Quantitative structure activity relationship studies of novel hydrazone derivatives as α-amylase inhibitors with index of ideality of correlation

Article abstract of DOI:10.1080/07391102.2020.1863861

The present manuscript describes the synthesis, α-amylase inhibition, in silico studies and in-depth quantitative structure–activity relationship (QSAR) of a library of aroyl hydrazones based on benzothiazole skeleton. All the compounds of the developed l

Full text of DOI:10.1080/07391102.2020.1863861

Journal of Biomolecular Structure and Dynamics
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/tbsd20
Quantitative structure activity relationship studies
of novel hydrazone derivatives as α-amylase
inhibitors with index of ideality of correlation
Meenakshi Duhan , Jayant Sindhu , Parvin Kumar , Meena Devi , Rahul
Singh , Ramesh Kumar , Sohan Lal , Ashwani Kumar , Sudhir Kumar & Khalid
Hussain
To cite this article: Meenakshi Duhan , Jayant Sindhu , Parvin Kumar , Meena Devi , Rahul
Singh , Ramesh Kumar , Sohan Lal , Ashwani Kumar , Sudhir Kumar & Khalid Hussain (2020):
Quantitative structure activity relationship studies of novel hydrazone derivatives as α-amylase
inhibitors with index of ideality of correlation, Journal of Biomolecular Structure and Dynamics, DOI:
10.1080/07391102.2020.1863861
To link to this article: https://doi.org/10.1080/07391102.2020.1863861
Published online: 24 Dec 2020.
Article views: 55
View supplementary material
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=tbsd20

JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS
https://doi.org/10.1080/07391102.2020.1863861
Quantitative structure activity relationship studies of novel hydrazone
derivatives as a-amylase inhibitors with index of ideality of correlation
Meenakshi Duhana, Jayant Sindhub, Parvin Kumara , Meena Devia, Rahul Singha, Ramesh Kumara, Sohan Lala,
Ashwani Kumarc, Sudhir Kumard and Khalid Hussaine
b
aDepartment of Chemistry, Kurukshetra University, Kurukshetra, India; Department of Chemistry, COBS&H, CCS Haryana Agricultural
c
University, Hisar, India; Department of Pharmaceutical Sciences, Guru Jambeshwar University of Science and Technology, Hisar, India;
e
dDepartment of MBB&B, COBS&H, CCS Haryana Agricultural University, Hisar, India; Department of Applied Sciences and Humanities,
Mewat Engineering College, Nuh, India
Communicated by Ramaswamy H. Sarma
ABSTRACT
ARTICLE HISTORY
Received 12 August 2020
Accepted 9 December 2020
The present manuscript describes the synthesis, a-amylase inhibition, in silico studies and in-depth
quantitative structure–activity relationship (QSAR) of a library of aroyl hydrazones based on benzothia-
zole skeleton. All the compounds of the developed library are characterized by various spectral techni-
ques. a-Amylase inhibitory potential of all compounds has been explored, where compound 7n
exhibits remarkable a-amylase inhibition of 87.5% at 50 mg/mL. Robust QSAR models are made by
using the balance of correlation method in CORAL software. The chemical structures at different con-
centration with optimal descriptors are represented by SMILES. A data set of 66 SMILES of 22 hydra-
zones at three distinct concentrations are prepared. The significance of the index of ideality of
correlation (IIC) with applicability domain (AD) is also studied at depth. A QSAR model with best
R2validation ¼ 0.8587 for split 1 is considered as a leading model. The outliers and promoters of increase
and decrease of endpoint are also extracted. The binding modes of the most active compound, that
is, 7n in the active site of Aspergillus oryzae a-amylase (PDB ID: 7TAA) are also explored by in silico
molecular docking studies. Compound 7n displays high resemblance in binding mode and pose with
the standard drug acarbose. Molecular dynamics simulations performed on protein–ligand complex for
100ns, the protein gets stabilised after 20 ns and remained below 2 Å for the remaining simulation.
Moreover, the deviation observed in RMSF during simulation for each amino acid residue with respect
to Ca carbon atom is insignificant.
KEYWORDS
Benzothiazole; aroyl
hydrazone; a-amylase
inhibition; molecular
docking; QSAR; IIC
Abbreviations: IIC: index of ideality of correlation; CW: correlation weight; OECD: Organization of
Economic Co-operation and Development; QSAR: quantitative structure–activity relationship; CORAL:
CORrelation And Logic; AD: applicability domain; MD: molecular dynamics
1. Introduction
main function is to hydrolyze the a-D-(1,4)-glycosidic linkage
in starch (Brayer et al., 1995; Shi et al., 2018; Souza &
Diabetes and obesity are a cluster of metabolic disorders
related to lifestyle and characterised by high blood glucose
over a prolonged time. The increasing occurrence of these
two disorders accelerated the discovery of new drugs.
a-Amylase (EC 3.2.1.1) is an endoamylase which mainly
occurs in plants, microorganism and higher organisms and
~
Magalhaes, 2010) and retaining a-anomeric configuration in
the products. The over-expression of a-amylase leads to
hyperglycaemia which results in the development of dia-
betes mellitus. This feature established a-amylase as a well-
known molecular target for type
2 diabetes mellitus.
Marketed drugs prescribed to treat type-II diabetes mellitus
belongs to 13th family of glycoside hydrolases (GH13). Its are associated with numerous side effects such as diarrhoea,
CONTACT Parvin Kumar
parvinjangra@gmail.com; parvinchem@kuk.ac.in
Department of Chemistry, Kurukshetra University, Kurukshetra, India
Supplemental data for this article can be accessed online at https://doi.org/10.1080/07391102.2020.1863861.
ß 2020 Informa UK Limited, trading as Taylor & Francis Group

2
M. DUHAN ET AL.
abdominal discomfort and flatulence. Therefore, there is a a-amylase inhibitors (Figure 1). Taha et al. recently reported
strong need to develop new drugs with lesser side effect hydrazones as a-amylase and a-glucosidase inhibitors with a
and better selectivity.
In recent years, N and S containing heterocyclic com-
pounds have attracted synthetic medicinal chemist owing to
wide range of IC50 value (Taha et al., 2015, 2017, 2019). The
presence of a thiazole, benzofuran and benzothiazole on one
side of hydrazone played an important role in the inhibition
of target. The effect of different substituents on the phenyl
ring of carbaldehyde was also investigated by the authors.
Using these tactics, we have designed and prepared a series
of new derivatives of benzothiazole linked with hydrazone
via an ether linkage. Further, we focused on optimizing the
aryl/heteroaryl region of the imine group and phenyl ring of
the ether linkage.
their
numerous
pharmacological
properties.
1,3-
Benzothiazole is one of the important N- and S-containing
heterocyclic scaffold which has gained the attention due to
its use in a wide range of industrial (Kaur et al., 2018;
Sharma et al., 2018) and biological applications (Bhutani
et al., 2018; Gollapalli et al., 2019; Kumar et al., 2017c; Mishra
et al., 2019; Murtuja et al., 2018). Similarly, aroyl hydrazone
has acquired a leading position in the pharmacophoric
framework of biologically active compounds (Angelova et al.,
2.2. Synthesis of a-amylase inhibitors
ꢀ
ꢀ
2016; Cardoso et al., 2017; Hernandez-Vazquez et al., 2016;
Kumar et al., 2017b; Thota et al., 2018).
The synthetic sequence of (E)-N0-(4-(4-(benzo[d]thiazol-2-yl)ar-
yloxy)benzylidene)-(aryl)hydrazide (7a–7v) is presented in
Scheme 2. The key intermediate, 4-(4-(benzo[d]thiazol-2-yl)ar-
yloxy)benzaldehyde (5a–5b) was prepared in two steps from
Quantitative structure–activity relationship (QSAR) is an
important technique for predicting the biological effects of
molecules based on mathematical and statistical relationship
(Krallinger et al., 2017; Le et al., 2012). CORAL software
(http://www.insilico.eu/CORAL) is a user-friendly freely avail-
able software and has been used to develop QSAR models
for various endpoints (Kumar & Chauhan, 2017; Kumar et al.,
2019c; Toropova & Toropov et al., 2019a). The optimal
descriptors, so-called correlation weights, are calculated from
SMILES using the Monte Carlo optimization method where
correlation coefficient between the endpoint and optimal
descriptor are used to compute the target function.
Literature survey reveals that the index of ideality of correl-
ation (IIC) has been applied to validate the developed QSAR
models (Toropov & Toropova, 2019; Toropov et al., 2019;
Toropova & Toropov, 2019b, 2019c). The IIC evaluates the
predictive capability of the QSAR model, which is not only
concerned with the correlation coefficient but to the residual
values of the endpoint and dot alignment of the image rela-
tive to the diagonal (Toropova & Toropov, 2019b, 2019c).
To the best of our knowledge, there is no scientific report
on benzothiazole clubbed hydrazone as a-amylase inhibitors.
Continuing our ongoing research on N/S containing hetero-
cyclic compounds (Bhatia et al., 2019; Kumar et al., 2017a,
2018) and in search of novel therapeutic agents, we, herein
report the synthesis and a-amylase inhibitory potential of
novel molecular hybrids of benzothiazole clubbed hydrazone
with QSAR study. The objectives of the present study are: (i)
Multistep synthesis of benzothiazole clubbed with hydrazone
via ether linkage (7a–7v) and in vitro a-amylase inhibition at
three different concentrations (50, 25, 12.5 mg/mL); (ii)
Building up and estimation of QSAR models using Monte
Carlo approach with IIC; (iii) Extraction of structural features
responsible for the increase and decrease of a-amylase inhib-
ition and (iv) Docking studies to explore the probable bind-
ing modes.
the
commercially
available
2-aminothiophenol
(1)
(1.00 mmol), p-hydroxybenzaldehyde (2a) (1.00 mmol)/vanillin
(2b) (1.00 mmol) and 4-fluorobenzaldehyde (4) (1.00 mmol).
The SiO2 supported HNO3 catalysed condensation of 2-ami-
nothiophenol (1) (1.00 mmol) with p-hydroxybenzaldehyde
(2a) (1.00 mmol)/vanillin (2b) (1.00 mmol) under the solvent-
free condition at room temperature resulted in the formation
of 2-arylbenzothiazole (3a–3b) (Kumar et al., 2015). Further,
compound 3 was arylated with p-fluorobenzaldehyde (4)
(1.00 mmol) using K2CO3 (1.20 mmol) in dimethyl sulfoxide
(DMSO) at 120 ꢀC for 4–5 h. The reaction resulted in the for-
mation of 4-(4-(benzo[d]thiazol-2-yl)aryloxy) benzaldehyde
(5a–5b) in 90% yield and confirmed by spectral analysis
(Scheme 1).
A condensation reaction of 4-(4-(benzo[d]thiazol-2-yl)ary-
loxy)benzaldehyde (5a–5b) was attempted with correspond-
ing aryl/heteroaryl hydrazide (6a–6k) in refluxing EtOH:THF
(4:1; 20 mL) using a catalytic amount of acetic acid. All the
reaction proceeded smoothly and afforded N0-(4-(4-(benzo[d]-
thiazol-2-yl)aryloxy)benzylidene)-(aryl)hydrazides
(7a–7v)
in 79–90% yield (Scheme 2, Table 1). The structure of all the
compounds (7a–7v) were confirmed by spectral analysis,
that is, IR, 1H NMR, 13C NMR and elemental analysis
(Figures S1–S49).
2.3. Spectral characterization
In the IR spectra of compounds 7a–7v, absorption signals
corresponding to N–H and C ¼ O stretching appeared at
3435–3209 cm− 1 and 1660–1638 cm− 1, respectively. In the
case of compound 7f, the appearance of a singlet at d
1
11.82 ppm in the H NMR spectrum confirmed the presence
of amidic –NH–. In addition to this, the appearance of two
singlets at d 8.48 ppm and 2.39 ppm revealed the presence
of imine [–N ¼ CH–(H23)] and methyl (–CH3) groups, respect-
ively. All the aromatic protons resonate in the region of d
8.16–7.21 ppm. A doublet appeared at d 7.35 ppm (J ¼ 8.0 Hz)
2. Results and discussion
2.1. Design and development of a-amylase inhibitors
Based on the activity reported in the literature (Taha et al., corresponding to H29/H31 protons showed a correlation with
2015, 2017, 2019), we have designed a series of new protons H28/H32 resonating at d 7.83 (J ¼ 12.1, 8.4 Hz) appears

JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS
3
Figure 1. Designing of target compounds as a-amylase inhibitors.
Scheme 1. Synthesis of 4-(4-(benzo[d]thiazol-2-yl)aryloxy)benzaldehydes (5a–5b).
Scheme 2. Synthesis of benzothiazole clubbed hydrazone (7a–7v) based molecular hybrids.
in the form of a doublet of doublet. A multiplet at d the form of multiplet at d 7.57–7.53 ppm. Proton H19/H21 res-
8.16–8.14 ppm corresponds to H11/H15 protons showed a cor- onating at d 7.83 (J ¼ 12.1, 8.4 Hz) in the form doublet of
relation with protons H12/H14 resonating in the form of the doublet showed a correlation with proton H18/H12 present in
triplet at d 7.23 ppm (J ¼ 8.6 Hz). H5 proton resonating at d the form of the triplet at d 7.23 (J ¼ 8.6 Hz) as shown in
8.16–8.14 ppm in the form of multiplet showed a correlation Figure 2.
with proton H6 resonating at d 7.49–7.45 ppm in the form of
The proton decoupled 13C NMR of compound 7f showed
the multiplet. A doublet at d 8.06 (J ¼ 7.7 Hz) corresponds to 24 peaks, in which signal at d 21.0 ppm corresponds to CH3
H8 proton showed a correlation with proton H7 appeared in proton. A singlet at d 2.42 ppm due to three protons

