Structure-based drug design: Synthesis, crystal structure, biological evaluation and docking studies of mono- and bis-benzo[b]oxepines as non-steroidal estrogens

Article abstract of DOI:10.1016/j.bmc.2003.08.026

Mono- and bis-benzo[b]oxepine derivatives have been rationally synthesized to meet the molecular requirement for interaction with estrogen receptor. Bis-benzo[b]oxepines (7 and 9) and mono-benzo[b]oxepine (10) acquire geometry with phenolic groups disposed in a fashion to stimulate estrogen receptor. Structure-based investigation, in vivo activity and docking studies have been described and correlated to demonstrate a practical approach for suitable ligand design.

Full text of DOI:10.1016/j.bmc.2003.08.026

Bioorganic & Medicinal Chemistry 11 (2003) 5025–5033
Structure-Based Drug Design: Synthesis, Crystal Structure,
Biological Evaluation and Docking Studies of Mono- and
Bis-benzo[b]oxepines as Non-steroidal Estrogens§
Sanjay Sarkhel,a,y Ashoke Sharon,a,y Vishal Trivedi,a Prakas R. Maulik,a,*
Man Mohan Singh,b Paloth Venugopaland and Suprabhat Rayc,*
aMolecular and Structural Biology Division, Central Drug Research Institute, Lucknow 226001, India
bEndocrinology Division, Central Drug Research Institute, Lucknow 226001, India
cMedicinal Chemistry Division, Central Drug Research Institute, Lucknow 226001, India
dDepartment of Chemistry, Punjab University, Chandigarh 160014, India
Received 17July 2003; revised 27August 2003; accepted 29 August 2003
Abstract—Mono- and bis-benzo[b]oxepine derivatives have been rationally synthesized to meet the molecular requirement for
interaction with estrogen receptor. Bis-benzo[b]oxepines (7 and 9) and mono-benzo[b]oxepine (10) acquire geometry with phenolic
groups disposed in a fashion to stimulate estrogen receptor. Structure-based investigation, in vivo activity and docking studies have
been described and correlated to demonstrate a practical approach for suitable ligand design.
# 2003 Elsevier Ltd. All rights reserved.
Introduction
ovascular systems, including raloxifene,8À10 tamox-
ifen11,12 and indole analogues.13
The central role of endogenous estrogen in the develop-
ment and maintenance of the female reproductive
organs and other sexual characteristics has been long
recognized. More recently their involvement in estrogen
deficient syndrome such as osteoporosis,1 Alzheimer
and cardiovascular diseases in females as well as males
has also been established.2À5 Endogenous estrogens are
also known to play an important role in lipid, choles-
terol metabolism, the skeletal as well as, the central
The development of nonsteroidal estrogens without
the usual drawback of endogenous estrogen for the
treatment and prevention of diseases associated with
estrogen deficiency, has therefore, attracted much
attention.
Structural studies of first nonsteroidal oral contra-
ceptive14 centchroman (III) reveals the importance of
the conformation–activity relationship.15,16 It is for this
reason; synthesis of novel estrogens and study of their
structural features with
nervous systems and reproductive functions.6,7 Various
classes of molecules are being developed as alternatives
to estradiol-replacement therapy on skeletal and cardi-
§CDRI Communication Number: 6434.
*Corresponding authors. Tel.: +91-522-2212411-18; fax: +91-522-
2223405; e-mail: maulik_prakas@yahoo.com (P.R. Maulik)
yBoth authors contributed equally to this work.
respect to their biological activity was undertaken.
Although molecular requirement for a ligand to interact
0968-0896/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2003.08.026

5026
S. Sarkhel et al. / Bioorg. Med. Chem. 11 (2003) 5025–5033
with estrogen receptor has been reasonably delineated,17
but finer aspects which make them tissue selective is
poorly understood. Since diethyl stilbestrol (DES) (I)
and d-Estradiol (II) bind to estrogen receptor (ER) with
high affinity, possess molecular similarity in having
hydroxy groups projected in space in a particular
orientation, it was conceived that mono-benzo[b]ox-
epine (IV) and bis-benzo[b]oxepine derivatives (V),
which resemble with DES, would interact with ER and
may provide estrogenic ligands of interest.
phenoxy)propane (1) and 1-(3-bromopropoxy)-3-meth-
oxybenzene (2) as minor and major products,
respectively. Compound 2 was treated with NaCN in
aqueous medium at elevated temperature to give the
corresponding nitrile 3 which on hydrolysis followed by
esterification in the presence of H2SO4 gave an ester 4
which on saponification led to yield corresponding car-
boxylic acid 5. Cyclization of 5 in polyphosphoric acid
under anhydrous condition gave 8-methoxy-3,4-dihy-
dro-1-benzoxepin-5(2H)-one (6). Reductive coupling of
6 with 4-benzyloxybenzaldehyde in the presence of Zn/
TiCl4 under anhydrous condition yielded 7 and 8. The
mechanism of above reductive coupling reaction is
probably that proposed by McMurray.27 It involved an
electron transfer and generation of an anion radical
followed by dimerization. Coordination of the pinaco-
late to the surface of a small zero valent Ti particle and
subsequent cleavage of the C–O bonds lead to the for-
mation of the alkene. Self dimerization resulted in 7
whereas reaction of 6 with benzyloxy benzaldehyde unit
produced 8. Pd/C-catalyzed hydrogenation of 7 and 8 at
30 psi gave 9 and 10, respectively. Condensation of 10
with 1-(2-chloroethyl)piperidine hydrochloride in ace-
tone in the presence of K2CO3 under reflux gave the
desired product which, on subsequent treatment with
dry ethereal HCl, afforded 11. All chiral compounds are
in the racemate forms.
