Synthesis, in vitro antitumor activity and molecular modeling studies of a new series of benzothiazole Schiff bases

Article abstract of DOI:10.1016/j.cclet.2015.12.033

A new series of benzothiazole Schiff bases 3-29 was synthesized and screened for antitumor activity against cervical cancer (Hela) and kidney fibroblast cancer (COS-7) cell lines. Results indicated that compounds 3, 14, 19, 27 and 28 have promising activity against Hela cell line with IC₅₀ values of 2.41, 3.06, 6.46, 2.22 and 6.25 μmol/L, respectively, in comparison to doxorubicin as a reference antitumor agent (IC₅₀ 2.05 μmol/L). In addition, compound 3 displayed excellent activity against COS-7 cell line with IC₅₀ value of 4.31 μmol/L in comparison to doxorubicin (IC₅₀ 3.04 μmol/L). In the present work, structure based pharmacophore mapping, molecular docking, protein-ligand interaction, fingerprints and binding energy calculations were employed in a virtual screening strategy to identify the interaction between the compounds and the active site of the putative target, EGFR tyrosine kinase. Molecular properties, toxicity, drug-likeness, and drug score profiles of compounds 3, 14, 19, 27, 28 and 29 were also assessed.

Full text of DOI:10.1016/j.cclet.2015.12.033

G Model
CCLET 3541 1–8
Chinese Chemical Letters xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Chinese Chemical Letters
journal homepage: www.elsevier.com/locate/cclet
1
2
Original article
3
4
Synthesis, in vitro antitumor activity and molecular modeling studies
of a new series of benzothiazole Schiff bases
a,b
a,
a
5
6
Q1 Moustafa T. Gabr , Nadia S. El-Gohary *, Eman R. El-Bendary ,
a
b
Mohamed M. El-Kerdawy , Nanting Ni
a
7
8
Department of Medicinal Chemistry, Faculty of Pharmacy, Mansoura University, Mansoura 35516, Egypt
Department of Chemistry, Georgia State University, Atlanta, GA 30303, USA
b
A R T I C L E I N F O
A B S T R A C T
Article history:
A new series of benzothiazole Schiff bases 3–29 was synthesized and screened for antitumor activity
against cervical cancer (Hela) and kidney fibroblast cancer (COS-7) cell lines. Results indicated that
compounds 3, 14, 19, 27 and 28 have promising activity against Hela cell line with IC50 values of 2.41,
Received 22 April 2015
Received in revised form 7 September 2015
Accepted 31 December 2015
Available online xxx
3
.06, 6.46, 2.22 and 6.25
.05 mol/L). In addition, compound 3 displayed excellent activity against COS-7 cell line with IC50 value
mol/L in comparison to doxorubicin as a reference (IC50 3.04 mol/L). In the present work,
mmol/L, respectively, in comparison to doxorubicin as a reference (IC50
2
m
of 4.31
m
m
Keywords:
structure based pharmacophore mapping, molecular docking, protein-ligand interaction, fingerprints
and binding energy calculations were employed in a virtual screening strategy to identify the interaction
between the compounds and the active site of the putative target, EGFR tyrosine kinase. Molecular
properties, toxicity, drug-likeness, and drug score profiles of compounds 3, 14, 19, 27, 28 and 29 were
also assessed.
Benzothiazole
Schiff bases
Antitumor activity
Molecular modeling
In silico studies
ß 2016 Chinese Chemical Society and Institute of Materia Medica, Chinese Academy of Medical Sciences.
Published by Elsevier B.V. All rights reserved.
9
1
0
1. Introduction
most widely studied of these, is the epidermal growth factor 28
receptor (EGFR) small molecule inhibitor erlotinib, which has been 29
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Protein kinases have become one of the most intensively
pursued classes of drug targets with a number of clinical success
stories [1]. Receptor protein tyrosine kinases play an important
role in signal transduction pathways that regulate cell division and
differentiation. Epidermal growth factor receptor tyrosine kinase
(EGFR-TK) and the related human epidermal growth factor
receptor are among the growth factor receptor kinases that have
been identified as important in cancer development [2]. EGFR
dependent aberrant signaling is associated with cancer cell
proliferation, apoptosis, angiogenesis and metastasis [3]. Anilino-
approved for the treatment of non-small cell lung cancer 30
[6,7]. Subsequent research aimed at further exploration of the 31
SAR of this novel template led to the discovery of highly selective 32
compounds that target EGFR. Benzothiazoles act via competing 33
with ATP for binding at the catalytic domain of EGFR-TK [8]. The 34
ATP-binding site (Fig. 1) has the following features: adenine region, 35
which contains two key hydrogen bonds formed by the interaction 36
of N and N of the adenine ring. Many potent inhibitors use one of 37
these hydrogen bonds; sugar pocket, which is a hydrophilic region; 38
hydrophobic regions and channels, which play an important role in 39
inhibitor selectivity and binding affinity; and phosphate binding 40
region, which is largely solvent exposed and can be used for 41
1
6
W
quinazoline-containing compounds, erlotinib (Tarceva ) [3,4] and
W
gefitinib (Iressa ) [5] have been approved for the treatment of
advanced non-small cell lung cancer. They contain the 4-
(substituted amino)pyrimidine pharmacophoric core that binds
to the hinge region of the kinase.
improving inhibitor selectivity [9].
42
Erlotinib binds to the active form of EGFR and exploits the 43
threonine gatekeeper (Thr-790) to penetrate in the EGFR back cleft 44
through this gate. The crystal structure of EGFR in complex with 45
erlotinib (PDB code: 1M17) (http://www.rcsb.org/pdb/home/ 46
Quinazolines have emerged as
a versatile template for
inhibitors of a diverse range of receptor tyrosine kinases. The
1
home.do) showed the binding modes of erlotinib (Fig. 2). N of 47
the quinazoline core accepts a hydrogen bond from the amide 48
nitrogen of Met-769 (numbering used in PDB code: 1M17) in the 49
hinge region. Moreover, the acetylene moiety at the 3-position of 50
*
Corresponding author.
