Synthesis, molecular docking, and QSAR study of bis-sulfonamide derivatives as potential aromatase inhibitors

Article abstract of DOI:10.1016/j.bmc.2019.08.001

A library of bis-sulfonamides (9–26) were synthesized and tested for their aromatase inhibitory activities. Interestingly, all bis-sulfonamide derivatives inhibited the aromatase with IC₅₀ range of 0.05–11.6 μM except for compound 23. The analo

Full text of DOI:10.1016/j.bmc.2019.08.001

Journal Pre-Proof
Synthesis, molecular docking, and QSAR study of bis-sulfonamide derivatives
as potential aromatase inhibitors
Ronnakorn Leechaisit, Ratchanok Pingaew, Veda Prachayasittikul, Apilak
Worachartcheewan, Supaluk Prachayasittikul, Somsak Ruchirawat, Virapong
Prachayasittikul
PII:
S0968-0896(19)31002-8
https://doi.org/10.1016/j.bmc.2019.08.001
BMC 15040
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
14 June 2019
30 July 2019
1 August 2019
Please cite this article as: Leechaisit, R., Pingaew, R., Prachayasittikul, V., Worachartcheewan, A., Prachayasittikul,
S., Ruchirawat, S., Prachayasittikul, V., Synthesis, molecular docking, and QSAR study of bis-sulfonamide
derivatives as potential aromatase inhibitors, Bioorganic & Medicinal Chemistry (2019), doi: https://doi.org/
1
0.1016/j.bmc.2019.08.001
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will
undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing
this version to give early visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
©
2019 Published by Elsevier Ltd.

JOURNAL PRE-PROOF
1
Synthesis, molecular docking, and QSAR study of bis-
sulfonamide derivatives as potential aromatase inhibitors
a
a,
b,
Ronnakorn Leechaisit , Ratchanok Pingaew *, Veda Prachayasittikul *, Apilak
Worachartcheewanb,c,d, Supaluk Prachayasittikul , Somsak Ruchirawat , Virapong
b
e,f
Prachayasittikulg
aDepartment of Chemistry, Faculty of Science, Srinakharinwirot University, Bangkok 10110,
Thailand
bCenter of Data Mining and Biomedical Informatics, Faculty of Medical Technology,
Mahidol University, Bangkok 10700, Thailand
cDepartment of Community Medical Technology, Faculty of Medical Technology, Mahidol
University, Bangkok 10700, Thailand
dDepartment of Clinical Chemistry, Faculty of Medical Technology, Mahidol University,
Bangkok 10700, Thailand
eLaboratory of Medicinal Chemistry, Chulabhorn Research Institute, and Program in
Chemical Biology, Chulabhorn Graduate Institute, Bangkok 10210,Thailand
fCenter of Excellence on Environmental Health and Toxicology, Commission on Higher
Education (CHE), Ministry of Education, Thailand
gDepartment of Clinical Microbiology and Applied Technology, Faculty of Medical
Technology, Mahidol University, Bangkok 10700, Thailand
*
Corresponding authors:
E-mail: ratchanok@swu.ac.th (R. Pingaew)
Tel: 662-649-5000 ext 18253; Fax: 662-260-0128
E-mail: veda.pra@mahidol.ac.th (V. Prachayasittikul)
Tel: 662-441-4376; Fax: 662-441-4380

JOURNAL PRE-PROOF
2
Abstract
A library of bis-sulfonamides (9-26) were synthesized and tested for their aromatase
inhibitory activities. Interestingly, all bis-sulfonamide derivatives inhibited the aromatase
with IC range of 0.05-11.6 μM except for compound 23. The analogs 15 and 16 bearing
5
0
hydrophobic chloro and bromo groups exhibited the potent aromatase inhibitory activity in
sub-micromolar IC50 values (i.e., 50 and 60 nM, respectively) with high safety index.
Molecular docking revealed that the chloro and bromo benzene sulfonamides (15 and 16)
may play role in the hydrophobic interaction with Leu477 of the aromatase to mimic steroidal
backbone of the natural substrate, androstenedione. QSAR study also revealed that the most
potent activity of compounds was governed by van der Waals volume (GATS6v) and mass
(Mor03m) descriptors. Finally, the two compounds (15 and 16) were highlighted as
promising compounds to be further developed as novel aromatase inhibitors.
Keywords:
Sulfonamide; Bis-sulfonamide; Xylylenediamine; Aromatase inhibitor;
Molecular docking; QSAR

JOURNAL PRE-PROOF
3
1
. Introduction
Estrogens (estrone, 17β-estradiol) are steroidal sex hormones which are
biosynthesized by the conversion of their corresponding androgens (androstenedione,
testosterone). This conversion reaction is catalyzed by the key rate-limiting enzyme called
1
-7
aromatase (CYP19), a member of the cytochrome P450 family. Estrogens have been noted
for their promoting roles in cancer cell growth, recurrence, and mestastasis of the hormone-
dependent breast cancers. A reduction of estrogen levels by targeting the aromatase inhibition
is one of effective current therapeutic strategies for management of breast cancer. Several
aromatase inhbitors (AIs) have been developed including steroidal AIs and nonsteroidal AIs.
Although clinical success stories have been noted for these available AIs, coping with their
side effects (such as bone loss and cardiovascular disease) and potential resistance in
prolonged use are still challenging.8-11 Therefore, the search for novel classes of AIs still
remains an essential on-going research issue.
Sulfonamides have been proven to be interesting scaffold that hold a wide range of
biological activities including anticancer, antimicrobial, antimalarial, and antiviral
activities.12-16 This is owing to their potential structural features which are capable of multiple
1
6
interactions with various biological targets. Furthermore, sulfonamides are generally stable,
easy to synthesize, while providing maximized pharmacological profiles such as oral
1
3
absorption and low side effects. Bis-sulfonamide comprising two groups of sulfonamide
that could act as apoptotic inducing agents by disrupting anionic homeostasis of cell in cancer
1
7
18-24
treatment. Significantly, sulfonamide pharmacophore is found in AIs (1-5)
as shown in
Figure 1 in which the molecular docking study revealed that both oxygen and nitrogen atoms
of such sulfonamide group could form hydrogen bond with the target protein.20-24

JOURNAL PRE-PROOF
4
CH3
H3CO
H3CO
O
N
S
O
N
N
O
O
S
N
N
CH3
N
H
O
O
O
O O
S
Br
N
N
N
OCH3
1
2
3
O
O
H3CO
O
OCH3
OH
O O
S
N
N
S
NH
NH
HN
HN
S
S
S O
O
N
OCH3
OCH3
NH
HN
H
O
H
O
O
N
R
R
NH HN
6a : R = Cl
O
6
b : R = NO2
4
5
Figure 1. Representative aromatase inhibitors containing sulfonamide (1-5) and thiourea (6)
moieties
Recently, our group has reported that bis-thioureas derived from meta-
xylylenediamine exert aromatase inhibitory activity such as compounds 6a and 6b with IC50
2
5
=
5.5 and 0.8 μM, respectively. The molecular docking study revealed that phenyl xylenyl
thiourea moiety could mimic steroidal backbone of androstenedione (ASD) through
hydrophobic interaction. In this article, it could be hypothesized that the replacement of
thiourea moiety with a bioisosteric sulfonamide group might provide the compound with
aromatase inhibitory activity.
To seek for a novel class of AIs, the design, synthesis and investigation on aromatase
inhibitory activities of bis-sulfonamide derivatives (9-26, Scheme 1) have been studied. In
addition, computational approaches, including molecular docking and QSAR, were further
carried out to provide beneficial information for guiding the rational design and discovery of
new potential AIs.
2
. Results and discussion
2
.1 Chemistry
Bis-sulfonamides (9-21, 23, 24) representing various substituents (R) on sulfonyl part
were readily synthesized by alkylations of meta- or para-xylylenediamine 7a,b with the
corresponding sulfonyl chlorides 8a-m in the presence of sodium carbonate in
dichloromethane at room temperature. However, bis-sulfonamides (25, 26) were obtained by
the alkylations of meta-phenylenediamine 7c with the appropriate sulfonyl chlorides 8a-m in