The accurate prediction of protein–ligand interaction
geometries is essential for the success of virtual screen-
ing approaches in structure-based drug design. It
requires docking tools that are able to generate suitable
configuration and conformations of a ligand within a
protein binding site and energetic measurements
describing the quality of the interaction.18À22 For this
purpose an ‘energy-driven’ docking method was
required that is solely based on the direct optimization
of an energy criterion23 and does not rely on other geo-
metric or combinatorial rules for generating docked
ligand orientations and conformations. Traditionally,
Monte Carlo simulated annealing has been used as
optimization algorithm with the AutoDock pro-
gram.24,25 In the most recent version available, a so-
called Lamarckian genetic algorithm AutoDock 3.0.5
has been implemented as more efficient alternative.26
Here, we describe a rational approach for the synthesis,
structural correlation, biological evaluation and dock-
ing studies of novel benzo[b]oxepine ER-ligands (IV
and V).
Result and Discussion
Ligand’s structural studies
The stereochemistry at C5 double bond in 7 could be
either syn or anti-conformation. The relative disposition
of the rings would decide whether the compound elicits
biological response and mimics estradiol (II). This
would then require the rings to be anti-conformation
with respect to each other so as to reach out to the D-
ring region of estradiol. A NOE experiment indicates
Chemistry
Preparations of the desired prototypes were carried out
starting from m-methoxy phenol (Scheme 1). Its con-
densation with 1,3-dibromopropane in aqueous medium
in the presence of base gave 1,3-bis-(3-methoxy-
Scheme 1. Reagents and conditions: (i) aq NaOH; (ii) NaCN/aq ethanol; (iii) abs. ethanol/conc H2SO4; (iv) aq NaOH; (v) PPA; (vi) 4-benzylox-
ybenzaldehyde/Zn/TiCl4/THF; (viii) H2-Pd/C; (ix) 1-(2-chlorohydrochloride/KOH, HCl.

S. Sarkhel et al. / Bioorg. Med. Chem. 11 (2003) 5025–5033
5027
for a possible E-conformation in compound 7. The
observed NOE H1–H3=2% and H3–H1=2% indi-
cated for a possible E-geometry. The observed H2–
H1=12% and H1–H2=12% indicated equal NOE for
these neighbouring protons. Diffraction quality crystals
were obtained only for compounds 7 and 9. Accord-
ingly, their crystallographic studies were undertaken to
ascertain the conformation of the two benzo[b]oxepine
moieties with respect to each other. The molecular
structure and conformation of 7, as assigned by single
crystal X-ray analysis28 is shown in Figure 1(a). There is
half a molecule in one asymmetric unit. The benzo
[b]oxepine moieties are anti- to each other. The aro-
matic ring is planar while the seven-membered oxepine
ring is puckered and adopts a distorted chair con-
formation. The crystal packing [Fig. 1(b)] reveals that
there are two intermolecular C–H. . .O interactions. The
O1 atom in the oxepine ring forms bifurcated hydrogen
bonds with H7and H11A of the other molecule. This is
an interesting observation in light of the documented
fact that the 3-hydroxy of estradiol forms three hydrogen
bonding interactions at the receptor site.17 From the
packing diagram [Fig. 1(b)], it is clear that methyl carbon
C11 and aromatic C7carbon atom may acts as a hydro-
gen bond donor. The H7. . .O1 distance of 2.6 A is smal-
ler compared to that of the H11A, indicates about more
acidic nature of H7than the methyl proton H11A. This
has some implications. It is possible that the H7atom also
takes part in H-bonding (weak) interactions along with the
stronger O–H. . . interactions at the receptor binding site.
The importance of these interactions towards the sta-
bilization of the protein structures29 has been recently
highlighted. The idea behind the synthesis and study of
the crystal structure of 928 [Fig. 2(a)] was to investigate
the effect of reduction of C5–C5B double bond on the
molecular conformation/interaction and biological
activity. The resulted single bond provide a certain
degree of flexibility to the molecule, which might lead to
a better binding at the receptor active site. Interestingly
the intermolecular interaction pattern for 9 is quite dif-
ferent but the overall conformations of the molecules (7
and 9) are quite similar.