E-mail address: dr.nadiaelgohary@yahoo.com (N.S. El-Gohary).
http://dx.doi.org/10.1016/j.cclet.2015.12.033
1
001-8417/ß 2016 Chinese Chemical Society and Institute of Materia Medica, Chinese Academy of Medical Sciences. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: M.T. Gabr, et al., Synthesis, in vitro antitumor activity and molecular modeling studies of a new series
of benzothiazole Schiff bases, Chin. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.cclet.2015.12.033

G Model
CCLET 3541 1–8
2
M.T. Gabr et al. / Chinese Chemical Letters xxx (2016) xxx–xxx
3
–29. The synthesis information and characterization data of
78
79
80
81
82
83
84
85
86
87
target compounds are deposited in Supporting information. The
antitumor screening of compounds 3–29 against cervical cancer
(
Hela) and kidney fibroblast cancer (COS-7) cell lines was carried
out adopting the MTT assay [22–24] and using doxorubicin as a
reference antitumor agent. In addition, structure based pharma-
cophore mapping, molecular docking, protein–ligand interaction,
fingerprints and binding energy calculations were employed in a
virtual screening strategy to identify the interaction between the
compounds and the active site of the putative target, EGFR-TK.
3. Results and discussion
88
The structures of all the synthesized compounds were
89
90
91
1
13
1
confirmed by IR, H NMR, C NMR and HRMS. H NMR spectra
of compounds 3–29 showed two characteristic singlets at 7.98–
d
Fig. 1. The molecular surface representation of the ATP-binding site that consists of
the adenine region, hydrophobic regions I and II, sugar pocket and phosphate
binding region [9].
9.35 and 11.40–13.00 for CH 55 N and NH, respectively. The
antitumor activity of the synthesized compounds 3–29 against
cervical cancer (Hela) and kidney fibroblast cancer (COS-7) cell
92
93
94
5
5
5
5
5
5
5
5
5
6
6
6
6
6
6
6
6
6
6
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
the phenyl ring is directed into the hydrophobic region I,
significantly improving the compound selectivity, whereas sub-
stituents at 6- and 7-positions of the quinazoline ring extend into
the entrance regions.
Benzothiazole derivatives constitute an important class of
therapeutic agents in medicinal chemistry. Literature survey
revealed that this nucleus is associated with diverse pharmaco-
logical effects, including antitumor [10–17] and antioxidant
[18,19] activities. In addition, benzothiazole Schiff bases were
reported to have antitumor activity [20]. Taking all these findings
lines was evaluated. Results are expressed in IC50
presented in Table 1. As shown in Table 1, compounds 3, 14, 19, 27
and 28 have promising activity against Hela cell line with IC50
(
m
mol/L) and
95
96
97
98
values of 2.41, 3.06, 6.46, 2.22 and 6.25
comparison to doxorubicin as a reference antitumor agent with
IC50 value of 2.05 mol/L. In addition, compound 3 displayed
excellent activity against COS-7 cell line with IC50 value of
4.31 mol/L in comparison to doxorubicin as a reference antitu-
mor agent with IC50 value of 3.04 mol/L.
m
mol/L, respectively, in
99
m
100
101
102
103
m
m
into consideration, we present herein
a
new subfamily of
3.1. Structure-activity relationship (SAR) studies
104
compounds containing the benzothiazole core for evaluation of
their antitumor activity. Our strategy is directed toward designing
a variety of ligands, which are structurally similar to the basic
skeleton, 4-anilinoquinazoline of tinibs (erlotinib and gefitinib)
with diverse chemical properties (Fig. 3). Accordingly, we replaced
quinazoline ring with benzothiazole since both rings are isosteric
with the adenine portion of ATP and can mimic the ATP-
competitive binding regions of EGFR-TK.
Structure activity correlation of compounds 3–29 based on the
tested cell lines, cervical cancer (Hela) and kidney fibroblast cancer
(COS-7) cell lines, is discussed. The presence of 2-(4-hydroxy-2-
methoxybenzylidene)hydrazino moiety at the 2-position of
benzothiazole nucleus greatly enhanced the activity against
cervical cancer (Hela) and kidney fibroblast cancer (COS-7) cell
lines (compound 3). On the other hand, replacement of 4-hydroxy
substituent in compound 3 with 4-methoxy resulted in decreased
activity against both cell lines (compound 4). Furthermore,
changing the positions of the hydroxy and methoxy substituents
on the benzylidene moiety led to decreased activity against both
cell lines (compounds 5 and 6). In addition, the presence of 2-(3-
methylbenzylidene)hydrazino moiety improved the hydrophobic
interaction with the receptor and resulted in considerable activity
against both cell lines (compound 7). On the other hand, the
presence of 2-(4-hydroxy-3-methylbenzylidene)hydrazino moiety
showed decreased activity against both cell lines (compound 9).
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
7
0
2. Experimental
71
72
73
74
75
76
77
A general approach for the synthesis of the designed
compounds is outlined in Scheme 1. The starting compound, 2-
amino-6-fluorobenzothiazole (1) was reacted with hydrazine
hydrate in refluxing ethylene glycol in the presence of hydrochloric
acid to produce the hydrazine derivative 2 [21]. Reaction of
compound 2 with the appropriate aromatic aldehyde in ethanol
under microwave irradiation gave the corresponding Schiff bases
Fig. 2. The 3D and 2D representation of EGFR crystal structure in complex with erlotinib (PDB code: 1M17). (A) The 3D representation of EGFR crystal structure in complex
with erlotinib; (B) the 2D representation of EGFR crystal structure in complex with erlotinib.