JOURNAL PRE-PROOF
5
refluxing pyridine (Scheme 1). Aminosulfonamide 22 was derived from the reduction of
nitrosulfonamide 21 using stannous chloride in refluxing ethanol.
Structures of the desired bis-sulfonamides (9-26) were confirmed based on their
1
NMR, IR and HRMS spectra. H NMR spectra of compounds (9-26) typically showed a
symmetrical signal of two sulfonamide parts (NHSO Ar), and a characteristic singlet or
2
doublet signal of methylene proton at δ in the range of 3.8–4.1 ppm was appeared in
compounds 9-24. In addition, infrared spectra of the synthesized compounds displayed strong
−
1
N–H absorptions at 3200–3400 cm , and all bis-sulfonamides had molecular ion peaks
1
corresponding to their molecular formula. H NMR data of known compounds 9, 11-13, 16
1
7
and 19 are consistent with that reported in the literature .
O
O
S
A
Cl
R
A
8a-m
H2N
NH2
R
NH
HN
R
Na2CO3, CH2Cl2
S
S
CH3
7a
O
O
O O
CH3
9-21
NO2
NO2
R
=
H3C
CH3
CF3
CH3
0, 23
OCH3
SnCl2 2H2O
EtOH, reflux
1
11
9
1
12, 25
NH
HN
S
S
R
NH
HN
R
S
S
O
O
O
O
O
O
O O
21
22
O
O
F
Cl
Br
CN
3
14
15, 26
16
S
17, 24
Cl
R
O
O
8b, 8i
O
O
NO2
NH2
S
NH
HN
O
S
H2N
NH2
Na2CO3, CH2Cl2
R
R
2
3-24
CH3
7
b
c
NO2 O2N
O
O
O
22
1
8
19
20
21
S
O
O
O
S
Cl
R
S
8d, 8g
R
N
N
R
H2N
NH2
H
H
pyridine
7
25-26
Scheme 1. Synthesis of bis-sulfonamide derivatives (9-26)
2
.2 Aromatase inhibitory activity / Structure-activity relationship
Eighteen bis-sulfonamide derivatives (9-26) were evaluated for their abilities to
inhibit the aromatase using letrozole (IC = 1.9 nM) as the reference drug. An overview
5
0
about the effects of these inhibitors on the aromatase activity have been provided in Table 1.
The investigated bis-sulfonamides comprised three different ring A (meta-xylylenediamine
ring (9-22), para-xylylenediamine ring (23-24), and m-phenylenediamine ring (25-26)) as a
core part on the nitrogen of sulfonamide whereas the sulfonyl groups of these sulfonamides
bearing various R groups as substituted benzene and naphthalene rings.

JOURNAL PRE-PROOF
6
Table 1 Aromatase inhibitory and cytotoxic activities (IC , μM) of bis-sulfonamides (9-26).
5
0
Compound
Aromatase
inhibitory activity
cytotoxic activity
T47-D
selectivity
indexa
MRC-5
(SI)
9
0.13±0.06
3.20±1.07
0.21±0.11
11.6±2.1
0.43±0.22
2.19±0.54
0.05±0.01
0.06±0.02
0.16±0.09
0.28±0.01
0.37±0.01
3.67±1.27
2.75±1.09
3.30±0.58
>12.5
43.93±0.85
30.92±0.71
Non-cytotoxic
75.27±2.04
75.38±5.29
–b
Non-cytotoxic
80.89±3.02
> 865.14
25.28
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
Non-cytotoxic
Non-cytotoxic
74.59±7.81
> 460.86
> 9.04
173.47
> 50.46
1370.4
> 1451.0
> 669.84
> 356.72
> 266.80
> 26.90
> 35.89
> 33.93
–
Non-cytotoxic
68.52±1.08
37.47±2.15
–b
Non-cytotoxic
Non-cytotoxic
Non-cytotoxic
Non-cytotoxic
Non-cytotoxic
Non-cytotoxic
Non-cytotoxic
16.27±0.86
69.28±3.55
Non-cytotoxic
39.29±2.02
75.36±2.15
Non-cytotoxic
90.61±0.95
Non-cytotoxic
Non-cytotoxic
Non-cytotoxic
87.22±4.39
–
0.66±0.16
0.53±0.11
0.78±0.12
0.0019±0.0002
–
92.83±4.24
140.65
> 210.34
80.90
Non-cytotoxic
63.10±1.24
Letrozolec
–
–
Doxorubicinc
0.88±0.021
2.19±0.37
–

JOURNAL PRE-PROOF
7
T47-D = hormone-dependent breast cancer cell line; MRC-5 = normal embryonic lung cell line.
Non-cytotoxic= IC > 50 µg/mL.
5
0
a
b
c
SI = IC for MRC-5/ IC for aromatase.
Insoluble in testing medium.
Letrozole and doxorubicin were used as reference drugs.
5
0
50
Results showed that all bis-sulfonamides (9-26) exhibited the aromatase inhibitory
activities with IC in the range of 0.05-11.6 μM, except for para- ring A derivative with R =
5
0
tetramethyl benzene 23 (inactive, IC > 12.5 μM).
5
0
Bis-sulfonamide 9 with R = 4-CH C H showed the aromatase inhibitory activity
3
6
4
with IC of 0.13 µM. Replacement of the R group of compound 9 with hindered R groups
5
0
(2,3,5,6-tetramethylbenzene and naphthalene) as seen in compounds 10 and 11 decreased the
inhibitory potency in 24.6-fold and 1.6-fold, respectively. It could be implied that steric effect
of R groups is responsible for inhibitory potency. Reduction of the activity was noted for
compound 12 with R = 4-OCH C H (IC = 11.6 µM), compound 13 with R = 4-CF C H
4
3
6
4
50
3
6
(
IC = 0.43 µM) and compound 14 with R = 4-FC H (IC = 2.19 µM) compared with the
50 6 4 50
compound 9 (R = 4-CH C H ). Promisingly, the inhibition of aromatase was dramatically
3
6
4
improved when hydrophobic halogen groups (Cl and Br) were introduced as seen in
compounds 15 (IC50 = 0.05 µM) and 16 (IC50 = 0.06 µM). Comparable activity with
compound 9 (R = 4-CH C H ) was noted for bis-sulfonamide 17 with R = 4-CNC H (IC =
3
6
4
6
4
50
0
.16 µM) whereas reduction of the potency was seen in compounds 18 (IC = 0.28 µM) and
50
1
9 (IC = 0.37 µM) having the polar electron withdrawing substituents (4-COCH and 4-
50 3
NO ) on the benzene ring. When 4-nitro group of the compound 19 was moved to 2- and 3-
2
positions led to compounds 20 and 21 with decreasing 9.9-fold and 7.4-fold aromatase
inhibitory potency, respectively. The polar electron donating amino compound 22 (R = 3-
NH C H ) diminished the aromatase inhibitory potency compared with the nitro compound
2
6
4
2
1. It could be concluded that in the meta-isomer ring A, both of steric and electronic effects
influenced the aromatase inhibitory potency. Apparently, the effect of R substituents
increased the potency in the following trend: 4-Cl phenyl (15) ~ 4-Br phenyl (16) > 4-CH3
phenyl (9) > 4-CN phenyl (17) > 2-naphthyl (11) > 4-CH CO phenyl (18) > 4-NO phenyl
3
2
(
19) > 4-CF phenyl (13) > 4-F phenyl (14) > 3-NO phenyl (21) > 2,3,5,6-tetraCH phenyl
3 2 3

JOURNAL PRE-PROOF
8
(
10) > 3-NH phenyl (22) > 2-NO phenyl (20) > 4-OCH phenyl (12). It should be noted that
2 2 3
the better activity may require higher lipophilicity with lower polarity as seen in compounds
5 (Cl) & 16 (Br) > 14 (F) and in 9 (CH ) > 13 (CF ).
1
3
3
Obviously, replacement of meta-xylene ring with para counterpart (compounds 10
vs 23 and compounds 17 vs 24) afforded the compounds with decreased aromatase inhibitory
potency. Considering the meta-isomers 15 and 26 bearing different hydrocarbon chain length
on ring A, the compound 15 (R = 4-ClC H ) displayed higher activity than the compound 26
6
4
(
R = 4-ClC H , IC = 0.78 μM). This could suggest that the meta-xylene (15) is an
6 4 50
appropriate size that can interact with the target site of action.
Previously, meta-bis-thioureas 6a (R = Cl, Figure 1) was reported to exert the
aromatase inhibitory activitiy with IC value of 5.5 μM, whereas 4-nitro compound 6b (R =
5
0
NO ) displayed the most potent activity (IC = 0.8 μM ).25 In this study, meta-bis-
2
50
sulfonamide 15 (R = 4-ClC H , Scheme 1) exhibited the most potent activity (IC = 0.05
6
4
50
μM) when compared with the nitro compound 19 (R = 4-NO C H , IC = 0.37 μM). The
2
6
4
50
results indicated that the replacement of meta-bis-thiourea moiety25 with meta-bis-
sulfonamide analogs, mostly, provided the compounds with improved activity (Table 1).
2
.3 Cytotoxic activity
The bis-sulfonamide derivatives (9-26) were assayed for their cytotoxic effects
against hormone-dependent breast cancer cell line (T47-D). The results (Table 1) showed that
compounds 9, 10, 12, 13, 15, 17, 19, 20, 22 and 26 exhibited weak to moderate cytotoxic
activity against the breast cancer cells (IC 30.92-90.61 µM).
5
0
These sulfonamides (9-26) were also tested against the normal embryonic lung
(MRC-5) cell line (Table 1). Most analogs were non-cytotoxic toward the normal cell
whereas compounds 10, 13, 15, 23, 24 and 26 showed low cytotoxic effect. The most active
compounds 15 and 16 displayed high safety index with the SI values of 1370.4 and >1451.0,
respectively.
Additionally, some of these bis-sulfonamides (i.e., 9, 12, 13, 16, and 19) have been
reported to induce apoptosis of several types of cancer cells including breast cancer MCF-7
cell line. The apoptosis induction mechanism of these compounds was proposed to be via
-
acting as an artificial Cl ion transporter leading to the disturbance of intracellular ion
homeostasis and excessive production of reative oxygen speices (ROS).17

