Towards β-selectivity in functional estrogen receptor antagonists

Article abstract of DOI:10.1039/c2ob26062j

Based on the benzo[b]naphtho[1,2-d]furan and benzo[b]naphtho[1,2-d] thiophene frameworks, a series of ligands with different basic side chains (BSCs) has been synthesized and pharmacologically evaluated. Also, their binding modes have been modelled using docking techniques. It was found that the introduction of a BSC in these systems brings about a decrease of affinity for both estrogen receptors α and β in an in vitro competitive binding assay. However, two full antagonists of the estrogen receptor β (9c and 9f) have been discovered, with potency in the low micromolar concentration in a cell-based luciferase reporter assay, and completely devoid of activity against the α receptor at the same concentration range. Differences in the ERα/ERβ binding modes have also been rationalized with the help of molecular modelling techniques. This interesting functional profile could be used to elucidate the physiological role of each ER subtype.

Full text of DOI:10.1039/c2ob26062j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 7334
www.rsc.org/obc
PAPER
Towards β-selectivity in functional estrogen receptor antagonists†
Jose Juan Rodríguez,a Kamila Filipiak,a,b Maciej Maslyk,a,b Jakub Ciepielski,a,c Sebastian Demkowicz,a,d
Sonia de Pascual-Teresa,e Sonsoles Martín-Santamaría,*a Beatriz de Pascual-Teresaa and Ana Ramos*a
Received 1st June 2012, Accepted 11th July 2012
DOI: 10.1039/c2ob26062j
Based on the benzo[b]naphtho[1,2-d]furan and benzo[b]naphtho[1,2-d]thiophene frameworks, a series of
ligands with different basic side chains (BSCs) has been synthesized and pharmacologically evaluated.
Also, their binding modes have been modelled using docking techniques. It was found that the
introduction of a BSC in these systems brings about a decrease of affinity for both estrogen receptors α
and β in an in vitro competitive binding assay. However, two full antagonists of the estrogen receptor β
(9c and 9f) have been discovered, with potency in the low micromolar concentration in a cell-based
luciferase reporter assay, and completely devoid of activity against the α receptor at the same
concentration range. Differences in the ERα/ERβ binding modes have also been rationalized with the help
of molecular modelling techniques. This interesting functional profile could be used to elucidate the
physiological role of each ER subtype.
for the development of more efficient drugs for the treatment of
Introduction
several disorders, such as cancer, cardiovascular disease, mul-
tiple sclerosis and Alzheimer’s disease.3,4 The use of the ERα
The biological actions of estrogens are manifested through two
genetically distinct estrogen receptors (ERα and ERβ) that
display nonidentical expression patterns in target tissues. Inter-
esting differences have been found in the physiological role of
both receptor subtypes. ERα is predominantly involved in the
development and function of the mammary gland and uterus,
and in the maintenance of metabolic and skeletal homeostasis.
ERβ has more pronounced effects on the central nervous system
and on cellular hyperproliferation.1
Since the discovery of ERβ in 1996, compounds that are selec-
tive in activating or inhibiting both ER subtypes are intensively
sought after.2 The data obtained suggest that the discovery of
compounds that selectively bind ERα or ERβ is of great interest
selective agonist propylpyrazoletriol (PPT) (Fig. 1) has shown
that several classical estrogen-induced tissue responses can be
effectively evoked via ERα alone.5 On the other hand, the
design of highly ERβ selective ligands has proved to be quite
challenging, and several groups have reported attempts to design
this kind of compound using different scaffolds. The highly ERβ
selective agonist ERB-041 (Fig. 1) has been used to demonstrate
that this receptor may be a useful target for certain inflammatory
diseases. This compound has a dramatic beneficial effect in the
HLA-B27 transgenic model of inflammatory bowel disease and
the Lewis rat adjuvant-induced arthritis model, while it is inac-
tive in several classic models of estrogen action.6 Other non-
steroidal scaffolds which have been developed as ERβ ligands are
diarylpropanenitriles,7 2-phenylnaphthalenes,8,9 and phenyl-2H-
indazoles.10
Interestingly, some substituted tetrahydrochrysene ligands
such as cis-(R,R)-diethyl (THC) (Fig. 1) have been described as
potent agonists on ERα, but more potent antagonists on
ERβ,11,12 in contrast with tamoxifen and raloxifene which are
partial antagonists on both ERα and ERβ (Fig. 1). The structure
of these ERβ antagonists is also different from the structure of
tamoxifen and raloxifene, where the bulky basic side chain
(BSC) is responsible for their antagonist activity through the
blockage of the ER helix-12 movement by interaction with
Asp351 carboxylate (ERα numbering).13 Crystallographic struc-
tures of the ERα ligand binding domain (LBD) bound to both
THC and a fragment of the transcriptional coactivator GRIP1,
and ERβ LBD bound to THC show that this compound anta-
gonizes ERβ through a novel mechanism termed “passive
aDepartamento de Química, Facultad de Farmacia, Universidad CEU
San Pablo, 28668-Boadilla del Monte, Madrid, Spain.
E-mail: aramgon@ceu.es, smsantamaria@ceu.es;
Fax: (+34) 913510496; Tel: (+34) 913724796
bDepartment of Molecular Biology, Faculty of Mathematics and Natural
Sciences, The John Paul II Catholic University of Lublin, 20-718
Lublin, Poland
cDepartment of Environmental Biochemistry and Chemistry, Faculty of
Mathematics and Natural Sciences, The John Paul II Catholic
University of Lublin, 20-718 Lublin, Poland
dDepartment of Organic Chemistry, Gdansk University of Technology,
11/12 G. Narutowicza St., 80-233 Gdańsk, Poland
eInstitute of Food Science, Food Technology and Nutrition (ICTAN),
Spanish National Research Council (CSIC), José Antonio Novais 10,
28040-Madrid, Spain
†Electronic supplementary information (ESI) available: δH and δC NMR
spectra of compounds 5b, 6b, 7b, 8a–j, 9a–j, 10 and 11. Data
from docking calculations and MD simulations. See DOI:
10.1039/c2ob26062j
7334 | Org. Biomol. Chem., 2012, 10, 7334–7346
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 2 Chemical structures of compounds 1–3 and 9a.
of the synthesized compounds and studied the transcriptional
efficacy and MCF-7 antiproliferative activity of the most interest-
ing of these. The most active compounds in the series behave as
ERβ-selective functional antagonists, which make them interest-
ing tools to elucidate the biological effects of this receptor
subtype. Molecular modelling studies have helped to rationalize
these findings.
Results and discussion
Chemistry
Fig. 1 Chemical structures of some known ERα- or ERβ-selective
ligands and tamoxifen and raloxifene.
In our previous work we described the synthesis of benzo-
naphthofurans and benzonaphthothiophenes 1–3, and we studied
their affinity towards both estrogen receptors α and β.17 These
compounds were designed bearing a cyano group in their struc-
ture to allow their transformation into antagonists by the intro-
duction of the appropriate BSC. In fact, a basic side chain was
introduced to give 9a (Table 1) via the five step synthetic
pathway depicted in Scheme 1, demonstrating that these are suit-
able scaffolds for the design of potential new SERMs.
In this paper we have extended this study to the synthesis of
compounds 9b–9j (Table 1) following the same synthetic
pathway. Thus, compound 5b was synthesized by reacting 4b
with 4-fluorophenylmagnesium bromide. Treatment of 5b with
BBr3 gave 6b, which was transformed into the benzyl-protected
derivative 7b by reaction with benzyl bromide in the presence of
K2CO3. Both 7a17 and 7b were useful intermediates for the syn-
thesis of the final compounds 9, through nucleophilic aromatic
substitution of the fluorine atom by the corresponding alcohol or
amine using NaH and K2CO3 respectively as a base, followed by
deprotection of the hydroxyl groups. Deprotection using H2 and
Pd/C 5% under different conditions of pressure and temperature
gave poor yields of the deprotected compound.17
antagonism”.14 Other series of ERβ-selective modulators based
on a 1,3,5-triazine scaffold, that behave as ERα partial agonists
and ERβ antagonists, have been identified.15
However, there are few examples of compounds that are more
potent antagonists of ERβ than of ERα.2 Pyrazolo[1,5-a]pyrimi-
dines (Fig. 1) possess this new profile: they are passive on both
ERs, with a distinct potency selectivity in favor of ERβ. In a
recent structure-based virtual screening, one antagonist for both
subtypes, showing significant selectivity for ERβ, was discov-
ered, which showed inhibitory activities on the proliferation of
the MCF-7 cell line.16
Compounds acting as ERβ antagonists are interesting to probe
the biology of this receptor subtype, and they can be useful to
understand the role that the ERβ play in several types of cancers
such as prostate, colon and lung cancers, where it is the predomi-
nant ER subtype.4
With the purpose of extending the available scaffolds useful
for the design of new selective estrogen receptor modulators
(SERMs), we initiated a program directed to the synthesis of tetra-
cyclic systems containing an oxygen or sulphur atom, and an
additional cyano substituent, that could be used to introduce the
appropriate basic chains (Fig. 2). We found that compounds 2 and
3, with modest selectivity for ERβ in a scintillation proximity
assay, behave as ERβ agonists and ERα antagonists,17 and present
interesting antitumor activity against two pancreatic cell lines.18
The aim of the present study is to transform these ERβ ago-
nists into antagonists by the introduction of the BSCs present in
tamoxifen, raloxifene and other antagonists described in the
literature. We have evaluated the ERα and ERβ binding affinities
However, the use of black palladium and ammonium formate
as the source of hydrogen gave excellent yields of compounds 9,
which were converted to the corresponding hydrochloride salts
to be tested in the biological assays.
Molecular modelling
Docking studies. Docking studies were performed on selected
final compounds (9a–9c and 9f–9j, Table 1) with the aim of
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7334–7346 | 7335

View Article Online
Table 1 Chemical structure and yields in the synthesis of compounds
8a–j and 9a–j
studying how different BSCs modulate the interaction with the
receptor. In order to explore different binding modes of the ER
in complex with agonists and antagonists, several crystallo-
graphic structures were used: ERα in complex with estradiol
(PDB 1A52), raloxifene (PDB 1ERR), genistein (PDB 1X7R)
and 4-hydroxytamoxifen (PDB 3ERT), and ERβ in complex
with genistein (PDB 1X7J), THC (PDB 1L2J) and 4-hydroxy-
tamoxifen (PDB 2FSZ). Two different docking programs were
used, AutoDock4 and Glide, to compare the results (see ESI†
for docking energies).
Although compound 6b, with a short side chain, is a synthetic
intermediate, we were also interested in the study of its binding
mode, so it was considered in the docking calculations, as well
as estradiol, genistein and 4-hydroxytamoxifen as reference com-
pounds to validate the computational protocols. Overall results
are here presented; focusing on the different features found in
the binding, therefore relevant biological information can be
inferred for the series of compounds.
Comp.
X
S
R
Yield 8/9 (%)
8a/9a
85/90
8b/9b
8c/9c
S
S
77/59
76/75
For compound 6b, in general, the results of the docking
studies predicted best binding poses towards ERβ, in terms of
docking energy (for example, the value of the Glide score was
ERα = 7.6 kcal mol−1; ERβ = −10.0 kcal mol−1), and were in
agreement with the experimental data of RBA (relative binding
affinity). Hydrogen bonds are established between the
HO-9 group with Arg346 and the carbonyl group of Leu339,
between the HO-3 group with Thr299, and between the carbonyl
group and the NH of His475 (Fig. 3). The side chain occupies a
hydrophobic region delimited by residues Leu380, Ile373 and
Phe377. An alternative binding mode is also observed in which
the aromatic ring provides a stacking interaction with the imida-
zole ring of His475. The docking calculations in ERα predicted
a binding mode in which the two hydroxyl groups establish
hydrogen bonds with Arg394–Glu353, and with Thr347, while
the side chain establishes a stacking interaction with the imida-
zole from His524. As expected, no binding pose with inter-
actions with ERα Asp351 (ERβ Asp303) was predicted (Fig. 3).
