Molecular docking, 3D-QSAR, molecular dynamics, synthesis and anticancer activity of tyrosine kinase 2 (TYK 2) inhibitors

Article abstract of DOI:10.1080/10799893.2019.1585453

A therapeutic rationale is proposed by selectively targeting tyrosine kinase 2 (TYK 2) to obtain potent TYK 2 inhibitors by molecular modeling studies. In the present study, we have taken tyrosine kinase (TYK 2) inhibitors and carried out molecular docking, 3 D quantitative structure–activity relationship (3D-QSAR) analysis and molecular dynamics (MD). Based on the 3D-QSAR results thirteen new compounds (R-1 to R-13) were designed and synthesized in good yields. The synthesized molecules were evaluated for their in vitro anticancer activity against LnCap and A549 cell lines. The molecules R-1, R-3, R-5, R-7, and R-10 exhibited considerable anti cancer activity.

Full text of DOI:10.1080/10799893.2019.1585453

Journal of Receptors and Signal Transduction
ISSN: 1079-9893 (Print) 1532-4281 (Online) Journal homepage: https://www.tandfonline.com/loi/irst20
Molecular docking, 3D-QSAR, molecular dynamics,
synthesis and anticancer activity of tyrosine
kinase 2 (TYK 2) inhibitors
Ramesh Itteboina, Srilata Ballu, Sree Kanth Sivan & Vijjulatha Manga
To cite this article: Ramesh Itteboina, Srilata Ballu, Sree Kanth Sivan & Vijjulatha Manga (2018)
Molecular docking, 3D-QSAR, molecular dynamics, synthesis and anticancer activity of tyrosine
kinase 2 (TYK 2) inhibitors, Journal of Receptors and Signal Transduction, 38:5-6, 462-474, DOI:
10.1080/10799893.2019.1585453
To link to this article: https://doi.org/10.1080/10799893.2019.1585453
View supplementary material
Published online: 30 Apr 2019.
Submit your article to this journal
Article views: 6
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=irst20

JOURNAL OF RECEPTORS AND SIGNAL TRANSDUCTION
2019, VOL. 38, NOS. 5-6, 462–474
https://doi.org/10.1080/10799893.2019.1585453
RESEARCH ARTICLE
Molecular docking, 3D-QSAR, molecular dynamics, synthesis and anticancer
activity of tyrosine kinase 2 (TYK 2) inhibitors
Ramesh Itteboina, Srilata Ballu, Sree Kanth Sivan and Vijjulatha Manga
Molecular Modeling and Medicinal Chemistry Group, Department of Chemistry, University College of Science, Osmania University,
Hyderabad, Telangana State, India
ABSTRACT
ARTICLE HISTORY
Received 5 October 2018
Revised 22 January 2019
Accepted 22 January 2019
A therapeutic rationale is proposed by selectively targeting tyrosine kinase 2 (TYK 2) to obtain potent
TYK 2 inhibitors by molecular modeling studies. In the present study, we have taken tyrosine kinase
(TYK 2) inhibitors and carried out molecular docking, 3 D quantitative structure–activity relationship
(3D-QSAR) analysis and molecular dynamics (MD). Based on the 3D-QSAR results thirteen new com-
pounds (R-1 to R-13) were designed and synthesized in good yields. The synthesized molecules were
evaluated for their in vitro anticancer activity against LnCap and A549 cell lines. The molecules R-1, R-
3, R-5, R-7, and R-10 exhibited considerable anti cancer activity.
KEYWORDS
CoMFA (Comparative
molecular field analysis);
CoMSIA (Comparative
molecular similarity indices
analysis); PLS (partial least
square) analysis; TYK
2(Tyrosine kinase 2)
1. Introduction
were also employed to elucidate the probable binding
modes of these antagonists at the binding site.
A wide range of biological functions is mediated by cyto-
kine-signaling pathways, which include several aspects of
inflammation and immunity [1–3]. Types I and II cytokine
receptors do not show kinase activity and instead transmit
their signals through the receptor-associated Janus kinases
(JAK1, JAK2, JAK3, and TYK2) [4,5]. Cytokine receptor com-
plexes activate specific combinations of JAKs in well-defined
patterns upon stimulation [6], leading to further activation of
signal transducer and activator of transcription (STAT) pro-
teins that are located in the cytoplasm. STATs upon tyrosine
phophorylation that is mediated by JAK undergoes dimeriza-
2. Material and methods
2.1. Molecular docking and 3 D QSAR
The crystal structure of TYK 2 (PDB id: 4GJ2) [21] was down-
loaded from the protein databank. GLIDE 5.6 [22] was used
for molecular was used for molecular docking studies. The
protein was prepared using protein preparation module
applying the default parameters; a grid was generated
around the active site of the TYK 2 with receptor van der
tion and translocated to the nucleus and regulate transcrip- Waals scaling for the nonpolar atoms as 0.9 [23]. A set of 54
tion of specific target genes [7–9]. The importance of the known TYK 2 inhibitors (Table 1) was selected from the lit-
JAK/STAT pathways in cytokine signaling, targeting of JAK erature [21,24]. These were built using Maestro build panel
€
kinases is envisioned to be useful in the treatment of a var- and prepared by Lig prep application in the Schrodinger
iety of diseases including rheumatoid arthritis (RA) [10], mye- 2010 suite. The molecular docking of the 54 molecules into
loproliferative disorders [11,12] and cancer [13].
the generated grid was performed by using the extra – pre-
cision (XP) docking mode [23]. The co-crystallized ligand was
redocked and its root mean square deviation (RMSD) was
calculated to validate the docking process. The dock poses
of all the molecules were analyzed for their hydrogen bond
interaction with the active site residue. The best dock pose
for each molecule was selected for CoMFA and CoMSIA ana-
Computational drug design approaches were employed in
the development and optimization of inhibitors. The molecu-
lar interactions of TYK 2 inhibitors with the receptor were
studied to design novel molecules with enhanced anticancer
activity. Our prime objective was to study the structural
requirements for TYK 2 inhibitors and to synthesize new mol-
ecules. In the present article, we report receptor based 3D- lysis without further alignment, docked based alignment is
QSAR [13–17] studies using CoMFA [18,19] and CoMSIA [20] shown in Figure 1. The molecules were imported into SYBYL-
on 54 N-(pyridin-4-yl) benzamide Scaffold derivatives. X 2.1 [25] Tripos Associates, St. Louis (MO) molecular model-
Docking analysis and molecular dynamics (MD) simulations ing program package, Gasteiger-Huckel charges [26] were
CONTACT Vijjulatha Manga
vijjulathamanga@gmail.com
Department of Chemistry, University College of Science, Osmania University, Hyderabad –
07, India
Supplemental data for this article can be accessed here.
ß 2019 Informa UK Limited, trading as Taylor & Francis Group