Interaction pattern from crystal packing of 9 [Fig. 2(b)]
reveals that there is a intermolecular C–H. . .O interac-
tion (O10. . .H11–C11) involving oxygen atom of meth-
oxy moiety leading to diad structure. It supports the
possibility for the formation of better complex of 9
through H-bonding with estrogen receptor. It is evident
from the biological data that 7 as well as 9 elicit slight
estrogenicity (Table 1).30 A flexible alignment of 9 with
estradiol were performed using advance program MOE-
Flex Align,31 clearly indicates the similarity in their
structures as far as disposing off the methoxy groups of
9 in place of hydroxy groups at the 17-positions of the
estradiol is concerned. Best scored superposition (Fig. 3)
shows matching the aromatic ring A of estradiol (II)
with the corresponding phenyl ring of 9 and the oxygen
atom of the methoxy group matched with that of
the hydroxyl group for estradiol. The hydroxyl group at
Figure 1. (a) ORTEP diagram showing the crystal structure of 7 with atomic numbering scheme; (b) partial crystal packing diagram of 7 showing
weak H-bonding (dotted line) involving O1 atom.
Figure 2. (a) ORTEP diagram showing the crystal structure of 9 with atomic numbering scheme; (b) partial crystal-packing diagram showing the
dimerization of molecules (9) through C–H. . .O bond involving O10 (Phenoxy) atom.

5028
S. Sarkhel et al. / Bioorg. Med. Chem. 11 (2003) 5025–5033
Table 1. Biological activity data (in vivo)30
Compd
Daily dose
(mg/kg)a
Uterine wt (mg)b
Vaginal
cornification
(%)
Absolute
10 g body weight
Ethynyl-estradiol
7
9
Vehicle
11
Vehicle
10
Vehicle
0.02
10
89.6Æ13.5
55.0Æ2.716.5
62.0Æ5.2
21.3Æ2.3
Æ0.8
100
30–40
30
Nil
Nil
Nil
80–90
Nil
10
10
10
10
10
10
19.8Æ2.0
5.2Æ0.3
12.1Æ0.5
Æ0.2
16.8Æ0.8
58.7Æ3.2
15.5Æ0.73.6
101.7Æ4.3
17.0Æ0.81
25.6Æ1.4
6.8Æ0.4
aOnce daily for 3 days, po; six animals per group.
bMeanÆSEM.
Figure 4. Chemical structure of compound 10.
Figure 3. A flexible overlay of the molecule 9 (green) and estradiol
(red) showing significant molecular alignment.31
3-position was considered as the most essential feature
(Pharmacophore) as far as the binding with the ER is
concerned. The flexible superposition (Fig. 3) of 9
clearly shows that a part of the puckered oxepine ring
projects into the 7a-region of estradiol whereas a part of
the ring (from the other half of the molecule) occupies
the 11b-position of estradiol. Both these positions are
known to be tolerant to steric bulk.8 The moderate
estrogenicity elicited by these compounds might then be
rationalized in light of the above observations. The
molecular structure of 7 and 9 thus explains the possible
mimicking of the binding requirements of estradiol
through hydrogen bonding at the receptor active site.
Figure 5. Superposition of the docked estradiol molecule (green) with
crystallographic estradiol molecule (yellow) complexed in the binding
cavity of ER. The H-bonding are shown by broken white lines.
the higher stability of the formed complex between ER
and 10 resulting from D-mode of binding (Fig. 4). This
was further well established by molecular modeling
studies. Regarding the estrogenic profile of 11, absence
of vaginal cornification but gain in weight of uterus may
be attributed to non-specific mobilization of water or
tissue-fluid in the uterus. It is postulated in the case of
11, where both hydroxyl groups protected, the amino
alkoxy side chain may not allow for adopt a suitable
conformation for mimicking the estradiol and that may
be the reason towards its poor estrogenecity. The pre-
sence of a free hydroxy group in compound 10 in place
of the alkyl amino alkoxy chain as in 11, would explain
its relatively higher estrogenic activity.
In light of the observation regarding the exceptionally
high estrogenic profile of 10, the free phenolic group
might mimic the A ring of estradiol so as to dispose the
hydroxyl group in the same space as that of the 3-OH
group of estradiol. The 3-OH of estradiol contributed
more significantly (3-fold) as compared to the 17b-OH
towards the binding stability with the ER.32 The high
estrogenecity elicited by 10 might be a consequence of

S. Sarkhel et al. / Bioorg. Med. Chem. 11 (2003) 5025–5033
5029
Table 2. Docking results
Compd
Cluster rank
RMSD from
estradiol (II)
Final intermolecular
energy (kcal/mol)
Final docked
energy (kcal/mol)
II (ref)
7
9
10
11
I
I
I
I
I
0.69
2.79
1.66
1.85
4.71
À10.71
À9.98
À10.71
À9.34
10.08
À10.98
À10.45
À12.91
À10.44
À7.77
Figure 6. (a) Docked compound 7 in the binding cavity of ER showing interaction with neighbouring residues through H-bonding (broken white
lines); (b) docking of the compound 9 in the binding cavity of ER showing the interaction with neighbouring residues through H-bonding (broken
white lines); (c) docking of the compound 10 in the binding cavity of ER showing the interaction with neighbouring residues through H-bonding
(broken white lines).