JOURNAL PRE-PROOF
9
2
.4 Molecular docking
Molecular docking was performed to reveal possible binding modes of bis-
sulfonamide analogs (9-26) and the target aromatase enzyme. The validation of docking
protocol was performed by redocking the co-crystallized ligand, ASD, towards its target
(
aromatase enzyme, PDB ID: 3EQM). The protocol provided acceptable accuracy as
indicated by the redocking root mean standard deviation (RMSD) value of 0.705 A° (Figure
A). The validated protocol was subsequently used for investigation of possible binding
2
modes of the bis-sulfonamides 9-26. All investigated bis-sulfonamides could bind within the
same binding site as shown in Figure 2B. The binding modality of the most potent compound
1
5 is provided in Figure 2C.
A
B
C
Figure 2. Heme and Fe atom are shown in orange and red, respectively. (A) Redocking of the
co-crystallized ligand androstenedione (ASD) towards aromatase enzyme providing RMSD
value of 0.705 A° (original and redocked poses are shown in green and yellow, respectively).
(B) Docking poses of all investigated compounds 9-26. (C) Binding modality of the most
potent bis-sulfonamide 15.
According to the experimental IC values, the active compounds could be classified
5
0
into three groups. Compounds 15 and 16 (IC values < 0.1 µM) are considered the highly
5
0
actives. The compounds with 0.1 < IC values < 1.0 µM (compounds 9, 17, 11, 18, 19, 13,
5
0
2
5, 24, and 26) are classified as moderately actives. The compounds with IC values > 1.0
50
µM are classified as weakly actives (compounds 14, 21, 10, 22, 20, and 12). To elucidate the
crucial chemical interactions which are essential for potent aromatase inhibitory activity, a

JOURNAL PRE-PROOF
1
0
comparison of the 2D protein-ligand interactions of the bis-sulfonamide representatives from
each group was also performed (Figure 3). Herein, the most potent compounds of each group
are selected as the representative compounds (compounds 15 and 16 for the highly actives,
compound 9 for the moderately active, and compound 14 for the weakly active).
In overview, the interaction diagrams revealed that the representative compounds
from all groups shared some common features in forming the hydrogen bonding with Ser478,
the π-π interaction with Phe221, and the hydrophobic interactions with Leu477, Phe221,
Val370, and Thr310 residues of the enzyme (Figures 3A-3D). In contrast, the unique binding
characteristics were observed for the two representative compounds from the highly active
group (compounds 15 and 16) including the hydrogen-bonding formation with His480 and
3
+
the cation-π interaction with the Fe of the enzyme (Figures 3A and 3B). The comparison of
binding interactions of these two compounds with that of the natural ASD also revealed that
the substitution of the halogen atoms on the terminal benzene rings (R groups) of the bis-
sulfonamides (i.e., Cl atom in compound 15, and Br atom in compound 16) may play role in
the formation of hydrophobic interaction with the Leu477 to mimic steroidal backbone of the
ASD (Figures 3A, 3B, and 3E). Additionally, these structural features may facilitate the
formation of cation-π interaction between one of the terminal substituted benzene ring (R
group) and the Fe3+ of the enzyme (Figures 3A, 3B). In summary, molecular docking
indicated that the hydrogen-bond formation with His480 as well as the cation-π interaction
3
+
with Fe of the enzyme are suggested to be crucial binding features for the highly potent
activities of the halogen bis-sulfonamide analogs 15 and 16. As the above mentioned, such
hydrophobic interaction with Leu477 and Phe221, hydrogen bonding with Ser478, π-π
3
+
interaction with Phe221, and cation-π interaction with Fe of the enzyme, except for H-bond
2
5
formation with His480, were also observed for the bis-thiourea 6b. The more potent activity
of bis-sulfonamides (15 and 16) than bis-thiourea 6b could be facilitated by the His480 in
forming H-bond with the aromatase enzyme.
Accordingly, His480 residue crucial for the aromatase inhibition has been supported
by mutagenesis studies26,27 and the compounds 15 and 16 could mimic nonsteroidal AIs such
as letrozole by the characteristic interaction with heme1,28
.

JOURNAL PRE-PROOF
1
1
A
B
C
D
E
Figure 3. The 2D ligand-protein interactions of the highly active compounds 15 and 16 (A
and B), the moderately active compound 9 (C), the weakly active compound 14 (D), and the
natural substrate ASD (E).
2
.5 QSAR
QSAR modelling was performed to gain insights into structure-activity relationships
of the tested compounds (9-26, scheme I) and their experimental activities (Table 1). A set of
7 active compounds along with their chemical structures were used as a data set for QSAR
1
analysis. The compounds were drawn, geometrically optimized, and calculated to obtain a
large set of descriptors (i.e., molecular descriptors and quantum chemical descriptors).
Correlation-based feature selection followed by stepwise multiple linear regression (MLR)
was performed to select a final set of important descriptors for QSAR modelling as shown in
Table 2. Values of informative molecular descriptors of the compounds in the data set are
provided in Table 3.

JOURNAL PRE-PROOF
1
2
Table 2 Definition of descriptors used for QSAR modelling
Descriptor Type
Definition
Mor03m
3D-MoRSE descriptors Signal 03 / weighted by mass 3D-MoRSE
descriptors
GATS6v
2D autocorrelations
Geary autocorrelation of lag 6 weighted by van der
Waals volume
Table 3 The informative molecular descriptor values of compounds 9-26
Compound
9
Mor03m
-5.754
-7.546
-4.948
-5.125
-5.761
-5.648
-3.091
-2.644
-5.113
-4.941
-5.22
GATS6v
1.249
0.872
1.581
0.628
1.047
0.588
1.249
1.705
1.269
1.164
0.869
0.872
0.526
0.896
0.974
1.368
0.402
0.925
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
-5.622
-7.521
-7.498
-8.481
-6.733
-5.927
-6.3

JOURNAL PRE-PROOF
1
3
The QSAR model was successfully constructed using MLR algorithm as shown in equation
1).
(
푝퐼퐶50 = 0.2081(푀표푟03푚) + 0.9138 (퐺퐴푇푆6푣) + 0.4406
(1)
The model provided an acceptable predictive performance as indicated by R = 0.7951 and
Tr
RMSE = 0.4009, and R = 0.6759 and RMSECV = 0.4992 for training and testing sets,
Tr
CV
respectively. The experimental and predicted activity (pIC ) of the compounds are shown in
5
0
Table 4 and Figure 4.
Table 4 Experimental and predicted aromatase inhibitory activities (pIC ) of compounds 9-
5
0
2
6
Compound
9
Experimental
0.886
-0.505
0.678
-1.064
0.367
-0.34
Predicted
0.327
-0.29
0.906
0.204
0.187
-0.167
0.794
1.601
0.511
0.47
Error a
-0.559
0.215
0.228
1.268
-0.18
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
0.173
-0.507
0.379
-0.285
-0.083
-0.312
0.681
-0.26
1.301
1.222
0.796
0.553
0.432
-0.565
-0.439
-0.519
-b
0.12
0.116
-0.699
-0.248
-b
0.271
-b
0.18
0.331
-0.685
-0.036
0.151
-0.961
-0.144
0.276
0.108
a
Error = Predicted pIC - Experimental pIC
50
5
0
-b Inactive compound was excluded from the QSAR modelling.

JOURNAL PRE-PROOF
1
4
Figure 4. Experimental aromatase inhibitory activities of the tested compounds were plotted
against their predicted activities obtained by the constructed QSAR model (experimental
versus predicted pIC50 values). Closed triangles and solid regression line represented
activities of the compounds generated by training set, whereas those of leave-one-out
validated testing set are presented by opened circles and dash regression line.
According to the constructed QSAR model, two descriptors including GATS6v (van
der Waals volume) and Mor03m (mass) with high regression coefficient values of 0.9138 and
0
.2081, respectively, played role in the aromatase inhibitory effect of the compounds.
In overview, the QSAR analysis provided information suggesting the general desired
characteristics of the compounds that are in concordance with the experimental findings.
Mostly, the more potent compound displayed the higher value of either GATS6v or Mor03m,
or both of the two descriptors, depending on the property of substituents and/or substitution
patterns on the ring A (Scheme I) i.e., meta and para isomers. Considering the most potent
meta-compound 16, which is considered highly active, possessed the highest values of
GATS6v (1.705) and Mor03m (-2.644) when compared to others (Table 3). Additionally, the
most potent meta-compound 15 had high values of both Mor03m (-3.091) and GATS6v
(1.249). In contrast, the rest of compounds with weaker activity and inactive compound (23),