JOURNAL OF RECEPTORS AND SIGNAL TRANSDUCTION
463
Table 1. Structure of the molecules.
R
Cl
Cl
O
NH
O
N
NH
Pred.pKi
(CoMFA)
Pred.pKi
(CoMSIA)
Glide score
(K.cal/mol)
Molecule
R
Exp. pKi
1
NH2
NHAc
OH
OCH3
CH2OH
9.000
8.553
8.509
8.108
8.268
8.923
8.515
8.717
8.229
8.279
8.885
8.349
8.533
8.218
8.423
ꢀ9.861
ꢀ8.564
ꢀ10.009
ꢀ9.365
ꢀ8.531
2
3t
4t
5
6t
8.119
8.228
8.218
ꢀ9.803
OH
H3C
7t
8
7.661
7.987
8.773
8.077
7.984
7.568
ꢀ9.369
OH
CH3
H3C
ꢀ10.308
OH
9
CH2NH2
8.356
8.481
7.216
8.108
8.745
8.321
8.506
7.165
8.152
8.731
8.137
8.137
7.634
7.915
8.608
ꢀ10.009
ꢀ9.884
ꢀ9.676
ꢀ8.395
ꢀ9.093
10t
11
12
13
CH2NHCH3
CH2N(CH3)2
COCH3
CN
14t
8.022
8.468
7.990
8.668
7.676
8.295
ꢀ9.737
CH
15
ꢀ7.806
N
N
H
16t
H
N
7.962
8.503
8.245
ꢀ8.877
N
N
Cl
O
Cl
NH
O
R
N
NH
Pred.pKi
(CoMFA)
Pred.pKi
(CoMSIA)
Glide score
(K.cal/mol)
Molecule
17
R
Exp. pKi
8.602
8.326
8.067
8.211
8.157
ꢀ8.128
ꢀ9.346
F
18
7.943
F
(continued)

464
R. ITTEBOINA ET AL.
Table 1. Continued.
Pred.pKi
(CoMFA)
Pred.pKi
(CoMSIA)
Glide score
(K.cal/mol)
Molecule
19
R
Exp. pKi
8.292
8.164
8.114
ꢀ9.929
ꢀ9.923
Cl
CH3
OH
20
8.070
8.056
8.014
21
22
7.835
8.041
7.681
8.279
7.824
8.500
ꢀ9.952
ꢀ9.440
N
23t
24
25
7.955
8.495
8.886
7.858
8.495
9.009
8.245
8.451
9.013
ꢀ9.930
NH2
O
ꢀ4.618
OH
O
ꢀ10.184
N
NH
N
N
26
7.824
7.787
7.885
ꢀ4.313
N
N
N
H
R2
R1
Cl
O
NH
O
F
N
NH
Pred.pKi
(CoMFA)
Pred.pKi
(CoMSIA)
Glide score
(K.cal/mol)
Molecule
R1
R2
Exp. pKI
27t
Cl
F
CN
CN
8.854
8.796
8.382
8.416
8.472
8.663
ꢀ7.876
ꢀ7.835
28t
R2
R1
O
NH
N
Pred.pKi
(CoMFA)
Pred.pKi
(CoMSIA)
Glide score
(K.cal/mol)
Molecule
R1
R2
Exp. pKi
29
30
31
32t
33
34
35
36
37
38
H
Cl
CH3
F
Cl
Cl
Cl
Cl
Cl
Cl
6-H
6-H
6-H
5.509
5.959
5.619
6.267
5.553
5.721
6.102
6.046
5.553
5.921
5.562
5.909
5.530
5.826
5.591
5.737
5.766
6.250
5.669
5.794
5.548
5.705
5.677
5.949
5.789
5.981
5.879
5.942
5.796
5.684
ꢀ3.733
ꢀ4.503
ꢀ6.577
ꢀ5.377
ꢀ7.329
ꢀ5.183
ꢀ6.027
ꢀ6.211
ꢀ5.402
ꢀ6.912
F
6-NH2
6-CF3
5-Cl
5-CF3
5-OCH3
4-Cl
(continued)