Docking studies
The AutoDock predicted conformation of II (estradiol,
green) is shown in Figure 5 with the X-ray crystal-
lographic obtained conformation17 (yellow) super-
position. The root mean square deviation (RMSD)
between these two conformations is ꢀ0.69 A, indicating
that the parameter set for the AutoDock simulation is
reasonable to reproduce the X-ray structure. The Auto-
Dock method and the parameter set could be extended
The docking values for these compounds were calcu-
lated as described in the experimental section and are
summarized in Table 2. From the docking experiment
and in vivo activity, it is evident that docking scores
indeed correlated to in vivo activity of the compounds
and X-ray structural studies of ligands.

5030
S. Sarkhel et al. / Bioorg. Med. Chem. 11 (2003) 5025–5033
Table 3. Docking results
Compd
Cluster rank
RMSD from
estradiol (II)
Final intermolecular
energy (kcal/mol)
Final docked
energy (kcal/mol)
II (ref)
12
13
14
15
I
I
I
I
I
I
0.69
2.51
2.18
1.85
1.96
2.28
À10.71
À11.90
À11.14
À10.88
À12.02
À10.42
À10.71
À11.67
À11.14
À9.72
À11.63
À10.63
16
increases receptor binding significantly but no addi-
tional advantage is seen with dihydroxy compounds (13,
15 and 16). Significantly lowered intermolecular ener-
gies (Table 3) of hydroxyl compounds 12–15 corre-
sponds to 7 and 9 suggest partial or slow in vivo
metabolism of 7 and 9 to its demethylated products.
Thus moderate in-vivo estrogenicity shown by
compound 7 and 9 justified.
Experimental
All reactions were monitored by thin-layer chromato-
graphy over precoated silica gel plates and visualized by
irradiation or exposure to iodine vapours. The melting
points were recorded on electrically heated block and
are uncorrected. NMR spectra were recorded on Bruker
Avance 200 and 300 MHz FT spectrometers using TMS
as internal reference. Chemical shifts and coupling con-
stants (J) were repported in d (ppm) and in Hz, respec-
tively. Mass spectra of the compounds were taken with
Jeol-JMS-D-300 instrument. Microanalysis was carried
out on Carlo Ebra model EA-11108 and Heraeus CHN
rapid instrument. IR spectra of liquid samples were run
neat, and solids as KBr pellets.
Scheme 2.
to search the binding conformations for other ligands
accordingly. Table 2 lists the calculated energies and
docked results from the compounds 7, 9, 10 and 11 with
comparison of docked estradiol.
Figure 6(a–c) shows the 3-D docked model of com-
pound 7, 9 and 10 with ER. Figure 6 illustrates the
probable binding conformational alignment of the syn-
thesized compounds. Figure 6(a) indicates that there is a
short contact between water (3FO) and C11 atom of 7
within an ER, as well as lack of good fit, represented by
poor energy parameters through docking results. Figure
6(b) also indicates the same result as Figure 6(a) but better
fit in compared to 7. Interestingly this result is also sup-
ported by initial crystallographic discussion, where pack-
ing diagram showed the involvement of O10 in hydrogen
bonding, only in the case of 9. Hydrogen-bonding is one of
the important characteristics of the ligand-receptor inter-
action. It was nicely found in the case of compound 10, as
shown in Figure 6(c), for which similar binding pattern
and energy parameters were obtained as were found in
original X-ray structure of ER–estradiol complex.
Preparation
1-(3-Bromopropoxy)-3-methoxybenzene (2). A 25% aqu-
eous solution of NaOH (35 mL) was added to a mixture
of 3-methoxyphenol (10 g, 0.081 mol), 1,3-dibromopro-
pane (32 g, 0.158 mol) in boiling water (70 mL) and
heating was continued with stirring for 10 h. The reac-
tion was monitored through TLC and extracted with
ether. The ethereal layer was dried over anhydrous
CaCl2 and concentrated to dryness. The crude product
was chromatographed over silica gel (60–120 mesh)
using 5% ethyl acetate–hexane mixture as an elutant.
The oily product was identified as 2.
From the above results it could be speculated that con-
formation of the compound 10 [Fig. 6(c)] (the most
potent ligand among the four compounds) is suitable
for interaction with ER. Interestingly only compound
10 contains a phenolic group, which plays an important
role in receptor binding. So we undertook modeling
experiments taking hydroxy compounds (12–16)
(Scheme 2) in place of methoxy. The result of virtual
docking of free phenolic compounds, shown in Table 3,
would suggest that the presence of mono-hydroxy group
Yield: 14.5 g (73.5%), oil, MS (EI): 245 (M+), 1H NMR
(CDCl3): d 2.307(m, 2H, CH ), 3.598 (m, 2H, CH2),
2
3.79 (s, 3H, OCH3), 4.087(m, 2H, OCH ), 6.493 (m,
3H, ArH), 7.181 (m, 1H, ArH).
2
1-Cyano-3-(3-mehtoxyphenoxy)propane (3). A mixture
of 2 (13 g, 0.053 mol), NaCN (6 g, 0.122 mol), 95%
Ethanol (100 mL) and water (30 mL) was heated under
reflux. After 40 h the TLC monitored showed two major
spots, the upper corresponding to starting material 1.