Please cite this article in press as: M.T. Gabr, et al., Synthesis, in vitro antitumor activity and molecular modeling studies of a new series
of benzothiazole Schiff bases, Chin. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.cclet.2015.12.033

G Model
CCLET 3541 1–8
M.T. Gabr et al. / Chinese Chemical Letters xxx (2016) xxx–xxx
3
Table 1
In vitro antitumor activity of compounds 3–29 against cervical cancer (Hela) and
kidney fibroblast cancer (COS-7) cell lines.
Comp. no
IC50
(
m
mol/L)
COS-7
4.31
Comp. no
IC50
(
m
mol/L)
COS-7
Hela
Hela
3
5
7
9
2.41
>50
4
>50
>50
>50
>50
>50
>50
>50
>50
45.6
6
22
>50
>50
>50
>50
8
>50
>50
>50
>50
>50
10
36.1
>50
3.06
11
13
15
17
19
21
23
25
27
29
12
16.01
46.5
>50
14
16
>50
>50
>50
>50
>50
36.4
6.46
32.8
>50
>50
2.22
>50
18
>50
>50
>50
>50
>50
>50
16.01
20
>50
>50
>50
>50
22
24
26
15.1
6.25
2.05
Fig. 3. Reported antitumor quinazolines and proposed compounds.
28
9.75
Doxorubicin
3.04
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
Introduction of trifluoromethyl substituent at the para position of
the benzylidene moiety increased the activity against Hela cell line
compared to the other electron-withdrawing substituents (Br, Cl,
Bold values used to point out the active compounds.
NO
2
) (compound 10 versus 11, 12 and 13), whereas the presence of
moiety with the 2-(2-chloro-6-fluorobenzylidene)hydrazino coun- 138
terpart resulted in decreased activity against the same cell line 139
(compound 18).
a chloro substituent at the para position of the benzylidene moiety
increased the activity against COS-7 cell line compared to the other
140
electron-withdrawing substituents (CF
3
, Br, NO
2
) (compound 12
The presence of 2-(naphthalen-2-ylmethylene)hydrazino moi- 141
ety at the 2-position of benzothiazole scaffold improved the 142
activity against Hela cell line compared to the 2-(naphthalen-1- 143
ylmethylene)hydrazino moiety (compound 21 versus 22). On the 144
other hand, the presence of a nitro substituent at 1-position of 145
naphthalene ring of 2-(naphthalen-2-ylmethylene)hydrazino moi- 146
ety led to decreased activity against the same cell line (compound 147
23 versus 21). Replacement of the 2-(naphthalen-2-ylmethylene)- 148
versus 10, 11 and 13). Moreover, the presence of 2-(5-bromo-2-
hydroxybenzylidene)hydrazino moiety enhanced the activity
against Hela cell line (compound 14). Replacement of 5-bromo
substituent with 5-nitro or 5-methyl substituents resulted in
decreased activity against the same cell line (compounds 15 and
16). In addition, the presence of 2-(2,6-dichlorobenzylidene)hy-
drazino moiety contributed to the considerable activity of
compound 17 against Hela cell line. On the other hand,
0
hydrazino moiety with the 2-(1,1 -biphenyl-4-ylmethylene)hy- 149
replacement
of
the
2-(2,6-dichlorobenzylidene)hydrazino
drazino counterpart improved the activity against both cell lines 150
Scheme 1. Reagents and reaction conditions: (a) hydrazine hydrate, hydrochloric acid, ethylene glycol, reflux, 5 h; (b) aromatic aldehydes, ethanol, microwave (20 W), 80 8C,
1
0 min.
Please cite this article in press as: M.T. Gabr, et al., Synthesis, in vitro antitumor activity and molecular modeling studies of a new series
of benzothiazole Schiff bases, Chin. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.cclet.2015.12.033

G Model
CCLET 3541 1–8
4
M.T. Gabr et al. / Chinese Chemical Letters xxx (2016) xxx–xxx
0
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
(compound 19). On the other hand, replacement of the 1,1 -
biphenyl moiety with 4-(morpholin-4-yl)phenyl decreased the
activity against both cell lines (compound 20). Furthermore,
replacement of the 2-(naphthalen-2-ylmethylene)hydrazino moi-
ety with the 2-(quinolin-7-ylmethylene)hydrazino or 2-(quinox-
alin-2-ylmethylene)hydrazino counterparts decreased the activity
against Hela cell line (compound 21 versus 24 and 25). On the other
hand, introduction of 2-(2-oxo-2H-chromen-6-yl)hydrazino coun-
terpart improved the activity against the same cell line (compound
26). Attempts to improve hydrophobic interaction with the target
receptor by adding an extra phenyl ring via replacing the 2-
(naphthalen-2-ylmethylene)hydrazino moiety with the 2-(anthra-
cen-9-ylmethylene)hydrazino counterpart greatly enhanced the
activity against Hela cell line (compound 27). Trials to introduce an
extra site for hydrophobic interaction in compound 27 by the
addition of chloro substituent at the 10-position of the anthracene
ring decreased the activity against the same cell line (compound
28). In addition, the presence of 2-(phenanthren-9-ylmethylene)-
hydrazino moiety decreased the activity against Hela cell line but
improved the activity against COS-7 cell line (compound 29).
However, the most common approaches are based on the 2D
fingerprints, with the similarity between a reference structure and
a database structure.
The most active compounds, 3, 14, 19, 27, 28 and 29 in Standard
Delay Format (SDF) were submitted to the ReverseScreen3D server
[35], the server uses a reverse virtual screening (VS) method called
ReverseScreen3D. It is a 2D fingerprint-based method to select a
ligand template from each unique binding site of each protein with
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
a
target database. The target database contains only the
structurally determined bioactive conformations of known
ligands. The 2D comparison is followed by a 3D structural
comparison to the selected query ligand using a geometric
matching method in order to priotrize each target binding site
in the database. The output in the form of a list of the 2D and 3D
scores for protein tyrosine kinase (cluster no. 14836) is listed in
Table 2. Compounds 3, 14, 19 and 27 with promising antitumor
activity, displayed the highest 3D score values of 0.519, 0.589,
0.631 and 0.524, respectively.
