2
S. Archana et al. / Bioorg. Med. Chem. Lett. xxx (2014) xxx–xxx
In order to get insight into the structure activity relationship,
deiodinated product 12 was also synthesized using Pd(OAc)2 and
HN
formic acid (Scheme 2).24
O
O
OH
OH
O
O
N
O
HO
H
O
OH
OH
Biology: Cytotoxicity of the synthesized compounds 11a–o and
12 were checked against ER positive breast cancer cell line MCF-7
and ER negative cell line MDA-MB-231 and control mammary epi-
thelial cells MEpiC by MTT assay. Even though the antiproliferative
activity of these compounds were not more than the reference
drug but as per our hypothesis it was observed that most of the
compounds were showing more potency toward ER positive cell
line MCF-7 reflecting the possible interaction of these molecules
with estrogen receptor (Table 3). It is also interesting to note that
compounds 11d, 11f–n and 12 were non-cytotoxic to control
mammary epithelial cells indicating the specificity of these com-
pounds toward the cancer cells. However, compounds 9–10,
11a–c and 11e were observed to be of comparable anti-prolifera-
tive capacity to both cancer cell lines as well as control cells. To
further understand the possible interactions a molecular modeling
study was performed.
O
OH
O
O
NH2 HCl
3
1
2
Doxorubicin.HCl;
R2
Tamoxifen;
O
Nakijiquinone;
R3
NH
HO
O
HO
O
O
O
O
O
5
HU-331; 4
Spirobacillene A;
6
Designed Prototype;
Figure 1. Pharmaceutically important quinones and related compounds.
The antiproliferative data reflects that the two precursors of
designed spiro products, compound 9 and 10 did not exhibit any
cytotoxicity against MCF-7 while they were showing potency
against MDA-MB-231 cell line. Most of the products obtained after
Suzuki coupling such as 11a–f, 11m–n and 12 were selective
toward MCF-7 whereas compounds 11g–k showed moderate
selectivity toward MCF-7. Compound 11l was more selective
toward MDA-MB-231. It was evident from the cytotoxicity data
of 11f and 11h that increasing electron donating groups such as
methoxy did not improve the cytotoxicity against MCF-7 but the
selectivity between MCF-7 and MDA-MB-231 was lost showing
that compound 11h might be having different mode of action.
Compounds with polar group at phenyl rings, for example, 11d,
11g and 11m were found more potent and compound 11d was
highly selective toward MCF-7.
Table 1
Optimization of Suzuki coupling
Entry Catalysta
Solvent
Base
Cs2CO3 85 °C
K2CO3 85 °C
Cs2CO3 85 °C
K2CO3 85 °C
Temp Time (h) Yield (%)
1
2
3
4
5
6
7
8
9
Pd(PPh3)4
Pd(PPh3)4
DMF
DMF
20
20
12
5
5
12
9
NDc
NDc
NDb
52b
15b
Pd(PPh3)2Cl2 DMF
PdCl2
PdCl2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
DMF
DMF
DMF
THF
Cs2CO3 85 °C
Cs2CO3 85 °C
b
20
K2CO3
K2CO3
60 °C
85 °C
85 °C
10b
68b
80b
DMF
9
4
DMF/water K2CO3
a
All the reactions were carried out by using 10 (0.41 mmol), catalyst
(0.002 mmol), base (0.82 mmol) and boronic acid (0.82 mmol) in solvent (3 mL).
b
Molecular docking studies on ER-a: An analysis of the available X-
Isolated yield.
c
Starting material (10) remained unreacted. ND-product not detected.
ray crystal structures for estrogen receptor revealed that the active
site is highly flexible and can accommodate various scaffolds
(Fig. 2). The binding site is dominated by hydrophobic residues
and thus indicates the importance of hydrophobic interactions in
ligand binding to this macromolecule.
The crystal structure of human estrogen receptor alpha ligand-
binding domain in complex with 4-hydroxytamoxifen (PDB ID:
3ERT) was used for the docking. The protein was prepared using
Maestro module of Schrodinger software (Schrödinger version
9.3, Schrödinger, LLC, New York, NY).During protein preparation
correct bond orders were assigned, hydrogen and other missing
atoms were added to the residues and charges were assigned to
atoms using OPLS-AA force field. The co-crystallized waters were
retained for docking. The docking was done using Glide module
of the Schrodinger software. The docking grid was generated using
co-crystallized ligand (4-hydroxytamoxifen) as grid center with
default settings. The ligands were sketched in Maestro module of
of Suzuki coupling of 10 with 2-benzothiophenyl boronic acid as
per reported procedure did not give very good yields hence we
optimized the reaction with different palladium catalysts and
bases (Table 1) (Scheme 1). Initial trials with 20 mol % of Pd(PPh3)4
or Pd(PPh3)2Cl2 in the presence of Cs2CO3 or K2CO3 as base was
unsuccessful (entries 1–3, Table 1). However, coupled product
was obtained with PdCl2 in presence of Cs2CO3 in low yields while
a better yield was observed with K2CO3 (entry 3–4, Table 1). With
10 mol % Pd(OAc)2 as catalyst and K2CO3 as base gave the best yield
of 80% in DMF–water (4:1) (entry 9, Table 1).
With the optimized condition23 at hand, 14 derivatives 11a–n
were synthesized (Table 2). To our delight, most of the reactions
gave good to excellent yields of the coupled product except
3,4,5-trimethoxyphenylboronic acid that gave poor yield. The
newly synthesized products were characterized by 1H, 13C NMR
and mass analysis.
Schrodinger software and were minimized to
a gradient of
0.001 kcal/Mol Å2. The energy minimized ligands were prepared
for docking using Ligprep module of Schrodinger software. The
O
I
O
O
S
S
OH
O
B(OH)2
CO2H
DIC/HOBt, DCM
O
Ph
I2, NaHCO3, ACN
Ipsocyclization
O
Ph
Ph
+
Suzuki Coupling
Conditions; Table 1
Ph
OMe
OMe
O
O
9
8
10
7
11a
Scheme 1. Preparation of compounds 11a–n.