4
M. DUHAN ET AL.
confirms the presence of the methyl (–CH3) group. A doublet we screened all the synthesised compounds for inhibitory
potential of a-amylase at three different concentrations (50,
at d 8.07 ppm (J ¼ 7.85 Hz) corresponds to H8 proton showed
25, 12.5 mg/mL) as shown in Table 2. The inhibitory potential
of all the compounds was also compared with standard drug
acarbose. The selectivity and inhibitory activity profile related
to the different heteroaryl ring and functional group in com-
pounds 7a–7v have also been studied.
The maximum inhibition potential at 50 mg/mL, 25 mg/mL
and 12.5 mg/mL were displayed by compounds 7n, 7c and
7f, respectively. Compound 7n was found to be the most
potent with 87.5% of inhibition at 50 mg/mL. As far as min-
imum inhibition potential is concerned, among hydrazones,
compound 7u exhibited the lowest inhibitory activity at all
concentrations, that is, 50 mg/mL (65.99%), 25 mg/mL (50.74%)
and 12.5 mg/mL (42.28%) (Table 2). A comparative graph of
% inhibition for compounds 7a–7v with standard drug acar-
bose is shown in Figure 3.
a correlation with proton H7 present at d 7.58–7.54 ppm in
the form of the multiplet.
A doublet at d 8.03 ppm
(J ¼ 8.19 Hz) corresponds to H29/H33 proton showed a correl-
ation with proton H30/H32 present at d 7.49–7.45 ppm in the
form of the multiplet. A doublet at d 7.36 ppm (J ¼ 8.75 Hz)
and 7.32 ppm (J ¼ 8.19 Hz) for protons H18/H22 and H12/H14
showed a correlation with proton present at d 8.21–8.15 ppm
for H19/H21 and H11/H15 in the form of the multiplet. The aro-
matic resonates from d 8.22 to7.30 ppm, as shown in Figure
2. In the ESI-MS mass spectra of compound 7f the m/z value
was observed at 464.1424 (m þ 1), respectively.
2.4. Biological studies
Continuing our efforts towards the development of novel
a-amylase inhibitors (Duhan et al., 2019; Kumar et al., 2017a),
3. QSAR studies
3.1. Data set for QSAR
Table 1. Synthesis of benzothiazole clubbed hydrazone (7a–7v) based
molecular hybrids.
QSAR/QSPR is a significant method to establish a link
between the structure and endpoint of the compounds
(Karelson et al., 1996). To find out the most contributing
molecular structure from the studied hydrazones as a-amyl-
ase inhibitors, QSAR models were developed using CORAL
software. Data set for Hydrazones (Table 3) with their %
inhibition at three different concentrations (50 lg/mL, 25 lg/
mL and 12.5 lg/mL) were selected for the generation of
QSAR models. Chemical structures of 22 newly synthesized
compounds (7a–7v) were sketched using ChemAxon soft-
ware (Version & 6.2.2, 2014) and SMILES were generated
from those. The symbols $, % and & are used as quasi-
SMILES notation for three different concentrations, that is,
50 lg/mL, 25 lg/mL and 12.5 lg/mL. The present data set of
66 SMILES comprising of both standard SMILES and quasi-
SMILES notation, in which standard SMILES has been used
before quasi SMILES. Four random splits were made. The
non-identical nature (Kumar et al., 2019c; Toropova et al.,
2017) of these splits were calculated using the reported
S. no.
–R1
–R2
Compounds
Yield (%)
1.
2.
–H
–H
–C6H5
4-NO2–C6H4
4-Cl–C6H4
4-Pyridyl
4-OCH3–C6H4
4-CH3–C6H4
4-Br–C6H4
3-Pyridyl
2-Furyl
2-Thenyl
3-CH3–C6H4
–C6H5
4-NO2–C6H4
4-Cl–C6H4
4-Pyridyl
4-OCH3–C6H4
4-CH3–C6H4
4-Br–C6H4
3-Pyridyl
2-Furyl
2-Thenyl
7a
7b
7c
7d
7e
7f
7g
7h
7i
7j
7k
7l
7m
7n
7o
7p
7q
7r
84
88
87
80
88
90
89
80
81
83
85
83
87
85
79
87
89
83
81
78
80
88
3.
–H
4.
–H
5.
–H
6.
7.
–H
–H
8.
–H
9.
–H
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
–H
–H
–OCH3
–OCH3
–OCH3
–OCH3
–OCH3
–OCH3
–OCH3
–OCH3
–OCH3
–OCH3
–OCH3
7s
7t
7u
7v
3-CH3–C6H4
Figure 2. Various characteristic peaks of compound 7f.

JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS
5
ꢀ
ꢁ
100−PIa−amylase
methodology (Table S1) (see Supplementary Information).
Each split was further divided into four sets having individual
responsibility such as (i) Training set: was used to prepare
the QSAR model by computing the correlation weights (CW);
(ii) Invisible-training set: was responsible for checking the fit-
ness of the compounds which were absent in training set;
(iii) Calibration set: used to identify the start of overtraining
and (iv) Validation sets: used to validate the QSAR models
(Achary et al., 2019; Kumar et al., 2019b). The percentage
inhibition activity of a-amylase is expressed in logarithmic
scale using the following equation
pPIa−amylase ¼ −log
(1)
PIa−amylase
Here, PIa−amylase stands for the percentage inhibition of
a-amylase at various concentrations.
3.2. QSAR building
The data set of 66 SMILES of 22 compounds at three differ-
ent concentrations was made to develop QSAR model. A
detailed explanation of the SMILES features is shown in
Table 4.
The dataset was further divided into four random splits
(Table 5). Four sets, that is, training, invisible training, calibra-
tion and validation sets were made from each split. The pre-
diction of a developed QSAR model was also assessed in the
context of IIC. To check the robustness and predictability of
the constructed QSAR models, various statistical parameters
(Kumar & Kumar, 2020) such as R2, CCC, IIC, Q2 (cross-vali-
Table 2. a-Amylase inhibition activity of (E)-N0-(4-(4-(benzo[d]thiazol-2-yl)ary-
loxy)benzylidene)-(aryl)hydrazides (7a–7v).
% Inhibition
Concentration (mg/mL)
S. no.
Compound
50
25
12.5
dated correlation coefficient), s, MAE, F, CR , Y-test, RMSEP,
2
p
Q2F1, Q2F2, Q2F3 and R2m (avg) were calculated and these
parameters are depicted in Table 5. By using the equation of
TF1 for splits 1, 2, 3 and 4, the best (T , N ) are (6, 7), (2, 8),
(1, 7), (5, 5). However, by considering the influence of IIC on
the activity (pPIa−amylase), the equation of TF2 for splits 1, 2, 3
1
2
3
4
5
6
7
8
7a
7b
7c
7d
7e
7f
7g
7h
7i
7j
7k
7l
7m
7n
7o
7p
78.68 1.98
82.35 1.87
84.38 1.73
77.21 1.82
81.62 1.92
82.9 1.62
71.14 1.85
74.45 1.71
76.10 1.54
68.75 1.80
66.36 1.58
73.16 1.55
70.77 1.38
64.15 1.80
65.99 1.76
58.27 1.41
62.50 1.37
75.55 1.98
68.01 1.74
70.40 1.04
58.64 1.24
67.10 1.51
74.45 1.45
63.60 1.53
71.87 1.41
57.54 1.25
50.74 1.29
60.66 1.75
71.17 1.74
57.17 1.63
64.15 1.49
56.43 1.13
54.96 1.47
57.54 1.57
68.03 1.33
65.99 1.61
48.35 1.25
44.12 1.21
49.08 1.14
48.53 1.01
60.29 1.73
61.76 1.75
52.57 1.11
43.38 1.03
56.25 1.10
66.91 1.14
57.72 1.20
65.81 1.25
51.47 1.23
42.28 1.25
56.07 1.47
ꢁ
ꢁ
86.76 1.88
76.10 1.96
80.33 1.80
79.23 1.85
75.92 1.57
83.09 1.89
76.29 1.86
87.50 1.09
71.69 1.69
73.71 1.45
81.43 1.40
79.78 1.64
79.23 1.52
72.06 1.74
65.99 1.94
68.38 1.80
77.96 1.63
ꢁ
ꢁ
and 4 resulted into different (T , N ) and these are (5, 13), (6,
13), (5, 17) and (3, 16). The above outcome clearly shows
that the preferable threshold and number of epochs for both
target functions are not identical. A total of eight QSAR mod-
els was built using the balance of correlation method with
IIC (four for each data set using TF2) and without IIC (four
for each data set using TF1). The IIC was also implemented
to improve the predictability and robustness of developed
QSAR models. The best QSAR models, along with other stat-
istical parameters, are presented in Table 5. The values of
these statistical parameters are a good sign of the robust
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
7q
7r
7s
7t
7u
7v
Acarbose (AC)
67.25 1.39 QSAR models. The value of R2 5 0.8587 was highest for the
validation set of split 1 using IIC, so this QSAR model was
Figure 3. Comparative analysis of percentage inhibition of compounds 7a–7v.

6
M. DUHAN ET AL.

JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS
7

8
M. DUHAN ET AL.
Table 4. The detailed description of SMILES attributes.
S. no.
SMILES notation
Comments
1
2
Sk
SSk
SMILES-atoms, that is, one or two symbols which cannot be examined separately
A combination of two SMILES atoms
3
4
5
6
7
8
9
10.
11.
SSSk
BOND
HALO
NOSP
PAIR
Cmax
Nmax
Omax
Smax
A combination of three SMILES atoms
The presence or absence of double (‘¼’), triple (‘#’) and stereochemical (‘@’) bonds
Presence or absence of halogens
Presence or absence of nitrogen, oxygen, sulfur and phosphorus
The simultaneous presence of two SMILES atoms from the BOND, NOSP and HALO
Total number of rings (the range 0 … 9)
Total number of nitrogen atoms in the molecular structure
Total number of oxygen atoms in the molecular structures
Total number of sulphur atoms in the molecular structure
considered as leading models. The mathematical representa- QSAR models build by TF2 using IIC are more robust and
tion of leading QSAR model developed by CORAL software
using target function TF1 (without IIC) and TF2 (with IIC) are
given below:
reliable than by TF1. In the Monte-Carlo optimization, the
ideal model was chosen on the basis of the higher coeffi-
cient of determination (R2) for validation set and higher IIC
for the calibration set. The QSAR model was rated right or
wrong according to the method given in the literature
(Kumar & Kumar, 2020; Manisha et al., 2019; Nimbhal et al.,
2020; Toropova & Toropov, 2019b). After fastidious assess-
ment of statistical outcomes depicted in Table 5 the right
and wrong rating was given to each statistical parameter.
All statistical parameters excluding Q2 and MAE showed a
better model for splits 1 correctly. Similarly, in the case of
split 2, only two parameters Q2 and DRm2 predicted model
wrongly. For split 3, all statistical criteria apart from Y-test
exhibited better models correctly. For split 4, only four criter-
ion Q2, F, MAE, DR2m showed a better model wrongly. All
models were correctly predicted by the IIC.
The AD is a significant parameter of OECD principles and
those compound which falls outside the AD are defined as
outliers. In the case of combined dataset, no outlier was
found for the developed QSAR models using TF1 and TF2. It
has been observed that hydrazones of 4-pyridylcarbaldehyde
(7d and 7o) at different concentrations (HYD04, HYD15,
HYD26, HYD37, HYD48 and HYD59) and compound 7s at
25 lg/mL (HYD41) of split 4 only was outside the AD for the
developed QSAR models using TF2. On the other hand, no
outlier was found for split 1 and split 4 using TF1, but com-
pounds 7b (HYD02, HYD24, HYD46) and 7m (HYD13, HYD35,
HYD57) of split 2 at each concentration and were outside
the AD for the developed QSAR models. In split 3, com-
pounds 7b (HYD24), 7i (HYD31), 7t (HYD42) at 25 lg/mL and
7m (HYD13, HYD35, HYD57) all concentrations were
found outlier.
pPIa−amylase ¼ −10:9652922 ð6 0:3110474Þ þ 0:2760362
ð6 0:0076075Þ ꢁ DCWð6, 7Þ using TF1
n ¼ 17, R2 ¼ 0.8392, CCC ¼ 0.9126, IIC ¼ 0.8143, Q2 ¼ 0.7981
(training); n ¼ 17, R2 ¼ 0.8357, CCC ¼ 0.8952, IIC ¼ 0.6095,
Q2 ¼ 0.7931 (invisible training); n ¼ 16, R2 ¼ 0.7879, CCC ¼
0.8749, IIC
¼
0.8123, Q2 ¼ 0.7244 (calibration); n ¼ 16,
R2 ¼ 0.7548,
CCC
¼
0.8668,
IIC
¼
0.6516,
Q2 ¼ 0.6921 (validation).
pPIa−amylase ¼ −7:6186542 ð6 0:1453246Þ þ 0:1407535
ð6 0:0025541Þ ꢁ DCWð5, 13Þ using TF2
n ¼ 17, R2 ¼ 0.8746, CCC ¼ 0.9331, IIC ¼ 0.6547, Q2 ¼ 0.8491
(training); n ¼ 17, R2 ¼ 0.8317, CCC ¼ 0.8976, IIC ¼ 0.4154,
Q2 ¼ 0.7866 (Invisible training); n ¼ 16, R2 ¼ 0.7890, CCC ¼
0.8812, IIC
¼
0.8882, Q2 ¼ 0.7072 (calibration); n ¼ 16,
R2 ¼ 0.8587,
CCC
¼
0.9237,
IIC
¼
0.8932,
Q2 ¼ 0.8091 (validation).
As we can see from the above and other QSAR models
given in Table 5, the standard error of coefficients of DCWs
and constant parameters is very low and each of these mod-
elled QSAR equations can be applied to predict pPIa−amylase
:
The numerical data of experimental pPIa−amylase, predicted
pPIa−amylase, their difference and applicability domain (AD) of
newly synthesized compounds are given in Table 3 (see
Supplementary Information also). It can be seen that the cal-
culated pPIa−amylase at different concentrations for the data-
set were in good correlation with its experimental
pPIa−amylase values from QSAR models calculated with TF1
and TF2 which are presented in Figure 4. The robustness and
predictability of developed QSAR models had been ascer-
tained by the various statistical parameters such as R2, CCC,
Q2F1, Q2F2, Q2F3, Q2, C and metric R2m (Table 5). The robust-
2
Rp
ness of QSAR models was also quantified by the Y-random-
ization test (Table 5) (for details see Supplementary
Information). For the Y-randomization test, the constructed
4. Interpretation of structure–activity relationship
by mechanistic interpretation
A mechanistic interpretation means that structural features
responsible for the increase and decrease of an endpoint can
be obtained from the constructed QSAR models. In Monte
Carlo optimization method, the SMILES structural attributes
(SAk) had been interpreted for investigating the molecular
information, which was involved in a-amylase inhibition
models were free from chance correlation if the value of CR2
was more than 0.5 (Ojha & Roy, 2011).
The IIC was applied as a final statistical parameter to
authenticate the build QSAR models. The role of IIC in devel-
oping a robust and reliable QSAR model was well cited in
the literature (Kumar et al., 2019c; Manisha et al., 2019;
p
Nimbhal et al., 2020; Toropova & Toropov, 2019c). Therefore, study through the analysis of the correlation weight, the

JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS
9

10
M. DUHAN ET AL.
Figure 4. Plots of observed versus calculated pPIa−amylase values for all splits with both TF1 and TF2.
optimal descriptor, obtained from QSAR modelling. The pro- increase (if all correlation weights are positive in three inde-
moters of increase and decrease of the a-amylase inhibition pendent Monte Carlo optimization runs) ; (ii) List of pro-
were extracted from these structural attributes (SAk).
This was done by splitting the data samples into the fol- in three independent Monte Carlo optimization runs); (iii)
lowing four classes: (i) List of promoters of pPIa−amylase Attributes with an unclear role (the correlation weights have
moters of pIC50 decrease (all correlation weights are negative

JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS
11
Table 6. Promoter of endpoint increase and endpoint decrease for hydrazones.
No.
SAk
Remarks
Promoter of endpoint increase
Branching
CWs Probe 1 CWs Probe 2 CWs Probe 3 NSs NSc NSv Defect [SAk]
1
2
3
4
5
6
7
8
( … … … ..
0.1264
0.2795
0.4449
0.5658
0.4642
0.3831
0.0836
0.4370
0.2313
0.4043
0.0604
0.4832
0.2169
0.3663
0.0284
0.6403
0.1445
0.7416
0.0946
0.3311
0.8671
0.4520
0.2960
0.2395
0.7018
0.5959
0.2336
0.6466
0.3143
0.9737
0.3386
0.5074
0.1701
0.2959
0.5515
0.9098
0.0862
0.3342
0.6048
0.5057
0.7436
0.5982
0.2115
0.9931
0.5732
0.1439
0.4558
0.9681
0.2986
0.4671
0.1755
0.4749
0.3337
1.0135
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
þþþþN—O==¼ Presence of nitrogen with oxygen
þþþþN—S==¼ Presence of nitrogen with sulphur
þþþþO—B2=¼ Presence of oxygen with double bond
þþþþO—S==¼ Presence of oxygen with sulphur
þþþþS—B2=¼ Presence of sulfur with double bond
/ … ( … … .
/ … … … ..
Trans bond with branching
Trans bond
One aromatic ring with branching
One aromatic ring
Two aromatic ring with branching
Two aromatic ring
2nd ring having aromatic carbon with branching
Combination of double bond, nitrogen and trans bond
Combination of double bond, oxygen and branching
Aliphatic carbon with branching on first ring
Aliphatic carbon with double bond
Presence of double bond
9
1 … ( … … .
1 … … … ..
2 … ( … … .
2 … … … ..
2 … c … ( …
¼ … N … / …
¼ … O … ( …
C … ( … 1 …
C … ¼ … … .
BOND10000000
10
11
12
13
14
15
16
17
18
Promoter of endpoint decrease
1
2
3
4
5
6
( … C … ( …
1 … c … ( …
c … ( … … .
c … 1 … … .
& … … … ..
1 … & … … .
Branching on both side of aliphatic carbon
Combination one ring, aromatic carbon and branching
aromatic carbon with branching
aromatic carbon with one aromatic ring
Activity performed at 25 mg/mL
Activity performed at 25 mg/mL for a
compound having one aromatic ring
–0.0921
–0.2164
–0.1215
–0.2709
–0.5493
–0.1856
–0.2289
–0.3896
–0.0034
–0.2359
–0.4821
–0.5110
–0.0813
–0.3913
–0.1775
–0.1524
–0.1239
–0.0030
17
17
17
17
6
17
17
17
17
6
16
16
16
16
5
0.0000
0.0000
0.0000
0.0000
0.0037
0.0018
5
5
5
both negative and positive value in three independent their activity. The most active compounds 7n were sketched
Monte Carlo optimization runs) and (iv) Blocked (rare) attrib- using ChemAxon software and these were changed into 3D
utes (not involved in three independent Monte Carlo opti- structures using the MMFF94 force field.
mization runs). Lists of the most significant promoters of
pPIa−amylase are presented in Table 6.
5.1. Docking analysis of a-amylase
The common structural attributes (SAk) for increasing the
endpoint pPIa−amylase in split 1 are: (i) the presence of nitro-
gen with oxygen; (ii) the presence of nitrogen with sulfur;
(iii) the presence of oxygen with sulfur; (iv) one aromatic ring
with branching; (v) aromatic ring surrounded by two sp2
hybridized carbon. The common SAk for a promoter of
decreasing the endpoint is the inhibition study performed at
12.5 lg/mL concentration. The inhibitory potential of com-
pound 7k (o-methyl group) was less than the 7a (Table 2).
The decrease in the activity of compound 7k was due to the
presence of an ortho-methyl group and was confirmed by
putting the above SMILES together to generate a sequence
… c(C)c1 of SMILES.
Binding pose with the highest binding affinity for docked
conformation of 7n is shown in Figure 5. The ligand fits
snugly in the active site of the protein. It has been revealed
from Figure 5 that various interactions in terms of hydrogen
bonding, hydrophobic and electrostatic interactions are
mainly responsible for anchoring of the compound 7n in the
active site of a-amylase. The main hydrophobic interaction
are observed with Leu173, Tyr82, Leu166, Leu232, Tyr155, Trp83,
Tyr75. The phenyl ring of ether linkage and the phenyl group
of carbonyl group are involved in p–p interactions with resi-
dues Hie80, Arg344, Trp83 and Tyr75. The NH of hydrazone is
involved in H-bond with Gln35. All other interactions
observed in the protein-ligand complex of 7TAA:7n are
shown in Figure 5. In case of a-amylase, Glu230 and Asp206
are the residues responsible for the catalytic activity in
hydrolytic reactions (Svensson, 1994). From the docking stud-
ies, it has been revealed that presence of interactions
between 7n and (Tyr82, Arg204 and Glu230) of a-amylase are
similar to those responsible for a-amylase inhibition by acar-
bose as shown in Figure 5.
5. Molecular docking studies
Recently, the computer-assisted drug design (CADD) has
been accepted as a successful methodology of drug design
& discovery to understand the structural design of a drug
candidate. It is quite fast, cost-effective and result-oriented
fruitful in silico technique (Ferreira et al., 2015; Macalino
et al., 2015). The active site of a target protein can be
assessed using molecular docking to explore the probable
binding modes of a drug. The molecular docking studies
were performed with Auto Dock Vina software to predict the
probable mode of binding for most potent compounds 7n.
Docking simulation of 7n was performed with the active site
6. Molecular dynamics
Molecular dynamics simulation was performed with the
docked conformation of 7n with high binding affinity and
low binding energy for 100 ns to analyse the stability of pro-
of Aspergillus oryzae a-amylase (PDB ID: 7TAA) to establish tein–ligand complex in physiological condition. Both RMSD
the binding conformation and interactions responsible for (Å) and RMSF (Å) values were analysed for the protein–ligand

12
M. DUHAN ET AL.
Figure 5. Interactions (dashed lines) of compounds with active site residues of Aspergillus oryzae a-amylase (a) 7n; and (b) Acarbose.
Figure 6. RMSD (Å) and RMSF (Å) of simulated protein 7TAA.pdb with 7n.
complex. The stability of the protein depends upon RMSD, to Lys209 and His210 to Arg340. The ligand is stabilised by
protein with lower RMSD in dynamic simulations exhibits majority of hydrophobic interactions with 5%–70% of simula-
tion time with residues Tyr75, Trp83, Ile152, Tyr155, Leu166
,
greater stability. The RMSD of free protein get stabilised after
20 ns and remained below 2 Å for the remaining simulation
(Figure 6). Moreover, the deviation observed in RMSF during
simulation for each amino acid residue with respect to Ca
carbon atom was insignificant. The flexibility of residue on
ligand binding was analysed using metrics of root mean
square fluctuations (RMSFs), highest flexibility was observed
with RMSF of Thr72 (Ca:2.53 Å; backbone:2.37 Å), Asp157 (Ca:
2.67 Å; backbone: 2.42 Å), Glu156 (Ca: 2.50 Å; backbone:
2.58 Å) and Gly234 (Ca: 2.53 Å; backbone: 2.16 Å), however
lower RMSF was observed for Phe13 (Ca: 0.37 Å; backbone:
0.39 Å), Tyr12 (Ca: 0.39 Å; backbone: 0.43 Å) and Trp61 (Ca:
0.40 Å; backbone: 0.43 Å) (Figure 6).
Val171 and Leu173 (Figure 7). The 2D interaction diagram
reveals various hydrophobic and hydrogen bonding interac-
tions present in this protein–ligand complex. The NH group
of hydrazone donated one hydrogen bond to Asp168 with
13%, the carbonyl group of hydrazide accepted one hydro-
gen bond from Gln71 with 11% of simulation time. The p–p
stacking of Trp83 and Tyr75 with phenyl ring of hydrazide is
31% and 12%, respectively (Figure 7).
7. Conclusion
In this study, 22 novel (E)-N’-(4-(4-(benzo[d]thiazol-2-yl)arylox-
y)benzylidene)-(aryl)hydrazide (7a–7v) compounds were syn-
The protein–ligand complex shows significant interactions
with different amino acids, from Gln35 to Leu69, Gln71 to thesized. The synthesis of compounds 7a–7v can be
Tyr82, Trp83 to Asn121, His122 to Glu156, Gln158 to Thr170, Val171 achieved by the condensation reaction of 4-(4-

JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS
13
Figure 7. Interaction fraction and 2D interaction diagram of 7TAA complexed with 7n.
(benzo[d]thiazol-2-yl)aryloxy)benzaldehyde (5a–5b) with cor- using Perkin Elmer FTIR spectrophotometer and Cary 660
responding aryl/heteroaryl hydrazide (6a–6k) in refluxing Agilent IR spectrophotometer. The 1H NMR and 13C NMR
EtOH:THF (80:20, v/v) in the presence of the catalytic amount
of acetic acid. All the newly synthesized compounds and the
standard drug, that is, acarbose were subjected to in vitro
a-amylase inhibition activity at three different concentrations
(50 lg/mL, 25 lg/mL and 12.5 lg/mL). Among these, com-
pound 7n at 50 lg/mL concentration showed a remarkable
a-amylase inhibition, that is, 87.5% when compared with
standard acarbose (77.96%, 71.17% and 67.24% at 50 mg/mL,
25 mg/mL and 12.5 mg/mL concentration), respectively. The
percentage inhibition at various concentration was also pre-
dicted by developing the robust QSAR models using the bal-
ance of correlation method in CORAL software. The chemical
structures at different concentration with optimal descriptors
were represented by SMILES. A data set of 66 SMILES of 22
hydrazones at three distinct concentrations were prepared.
The significance of the IIC with AD was also studied at
depth. A QSAR models with best R2validation ¼ 0.8587 for split
1 was considered as a leading model. The outliers and pro-
moters of increase and decrease of endpoint were also
extracted independently from the leading model. Docking
studies of compounds 7n were performed against the active
site of A. oryzae a-amylase (PDB ID: 7TAA). The docking
studies revealed that the binding interactions found between
7n and a-amylase are similar to those responsible observed
in case of acarbose. MD simulations performed on protein–li-
gand complex for 100 ns have strengthened the findings of
docking studies. The RMSD of free protein get stabilised after
20 ns and remained below 2 Å for the remaining simulation.
Moreover, the deviation observed in RMSF during simulation
for each amino acid residue with respect to Ca carbon atom
is insignificant.
spectra were recorded on Bruker (Avance-II) at 400 MHz &
500 MHz in DMSO-d6 using tetramethylsilane (TMS) as an
internal standard. Chemical shifts (d) and coupling constants
(J) are expressed in ppm and Hz, respectively. HRMS analysis
was performed using LC–MS on SCIEX 5600þQTOF operating
at positive scan mode (120,000 FWMH) in a range of
100–1000 m/z. All chemicals used were supplied by Sigma-
Aldrich, Avra and Merck chemical companies.
Synthesis of 2-arylbenzothiazole (3a–3b)
To a mixture of 2-aminothiophenol 1 (1.0 mmol) and p-
hydroxybenzaldehyde (2a) (1.0 mmol) or vanillin (2b)
(1.0 mmol) in a vial, added SiO2–HNO3 (2 wt% of the alde-
hyde) and the contents were shaken at room temperature
(Kumar et al., 2015) (Scheme 1). An instant exothermic reac-
tion occurred with the completion of the reaction (as moni-
tored on TLC) and reaction mixture became yellowish-orange
solid that was directly chromatographed over silica gel col-
umn using petroleum ether:ethyl acetate (80:20, v/v) with
the increasing proportion of ethyl acetate to afford pure 2-
arylbenzothiazole (3a–3b).
Synthesis of 4-(4-(benzo[d]thiazol-2-yl)aryloxy)
benzaldehyde (5a–5b)
A mixture of 2-arylbenzothiazole (3a–3b) (1.0 mmol), potas-
sium carbonate (K2CO3) (1.2 mmol) in DMSO (10 mL) and 4-
fluorobenzaldehyde (4) (1.1 mmol) was taken in a round bot-
tom flask (50 mL) and heated at 120 ꢀC for 4–5 h on an oil
bath. After completion of the reaction as monitored by TLC
using petroleum ether:ethylacetate (80:20, v/v), the reaction
was quenched using ice and the solid, so formed, was fil-
tered under suction and recrystallized from ethanol to get
the desired product (5a–5b).
8. Experimental
Structures of all the compounds were identified by their
spectral data. Silica gel (60 F254) plates (precoated aluminium
plates) from Merck were used to monitor the reaction pro-
gress. All the melting points were determined in open glass 4-(4-(Benzo[d]thiazol-2-yl)phenoxy)benzaldehyde (5a)
Yield: 87%; M.pt: 118–120 ꢀC; IR (vmax cm− 1, KBr): 3061,
capillary tubes and are uncorrected. In IR absorption spectra,
the values are expressed as ꢀmax cm− 1 and were obtained 2732, 1702, 1593; 1H NMR (400 MHz, d (ppm) DMSO-d6):