3.2.2. 3D Pharmacophore elucidation
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
3
D Pharmacophore designing methods take into account both
1
71
3.2. Molecular modeling and computational studies
the 3-dimensional structures and binding modes of receptors and
inhibitors in order to identify regions that are favorable for specific
receptor–inhibitor interaction. The description of the receptor–
inhibitor interaction pattern is determined by a correlation
between the characteristic properties of the inhibitors and their
biochemically determined enzymatic activity.
LigandScout, a program that allows the automatic construction
and visualization of 3D pharmacophore from structural data of
protein-ligand complexes, was used in this study to create a
pharmacophore for the mode of action of erlotinib, which
prevents the activation of EGFR kinase [36]. The model (Fig. 4)
was created by automatically overlaying pharmacophoric fea-
tures gathered from the crystal structure of EGFR kinase domain
(PDB ID: 1M17) in complex with erlotinib (http://www.rcsb.org/
pdb/home/home.do).
The investigated pharmacophoric features included hydrogen
bond donors and acceptors as directed vectors, positive and
negative ionizable regions as well as lipophilic areas that are
represented by spheres. According to the pharmacophore gener-
ated by LigandScout, the minimal structural requirements for
antitumor activity consist of a hydrophobic region attached to a
heterocyclic ring that fits into the ATP-binding site, two hydrogen
bond acceptors and one hydrogen bond donor. The 3D alignment of
the pharmacophoric features of each of the synthesized com-
pounds and the 3D pharmacophore of erlotinib binding pose
showed that these compounds possess similar pharmacophoric
features required for activity. The pharmacophore score listed in
Table 2 was calculated for the alignment of compounds 3, 14, 19,
27, 28 and 29 into the 3D pharmacophore of erlotinib binding pose
generated by LigandScout. This score reflects the similarity of the
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
Overactivation of receptor tyrosine kinase (RTK) signaling
pathways is strongly associated with carcinogenesis. Thus, it is
becoming increasingly clear that impaired deactivation of RTKs
may be an oncogenic driver of cancer [25]. On this basis, Computer-
Aided Drug Design (CADD) tools were used to identify the
interaction between the newly synthesized compounds and the
active site of EGFR-TK in comparison to erlotinib as a reference
EGFR-TK inhibitor. Kinase inhibitors should contain the following
features to gain selectivity and potency [26]: A portion that closely
mimics the ATP molecule and one to three hydrogen bonds with
the amino acids located in the hinge region of the target kinases, as
in erlotinib [27], lapatinib [28] and gefitinib [29]. An additional
hydrophobic binding site (allosteric site), which is directly
adjacent to the ATP-binding site, as in imatinib [30] and sorafenib
[31]. However, other mechanism could be achieved through
binding outside the ATP-binding site at an allosteric site [32] and
by forming an irreversible covalent bond to the kinase active site
[33,34]. In the present work, erlotinib binding mode to EGFR-TK
was studied and the design of the newly synthesized compounds is
based on the essential chemical features required for erlotinib
binding affinity to EGFR-TK.
1
1
1
1
1
1
93
94
95
96
97
98
3.2.1. Similarity-based virtual screening
Similarity methods may be the simplest and most widely used
tools for ligand-based virtual screening of chemical databases,
where functionally similar molecules are sought by searching
molecular databases for structurally similar molecules. These
methods can be categorized as 2D and 3D similarity methods.
Table 2
Results of molecular docking analysis of the most active compounds in the EGFR-TK active site.
b
c
c
Comp. no
ReverseScreen3D
Pharmacophore score
Simple fitness kcal/mol
Full fitness kcal/mol
a
a
2D score
3D score
3
0.451
0.425
0.413
0.375
0.387
0.378
0.421
0.519
0.589
0.631
0.524
0.413
0.427
0.516
97.91
116.52
86.49
86.07
ꢀ7.72
ꢀ8.05
ꢀ8.32
ꢀ8.37
ꢀ8.57
ꢀ8.34
ꢀ7.32
ꢀ2009.21
ꢀ2011.98
ꢀ1985.95
ꢀ1977.91
ꢀ1964.48
ꢀ1977.39
ꢀ1908.77
14
19
27
28
29
79.97
81.86
104.29
Erlotinib
a
2
D and 3D scores were calculated using ReverseScreen 3D software [35].
b
c
Pharmacophore scores were calculated using LigandScout software [36].
Simple fitness and full fitness (kcal/mol) were calculated using SwissDock software [37].
Please cite this article in press as: M.T. Gabr, et al., Synthesis, in vitro antitumor activity and molecular modeling studies of a new series
of benzothiazole Schiff bases, Chin. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.cclet.2015.12.033

G Model
CCLET 3541 1–8
M.T. Gabr et al. / Chinese Chemical Letters xxx (2016) xxx–xxx
5
Fig. 4. (A) The crystal structure of EGFR kinase domain (PDB ID: 1M17) in complex with erlotinib was obtained from the protein data bank (PDB); (B) LigandScout 3D proposed
docking pose of erlotinib in the ATP binding site of EGFR kinase domain; (C) The 3D pharmacophore of erlotinib (in ball and stick representation); The pharmacophore color
coding is red for hydrogen bond acceptor, yellow for hydrophobic regions and green for hydrogen bond donors; (D) The final 3D pharmacophore model for EGFR kinase
domain (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.).