For compounds 9a–9c and 9f–9j, docking studies allowed the
identification of binding poses in agreement with the obtained
affinity values (Table 2), noticing key interactions of the tetra-
cycle into the LBD similar to those for genistein and 4-hydroxy-
tamoxifen: hydroxyl groups establish hydrogen bonds with
Glu353, Arg394 and/or the Leu387 carbonyl group in one of the
8d/9d
S
85/90
8e/9e
8f/9f
S
85/99
96/85
O
8g/9g
8h/9h
O
O
93/95
85/71
8i/9i
8j/9j
O
O
85/87
85/79
Scheme 1 Reaction conditions for the synthesis of compounds 8 and 9. (a) 4-Fluorophenylmagnesium bromide, THF reflux for 24 h, 98% for 5b;
(b) BBr3, DCM, 95%; (c) benzyl bromide, K2CO3, EtOH, 95% for 6b; (d) RH/NaH, DMF for 8a–c and 8f–h or RH/K2CO3, DMF for 8d,e and 8i,j;
(e) H2, black palladium, ammonium formate.
7336 | Org. Biomol. Chem., 2012, 10, 7334–7346
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 3 Agonistic and antagonistic profile and antiproliferative
activities of selected compounds
Agonistic activity Antagonistic activity
(EC50) (μM)
(IC50)a (μM)
b
MCF-7 IC50
Ligand
ERα
ERβ
ERα
ERβ
0.592 0.04
0.148 0.0068 3.34 0.71
0.183 0.007 2.25 0.43
(μM)
6b
9c
9f
Weak
—
—
—
—
—
—
—
—
2.42 0.38
a Experimental values represent the average of 2 experiments performed
in triplicate along with standard deviation (SD) between assay values.
b Experimental values represent the average of 3 experiments performed
in quintuplicate along with standard deviation (SD) between assay
values.
Fig. 3 Docked binding mode obtained with Glide for compound 6b in
ERβ.
Table 2 Estrogen receptor relative binding affinity (RBA) and IC50 for
compounds 2, 3, 6b, 9a–c, 9f, 9i, 9j, 10 and 11
a
a
ERα IC50
Ligand (μM)
ERβ IC50
(μM)
RBA
(ERα)
RBA
(ERβ)
β/α
ratio
2
3
7.02
30.83
68.0
1.76
6.0
25.0
49.27
25.29
18.07
12.19
7.38
>50
7.06
NA
31.59
0.128
0.029
0.013
—
0.042
0.07
0.036
0.120
0.025
0.077
NA
0.39
3.05
3.93
2.08
—
0.65
0.54
1.56
0.77
—
0.115
0.027
0.013
0.027
0.038
0.056
0.093
—
6b
9a
9b
9c
9f
9h
9i
9j
10
11
>100
21.40
12.78
24.92
7.46
34.8
11.66
0.097
NA
0.021
1.26
—
0.56
Fig. 4 Docked binding mode obtained with Glide for compound 9f in
ERβ.
NA
23.91
0.037
a Values are an average of at least 3 experiments with typical standard
errors below 15%; NA – not achieving binding at the assayed
concentration.
similar results in both receptors (Fig. 4). These results could
point towards a theoretical ERβ selectivity for 9j and 9f. As will
be shown below, affinity data confirmed these predictions. More-
over, both compounds were shown to be ERβ selective in the
functional characterization studies. Additional MD simulations
were performed on the ERβ–9f complex (see below).
binding site ends, and with His524 in the other end (ERα
numbering).
Careful study of the results reveals some differences depend-
ing on the nature of the tetracycle. For benzothiophene deriva-
tives (9a, 9b and 9c), only docked binding poses into ERβ
showed the characteristic interactions together with one inter-
action of the BSC with Asp303, putting forward the possibility
of antagonist behaviour. Remarkably, compound 9c clearly
showed ERβ selective antagonism in the functional characteriz-
ation studies (Table 3).
Regarding benzofuran derivatives, for compounds 9g, 9h and
9i, several binding poses were predicted without remarkable
ERα/ERβ differences in terms of energy, either with AutoDock
or Glide. Although in the case of ERβ a larger number of
binding poses were obtained, the theoretical binding energies
were very similar, not pointing to a clear prediction of selectivity
for these three compounds. However, this was not the case for
the other benzofuran compounds, 9j and 9f. It is worth mention-
ing that for compound 9j, both docking programs only led to
binding solutions in ERβ. For compound 9f only ERβ binding
solutions were predicted by AutoDock, while Glide showed
Superimposition of docked poses of compound 9f and the
thioderivative 9a led to the observation of a shift of 9a relative
to the position occupied by 9f within the LBD, which is justified
by repulsion of the sulphur atom with an Ala302 side chain.
This shifting also prevents the hydrogen bond formation
between the OH-9 group and Arg346, and might explain the
poor ERβ affinity found for this compound. The larger volume
of the thioligand and this repulsion could also justify that the
docking studies did not provide any binding solution in the ERα
for compound 9a. A similar repulsive interaction, involving the
sulphur atom, was observed in the thioderivative 3,17 thus
accounting for the ERβ selectivity. It is worth mentioning that
for compounds 9a, 9c and 9j docking results were only obtained
in the ERβ. In the agonism/antagonism profile assays, compound
9c showed an IC50 value of 0.148
0.0068 μM as an ERβ
antagonist (see below), which is in agreement with the
prediction.
Regarding the nature of the side chains, docking studies led to
better results for compounds with flexible chains in terms of
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7334–7346 | 7337

View Article Online
energy, number of binding poses and geometry of the interaction
(data not shown), highlighting the interaction of the protonated
BSC nitrogen with the Asp303 carboxylate from helix-12.
However, it is important to mention that this interaction was also
observed in scarce solutions for the piperidine derivatives. In any
case, the best docking results corresponded to the ERβ binding.
Molecular dynamics simulations
Fig. 5 Chemical structures of compounds 10 and 11.
Prompted by the experimentally found ERβ selectivity for com-
pound 9f, molecular dynamics simulations were performed on
the ERβ–9f complex. The starting structure was obtained from
the docking studies (see Experimental). Initially, during the equi-
libration period, positional restraints to α-carbon atoms were
applied, together with a distance restraint to keep the interaction
between the Asp303 carboxylate moiety and the piperidinium
NH group. Then, restrictions were gradually lowered until no
restriction was applied. As a further step, two independent MD
simulations were then continued: the first one maintained the
COO⋯HN distance restraint during the initial 500 ps followed
by 1 ns of simulation with no restriction, while the second one
was run without any restrictions during 2 ns. The analysis of the
results showed that the complex remained fairly stable during the
simulations, maintaining a variety of stabilizing interactions
within the LBD. However, in both cases, the above mentioned
salt bridge was lost within a few ps when the corresponding
restriction was removed. Indeed, it was observed that the Asp303
side chain suffers a conformational change, exposing the carboxy-
late moiety to the solvent (see ESI†). Then, it fluctuates during
the rest of the simulation time, establishing transient hydrogen
bonds with water molecules. In any case, rotation of the helix-12
would be prevented due to the stability of the receptor–ligand
complex, and to the presence of the BSC.
In our previous work17 on ER ligands based on novel tetra-
cyclic scaffolds, the most interesting result was found for com-
pounds 2 and 3, which behaved as ERβ agonists and ERα
antagonists, and presented a 3.5-fold higher affinity toward ERβ.
This result was rationalized, based on molecular modelling
studies, by an additional interaction between the cyano group
present in these compounds and ERβ Thr299, not present in the
docked complex with ERα. In an attempt to improve the affinity
and selectivity of this type of compound, and to get more infor-
mation on the importance of this interaction, we carried out
docking studies of twelve analogues of 2 and 3, by substituting
the cyano group by functional groups able to establish hydrogen
bonds with Thr299.
These functional groups were COOH, COOCH3, CONH2,
CHO, COCH3, and CH2OH, and the docking was carried out in
both alpha and beta ERs, in agonist (PDB codes 1X7E and
1YYE) and antagonist (PDB codes 1ERR and 1L2J) confor-
mations. Docking results were able to highlight proper inter-
actions with Thr299 in the antagonist conformation of ERβ,
without showing remarkable differences between furan/thio-
phene analogues. Finally, compounds 10 and 11 (Fig. 5) were
synthesized to test their biological behaviour. The competitive
binding assay showed that the substitution of the cyano group by
a carboxylic acid (10) brings about a complete loss of affinity. In
the case of ester 11, a decrease in the affinity for ERβ is
observed, while the affinity for ERα is little affected.
Estrogen receptor binding affinity
The RBAs of compounds 6b, 9a–c, 9f, 9h–j, 10 and 11 were
determined in an in vitro competitive binding assay following a
reported method19 with some modifications. To compare the
affinity of these compounds with their parent tetracyclic ana-
logues lacking the basic chain, we have also measured the RBAs
of 2 and 3 in this assay. Table 2 shows a summary of the results
obtained (E2 has an RBA of 100%). Under these conditions, 2
was found to have the highest RBA for ERα (0.128) and ERβ
(0.39), with a slight selectivity for ERβ (β/α = 3.05), which is in
agreement with our previously published affinity results using a
scintillation proximity assay.17 In the case of the benzonaphtho-
furan series, compound 3 showed less affinity than the oxygen
analogue 2, with an RBA of 0.029 for ERα and 0.115 for ERβ,
and similar selectivity (β/α = 3.05). The RBA values for 6b and
9 vary from 0.013 to 0.120 for ERα and from 0.027 to 0.097 for
ERβ, showing that the introduction of a short side chain (com-
pound 6b) or a BSC brings about a decrease in the affinity of
this type of compound and a decrease in selectivity. It should be
noted that under our experimental conditions the binding between
estradiol and ERα is more effective than that of ERβ. Since data
are expressed as a percentage related to estradiol this means that
the actual ratios ERβ/ERα may be even higher than calculated.
In vitro functional activity of selected ER ligands
With the purpose of characterizing the agonist/antagonist profile
of these compounds, we have selected compounds 6b and 9f as
representative ERβ binders and 9c as a representative ERα
binder. The agonistic and antagonistic activities of the com-
pounds were evaluated using a commercially available cell-based
assay (INDIGO Bioscience’s ER Reporter Assay), which allows
quantifying functional activities of the tested compounds,
against ERα and ERβ. The system utilizes non-human mamma-
lian cells engineered to provide constitutive high-level
expression of ERα and ERβ. Additionally, these cells contain
either the ERα or ERβ-responsive luciferase reporter gene. Thus,
quantification of luciferase activity provides a surrogate measure
of ERα and ERβ activation in the treated reporter cells. Although
the compounds possessed only low RBAs, they proved to have a
significant antagonistic effect on ERβ, and almost no effect on
ERα. Compound 6b showed antagonistic activity in ERβ,
already pointing at the potential of benzonaphthofuran 3 to be
7338 | Org. Biomol. Chem., 2012, 10, 7334–7346
This journal is © The Royal Society of Chemistry 2012

View Article Online
converted into an ERβ antagonist through the introduction of the
BSC.
Perkin-Elmer 1330 infrared spectrophotometer. 1H and 13C
NMR were recorded on a Bruker 300-AC instrument. Chemical
shifts (δ) are expressed in parts per million; coupling constants
(J) are in Hertz. Mass spectra were run on a Bruker Esquire 3000
spectrometer. Thin-layer chromatography (TLC) was run on
Merck silica gel 60 F-254 plates. Unless stated otherwise, start-
ing materials used were high-grade commercial products. Com-
pounds 9a–c, 9f and 9h–j were tested as hydrochlorides and
their purities were determined by HPLC on an Agilent 1100
HPLC system with a UV detector, using a C18 reversed-phase
Discovery column (15 × 4.6 mm ID, 5 μm), eluting with an iso-
cratic pH 7 phosphate buffer–methanol (80 : 20 v/v) mobile
phase.