JOURNAL PRE-PROOF
1
5
mostly, possessed the lower values of both descriptors (Table 3) compared with the most
potent compounds (15 and 16).
It should be noted that meta-isomer 26, the analog of 15 bearing shorter hydrocarbon
chain on ring A, displayed markedly decreased activity (pIC = 0.108) with lower values of
5
0
GATS6v (0.925) and Mor03m (-6.300) compared with the most potent compound 15 (pIC50
=
1.301, GATS6v = 1.249, Mor03m = -3.091).
In case of moderately active compound 9 ((pIC = 0.886, R = 4-CH C H ), it had
5
0
3
6
4
lower van der Waals volume (GATS6v = 1.249) and mass (Mor03m = -5.754) compared with
the most potent compound 16. Comparing with the most potent compound 15 (Mor03m = -
3
.091), compound 9 displayed the lower Mor03m value but with the same GATS6v value
1.249).
For the weakly active compound 14 (pIC = -0.34, R = 4-FC H ), its lower values of
(
5
0
6
4
both descriptors (GATS6v = 0.588, Mor03m = -5.648) were noted when compared with the
most potent 4-Cl (15) and 4-Br (16) containing compounds with the higher values of
GATS6v and Mor03m descriptors. This could be due to the effect of atomic size of these
halogens (Br > Cl > F).
In addition, the higher values of both van der Waals volume and mass descriptors
involved in the higher activity of compounds such as compounds 9 (R = 4-CH C H ) > 13 (R
3
6
4
=
4-CF C H ), and 17 (R = 4-CNC H ) > 19 (R = 4-NO C H ) as shown in Table 3. On the
3 6 4 6 4 2 6 4
other hands, the higher van der Waals volume was seen in the more potent compounds i.e., 13
14 and 19 > 12, R = 4-OCH C H (Table 3).
>
3
6
4
The position where both sulfonamides moieties are attached on the central benzene
ring A also affects aromatase inhibitory ability of the compound. The compounds possessing
the meta-bis-sulfonamides exhibited more potent activities when compared to their para-
derivatives (i.e., tetra-CH derivatives (10 > 23) and CN-derivatives (17 > 24)). Regarding
3
the QSAR model, the loss of activity of these para-compounds (23 and 24) could be due to
the decreased value of mass descriptor Mor03m (Mor03m: 10 = -7.546, 23 = -8.481, 17 = -
5
.113, 24 = -6.733, Table 3). The influence of mass descriptor Mor03m was also observed for
the benzene sulfonamide ring substituted with NO moiety. Among three NO -derivatives
2
2
(
19-21), the 4-NO -substituted compound 19 with the highest Mor03m value provided more
2
potent activity (pIC = 0.432, Mor03m = -5.22) than those of others i.e., ortho-compound 20
5
0

JOURNAL PRE-PROOF
1
6
(
pIC = -0.565, Mor03m = -5.622) and meta-compound 21 (pIC = -0.439, Mor03m = -
5
0
5
0
7
3
.521).
. Conclusions
Eighteen bis-sulfonamides (9–26) were synthesized by a simple one step sulfonylation
of amine compounds using commercially available starting materials. Their aromatase
inhibitory activities, molecular docking, and QSAR were investigated. All bis-sulfonamides
displayed the aromatase inhibitory activity with IC50 range of 0.05-11.6 M except for
compound 23. Promisingly, the bis-sulfonamides 15 and 16 were shown to be the most potent
inhibitors (IC of 0.05 and 0.06 M, respectively) with high safety index. The molecular
5
0
docking revealed that the 4-Cl and 4-Br benzene substitutents (R) on the compounds 15 and
6 may play crucial role in the hydrophobic interaction with Leu477 of the aromatase to
1
mimic steroidal backbone of the natural substrate (ASD). In addition, the QSAR study
indicated that both compounds displayed high values of van der Waals volume (GATS6v)
and mass (Mor03m) descriptors in governing the aromatase inhibitory activity. In summary,
this study demonstrated the integrative approach towards discovering a novel class of
aromatase inhibitors where two compounds (15 and 16) were highlighted as promising
compounds for further development. Additionaly, the computational results would benefits
further rational design of the related compounds for therapeutic applications.
4
. Experimental section
4
.1 Chemistry
Column chromatography was carried out using silica gel 60 (70–230 mesh ASTM).
Analytical thin-layer chromatography (TLC) was performed using silica gel 60 F254
1
13
aluminum sheets. H- and C- NMR spectra were recorded on a Bruker AVANCE 300 NMR
1
13
spectrometer (operating at 300 MHz for H and 75 MHz for C). The following standard
abbreviations were used for signal multiplicities: singlet (s), doublet (d), triplet (t), quartet
(
q), multiplet (m), and broad (br). Mass spectra were recorded on a Bruker Daltonics
microTOF). IR spectra were obtained using a universal attenuated total reflectance attached
(
on a Perkin–Elmer Spectrum One spectrometer. Melting points were determined using a
Griffin melting point apparatus and were uncorrected.
4
.2 General procedure for the synthesis of sulfonamides (9-24)

JOURNAL PRE-PROOF
1
7
A solution of xylylenediamine 7a,b (2.5 mmol) in dichloromethane (50 mL) was
added in dropwise manner to a stirred mixture of appropriate benzenesulfonyl chloride 8a-m
(
5 mmol) and sodium carbonate (10 mmol) in dichloromethane (20 mL). The reaction
mixture was stirred at room temperature for 15-24 h (monitored by TLC), and distilled water
20 mL) was added. The organic phase was separated and the aqueous phase was extracted
(
with dichloromethane (2  30 mL). The organic extracts were combined and washed with
water (30 mL). The organic layer was dried over anhydrous sodium sulfate, filtered and
evaporated to dryness under reduced pressure. The crude product was further purified by
recrystallized or column chromatography on silica gel to obtain the pure product.
4
.2.1 N,N'-(1,3-phenylenebis(methylene))bis(4-methylbenzenesulfonamide) (9)17
From m-xylylenediamine 7a and 4-methylbenzenesulfonyl chloride 8a. Off-white
o
-1
1
solid. 80% yield; mp 151-152 C; IR (UATR) cm : 3298, 3268, 1597, 1323, 1152. H NMR
(
300 MHz, DMSO-d ) δ 2.37 (s, 6H, 2 × CH ), 3.86 (d, J = 6.3 Hz, 4H, 2 × CH NH), 7.06 (s,
6 3 2
1
4
H, ArH), 7.09 (d, J = 7.8 Hz, 2H, ArH), 7.19 (t, J = 7.5 Hz, 1H, ArH), 7.37 (d, J = 8.1 Hz,
1
3
H, ArH), 7.68 (d, J = 8.1 Hz, 4H, ArH), 8.06 (t, J = 6.3 Hz,, 2H, 2 × NH). C NMR (75
MHz, DMSO-d ) δ 21.4, 46.5, 126.9, 127.0, 127.2, 128.7, 130.1, 138.1, 143.1. TOF-MS m/z:
6
4
45.1237 (Calcd for C H N O S : 445.1250).
22 25 2 4 2
4
.2.2 N,N'-(1,3-phenylenebis(methylene))bis(2,3,5,6-tetramethylbenzenesulfonamide) (10)
From m-xylylenediamine 7a and 2,3,5,6-tetramethylbenzenesulfonyl chloride 8b. Off-
o
-1
1
white solid. 88% yield; mp 170-172 C; IR (UATR) cm : 3311, 1463, 1318, 1143. H NMR
300 MHz, DMSO-d ) δ 2.15 (s, 12H, 4 × CH ), 2.39 (s, 12H, 4 × CH ), 3.82 (s, 4H, 2 ×
(
6
3
3
CH NH), 6.70 (s, 1H, ArH), 6.98 (d, J = 7.3 Hz, 2H, ArH), 7.08 (t, J = 6.6 Hz, 1H, ArH),
2
1
3
7
1
.11 (s, 2H, ArH), 7.90 (s, 2H, 2 × NH). C NMR (75 MHz, DMSO-d ) δ 18.0, 20.9, 45.8,
6
26.7, 127.1, 128.1, 134.6, 135.3, 135.7, 137.8, 139.4. TOF-MS m/z: 551.2003 (Calcd for
C H N NaO S : 551.2009).
2
8
36
2
4 2
4
.2.3 N,N'-(1,3-phenylenebis(methylene))bis(naphthalene-2-sulfonamide) (11)17
From m-xylylenediamine 7a and 2-naphthalenebenzenesulfonyl chloride 8c. Off-
o
-1
1
white solid. 72% yield; mp 148-150 C; IR (UATR) cm : 3279, 1592, 1322, 1154. H NMR
(
300 MHz, DMSO-d ) δ 3.86 (s, 4H, 2 × CH NH), 7.00-7.13 (m, 4H, ArH), 7.58-7.70 (m,
6 2
4
4
H, ArH), 7.78 (dd, J = 8.6, 1.6 Hz, 2H, ArH), 8.00 (d, J = 7.5 Hz, 2H, ArH), 8.05-8.16 (m,
1
3
H, ArH), 8.39 (s, 2H, ArH). C NMR (75 MHz, DMSO-d ) δ 46.5, 122.7, 126.9, 127.3,
6