2
2
2
2
49
50
51
52
compounds to the reference pharmacophore. Compound 14
showed the highest pharmacophore score value of 116.52. The
detailed 2D mapping of the pharmacophore model with the
structural features of compound 14 is depicted in Fig. 5.
EGFR-TK active site as illustrated in Fig. 7a. The important residues 277
in the hydrophobic regions that interact with the hit compounds 278
are (Phe-699, Ala-719, Val-702, Lys-721 and Asp-831). The docking 279
results showed that hydrogen bonding interactions of the newly 280
synthesized compounds with EGFR-TK binding site greatly 281
enhanced the affinity toward the enzyme. Analysis of binding 282
mode revealed that compounds 3 and 14 are involved in the 283
formation of two hydrogen bonds with the EGFR-TK active site, 284
which indicates a correlation between the hydrogen bonding 285
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
3.2.3. Docking
The crystal structure of EGFR kinase domain (PDB ID: 1M17) in
complex with an irreversible inhibitor was obtained from the
protein data bank (PDB) (http://www.rcsb.org/pdb/home/home.
do). Docking simulation was done by using the SwissDock software
[37]. All the conformers were virtually docked at the defined cavity
of the receptor. The docking scores of the best conformers for each
ligand are listed (Table 2). The ligand forming the most stable drug
receptor complex is the one having the lowest docking score value.
The six active compounds are evaluated using two scoring
functions: simple fitness and full fitness. Simple fitness is a fast
and efficient method to evaluate the individual binding modes but
neglects the solvent effect and is used to drive the search.
Simultaneously, clusters of binding modes are evaluated by the
more selective yet slower, full fitness method, which accounts for
the solvation free energy. Compound 14 showed a relatively low
full fitness score value of ꢀ2011.98 kcal/mol. 3D Interactions of
compound 14 with the binding site of EGFR-TK are shown in Fig. 6.
interactions and the antitumor activity (Fig. 7b and c).
286
3.3. Molecular properties and drug-likeness
287
Drug-likeness is a complex balance of various structural 288
features that determine whether a particular molecule is similar 289
to the known drugs or not. It generally means ‘‘molecules that 290
contain functional groups and/or have physical properties consis- 291
tent with most of the known drugs’’. Hydrophobicity, molecular 292
size, flexibility and presence of various pharmacophoric features 293
are the main physical properties that influence the behavior of 294
molecules in a living organism. Computational chemists have a 295
wide array of tools and approaches available for the assessment of 296
molecular diversity. Diversity analysis has been shown to be an 297
important ingredient in designing drugs. So, computational 298
sensitivity analysis and structural analysis have been used to 299
study the drug-likeness of the candidate drugs. As good 300
271
272
273
274
275
276
3.2.4. Analysis of the binding mode
Analysis of the docking results (Table 2) using the Lead IT
software revealed that the main interaction forces of the candidate
compounds with the EGFR-TK active site are hydrophobic in nature
[38]. Enhancement of the antitumor activity of compound 19 may
be explained by the improved hydrophobic interactions with
Fig. 6. 3D Interactions of compound 14 with EGFR-TK binding site. The atoms are
colored as following: red for oxygen atoms, blue for nitrogen atoms, yellow for
sulfur atoms, white for hydrogen atoms, cyan for carbon atoms and green for
chlorine atoms (For interpretation of the references to color in this figure legend, the
reader is referred to the web version of this article.).
Fig. 5. The 2D representation of the structural features of compound 14 that can be
aligned with the pharmacophore hypothesis. HBA, hydrogen bond acceptor; H,
hydrophobic center; HBD, hydrogen bond donor.
Please cite this article in press as: M.T. Gabr, et al., Synthesis, in vitro antitumor activity and molecular modeling studies of a new series
of benzothiazole Schiff bases, Chin. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.cclet.2015.12.033

G Model
CCLET 3541 1–8
6
M.T. Gabr et al. / Chinese Chemical Letters xxx (2016) xxx–xxx
Fig. 7. 2D Interactions of compounds 19 (a), 3 (b) and 14 (c) with EGFR-TK binding site. Dashed lines represent hydrogen bonds.
3
3
3
3
3
01
02
03
04
05
bioavailability can be achieved with an appropriate balance
between solubility and partitioning properties, the six active
compounds, 3, 14, 19, 27, 28 and 29 were analyzed for the
prediction of Lipinski’s rule of five [39] as well as other properties
(Tables 3 and 4).
Molecular properties (TPSA, nrotb, miLogP, OH–NH interaction,
O–N interaction, molecular weight and number of violations from
Lipinski’s rule) of the six active compounds, 3, 14, 19, 27, 28 and 29
were calculated using the molinspiration software (Table 3).
Topological polar surface area (TPSA) and lipophilicity (logP)
values are two important properties for the prediction of oral
bioavailability of drug molecules [41–44]. TPSA is calculated based
on the methodology published by Ertl et al. [44] as the surface
areas that are occupied by oxygen and nitrogen atoms and by
hydrogen atoms attached to them. Thus, it is closely related to the
hydrogen bonding potential of a compound [41–44]. TPSA has been
shown to be a very good descriptor characterizing drug absorption,
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
306
307
308
309
310
311
312
313
314
3.3.1. Molinspiration calculations
As a part of our study; the compliance of compounds to the
Lipinski’s rule of five was evaluated [39]. This simple rule is based
on the observation that most drugs that passed phase 2 clinical
trials have molecular weight of 500 or less, clogP values lower than
5, hydrogen bond donors fewer than 5 and hydrogen bond
acceptors fewer than 10. In addition, topological polar surface area
(TPSA) and number of rotatable bonds have also been linked to
drug bioavailability [40].
including intestinal absorption, bioavailability and blood-brain
2
˚
barrier penetration. Molecules with TPSA values of 140 A or more
are expected to exhibit poor intestinal absorption [40]. Results
Table 3
Topological polar surface area, number of rotatable bonds and calculated Lipinski’s rule of five for the most active compounds.
a
Comp. no
Molecular properties
b
c
d
TPSA
Nrotb
miLogP
OH–NH interact
O–N interact
M. wt.
No. of violations
3
66.745
57.511
37.283
37.283
37.283
37.283
4
3
4
3
3
3
3.432
4.652
5.722
6.222
6.828
6.222
2
2
1
1
1
1
5
4
3
3
3
3
317.345
366.215
347.418
371.44
0
0
1
1
1
1
14
19
27
28
29
405.885
371.44
a
Molecular properties (TPSA, nrotb, miLogP, OH–NH interaction, O–N interaction, molecular weight and number of violations from Lipinski’s rule) were calculated using
molinspiration software [41].
b
TPSA: topological polar surface area.
c
Nrotb: number of rotatable bonds.
d
miLogP: the parameter of lipophilicity.