Interestingly, when a BSC was introduced, as in 9c and 9f,
full ERβ antagonists were obtained, with an IC50 of 0.183
0.007 and 0.148 0.0 068 μM, respectively, while they were
completely devoid of activity against the α receptor at the same
concentration range. These results are in agreement with the pre-
dicted lack of interaction of the BSC with ERα Asp351 carboxy-
late for most of the here reported compounds (see docking
studies). Thus, although these ligands possessed low RBAs, and
did not show any significant subtype preference in binding
assays, they proved to share significant degrees of antagonism on
ERβ. Given the limited number of examples of ERβ-selective
antagonists available,4 these compounds could be used as poten-
tial molecular probes to differentiate the biological roles of both
ER subtypes.
(3,9-Methoxybenzo[b]naphtho[1,2-d]furan-5-yl)(4-fluorophenyl)-
methanone 5b. To a solution of 4b17 (0.98 g, 3.23 mmol) in
THF (20 cm3) was added a solution of 4-fluorophenylmagne-
sium bromide (65 cm3, 1 M in THF, 65 mmol) at RT, and the
mixture was refluxed for 24 h. After cooling, HCl 3 N (20 cm3)
was added to the crude, and a red solid, which was characterized
as the imine of 5b, precipitated. The solid was isolated by fil-
tration and after adding HCl 3 N (125 cm3), the suspension was
refluxed for 48 h. The new yellow precipitate formed was
extracted with DCM (3 × 20 cm3) and the organic extracts were
washed with brine, dried (MgSO4), and evaporated to give 5b
(1.26 g, 98%) as a yellow solid, mp 170–171 °C; νmax (KBr)/
cm−1 1620; δH (300 MHz, CDCl3) 3.68 (3 H, s, OMe), 3.93
(3 H, s, OMe), 7.07 (1 H, dd, J 2.4, 8.8, ArH), 7.15–7.20 (3 H,
m, ArH), 7.39 (1 H, dd, J 2.4, 9.3, ArH), 7.76 (1 H, d, J 2.4,
ArH), 7.81 (1 H, s, ArH), 7.94 (2 H, d, J 8.8, ArH), 8.09 (1 H,
d, J 8.8, ArH) and 8.36 (1 H, d, J 8.8, ArH); δC (75.4 MHz;
CDCl3) 55.1, 55.6, 96.3, 105.7, 112.1, 115.4, 115.6, 115.7,
117.1, 119.6, 121.2, 122.6, 123.9, 125.0, 129.7, 132.1, 132.9,
133.0, 135.0, 135.02, 150.6, 157.3, 158.2, 159.8, 163.9, 167.3
and 195.9; EIMS (m/z) 400 [M]+.
In vitro antiproliferative activity
Compounds 6b, 9c and 9f, characterized as full antagonists at
low micromolar concentration, were selected for the assessment
of antiproliferative potencies on the human MCF-7 breast cancer
line (Table 3). All of them displayed activity at higher concen-
trations, with IC50 within the range of 2.25 to 3.34 μM. This
result is in accordance with their antagonistic character. The
modest antiproliferative activity observed was expected, as
MCF-7 is a human breast cancer cell line showing a low level of
ERβ expression.
Conclusions
We have used the benzo[b]naphtho[1,2-d]furan and benzo[b]-
naphtho[1,2-d]thiophene systems, previously described by us as
selective β-agonists,17 to generate a series of antagonists by the
introduction of the BSCs present in tamoxifen, raloxifene and
other antagonists described in the literature. The antagonism is
thus produced through the establishment of an interaction
between this BSC and Asp303 carboxylate (ERβ numbering),
blocking the ER helix-12 movement. Docking studies on our
compounds have pointed towards a clearly preferred ERβ
binding, and have allowed proposing binding modes exhibiting
the interaction of the BSC and Asp303 carboxylate from
helix-12. Among these novel ligands, compounds 9c and 9f pre-
sented a promising profile, with low micromolar activity as ERβ
antagonists in a cell-based functional assay, and no activity in
ERα at the same concentration range.
(3,9-Dihydroxybenzo[b]naphtho[1,2-d]furan-5-yl)(4-fluorophenyl)-
methanone 6b. To a solution of 5b (0.8 g, 2 mmol) in dry DCM
(20 cm3) was added BBr3 (40 cm3, 1 M in DCM, 40 mmol) at
0 °C. The mixture was stirred in a sealed tube at 70 °C for 48 h.
After cooling to room temperature, the crude reaction mixture
was quenched carefully with ice, water and 1 N HCl. The
aqueous layer was extracted with AcOEt (3 × 50 cm3) and the
combined organic extracts were washed with saturated aqueous
NaHCO3 and brine, dried (MgSO4) and concentrated to dryness
to give 6b (0.71 g, 95%) as a solid, mp 224–225 °C; νmax (KBr)/
cm−1 1595 and 3300; δH (300 MHz, MeOD) 6.95 (1 H, dd,
J 2.4, 8.6, ArH), 7.00–7.04 (2 H, m, ArH), 7.09–7.14 (2 H, m,
ArH), 7.25 (1 H, dd, J 2.4, 9.2, ArH), 7.59 (1 H, s, ArH),
7.66–7.71 (2 H, m, ArH), 8.15 (1 H, d, J 8.6, ArH) and 8.45
(1 H, d, J 9.2, ArH); δC (75.4 MHz; MeOD) 99.3, 109.8, 113.5,
116.4, 116.8, 117.7, 120.3, 123.6, 124.2, 126.6, 130.5, 131.9,
132.1, 135.5, 136.0, 136.02, 152.5, 155.8, 158.8, 159.3, 164.4,
167.8 and 179.1; EIMS (m/z) 395 [M + Na]+; HPLC purity:
96.52.
There are few examples of ligands with this β-selective
antagonistic profile. The ability of these compounds to annul the
estrogen action through ERβ, without having any effect on its
activity through ERα, could be used to differentiate the biologi-
cal roles of both ER subtypes.
Experimental
(3,9-Dibenzyloxybenzo[b]naphtho[1,2-d]furan-5-yl)(4-fluoro-
phenyl)methanone 7b. A solution of 6b (0.5 g, 1.34 mmol),
K2CO3 (3.2 g, 23 mmol) and benzyl bromide (0.96 cm3,
8.1 mmol) in EtOH (3 cm3) was refluxed for 24 h. The reaction
General methods
Melting points (uncorrected) were determined on a Stuart Scien-
tific SMP3 apparatus. Infrared (IR) spectra were recorded with a
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7334–7346 | 7339

View Article Online
was then diluted with AcOEt, washed with water and brine,
dried (MgSO4) and evaporated to dryness. The residue was
purified by flash column chromatography using hexane–AcOEt
(9 : 1) as an eluent to give 7b (0.72 g, 82%) as a yellow solid,
mp 155–157 °C; νmax (KBr)/cm−1 1590 and 3020; δH
(300 MHz, CDCl3) 5.11 (2 H, s, CH2O), 5.12 (2 H, s, CH2O),
7.14–7.20 (3 H, m, ArH), 7.36–7.50 (12 H, m, ArH), 7.82–7.86
(2 H, m, ArH), 7.93–7.95 (2 H, m, ArH), 8.27 (1 H, d, J 7.8,
ArH) and 8.56 (1 H, d, J 8.8, ArH); δC (75.4 MHz; CDCl3)
69.8, 70.3, 97.4, 106.9, 112.7, 115.4, 115.5, 115.7, 117.3, 119.9,
121.1, 122.7, 124.0, 125.0, 127.6, 127.9, 128.1, 128.5, 129.6,
132.2, 132.9, 133.0, 134.8, 134.9, 136.3, 136.5, 150.76, 156.5,
158.1, 158.8, 163.9, 167.3 and 195.9; EIMS (m/z) 553
[M + H]+.
added and the stirring was continued for 6 h at 100 °C. To the
solution was added water, and the aqueous layer was extracted
with AcOEt, washed with brine, dried (MgSO4) and evaporated.
The residue was purified by flash column chromatography using
AcOEt as an eluent to give 8d (0.101 g, 85%) as a yellow oil.
νmax (KBr)/cm−1 1591, 2925 and 3356; δH (300 MHz, CDCl3)
1.40–1.55 (2 H, m, CH2), 1.56–1.64 (4 H, m, 2CH2), 2.32–2.49
(4 H, m, 2CH2N), 2.56–2.58 (2 H, m, CH2N), 3.19–3.20 (2 H,
m, CH2N), 5.07 (2 H, s, CH2O), 5.15 (2 H, s, CH2O), 5.21 (1 H,
s, NH), 6.57 (2 H, d, J 8.8, ArH), 7.23–7.49 (13 H, m, ArH),
7.74 (1 H, d, J 2.5, ArH), 7.83 (2 H, d, J 8.8, ArH), 8.02 (1 H, s,
ArH), 8.67 (1 H, d, J 8.8, ArH), and 8.89 (1 H, d, J 9.8, ArH);
δC (75.4 MHz; CDCl3) 24.21, 25.77, 39.27, 54.03, 56.61, 69.75,
70.11, 106.93, 107.52, 111.34, 114.60, 119.10, 122.26, 124.58,
125.56, 126.49, 127.43, 127.67, 127.83, 127.99, 128.38, 128.52,
129.89, 130.57, 131.06, 133.07, 133.46, 134.91, 136.42, 136.49,
142.28, 152.73, 156.01, 157.03 and 195.75; EIMS (m/z) 677
[M + H]+.
(3,9-Dibenzyloxybenzo[b]naphtho[1,2-d]thiophen-5-yl)(4-[2-(N,N-
dimethyl)ethoxy]phenyl)methanone 8b. To
a
solution of
2-(dimethylamino)ethanol (0.31 g, 3.5 mmol) in DMF (5 cm3)
was added NaH (0.08 g, 3.5 mmol) and the mixture was stirred
at RT for 30 min. Then 7a17 (0.1 g, 0.18 mmol) was added, and
the stirring was continued for 6 h. To the solution was added
water, and the aqueous layer was extracted with AcOEt, washed
with brine, dried (MgSO4) and evaporated. The residue was
purified by flash column chromatography using AcOEt–MeOH
(9 : 1) as an eluent to give 8b (0.086 g, 77%) as a yellow oil;
νmax (KBr)/cm−1 1597 and 2931; δH (300 MHz, CDCl3) 2.37
(6 H, s, 2CH3), 2.75–2.84 (2 H, m, CH2N), 4.11–4.20 (2 H, m,
CH2O), 5.08 (2 H, s, CH2O), 5.18 (2 H, s, CH2O), 6.99 (2 H, d,
J 7.8, ArH), 7.24–7.49 (13 H, m, ArH), 7.76 (1 H, s, ArH),
7.90–7.7.92 (3 H, m, ArH), 8.68 (1 H, d, J 8.8 Hz, ArH) and
8.90 (1 H, d, J 9.3 Hz, ArH); δC (75.4 MHz; CDCl3) 45.8, 58.0,
66.1, 69.8, 70.2, 106.9, 107.4, 114.2, 114.8, 119.3, 123.3, 124.7,
125.6, 125.8, 127.5, 127.7, 127.9, 128.1, 128.5, 128.6, 129.8,
131.0, 131.1, 131.3, 132.7, 133.3, 133.7, 136.4, 136.5, 142.6,
156.3, 157.3, 163.0 and 196.4; EIMS (m/z) 638 [M + H]+.