JOURNAL OF RECEPTORS AND SIGNAL TRANSDUCTION
465
Table 1. Continued.
Pred.pKi
(CoMFA)
Pred.pKi
(CoMSIA)
Glide score
(K.cal/mol)
Molecule
R1
R2
Exp. pKi
39
40
Cl
Cl
4-NH2
4-OCH3
6.420
5.495
6.498
5.652
6.154
5.739
ꢀ7.873
ꢀ3.829
41
5.959
5.936
5.659
ꢀ7.356
Cl
Cl
O
HN
N
Cl
Cl
HN
O
N
R
Pred.pKi
(CoMFA)
Pred.pKi
(CoMSIA)
Glide score
(K.cal/mol)
Molecule
R
Exp. pKi
42
NH2
NHC(O)CH3
NH-Phenyl
6.031
7.018
6.854
6.053
7.289
7.624
6.233
7.906
7.275
ꢀ7.875
ꢀ9.564
ꢀ9.428
43t
44t
45t
46
47
7.538
8.309
8.000
8.369
8.168
7.979
7.762
8.014
8.476
ꢀ7.920
ꢀ9.528
ꢀ9.544
N
NH
N
N
NH
N
N
NH
48t
7.721
9.319
8.161
ꢀ9.376
CH3
N
N
NH
49
7.001
6.619
6.921
6.070
7.009
6.146
ꢀ8.114
ꢀ8.148
NH
OH
50t
Cl
Cl
O
NH
O
N
NH
R
Pred. pKi
(CoMFA)
Pred.pKi
(CoMSIA)
Glide score
(K.cal/mol)
Molecule
R
Exp. pKi
51t
52t
53
Cyclo propyl
Cyclopentyl
Phenyl
8.319
6.619
5.699
7.745
8.196
7.106
5.819
7.794
8.070
6.865
5.982
8.085
ꢀ8.694
ꢀ8.793
ꢀ8.459
ꢀ9.664
54
NHCH2CH3
t ¼ Test set molecule.