Table 4
Toxicity risks, drug-likeness and drug score for the most active compounds.
a
a
a
Comp. no
Toxicity risks
Mutagenicity
Drug-likeness
Drug score
Tumorogenicity
Irritancy
Reproductive effects
3
+
+
+
+
+
+
+
+
+
1.05
ꢀ0.11
1.30
0.45
0.29
0.25
0.04
0.06
0.11
1
1
2
2
2
4
9
7
8
9
+
+
+
+
+
+
+++
++
++
+++
+++
+++
+++
+++
+
ꢀ0.51
1.30
1.41
+
: low risk; ++: moderate risk; +++: high risk.
a
Toxicity risks (mutagenicity, tumorogenicity, irritancy and reproductive effects) and physicochemical properties (drug-likeness and drug scores) were calculated by the
methodology developed by Osiris software [41].
Please cite this article in press as: M.T. Gabr, et al., Synthesis, in vitro antitumor activity and molecular modeling studies of a new series
of benzothiazole Schiff bases, Chin. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.cclet.2015.12.033

G Model
CCLET 3541 1–8
M.T. Gabr et al. / Chinese Chemical Letters xxx (2016) xxx–xxx
7
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
shown in Table 3 indicated that all the analyzed compounds have
Appendix A. Supplementary data
391
2
˚
TPSA values < 140 A ; thus, they are expected to have good
intestinal absorption. Molecules with more than 10 rotatable
bonds may have problems with bioavailability [40]. All the
analyzed compounds have 3 or 4 rotatable bonds and they might
not have problems with bioavailability (Table 3). MiLogP is
calculated using the methodology developed by Molinspiration as
a sum of fragment-based contributions and correction factors
(http://www.molinspiration.com) (Table 3). It has been shown
that for a compound to have a reasonable probability of being well
absorbed, miLogP value must be in the range of ꢀ0.4 to +5.6
[40]. On this basis, compounds 3 and 14 were found to have
miLogP values within the acceptable criteria. It is worth
mentioning that all the analyzed compounds have one or zero
violation of Lipinski’s rule and they are expected to have
reasonable oral absorption.
Supplementary data associated with this article can be found, in 392
the online version, at http://dx.doi.org/10.1016/j.cclet.2015.12. 393
033.
394
References
395
[
[
1] Q. Liu, Y. Sabnis, Z. Zhao, et al., Developing irreversible inhibitors of the protein 396
kinase cysteinome, Chem. Biol. 21 (2013) 146–159.
2] L.V. Peng-Cheng, C.F. Zhou, J. Chen, et al., Design, synthesis and biological 398
evaluation of thiazolidinone derivatives as potential EGFR and HER-2 inhibitors, 399
Bioorg. Med. Chem. Lett. 18 (2010) 314–319.
397
400
[
3] M. Reck, N.V. Zandwijk, C. Gridelli, et al., Erlotinib in advanced non-small cell lung 401
cancer: efficacy and safety findings of the global phase IV Tarceva lung cancer 402
survival treatment study, J. Thorac. Oncol. 5 (2010) 1616–1622.
[4] J. Smith, Erlotinib: small-molecule targeted therapy in the treatment of non small- 404
cell lung cancer, Clin. Ther. 27 (2005) 1513–1534.
5] K. Tamura, M. Fukuoka, Gefitinib in non-small cell lung cancer, Expert Opin. 406
Pharmacother. 6 (2005) 985–993.
403
405
407
[
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
3.3.2. Osiris calculations
[6] P. Ballard, R.H. Bradbury, C.S. Harris, et al., Inhibitors of epidermal growth factor 408
receptor tyrosine kinase: optimisation of potency and in vivo pharmacokinetics, 409
Toxicity risks (mutagenicity, tumorogenicity, irritancy and
reproductive effects) and physicochemical properties (drug-
likeness and drug score) of the synthesized compounds were
calculated by the methodology developed by Osiris [41]. The
toxicity risk predictor locates fragments within a molecule that
indicate a potential toxicity risk. Toxicity risk alerts are an
indication that the drawn structure may be harmful concerning the
risk category specified. From the data presented in Table 4, it is
obvious that compounds 3, 14 and 19 are expected to be non-
mutagenic and non-tumorigenic. Compounds 3, 14, 19 and 29 are
expected to be non-irritant. In addition, all analyzed compounds
were found to have non-reproductive effects.
Drug-likeness is defined as a complex balance of various
molecular properties and structural features that indicates
whether a particular molecule is similar to the known drugs or
not [45]. Osiris program was used in calculating the fragment-
based drug-likeness of the synthesized compounds, where a
positive value indicates that the designed molecule contains
fragments that are frequently present in commercial drugs. Results
shown in Table 4 indicated that compounds 3, 19, 28 and 29 have
positive drug-likeness values. The drug score combines drug-
likeness, miLogP, solubility, molecular weight and toxicity risks in
one handy value that may be used to judge the compound’s overall
potential to qualify for a drug [41]. A value of 0.5 or more makes the
compound a promising lead for future development of safe and
efficient drugs. The overall drug score values for the synthesized
compounds were calculated (Table 4).