(3,9-Dibenzyloxybenzo[b]naphtho[1,2-d]thien-5-yl)(4-[4-isopropyl-
piperazin-1-yl]phenyl)methanone 8e. The procedure described
above for 8d was used for the synthesis of 8e. From 7a (0.1 g,
0.18 mmol), 4-isopropylpiperazine (0.451 g, 3.52 mmol) and
K2CO3 (0.486 g, 3.5 mmol), 8e (0.1 g, 85%) was obtained as
a yellow oil. νmax (KBr)/cm−1 1591, 2912 and 3443; δH
(300 MHz, CDCl3) 1.09 (6 H, d, J 6.4, 2CH3), 2.63–2.66 (5 H,
m, 2CH2N and CH), 3.37–3.39 (4 H, m, 2CH2N), 5.06 (2 H, s,
CH2O), 5.14 (2 H, s, CH2O), 6.84 (2 H, d, J 8.8, ArH), 7.24
(1 H, dd, J 2.4, 9.3, ArH), 7.37–7.49 (12 H, m, ArH), 7.77 (1 H,
d, J 2.9, ArH), 7.86 (2 H, d, J 8.8, ArH), 7.91 (1 H, s, ArH),
8.67 (1 H, d, J 9.3, ArH) and 8.89 (1 H, d, J 9.8, ArH); δC
(75.4 MHz; CDCl3) 18.39, 47.20, 48.22, 54.36, 69.77, 70.13,
106.91, 107.46, 112.89, 114.66, 119.17, 122.68, 124.62, 125.59,
125.65, 127.44, 127.62, 127.69, 127.86, 128.02, 128.39, 128.54,
129.84, 130.81, 131.07, 132.62, 133.37, 134.43, 136.40, 136.48,
142.39, 154.23, 156.11, 157.09 and 195.86; EIMS (m/z) 677
[M + H]+.
(3,9-Dibenzyloxybenzo[b]naphtho[1,2-d]thien-5-yl)(4-[1-methyl-
piperidine-4-yloxy]phenyl)methanone
8c. The
procedure
described above for 8b was used for the synthesis of 8c. From
7a (0.1 g, 0.18 mmol), 1-methylpiperidine-4-ol (0.41 g,
3.5 mmol) and NaH (0.08 g, 3.5 mmol), 8c (0.088 g, 76%) was
obtained as a yellow oil. νmax (KBr)/cm−1 1597 and 2925;
δH (300 MHz, CDCl3) 1.84–1.99 (2 H, m, CH2), 2.00–2.15
(2 H, m, CH2), 2.26–2.40 (5 H, m, CH2 and CH3), 2.64–2.79
(2 H, m, CH2), 4.42–4.54 (1 H, m, CH), 5.08 (2 H, s, CH2O),
5.20 (2 H, s, CH2O), 6.95 (2 H, d, J 7.8, ArH), 7.26–7.52 (13 H,
m, ArH), 7.77 (1 H, s, ArH), 7.88–7.93 (3 H, m, ArH), 8.70
(1 H, d, J 8.3, ArH) and 8.91 (1 H, d, J 8.8, ArH); δC
(75.4 MHz; CDCl3) 14.11, 20.99, 29.60, 30.39, 45.95, 52.19,
60.31, 69.80, 70.16, 106.91, 107.39, 114.79, 115.10, 119.26,
123.34, 124.71, 125.61, 125.78, 127.46, 127.68, 127.93, 128.08,
128.45, 128.58, 129.76, 130.89, 131.03, 131.30, 132.81, 133.23,
133.66, 136.38, 136.44, 142.57, 156.25, 157.24, 161.75 and
196.34; EIMS (m/z) 664 [M + H]+.
(3,9-Dihydroxybenzo[b]naphtho[1,2-d]furan-5-yl)(4-[2-(piperi-
din-1-yl)ethoxy]phenyl)methanone 8f. The procedure described
above for 8b was used for the synthesis of 8f. From 2-(piperi-
din-1-yl)ethanol (0.47 g, 3.6 mmol), NaH (0.09 g, 3.6 mmol)
and 7b (0.1 g, 0.18 mmol), 8f (0.114 g, 96%) was obtained as a
yellow oil. νmax (KBr)/cm−1 1590 and 3020; δH (300 MHz,
CDCl3) 1.40–1.50 (2 H, m, CH2), 1.58–1.61 (4 H, m, CH2),
2.45–2.64 (4 H, m, CH2), 2.78–2.89 (2 H, m, CH2N), 4.15–4.27
(2 H, m, CH2O), 5.08 (2 H, s, CH2O), 5.18 (2 H, s, CH2O), 6.97
(2 H, d, J 7.3, ArH), 7.17 (1 H, m, ArH), 7.25 (1 H, m, ArH),
7.35–7.52 (11 H, m, ArH), 7.79–7.80 (2 H, m, ArH), 7.90 (2 H,
d, J 7.3, ArH), 8.24 (1 H, d, J 7.83, ArH) and 8.53 (1 H, d,
J 8.28, ArH); δC (75.4 MHz; CDCl3) 24.01, 25.78, 54.96, 57.58,
66.15, 69.82, 70.35, 97.49, 107.06, 112.61, 114.18, 114.70,
117.55, 119.90, 120.38, 122.56, 124.05, 124.99, 127.46, 127.66,
127.90, 128.05, 128.43, 128.57, 129.58, 131.07, 132.75, 133.53,
136.36, 136.50, 151.05, 156.27, 157.95, 158.66, 162.92 and
196.17; EIMS (m/z) 662 [M + H]+.
(3,9-Dibenzyloxybenzo[b]naphtho[1,2-d]thien-5-yl)(4-[2-(piperi-
dine-1-yl)ethylamino]phenyl)methanone 8d. To a solution of
2-(piperidine-1-yl)ethylamine (0.451 g, 3.52 mmol) in DMF
(5 cm3) was added K2CO3 (0.486 g, 3.5 mmol) and the mixture
was stirred at RT for 15 min. Then 7a (0.1 g, 0.18 mmol) was
(3,9-Dibenzyloxybenzo[b]naphtho[1,2-d]furan-5-yl)(4-[2-(N,N-
dimethyl)ethoxy]phenyl)methanone 8g. The same procedure
described above for 8b was used for the synthesis of 8g. From
7340 | Org. Biomol. Chem., 2012, 10, 7334–7346
This journal is © The Royal Society of Chemistry 2012

View Article Online
7b (0.1 g, 0.18 mmol), 2-(dimethylamino)ethanol (0.32 g,
3.6 mmol) and NaH (0.09 g, 3.6 mmol), 8g (0.104 g, 93%) was
obtained as a yellow oil. νmax (KBr)/cm−1 1594 and 2931; δH
(300 MHz, CDCl3) 2.37 (6 H, s, 2CH3), 2.73–2.86 (2 H, m,
CH2N), 4.12–4.21 (2 H, m, CH2), 5.08 (2 H, s, CH2O), 5.16
(2 H, s, CH2O), 6.99 (2 H, d, J 7.8, ArH), 7.14–7.17 (1 H, m,
ArH), 7.23 (1 H, s, ArH), 7.36–7.51 (11 H, m, ArH), 7.80–7.82
(2 H, m, ArH), 7.91 (2 H, d, J 7.3, ArH), 8.23 (1 H, d, J 7.80,
ArH) and 8.51 (1 H, d, J 8.31, ArH); δC (75.4 MHz; CDCl3)
45.78, 57.93, 66.10, 69.79, 70.31, 97.45, 107.05, 112.59,
114.15, 114.73, 117.51, 119.87, 120.38, 122.54, 124.03, 124.97,
127.44, 127.65, 127.88, 128.03, 128.42, 128.55, 129.57, 131.12,
132.72, 133.47, 136.34, 136.49, 151.01, 156.25, 157.92, 158.64,
162.89 and 196.13; EIMS (m/z) 621 [M + H]+.
CDCl3) 1.10 (6 H, d, J 6.4, 2CH3), 2.66–2.78 (5 H, m, 2CH2
and CH), 3.40–3.43 (4 H, m, 2CH2), 5.08 (2 H, s, CH2O),
5.19 (2 H, s, CH2O), 6.88 (2 H, d, J 9.3, ArH), 7.17 (1 H, dd,
J 2.5, 8.8, ArH), 7.27 (1 H, d, J 2.0, ArH), 7.37–7.50 (11 H, m,
ArH), 7.77 (1 H, d, J 2.4, ArH), 7.80 (1 H, s, ArH), 7.84 (2 H,
d, J 8.8, ArH), 8.26 (1 H, d, J 8.8, ArH) and 8.53 (1 H, d, J 9.3,
ArH); δC (75.4 MHz; CDCl3) 18.41, 47.29, 48.27, 54.39, 69.85,
70.37, 97.57, 107.24, 112.50, 112.92, 114.08, 117.70, 119.82,
122.46, 124.07, 124.92, 127.45, 127.69, 127.86, 128.02, 128.41,
128.55, 129.59, 132.67, 134.46, 136.43, 136.58, 151.26, 154.26,
156.13, 157.83, 158.53 and 195.66; EIMS (m/z) 661 [M + H]+.
(3,9-Dihydroxybenzo[b]naphtho[1,2-d]thien-5-yl)(4-[2-(piperi-
dine-1-yl)ethoxy]phenyl)methanone 9a. To a solution of 8a17
(0.096 g, 0.142 mmol) in EtOH–AcOEt–H2O 7 : 3 : 1 (10 cm3)
was added ammonium formate (0.29 g, 4.26 mmol) and black
palladium (0.015 g, 0.142 mmol) and the mixture was stirred at
reflux for 3 h. The black palladium was eliminated by filtration
and the solvent was evaporated. The residue was purified by
flash column chromatography using DCM–MeOH 9 : 1 as an
eluent to give 9a (0.064 g, 90%) as a yellow oil; δH (300 MHz,
MeOD) 1.43–1.47 (2 H, m, CH2), 1.64–1.68 (4 H, m, 2CH2),
3.00–3.06 (4 H, m, 2CH2N), 3.20–3.25 (2 H, m, CH2N),
4.09–4.12 (2 H, m, CH2O), 6.77 (2 H, d, J 9.2, ArH), 7.02 (1 H,
dd, J 2.4, 8.6, ArH), 7.22–7.26 (2 H, m, ArH), 7.32 (1 H, d,
J 2.4, ArH), 7.62 (1 H, s, ArH), 7.65 (2 H, d, J 8.6, ArH), 8.42
(2 H, m, 2OH), 8.55 (1 H, d, J 9.2, ArH) and 8.79 (1 H, d, J 9.2,
ArH); EIMS (m/z) 498 [M + H]+. To 9a was added a solution of
HCl saturated ether and the solution was stirred overnight to give
the hydrochloride compound as a solid which was isolated by fil-
tration, mp 217–218 °C; νmax (KBr)/cm−1 1600, 2680 and 3200;
δH (300 MHz, DMSO) 1.67–1.79 (6 H, m, CH2–piperidine),
2.98–3.01 (2 H, m, CH2–piperidine), 3.45–3.58 (4 H, m, CH2–
piperidine, CH2N), 4.45–5.52 (2 H, m, CH2O), 7.11–7.16 (3 H,
m, ArH), 7.27 (1 H, d, J 2.5, ArH), 7.33 (1 H, dd, J 2.5, 9.2,
ArH), 7.48 (1 H, dd, J 1.8, ArH), 7.81–8.03 (2 H, m, ArH), 8.03
(1 H, s, ArH), 8.77 (1 H, d, J 9.2, ArH), 8.95 (1 H, d, J 9.2,
ArH), 9.94 (1 H, s, OH), 10.05 (1 H, s, NH) and 10.16 (1 H, s,
OH); δC (75.4 MHz; DMSO) 20.6, 21.9, 52.2, 54.1, 62.1, 108.0,
108.3, 114.4, 114.6, 118.7, 121.8, 123.2, 127.3, 130.2, 130.2,
130.6, 131.1, 131.9, 132.6, 141.6, 154.6, 155.9, 161.3 and
195.3; EIMS (m/z) 498 [M + H]+; HPLC > 94.38%.