466
R. ITTEBOINA ET AL.
For MTT assay, each test compounds were weighed separ-
ately and dissolved in DMSO. With media make up the final
concentration to 1 nM/ml and the cells were treated with ser-
ies of concentrations from 10 to 100 nM/ml.
2.2.1. Cell viability by MTT assay
Cell viability was evaluated by the MTT Assay with three
independent experiments with six concentrations of com-
pounds in triplicates. Cells were trypsinized and preform the
thetryphan blue assay to know viable cells in cell suspension.
Cells were counted by hemocytometer and seeded at density
of 5.0 ꢂ 103 cells/well in 100 ll media in 96-well plate culture
medium and incubated overnight at 37ꢁC. After incubation,
take off the old media and add fresh media 100 ml with dif-
ferent concentrations of test compound in representative
wells in 96 plates. After 48 h, discard the drug solution and
add the fresh medic with MTT solution (0.5 mg/mLꢀ1) was
added to each well and plates were incubated at 37ꢁ C for
3 h. At the end of incubation time, precipitates are formed as
a result of the reduction of the MTT salt to chromophorefor-
mazan crystals by the cells with metabolically active mito-
chondria. The optical density of solubilized crystals in DMSO
was measured at 570 nm on a microplate reader. The per-
centage growth inhibition was calculated using the following
formula and concentration of test drug needed to inhibit cell
growth by 50% values is generated from the dose-response
curves for each cells using origin software.
Figure 1. Docked base alignment of TYK 2 inhibitors.
assigned. A standard protocol was followed for calculating
steric, electrostatic, hydrophobic, acceptor and donor param-
eters required for CoMFA and CoMSIA analyses using Tripos
force fields [27].
A partial least squares (PLS) regression was used to pro-
duce a linear relationship that correlates changes in the com-
puted fields with changes in the corresponding experimental
values of biological activity (pKi) for the data set of ligands.
The data set was divided into a training set consisting of 36
molecules and a test set of 18 molecules. Biological activity
values of ligands were used as dependent variables in a PLS
statistical analysis. The column filtering value (s) was set to
2.0 kcal molꢀ1 to improve the signal-to-noise ratio by omit-
ting those lattice points whose energy variation were below
this threshold. Cross-validation was performed by the leave-
one-out (LOO) procedure to determine the optimum number
of components (ONC) and the coefficient r2
.
100 Control ꢀ Treatment
ð
Þ
loo
%Inhibition ¼
An optimum number of components obtained are then
used to derive the final QSAR model using all of the training
set molecules with noncross validation and to get the con-
ventional regression coefficient (r2). To validate the CoMFA-
and CoMSIA-derived models, the predictive ability of the test
set molecules (expressed as r2pred) was determined [27].
Since the statistical parameters were found to be best for
the model from the LOO method, it was employed for fur-
ther prediction of the designed molecules. The designed
molecules were also constructed, minimized and docked into
the protein active site.
Control
2.3. Molecular dynamics simulations
Molecular dynamics (MD) simulations were performed with
Desmond 3.8 [28] to explore the conformational changes on
the binding interaction and stability of docked molecule. The
OPLS 2005 force field [29] was employed to compare the
selective binding modes of the co-crystal structure ligand
(molecule 17; PDB ID: 4GJ2) and newly synthesized molecule
R5 utilizing their 3 D structures obtained from docking
results. The coordinates for MD simulations have been
achieved from the docking results. The two complexes were
soaked using simple point charge (SPC) water molecules [30]
in an orthorhombic box of 15 Å ꢂ 15 Å ꢂ 15 Å size. The
molecular dynamics simulation methodology utilized in our
previous works has been adopted. [31,32]
2.2. Biological assay
Measurement of cell viability and proliferation forms the
basis for numerous in vitro assays of a cell population’s
response to external factors. The MTT Cell Proliferation Assay
measures the cell proliferation rate and conversely, when
metabolic events lead to apoptosis or necrosis, the reduction
in cell viability. DMEM (Dulbecco’s modified Eagles medium),
MTT [3–(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium
bromide], trypsin, EDTA phosphate-buffered saline (PBS) and
were purchased from Sigma Chemicals Co. (St. Louis, MO)
and fetal bovine serum (FBS) were purchased from Gibco.
3. Results and discussion
3.1. 3D-qsar
Accuracy of extra precision glide docking procedure
employed is determined by measuring root mean square
25-cm2 and 75-cm2 flask and 96-well plated purchased from deviation (RMSD) between best binding conformation pre-
Eppendroff India. The cancer cells were purchased from dicted by the object scoring function (Glide score) and
NCCS, Pune and the cells were maintained in MEM supple- experimental binding mode as determined by X-ray crystal-
mented with 10% FBS and the antibiotics penicillin/strepto- lography [33]. Crystal structure ligand 2, 6-dichloro-N-[2-
mycin (0.5 mlꢀ1), in atmosphere of 5% CO2/95% air at 37 ꢁC. ({[(1S,2S)-2-fluorocyclopropyl]
carbonyl}amino)pyridin-4-yl]

