein. Analysis of the enzyme kinetic
data was performed with the weighted linear regression
analysis previously described by Cleland.13
3-Phenyl-7-(20-pyridylmethoxy)-2-[(40-pyridylmethyl)thio]-
4H-1-benzopyran-4-one (5n). With 2-(bromomethyl)pyri-
dine hydrobromide as an alkyl halide, 49 mg (54%) of
the title compound was obtained as a pale yellow powder:
1H NMR (250 MHz, DMSO-d6) d 8.64 (d, J = 3.8 Hz,
1H), 8.49 (d, J = 5.0 Hz, 2H), 7.88–7.96 (m, 2H), 7.58
(d, J = 7.8 Hz, 1H), 7.40–7.44 (m, 7H), 7.22 (m, 3H),
5.39 (s, 2H), 4.44 (s, 2H); HRMS Calcd for C27H20N2Na
O3S (M+Na)+ 475.1092. Found: 475.1092.
4.2.4. Tritiated water-release assay in SK-BR-3 cell lines.
Measurement of aromatase enzyme activity was based
on the tritium water release assay.14 Cells in T-25 flasks
were treated with 0.1% DMSO (control) and inhibitors
at the indicated concentrations. After 24 h, the cells were
incubated for 6 h with fresh media along with 50 nM
androstenedione including 2 lCi [1b-3H]androst-4-ene-
3,17-dione. Subsequently, the reaction mixture was re-
moved, and proteins were precipitated using 10% tri-
chloroacetic acid at 42 ꢁC for 20 min. After a brief
centrifugation, the media were extracted three times
with an equal amount of chloroform to extract unused
substrate and further dextran-treated charcoal. After
centrifugation, a 250-ll aliquot containing the product
was counted in 5 ml of liquid scintillation mixture. Re-
sults were corrected for blanks and for the cell contents
of culture flasks, and results were expressed as picomoles
4.2. Biological study
4.2.1. Preparation of human placental microsomes. Human
term placentas were processed immediately after delivery
from The Ohio State University Hospitals at 4 ꢁC. The
placenta was washed with normal saline, and connective
and vascular tissue was removed. Microsomes were
prepared from the remaining tissue using the method
previously described.12 Microsomal suspensions were
stored at ꢀ80 ꢁC until required.
3
of H2O formed per hour incubation time per million
4.2.2. Inhibition study. Inhibition of human placental
aromatase was determined by monitoring the amount
live cells (pmol/h/106 cells). To determine the amount
of cells in each flask, the cells were trypsinized and ana-
lyzed using the diphenylamine DNA assay adapted to a
96-well plate.14,15
3
3
of H2O released as the enzyme converts [1bꢀ H]and-
rost-4-ene-3,17-dione to estrone. Ten inhibitor concen-
trations ranging from 0.2 nM to 50 lM were
evaluated. Aromatase activity assays were carried out
in 0.1 M potassium phosphate buffer (pH 7.0) with 5%
propylene glycol. All samples contained a NADPH-
regenerating system consisting of 2.85 mM glucose-6-
phosphate, 1.8 mM NADP+, and 1.5 U of glucose-6-
phosphate dehydrogenase (Sigma, St. Louis, MO). Sam-
ples contained 100 nM androst-4-ene-3,17-dione
(400,000–450,000 dpm). Reactions were initiated with
the addition of 50 lg microsomal protein. The total
incubation volume was 2.0 mL. Incubations were al-
lowed to proceed for 15 min in a shaking water bath
at 37 ꢁC. Reactions were quenched by the addition of
2.0 mL chloroform. Samples were then vortexed and
centrifuged for 5 min and the aqueous layer was re-
moved. The aqueous layer was subsequently extracted
twice in the same manner with 2.0 mL chloroform. A
0.5 mL aliquot of the final aqueous layer was combined
with 5 mL of 3a70B scintillation cocktail (Research
Products International Corp., Mt. Prospect, IL) and
the amount of radioactivity was determined. Each sam-
Acknowledgments
This research was supported by NIH Grant CA73698
(R.W.B.), USAMRMC Breast Cancer Program Idea
Grant DMAD-17-00-1-0388 (R.W.B.), and USAMRMC
Breast Cancer Program Predoctoral Fellowships
DMAD17-02-1-0529 (J.C.H.) and DMAD17-99-1-9342
(Y.W.K.).
References and notes
1. American Cancer Society, Facts and Figures, 2005.
2. Simpson, E. R.; Mahendroo, M. S.; Means, G. D.;
Kilgore, M. W.; Hinshelwood, M. M.; Graham-Lorence,
S.; Amarneh, B.; Ito, Y.; Fisher, C. R.; Michael, M. D.
Endocr. Rev. 1994, 15, 342–355.
3. Brodie, A.; Lu, Q.; Long, B. J. Steroid. Biochem. Mol.
Biol. 1999, 69, 205–210.

B. Su et al. / Bioorg. Med. Chem. 13 (2005) 6571–6577
6577
4. Santen, R. J.; Harvey, H. A. Endocr. Relat. Cancer 1999,
6, 75–92.
5. Brueggemeier, R. W. Am. J. Ther. 2001, 8, 333–
344.
6. Brueggemeier, R. W. Expert Rev. Anticancer Ther. 2002, 2,
181–191.
7. Adlercreutz, H. Environ. Health Perspect. 1995, 103, 112–
130.
10. Darby, M. V.; Lovett, J. A.; Brueggemeier, R. W.;
Groziak, M. P.; Counsell, R. E. J. Med. Chem. 1985, 28,
803–807.
11. Li, P.-K.; Brueggemeier, R. W. J. Enzyme Inhib. 1990,
113–120.
12. Kellis, J. T., Jr.; Vickery, L. E. J. Biol. Chem. 1987, 262,
4413–4420.
13. Cleland, W. W. Method Enzymol. 1979, 63, 103–138.
14. Richards, J. A.; Brueggemeier, R. W. J. Clin. Endocrinol.
Metab. 2003, 88, 2810–2816.
8. Kim, Y.-W.; Hackett, J. C.; Brueggemeier, R. W. J. Med.
Chem. 2004, 47, 4032–4040.
9. OꢀReilly, J. M.; Li, N.; Duax, W. L.; Brueggemeier, R. W.
J. Med. Chem. 1995, 38, 2842–2850.
15. Natarajan, N.; Shambaugh, G. E.; Elseth, K. M.; Haines,
G. K.; Radosevich, J. A. Biotechniques 1994, 17, 166–171.