(3,9-Dibenzyloxybenzo[b]naphtho[1,2-d]furan-5-yl)(4-[1-methyl-
piperidine-4-yloxy]phenyl)methanone
8h. The
procedure
described above for 8b was used for the synthesis of 8h. From
7b (0.1 g, 0.18 mmol), 1-methylpiperidine-4-ol (0.42 g,
3.6 mmol) and NaH (0.09 g, 3.6 mmol), 8h (0.1 g, 85%) was
obtained as a yellow oil. νmax (KBr)/cm−1 1594 and 2931;
δH (300 MHz, CDCl3) 1.82–2.00 (2 H, m, CH2), 2.01–2.16
(2 H, m, CH2), 2.28–2.46 (5 H, m, CH3 and CH2N), 2.67–2.79
(2 H, m, CH2N), 4.42–4.52 (1 H, m, CH), 5.08 (2 H, s, CH2O),
5.17 (2 H, s, CH2O), 6.95 (2 H, d, J 7.8, ArH), 7.15–7.16 (1 H,
m, ArH), 7.24 (1 H, s, ArH), 7.35–7.51 (11 H, m, ArH),
7.80–7.82 (2 H, m, ArH), 7.90 (2 H, d, J 7.8, ArH), 8.24 (1 H,
d, J 8.28, ArH) and 8.52 (1 H, d, J 8.28, ArH); δC (75.4 MHz;
CDCl3) 30.43, 45.99, 52.27, 69.81, 70.34, 97.48, 107.07,
112.61, 114.72, 115.08, 117.54, 119.87, 120.39, 122.57, 124.06,
124.99, 127.46, 127.67, 127.91, 128.06, 128.44, 128.57, 129.58,
130.93, 132.84, 133.49, 136.35, 136.50, 151.04, 156.27, 157.94,
158.66, 161.70 and 196.10; EIMS (m/z) 648 [M + H]+.
(3,9-Dibenzyloxybenzo[b]naphtho[1,2-d]furan-5-yl)(4-[2-(piperi-
dine-1-yl)ethylamino]phenyl)methanone 8i. The procedure
described above for 8d was used for the synthesis of 8i. From 7b
(0.1 g, 0.18 mmol), 2-(piperidine-1-yl)ethylamine (0.463 g,
3.62 mmol) and K2CO3 (0.5 g, 3.62 mmol), 8i (0.101 g, 85%)
was obtained as a yellow oil. νmax (KBr)/cm−1 1591, 2925 and
3356; δH (300 MHz, CDCl3) 1.45–1.48 (2 H, m, CH2),
1.58–1.62 (4 H, m, CH2), 2.38–2.50 (4 H, m, CH2), 2.58–2.62
(2 H, m, CH2N), 3.20–3.26 (2 H, m, CH2N), 5.08 (2 H, s,
CH2O), 5.19 (3 H, m, CH2O and NH), 6.59 (2 H, d, J 8.8, ArH),
7.17 (1 H, dd, J 2.4, 8.8, ArH), 7.26–7.28 (1 H, m, ArH),
7.31–7.55 (11 H, m, ArH), 7.74 (1 H, d, J 2.4, ArH), 7.80–7.83
(3 H, m, ArH), 8.25 (1 H, d, J 8.8, ArH) and 8.53 (1 H, d, J 8.8,
ArH); δC (75.4 MHz; CDCl3) 24.22, 25.77, 39.28, 54.06, 56.62,
69.81, 70.34, 97.52, 107.19, 111.34, 112.42, 113.71, 117.73,
119.52, 119.76, 122.39, 124.043, 124.90, 126.51, 127.45,
127.69, 127.84, 128.01, 128.40, 128.54, 129.54, 133.14, 134.94,
136.42, 136.57, 151.33, 152.71, 156.01, 157.73, 158.43 and
195.57; EIMS (m/z) 661 [M + H]+.
(3,9-Dihydroxybenzo[b]naphtho[1,2-d]thien-5-yl)(4-[2-(N,N-
dimethyl)ethoxy]phenyl)methanone
9b. The
procedure
described above was used for the synthesis of 9b. From 8b
(0.086 g, 0.135 mmol), ammonium formate (0.255 g,
4.05 mmol) and black palladium (0.014 g, 0.135 mmol) in
EtOH (20 cm3), 9b (0.036 g, 59%) was obtained as a yellow oil.
δH (300 MHz, MeOD) 2.88 (6 H, s, 2CH3), 3.45–3.55 (2 H, m,
CH2N), 4.27–4.36 (2 H, m, CH2O), 6.96–6.99 (2 H, d, J 9.2,
ArH), 7.09 (1 H, dd, J 2.4, 8.6, ArH), 7.29–7.33 (2 H, m, ArH),
7.38 (1 H, d, J 2.5, ArH), 7.75–7.79 (3 H, m, ArH), 8.44 (2 H,
m, 2OH), 8.64 (1 H, d, J 9.2, ArH) and 8.87 (1 H, d, J 9.8 Hz,
ArH); δC (75.4 MHz; MeOD) 44.03, 57.48, 63.79, 109.28,
110.35, 115.52, 115.84, 119.83, 123.98, 125.90, 126.26, 127.18,
129.91, 132.57, 132.83, 133.00, 133.43, 133.84, 134.45, 144.14,
156.39, 157.70, 163.39 and 198.46. To 9b was added a solution
of HCl saturated ether and the solution was stirred overnight to
(3,9-Dibenzyloxybenzo[b]naphtho[1,2-d]furan-5-yl)(4-[4-isopropyl-
piperazin-1-yl]phenyl)methanone 8j. The procedure described
above for 8d was used for the synthesis of 8j. From 7b (0.1 g,
0.18 mmol), 4-isopropylpiperazine (0.463 g, 3.62 mmol) and
K2CO3 (0.5 g, 3.62 mmol), 8j (0.101 g, 85%) was obtained as a
yellow oil. νmax (KBr)/cm−1 1588 and 2962; δH (300 MHz,
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7334–7346 | 7341

View Article Online
give the hydrochloride compound as a solid which was isolated
by filtration, mp 188–189 °C; νmax (KBr)/cm−1 1600, 2710 and
3220; δH (300 MHz, DMSO) 2.84 (6 H, s, 2CH3), 3.30–3.49
(2 H, m, CH2N), 4.41–4.43 (2 H, m, CH2O), 7.12–7.15 (3 H, m,
ArH), 7.24–7.25 (1 H, m, ArH), 7.30–7.35 (1 H, m, ArH),
7.46–7.49 (1 H, m, ArH), 7.81 (2 H, m, ArH), 8.02 (1 H, s,
ArH), 7.72–7.80 (1 H, m, ArH), 8.92–8.98 (1 H, m, ArH), 9.70
(1 H, s, NH), 9.92 (1 H, s, OH) and 10.13 (1 H, s, OH); EIMS
(m/z) 458 [M + H]+; HPLC purity: 91.27%.
obtained as a yellow oil. δH (300 MHz, MeOD) 1.02 (6 H, d,
J 6.7, 2CH3), 2.52–2.53 (5 H, m, 2CH2, CH), 3.23–3.25 (4 H,
m, 2CH2N), 6.76 (2 H, d, J 9.2, ArH), 7.11 (1 H, dd, J 2.4, 8.6,
ArH), 7.28–7.34 (3 H, m, ArH), 7.67 (2 H, d, J 8.6, ArH), 7.74
(1 H, s, ArH), 8.68 (1 H, d, J 8.6, ArH) and 8.90 (1 H, d, J 9.2,
ArH); δC (75.4 MHz; MeOD) 18.09, 47.08, 49.24, 56.99,
109.35, 110.47, 114.45, 115.81, 119.82, 123.28, 125.92, 126.15,
126.20, 127.08, 129.04, 130.06, 132.39, 132.62, 133.63, 133.79,
135.31, 144.00, 155.48, 156.22, 157.58 and 198.35. To 9e was
added a solution of HCl saturated ether and the solution was
stirred overnight to give the hydrochloride compound as a solid
which was isolated by filtration; νmax (KBr)/cm−1 1595, 2920
and 3185; δH (300 MHz, DMSO) 1.29 (6 H, d, J 6.6, 2 × CH3),
3.10–3.35 (4 H, m, 2CH2N), 3.49–3.53 (3 H, m, CH2N, CH),
4.10–4.14 (2 H, m, CH2N), 7.08–7.14 (3 H, m, ArH), 7.24 (1 H,
d, J 2.5, ArH), 7.32 (1 H, dd, J 2.6, 9.1, ArH), 7.47 (1 H, d,
J 2.5, ArH), 7.71 (2 H, d, J 8.9, ArH), 7.98 (1 H, s, ArH), 8.75
(1 H, d, J 9.2, ArH), 8.93 (1 H, d, J 9.2, ArH), 9.84 (1 H, s,
OH), 10.06 (1 H, s, OH) and 10.11 (2 H, m, 2NH). HPLC
purity: 92.49%.
(3,9-Dihydroxybenzo[b]naphtho[1,2-d]thien-5-yl)(4-[1-methyl-
piperidine-4-yloxy]phenyl)methanone
9c. The
procedure
described above was used for the synthesis of 9c. From 8c
(0.088 g, 0.133 mmol), ammonium formate (0.251 g,
3.99 mmol) and black palladium (0.014 g, 0.133 mmol) in
EtOH–AcOEt–H2O 7 : 3 : 1 (20 cm3), 9c (0.048 g, 75%) was
obtained as a yellow oil. δH (300 MHz, MeOD) 1.90–2.08 (2 H,
m, CH2), 2.09–2.18 (2 H, m, CH2), 2.72 (3 H, s, CH3),
3.02–3.15 (2 H, m, CH2N), 3.16–3.27 (2 H, m, CH2N),
4.60–4.69 (1 H, m, CH), 6.95 (2 H, d, J 7.8, ArH), 7.09–7.11
(1 H, m, ArH), 7.29–7.39 (3 H, m, ArH), 7.39 (1 H, s, ArH),
7.74–7.77 (3 H, m, ArH), 8.53 (2 H, m, 2OH), 8.65 (1 H, d,
J 9.2, ArH) and 8.88 (1 H, d, J 8.6, ArH). To 9c was added a
solution of HCl saturated ether and the solution was stirred over-
night to give the hydrochloride compound as a solid which was
isolated by filtration, mp 187–188 °C; νmax (KBr)/cm−1 1600,
2715 and 3230; δH (300 MHz, DMSO) 1.87–2.28 (4 H, m,
2CH2–piperidine), 2.76 (3 H, s, CH3), 3.00–3.40 (4 H, m, 2 ×
CH2–piperidine), 4.90 (1 H, m, CH–piperidine), 7.11–7.18 (3 H,
m, ArH), 7.28 (1 H, s, ArH), 7.33 (1 H, dd, J 2.4, 9.2, ArH),
7.48 (1 H, d, J 2.4, ArH), 7.78–7.82 (2 H, m, ArH), 8.04 (1 H,
s, ArH), 8.76 (1 H, d, J 9.2, ArH), 8.95 (1 H, d, J 9.2, ArH),
9.94 (1 H, s, OH), 10.06 (1 H, s, OH) and 10.42 (1 H, s, NH);
δC (75.4 MHz; MeOD) 41.47, 41.83, 47.99, 51.14, 65.99, 69.96,
107.98, 108.00, 108.31, 113.70, 114.56, 114.56, 114.93, 115.33,
118.66, 121.67, 123.19, 127.28, 130.16, 130.25, 131.07, 131.97,
132.65, 132.67, 141.61, 154.57, 155.89, 160.45, 160.68 and
195.15; HPLC purity: 94.38%.