JOURNAL OF RECEPTORS AND SIGNAL TRANSDUCTION
467
Scotter plot of Exp.pKI vs
Scotter plot of Exp.pKI vs
Pred.pKI- CoMFA
10
9
10
9
Pred.pKI- CoMSIA
8
8
7
7
y = 0.9905x + 0.0698
R² = 0.9898
y = 0.9598x + 0.2927
R² = 0.9592
6
6
5
5
5
7
9
11
5
7
9
11
Experimental pKI
Experimental pKI
Figure 3. Scatter plot of predicted vs. experimental pKI values (test set is repre-
sented as triangles and training set represented as squares).
Figure 2. Superimposition of crystal structure pose (green) on dock pose (light
pink) of co-crystallized ligand. The RMS deviation is 1.3051 A0.
Molecular fields define the favorable or unfavorable inter-
action energies of molecules with a probe atom traversing
across the lattice grid points surrounding the molecules. The
3 D colored plots suggested the modification required to
design new molecules. The CoMFA/CoMSIA results were
graphically interpreted by field contribution maps using the
Table 2. PLS Result Summary.
Statistical parameters
CoMFA
CoMSIA
q2
0.74
36
0.71
36
18
a
loo
Number of molecules in training set
Number of molecules in test set
18
7
0.139
0.99
95.389
0.58
ꢃ
ONCb
SEEc
r2d
4
“STDEV COEFF” field type. All the contours represented
0.265
0.959
181.329
0.6
default 80% and 20% level contribution for favored and dis-
favored regions respectively. Most active molecule in data
set (molecule 1) among the series was displayed on the map
to assist in visualization.
e
Fratio
f
r2
pred
Fraction of fields contributions
Steric
49.30%
13.90%
28.40%
13.50%
23.30%
20.80%
Figure 4(a,b) shows the contour maps derived from the
CoMFA and CoMSIA steric and electrostatic fields, respect-
ively, steric and electrostatic maps of these two models were
almost similar. In steric contour, green and yellow represent
favored and disfavored regions respectively. Large green
contour is seen on the cyclo propane ring. In electrostatic
contour, red represents electro negative favored region and
blue indicates electro positive favored region, in these con-
tour red counter is seen on carbonyl group and amino group
nearer to the pyridine ring, and –NH2 group on benzene
indicates electro negative groups increases the activity.
Figure 5(a) shows the hydrophobic contour maps derived
from the CoMSIA PLS model for molecule 1. Yellow contour
indicates hydrophobic favored region and white contour
indicates hydrophilic favored region. In these contour white
contour is seen around the pyridine ring and –NH2 group on
benzene ring, indicates hydrophilic groups at these regions
increases the activity, and yellow contour is seen on –NH of
the amide group indicates hydrophobic groups at this
regions increases the activity. Figure 5(b) represents hydro-
gen bond donor and acceptor contours, where cyan and pur-
ple contour represents favorable and disfavorable hydrogen
bond donor groups respectively and magenta and red con-
tours represents favorable and unfavorable H-bond acceptor
groups respectively. In this contour, purple counter is seen
on –NH group attached to the pyridine ring indicates hydro-
gen bond donor groups increases the activity at this region.
Magenta contour is seen near to the NH2 group attached to
the benzene ring indicates H-bond acceptor groups will
increase the activity at this region.
Electrostatic
Hydrophobic
Acceptor
Donor
50.70%
–
–
–
aCross-validation correlation coefficient by leave one out method,.
bOptimum number of components.
cStandard error of estimate.
dConventional correlation coefficient.
eFisher test value.
fCross-validation correlation coefficient on test set.
benzamide was redocked into the binding site of TYK 2 and
RMSD between the docked conformation and X-ray crystallo-
graphic conformation of molecule was found to be equal to
1.3051 Å (Figure 2), this value signifies the reliability of
parameters set for Glide docking in reproducing the experi-
mentally observed binding mode for TYK 2 inhibitor.
3D-QSAR CoMFA and CoMSIA studies were carried out on
TYK 2 inhibitors. Molecules having inhibitory activity against
TYK 2 enzyme with exact Ki values were selected. A total of
54 molecules divided into a training set of 36 and test set of
18 were used for derivation of the model. The CoMFA and
CoMSIA statistical analysis is compiled in (Table 2). Statistical
data show q2 of 0.740 for CoMFA, 0.710 for the CoMSIA
loo
models, r2
of 0.990 for CoMFA and 0.959 for CoMSIA
ncv
which signify a good internal predictive ability of the models.
To examine the predictive ability of the models, a test set of
18 molecules excluded from the model derivation was used.
The predictive correlation coefficient r2
of 0.58 for CoMFA
pred
and 0.60 for the CoMSIA models indicate good external pre-
dictive capability of the models. The experimental and pre-
dicted activity from CoMFA and CoMSIA model is given in
Table 1. Scatter plot of experimental pKi vs. predicted pKi
values is represented in (Figure 3).
The complete contour map analysis of both CoMFA and
CoMSIA models permit us to recognize structural require-
ments for the inhibitory activity. The molecules were modi-
fied further to improve the inhibitory activity toward TYK-2.
CoMFA and CoMSIA contour maps depict different
molecular field values at lattice points that strongly con- Based on the steric and hydrophobic contour maps –NH of
nected with difference in the receptor binding affinity. the amide group hydrogen’s is substituted with benzyl

468
R. ITTEBOINA ET AL.
Figure 4. CoMFA and CoMSIA Steric and electrostatic fields.
Figure 5. CoMSIA Hydrophobic, Hydrogen bond donor and acceptor fields.
CoMSIA values are represented in Table 3. All newly
designed molecules had good predicted activity and dock
score values. An acceptable correlation between predicted
activity and experimental activity values (70% of CoMFA with
Lncap cell lines, 61% of CoMSIA with Lncap cell line, and
55% of CoMSIA with A 549 cell lines- Supplementary infor-
mation Figure S1(a–c)) and a good correlation (56%) of dock
scores with LnCap cell lines (Supplementary information
Figure S1(d)). From the docking results, we have observed
that the highest active molecule in dataset (Molecule-1)
showed two hydrogen bond interaction with ASN 1028 and
VAL 981 in the protein active site, the best active com-
pounds R-5 showed hydrogen bond interactions with SER
985 and R-10 showed hydrogen bond interactions with ASP
988 and ASN 1028 residues along with hydrophobic interac-
tions with both R-5 and R-10. The other compounds also
showed hydrogen bond interactions with active site amino
acid residues. We synthesized the designed molecules (R1-
R11) in two steps, in step 2 we obtained two novel inter-
mediates R-12&R-13 for which anticancer activity against
LnCap and A549 cell lines. All the synthesized molecules
showed activity on both cell lines. The anticancer activity val-
ues (pIC50) of these molecules are provided in Table 4 and
the bar graph is shown in Figure 8 showing comparison
between R1 to R13 and Doxorubicin. The dock pose of high
active molecule in dataset (molecule 1) and high active mol-
ecule in newly synthesized set (molecule R-5) shown in
Figure 9. docking results were validated by MD simula-
Electronegitive,hydrophilic,hydrogenbond acceptor and
steric favoured region
R
Cl
Cl
Hydrophobic and steric
favoured region
O
NH
O
N
NH
steric,hydrophilic,electronegitive favoured
region
Figure 6. Structural requirements for TYK 2 inhibitors obtained from CoMFA
and CoMSIA contourmap analysis.
group, and based on electrostatic counter map –NH of pyri-
dine attached group is substituted with chlorine group.
Based on hydrogen bond acceptor contour map –NH2 on
the benzene ring is substituted with methoxy and nitro
group. Molecule 1 having best activity was chosen as a refer-
ence structure to design new potent molecules (Figure 6).
The newly designed molecules were docked into the pro-
tein (PDB ID: 4GJ2) active site. Dock poses were used to pre-
dict the activity by applying the 3D-QSAR model. A set of 13
molecules with good predicted activity were designed and
assessed by previously established CoMFA and CoMSIA mod-
els, Structures (Figure 7), dock scores and predicted CoMFA- tion studies.