14
M. DUHAN ET AL.
(E)-N0-(4-(4-(benzo[d]thiazol-2-yl)phenoxy)benzylidene)-
4-nitrobenzohydrazide (7b)
9.98 (s, 1H, CHO), 8.20–8.14 (m, 3H, H19/H21, H5), 8.07 (dd,
J ¼ 7.9 Hz, 1.1 Hz, 1H, H8), 7.99 (d, J ¼ 8.8 Hz, 2H, H11/H15),
7.58–7.54 (m, 1H, H7), 7.49–7.45 (m, 1H, H6), 7.31 (d,
J ¼ 8.8 Hz, 2H, H18/H22), 7.28 (d, J ¼ 8.8 Hz, 2H, H12/H14); 13C
NMR (100 MHz, d (ppm) DMSO-d6): 191.59, 166.35, 161.17,
157.65, 153.57, 134.47, 132.06, 131.99, 129.41, 129.16, 126.66,
125.47, 122.79, 122.34, 120.26, 118.66; Anal. Calc. For
C20H13NO2S: C, 72.49; H, 3.95; N, 4.23. Found: C, 72.44; H,
3.91; N, 4.19.
Yield: 88%; M.pt: 276–278 ꢀC; IR (vmax cm− 1, KBr): 3248,
3055, 1654, 1603, 1502, 1345; 1H NMR (400 MHz, d (ppm)
DMSO-d6): 12.17 (s, 1H, NH), 8.50 (s, 1H, H23), 8.40 (d,
J ¼ 8.9 Hz, 2H, H29/H31), 8.18–8.16 (m, 5H, H28/H32, H11/H15,
H5), 8.06 (d, J ¼ 7.7 Hz, 1H, H8), 7.85 (d, J ¼ 8.8 Hz, 2H, H19/
H21), 7.58–7.54 (m, 1H, H7), 7.49–7.45 (m, 1H, H6), 7.28–7.23
(m, 4H, H18/H22, H12/H14); Anal. Calc. For C27H18N4O4S: C,
65.58; H, 3.67; N, 11.33. Found: C, 65.51; H, 3.59; N, 11.28.
4-(4-(Benzo[d]thiazol-2-yl)-2-methoxyphenoxy)benzalde-
hyde (5b)
(E)-N0-(4-(4-(benzo[d]thiazol-2-yl)phenoxy)benzylidene)-
4-chlorobenzohydrazide (7c)
Yield: 89%; M.pt: 123-125 ꢀC; IR (vmax cm− 1, KBr): 1685,
1
1580, 1500, 1482, 1275; H NMR (400 MHz, d (ppm) DMSO-
Yield: 87%; M.pt: 288–290 ꢀC; IR (vmax cm− 1, KBr): 3245,
d6): 9.92 (s, 1H, –CHO), 8.16 (d, J ¼ 7.72 Hz, 1H, H5), 8.10 (d,
J ¼ 7.96 Hz, 1H, H8), 7.93–7.88 (m, 3H, H19/H21, H15), 7.72 (dd,
J ¼ 8.24, 2.04 Hz, 1H, H11), 7.59–7.55 (m, 1H, H7), 7.51–7.47
(m, 1H, H6), 7.36 (d, J ¼ 8.12 Hz, 1H, H12), 7.10 (d, J ¼ 8.74, 2H,
H18/H22), 3.89 (s, 3H, –OCH3); 13C NMR (100 MHz, d (ppm)
DMSO-d6): 191.45, 166.445, 162.19, 153.46, 151.73, 144.68,
134.59, 131.92, 131.12, 131.06, 126.73, 125.60, 122.99, 122.87,
122.36, 120.87, 116.23, 111.20, 55.95; Anal. Calc. For
C21H15NO3S: C, 69.79; H, 4.18; N, 3.88. Found: C, 69.75; H,
4.17; N, 3.85.
1
3061, 1654, 1484, 1244; H NMR (400 MHz, d (ppm) DMSO-
d6): 11.91 (s, 1H, NH), 8.48 (s, 1H, H23), 8.17–8.14 (m, 3H, H11/
H15, H5), 8.06 (d, J ¼ 7.92 Hz, H8), 7.96 (d, J ¼ 8.44 Hz, 2H, H28/
H32), 7.83 (d, J ¼ 8.52 Hz, 2H, H19/H21), 7.62 (d, J ¼ 8.4 Hz, 2H,
H29/H31), 7.58–7.53(m, 1H, H7), 7.49–7.45 (m, 1H, H6),
7.26–7.22 (m, 4H, H18/H22, H12/H14); 13C NMR (100 MHz, d
(ppm) DMSO-d6): 166.48, 161.99, 158.82, 157.28, 153.58,
147.33, 136.55, 134.41, 132.11, 130.24, 129.51, 129.30, 129.16,
128.55, 128.34, 126.63, 125.39, 122.71, 122.30, 119.47, 119.23;
Anal. Calc. For C27H18ClN3O2S: C, 67.01; H, 3.75; N, 8.68.
Found: C, 66.95; H, 3.70; N, 8.61.
Synthesis of N0-(4-(4-(benzo[d]thiazol-2-yl)aryloxy)
benzylidene)-(aryl)hydrazide (7a–7v)
(E)-N0-(4-(4-(benzo[d]thiazol-2-yl)phenoxy)
benzylidene)isonicotinohydrazide (7d)
To the stirred solution of 4-(4-(benzo[d]thiazol-2-yl)aryloxy)-
benzaldehyde (7a–7b) (1.0 mmol) in EtOH:THF (4:1; 20 mL),
aryl/heteroaryl hydrazide (6a–6k) (1.0 mmol) was added in
the presence of a catalytic amount of acetic acid and the
solution was refluxed for 6 h. The extent of the reaction was
examined by TLC using petroleum ether:ethyl acetate (80:20,
v/v). On completion of the reaction, the mixture was allowed
to cool at room temperature and the solid thus obtained
was filtered and recrystallized from absolute ethanol to
Yield: 80%; M.pt: 264–268 ꢀC; IR (vmax cm− 1, KBr): 3209,
3055, 1645, 1603, 1507, 1411, 1266; 1H NMR (400 MHz, d
(ppm) DMSO-d6): 12.10 (s, 1H, NH), 8.81–8.79 (m, 2H, H29/
H31), 8.49 (s, 1H, H23), 8.18–8.15 (m, 3H, H11/H15, H5), 8.06 (d,
J ¼ 7.8, 1H, H8), 7.86–7.83 (m, 4H, H28/H32, H19/H21), 7.58–7.54
(m, 1H, H7), 7.49–7.45 (m, 1H, H6), 7.28–7.22 (m, 4H, H18/H22,
H12/H14);13C NMR (100 MHz, d (ppm) DMSO-d6): 166.47,
161.51, 158.73, 157.47, 153.56, 150.31, 149.67, 148.19, 140.42,
134.39, 129.96, 129.32, 128.37, 126.64, 125.40, 122.71, 122.32,
121.49, 119.44, 119.29; Anal. Calc. For C26H18N4O2S: C,
69.32; H, 4.03; N, 12.44. Found: C, 69.26; H, 3.97; N, 12.39.
afford
N0-(4-(4-(benzo[d]thiazol-2-yl)aryloxy)benzylidene)-
(aryl)hydrazides (7a–7v) in good to excellent yield.
(E)-N0-(4-(4-(benzo[d]thiazol-2-yl)phenoxy)benzylidene)
benzohydrazide (7a)
(E)-N0-(4-(4-(benzo[d]thiazol-2-yl)phenoxy)benzylidene)-
4-methoxybenzohydrazide (7e)
Yield: 84%; M.pt: 246–248 ꢀC; IR (vmax cm− 1, KBr): 3214,
1
Yield: 88%; M.pt: 250–254 ꢀC; IR (vmax cm− 1, KBr): 3247,
3056, 1648, 1605, 1548, 1484, 1285; 1H NMR (400 MHz, d
(ppm) DMSO-d6): 11.79 (s, 1H, NH), 8.49 (s, 1H, H23),
8.18–8.14 (m, 3H, H11/H15, H5), 8.06 (d, J ¼ 7.88 Hz, 1H, H8),
7.94 (d, J ¼ 8.72 Hz, 2H, H28/H32), 7.81 (d, J ¼ 8.4 Hz, 2H, H19/
H21), 7.58–7.54 (m, 1H, H7), 7.49–7.45 (m, 1H, H6), 7.21–7.26
(m, 4H, H18/H22, H12/H14), 7.07 (d, J ¼ 8.8 Hz, 2H, H29/H31),
3055, 1638, 1478, 1250; H NMR (400 MHz, d (ppm) DMSO-
d6): 11.89 (s, 1H, NH), 8.49 (s, 1H, H23), 8.17–8.15 (m, 3H, H11/
H15, H5), 8.06 (d, J ¼ 7.92 Hz, 1H, H8), 7.93 (d, J ¼ 7.2 Hz, 2H,
H28/H32), 7.83 (d, J ¼ 8.7 Hz, 2H, H19/H21), 7.63–7.53 (m, 4H,
H7, H29/H31, H30), 7.49–7.45 (m, 1H, H6), 7.26–7.21 (m, 4H,
H18/H22, H12/H14); 13C NMR (100 MHz, d (ppm) DMSO-d6):
166.49, 158.84, 157.19, 153.56, 146.94, 142.64, 135.88, 134.39,
133.37, 131.74, 130.33, 129.31, 129.11, 128.47, 127.57, 126.65, 3.85 (s, 3H, OCH3); 13C NMR (400 MHz, d (ppm) DMSO-
125.41, 122.70, 122.33, 119.49, 119.22; Anal. Calc. For d6):166.49, 162.46, 158.89, 157.03, 153.56, 146.31, 143.37,
C27H19N3O2S: C, 72.14; H, 4.26; N, 9.35. Found: C, 72.09; H, 134.39, 130.49, 129.50, 129.29, 129.00, 128.25, 126.63, 125.39,
4.21; N, 9.29.
125.00, 122.69, 122.31, 119.50, 119.16, 113.68, 55.39; Anal.

JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS
15
Calc. For C28H21N3O3S: C, 70.13; H, 4.41; N, 8.76. Found: C, (400 MHz, d (ppm) DMSO-d6): 11.88 (s, 1H, NH), 8.48 (s, 1H,
70.09; H, 4.37; N, 8.69.
H23), 8.17–8.14 (m, 3H, H11/H15, H5), 8.06 (d, J ¼ 8.0 Hz, 1H,
H8), 7.97 (s, 1H, H30), 7.80 (d, J ¼ 8.76 Hz, 2H, H19/H21),
7.58–7.53 (m, 1H, H7), 7.49–7.45 (m, 1H, H6), 7.32–7.31 (d,
J ¼ 2.32, 1H, H28), 7.26–7.21 (m, 4H, H18/H22, H12/H14), 6.72
(dd, J ¼ 3.49, 1.74 Hz, 1H, H29); 13C NMR (100 MHz, d (ppm)
DMSO-d6): 166.48, 158.84, 157.17, 154.13, 153.56, 147.01,
146.60, 145.81, 134.39, 130.25, 129.29, 129.10, 128.28, 126.63,
125.39, 122.70, 122.31, 119.48, 119.18, 114.88, 112.07; Anal.
Calc. For C25H17N3O3S: C, 68.32; H, 3.90; N, 9.56. Found: C,
68.27; H, 3.82; N, 9.48.
(E)-N0-(4-(4-(benzo[d]thiazol-2-yl)phenoxy)benzylidene)-
4-methylbenzohydrazide (7f)
Yield: 90%; M.pt: 246–248 ꢀC; IR (vmax cm− 1, KBr): 3246,
3055, 1651, 1598, 1502, 1484, 1244; 1H NMR (400 MHz, d
(ppm) DMSO-d6): 11.82 (s, 1H, NH), 8.48 (s, 1H, H23),
8.16–8.14 (m, 3H, H11/H15, H5), 8.06 (d, J ¼ 7.7 Hz, 1H, H8),
7.83 (dd, J ¼ 12.1, 8.4 Hz, 4H, H28/H32, H19/H21), 7.57–7.53 (m,
1H, H7), 7.49–7.45 (m, 1H, H6), 7.35 (d, J ¼ 8.0 Hz, 2H, H29/
H31), 7.23 (t, J ¼ 8.6 Hz, 4H, H12/H14, H18/H22), 2.39 (s, 3H,
–CH3); 13C NMR (100 MHz, d (ppm) DMSO-d6): 166.49,
162.86, 158.86, 157.11, 153.56, 146.67, 141.77, 134.39, 130.47,
130.41, 129.29, 129.06, 128.98, 128.27, 127.60, 126.63, 125.39,
122.70, 122.31, 119.49, 119.18, 21.01; Anal. Calc. For
C28H21N3O2S: C, 72.55; H, 4.57; N, 9.06. Found: C, 72.49; H,
4.51; N, 9.02.
(E)-N’-(4-(4-(benzo[d]thiazol-2-yl)phenoxy)benzylidene)
thiophene-2-carbohydrazide (7j)
Yield: 83%; M.pt: 258–260 ꢀC; IR (vmax cm− 1, KBr): 3260,
3145, 1638, 1597, 1481, 1251; 1H NMR (400 MHz, d (ppm)
DMSO-d6): 11.87 (s, 1H, NH), 8.46 (s, 1H, H23), 8.17–8.14 (m,
3H, H11/H15, H5), 8.07–8.05 (m, 1H, H8), 7.98–7.82 (m, 4H, H19/
H21 H30, H28), 7.58–7.54 (m, 1H, H7), 7.49–7.45 (m, 1H, H6),
7.27–7.23 (m, 5H, H29, H18/H22, H12/H14); 13C NMR (100 MHz,
d (ppm) DMSO-d6): 166.48, 161.21, 158.86, 154.37, 153.56,
146.73, 143.25, 134.74, 134.39, 132.89, 132.30, 130.26, 129.29,
128.27, 127.58, 126.63, 125.39, 122.70, 122.31, 119.68, 119.15;
Anal. Calc. For C25H17N3O2S2: C, 65.91; H, 3.76; N, 9.22.
Found: C, 65.83; H, 3.70; N, 9.18.
(E)-N0-(4-(4-(benzo[d]thiazol-2-yl)phenoxy)benzylidene)-
4-bromobenzohydrazide (7g)
Yield: 89%; M.pt: 254–258 ꢀC; IR (vmax cm− 1, KBr): 3288,
3058, 1651, 1590, 1483, 1549, 1245; 1H NMR (400 MHz, d
(ppm) DMSO-d6): 11.95 (s, 1H, NH), 8.47 (s, 1H, H23),
8.17–8.15 (m, 3H, H11/H15, H5), 8.06 (d, J ¼ 8.00 Hz, 1H, H8),
7.89 (d, J ¼ 8.6 Hz, 2H, H28/H32), 7.83 (d, J ¼ 8.68 Hz, 2H, H19/
H21), 7.77 (d, J ¼ 8.48 Hz, 2H, H29/H31), 7.58–7.54 (m, 1H, H7),
7.49–7.45 (m, 2H, H6), 7.27–7.22 (m, 4H, H18/H22, H12/H14); 13C
NMR (100 MHz, d (ppm) DMSO-d6): 166.49, 162.08, 157.27,
153.56, 147.31, 134.39, 132.42, 131.50, 130.21, 129.69, 129.31,
129.18, 128.32, 126.65, 125.51, 125.41, 122.70, 122.32, 119.47,
119.24; Anal. Calc. For C27H18BrN3O2S: C, 61.37; H, 3.43; N,
7.95. Found: C, 61.31; H, 3.35; N, 7.86.
(E)-N’-(4-(4-(benzo[d]thiazol-2-yl)phenoxy)benzylidene)-
3-methylbenzohydrazide (7k)
Yield: 85%; M.pt: 198–200 ꢀC; IR (vmax cm− 1, KBr): 3429,
3233, 1653, 1482, 1289, 1244; 1H NMR (400 MHz, d (ppm)
DMSO-d6): 11.84 (s, 1H, NH), 8.48 (s, 1H, H23), 8.17–8.15 (m,
3H, H11/H15, H5), 8.06 (d, J ¼ 7.96 Hz, 1H, H8), 7.82 (d,
J ¼ 8.60 Hz, 2H, H19/H21), 7.75–7.72 (m, 2H, H28, H32),
7.58–7.54 (m, 1H, H7), 7.49–7.42 (m, 3H, H6, H29, H30),
7.26–7.22 (m, 4H, H18/H22, H12/H14), 2.41 (s, 3H, –CH3); 13C
NMR (100 MHz, d (ppm) DMSO-d6): 166.50, 161.11, 158.86,
157.16, 153.57, 146.80, 137.79, 134.40, 133.36, 132.32, 133.37,
129.31, 129.09, 128.36, 128.29, 128.07, 126.65, 125.41, 124.71,
122.71, 122.35, 119.50, 119.21, 20.93; Anal. Calc. For
(E)-N0-(4-(4-(benzo[d]thiazol-2-yl)phenoxy)benzylidene)
nicotinohydrazide (7h)
Yield: 80%; M.pt: 282–284 ꢀC; IR (vmax cm− 1, KBr): 3429,
3198, 3052, 1644, 1602, 1507, 1481, 1264; 1H NMR
(400 MHz, d (ppm) DMSO-d6): 12.05 (s, 1H, NH), 9.08 (d,
J ¼ 1.76 Hz, 1H, H32), 8.78 (dd, J ¼ 4.80, 1.52 Hz, 1H, H30), 8.47
(s, 1H, H23), 8.27 (dt, J ¼ 8.0, 1.8 Hz, 1H, H28), 8.18–8.14 (m,
3H, H11/H15, H5), 8.06 (d, J ¼ 7.96 Hz, 1H, H8), 7.84 (d,
J ¼ 8.72 Hz, 2H, H19/H21), 7.61–7.54 (m, 2H, H29, H7), 7.49–7.45
(m, 1H, H6), 7.27–7.23 (m, 4H, H18/H22, H12/H14); 13C NMR
C
28H21N3O2S: C, 72.55; H, 4.57; N, 9.06. Found: C, 72.50; H,
4.52; N, 9.01
(E)-N’-(4-(4-(benzo[d]thiazol-2-yl)-2-methoxyphenoxy)
benzylidene)benzohydrazide (7 l)
Yield: 83%; M.pt: 240–244 ꢀC; IR (vmax cm− 1, KBr): 3258,
3058, 1651, 1609, 1505, 1483, 1270; 1H NMR (500 MHz, d
(ppm) DMSO-d6): 11.81 (s, 1H, NH), 8.44 (s, 1H, H23), 8.15 (d,
J ¼ 8.1 Hz, 1H, H5), 8.09 (d, J ¼ 8.1 Hz, 1H, H8), 7.92 (d,
J ¼ 7.5 Hz, 2H, H28/H32), 7.87–7.86 (m, 1H, H15), 7.74 (d,
J ¼ 8.5 Hz, 2H, H19/H21), 7.70 (dd, J ¼ 8.3, 2.0 Hz, 1H, H11),
7.61–7.52 (m, 4H, H29/H31, H30, H7), 7.49–7.46 (m, 1H, H6),
7.26 (d, J ¼ 8.3 Hz, 1H, H12), 7.04 (d, J ¼ 8.5 Hz, 2H, H18/H22),
3.91 (s, 3H, –OCH3); 13C NMR (125 MHz, d (ppm) DMSO-d6):
(100 MHz,
d (ppm) DMSO-d6): 166.48, 161.61, 158.77,
157.36, 153.56, 152.27, 148.53, 147.59, 135.43, 134.39, 130.08,
129.31, 129.24, 129.13, 128.33, 126.64, 125.40, 123.60, 122.70,
122.32, 119.45, 119.26; Anal. Calc. For C26H18N4O2S: C,
69.32; H, 4.03; N, 12.44. Found: C, 69.28; H, 3.97; N, 12.38.
(E)-N0-(4-(4-(benzo[d]thiazol-2-yl)phenoxy)benzylidene)
furan-2-carbohydrazide (7i)
Yield: 81%; M.pt: 240–246 ꢀC; IR (vmax cm− 1, KBr): 3229, 167.08, 163.56, 159.10, 153.97, 152.09, 147.74, 146.43, 135.05,
3057, 1655, 1590, 1503, 1481, 1289, 1017; 1H NMR 133.91, 132.18, 130.82, 129.51, 129.39, 128.94, 128.04, 127.19,

16
M. DUHAN ET AL.
126.02, 123.30, 122.80, 122.54, 121.34, 117.36, 111.66, 56.44;
Anal. Calc. For C28H21N3O3S: C, 70.13; H, 4.41; N, 8.76.
Found: C, 70.09; H, 4.35; N, 8.72.
(E)-N’-(4-(4-(benzo[d]thiazol-2-yl)-2-methoxyphenoxy)
benzylidene)-4-methoxybenzohydrazide (7p)
Yield: 87%; M.pt: 284–288 ꢀC; IR (vmax cm− 1, KBr): 3238,
1645, 1605, 1500, 1417, 1270; 1H NMR (500 MHz, d (ppm)
DMSO-d6): 11.68 (s, 1H, NH), 8.43 (s, 1H, H23), 8.17 (d,
J ¼ 7.93 Hz, 1H, H5), 8.10 (d, J ¼ 8.06 Hz, 1H, H8), 7.91 (d,
J ¼ 8.76 Hz, 2H, H28/H32), 7.87 (d, J ¼ 1.84 Hz, 1H, H15),
7.75–7.70 (m, 3H, H19/H21, H11), 7.60–7.54 (m, 1H, H7),
7.51–7.46 (m, 1H, H6), 7.27 (d, J ¼ 8.25 Hz, 1H, H12), 7.05 (m,
4H, H18/H22, H29/H31,), 3.92 (s, 3H, –OCH3), 3.84 (s, 3H,
–OCH3); 13C NMR (125 MHz, d (ppm) DMSO-d6): 167.07,
162.89, 162.44, 158.97, 153.99, 152.09, 147.06, 146.48, 135.06,
130.80, 129.96, 129.67, 129.26, 127.19, 126.02, 125.95, 123.32,
122.84, 122.52, 121.35, 117.38, 114.18, 111.65, 56.45, 55.89;
Anal. Calc. ForC29H23N3O4S: C, 68.35; H, 4.55; N, 8.25.
Found: C, 68.31; H, 4.54; N, 8.21.
(E)-N’-(4-(4-(benzo[d]thiazol-2-yl)-2-methoxyphenoxy)
benzylidene)-4-nitrobenzohydrazide (7m)
Yield: 87%; M.pt: 188–190 ꢀC; IR (vmax cm− 1, KBr): 3230,
3067, 1654, 1600, 1507, 1482, 1346, 1270; 1H NMR
(500 MHz, d (ppm) DMSO-d6):12.09 (s, 1H, NH), 8.46 (s, 1H,
H23), 8.37 (d, J ¼ 8.5 Hz, 2H, H29/H31), 8.16–8.14 (m, 3H, H28/
H32, H5), 8.09 (d, J ¼ 8.20 Hz, 1H, H8), 7.87 (d, J ¼ 2.02 Hz, 1H,
H15), 7.76 (d, J ¼ 8.82 Hz, 2H, H19/H21), 7.70 (dd, J ¼ 8.26,
2.05 Hz, 1H, H11), 7.58–7.55 (m, 1H, H7), 7.49–7.46 (m, 1H, H6),
7.28 (d, J ¼ 8.21 Hz, 1H, H12), 7.05 (d, J ¼ 8.76 Hz, 2H, H18/H22),
3.91 (s, 3H, –OCH3); 13C NMR (125 MHz, d (ppm) DMSO-d6):
167.05, 161.87, 159.35, 153.97, 152.11, 149.69, 148.85, 146.31,
139.59, 135.05, 131.13, 130.90, 129.59, 129.19, 127.18, 126.02,
124.10, 123.31, 122.81, 122.63, 121.34, 117.33, 111.67, 56.44;
Anal. Calc. For C28H20N4O5S: C, 64.11; H, 3.84; N, 10.68.
Found: C, 64.11; H, 3.84; N, 10.68.
(E)-N’-(4-(4-(benzo[d]thiazol-2-yl)-2-methoxyphenoxy)
benzylidene)-4-methylbenzohydrazide (7q)
Yield: 89%; M.pt: 272–278 ꢀC; IR (vmax cm− 1, KBr): 3266,
2958, 1651, 1611, 1499, 1418, 1270; 1H NMR (500 MHz, d
(ppm) DMSO-d6): 11.74 (s, 1H, NH), 8.44 (s, 1H, H23), 8.16 (d,
J ¼ 7.94 Hz, 1H, H5), 8.10 (d, J ¼ 8.07 Hz, 1H, H8), 7.87 (d,
J ¼ 1.65 Hz, 2H, H15), 7.83 (d, J ¼ 7.91 Hz, 2H, H28/H32), 7.73 (d,
J ¼ 8.51 Hz, 2H, H19/H21), 7.70 (dd, J ¼ 8.29, 1.75 Hz, 1H, H11),
7.58–7.55 (m, 1H, H7), 7.50–7.46 (m, 1H, H6), 7.33 (d,
J ¼ 7.90 Hz, 2H, H29/H31), 7.27 (d, J ¼ 8.23 Hz, 1H, H12), 7.04 (d,
J ¼ 8.52 Hz, 2H, H18/H22), 3.91 (s, 3H, –OCH3), 2.38 (s, 3H,
–CH3); 13C NMR (125 MHz, d (ppm) DMSO-d6): 167.07,
163.33, 159.04, 153.98, 152.09, 147.42, 146.44, 142.21, 135.05,
131.03, 130.81, 129.59, 129.46, 129.33, 128.06, 127.19, 126.02,
123.31, 122.82, 122.53, 121.34, 117.36, 111.64, 56.44, 21.50;
Anal. Calc. For C29H23N3O3S: C, 70.57; H, 4.70; N, 8.51.
Found: C, 70.54; H, 4.63; N, 8.48.
(E)-N’-(4-(4-(benzo[d]thiazol-2-yl)-2-methoxyphenoxy)
benzylidene)-4-chlorobenzohydrazide (7n)
Yield: 85%; M.pt: 246–250 ꢀC; IR (vmax cm− 1, KBr): 3285,
3067, 1655, 1596, 1483, 1272; 1H NMR (500 MHz, d (ppm)
DMSO-d6): 11.87 (s, 1H, NH), 8.44 (s, 1H, H23), 8.15 (d,
J ¼ 8.00 Hz, 1H, H5), 8.09 (d, J ¼ 8.07 Hz, 1H, H8), 7.94 (d,
J ¼ 8.47 Hz, 2H, H28/H32), 7.87 (d, J ¼ 1.74 Hz, 1H, H15), 7.74 (d,
J ¼ 8.65 Hz, 2H, H19/H21), 7.69 (dd, J ¼ 8.25, 1.81 Hz, 1H, H11),
7.61 (d, J ¼ 8.47 Hz, 2H, H29/H31), 7.58–7.54 (m, 1H, H7),
7.49–7.46 (m, 1H, H6), 7.26 (d, J ¼ 8.25 Hz, 1H, H12), 7.04 (d,
J ¼ 8.62 Hz, 2H, H18/H22), 3.91 (s, 3H, –OCH3); 13C NMR
(125 MHz,
d (ppm) DMSO-d6): 167.06, 162.44, 159.18,
153.98, 152.09, 148.04, 146.38, 137.00, 135.05, 132.62, 130.84,
129.99, 129.44, 129.40, 129.03, 127.17, 126.01, 123.31, 122.80,
122.55, 121.33, 117.35, 111.65, 56.43; Anal. Calc. For
C28H20ClN3O3S: C, 65.43; H, 3.92; N, 8.18. Found: C, 65.37; H,
3.90; N, 8.15.
(E)-N’-(4-(4-(benzo[d]thiazol-2-yl)-2-methoxyphenoxy)
benzylidene)-4-bromobenzohydrazide (7r)
(E)-N’-(4-(4-(benzo[d]thiazol-2-yl)-2-methoxyphenoxy)
benzylidene)isonicotinohydrazide (7o)
Yield: 83%; M.pt: 210–212 ꢀC; IR (vmax cm− 1, KBr): 3209,
3057, 1647, 1608, 1492, 1270; 1H NMR (500 MHz, d (ppm)
DMSO-d6): 11.87 (s, 1H, NH), 8.44 (s, 1H, H23), 8.15 (d,
J ¼ 7.92 Hz, 1H, H5), 8.09 (d, J ¼ 8.08 Hz, 1H, H8), 7.88–7.86 (m,
3H, H28/H32, H15), 7.76–7.68 (m, 5H, H19/H21, H29/H31), 7.69
(dd, J ¼ 8.23, 1.68 Hz, 1H, H11), 7.58–7.54 (m, 1H, H7),
7.49–7.46 (m, 1H, H6), 7.26 (d, J ¼ 8.25 Hz, 1H, H12), 7.04 (d,
J ¼ 8.61 Hz, 2H, H18/H22), 3.91 (s, 3H, OCH3); 13C NMR
Yield: 79%; M.pt: 274–278 ꢀC; IR (vmax cm− 1, KBr): 3261,
3048, 1660, 1508, 1487, 1409, 1272; 1H NMR (500 MHz, d
(ppm) DMSO-d6): 12.03 (s, 1H, NH), 8.79 (d, J ¼ 6.00 Hz, 2H,
H29/H31), 8.45 (s, 1H, H23), 8.17 (d, J ¼ 8.00 Hz, 1H, H5), 8.10
(d, J ¼ 8.00 Hz, 1H, H8), 7.88 (d, J ¼ 1.69 Hz, 1H, H15), 7.82 (d,
J ¼ 6.00 Hz, 2H, H28/H32), 7.76 (d, J ¼ 8.65 Hz, 2H, H19/H21),
7.71 (dd, J ¼ 8.26, 1.78 Hz, 1H, H11), 7.57 (t, J ¼ 7.64 Hz, 1H,
H7), 7.48 (t, J ¼ 7.60 Hz, 1H, H6), 7.29 (d, J ¼ 8.26 Hz, 1H, H12),
7.05 (d, J ¼ 8.63 Hz, 2H, H18/H22), 3.92 (s, 3H, –OCH3); 13C
NMR (125 MHz, d (ppm) DMSO-d6): 167.06, 161.97, 159.37,
153.97, 152.12, 150.79, 148.94, 146.30, 140.98, 135.05, 130.91,
129.60, 129.15, 127.20, 126.04, 123.32, 122.83, 122.67, 121.98,
121.35, 117.33, 111.68, 56.46; Anal. Calc. For C27H20N4O3S:
C, 67.49; H, 4.20; N, 11.66. Found: C, 67.46; H, 4.17; N, 11.64.
(125 MHz,
d (ppm) DMSO-d6): 167.04, 162.52, 159.18,
153.99, 152.09, 148.04, 146.38, 135.06, 132.99, 131.96, 130.85,
130.16, 129.44, 129.41, 127.16, 126.00, 125.94, 123.31, 122.80,
122.55, 121.32, 117.35, 111.64, 56.43; Anal. Calc. For
C28H20BrN3O3S: C, 60.22; H, 3.61; N, 7.52. Found: C, 60.19;
H, 3.59; N, 7.50.

JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS
17
DMSO-d6): 11.77 (s, 1H, NH), 8.44 (s, 1H, H23), 8.15 (d,
J ¼ 7.93 Hz, 1H, H5), 8.09 (d, J ¼ 8.10 Hz, 1H, H8), 7.88–7.86 (m,
1H, H15), 7.74–7.68 (m, 5H, H19/H21, H11, H32, H28), 7.58–7.55
(m, 1H, H7), 7.49–7.46 (m, 1H, H6), 8.43–8.40 (m, 2H, H29, H30),
7.26 (d, J ¼ 8.24 Hz, 1H, H12), 7.04 (d, J ¼ 8.54 Hz, 2H, H18/H22),
3.91 (s, 3H, –OCH3), 2.51 (s, 3H, CH3); 13C NMR (125 MHz, d
(ppm) DMSO-d6): 167.06, 163.58, 159.07, 153.99, 152.09,
147.55, 146.42, 138.25, 135.06, 133.93, 132.74, 130.82, 129.55,
129.36, 128.82, 128.54, 127.17, 126.00, 125.20, 123.31, 122.80,
122.53, 121.33, 117.36, 111.64, 56.43, 21.42; Anal. Calc. For
(E)-N’-(4-(4-(benzo[d]thiazol-2-yl)-2-methoxyphenoxy)
benzylidene)nicotinehydrazide (7s)
Yield: 81%; M.pt: 205–208 ꢀC; IR (vmax cm−1, KBr): 3248, 1647,
1
1505, 1482, 1417, 1270; H NMR (500 MHz, d (ppm) DMSO-
d6): 11.97 (s, 1H, NH), 9.09–9.07 (m, 1H, H32), 8.78–8.75 (m, 1H,
H30), 8.43 (s, 1H, H23), 8.27–8.25 (m, 1H, H28), 8.16 (d,
J ¼ 7.88 Hz, 1H, H5), 8.10 (d, J ¼ 7.99 Hz, 1H, H8), 7.88 (d,
J ¼ 1.83 Hz, 1H, H15), 7.76 (d, J ¼ 8.72 Hz, 2H, H19/H21), 7.71 (dd,
J ¼ 8.25, 1.92 Hz, 1H, H11), 7.59–7.55 (m, 2H, H29, H7), 7.50–7.47
(m, 1H, H6), 7.28 (d, J ¼ 8.25 Hz, 1H, H12), 7.05 (d, J ¼ 8.70 Hz,
2H, H18/H22), 3.92 (s, 3H, –OCH3); 13C NMR (125 MHz, d (ppm)
DMSO-d6): 167.05, 162.04, 159.27, 153.98, 152.71, 152.12,
149.02, 148.32, 146.34, 135.89, 135.06, 130.88, 129.70, 129.51,
129.28, 127.18, 126.02, 124.07, 123.32, 122.83, 122.63, 121.34,
117.34, 111.66, 56.45; Anal. Calc. For C27H20N4O3S: C, 67.49;
H, 4.20; N, 11.66. Found: C, 67.45; H, 4.17; N, 11.64.
C
29H23N3O3S: C, 70.57; H, 4.70; N, 8.51. Found: C, 70.54; H,
4.68; N, 8.49.
a-Amylase inhibition studies
The inhibition studies of compounds 7a–7v for a-amylase
were carried out by adopting the assay procedure as
described in literature (Keharom et al., 2016). a-Amylase
inhibitory activity was carried out by DNSA method on
Systronics Spectrophotometer 166 using acarbose as a refer-
ence drug.
All the compounds and standard drug were dissolved in
dimethyl sulphoxide (DMSO). A stock solution of 1 mg/mL
concentration was prepared by using the DMSO solvent.
Inhibitory potential of all the synthesized derivatives toward
(E)-N’-(4-(4-(benzo[d]thiazol-2-yl)-2-methoxyphenoxy)
benzylidene)furan-2-carbohydrazide (7t)
Yield: 78%; M.pt: 200–203 ꢀC; IR (vmax cm− 1, KBr): 3233,
1655, 1606, 1506, 1479, 1413, 1272, 1231; 1H NMR
(500 MHz, d (ppm) DMSO-d6): 11.80 (s, 1H, NH), 8.44 (s, 1H,
H23), 8.16 (d, J ¼ 7.91 Hz, 1H, H5), 8.09 (d, J ¼ 8.06 Hz, 1H, H8),
7.95 (s, 1H, H30), 7.87 (d, J ¼ 1.84 Hz, 1H, H15), 7.73–7.69 (m,
3H, H19/H21, H11), 7.56 (t, J ¼ 7.56 Hz, 1H, H7), 7.48 (t, a-amylase was tested at three different concentration, that is,
J ¼ 7.58 Hz, 1H, H6), 7.30–7.26 (m, 2H, H28, H12), 7.03 (d,
50 mg/mL, 25 mg/mL and 12.5 mg/mL concentrations, respect-
J ¼ 8.69 Hz, 2H, H18/H22), 6.70 (dd, J ¼ 3.38, 1.65 Hz, 1H, H29),
ively. Acarbose was used as a reference drug and the solu-
3.91 (s, 3H, –OCH3); 13C NMR (500 MHz, d (ppm) DMSO-d6):
tions without test samples were used as a control. The
167.05, 159.10, 154.59, 153.98, 152.08, 147.75, 147.15, 146.41,
146.23, 135.06, 130.83, 129.43, 129.37, 127.18, 126.01, 123.31,
122.82, 122.54, 121.33, 117.36, 115.29, 112.54, 111.64, 56.43;
Anal. Calc. For C26H19N3O4S: C, 66.51; H, 4.08; N, 8.95.
Found: C, 66.48; H, 4.03; N, 8.91.
a-amylase (A. oryzae) enzyme solution having 50 mg/mL con-
centration was prepared using 20 mM sodium-potassium buf-
fer (pH 6.9). A total of 1 mL of sample solution 7a–7v at
different concentration (12.5, 25, 50 mg/mL) and 1 mL of
enzyme solution were mixed and the solution was incubated
at 37 ꢀC for 30 min. A 1 mL of starch solution (prepared by
dissolving starch (500 mg) in 25 mL of 0.5 N NaOH and heat
the solution for 5 min at 100 ꢀC) was then added to each test
tubes and the mixture was further incubated for 15 min at
37 ꢀC. The reaction was stopped by adding 1 mL of 96 mM of
DNSA solution to each test tube. After shaking the reaction
mixture, the closed test tubes were further heated on the
water bath for 15 min at 85 ꢀC. Later on, test tubes were fur-
ther cooled at room temperature and absorbance was meas-
ured at 650 nm. Blanks were prepared without enzymes. A
control experiment was conducted in the same manner by
replacing the drug sample with 1 mL DMSO. Inhibition per-
centage of a-amylase was calculated by the following math-
ematical equation (Nickavar & Yousefian, 2010)
(E)-N’-(4-(4-(benzo[d]thiazol-2-yl)-2-methoxyphenoxy)
benzylidene)thiophene-2-carbohydrazide (7u)
Yield: 80%; M.pt: 220–222 ꢀC; IR (vmax cm− 1, KBr): 3435,
3054, 1651, 1611, 1515, 1396, 1271; 1H NMR (500 MHz, d
(ppm) DMSO-d6): 11.83 (s, 1H, NH), 8.42 (s, 1H, H23), 8.17 (d,
J ¼ 7.82 Hz, 1H, H5), 8.11–8.05 (m, 2H, H8, H28), 7.94–7.91 (m,
1H, H30), 7.87 (d, J ¼ 1.93 Hz, 1H, H15), 7.81–7.72 (m, 2H, H19/
H21), 7.71 (dd, J ¼ 8.26, 1.98 Hz, 1H, H11), 7.59–7.55 (m, 1H,
H7), 7.50–7.47 (m, 1H, H6), 7.27 (d, J ¼ 8.25 Hz, 1H, H12),
7.24–7.21 (m, 1H, H29), 7.06 (d, J ¼ 6.96 Hz, 2H, H18/H22), 3.92
(s, 3H, –OCH3); 13C NMR (125 MHz, d (ppm) DMSO-d6):
167.06, 161.67, 159.12, 158.98, 153.98, 152.08, 146.43, 143.86,
135.05, 132.24, 130.83, 129.50, 129.40, 128.56, 127.19, 126.03,
123.32, 122.84, 122.52, 121.34, 117.53, 117.37, 111.63, 56.45;
Anal. Calc. For C26H19N3O3S2: C, 64.31; H, 3.94; N, 8.65.
Found: C, 64.29; H, 3.90; N, 8.62.
% Inhibition ¼ (Control-Test)/Control ꢂ 100
8.1. QSAR building
8.1.1. Method for QSAR model development
(E)-N’-(4-(4-(benzo[d]thiazol-2-yl)-2-methoxyphenoxy)
benzylidene)-3-methylbenzohydrazide (7v)
The predictive QSAR model between pPIa−amylase and correl-
ation weight was built by using Monte Carlo method present
in CORAL software. The SMILES and quasi SMILES based
Yield: 88%; M.pt: 188–190 ꢀC; IR (vmax cm− 1, KBr): 3230,
1648, 1609, 1417, 1482, 1271; 1H NMR (500 MHz, d (ppm) descriptor can be calculated by the summation of the

18
M. DUHAN ET AL.
þN
correlation weights (CWs) of diverse SMILES attributes. The
balance of correlation approach was used to build QSAR
models and the endpoint (pPIa−amylase) was calculated by the
following mathematical equations (Kumar & Kumar, 2019;
Veselinovic et al., 2013)
X
1
þMAEset ¼
jDkjDk ꢃ 0, þN is the number of Dk ꢃ 0
þN k¼1
(9)
Here
ð
Þ
ð
Þ
ð
Þ
pPIa−amylase ¼ f SMILES$ or f SMILES% or f SMILES&
Dk ¼ observedk−Predictedk ¼ Yobs−Ypred
The values from 1 to 15 for threshold (T) and 1 to 50 for
Nepoch were used to attain the most predictive arrangement
of T and Nepoch for all splits. The dRweight and Dlimit were 0.1.
(10)
(2)
SMILES$ or SMILES% or SMILES%
pPIa−amylase ¼ C0 þ C1ꢂ
DCW ðT, Nepoch
Þ
(3)
The weight of the index of ideality of correlation (IICweight
)
Here, $, % and & are quasi-SMILES notation for 50, 25 and
12.5 mg/mL concentration.
The optimal descriptor DCW was calculated by the follow-
ing equation:
ꢁ
was 0.2. The optimization Dstart step was 0.1 CW(SA). Here,
CW(SA) is the correlation weight of the structural attribute
(SA) at the start. The number of probes of optimization
was 3.
X
X
ð
Þ
ð
Þ
SMILESDCWðT,
¼
CW SK
þ
CW SSK
Nepoch
Þ
X
ð
Þ
ð
Þ
þ
CWðSSSKÞ þ CW BOND þ CW HALO
8.1.2. Validation
The prime goal of the developed QSAR model is to form a
robust model that is capable of predicting the properties of
a novel molecule in a purposeful, authentic and accur-
ate way.
ð
Þ
ð
Þ
Þ
þ CW NOSP þ CW PAIR þ CWðCmaxÞ þ CWðNmax
Þ
þ CWðOmaxÞ þ CWðSmax
(4)
The robustness and reliability of a developed QSAR model
can be decided based on the following methods: (a) cross-
validation or internal validation with the training set data; (b)
external validation reckon with the test set data and (c) data
randomization or Y-scrambling. The Monte Carlo based QSAR
Here, CW(X) denotes the correlation weight for the SMILES
and quasi-SMILES attributes. T and Nepoch denote threshold
and number of epoch, respectively. The threshold ‘T’ is
applied to classify the molecular attributes into two classes
(i) rare (noise) and (ii) active. A detailed description of global
SMILES attributes is outlined in Table 4 (Ahmadi et al., 2020; models are validated by successfully implementing these
Toropova & Toropov, 2019a).
methods. The certain statistical parameters such as correl-
ation coefficient (r2), concordance correlation coefficient
(CCC), cross-validated correlation coefficient (q2), the correl-
ation coefficient for external data (r2ext), Q2F1, Q2F2, Q2F3, val-
idation metrics, mean absolute error (MAE), Fischer ratio (F)
and novel metrics (Rm2 and MAE based metric) were used to
validate the developed QSAR models (Hossain & Roy, 2018;
Khan et al., 2019a, 2019b; Kumar et al., 2019c; Ojha & Roy,
2018; Roy, 2015). These parameters were also compared with
IIC to review the predictive potential of developed QSAR
models (Ahmadi, 2020; Kumar et al., 2019a).
Balance of correlation (TF1) and balance of correlation
with IIC (TF2) are the two target functions in Monte Carlo
Optimization used for QSAR building (Toropova et al., 2015;
Veselinovic et al., 2015). In the literature, the role of IIC has
been suggested as a novel yardstick to predict the predictive
potential of developed QSAR models
TF1 ¼ R training þ RinvTraining − │Rtraining − RinvTraining│ ꢂ Const
(5)
(6)
TF2 ¼ TF1 þ IICSET ꢂ WIIC
The Rtraining and RinvTraining are correlation coefficients for
the training and invisible training sets, respectively. The empir-
ical constant (Const) and weight of IIC (WIIC) are usually con-
stant (Ahmadi et al., 2019; Toropova et al., 2015; Veselinovic
et al., 2015). The following equation is implemented to com-
pute the index of IIC of a particular set (Kumar et al., 2019a;
Toropov & Toropova, 2018; Toropova & Toropov, 2017;
Toropova & Toropov, 2019a; Toropova et al., 2019)
8.1.3. Applicability domain
The parameters or descriptions used in the development of
QSAR models must be explanatory. Mechanistic elucidation
of established QSAR models assists in the comprehension of
the effect of descriptors in the predictive activity. The AD
analysis allows us to explore whether the constructed QSAR
model can be applied to any set of molecules. In the formu-
lation of the QSAR model, the AD of molecules executes a
decisive role in assessing the uncertainty in the prediction of
a specific molecule in terms of how alike it is to the mole-
cules used to formulate the model. Therefore, the prediction
of a modelled endpoint using QSAR is only significant when
the molecule being predicted falls within the AD of the
developed QSAR model. Therefore, AD can be expressed as
physicochemical, biological or structural space information,
based on which a training set of the model is constructed,
and the model is applied to make predictions for new mole-
cules within the specific domains (Kumar et al., 2019a). In
−
minð MAESET , þMAESET
Þ
Þ
IICSET ¼ RSET
Here, the
ꢂ
(7)
−
maxð MAESET , þMAESET
R
is the correlation coefficient value
SET
between the observed and predicted endpoint of a typical
set. MAE is mean absolute error which can be computed by
the following mathematical equation:
−N
X
1
−
−MAEset ¼
jDkjDk< 0, N is the number of Dk< 0
−N k¼1
(8)

JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS
19
the QSAR model built by CORAL Software, AD is calculated OPLS2005 force field was used for this simulation study.
by arranging SMILES attributes in the training and calibration
sets. If a compound falls outside the range of AD, it is
labelled as an outlier. In the CORAL QSAR model, the AD is
defined in consonance with the distribution of SMILES char-
acteristics in training and calibration sets as two steps:
Step 1: The definition of statistical defects (d(FK)) for each
of the SMILES attributes included to construct the model:
Orthorhombic periodic boundary condition for 10 Å buffer
region was used for solvation of molecular system with crys-
tallographic water (TIP3P) (Jorgensen et al., 1983) and the
system was neutralised by adding Naþ as counter ions. An
ensemble (NPT) of Nose–Hoover thermostat (Martyna et al.,
1992, 1994) and barostat was applied to maintain the con-
stant temperature (300 K) and pressure (1 bar) of the systems,
respectively.
A
limited memory algorithm (Broyden–
PTðAKÞ−PCðAKÞ
dðFKÞ ¼
(11)
Fletcher–Goldfarb–Shanno (LBFGS)) was employed with con-
vergence threshold gradient of 1 kcal/mol/Å for energy mini-
mization. The data were collected for every 100 ps, and the
obtained trajectory was analyzed with Desmond interphase.
NTðAKÞ þ NCðAKÞ
where PT(AK) and PC(AK) are probabilities of attributes AK in
training and calibration set, respectively; NT(AK) and NC(AK)
are the frequency of attributes AK in the training set and cali-
bration sets, respectively.
Step 2: the calculation for all substances the statistical
SMILES defect (Dj):
Acknowledgements
The authors are thankful to Dr. Andrey A. Toropov and Dr. Alla P.
Toropova for providing CORAL software. The authors are also thankful to
their respective universities for providing the infrastructure. Financial
support as senior research fellowship (SRF) to M. Duhan by Haryana
State Council of Science and Technology is acknowledged for accom-
plishing this work.
NA
X
Dj ¼
dðFKÞ
(12)
K¼1
where NA is the number of non-blocked SMILES attributes in
the SMILES.
In the current statistical calculation, a compound falls in
AD if
Author contributions
Dj < 2 ꢂ DefectNS
(13) Authors have done equivalent contributions to this work.
Here, DefectNS is the average of the statistical SMILES
defect for the training set.
Disclosure statement
The authors reported no potential conflict of interest.
8.2. Docking studies
Funding
Marvin sketch was used for preparing the optimized 3D
structure of compounds 7n. The protein data bank was
assessed for the PDB structure of a-amylase for A. oryzae
(PDB ID: 7TAA) (http://www.rcsb.org/pdb). The protein was
prepared by using UCSF Chimera 1.10 (Pettersen et al., 2004)
in which co-crystallized ligand and solvent molecules were
removed to avoid interference in binding interactions.
Missing side-chain gaps were filled using Dun Brack Rotamer
Library (Dunbrack, 2002). Gasteiger charges were calculated
using AMBERf14SB and antechamber (Wang et al., 2006) and
hydrogens were added. The docking studies were performed
using Auto Dock Vina 1.1.2 (Trott & Olson, 2010). Grid center
with following size center_x ¼ 38.1433640994, center_-
y ¼ 39.1685534078, center_z ¼ 31.0477751774, size_x ¼ 25.0,
size_y ¼ 25.0 and size_z ¼ 25.0 was placed on the active site.
The results of docking studies were analysed using
Desmond interface.
Meenakshi Duhan thanks Haryana State Council of Science and
Technology (HSCST) for the award of the research fellowship.
ORCID
Parvin Kumar
http://orcid.org/0000-0002-2635-6465
References
Achary, P. G. R., Toropova, A. P., & Toropov, A. A. (2019). Combinations
of graph invariants and attributes of simplified molecular input-line
entry system (SMILES) to build up models for sweetness. Food
Research International (Ottawa, Ont.), 122, 40–46. https://doi.org/10.
1016/j.foodres.2019.03.067
Ahmadi, S. (2020). Mathematical modeling of cytotoxicity of metal oxide
nanoparticles using the index of ideality correlation criteria.
Chemosphere, 242, 125192. https://doi.org/10.1016/j.chemosphere.
2019.125192
Ahmadi, S., Ghanbari, H., Lotfi, S., & Azimi, N. (2020). Predictive QSAR
modeling for the antioxidant activity of natural compounds deriva-
tives based on Monte Carlo method. Molecular Diversity. https://doi.
org/10.1007/s11030-019-10026-9
8.3. MD simulations
The molecular dynamics simulation of the docked complex
of 7TAA.pdb with 7n was performed for 100 ns using
€
Desmond module of Schrodinger 2019-4 to establish the sta-
Ahmadi, S., Mehrabi, M., Rezaei, S.,
&
Mardafkan, N. (2019).
bility of the docked complex (Guo et al., 2010). The docked
poses of protein ligand complexes were used as input struc-
tures and each complex was prepared by system setup
option in Desmond module. Explict solvent system with
Structure–activity relationship of the radical scavenging activities of
some natural antioxidants based on the graph of atomic orbitals.
Journal of Molecular Structure, 1191, 165–174. https://doi.org/10.1016/
j.molstruc.2019.04.103

20
M. DUHAN ET AL.
Angelova, V., Karabeliov, V., Andreeva-Gateva, P. A., & Tchekalarova, J. (2016).
Recent developments of hydrazide/hydrazone derivatives and their analogs
as anticonvulsant agents in animal models. Drug Development Research,
77(7), 379–392. https://doi.org/10.1002/ddr.21329
Bhatia, R., Kadyan, K., Duhan, M., Devi, M., Singh, R., Kamboj, R. C., &
Kumar, P. (2019). A Serendipitous synthesis: SiO2–HNO3 mediated oxi-
dative aromatization and regioselective nitration of 1,3,5-trisubsti-
tuted-4,5-dihydro-1H-pyrazoles. ChemistrySelect, 4(35), 10417–10424.
https://doi.org/10.1002/slct.201902285
Bhutani, R., Pathak, D. P., Kapoor, G., Husain, A., Kant, R., & Iqbal, M. A.
(2018). Synthesis, molecular modelling studies and ADME prediction
of benzothiazole clubbed oxadiazole-Mannich bases, and evaluation
of their anti-diabetic activity through in vivo model. Bioorganic
Chemistry, 77, 6–15. https://doi.org/10.1016/j.bioorg.2017.12.037
Brayer, G. D., Luo, Y., & Withers, S. G. (1995). The structure of human
pancreatic alpha-amylase at 1.8 A resolution and comparisons with
related enzymes . Protein Science: A Publication of the Protein Society,
4(9), 1730–1742. https://doi.org/10.1002/pro.5560040908
surface methodology by dinitrosalicylic acid method. International
Food Research Journal, 23(1), 10–17.
Khan, K., Benfenati, E., & Roy, K. (2019a). Consensus QSAR modeling of
toxicity of pharmaceuticals to different aquatic organisms: Ranking
and prioritization of the DrugBank database compounds.
Ecotoxicology and Environmental Safety, 168, 287–297. https://doi.org/
10.1016/j.ecoenv.2018.10.060
Khan, P. M., Roy, K., & Benfenati, E. (2019b). Chemometric modeling of
Daphnia magna toxicity of agrochemicals. Chemosphere, 224,
470–479. https://doi.org/10.1016/j.chemosphere.2019.02.147
Krallinger, M., Rabal, O., Lourenco, A., Oyarzabal, J., & Valencia, A. (2017).
Information retrieval and text mining technologies for chemistry.
Chemical Reviews, 117(12), 7673–7761. https://doi.org/10.1021/acs.
chemrev.6b00851
Kumar, A., & Chauhan, S. (2017). Use of the Monte Carlo method for
OECD principles-guided QSAR modeling of SIRT1 inhibitors. Archiv der
Pharmazie (Weinheim), 350(1), e1600268. https://doi.org/10.1002/ardp.
201600268
Kumar, P., Bhatia, R., Kumar, D., Kamboj, R. C., Kumar, S., Kamal, R., &
Kumar, R. (2015). An economic, simple and convenient synthesis of 2-
aryl/heteroaryl/styryl/alkylbenzothiazoles using SiO2–HNO3. Research
on Chemical Intermediates, 41(7), 4283–4292. https://doi.org/10.1007/
s11164-013-1529-x
Cardoso, L. N. F., Nogueira, T. C. M., Rodrigues, F. A. R., Oliveira, A. C. A.,
dos Santos Luciano, M. C., Pessoa, C., & de Souza, M. V. N. (2017). N-
acylhydrazones containing thiophene nucleus: A new anticancer class.
Medicinal Chemistry Research, 26(8), 1605–1608. https://doi.org/10.
1007/s00044-017-1832-y
Kumar, P., Duhan, M., Kadyan, K., Bhardwaj, J. K., Saraf, P., & Mittal, M.
(2018). Multicomponent synthesis of some molecular hybrid contain-
ing thiazole pyrazole as apoptosis inducer. Drug Research, 68(2),
72–79. https://doi.org/10.1055/s-0043-116947
Kumar, P., Duhan, M., Kadyan, K., Sindhu, J., Kumar, S., & Sharma, H.
(2017a). Synthesis of novel inhibitors of a-amylase based on the thia-
zolidine-4-one skeleton containing a pyrazole moiety and their con-
figurational studies. MedChemComm, 8(7), 1468–1476. https://doi.org/
10.1039/c7md00080d
Kumar, P., Kadyan, K., Duhan, M., Sindhu, J., Singh, V., & Saharan, B. S.
(2017b). Design, synthesis, conformational and molecular docking
study of some novel acyl hydrazone based molecular hybrids as anti-
malarial and antimicrobial agents. Chemistry Central Journal, 11(1),
115. https://doi.org/10.1186/s13065-017-0344-7
Kumar, P., & Kumar, A. (2019). Nucleobase sequence based building up
of reliable QSAR models with the index of ideality correlation using
Monte Carlo method. Journal of Biomolecular Structure and Dynamics,
38(11), 3296–3306. https://doi.org/10.1080/07391102.2019.1656109
Kumar, P., & Kumar, A. (2020). CORAL: QSAR models of CB1 cannabinoid
receptor inhibitors based on local and global SMILES attributes with
the index of ideality of correlation and the correlation contradiction
index. Chemometrics and Intelligent Laboratory Systems, 200, 103982.
https://doi.org/10.1016/j.chemolab.2020.103982
Duhan, M., Singh, R., Devi, M., Sindhu, J., Bhatia, R., Kumar, A., & Kumar,
P. (2019). Synthesis, molecular docking and QSAR study of thiazole
clubbed pyrazole hybrid as a-amylase inhibitor. Journal of
Biomolecular Structure and Dynamics. 1–31. https://doi.org/10.1080/
07391102.2019.1704885
Dunbrack, R. L. Jr. (2002). Rotamer libraries in the 21st century. Current
Opinion in Structural Biology, 12(4), 431–440. https://doi.org/10.1016/
S0959-440X(02)00344-5
Ferreira, L. G., Dos Santos, R. N., Oliva, G., & Andricopulo, A. D. (2015).
Molecular docking and structure-based drug design strategies.
Molecules (Basel, Switzerland), 20(7), 13384–13421. https://doi.org/10.
3390/molecules200713384
Gollapalli, M., Taha, M., Javid, M. T., Almandil, N. B., Rahim, F., Wadood,
A., Mosaddik, A., Ibrahim, M., Alqahtani, M. A., & Bamarouf, Y. A.
(2019). Synthesis of benzothiazole derivatives as a potent a-glucosi-
dase inhibitor. Bioorganic Chemistry, 85, 33–48. https://doi.org/10.
1016/j.bioorg.2018.12.021
Guo, Z., Mohanty, U., Noehre, J., Sawyer, T. K., Sherman, W., & Krilov, G.
(2010). Probing the alpha-helical structural stability of stapled p53
peptides: Molecular dynamics simulations and analysis. Chemical
Biology & Drug Design, 75(4), 348–359. https://doi.org/10.1111/j.1747-
0285.2010.00951 https://doi.org/10.1111/j.1747-0285.2010.00951.x
ꢀ
ꢀ
ꢀ
Hernandez-Vazquez, E., Salgado-Barrera, S., Ramırez-Espinosa, J. J.,
ꢀ
Kumar, P., Kumar, A., & Sindhu, J. (2019a). Design and development of
novel focal adhesion kinase (FAK) inhibitors using Monte Carlo
method with index of ideality of correlation to validate QSAR. SAR
and QSAR in Environmental Research, 30(2), 63–80. https://doi.org/10.
1080/1062936X.2018.1564067
Estrada-Soto, S., & Hernandez-Luis, F. (2016). Synthesis and molecular
docking of N0-arylidene-5-(4-chlorophenyl)-1-(3,4-dichlorophenyl)-4-
methyl-1H-pyrazole-3-carbohydrazides as novel hypoglycemic and
antioxidant dual agents . Bioorganic & Medicinal Chemistry, 24(10),
2298–2306. https://doi.org/10.1016/j.bmc.2016.04.007
Kumar, P., Kumar, A., & Sindhu, J. (2019b). In silico design of diacylgly-
cerol acyltransferase-1 (DGAT1) inhibitors based on SMILES descriptors
using Monte-Carlo method. SAR and QSAR in Environmental Research,
30(8), 525–541. https://doi.org/10.1080/1062936X.2019.1629998
Kumar, P., Kumar, A., Sindhu, J., & Lal, S. (2019c). QSAR models for nitro-
gen containing monophosphonate and bisphosphonate derivatives as
human farnesyl pyrophosphate synthase inhibitors based on Monte
Carlo method. Drug Research, 69(3), 159–167. https://doi.org/10.1055/
a-0652-5290
Hossain, K. A., & Roy, K. (2018). Chemometric modeling of aquatic tox-
icity of contaminants of emerging concern (CECs) in Dugesia japonica
and its interspecies correlation with daphnia and fish: QSTR and
QSTTR approaches. Ecotoxicology and Environmental Safety, 166,
92–101. https://doi.org/10.1016/j.ecoenv.2018.09.068
Jorgensen, W. L., Chandrasekhar, J., Madura, J. D., Impey, R. W., & Klein,
M. L. (1983). Comparison of simple potential functions for simulating
liquid water. The Journal of Chemical Physics, 79(2), 926–935. https://
doi.org/10.1063/1.445869
Kumar, S., Rathore, D. S., Garg, G., Khatri, K., Saxena, R., & Sahu, S. K.
(2017c). Synthesis and evaluation of some benzothiazole derivatives
as antidiabetic agents. International Journal of Pharmacy and
Pharmaceutical Sciences, 9(2), 60–68. https://doi.org/10.22159/ijpps.
2017v9i2.14359
Karelson, M., Lobanov, V. S., & Katritzky, A. R. (1996). Quantum-chemical
descriptors in QSAR/QSPR Studies. Chemical Reviews, 96(3),
1027–1044. https://doi.org/10.1021/cr950202r
Kaur, I., Khajuria, A., Ohri, P., Kaur, P., & Singh, K. (2018). Benzothiazole based
Schiff-base-A mechanistically discrete sensor for HSO4 and I−: Application
−
Le, T., Epa, V. C., Burden, F. R., & Winkler, D. A. (2012). Quantitative struc-
ture–property relationship modeling of diverse materials properties.
Chemical Reviews, 112(5), 2889–2919. https://doi.org/10.1021/
cr200066h
to bioimaging and vapour phase sensing of ethyl acetate. Sensors and
Actuators B: Chemical, 268, 29–38. https://doi.org/10.1016/j.snb.2018.04.072
Keharom, S., Mahachai, R., & Chanthai, S. (2016). The optimization study
of a-amylase activity based on central composite design–response