(3,9-Dihydroxybenzo[b]naphtho[1,2-d]furan-5-yl)(4-[2-(piperi-
dine-1-yl)ethoxy]phenyl)methanone 9f. The procedure described
above was used for the synthesis of 9f. From 8f (0.113 g,
0.170 mmol), ammonium formate (0.321 g, 5.1 mmol) and
black palladium (0.018 g, 0.170 mmol), 9f (0.100 g, 85%) was
obtained as a yellow oil. δH (300 MHz, MeOD) 1.15–1.26 (2 H,
m, CH2), 1.30–1.44 (4 H, m, CH2), 2.45–2.61 (4 H, m, CH2),
2.70–2.80 (2 H, m, CH2N), 3.75–3.89 (2 H, m, CH2O), 6.55
(2 H, d, J 7.9, ArH), 6.64–6.71 (2 H, m, ArH), 6.98–7.01 (1 H,
m, ArH), 7.14–7.16 (1 H, m, ArH), 7.26–7.28 (1 H, m, ArH),
7.41 (2 H, d, J 7.3, ArH), 7.85 (1 H, d, J 7.95, ArH), 8.15 (1 H,
d, J 9.15, ArH) and 8.26 (2 H, m, 2OH); δC (75.4 MHz; MeOD)
23.63, 25.07, 55.17, 57.61, 64.78, 99.33, 110.10, 113.71,
115.32, 117.47, 120.35, 121.88, 123.89, 124.29, 126.46, 131.16,
132.59, 133.81, 134.02, 151.59, 156.24, 159.17, 159.65, 163.71
and 198.05. To 9f was added a solution of HCl saturated ether
and the solution was stirred overnight to give the hydrochloride
compound as a solid which was isolated by filtration, mp
267–268 °C; νmax (KBr)/cm−1 1620, 2720 and 3180; δH
(300 MHz, DMSO) 1.67–1.80 (6 H, m, CH2–piperidine),
2.92–3.10 (2 H, m, CH2–piperidine), 3.45–3.59 (4 H, m, CH2–
piperidine, CH2N), 4.50–4.58 (2 H, m, CH2O), 7.02 (1 H, dd,
J 1.9, 8.6, ArH), 7.12–7.15 (3 H, m, ArH), 7.33–7.37 (2 H, m,
ArH), 7.81–7.84 (3 H, m, ArH), 8.39 (1 H, d, J 8.8, ArH), 8.63
(1 H, d, J 8.8, ArH), 9.86 (1 H, s, NH), 10.19 (1 H, s, OH) and
10.70 (1 H, s, OH); δC (75.4 MHz; MeOD) 21.50, 22.30, 52.57,
54.50, 62.69, 98.26, 108.47, 112.86, 113.32, 114.72, 115.34,
119.61, 119.72, 122.03, 122.92, 125.45, 129.26, 131.02, 132.35,
132.81, 149.74, 154.94, 157.57, 157.89, 161.70 and 195.29;
EIMS (m/z) 482 [M + H]+. HPLC purity: 97.59%.
(3,9-Dihydroxybenzo[b]naphtho[1,2-d]thien-5-yl)(4-[2-(piperi-
dine-1-yl)ethylamino]phenyl)methanone 9d. The procedure
described above was used for the synthesis of 9d. From 8d
(0.093 g, 0.137 mmol), ammonium formate (0.258 g,
4.12 mmol) and black palladium (0.015 g, 0.137 mmol) in
MeOH (20 cm3), 9d (0.060 g, 90%) was obtained as a yellow
oil. δH (300 MHz, MeOD) 1.48–1.50 (2 H, m, CH2), 1.65–1.69
(4 H, m, 2CH2), 2.83–2.90 (6 H, m, 3CH2N), 3.34–3.41 (2 H,
m, CH2N), 6.56 (2 H, d, J 8.6, ArH), 7.10 (1 H, dd, J 2.4, 9.2,
ArH), 7.28–7.34 (3 H, m, ArH), 7.64 (2 H, d, J 9.2, ArH), 7.69
(1 H, s, ArH), 8.65 (1 H, d, J 9.2, ArH) and 8.88 (1 H, d, J 9.8,
ArH); δC (75.4 MHz; MeOD) 22.81, 24.26, 38.56, 54.55, 56.51,
109.36, 110.48, 112.63, 115.80, 119.80, 122.83, 125.87, 126.17,
126.18, 126.99, 127.78, 130.03, 132.13, 132.57, 133.64, 134.34,
135.63, 143.85, 154.18, 156.14, 157.50 and 198.28.
(3,9-Dihydroxybenzo[b]naphtho[1,2-d]furan-5-yl)(4-[2-(N,N-
dimethyl)ethoxy]phenyl)methanone
9g. The
procedure
(3,9-Dihydroxybenzo[b]naphtho[1,2-d]thien-5-yl)(4-[4-isopropyl-
piperazin-1-yl]phenyl)methanone 9e. The procedure described
above was used for the synthesis of 9e. From 8e (0.095 g,
0.14 mmol), ammonium formate (0.265 g, 4.21 mmol) and
black palladium (0.015 g, 0.140 mmol), 9e (0.069 g, 99%) was
described above was used for the synthesis of 9g. From 8g
(0.101 g, 0.161 mmol), ammonium formate (0.304 g,
4.83 mmol) and black palladium, 9g (0.067 g, 95%) was
obtained as a yellow oil. δH (300 MHz, MeOD) 2.88 (6 H, s,
2CH3), 3.42–3.55 (2 H, m, CH2N), 4.33–4.44 (2 H, m, CH2O),
7342 | Org. Biomol. Chem., 2012, 10, 7334–7346
This journal is © The Royal Society of Chemistry 2012

View Article Online
6.97–7.03 (4 H, m, ArH), 7.30–7.33 (1 H, m, ArH), 7.43–7.45
(1 H, m, ArH), 7.66 (1 H, s, ArH), 7.81–7.84 (2 H, m, ArH),
8.23 (1 H, d, J 9.15, ArH), 8.51 (2 H, m, 2OH) and 8.53 (1 H,
d, J 9.15, ArH). To 9g was added a solution of HCl saturated
ether and the solution was stirred overnight to give the hydro-
chloride compound as a solid which was isolated by filtration,
mp 136–138 °C; νmax (KBr)/cm−1 1623, 2700 and 3370; δH
(300 MHz, DMSO) 2.84 (6 H, s, 2CH3), 3.49–3.56 (2 H, m,
CH2), 4.41–4.52 (2 H, m, CH2), 7.00–7.02 (1 H, m, ArH),
7.10–7.20 (3 H, m, ArH), 7.32–7.35 (2 H, m, ArH), 7.81–7.84
(3 H, m, ArH), 8.39 (1 H, d, J 8.6, ArH), 8.63 (1 H, d, J 8.6,
ArH), 9.89 (1 H, s, OH), 10.22 (1 H, s, OH) and 10.34 (1 H, s,
NH); δC (75.4 MHz; MeOD) 42.75, 55.09, 62.71, 98.32, 108.52,
112.95, 113.41, 114.83, 115.40, 119.70, 119.78, 122.09, 123.03,
125.56, 129.32, 131.10, 132.42, 132.87, 149.81, 155.00, 157.63,
157.96, 161.77 and 195.41; EIMS (m/z) 515 [M + H]+.
155.91, 158.91, 159.50 and 198.04. To 9i was added a solution
of HCl saturated ether and the solution was stirred overnight to
give the hydrochloride compound as a solid which was isolated
by filtration, mp 235–236 °C; δH (300 MHz, DMSO) 1.67–1.76
(6 H, m, 3CH2), 2.70–3.00 (2 H, m, CH2), 3.12–3.25 (2 H, m,
CH2), 3.55–3.60 (4 H, m, 2CH2), 6.70 (2 H, d, J 8.5, ArH), 7.00
(1 H, dd, J 8.5, ArH), 7.14 (1 H, m, ArH), 7.24 (1 H, m, ArH),
7.31 (1 H, dd, J 8.5, ArH), 7.62 (2 H, d, J 8.5, ArH), 7.24 (1 H,
s, ArH), 8.36 (1 H, d, J 8.5, ArH), 8.59 (1 H, d, J 9.2, ArH),
9.81 (1 H, s, OH) and 10.10–10.12 (3 H, m, OH, 2 × NH);
δC (75.4 MHz; DMSO) 21.35, 22.55, 36.92, 52.42, 54.26,
98.43, 108.74, 111.54, 112.36, 112.90, 115.69, 118.96, 119.68,
122.17, 122.98, 125.56, 125.84, 129.39, 132.67, 134.47, 150.14,
152.73, 154.75, 157.52, 157.74 and 194.52; EIMS (m/z) 481
[M + H]+; HPLC purity: 93.23%.
(3,9-Dihydroxybenzo[b]naphtho[1,2-d]furan-5-yl)(4-[4-isopropyl-
piperazin-1-yl]phenyl)methanone 9j. The procedure described
above was used for the synthesis of 9j. From 8j (0.101 g,
0.153 mmol), ammonium formate (0.057 g, 0.918 mmol) and
black palladium (0.016 g, 0.153 mmol) in MeOH–AcOEt 1 : 1
(10 cm3), 9j (0.058 g, 79%) was obtained as a yellow oil. δH
(300 MHz, MeOD) 0.96 (6 H, d, J 6.7, 2CH3), 2.51–2.61 (5 H,
m, 2CH2N, CH), 3.15–3.29 (4 H, m, 2CH2N), 6.71 (2 H, d,
J 8.5, ArH), 6.96 (1 H, dd, J 1.8, 8.6, ArH), 7.03 (1 H, d, J 2.4,
ArH), 7.29 (1 H, dd, J 1.8, 8.5, ArH), 7.37 (1 H, d, J 1.8, ArH),
7.59 (1 H, s, ArH), 7.62 (2 H, d, J 9.2, ArH), 8.18 (1 H, d, J 8.6,
ArH) and 8.49 (1 H, d, J 9.2, ArH); δC (75.4 MHz; MeOD)
18.47, 47.54, 49.39, 56.07, 99.37, 110.18, 113.64, 114.16,
114.57, 117.70, 120.34, 121.31, 123.79, 124.34, 126.40, 128.63,
131.15, 133.81, 135.17, 151.93, 155.77, 156.01, 158.99, 159.57
and 198.05. To 9j was added a solution of HCl saturated ether
and the solution was stirred overnight to give the hydrochloride
compound as a solid which was isolated by filtration, mp
251 °C; νmax (KBr)/cm−1 1620, 2680 and 3180; δH (300 MHz,
DMSO) 1.30 (6 H, d, J 6.72, 2 × CH3), 3.09–3.16 (2 H, m,
CH2N), 3.31–3.60 (5 H, m, 2CH2N, CH), 4.09–4.13 (2 H, m,
CH2N), 7.01 (1 H, dd, J 8.5, ArH), 7.08 (2 H, d, J 8.5, ArH),
7.15 (1 H, d, J 1.8, ArH), 7.27 (1 H, s, ArH), 7.33 (1 H, dd,
J 1.8, 9.2, ArH), 7.71 (2 H, d, J 8.5, ArH), 7.79 (1 H, s, ArH),
8.37 (1 H, d, J 9.2, ArH), 8.61 (1 H, d, J 9.2, ArH), 9.84 (1 H, s,
OH), 10.19 (1 H, s, OH) and 10.58–10.67 (2 H, m, 2 × NH); δC
(75.4 MHz; DMSO) 21.30, 22.49, 36.90, 52.38, 54.23, 98.39,
108.70, 111.50, 112.33, 112.85, 115.65, 118.91, 119.63, 122.13,
122.92, 125.82, 129.35, 132.61, 134.43, 150.10, 152.67, 154.70,
157.48, 157.69 and 194.46; EIMS (m/z) 481 [M + H]+; HPLC
purity: 94.32%.