S. Sarkhel et al. / Bioorg. Med. Chem. 11 (2003) 5025–5033
5031
Further refluxing did not show any appreciable change
in TLC. The reaction mixture was concentrated,
extracted with ethyl acetate and the organic layer was
dried over anhydrous sodium sulphate and removed the
solvent. The crude reaction mixture was chromato-
graphed over silica gel (60–120 mesh) using 20% ben-
zene-hexane mixture as eluent. The pure compound was
obtained as an oil.
5,50 -Bi[8-methoxy-2,3,4,5-tetrahydro-1-benzoxepin]yli-
dene (6) and 4-{[(8-methoxy-3,4-dihydro-1-benzoxepin-
5(2H)-ylidene]methyl}phenol (7). To a mixture of Zn
(3.6 g) and dry THF (70 mL) at 40–45 ꢁC, TiCl4 (3.2
mL) was added with and refluxed under anhydrous
condition for 2 h. To this a mixture of 6 (1.0 g, 5 mmol)
and p-benzyloxybenzaldehyde (1.65 g, 8 mmol), was
added slowly, refluxed for 5 h, filtered and the filtrate
poured into water. It was acidified with HCl and
extracted with ethyl acetate. The organic layer was dried
over anhydrous Na2SO4 and concentrated. The crude
extract was flash chromatographed over silica gel to
obtain 7 and 8. Compound 7 and 8 were recrystallized
from EtOAc–hexane mixture.
1
Yield 5.4 g (53%), oil, MS (EI): 191 (M+), H NMR
(CDCl3): d 2.127(m, 2H, CH 2), 2.582 (t, 2H, CH2),
3.790 (s, 3H, OCH3), 4.054 (t, 2H, OCH2), 6.477 (m,
3H, ArH), 7.186 (t, 1H, ArH).
Ethyl-4-(3-methoxyphenoxy)butyrate (4). A mixture of
3 (5 g, 0.026 mol), absolute ethanol (30 mL) and
concd H2SO4 (4 mL) was heated under reflux for 25
h. The reaction mixture was extracted with ethyl
acetate. The organic layer was washed with saturated
NaHCO3 solution and dried over anhydrous sodium
sulphate. The crude product was then chromato-
graphed over silica gel using 3% EtOAc–hexane as
the eluent.
Yield 7: 0.35 g (19%), mp: 132 ꢁC, MS (EI): 352 (M+),
1H NMR (CDCl3): d 1.64–1.75 (m, 4H, CH2), 1.78–1.82
(m, 4H, CH2), 3.74 (s, 6H, OCH3), 4.22–4.34 (t, 4H,
OCH2), 6.21–6.65 (m, 4H, ArH), 7.09–7.13 (d, 2H,
ArH). Anal. calcd for C22H24O4: C, 74.98; H, 6.86; O,
18.16. Found: C, 74.74; H, 6.96; O, 18.30.
Yield 8: 0.25 g (13%), Oil, MS (EI): 372 (M+), Anal.
calcd for C25H24O3: C, 80.62; H, 6.49; O, 12.89. Found:
C, 80.32; H, 6.89.
1
Yield 4.67g (75%), oil, MS (EI): 238 (M +), H NMR
(CDCl3): d 1.255 (t, 3H, CH3), 2.097(m, 2H, CH 2),
2.507(t, 2H, CH ), 3.777 (s, 3H, OCH3), 3.983 (t, 2H,
OCH2), 4.130 (q, 2H, CH2), 6.480 (m, 3H, ArH), 7.163
(t, 1H, ArH).
5,50-Bis[8-methoxy-2,3,4,5-tetrahydro-1-benzoxepin] (9).
Hydrogenation of 7 (200 mg, 0.568 mmol) was carried
out over Pd/C at 30 psi pressure for 7h in methanol.
The mixture was filtered and the filtrate was con-
centrated to yield a semi-solid residue. Which was crys-
tallized from ethyl acetate to yield 9.
2
4-(3-Mehtoxyphenoxy)butyric acid (5). A mixture of 4
(4 g, 0.168 mol) and a solution of 20% KOH was
heated under reflux for 2 h. The reaction mixture was
concentrated and poured into water. It was acidified
with dil. HCl and extacted with ethyl acetate. The
organic layer was washed with water and dried over
anhydrous sodium sulphate. Concentration of the
organic layer yielded 4 which was recrystallized from
hexane.
Yield: 90 mg (45%), mp: 186–188 ꢁC, MS (EI): 354
(M+), 1H NMR (CDCl3): d 1.48 (m, 4H, CH2), 1.55 (m,
4H, CH2), 2.22–2.24 (m, 4H, CH2), 3.79 (s, 6H, OCH3),
4.38–4.44 (t, 4H, OCH2), 6.55–6.66 (m, 4H, ArH), 7.15
(d, 2H, ArH). Anal. calcd for C22H26O4: C, 74.55; H,
7.39; O, 18.06. Found: C, 74.46; H, 7.67.
Yield 3.3 g (93.48%), solid, mp: 70 ꢁC, MS (EI): 210
4-[(8-Methoxy-2,3,4,5-tetrahydro-1-benzoxepin-5-yl)me-
thyl]phenol (10). Hydrogenation of 8 (200 mg, 0.537
mmol) was carried out under similar condition as in
case of 7. Recrystallization form ethyl acetate gave the
crude product as a solid.