JOURNAL OF BIOMOLECULAR STRUCTURE AND DYNAMICS
21
Macalino, S. J. Y., Gosu, V., Hong, S., & Choi, S. (2015). Role of computer-
aided drug design in modern drug discovery. Archives of Pharmacal
Research, 38(9), 1686–1701. https://doi.org/10.1007/s12272-015-0640-5
containing benzohydrazide moiety and their molecular docking stud-
ies. European Journal of Medicinal Chemistry, 92, 387–400. https://doi.
org/10.1016/j.ejmech.2015.01.009
Manisha, Chauhan, S., Kumar, P., & Kumar, A. (2019). Development of Taha, M., Shah, S. A. A., Imran, S., Afifi, M., Chigurupati, S., Selvaraj, M.,
prediction model for fructose-1,6-bisphosphatase inhibitors using the
Monte Carlo method. SAR and QSAR in Environmental Research, 30(3),
145–159. https://doi.org/10.1080/1062936X.2019.1568299
Rahim, F., Ullah, H., Zaman, K., & Vijayabalan, S. (2017). Synthesis and
in vitro study of benzofuran hydrazone derivatives as novel alpha-
amylase inhibitor. Bioorganic Chemistry, 75, 78–85. https://doi.org/10.
1016/j.bioorg.2017.09.002
Martyna, G. J., Klein, M. L., & Tuckerman, M. (1992). Nose–Hoover chains
– The canonical ensemble via continuous dynamics. The Journal of Thota, S., Rodrigues, D. A., Pinheiro, P. D. S. M., Lima, L. M., Fraga, C. A.,
Chemical Physics, 97(4), 2635–2643. https://doi.org/10.1063/1.463940
Martyna, G. J., Tobias, D. J., & Klein, M. L. (1994). Constant-pressure
molecular dynamics algorithms. The Journal of Chemical Physics,
101(5), 4177–4189. https://doi.org/10.1063/1.467468
Mishra, V. R., Ghanavatkar, C. W., Mali, S. N., Qureshi, S. I., Chaudhari,
H. K., & Sekar, N. (2019). Design, synthesis, antimicrobial activity and
computational studies of novel azo linked substituted benzimidazole,
& Barreiro, E. J. (2018). N-acylhydrazones as drugs. Bioorganic &
Medicinal Chemistry Letters, 28(17), 2797–2806. https://doi.org/10.
1016/j.bmcl.2018.07.015
Toropova, A. P., & Toropov, A. A. (2017). The index of ideality of correl-
ation: A criterion of predictability of QSAR models for skin permeabil-
ity? The Science of the Total Environment, 586, 466–472. https://doi.
org/10.1016/j.scitotenv.2017.01.198
benzoxazole and benzothiazole derivatives. Computational Biology Toropova, A. P., & Toropov, A. A. (2019a). QSPR and nano-QSPR: What is
and Chemistry, 78, 330–337. https://doi.org/10.1016/j.compbiolchem.
the difference? Journal of Molecular Structure, 1182, 141–149. https://
2019.01.003
doi.org/10.1016/j.molstruc.2019.01.040
Murtuja, S., Shaquiquzzaman, M., & Amir, M. (2018). Design, synthesis, Toropova, A. P., & Toropov, A. A. (2019b). Does the index of ideality of
and screening of hybrid benzothiazolyl-oxadiazoles as anticonvulsant
agents. Letters in Drug Design & Discovery, 15(4), 398–405. https://doi.
org/10.2174/1570180814666170526154914
Nickavar, B., & Yousefian, N. (2010). Inhibitory effects of six allium species
on a-Amylase enzyme activity. Iranian Journal of Pharmaceutical
Research, 8, 53–57.
Nimbhal, M., Bagri, K., Kumar, P., & Kumar, A. (2020). The index of ideality
of correlation: A statistical yardstick for better QSAR modeling of glu-
cokinase activators. Structural Chemistry, 31(2), 831–839. https://doi.
org/10.1007/s11224-019-01468-w
correlation detect the better model correctly? Molecular Informatics,
38(8–9), 1800157. https://doi.org/10.1002/minf.201800157
Toropova, A. P., & Toropov, A. A. (2019c). The index of ideality of correlation:
Improvement of models for toxicity to algae. Natural Product Research,
33(15), 2200–2207. https://doi.org/10.1080/14786419.2018.1493591
Toropova, A. P., Toropov, A. A., Beeg, M., Gobbi, M., & Salmona, M.
(2017). Utilization of the Monte Carlo method to build up QSAR mod-
els for hemolysis and cytotoxicity of antimicrobial peptides. Current
Drug Discovery Technologies, 14(4), 229–243. https://doi.org/10.2174/
1570163814666170525114128
Ojha, P. K., & Roy, K. (2011). Comparative QSARs for antimalarial endo- Toropova, A. P., Toropov, A. A., Benfenati, E., Leszczynska, D.,
&
chins: Importance of descriptor-thinning and noise reduction prior to
feature selection. Chemometrics and Intelligent Laborary Systems,
109(2), 146–161. https://doi.org/10.1016/j.chemolab.2011.08.007
Leszczynski, J. (2015). QSAR model as a random event: A case of rat
toxicity. Bioorganic & Medicinal Chemistry, 23(6), 1223–1230. https://
doi.org/10.1016/j.bmc.2015.01.055
Ojha, P. K., & Roy, K. (2018). Development of a robust and validated 2D- Toropova, A. P., Toropov, A. A., Veselinovic, A. M., Veselinovic, J. B.,
QSPR model for sweetness potency of diverse functional organic mol-
ecules. Food and Chemical Toxicology: An International Journal
Published for the British Industrial Biological Research Association, 112,
551–562. https://doi.org/10.1016/j.fct.2017.03.043
Leszczynska, D., & Leszczynski, J. (2019). Semi-correlations combined
with the index of ideality of correlation: A tool to build up model of
mutagenic potential. Molecular and Cellular Biochemistry, 452(1–2),
133–140. https://doi.org/10.1007/s11010-018-3419-4
Pettersen, E. F., Goddard, T. D., Huang, C. C., Couch, G. S., Greenblatt, Toropov, A. A., & Toropova, A. P. (2018). Predicting cytotoxicity of 2-phe-
D. M., Meng, E. C., & Ferrin, T. E. (2004). UCSF Chimera – A visualiza-
tion system for exploratory research and analysis. Journal of
Computational Chemistry, 25(13), 1605–1612. https://doi.org/10.1002/
jcc.20084
Roy, K. (2015). Application of chemometrics and cheminformatics in antimalarial
drug research. Combinatorial Chemistry & High Throughput Screening, 18(2),
89–90. https://doi.org/10.2174/138620731802150215154014
nylindole derivatives against breast cancer cells using index of ideality
of correlation. Anticancer Research, 38(11), 6189–6194. https://doi.org/
10.21873/anticanres.12972
Toropov, A. A., & Toropova, A. P. (2019). Use of the index of ideality of
correlation to improve predictive potential for biochemical endpoints.
Toxicology Mechanisms and Methods, 29(1), 43–52. https://doi.org/10.
1080/15376516.2018.1506851
Sharma, A. K., Sharma, R., & Gangwal, A. (2018). Surface tension studies of Toropov, A. A., Toropova, A. P., & Benfenati, E. (2019). The index of ideal-
ternary system: Cu(II) surfactants-2-amino-6-methyl benzothiazole com-
plex plus methanol plus benzene at 311 K. Current Physical Chemistry,
8(2), 151–161. https://doi.org/10.2174/1877946808666180914164134
Shi, J., Deng, Q., Li, Y., Zheng, M., Chai, Z., Wan, C., Zheng, Z., Li, L.,
Huang, F., & Tang, B. (2018). A rapid and ultrasensitive tetraphenyl-
ethylene-based probe with aggregation-induced emission for direct
detection of a-amylase in human body fluids. Analytical Chemistry,
90(22), 13775–13782. https://doi.org/10.1021/acs.analchem.8b04244
ity of correlation: QSAR model of acute toxicity for Zebrafish (Danio
rerio) embryo. International Journal of Environmental Research, 13(2),
387–394. https://doi.org/10.1007/s41742-019-00183-y
Trott, O., & Olson, A. J. (2010). AutoDock Vina: Improving the speed and
accuracy of docking with a new scoring function, efficient optimiza-
tion, and multithreading. Journal of Computational Chemistry, 31(2),
455–461. https://doi.org/10.1002/jcc.21334
Version, M., & 6.2.2. (2014). calculation module developed by ChemAxon.
http://www.chemaxon.com/products/marvin/marvinsketch/
~
Souza, P. M. D., & Magalhaes, P. de O. e. (2010). Application of microbial
a-amylase in industry – A review. Brazilian Journal of Microbiology, Veselinovic, A. M., Milosavljevic, J. B., Toropov, A. A., & Nikolic, G. M. (2013).
41(4), 850–861. https://doi.org/10.1590/S1517-83822010000400004
Svensson, B. (1994). Protein engineering in the alpha-amylase family:
Catalytic mechanism, substrate specificity, and stability. Plant
Molecular Biology, 25(2), 141–157. https://doi.org/10.1007/BF00023233
Taha, M., Irshad, M., Imran, S., Rahim, F., Selvaraj, M., Almandil, N. B.,
Mosaddik, A., Chigurupati, S., Nawaz, F., Ismail, N. H., & Ibrahim, M.
(2019). Thiazole based carbohydrazide derivatives as a-amylase inhibi-
tor and their molecular docking study. Heteroatom Chemistry, 2019,
1–8. https://doi.org/10.1155/2019/7502347
SMILES-based QSAR model for arylpiperazines as high-affinity 5-HT(1A)
receptor ligands using CORAL . European Journal of Pharmaceutical
Sciences: Official Journal of the European Federation for Pharmaceutical
Sciences, 48(3), 532–541. https://doi.org/10.1016/j.ejps.2012.12.021
Veselinovic, J. B., Nikolic, G. M., Trutic, N. V., Zivkovic, J. V., & Veselinovic, A. M.
(2015). Monte Carlo QSAR models for predicting organophosphate inhib-
ition of acetycholinesterase. SAR and QSAR in Environmental Research, 26(6),
449–460. https://doi.org/10.1080/1062936x.2015.1049665
Wang, J., Wang, W., Kollman, P. A., & Case, D. A. (2006). Automatic atom
type and bond type perception in molecular mechanical calculations.
Journal of Molecular Graphics & Modelling, 25(2), 247–260. https://doi.
org/10.1016/j.jmgm.2005.12.005
Taha, M., Ismail, N. H., Lalani, S., Fatmi, M. Q., Atia Tul, W., Siddiqui, S.,
Khan, K. M., Imran, S., & Choudhary, M. I. (2015). Synthesis of novel
inhibitors of a-glucosidase based on the benzothiazole skeleton