(3,9-Dihydroxybenzo[b]naphtho[1,2-d]furan-5-yl)(4-[1-methyl-
piperidine-4-yloxy]phenyl)methanone
9h. The
procedure
described above was used for the synthesis of 9h. From 8h
(0.105 g, 0.162 mmol), ammonium formate (0.306 g,
4.86 mmol) and black palladium (0.017 g, 0.162 mmol), 9h
(0.053 g, 71%) was obtained as a yellow oil. δH (300 MHz,
MeOD) 1.85–1.90 (2 H, m, CH2), 1.89–2.10 (2 H, m, CH2),
2.29 (3 H, s, CH3), 2.32–2.49 (2 H, m, CH2N), 2.64–2.76 (2 H,
m, CH2N), 4.50–4.62 (1 H, m, CH), 6.99–7.06 (4 H, m, ArH),
7.31–7.34 (1 H, m, ArH), 7.40–7.42 (1 H, m, ArH), 7.70 (1 H,
s, ArH), 7.82 (2 H, d, J 8.0, ArH), 8.65 (1 H, d, J 7.9, ArH) and
8.88 (1 H, d, J 8.5, ArH). To 9h was added a solution of HCl
saturated ether and the solution was stirred overnight to give the
hydrochloride compound as a solid which was isolated by fil-
tration, mp 269–270 °C; νmax (KBr)/cm−1 1625, 2740 and 3160;
δH (300 MHz, DMSO) 1.95–2.05 (2 H, m, CH2), 2.10–2.25
(2 H, m, CH2), 2.73 (3 H, s, CH3), 3.10–3.50 (4 H, m, 2CH2),
4.78–4.82 (1 H, m, CH), 6.99–7.06 (1 H, m, ArH), 7.11–7.20
(3 H, m, ArH), 7.30–7.40 (2 H, m, ArH), 7.75–7.90 (3 H, m,
ArH), 8.38 (1 H, d, J 7.9, ArH), 8.62 (1 H, d, J 8.5, ArH), 9.91
(1 H, s, OH) and 10.23–10.25 (2 H, m, OH, NH); δC
(75.4 MHz; DMSO) 30.08, 45.42, 52.01, 98.33, 108.48, 112.87,
113.26, 115.46, 119.59, 119.66, 122.08, 123.03, 125.56, 129.30,
130.17, 130.20, 132.53, 133.08, 149.86, 154.94, 157.60, 157.85,
161.64 and 195.26; EIMS (m/z) 468 [M + H]+; HPLC purity:
98.60%.
(3,9-Dihydroxybenzo[b]naphtho[1,2-d]furan-5-yl)(4-[2-(piperi-
dine-1-yl)ethylamino]phenyl)methanone 9i. The procedure
described above was used for the synthesis of 9i. From 8i
(0.086 g, 0.130 mmol), ammonium formate (0.049 g,
0.78 mmol) and black palladium (0.014 g, 0.130 mmol), 9i
(0.054 g, 87%) was obtained as a yellow oil. δH (300 MHz,
MeOD) 1.39–1.41 (2 H, m, CH2), 1.54–1.57 (4 H, m, 2CH2),
2.56–2.64 (6 H, m, 3CH2N), 3.22–3.27 (2 H, m, CH2N), 6.49
(2 H, d, J 8.6, ArH), 6.92 (1 H, dd, J 1.8, 8.6, ArH), 7.00 (1 H,
d, J 2.4, ArH), 7.25 (1 H, dd, J 2.4, 9.2, ArH), 7.31 (1 H, d,
J 2.4, ArH), 7.55 (1 H, s, ArH), 7.60 (2 H, d, J 8.5, ArH), 8.14
(1 H, d, J 8.6, ArH) and 8.44 (1 H, d, J 9.2, ArH); δC
(75.4 MHz; MeOD) 24.23, 25.68, 39.88, 55.14, 57.72, 99.33,
110.14, 112.34, 113.57, 114.10, 117.73, 120.28, 121.01, 123.71,
124.33, 126.38, 127.19, 131.11, 134.42, 135.64, 152.00, 154.76,
3,9-Dihydroxybenzo[b]naphtho[1,2-d]thiophene-5-carboxylic
acid 10. Nitrile 3 (60 mg, 0.206 mmol) was treated with a 6 N
aqueous solution of NaOH (0.72 g in 3 cm3) and the mixture
was heated for 24 h. After cooling, HCl 6 N was added and the
solid formed was isolated by filtration and purified by column
chromatography on silica gel using DCM–MeOH 9 : 1 as an
eluent to give 10 (62 mg, 97%) as a white solid. δH (300 MHz,
DMSO) 7.13 (1 H, d, J 8.5, ArH), 7.34 (1 H, d, J 8.5, ArH),
7.49 (1 H, s, ArH), 8.47 (1 H, s, ArH), 8.60 (1 H, s, ArH), 8.77
(1 H, d, J 9.1, ArH), 8.94 (1 H, d, J 9.1, ArH), 10.02 (1 H, s,
OH), 10.21 (1 H, s, OH), 13.11 (1 H, br s, COOH);
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7334–7346 | 7343

View Article Online
δC (75.4 MHz; DMSO) 104.5, 107.7, 108.5, 115.6, 118.4, 120.2,
122.8, 126.2, 127.2, 129.0, 129.2, 131.6, 132.1, 133.3, 143.3,
156.6 and 157.3; EIMS (m/z) 290 [M − H]+.
crystallographic structure. Box size was able to enclose the LBD
together with helix-12, and was similar to the box defined with
AutoDock. The docking procedure was performed with the XP
(extra precision) mode and a van der Waals radii scale factor of
1.0/0.8 for the receptor and the ligand, respectively. Induced Fit
Docking was also used, and contained constrained minimization
of the receptor with an RMSD cutoff of 0.18 Å, and Prime-side-
chain prediction on residues within 5 Å of any ligand pose.
Glide redocking was performed in structures within 30 kcal
mol−1 of the lowest energy structure with a van der Waals
scaling of 1.0/0.8 for the receptor and the ligand, respectively.
Methyl 3,9-dihydroxybenzo[b]naphtho[1,2-d]thiophene-5-
carboxylate 11. A solution of acid 10 (39 mg, 0.126 mmol) in
MeOH (3 cm3), in the presence of a catalytic amount of H2SO4,
was heated for 24 h. After evaporation of MeOH and purification
by column chromatography on silica gel using DCM–MeOH
15 : 1 as an eluent, the ester 11 was obtained (25 mg, 62%) as an
oil. δH (300 MHz, DMSO) 3.95 (3 H, s, CH3), 7.12 (1 H, dd,
J 9.2 and 1.8, ArH), 7.35 (1 H, dd, J 9.2 and 2.4, ArH), 7.49
(1 H, d, J 1.8, ArH), 8.36 (1 H, d, J 2.4, ArH), 8.62 (1 H,
s, ArH), 8.78 (1 H, d, J 9.2, ArH), 8.95 (1 H, d, J 9.2, ArH),
10.07 (1 H, br s, OH) and 10.17 (1 H, br s, OH); δC (75.4 MHz;
DMSO) 167.2, 156.8, 155.8, 143.0, 132.8, 131.3, 131.1, 127.5,
126.6, 126.0, 125.3, 123.9, 122.7, 118.9, 115.2, 109.0 and
108.5.
Molecular dynamics simulations. The protein–ligand system
was first minimized in vacuum for 1000 steps of steepest descent
followed by 4000 steps of conjugate gradient using the Amber11
program.24 All Cα atoms were restrained to their initial coordi-
nates. The resulting minimized complex was solvated by a box
of TIP3P waters which extended at least 8 Å away from any
given protein atom. The SHAKE algorithm was applied to all
hydrogen-containing bonds25 and a 1 fs integration step was
used. The simulation used periodic boundary conditions, and the
electrostatic interactions were represented using the smooth par-
ticle mesh Ewald method,26 with a grid spacing of 1 Å. Each
system was gently annealed from 100 to 300 K over a period of
25 ps. The systems were then maintained at a temperature of
300 K during 50 ps with positional restraints (α-carbon atoms),
together with a distance restraint to the hydrogen bond between
the Asp303 carboxylate and the piperidinium NH group, and
progressive energy minimizations, gradually releasing the
restraints of the solute followed by a 20 ps heating phase from
100 to 300 K, whereafter restraints were removed. Finally, two
production simulations were continued. The first one maintained
the distance restraint between the Asp303 carboxylate and the
piperidinium NH during 500 ps, plus one additional nanosecond
without restriction. The second one continued the MD simu-
lation during 2 ns with no restriction. Coordinate trajectories
were recorded each 2 ps throughout all equilibration and pro-
duction runs (RMSD, ESI†).
Computational methods
The theoretical study of the binding mode has been carried out
in both subtypes of estrogen receptor, α and β. As macro-
molecules, the crystallographic structures of the two receptor
subtypes in complex with several ligands have been selected:
ERα in complex with estradiol (PDB 1A52), raloxifene (PDB
1ERR), genistein (PDB 1X7R), WAY-244 (PDB 1X7E), and
4-hydroxytamoxifen (PDB 3ERT), and ERβ in complex with
genistein (PDB 1X7J), THC (PDB 1L2J), WAY-202196 (PDB
1YYE), and 4-hydroxytamoxifen (PDB 2FSZ). Water molecules
close to the amino acids Arg394 (ERβ Arg346) and Glu353
(Glu305 ERβ) have been kept for the docking procedures.
Ligands were built using the Maestro LigPrep module (www.
schrodinger.com). Optimization of the geometry and the charges
calculation were performed by using program Gaussian 0320 at
the B3LYP/3-21G* level. Once the compounds were optimized,
atom types and bond types were assigned, and mol2 files were
generated. Macromolecules geometry was refined by using the
Protein Preparation module in Maestro. Two different docking
programs were employed in order to contrast and compare the
results: AutoDock421 and Glide.22,23 General protocols are
described as follows.
Biological assays
Chemicals. 17-β-Estradiol, neutral red, dextran coated char-
coal, PSB, Tween-20, BSA, ICI 180.780, 4-hydroxytamoxifen,
insulin were purchased from Sigma-Aldrich. Estradiol
[2,4,6,7,16,17-3H(N)], scintillation counting liquid (Optifase
HiSafe2) were obtained from Perkin-Elmer (Salem, MA). Estro-
gen receptors α and β produced in insect cells and sodium pyru-
vate were purchased from Invitrogen. Cell culture medium
DMEM, EMEM, FBS, antibiotics, trypsin-EDTA, amino acids,
L-glutamine were purchased from Lonza. DCC-FBS was
obtained from Hyclone (Erembodegem, Aalst, Belgium).
Docking studies with AutoDock4. Docking regions were
defined considering a box of 80, 80 and 90 points in the x, y,
and z axes. The grids were built by focusing on the ER binding
site. For the calculation of energy maps, a grid spacing of
0.375 Å and a distance-dependent dielectric constant were used
by means of AutoGrid4. The docking was carried out using the
Lamarckian genetic algorithm, leaving all the bonds as rotatable.
The program searched until a maximum of 100 conformations
and the procedure was repeated 100 times (runs). After docking,
the 100 solutions were clustered in groups with RMSD less than
1.0 Å. For all other parameters, the default values were used
with AutoDock Tools.
Receptor binding studies: in vitro competitive binding assay.