1
(M+), H NMR (CDCl3): d 2.012 (m, 2H, CH2), 2.497
(t, 2H, CH2), 3.782 (s, 3H, OCH3), 3.88 (t, 2H, OCH2),
6.461 (m, 3H, ArH), 7.083 (m, 1H, ArH).
8-Methoxy-3,4-dihydro-1-benzoxepin-5-(2H)-one (6). A
mixture of polyphosphoric acid (PPA) (35 g) and 5 (3g,
0.014 mol) was taken in a dried R. B. flask fitted with a
guard tube. The mixture was heated for 3 h on a steam
bath. The colour of the mixture turned red. The reac-
tion mixture was poured into water (70 mL) and left
overnight. It was extracted with ethyl acetate, washed
with NaHCO3 solution and the remaining insoluble
gummy residue was dissolved in dry methanol. The
combined ethyl acetate and methanol extracts were
concentrated to yield the crude product, which was
purified from column chromatography using 5%
EtOAc–hexane as the eluent.
Yield: 130 mg (85%), mp: 97 ꢁC, MS (EI): 284 (M+), 1H
NMR (CDCl3): d 1.41 (m, 2H, CH2), 1.62–1.72 (m, 2H,
CH2), 2.75 (m, 1H, CH2), 2.83–2.95 (m, 2H, CH2), 3.72
(s, 3H, OCH3), 4.15 (t, 2H, OCH2), 6.35–6.79 (m, 7H,
ArH).
1-(2-{4-[(8-Methoxy-2,3,4,5-tetrahydro-1-benzoxepin-5-
yl)methyl]phenoxyl}ethyl)piperidine hydrochloride (11).
A mixture of (10) (130 mg, 0.458 mmol), chloroethyl
piperidine hydrochloride (120 mg), and a pellet of KOH
was heated under reflux for 3 h. The reaction mixture
was extracted with ethyl acetate, dried over anhydrous
Na2SO4 and concentrated. The crude residue was chro-
matographed over silica gel and the product was
obtained as a viscous oil. The hydrochloride salt was
prepared by treating with ethereal solution of HCl.
Removal of the ether under vacuum yielded the product
1
Yield 1.35 g (49.5%), MS (EI): 192 (M+), H NMR
(CDCl3): d 2.192 (m, 2H, CH2), 2.867(t, 2H, CH 2),
3.833 (s, 3H, OCH3), 4.241 (t, 2H, OCH2), 6.55–6.68 (m,
3H, ArH), 7.765 (d, 1H, ArH).

5032
S. Sarkhel et al. / Bioorg. Med. Chem. 11 (2003) 5025–5033
11 as a solid. Which was recrystallized from absolute
ethanol–ether mixture.
score function at the level of binding free energy was
derived and adopted in the version of AutoDock3.0.25
The total binding free energy was empirically calibrated
based on the above-stated terms and set of coefficient
factors. The same rational was applied to the system of
benzo[b]oxepine compounds and estrogen receptor in
order to evaluate the binding properties more precisely
than the traditional molecular mechanics method did.
The total binding free energy and corresponding inhibi-
tory constant between compound and ER were calcu-
lated according to the algorithm in the AutoDock 3.0
program.25
1
Yield 26 mg, MS (EI): 395 (M+), H NMR (CDCl3): d
1.27–1.35 s(m, 6H, CH2), 1.47(m, 2H, CH ), 2.20–2.22
2
(m, 4H, CH2), 2.641 (m, 4H, CH2), 2.777 (m, 1H, CH),
2.847(m, 2H, CH 2), 3.766 (s, 3H, OCH3), 4.112 (m, 2H,
OCH2), 4.274 (t, 2H, OCH2), 6.47–6.59 (m, 7H, ArH).
Anal. calcd for C25H33NO3. HCl: C, 69.51; H, 7.93; N,
3.24. Found: C, 69.12; H, 7.91; N, 2.83.
Computational method
Docking, molecular dynamics, energy minimization and
molecular graphics work were performed on a Silicon
Graphics Octane workstation. The molecular alignment
was performed using MOE 2002.03 program. The
genetic algorithm of AutoDock 3.0 has been employed
for docking the benzo[b]oxipene class compounds into
the active sites of estrogen receptor.
Biological activity method
For determination of estrogenic activity,30 immature
female rats ovariectomized 7days earlier were treated
orally with the test agent or the vehicle (Tween-80 in
glass distilled water). 17a-Ethynylestradiol (EE; Sigma
Chemical Co., USA) was dissolved in 2–3 drops of
redistilled ethanol and diluted to the desired concen-
tration with glass distilled water. All treatments were
done by the oral route once daily for 3 consecutive days
and at autopsy 24 h after the last treatment, uterine
fresh weight was taken and premature opening of
vagina and the extent of vaginal cornification, if any,
were recorded.