JOURNAL OF RECEPTORS AND SIGNAL TRANSDUCTION
469
Figure 7. Structures of the synthesized molecules (R1-R13).
Step 1. Synthesis of substituted amines.
3.2. Chemistry
potassium carbonate and tetra butyl ammonium bromide by
microwave irradiated for 10 min. Completion of the reaction
was checked by TLC and purified by column chromotogra-
phy with hexane:ethyl acetate (90:10, v/v)mobile phase. [34]
Step 2: In step 2, the solution of substituted benzoic acid
The desired benzamide Scaffold derivatives were synthesized
in two steps as outlined in Step 1 and 2.
Step1: The synthesis of desired benzamide scaffold was
performed in two steps starting from primary amine reacting
with substituted benzyl chlorides in the presence of derivatives in DCM was added a solution of DCC in DCM.

470
R. ITTEBOINA ET AL.
Step 2. Synthesis of substituted amides.
key features of the binding mechanism for these N-(pyridin-
4-yl) benzamide Scaffold derivatives. Based on the docking
results and 3D-QSAR, we designed and synthesized new mol-
Table 3. Predicted activity and dock scores of the designed molecules.
Pred.pKi
Molecule
CoMFA
CoMSIA
Glidescore (Kcal/mol)
1
ecules and were characterized by IR, H-NMR, and MS spec-
R-1
R-2
R-3
R-4
R-5
R-6
R-7
R-8
6.847
6.128
7.087
6.932
7.584
6.263
7.487
6.507
6.831
6.941
6.551
6.425
7.124
6.656
6.377
7.104
6.816
7.205
6.157
7.018
6.874
6.979
7.106
6.523
6.845
6.912
ꢀ7.138
ꢀ6.534
ꢀ7.248
ꢀ7.096
ꢀ7.637
ꢀ6.928
ꢀ7.154
ꢀ6.659
ꢀ7.085
ꢀ7.146
ꢀ7.225
ꢀ7.087
ꢀ7.249
tral data and elemental analyses. The synthesized molecules
R1-R13 were evaluated for their in vitro anticancer activity
against two human cell lines such as LnCap (prostate) and
A549 (lung) using a MTT Cell Proliferation Assay to estimate
the viability or growth. Measurement of cell viability and pro-
liferation forms the basis for numerous in vitro assays of a
cell population’s response to external factors. The MTT Cell
Proliferation Assay measures the cell proliferation rate and
conversely, when metabolic events lead to apoptosis or
necrosis, the reduction in cell viability. Significantly, the mol-
ecules R-5, R-7, and R-10 showed promising anti cancer
activity with pIC50 values ranging from 7.98 to 7.97, 7.92 to
7.37, and 7.87 to 7.76 respectively, against the two
cell Lines.
R-9
R-10
R-11
R-12
R-13
Table 4. Anticancer activity of the synthesized molecules.
IC50 (nM/ml)
pIC50
Molecule
LnCap
A549
LnCap
A549
R-1
R-2
R-3
R-4
R-5
R-6
R-7
R-8
18.687
60.4
16.07
33.921
57.021
39.675
31.241
10.633
43.896
41.741
23.591
16.312
13.239
17.253
26.666
37.532
3.42
7.7284
7.2189
7.7939
7.4894
7.9848
7.3535
7.9248
7.3705
7.6386
7.7605
7.5881
7.3268
7.8167
8.3575
7.4695
7.2439
7.4014
7.5052
7.9733
7.3575
7.3794
7.6272
7.7874
7.8781
7.7631
7.5740
7.4255
8.4659
3.3. Molecular dynamics simulations
32.397
10.354
44.301
11.889
42.608
22.981
17.356
25.811
47.111
15.251
4.39
Molecular dynamics (MD) simulation is a helpful precise
mechanism to find out molecular interactions possible in a
protein-ligand complex that is responsible for its stability. To
gauge and compare the stability and molecular interactions
of newly synthesized 3 D QSAR based molecule R-5 and crys-
tal ligand (Molecule17), MD simulations were performed for
10 ns. Figure 10 shows the root mean square deviations
(RMSD) of protein backbone with simulation time for each
complex from the initial structure. Initially, RMSD reached a
value of 1.553 Å from 0.722 Å during 0–500ps and then
retained between 1.162–2.382 Å throughout the simulation
up to 10 ns for co crystal ligand complex whereas RMSD of
molecule R-5 complex reached a value of 1.671 Å initially
and then retained between 1.008–2.316 Å during the entire
10 ns simulation time. The averaged RMSD of the crystallized
complex was found to be around 1.881 Å, whereas for R-5-
protein complex, it was around 1.665 Å shows similar stability
for both the complexes. Change in binding energies during
the simulation period using MD simulation trajectory was cal-
culated. Figure 11 shows changes in binding energies (dG)
with simulation time for both the protein–ligand complexes.
R-9
R-10
R-11
R-12
R-13
Doxorubicin
The solution was stirred at room temperature for 1 h, the
resulting suspension was added to a solution of the corre-
sponding amine and DMAP in DCM, and the resulting crude
was stirred for 24 h at room temp. The solid was filtered and
the solution was washed with water, 6 N HCl, 6 N NaOH and
finally again with water. The organic layer was dried over
MgSO4, filtered and the solvent was removed in vacuo. The
residue thus obtained was the desired amide derivatives [35].
Based on the above scheme, we synthesized R-1 to R-11
Molecules with good yields, and in step 2, we are getting
two novel intermediates (R-12&R-13) were obtained.
The good concordance between the 3 D contour maps From Figure 11, it is noticed that both the complexes
and the docking results provided insight to identify several showed stable dG during simulation time, dG bind values