The relative binding affinity (RBA) of the compounds for the
estrogen receptors was determined by a competition assay,
according to the method described by Arcaro with some modifi-
cations.19 Purified full-length human estrogen receptors α and β
were incubated for 4 h at 23 °C with different concentrations of
compounds in the presence of 5 nM [2,4,6,7,16,17-3H]-estradiol
Docking studies with Glide. Interaction maps of the binding
site were generated using the application “Receptor grid gen-
eration”, included in the Glide module, positioning the center
of the box on the center of the bound ligand present in the
7344 | Org. Biomol. Chem., 2012, 10, 7334–7346
This journal is © The Royal Society of Chemistry 2012

View Article Online
in 150 μl of total volume. The stocks of test compounds were
prepared in DMSO. All these compounds, including
[2,4,6,7,16,17-3H]-estradiol and receptors, were diluted in
Tween/PBS buffer (99.85 : 0.15 w/v). A vehicle control con-
tained 0.1% of DMSO. After incubation, the non-bound
[2,4,6,7,16,17-3H]-estradiol was removed by adding a mixture
of 10% DCC and 2% albumin bovine serum, incubating for
15 min at 4 °C, followed by centrifugation at 6000 g for 5 min at
4 °C. 150 μl of the supernatant was added to 4 cm3 of scintil-
lation liquid and the radioactivity of bound estradiol was
measured in a liquid scintillation counter Beckman LS 6500
(Beckman Coulter, Inc.). Three independent experiments with
three repetitions for each compound were performed. Results
were expressed as the percentage of specific binding of
[2,4,6,7,16,17-3H]-estradiol to ER versus log of competitor con-
centration. Graph Pad Prism software (non-linear regression
analysis) was used to calculate the concentration needed to dis-
place 50% of [2,4,6,7,16,17-3H]-estradiol (IC50). To compare
binding affinities of the test compounds to those reported in the
literature, IC50 values were converted to RBA values using
estradiol as a standard. The values of IC50 for estradiol were
8.98 and 6.87 nM for ERα and ERβ, respectively. The RBA of
estradiol was arbitrarily set at 100 (RBA = (IC50 of E2)/IC50 of
ligand × 100).
curves. In order to obtain agonist dose–response curves, ER
reporter cells were treated with media alone (estradiol was used
as a positive control agonist). To perform receptor inhibition
studies a co-mix of a known agonist (estradiol: 150 and 111 pM
corresponding to ∼EC70 for the ERα and ERβ assay, respec-
tively) and a dilution series of the test compounds was prepared.
IC182 780 was used as an antagonist positive control. The final
solvent control did not exceed 0.1% of DMSO. All measure-
ments were performed in triplicate.
Abbreviations
BSC
PDB
DCM
THF
DMF
RT
basic side chain
Protein Data Bank
dichloromethane
tetrahydrofuran
dimethylformamide
room temperature
molecular dynamics
MD
RMSD root-mean-square deviation
RBA relative binding affinity
Acknowledgements
This work was supported by the Spanish Ministry of Science
and Innovation (SAF2008-00945, CTQ2011-24741 and
CSD2007-00063) and Fundación Universitaria San Pablo CEU
(USP-PC 13/10). Grant to J. J. R. from Fundación Universitaria
San Pablo CEU is also acknowledged. We thank EADS-CASA
for fellowships to K. F., M. M., J. C. and S. D.
Estrogenic and antiestrogenic activities. MCF-7 cells were
seeded in 96-well plates at 5 × 103 cells per well in DMEM con-
taining 10% FBS, 0.01 mg cm−3 of insulin solution and 0.1 mM
nonessential amino acids. After 24 h, the medium was changed
to EMEM without phenol red, containing 5% dextran-coated
charcoal stripped FBS (DCC-FBS), 0.1 mM nonessential amino
acids, 1 mM sodium pyruvate and 2 mM L-glutamine, and was
preincubated for 3 days prior to treatment. Afterwards, different
concentrations of the test compounds (0.1–20 μM) were added
to the cells with/without 1 pM estradiol, in order to test the
capacity to induce or prevent the proliferation of MCF-7 cells.
The final vehicle concentration of maximally 0.1% of DMSO
(and 0.1% of ethanol in the case of treatment with estradiol)
served as a solvent control. On day 4, the medium in the plates
containing the compounds was refreshed. On day 8, cell proli-
feration was determined by the Neutral Red uptake assay, which
provides a quantitative estimation of the number of viable cells.
Briefly, the medium was removed and 200 μl of neutral red solu-
tion (50 μg cm−3) was added, and incubated for 2 h. Afterwards,
the cultures were carefully washed twice with PBS, and the
extraction solution was added (250 μL of 1% acetic acid, 50%
ethanol), and incubated for 15 min at room temperature. The
absorbance was measured at 540 nm wavelength in a plate
reader (Biotec). The viability was calculated considering the
controls without test substance as 100% viable.
References
1 S. Nilsson and J. A. Gustafsson, Clin. Pharmacol. Ther. (St. Louis),
2011, 89, 44–55.
2 J. A. Katzenellenbogen, R. Muthyala and B. S. Katzenellenbogen, Pure
Appl. Chem., 2003, 75, 2397–2403.
3 S. Nilsson, K. F. Koehler and J. A. Gustafsson, Nat. Rev. Drug Discovery,
2011, 10, 778–792.
4 F. Minutolo, M. Macchia, B. S. Katzenellenbogen and
J. A. Katzenellenbogen, Med. Res. Rev., 2011, 31, 364–442.
5 H. A. Harris, J. A. Katzenellenbogen and B. S. Katzenellenbogen, Endo-
crinology, 2002, 143, 4172–4177.
6 H. A. Harris, L. M. Albert, Y. Leathurby, M. S. Malamas,
R. E. Mewshaw, C. P. Miller, Y. P. Kharode, J. Marzolf, B. S. Komm,
R. C. Winneker, D. E. Frail, R. A. Henderson, Y. Zhu and J. C. Keith,
Endocrinology, 2003, 144, 4241–4249.
7 M. J. Meyers, J. Sun, K. E. Carlson, G. A. Marriner,
B. S. Katzenellenbogen and J. A. Katzenellenbogen, J. Med. Chem.,
2001, 44, 4230–4251.
8 R. E. Mewshaw, J. Edsall, J. Richard, C. Yang, E. S. Manas, Z. B. Xu,
R. A. Henderson, J. C. Keith Jr. and H. A. Harris, J. Med. Chem., 2005,
48, 3953–3979.
9 E. S. Manas, R. J. Unwalla, Z. B. Xu, M. S. Malamas, C. P. Miller,
H. A. Harris, C. Hsiao, T. Akopian, W.-T. Hum, K. Malakian,
S. Wolfrom, A. Bapat, R. A. Bhat, M. L. Stahl, W. S. Somers and
J. C. Alvarez, J. Am. Chem. Soc., 2004, 126, 15106–15119.
10 M. De Angelis, F. Stossi, K. A. Carlson, B. S. Katzenellenbogen and
J. A. Katzenellenbogen, J. Med. Chem., 2005, 48, 1132–1144.
11 J. Sun, M. J. Meyers, B. E. Fink, R. Rajendran, J. A. Katzenellenbogen
and B. S. Katzenellenbogen, Endocrinology, 1999, 140, 800–804.
12 M. J. Meyers, J. Sun, K. E. Carlson, B. S. Katzenellenbogen and
J. A. Katzenellenbogen, J. Med. Chem., 1999, 42, 2456–2468.
13 A. M. Brzozowski, A. C. W. Pike, Z. Dauter, R. E. Hubbard, T. Bonn,
O. Engström, L. Öhman, G. L. Greene, J. A. Gustafsson and
M. Carlquist, Nature, 1997, 389, 753–758.
Agonist/antagonist profile. Reporter cells were dispensed in a
96-well plate and then immediately dosed with the test com-
pounds. Following overnight incubation, the treatment media
were discarded and the luciferase detection reagent was added.
The intensity of light emission from the ensuing luciferase
reaction provides a measure that is directly proportional to the
level of ER activation in the reporter cells. The assays were
configured to perform agonist and antagonist dose–response
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7334–7346 | 7345

View Article Online
14 A. K. Shiau, D. Barstad, J. T. Radek, M. J. Meyers, K. W. Nettles,
B. S. Katzenellenbogen, J. A. Katzenellenbogen, D. A. Agard and
G. L. Green, Nat. Struct. Biol., 2002, 9, 359–364.
15 B. R. Henke, T. G. Consler, N. Go, R. L. Hale, D. R. Hohman,
S. A. Jones, A. T. Lu, L. B. Moore, J. T. Moore, L. A. Orband-Miller,
R. G. Graham Robinett, J. Shearin, P. K. Spearing, E. L. Stewart,
P. S. Turnbull, S. L. Weaver, S. P. Williams, G. B. Wisely and
M. H. Lambert, J. Med. Chem., 2002, 45, 5492–5505.
G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich,
A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul,
S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-
Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill,
B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople,
GAUSSIAN 03 (Revision C.02), Gaussian, Inc., Wallingford, CT, 2004.
21 R. Huey, G. M. Morris, A. J. Olson and D. S. Goodsell, J. Comput.
Chem., 2007, 28, 1145–1152.
16 J. Shen, C. F. Tan, Y. Y. Zhang, X. Li, W. H. Li, J. Huang, X. Shen and
Y. Tang, J. Med. Chem., 2010, 53, 5361–5365.
17 S. Martín-Santamaría, J. J. Rodríguez, S. de Pascual-Teresa, S. Gordon,
M. Bengtsson, I. Garrido-Laguna, B. Rubio-Viqueira, P. P. López-Casas,
M. Hidalgo, B. de Pascual-Teresa and A. Ramos, Org. Biomol. Chem.,
2008, 6, 3486–3496.
22 R. A. Friesner, J. L. Banks, R. B. Murphy, T. A. Halgren, J. J. Klicic,
D. T. Mainz, M. P. Repasky, E. H. Knoll, M. Shelley, J. K. Perry,
D. E. Shaw, P. Francis and P. S. Shenkin, J. Med. Chem., 2004, 47,
1739–1749.
18 A. M. Ramos, J. J. Rodríguez, S. Martín-Santamaría, B. de Pascual-
Teresa, M. Hidalgo, P. P. López-Casas and B. Rubio-Viqueira, España,
Patente de Invención ES P200801961, 2008.
19 K. F. Arcaro, Y. Yang, D. D. Vakharia and J. F. Gierthy, J. Toxicol.
Environ. Health, Part A, 2000, 59, 197–210.
20 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. Montgomery, J. A. T. Vreven, K. N. Kudin,
J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone,
B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson,
H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li,
J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo,
J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma,
23 T. A. Halgren, R. B. Murphy, R. A. Friesner, H. S. Beard, L. L. Frye,
W. T. Pollard and J. L. Banks, J. Med. Chem., 2004, 47, 1750–1759.
24 D. A. Case, T. A. Darden, T. E. Cheatham III, C. L. Simmerling, J. Wang,
R. E. Duke, R. Luo, R. C. Walker, W. Zhang, K. M. Merz, B. Roberts,
B. Wang, S. Hayik, A. Roitberg, G. Seabra, I. Kolossváry, K. F. Wong,
F. Paesani, J. Vanicek, J. Liu, X. Wu, S. R. Brozell, T. Steinbrecher,
H. Gohlke, Q. Cai, X. Ye, J. Wang, M.-J. Hsieh, G. Cui, D. R. Roe,
D. H. Mathews, M. G. Seetin, C. Sagui, V. Babin, T. Luchko, S. Gusarov,
A. Kovalenko and P. A. Kollman, AMBER 11, University of California,
San Francisco, 2010.
25 J. P. Ryckaert, G. Ciccotti and H. J. C. Berendsen, J. Comput. Phys.,
1977, 23, 327–341.
26 T. Darden, D. York and L. Pedersen, J. Chem. Phys., 1993, 98,
10089–10092.
7346 | Org. Biomol. Chem., 2012, 10, 7334–7346
This journal is © The Royal Society of Chemistry 2012