Docking studies/protocol
The crystal structure of the LBD of estrogen receptor in
complex with the endogenous oestrogen, 17b-oestradiol
(II)17 was recovered from the Brookhaven Protein Data
Bank (http;//http://www.rcsb.org/pdb/) (entry codes
1ERE). Ligand structures of 7 and 9 were obtained in
pdb-format from its crystal structure. The ligand struc-
ture of 10 and 11 were generated in pdb-format from
and energy minimized by 3DChem draw modeling soft-
ware. The advanced docking program AutoDock 3.0
was used to perform the automated molecular docking
for 7–16. The whole docking operation could be stated
as follows. First, the ligand molecules were checked for
polar hydrogens and assigned for Gasteiger–Huckel33
partial atomic charges. Flexible torsions were defined
with the help of AutoTors. This allowed the conforma-
tional search of ligands during the process of docking.
The PDBQ file was created for ligands. For macro-
molecules ER, polar hydrogens were added and Koll-
man-all-atom34 atomic charges were taken and the
atomic solvation parameters were also assigned using
the ADDSOL utility of AutoDock 3.0.5 program. Sec-
ond, the 3-D grid of 0.3 A resolutions was centered on
the active site using AutoGrid algorithm26 to evaluate
the interaction energies between the ligands and the ER-
LBD. In this stage, the ER was embedded in the 3-D
grid and a probe atom was placed at each grid point.
The affinity and electrostatic potential grid were calcu-
lated for each type of atom in the ligands. Third, a series
of the docking parameters were set on. Not only the
atom types but also the generations and the number of
runs for GA algorithm were edited and properly
assigned. The number of generations, energy evalua-
tions, and GA runs were set of 27,000, 250,000 and 20,
respectively. Finally, the docked complexes of ligand-
receptor were selected according to the criteria of inter-
acting energy combined with geometrical matching
quality. These complexes were used for comparative
study and correlation between and activity and its
structural conformations. On the basis of the traditional
molecular force field model of interaction energy, a new
Acknowledgements
Ashoke Sharon is thankful to CSIR for senior research
fellowship. We thank Dr. R. Ravishankar and Amit
Luthra for helpful discussion. Authors are thankful to
RSIC, CDRI, Lucknow for providing spectroscopic
data and elemental analyses.
References and Notes
1. Gowen, M.; Emery, J. G.; Kumanr, S. Emerg. Drugs 2000,
5, 1.
2. Smith, E. P.; Boyd, J.; Frank, G. R.; Takahashi, H.;
Cohen, R. M.; Specker, B.; Williams, T. C.; Lubahn, D. B.;
Korach, K. S. N. Engl. J. Med. 1994, 331, 1056.
3. Carani, C.; Qin, K.; Simoni, M.; Faustinifustini, M.; Ser-
pente, S.; Boyd, J.; Korach, K. S.; Simpson, E. R. N. Engl. J.
Med. 1997, 337, 91.
4. Cosman, F., Dempster, D., Lindsay, R. In Estrogen and
Antiestrogens. Basic and Clinical Aspects; Lindsay, D. W.,
Dempster, V. C., Jordan, Eds.; Lippincott-Raven: Philadel-
phia, 1997; p 151.
5. Farhat, M. Y.; Lavigne, M. C.; Ramwell, P. W. FASEB J.
1996, 10, 615.
6. Jordan, V. C. Breast Cancer Res. Treat. 1995, 36, 267.
7. Gradishar, W. J.; Jordan, V. C. J. Clin. Oncol. 1997, 15,
840.
8. McDonnell, D. P. TEM 1999, 10, 301.
9. Grese, T. A.; Pennington, J. P.; Sluka, J. P.; Adrian, M. D.;
Cole, H. W.; Fuson, T. R.; Magee, D. E.; Phillips, D. L.;
Rowley, E. R.; Shetler, P. K.; Short, L. L.; Venugopalan, M.;
Yang, N. N.; Sato, M.; Blasebrook, A. L.; Bryant, H. U. J.
Med. Chem. 1998, 41, 1272.
10. Black, L. J.; Jones, C. D.; Falcone, J. F. Life Sci. 1983, 28,
1031.

S. Sarkhel et al. / Bioorg. Med. Chem. 11 (2003) 5025–5033
5033
11. Foster, A. B.; Jarman, M.; Leung, O. T.; McCague, R.;
Leclercq, G.; Devleeschouwer, N. J. J. Med. Chem. 1985, 28,
1491.
Hart, W. E.; Belew, R. K.; Olson, A. J. J. Comput. Chem.
1998, 19, 1639.
27. McMurray, J. E. Chem. Rev. 1989, 89, 320.
12. Chander, S. K.; McCague, R.; Luqmani, Y.; Newton, C.;
Dowstt, M.; Jarman, M.; Coombes, R. C. Cancer Res. 1991,
51, 5851.