Bioorg. Med. Chem. Lett. 16 (2006) 4908–4912.
410
[
[
[
7] M. Ranson, Epidermal growth factor receptor tyrosine kinase inhibitors, Br. J. 411
Cancer 90 (2004) 2250–2255.
8] M.N. Noolvi, H.M. Patel, M. Kaur, Benzothiazoles: search for anticancer agents, 413
Eur. J. Med. Chem. 54 (2012) 447–462.
9] D. Fabbro, S. Ruetz, E. Buchdunger, et al., Protein kinases as targets for anticancer 415
agents: from inhibitors to useful drugs, Pharmacol. Ther. 93 (2002) 79–98.
412
414
416
[
10] X.H. Shi, Z. Wang, Y. Xia, et al., Synthesis and biological evaluation of novel 417
benzothiazole-2-thiol derivatives as potential anticancer agents, Molecules 17 418
(2012) 3933–3944.
419
[11] H.S. Elzahabi, Synthesis, characterization of some benzazoles bearing pyridine 420
moiety: search for novel anticancer agents, Eur. J. Med. Chem. 46 (2011) 4025– 421
4034.
422
[
12] S. Saeed, N. Rashid, P.G. Jones, M. Ali, R. Hussain, Synthesis, characterization and 423
biological evaluation of some thiourea derivatives bearing benzothiazole moiety 424
as potential antimicrobial and anticancer agents, Eur. J. Med. Chem. 45 (2010) 425
1323–1331.
426
[13] W.P. Hu, Y.K. Chen, C.C. Liao, et al., Synthesis, and biological evaluation of 2-(4- 427
aminophenyl)benzothiazole derivatives as photosensitizing agents, Bioorg. Med. 428
Chem. 18 (2010) 6197–6207.
14] Y.A. Al-Soud, H.H. Al-Sa’doni, B. Saeed, et al., Synthesis and in vitro antiprolifera- 430
tive activity of new benzothiazole derivatives, ARKIVOC xv (2008) 225–238.
429
[
[
431
15] G.M. Catriona, W. Geoffrey, C.P. Jean, et al., Antitumor benzothiazoles. 26. 2- 432
(3,4-dimethoxyphenyl)-5-fluorobenzothiazole (GW 610 NSC 721648), a sim- 433
ple fluorinated 2-arylbenzothiazole, shows potent and selective inhibitory 434
activity against lung, colon and breast cancer cell lines, J. Med. Chem. 49 435
(
2006) 179–185.
436
[16] E. Brantley, S. Antony, G. Kohlhagen, et al., Anti-tumor drug candidate 2-(4- 437
amino-3-methylphenyl)-5-fluorobenzothiazole induces single-strand breaks and 438
DNA-protein cross-links in sensitive MCF-7 breast cancer cells, Cancer Che- 439
mother. Pharmacol. 58 (2006) 62–72.
440
[17] C.J. Lion, C.S. Matthews, G. Wells, et al., Antitumour properties of fluorinated 441
benzothiazole-substituted hydroxycyclohexa-2, 5-dienones (‘quinols’), Bioorg. 442
Med. Chem. Lett. 16 (2006) 5005–5008.
443
[
18] N. Karali, O. G u¨ zel, N. Ozsoy, S. Ozbey, A. Salman, Synthesis of new spiroindo- 444
linones incorporating a benzothiazole moiety as antioxidant agents, Eur. J. Med. 445
Chem. 45 (2010) 1068–1077.
446
3
74
4. Conclusion
[
19] D. Cressier, C. Prouillac, P. Hernandez, et al., Synthesis, antioxidant properties and 447
radioprotective effects of new benzothiazoles and thiadiazoles, Bioorg. Med. 448
375
376
377
378
379
380
381
382
383
384
385
Compounds 3, 14, 19, 27 and 28 are the most active antitumor
agents in this study against cervical cancer (Hela) cell line. In
addition, compounds 3 and 29 are the most active members
against kidney fibroblast cancer (COS-7) cell line. From these
results we can conclude that the actual antitumor activity of
compounds 3, 19, 28 and 29 is correlated to the drug-likeness
prediction results as these compounds have positive drug-
likeness values. These encouraging preliminary results of
biological screening of the newly synthesized compounds could
offer a good framework toward the discovery of new potent
antitumor agents.
Chem. 17 (2009) 5275–5284.
449
[20] S.E. Etaiw, D.M. Abd El-Aziz, E.H. Abd El-Zaher, E.A. Ali, Synthesis, spectral, 450
antimicrobial and antitumor assessment of Schiff base derived from 2-amino- 451
benzothiazole and its transition metal complexes, Spectrochim. Acta A: Mol. 452
Biomol. Spectrosc. 79 (2011) 1331–1337.
[21] P. Yadav, D. Chauhan, N.K. Sharma, S. Singhal, 2-Substituted hydrazino-6-fluoro- 454
,3-benzothiazole: synthesis and characterization of new novel antimicrobial 455
agents, Int. J. ChemTech Res. 2 (2010) 1209–1213.
453
1
456
[
22] T. Mosmann, Rapid colorimetric assay for cellular growth and survival: appli- 457
cation to proliferation and cytotoxicity assays, J. Immunol. Methods 65 (1983) 458
5
5–63.
459
[23] F. Denizot, R. Lang, Rapid colorimetric assay for cell growth and survival. Mod- 460
ifications to the tetrazolium dye procedure giving improved sensitivity and 461
reliability, J. Immunol. Methods 89 (1986) 271–277.
24] D. Gerlier, T. Thomasset, Use of MTT colorimetric assay to measure cell activation, 463
J. Immunol. Methods 94 (1986) 57–63.
25] G.B. Kristi, S. Thomas, S. Herald, Defective down regulation of receptor tyrosine 465
kinases in cancer, Eur. Mol. Biol. Organ. 23 (2004) 2707–2712.