JOURNAL PRE-PROOF
1
8
1
5
4
27.8, 127.9, 128.2, 128.6, 129.1, 129.6, 129.7, 132.2, 134.6, 138.0, 138.2. TOF-MS m/z:
39.1060 (Calcd for C H N NaO S : 539.1070).
2
8
24
2
4 2
.2.4 N,N'-(1,3-phenylenebis(methylene))bis(4-methoxybenzenesulfonamide) (12)17
From m-xylylenediamine 7a and 4-methoxybenzenesulfonyl chloride 8d. Off-white solid.
o
-1
1
8
1% yield; mp 128-130 C; IR (UATR) cm : 3265, 1596, 1325, 1151. H NMR (300 MHz,
DMSO-d ) δ 3.82 (s, 6H, 2 × OCH ), 3.86 (s, 4H, 2 × CH NH), 7.03-7.20 (m, 8H, ArH), 7.71
6
3
2
1
3
(
d, J = 8.8 Hz, 4H, ArH). C NMR (75 MHz, DMSO-d ) δ 46.5, 56.1, 114.8, 126.9, 127.3,
6
1
4
4
28.6, 129.1, 132.8, 138.1, 162.6. TOF-MS m/z: 499.0962 (Calcd for C H N NaO S :
22 24 2 6 2
99.0968).
.2.5 N,N'-(1,3-phenylenebis(methylene))bis(4-trifluoromethylbenzenesulfonamide) (13)17
From m-xylylenediamine 7a and 4-trifluoromethylbenzenesulfonyl chloride 8e. Off-white
o
-1
1
solid. 71% yield; mp 176-178 C; IR (UATR) cm : 3267, 1592, 1324, 1138. H NMR (300
MHz, DMSO-d ) δ 3.94 (d, J = 6.2 Hz, 4H, 2 × CH NH), 7.00-7.14 (m, 4H, ArH), 7.91 (d, J
6
2
1
3
=
8.9 Hz, 4H, ArH), 7.96 (d, J = 8.9 Hz, 4H, ArH), 8.42 (t, J = 6.2 Hz, 2H, 2 × NH). C
1
3
NMR (75 MHz, DMSO-d ) δ 46.0, 123.5 (q, J =271), 126.3 (q, J =4), 126.5, 127.0,
6
CF
CF
2
1
27.4, 128.2, 132.0 (q, J =32), 137.2, 144.7. TOF-MS m/z: 553.0710 (Calcd for
CF
C H F N O S : 553.0685).
2
2
19
6
2
4 2
4
.2.6 N,N'-(1,3-phenylenebis(methylene))bis(4-fluorobenzenesulfonamide) (14)
From m-xylylenediamine 7a and 4-fluorobenzenesulfonyl chloride 8f. Off-white solid. 89%
o
-1
1
yield; mp 247-248 C; IR (UATR) cm : 3285, 1592, 1327, 1153. H NMR (300 MHz,
DMSO-d ) δ 3.92 (s, 4H, 2 × CH NH), 7.03-7.20 (m, 4H, ArH), 7.40 (t, J = 8.9 Hz, 4H,
6
2
1
3
ArH), 7.84 (dd, J = 5.3, 8.6 Hz, 4H, ArH), 8.23 (br s, 2H, 2 × NH). C NMR (75 MHz,
2
3
DMSO-d ) δ 46.5, 116.7 (d, J =23), 126.9, 127.3, 128.6, 130.0 (d, J =9), 137.6, 138.0,
6
CF
CF
1
1
64.5 (d, J =249). TOF-MS m/z: 475.0569 (Calcd for C H F N NaO S : 475.0568).
CF 20 18 2 2 4 2
4
.2.7 N,N'-(1,3-phenylenebis(methylene))bis(4-chlorobenzenesulfonamide) (15)
From m-xylylenediamine 7a and 4-chlorobenzenesulfonyl chloride 8g. Off-white solid. 82%
o
-1
1
yield; mp 157-158 C; IR (UATR) cm : 3279, 1587, 1327, 1159. H NMR (300 MHz,
DMSO-d ) δ 3.92 (d, J = 4.8 Hz, 4H, 2 × CH NH), 7.06 (s, 1H, ArH), 7.07 (d, J = 7.3 Hz, 2H,
6
2
ArH), 7.16 (t, J = 6.8 Hz, 1H, ArH), 7.62 (d, J = 8.5 Hz, 4H, ArH), 7.77 (d, J = 8.5 Hz, 4H,
1
3
ArH), 8.28 (br t, 2H, 2 × NH). C NMR (75 MHz, DMSO-d ) δ 46.4, 127.0, 127.3, 128.7,
6
1
4
28.9, 129.7, 137.7, 137.8, 140.0. TOF-MS m/z: 485.0159 (Calcd for C H Cl N O S :
20 19 2 2 4 2
85.0158).

JOURNAL PRE-PROOF
1
9
4
.2.8 N,N'-(1,3-phenylenebis(methylene))bis(4-bromobenzenesulfonamide) (16)17
From m-xylylenediamine 7a and 4-bromobenzenesulfonyl chloride 8h. Off-white solid. 88%
o
-1
1
yield; mp 169-170 C; IR (UATR) cm : 3276, 1574, 1327, 1155. H NMR (300 MHz,
DMSO-d ) δ 3.91 (s, 4H, 2 × CH NH), 7.06 (s, 1H, ArH), 7.07 (d, J = 6.8 Hz, 2H, ArH), 7.16
6
2
1
3
(
t, J = 6.8 Hz, 1H, ArH), 7.70 (d, J = 8.6 Hz, 4H, ArH), 7.76 (d, J = 8.6 Hz, 4H, ArH). C
NMR (75 MHz, DMSO-d ) δ 46.5, 126.5, 126.9, 127.4, 128.6, 129.0, 132.6, 137.9, 140.6.
6
TOF-MS m/z: 594.8967 (Calcd for C H Br N NaO S : 594.8967).
2
0
18
2
2
4 2
4
.2.9 N,N'-(1,3-phenylenebis(methylene))bis(4-cyanobenzenesulfonamide) (17)
From m-xylylenediamine 7a and 4-cyanobenzenesulfonyl chloride 8i. Yellow solid. 70%
o
-1
1
yield; mp 154-155 C; IR (UATR) cm : 3284, 2233, 1333, 1159. H NMR (300 MHz,
DMSO-d ) δ 3.96 (s, 4H, 2 × CH NH), 7.04 (d, J = 8.0 Hz, 2H, ArH), 7.05 (s, 1H, ArH), 7.12
6
2
(t, J = 7.4 Hz, 1H, ArH), 7.90 (d, J = 8.5 Hz, 4H, ArH), 8.02 (d, J = 8.4 Hz, 4H, ArH), 8.29
1
3
(
br s, 2H, 2 × NH). C NMR (75 MHz, DMSO-d ) δ 46.5, 115.2, 118.2, 127.0, 127.4, 127.7,
6
1
28.7, 133.8, 137.7, 145.4. TOF-MS m/z: 489.0650 (Calcd for C H N NaO S : 489.0662).
22 18 4 4 2
4
.2.10 N,N'-(1,3-phenylenebis(methylene))bis(4-acetylbenzenesulfonamide) (18)
From m-xylylenediamine 7a and 4-acetylbenzenesulfonyl chloride 8j. Off-white solid. 72%
o
-1
1
yield; mp 270-271 C; IR (UATR) cm : 3292, 1688, 1329, 1159. H NMR (300 MHz,
DMSO-d ) δ 2.62 (s, 6H, 2 × CH CO), 3.92 (s, 4H, 2 × CH NH), 7.07 (s, 1H, ArH), 7.05 (d, J
6
3
2
=
7.3 Hz, 2H, ArH), 7.13 (t, J = 6.7 Hz, 1H, ArH), 7.88 (d, J = 8.2 Hz, 4H, ArH), 8.08 (d, J =
1
3
8
1
4
.2 Hz, 4H, ArH). C NMR (75 MHz, DMSO-d ) δ 27.5, 46.6, 126.9, 127.3, 128.6, 129.3,
6
38.2, 139.7, 145.3, 197.8. TOF-MS m/z: 523.0956 (Calcd for C H N NaO S : 523.0968).
2
4
24
2
6 2
.2.11 N,N'-(1,3-phenylenebis(methylene))bis(4-nitrobenzenesulfonamide) (19)17
From m-xylylenediamine 7a and 4-nitrobenzenesulfonyl chloride 8k. Yellow solid. 80%
o
-1
1
yield; mp 170-171 C; IR (UATR) cm : 3306, 1526, 1350, 1161. H NMR (300 MHz,
DMSO-d ) δ 3.99 (s, 4H, 2 × CH NH), 7.05 (d, J = 8.0 Hz, 2H, ArH), 7.06 (s, 1H, ArH), 7.14
6
2
(t, J = 6.5 Hz, 1H, ArH), 8.00 (d, J = 8.9 Hz, 4H, ArH), 8.37 (d, J = 8.9 Hz, 4H, ArH), 8.55 (s,
1
3
2
1
4
H, 2 × NH). C NMR (75 MHz, DMSO-d ) δ 46.4, 124.9, 127.1, 127.4, 128.5, 128.8,
6
37.7, 146.7, 149.9. TOF-MS m/z: 529.0447 (Calcd for C H N NaO S : 529.0458).
2
0
18
4
8 2
.2.12 N,N'-(1,3-phenylenebis(methylene))bis(2-nitrobenzenesulfonamide) (20)
From m-xylylenediamine 7a and 2-nitrobenzenesulfonyl chloride 8l. Off-white solid. 70%
o
-1
1
yield; mp 154-155 C; IR (UATR) cm : 3345, 1535, 1339, 1162. H NMR (300 MHz,
1
3
DMSO-d ) δ 4.05 (s, 4H, 2 × CH NH), 7.04-7.15 (m, 4H, ArH), 7.69-7.92 (m, 8H, ArH). C
6
2