13. Miller, C. P.; Collini, M. P.; Tran, B. D.; Harris, H. A.;
Kharode, Y. P.; Marzolf, J. T.; Moran, R. A.; Henderson,
R. A.; Bender, R. H. W.; Unwalla, R. J.; Greenberger, L. M.;
Yardley, J. P.; Abou-Gharbia, M. A.; Lyttle, C. R.; Komm,
B. S. J. Med. Chem. 2001, 44, 1654.
28. Crystal data of 7: C22H24O4, M=352.41, monoclinic,
P2(1)/n, a=6.976(1), b=10.951(1), c=12.323(1) A,
b=100.519(10), V=925.58 A3, Z=2, Dc=1.264 g cmÀ1, m
(Mo-Ka)=0.07mm À1, F(000)=376.0, colorless rectangular
crystal, size 0.20ꢂ0.17ꢂ0.12 mm, 1764 reflections measured
(1623 unique), Rw=0.11, conventional R=0.0376 on F values
of 1450 reflections with I>2s(I), S=1.078 for all data and
120 parameters. Crystal data of 9: C22H26O4, M=354.43,
triclinic, P-1, a=5.521(1), b=7.544(1), c=11.358(3) A,
a=105.06(1), b=95.48(2), g=94.52(1), V=452.08 (16) A3,
14. Kamboj, V. P.; Ray, S.; Dhawan, B. N. Drugs Today
1992, 28, 227.
À1
,
15. Ray, S.; Tandon, A.; Dwivedy, I.; Wilson, S. R.; O’Neil,
J. P.; Katzenellenbogen, J. A. J. Med. Chem. 1994, 37, 696.
16. Srivastava, N.; Ray, S.; Dwivedy, I.; Wilson, S. R.; Hom,
R. K.; Katzenellenbogen, J. A. Bioorg. Med. Chem. Lett. 1996,
6, 1747.
17. Brzozowski, A. M.; Pike, A. C. W.; Dauter, Z.; Hubbard,
R. E.; Bonn, T.; Engstrom, O.; Ohman, L.; Greene, L.; Gus-
tafsson, J.-A.; Carlquist, M. Nature 1997, 389, 753.
18. Abagyan, R.; Totrov, M. Curr. Opin. Chem. Biol. 2001, 5,
375.
19. Kuntz, I. D.; Meng, E. C.; Shiochet, B. K. Acc. Chem.
Res. 1994, 27, 117.
20. Lengauer, T.; Rarey, M. Curr. Opin. Struct. Biol. 1996, 6,
402.
21. Dixon, J. S., Blaney, J. M. In Design Bioactive Molecules:
Three-dimensional Techniques and Interactions; Martin, Y. C.,
Willett, P., Eds.; American Chemical Society: Washington,
DC, 1997; p 175.
Z=1, Dc=1.302
g
cmÀ1
colorless
,
m
(Mo-Ka)=0.07mm
F(000)=190.0,
rectangular crystal,
size
0.25ꢂ0.16ꢂ0.06 mm, 2114 reflections measured (1569
unique), Rw=0.11, conventional R=0.0566 on F values of
894 reflections with I>2s(I), S=1.025 for all data and 123
parameters. Unit cell determination and intensity data col-
lection (2y=50ꢁ) was performed on a Siements P4 dif-
fractometer at 293(2) K. Structure solutions by direct
methods and refinements by full-matrix least-squares methods
on F2. Programs: XSCANS (Siemens Analytical X-ray
Instrument Inc.: Madision, WI, USA, 1996) SHELXTL-NT
(Bruker AXS Inc.: Madision, WI, USA, 1997). Further
details of the crystal structure investigation can be obtained
from the Cambridge Crystallographic Data Centre, 12 Union
Road, Cambridge CB2 1EZ, UK (CCDC deposition no. of 7:
215745, CCDC deposition no. of 9: 215746).
29. Vargas, R.; Garza, J.; Dixon, D. A.; Hay, B. P. J. Am.
Chem. Soc. 2000, 122, 4750.
22. Muegge, I., Rarey, M. In Computational Chemistry; Lip-
kowitz, K. B., Boyd, D. B., Eds.; Wiley-VCH: New York,
2001; Vol. 17, p 1.
30. Singh, M. M.; Agnihotri, A.; Garg, S. N.; Agrwal, S. M.;
Gupta, D. N.; Keshri, G.; Kamboj, V. P. Planta Med. 1988, 6,
492.
23. Totrov, M., Abagyan, R. In Drug-receptorthermody-
namics; Induction and Applications. Raffa, R. B., Ed; Wiley:
Chichester, 2001; p 603.
24. Goodsell, D. S.; Olson, A. J. Automated Docking of
Substrates to Proteins by Simulated Annealing. Proteins
Struct. Funct. Genet. 1990, 8, 195.
31. New MOE version 2002.03, Chemical Computing Group,
Inc., Montreal, Quebec, Canada: http://www.chemcomp.com.
32. Anstead, G. M.; Carlon, K. E.; Katzenellenbogen, J. A.
Steroids 1997, 62, 268.
33. Purcel, W. P.; Singer, J. A. J. Chem. Eng. Data 1967, 12,
235.
25. Morris, G. M.; Goodsell, D. S.; Huey, R.; Olson, A. J. J.
Computer-Aided Mol. Des. 1996, 10, 293.
26. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.;
34. Weiner, S. J.; Kollman, P. A.; Case, D. A.; Singh, C.;
Ghio, G.; Alagona, S.; Profeta, P.; Weiner, P. J. Am. Chem.
Soc. 1984, 106, 765.