[26] J. Zhang, P.L. Yang, N.S. Gray, Targeting cancer with small molecule kinase 467
inhibitors, Nat. Rev. Cancer 9 (2009) 28–39.
462
464
466
468
[
[
3
86
Acknowledgments
387
388
389
390
The authors thank Professor Binghe Wang at Georgia State
University, USA, for providing all required facilities and carrying
out the spectral and elemental analyses as well as MTT assay for
antitumor screening.
[
27] B. Liu, B. Bernard, J.H. Wu, Impact of EGFR point mutations on the sensitivity to 469
gefitinib: insights from comparative structural analyses and molecular dynamics 470
simulations, Proteins 65 (2006) 331–346.
471
Please cite this article in press as: M.T. Gabr, et al., Synthesis, in vitro antitumor activity and molecular modeling studies of a new series
of benzothiazole Schiff bases, Chin. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.cclet.2015.12.033

G Model
CCLET 3541 1–8
8
M.T. Gabr et al. / Chinese Chemical Letters xxx (2016) xxx–xxx
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97 Q2
98
[28] Z.A. Wainberg, A. Anghel, A.J. Desai, et al., Lapatinib, a dual EGFR and HER2 kinase
inhibitor, selectively inhibits HER2-amplified human gastric cancer cells and is
synergistic with trastuzumab in vitro and in vivo, Clin. Cancer Res. 16 (2010)
1509–1519.
[29] C. Yun, T.J. Boggon, Y. Li, et al., Structures of lung cancer-derived EGFR mutants
and inhibitor complexes: mechanism of activation and insights into differential
inhibitor sensitivity, Cancer Cell 11 (2007) 217–227.
[30] M. Cherry, D.H. Williams, Recent kinase inhibitor X-ray structures: mechanisms
of inhibition and selectivity insights, Curr. Med. Chem. 11 (2004) 663–673.
[31] S. Whittaker, R. Kirk, R. Hayward, et al., Gatekeeper mutations mediate resistance
to BRAF-targeted therapies, Sci. Transl. Med. 2 (2010) 35–41.
[32] E. Weisberg, P.W. Manley, S.W. Cowan-Jacob, A. Hochhaus, J.D. Griffin, Second
generation inhibitors of BCR-ABL for the treatment of imatinib resistant chronic
myeloid leukemia, Nat. Rev. Cancer 7 (2007) 345–356.
[33] S. Sridhar, L. Seymour, F.A. Shepherd, Inhibitors of epidermal growth factor
receptors: a review of clinical research with a focus on non small-cell lung
cancer, Lancet Oncol. 4 (2003) 397–406.
[34] F.A. Sharma, R. Sharma, T. Tyagi, Receptor tyrosine kinase inhibitors as potent
weapons in war against cancers, Curr. Pharm. Des. 15 (2009) 758–776.
[35] S.L. Kinnings, R.M. Jackson, ReverseScreen3D: a structure-based ligand matching
method to identify protein targets, J. Chem. Inf. Model. 51 (2011) 624–634.
[36] G. Wolber, T. Langer, LigandScout: 3D pharmacophores derived from protein-
bound ligands and their use as virtual screening filters, J. Chem. Inf. Comput. Sci.
45 (2005) 160–169.
[38] P. Maass, T. Schulz-Gasch, M. Stahl, M. Rarey, ReCore: a fast and versatile method
for scaffold hopping based on small molecule crystal structure conformations, J.
Chem. Inf. Model. 47 (2007) 390–399.
[39] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental and computa-
tional approaches to estimate solubility and permeability in drug discovery and
development settings, Adv. Drug Deliv. Rev. 46 (2001) 3–26.
[40] D.F. Veber, S.R. Johnson, H.Y. Cheng, et al., Molecular properties that influence the
oral bioavailability of drug candidates, J. Med. Chem. 45 (2002) 2615–2623.
[41] A. Jarrahpour, J. Fathi, M. Mimouni, et al., Petra, Osiris and molinspiration (POM)
together as a successful support in drug design: antibacterial activity and bio-
pharmaceutical characterization of some azo Schiff bases, Med. Chem. Res. 19
(2011) 1–7.
[42] A. Parvez, M. Jyotsna, M.H. Youssoufi, T. Ben Hadda, Theoretical calculations and
experimental verification of the antibacterial potential of some monocyclic beta-
lactames containing two synergetic buried antibacterial pharmacophore sites,
Phosphorus Sulfur Silicon Relat. Elem. 7 (2010) 1500–1510.
[43] A. Parvez, J. Meshram, V. Tiwari, et al., Pharmacophores modeling in terms of
prediction of theoretical physicochemical properties and verification by experi-
mental correlations of novel coumarin derivatives produced via Betti’s protocol,
Eur. J. Med. Chem. 45 (2010) 4370–4378.
[44] P. Ertl, B. Rohde, P. Selzer, Fast calculation of molecular polar surface area (PSA) as
a sum of fragment-based contributions and its application to the prediction of
drug transport properties, J. Med. Chem. 43 (2000) 3714–3717.
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
[45] O. Ursu, A. Rayan, A. Goldblum, T. Oprea, Understanding drug-likeness, WIREs
Comput. Mol. Sci. 1 (2011) 760–781.
[37] A. Grosdidier, V. Zoete, O. Michielin, SwissDock, a protein-small molecule docking
web service based on EADock DSS, J. Comput. Chem. Nucleic Acids Res. 39 (2011)
270–277.
Please cite this article in press as: M.T. Gabr, et al., Synthesis, in vitro antitumor activity and molecular modeling studies of a new series
of benzothiazole Schiff bases, Chin. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.cclet.2015.12.033