JOURNAL PRE-PROOF
2
0
NMR (75 MHz, DMSO-d ) δ 46.8, 124.6, 126.8, 127.3, 128.6, 130.0, 132.8, 134.0, 138.2,
6
1
48.0. TOF-MS m/z: 529.0456 (Calcd for C H N NaO S : 529.0458).
20 18 4 8 2
4
.2.13 N,N'-(1,3-phenylenebis(methylene))bis(3-nitrobenzenesulfonamide) (21)
From m-xylylenediamine 7a and 3-nitrobenzenesulfonyl chloride 8m. Off-white solid. 74%
o
-1
1
yield; mp 173-174 C; IR (UATR) cm : 3295, 1531, 1349, 1166. H NMR (300 MHz,
DMSO-d ) δ 3.94 (s, 4H, 2 × CH NH), 6.92-7.10 (m, 4H, ArH), 7.82 (t, J = 7.8 Hz, 2H,
6
2
1
3
ArH), 8.12 (d, J = 7.6 Hz, 2H, ArH), 8.36-8.42 (m, 4H, ArH), 8.48 (br s, 2H, 2 × NH). C
NMR (75 MHz, DMSO-d ) δ 46.0, 121.3, 126.6, 126.9, 127.0, 128.2, 131.2, 132.5, 137.0,
6
1
42.5, 147.7. TOF-MS m/z: 529.0465 (Calcd for C H N NaO S : 529.0458).
20 18 4 8 2
4
.2.14 N,N'-(1,4-phenylenebis(methylene))bis(2,3,5,6-tetramethylbenzenesulfonamide) (23)
From p-xylylenediamine 7b and 2,3,5,6-tetramethylbenzenesulfonyl chloride 8b. Off-white
o
-1
1
solid. 65% yield; mp 225-227 C; IR (UATR) cm : 3279, 1322, 1153. H NMR (300 MHz,
DMSO-d ) δ 2.18 (s, 12H, 4 × CH ), 2.41 (s, 12H, 4 × CH ), 3.89 (d, J = 6.2 Hz, 4H, 2 ×
6
3
3
1
3
CH NH), 6.99 (s, 4H, ArH), 7.15 (s, 2H, ArH), 7.93 (t, J = 6.2 Hz, 2H, 2 × NH). C NMR
2
(
75 MHz, DMSO-d ) δ 18.1, 20.9, 45.5, 127.6, 134.7, 135.3, 135.8, 137.1, 139.3. TOF-MS
6
m/z: 551.2002 (Calcd for C H N NaO S : 551.2009).
2
8
36
2
4 2
4
.2.15 N,N'-(1,4-phenylenebis(methylene))bis(4-cyanobenzenesulfonamide) (24)
From p-xylylenediamine 7b and 4-cyanobenzenesulfonyl chloride 8i. Off-white solid. 87%
o
-1
1
yield; mp 244-245 C; IR (UATR) cm : 3252, 2235, 1335, 1159. H NMR (300 MHz,
DMSO-d ) δ 3.97 (d, J = 6.2 Hz, 4H, 2 × CH NH), 7.09 (s, 4H, ArH), 7.90 (d, J = 8.2 Hz,
6
2
1
3
4
1
H, ArH), 8.03 (d, J = 8.2 Hz, 4H, ArH). C NMR (75 MHz, DMSO-d ) δ 46.3, 115.1,
6
18.2, 127.3, 127.7, 128.0, 133.7, 136.9, 145.6. TOF-MS m/z: 489.0667 (Calcd for
C H N NaO S : 489.0662).
2
2
18
4
4 2
4
.3 General procedure for the synthesis of sulfonamides (25-26)
Benzenesulfonyl chloride (5 mmol) was added to a solution of m-phenylenediamine
(2.5 mmol) in pyridine (5 mL), and the mixture was stirred under reflux for 6 h then
concentrated under reduced pressure. Water (20 mL) was added and extracted with
dichloromethane (3 × 20 mL). The organic layer was dried over anhydrous sodium sulfate,
filtered and evaporated to dryness under reduced pressure. The crude product was further
purified by column chromatography on silica gel to obtain the pure product.

JOURNAL PRE-PROOF
2
1
4
.3.1 N,N'-(1,3-phenylene)bis(4-methoxybenzenesulfonamide) (25)
From m-phenylenediamine 7c and 4-methoxybenzenesulfonyl chloride 8d. Brown solid. 64%
o
-1
1
yield; mp 155-156 C; IR (UATR) cm : 3256, 1596, 1303, 1150. H NMR (300 MHz,
DMSO-d ) δ 3.79 (s, 6H, 2 × OCH ), 6.65 (dd, J = 8.1, 2.0 Hz, 2H, ArH), 6.97-7.07 (m, 5H,
6
3
ArH), 7.09 (t, J = 1.9 Hz, 1H, ArH), 7.62 (d, J = 8.9 Hz, 4H, ArH), 10.15 (br s, 2H, 2 × NH).
1
3
C NMR (75 MHz, DMSO-d ) δ 56.1, 110.8, 114.7, 115.2, 129.3, 130.1, 131.5, 139.1,
6
1
62.9. TOF-MS m/z: 471.0662 (Calcd for C H N NaO S : 471.0655).
2
0
20
2
6 2
4
.3.2 N,N'-(1,3-phenylene)bis(4-chlorobenzenesulfonamide) (26)
From m-phenylenediamine 7c and 4-chlorobenzenesulfonyl chloride 8g. Yellow solid. 66%
o
-1
1
yield; mp 158-159 C; IR (UATR) cm : 3231, 1608, 1317, 1154. H NMR (300 MHz,
DMSO-d ) δ 6.70 (dd, J = 8.1, 2.0 Hz, 2H, ArH), 7.02 (t, J = 1.9 Hz, 1H, ArH), 7.08 (t, J =
6
8
2
1
4
.1 Hz, 1H, ArH), 7.60 (d, J = 8.8 Hz, 4H, ArH), 7.66 (d, J = 8.8 Hz, 4H, ArH), 10.42 (br s,
1
3
H, 2 × NH). C NMR (75 MHz, DMSO-d ) δ 111.6, 116.1, 129.0, 129.8, 130.4, 138.3,
6
38.5, 138.7. TOF-MS m/z: 478.9663 (Calcd for C H Cl N NaO S : 478.9664).
1
8
14
2
2
4 2
.4 The synthesis of N,N'-(1,3-phenylenebis(methylene))bis(3-aminobenzenesulfonamide)
(22)
A mixture of nitrobissulfonamide 21 (4 mmol) and SnCl 2H O (32 mmol) in absolute
2
2
ethanol (20 mL) was stirred under reflux for 6 h then concentrated under reduced pressure.
Water (20 mL) was added and extracted with EtOAc (3 × 20 mL). The organic extracts were
combined and washed with water (20 mL) and brine (20 mL). The organic layer was dried
over anhydrous sodium sulfate, filtered and concentrated. The crude product was further
purified by column chromatography on silica gel to obtain the pure product.
o
-1
Yellow solid. 80% yield; mp 148-150 C; IR (UATR) cm : 3467, 3372, 3278, 1598, 1323,
1
1
155. H NMR (300 MHz, DMSO-d ) δ 3.88 (d, J = 4.5 Hz, 4H, 2 × CH NH), 5.56 (s, 4H, 2
6
2
×
NH ), 6.75 (dd, J = 8.0, 1.9 Hz, 2H, ArH), 6.91 (d, J = 7.8 Hz, 2H, ArH), 7.00 (t, J = 1.9
2
1
3
Hz, 2H, ArH), 7.09-7.26 (m, 6H, ArH), 7.92 (br t, 2H, 2 × NH). C NMR (75 MHz, DMSO-
d6) δ 46.6, 111.6, 113.7, 117.6, 126.8, 127.1, 128.6, 130.0, 138.4, 141.5, 149.8. TOF-MS m/z:
4
69.0959 (Calcd for C H N NaO S : 469.0975).
20 22 4 4 2
4
.5 Aromatase inhibition assay
Aromatase inhibitory activity was evaluated using the modified method of Stressor et
al.. The experiment was performed using the Gentest kit. The activity of aromatase enzyme
2
9