JOURNAL OF RECEPTORS AND SIGNAL TRANSDUCTION
471
Figure 8. Bar graph showing inhibitory activities of synthesized molecules and standard doxorubicin.
Figure 9. (a) Docked pose of molecule-1 in the protein active site. Showing the hydrogen bond interaction (yellow lines) with ASN 1028 and VAL 981 (b) Docked
pose of molecule R-5 in the protein active site. Showing the hydrogen bond interaction (yellow lines) with SER 985.
3
2.5
2
-100
-80
-60
-40
-20
0
1.5
1
0
2000
4000
6000
8000
10000
Time(ps)
0.5
0
Figure 11. Change in Binding energy (dG) with respect to simulation time.
Blue color represents protein complexed with crystal structure ligand and red
color represents protein complexed with molecule R-5.
0
2,399
4,799
Time (ps)
7,198
9,598
to be ꢀ88.605 kcal/mol, and for molecule R-5, it was around
ꢀ90.279 kcal/mol. An identical trend of binding energies
over a period of simulation for both the complexes shows
their relative stability, strengthening the assumption that
molecule R-5 can act as a potent inhibitor against TYK 2. To
know the mode of interaction of the ligand with protein, lig-
and interaction was analyzed throughout the course of
Figure 10. Root mean square deviations (RMSD) of protein back bone as func-
tion of simulation time of each complex with respect to initial structure. Blue
color shows RMSD of protein back bone of crystal structure ligand where as
Red color shows RMSD of protein back bone of molecule R-5.
ranged from ꢀ81.190 to ꢀ93.272 kcal/mol for crystal struc-
ture ligand and ꢀ87.149 to ꢀ94.215 kcal/mol for molecule R- dynamics run. Figure 12 shows ligand interaction diagram of
5. Average dG bind for the crystal structure ligand was found crystal structure and synthesized molecule R-5 complex

472
R. ITTEBOINA ET AL.
Figure 12. Ligand interaction diagram of (a) co-crystal ligand and (b) newly designed molecule R5 with protein active site during the course of dynamic simulation
showing average interaction percentage with activesite residuces.
Figure 13. Protein ligand contact interaction over trajectory with respect to (a) Co-crystal ligand (molecule 17) and (b) Molecule R-5.
during the course of the simulation. The binding site of TYK water solubility (QPlogS) and partition coefficient (QPlogPo/
2 protein has the following amino acid residues ARG 901, w) that is critical for estimating the absorption and distribu-
LEU 903, GLY 906, VAL 911, ILE 960, GLU979, TYR 980, VAL tion of drugs within the body, ranged between 4.769–6.365
981, PRO 982, SER 985, ARG 987, ASP 988, ARG 1027, ASN and 4.16–5.375, respectively. Crossing the blood-brain barrier
1028, LEU 1030, GLY 1040, ASP 1041. The Newly synthesized (BBB), which is a mandatory for the entry of drugs into CNS,
molecule (R-5) showed 71% hydrogen bond interaction with was found to be in the satisfactory range (ꢀ1.308–0.146),
ASP 1041, 57% with ASN 1028, and 57% with ASP 988 via indicating that the molecules may be considered for further
water bridges. The crystal ligand showed 99%, 89% with VAL development in treating gliomas. MDCK cell permeability
981 and 81%, 38% hydrogen bond interaction with ASP 988 (QPPMDCK), a model that mimics the blood-brain barrier,
and ARG 901 via water bridges respectively in simulation ranges from 148.533 to 6522.896. Further, the predicted per-
time (Figure 13). The interaction diagrams clearly indicate centage human oral absorption is 100 for all molecules.
that the synthesized molecule has similar mode of inter- (Supplementary information-Table S1).
action with the receptor.
5. General information
4. Predicted ADME properties
All starting materials, reagents, and solvents were commer-
Newly synthesized molecules were analyzed for drug-likeness cially available and used after purification. Microwave irradi-
by calculating their physiochemical properties and by apply- ation was performed using synthetic microwave oven (Cata
ing Lipinski’s rule of five. The Lipinski’s rule for the drug-like 2R). Infrared (IR) spectra were recorded on a Perkin Elmer FT-
molecules, states that the molecule should have molecular IR 400 Spectrometer as KBr pellets and data reported in
weight <650 Daltons, H-bond donors <5, H-bond acceptors wave numbers (cmꢀ1). 1H NMR and 13C NMR spectra were
<10, and a log P of <5. For the synthesized molecules, the determined in CDCl3 solution by using 400 and 100 MHz