JOURNAL PRE-PROOF
2
2
(CYP19) is indicated by the conversion of the fluorometric substrate, O-benzyl fluorescein
benzyl ester (DBF), to the fluorescein product. Initially, 100 µL of cofactor (containing 78.4
µL of 50 mM phosphate buffer (pH 7.4); 20 µL of 20x NADPH-generating system (26 mM
+
NADP , 66 mM glucose-6-phosphate, and 66 mM MgCl ); and 1.6 µL of 100 U/mL glucose-
2
6
-phosphate dehydrogenase) were added into a 96-well black plate and preincubated for 10
o
min in 37 C water bath. Subsequently, 100 µL of enzyme/substrate mixture containing 77.3
µL of 50 mM phosphate buffer (pH 7.4); 12.5 µL of 16 pmol/mL CYP19; 0.2 µL of 0.2 mM
DBF, and 10 µL of tested compound or negative control (10% DMSO) or positive control
(
letrozole) were added to initiate the reaction. Then, fluorescence signal was recorded
excitation wavelength of 490 nm and emission wavelength of 530 nm) with cutoff 515 nm.
(
The measured absorbance values were used to calculate % inhibition according to the
equation (2). The compounds giving the % inhibition greater than 50 were further
investigated to obtain their IC values.
5
0
푖푛ℎ푖푏푖푡푖표푛 = 100 ‒ [(푠푎푚푝푙푒 ‒ 푏푙푎푛푘) (퐷푀푆푂 ‒ 푏푙푎푛푘
)
%
× 100]
(2)
4
.6 Cytotoxicity assay
The cytotoxic activity of compounds (9-26) was tested using hormone-dependent
breast cancer cell line (T47-D) and normal embryonic lung cell line (MRC-5). T47-D cells
were grown in RPMI-1640 medium supplemented with 2 mM L-gluthamine, 100 U/mL
penicillin-streptomycin, 0.2 U/mL insulin, 4.5 g/L glucose and 10% FBS whereas MRC-5
cells were grown in DMEM medium supplemented with 100 U/mL penicillin streptomycin
and 10% FBS. Briefly, the cells suspended in the corresponding culture medium were
inoculated in 96-well microtiter plates (Corning Inc., NY, USA) at a density of 10,000-
o
2
0,000 cells per well, and then incubated at 37 C under a humidified atmosphere with 95%
air and 5% CO for 24 h. An equal volume of additional medium containing either the serial
2
dilutions of the tested compounds, positive control (etoposide and/or doxorubicin) or negative
control (DMSO) was added to the desired final concentrations. The microtiter plates were
further incubated for 48 h. Cell viability in each well was determined by staining with MTT
assay.30-32 The MTT solution (10 mL/100 mL medium) was added to all wells of the assay,
and the plates were incubated for 2-4 h. Subsequently, DMSO was added to dissolve the
resulting formazan by sonication. The plates were read on a microplate reader (Molecular
Devices, USA) using a test wavelength of 550 nm and a reference wavelength of 650 nm.

JOURNAL PRE-PROOF
2
3
The IC value was determined as the compound concentration that inhibited cell growth by
5
0
5
0%. The compounds exhibited IC > 50 µg/mL were considered as noncytotoxic.
50
4
.7 Molecular Docking
Crystal structure of the target protein, human placental aromatase co-crystallized with
natural substrate ASD, was retrieved from RSCB protein data bank (PDB ID: 3EQM,
http://www.rcsb.org/). The preparation of the protein was initially performed by adding
essential hydrogen atoms and repairing missing side chains, using the WHAT IF web server
version 10.1.33 Subsequently, non-polar hydrogen atoms were merged, Gasteiger atomic
charges were assigned, and atom type of the protein structures were specified using
AutoDock Tools version 1.5.6..34,35 The chemical structures of the investigated ligands (9-26)
were drawn using Marvin Sketch version 6.1.4,36 and were geometrically optimized by
Gaussian 0937 using Becke’s three-parameter hybrid method with the Lee-Yang-Parr
correlation functional (B3LYP) together with the 6-31g(d) basis set. The geometrically
optimized compounds then were prepared by merging non-polar hydrogen atoms, assigning
partial atomic charge, and defining rotatable bonds using AutoDock Tools version 1.5.6.34,35
Molecular docking simulation was conducted using AutoDock Vina, as a part of the PyRx 0.8
3
8
software. The grid boxes sizing 25 × 25 × 25 A° were generated and the center of binding
cavity was allocated using x, y and z coordinates of 85.9275, 54.2004, and 46.0769,
respectively. The validation of docking protocol was performed by redocking of the co-
crystallized ligand ASD to the aromatase protein. The accuracy of the docking protocol was
evaluated by RMSD value, which was calculated by the difference between original and
3
9
redocking poses using the Chimera software. Docking poses of the investigated compounds
4
0
were visualized using PyMOL, and two-dimensional ligand-protein interaction diagrams
were generated using PoseView (http://proteins.plus).41
4
.8 QSAR
Data set
To obtain distribution of data points, the experimental activity (IC values) of the
5
0
compounds were converted to pIC values by taking the negative logarithm to the base of 10
5
0
(
-log IC ). Herein, the inactive compound 23 (IC > 12.5 μM) were excluded from the data
50 50
set. Finally, a set of 17 compounds were included for model construction.

JOURNAL PRE-PROOF
2
4
Molecular structure optimization and descriptor calculation
Geometrical optimization was performed to obtain the low-energy conformers that
were used as input files for descriptor extraction. The initial optimization with semi-empirical
Austin Model 1 (AM1) level was performed followed by the density functional theory (DFT)
calculation using the Becke’s three-parameter hybrid method with the Lee-Yang-Parr
3
7
correlation functional (B3LYP) together with the 6-31g(d) level using Gaussian 09. A set of
quantum chemical descriptors were subsequently extracted from the optimized structures
using an in-house developed script (i.e., mean absolute atomic charge (Q ), total energy
m
(
Etotal), total dipole moment (μ), highest occupied molecular orbital energy (EHOMO), lowest
unoccupied molecular orbital energy (ELUMO), energy difference of HOMO and LUMO
HOMO-LUMOGap), electron affinify (EA), ionization potential (IP), Mulliken
electronegativity (χ), hardness (η), softness (S), electrophilic index (ω ), and electrophilicity
(
i
(ω)). Additionally, the optimized structures were underwent the calculation to obtain a set of
4
2
3
,224 molecular descriptors using Dragon software (version 5.5). The obtained molecular
descriptors included 22 classes i.e., Constitutional descriptors, Topological descriptors, Walk
and path counts, Connectivity indices, Information indices, 2D autocorrelation, Edge
adjacency indices, Burden eigenvalues, Topological charge indices, Eigenvalue-based
indices, Randic molecular profiles, Geometrical descriptors, RDF descriptors, 3D-MoRSE
descriptors, WHIM descriptors, GETAWAY descriptors, Functional group counts, Atom-
centred fragments, Charge descriptors, Molecular properties, 2D binary fingerprints and 2D
frequency fingerprints.
Feature selection
Initially, correlation-based feature selection was performed to filter a set of relevant
descriptors from the whole large set. Pearson’s correlation coefficient (r) values were
calculated for each pair of a descriptor and bioactivity (pIC ). The descriptors whose |r| ≥
5
0
|0.7| then were selected, whereas the rest having |r| < 0.7 were excluded for further feature
selection process. Subsequent feature selection was performed using stepwise multiple linear
regression (MLR) algorithm as implemented in SPSS statistics 18.0 to obtain a final
descriptor set for QSAR modelling.
Multivariate analysis using multiple linear regression

JOURNAL PRE-PROOF
2
5
Multivariate analysis was performed to elucidate relationships of the selected
descriptors and the compound’s activity. Multiple linear regression (MLR) algorithm was
4
3
performed using Waikato Environment for Knowledge Analysis (WEKA) version 3.4.5 to
construct the QSAR model according equation (3), where selected descriptor values and
pIC50 values were assigned as independent variables (X) and dependent variable (Y),
respectively.
Y  B
0

BnXn
(3)
where Y is the pIC values of compounds, B is the intercept and B are the regression
5
0
0
n
coefficient of descriptors Xn.
Data sampling
Data set was separated into 2 subsets (i.e., training set and testing set) by means of
leave-one-out cross validation (LOO-CV). One compound was removed from the whole data
set (N) to be used as the testing set (whose bioactivity was predicted) while the remaining
compounds (N-1) were used as the training set (whose relationships between X and Y
variables were learned). The same sampling process was continued until every compound in
the data set was used as the testing set to predict its Y variable (activity).
Evaluating the performance of QSAR model
The predictive performance of the constructed QSAR model was assessed by two
statistical parameters i.e., correlation coefficient (R) representing the predictive performance
and root mean square error (RMSE) representing the predictive error of the model.
Acknowledgments
This project is financially supported by Srinakharinwirot University (grant no.
0
11/2562). Great supports from the annual budget grant (BE 2562-2563) of Mahidol
University are appreciated. We are also indebted to Chulabhorn Research Institute for
recording mass spectra and bioactivity testing.

JOURNAL PRE-PROOF
2
6
References
1
. Ahmad I, Shagufta. Recent developments in steroidal and nonsteroidal aromatase
inhibitors for the chemoprevention of estrogen-dependent breast cancer. Eur J Med
Chem. 2015;102:375–386.
2
3
. Ghosh D, Lo J, Egbuta C. Recent progress in the discovery of next generation inhibitors of
aromatase from the structure−function perspective. J Med Chem. 2016;59:5131–5148.
. Yadav MR, Barmade MA, Tamboli RS, Murumkar PR. Developing steroidal aromatase
inhibitors-an effective armament to win the battle against breast cancer. Eur J Med
Chem. 2015;105:1–38.
4
5
6
. Palmieri C, Patten DK, Januszewski A, Zucchini G, Howell SJ. Breast cancer: current and
future endocrine therapies. Mol Cell Endocrinol. 2014;382:695–723.
. Sainsbury R. The development of endocrine therapy for women with breast cancer. Cancer
Treat Rev. 2013;39:507–517.
. Adhikari N, Amin SA, Saha A, Jha T. Combating breast cancer with non-steroidal
aromatase inhibitors (NSAIs): Understanding the chemico-biological interactions
through comparative SAR/QSAR study. Eur J Med Chem. 2017;137:365–438.
. Bruno RD, Njar VCO. Targeting cytochrome P450 enzymes: A new approach in anti-
cancer drug development. Bioorg Med Chem. 2007;15: 5047–5060.
7
8
9
1
. Lonning PE, Eikesdal HP. Aromatase inhibition 2013: clinical state of the art and
questions that remain to be solved. Endocr Relat Cancer. 2013;20:R183–201
. Sobral AF, Amaral C, Correia-da-Silva G, Teixeira N. Unravelling exemestane: from
biology to clinical prospects. J Steroid Biochem Mol Biol. 2016;163:1–11.
0. Coates A, Keshaviah A, Thürlimann B, Mouridsen H, Mauriac L, Forbes JF, Paridaens R,
Castiglione-Gertsch M, Gelber RD, Colleoni M, Láng I, Del Mastro L, Smith I,
Chirgwin J, Nogaret JM, Pienkowski T, Wardley A Jakobsen, EH, Price KN, Goldhirsch
A. Five years of letrozole compared with tamoxifen as initial adjuvant therapy for
postmenopausal women with endocrine-responsive early breast cancer: update of study
BIG 1–98. J Clin Oncol. 2007;25:486–492.
1
1. Amir W, Seruga B, Niraula S, Carlsson L, Ocaña A. Toxicity of adjuvant endocrine
therapy in postmenopausal breast cancer patients: A systematic review and meta-
analysis. J Natl Cancer Inst. 2011;103:1299–1309.

JOURNAL PRE-PROOF
2
7
1
2. Zhao C, Rakesh KP, Ravidar L, Fang W-Y, Qin H-L. Pharmaceutical and medicinal
significance of sulfur (SVI)-containing motifs for drug discovery: A critical review. Eur
J Med Chem. 2019;162:679–734.
1
1
3. Hu L, Li ZR, Jiang JD, Boykin DW. Novel diaryl or heterocyclic sulfonamides as
antimitotic agents. Anticancer agents Med Chem. 2008;8:739–745.
4. Khan FA, Mushtaq S, Naz S, Farooq U, Zaidi A, Bukhari SM, Rauf A, Mubarak MS.
Sulfonamides as potential bioactive scaffolds. Curr Org Chem. 2018;22:818–830.
1
5. Ammazzalorso A, De Filippis B, Giampietro L, Amoroso R. N-acylsulfonamides:
Synthetic routes and biological potential in medicinal chemistry. Chem Biol Drug Des.
2
017;90:1094–1105.
1
1
6. Supuran CT. Special issue: Sulfonamides. Molecules. 2017;22:1642.
7. Saha T, Hossain MS, Saha D, Lahiri M, Talukdar P. Chloride-mediated apoptosis-
inducing activity of bis(sulfonamide) anionophores. J Am Chem Soc. 2016;138:7558–
7
567.
1
1
8. Su B, Tian R, Darby MV, Brueggemeier RW. Novel sulfonanilide analogs decrease
aromatase activity in breast cancer cells: Synthesis, biological evaluation, and ligand-
based pharmacophore identification. J Med Chem 51 2008;51:1126–1135.
9. Matteo MD, Ammazzalorso A, Andreoli F, Caffa I, Filippis BD, Fantacuzzi M,
Giampietro L, Maccallini C, Nencioni A, Parenti MD, Soncini D, Rio AD. Amoroso R.
Synthesis and biological characterization of 3-(imidazol-1-ylmethyl)piperidine
sulfonamides as aromatase inhibitors, Bioorg Med.Chem Lett. 2016;26:3192–3194.
0. Pingaew R, Prachayasittikul V, Mandi P, Nantasenamat C, Prachayasittikul S, Ruchirawat
S, Prachayasittikul V. Synthesis and molecular docking of 1,2,3-triazole-based
sulfonamides as aromatase inhibitors. Bioorg Med Chem. 2015;23:3472–3480.
1. Ghorab MM, Alsaid MS, Al-Ansary GH, Abdel-Latif GA, Abou El Ella DA. Analogue
based drug design, synthesis, molecular docking and anticancer evaluation of novel
chromene sulfonamide hybrids as aromatase inhibitors and apoptosis enhancers. Eur J
Med Chem. 2016;124:946–958.
2
2

JOURNAL PRE-PROOF
2
8
2
2
2. Pingaew R, Mandi P, Prachayasittikul V, Prachayasittikul S, Ruchirawat S,
Prachayasittikul V. Synthesis, molecular docking, and QSAR study of sulfonamide-
based indoles as aromatase inhibitors. Eur J Med Chem. 2018;143:1604–1615.
3. Ghorab MM, Alsaid MS, Samir N, Abdel-Latif GA, Soliman AM, Ragab FA, Abou El
Ella DA. Aromatase inhibitors and apoptotic inducers: Design, synthesis, anticancer
activity and molecular modeling studies of novel phenothiazine derivatives carrying
sulfonamide moiety as hybrid molecules. Eur J Med Chem. 2017;134:304–315.
2
2
4. Caporuscio F, Rastelli G, Imbriano C, Del Rio A. Structure-based design of potent
aromatase inhibitors by high-throughput docking. J Med Chem. 2011;54:4006–4017.
5. Pingaew R, Prachayasittikul V, Anuwongcharoen N, Prachayasittikul S, Ruchirawat S,
Prachayasittikul V. Synthesis and molecular docking of N,N'-disubstituted thiourea
derivatives as novel aromatase inhibitors. Bioorg Chem. 2018;79:171–178.
2
2
2
2
6. Kao Y-C, Korzekwa KR, Laughton CA, Chen S. Evaluation of the mechanism of
aromatase cytochrome P450 A site-directed mutagenesis study. Eur J Biochem.
2
001;268:243–251.
7. Loge C, Le Borgne M, Marchand P, Robert J-M, Le Baut G, Palzer M, Hartmann RW.
Three-dimensional model of cytochrome P450 human aromatase. J. Enzyme Inhib Med
Chem., 2005;20:581–585.
8. Wood PM, Woo LWL. Thomas MP. Mahon MF. Purohit A. Potter BVL. Aromatase and
dual aromatase-steroid sulfatase inhibitors from the letrozole and vorozole Templates.
ChemMedChem. 2011;6:1423–1438.
9. Stresser DM, Turner SD, McNamara J, Stocker P, Miller VP, Crespi CL, Patten CJ. A
high-throughput screen to identify inhibitors of aromatase (CYP19). Anal Biochem.
2
000;284:427–430.
3
3
0. Mosmann T. Rapid colorimetric assay for cellular growth and survival: application to
proliferation and cytotoxicity assays. J Immunol Methods. 1983;65:55–63.
1. Tominaga H, Ishiyama M, Ohseto F, Sasamoto K, Hamamoto T, Suzuki K, Watanabe M.
A water-soluble tetrazolium salt useful for colorimetric cell viability assay. Anal
Commun. 1999;36:47–50.

JOURNAL PRE-PROOF
2
9
3
2. Carmichael J, DeGraff WG, Gazdar AF, Minna JD, Mitchell JB. Evaluation of a
tetrazolium-based semiautomated colorimetric assay: assessment of radiosensitivity.
Cancer Res. 1987;47:943–946.
3
3
3
3. Vriend G. WHAT IF: A molecular modeling and drug design program. J Mol Graph.
1
990;8:52–56.
4. Sanner MF. Python: a programming language for software integration and development. J
Mol Graph Model. 1999;17:57–61.
5. Morris GM, Huey R, Lindstrom W, Sanner MF, Belew RK, Goodsell DS, Olson AJ.
AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility,
J Comput Chem. 2009;30:2785–2791.
3
3
6. ChemAxon. Marvin Sketch Version 6.0, in, Budapest, Hungary, 2013.
7. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, Scalmani
G, Barone V, Mennucci B, Petersson GA, Nakatsuji H, Caricato M, Li X, Hratchian HP,
Izmaylov AF, Bloino J, Zheng G, Sonnenberg JL, Hada M, Ehara M, Toyota K, Fukuda
R, Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O, Nakai H, Vreven T, Jr. JAM,
Peralta JE, Ogliaro F, Bearpark M, Heyd JJ, Brothers E, Kudin KN, Staroverov VN,
Kobayashi R, Normand J, Raghavachari K, Rendell A, Burant JC, Iyengar SS, Tomasi J,
Cossi M, Rega N, Millam JM, Klene M, Knox JE, Cross JB, Bakken V, Adamo C,
Jaramillo J, Gomperts R, Stratmann RE, Yazyev O, Austin AJ, Cammi R, Pomelli C,
Ochterski JW, Martin RL, Morokuma K, Zakrzewski VG, Voth GA, Salvador P,
Dannenberg JJ, Dapprich S, Daniels AD, Farkas Ö, Foresman JB, Ortiz JV, Cioslowski
J, Fox DJ. Gaussian Inc., Wallingford CT, 2009.
3
3
8. Dallakyan S. PyRx Version 0.8, 2013. http://pyrx.scripps.edu.
9. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE.
UCSF Chimera--a visualization system for exploratory research and analysis. J Comput
Chem. 2004;25:1605–1612.
4
4
0. Delano W. PyMOL Release 0.99, DeLano Scientific LLC, Pala Alto, Ca, 2002.
1. Fährrolfes R, Bietz S, Flachsenberg F, Meyder A, Nittinger E, Otto T, Volkamer A,
Rarey M. ProteinsPlus: a web portal for structure analysis of macromolecules. Nucleic
Acids Res. 2017;45:W337–W343.