JOURNAL OF RECEPTORS AND SIGNAL TRANSDUCTION
473
spectrometers, respectively (Instrument Bruker Ultra shield
Arance III Nano, 400.13 MHz for 1H; 100.61 MHz for 13C).
Chemical shift signals are given in d (parts per million) rela-
tive to TMS, and coupling constants (J) are expressed in
Hertz (Hz). Combinations of the following abbreviations are
used to describe NMR spectra: s (singlet), d (doublet), t (trip-
let) q (quartet), and m (multiplet). Flash column chromatog-
raphy was performed using silica gel (Merck, 60–120 Mesh).
Commercially available reagents were used as supplied and
some of them were distilled before use. All reactions were
performed in oven-dried glassware. All the synthesized com-
pounds are characterized by spectroscopic (1H NMR, 13 C
NMR and ESI-MS) data and were found to be in good com-
pliance with their depicted structures. Spectroscopic data for
these compounds is provided. (Supplementary information).
References
[1]
[2]
[3]
[4]
[5]
[6]
Yoshiya T, Shigeru I, Kunihiro Y. Jak and Syk: emerging their rele-
vance to the treatment of inflammatory diseases. Inflamm Regen.
2011;31:237–244.
O’Shea JJ, Pesu M, Borie DC. A new modality for immunosuppres-
sion: targeting the JAK/STAT pathway . Nat Rev Drug Discov.
2004;3:555–564.
Malemud CJ, Pearlman E. Targeting JAK/STAT signaling pathway
in inflammatory diseases. Curr Signal Transduction Ther. 2009;4:
201–221.
Ghoreschi K, Laurence A, O’Shea JJ. Selectivity and therapeutic
inhibition of kinases: to be or not to be? Nat Immunol. 2009;10:
356–360.
Kisseleva T, Bhattacharya S, Braunstein J. Signaling through the
JAK/STAT pathway, recent advances and future challenges. Gene.
2002;285:1–24.
ꢀ
Igaz P, Toth S, Falus A. Biological and clinical significance of the
JAK-STAT pathway; lessons from knockout mice. Inflamm Res.
2001;50:435–441.
[7]
[8]
[9]
Leonard WJ, O’Shea JJ. JAKS AND STATS: biological implications.
Annu Rev Immunol. 1998;16:293–322.
Christian S, David EL, Thomas D. JAK-STAT signaling: from inter-
ferons to cytokines. J Biol Chem. 2007:282:20059–263.
Levy DE, Darnell JE. Jr. Stats: transcriptional control and biological
impact. Nat Rev Mol Cell Biol. 2002;3:651–662.
6. Conclusions
Molecular modeling tools like docking based 3D-QSAR and
MD simulation studies are useful to explain binding mode of
molecules to receptors and rationalize structural require-
ments for inhibition. In present study, molecular docking [10] Stump KL, Lu LD, Dobrzanski P, et al. A highly selective, orally
active inhibitor of Janus kinase 2, CEP-33779, ablates disease in
two mouse models of rheumatoid arthritis. Arthritis Res Ther.
2011;13:R68
Quint_as-Cardama A, Hagop K, Jorge C, et al. Janus kinase inhibi-
tors for the treatment of myeloproliferative neoplasias and
beyond. Nat Rev Drug Discov. 2011;10:127–140.
(XP) and MD simulations were performed to understand the
binding mode of inhibitors with TYK 2. CoMFA and CoMSIA
methodologies were used to frame models for TYK 2 inhibi-
[11]
[12]
[13]
[14]
[15]
[16]
[17]
tors. The developed models Showed good statistical reliabil-
ity that is evident from high r2 and q2 values for all the
models, this indicates the ability of QSAR model to predict
potent inhibitors from designed molecules. Based on
detailed contour map analysis new molecules were designed
and synthesized, these molecules showed good dock scores
anti cancer activity against LnCap and A 549 cell lines.
Molecules R-5, R-7, and R-10 showed promising anti cancer
activity with pIC50 values against two cell lines as 7.98, 7.97,
and 7.92, 7.37, and 7.87, 7.76 respectively. Further MD sim-
ulations of protein-ligand complex also revealed that synthe-
sized molecules retained hydrogen bond interactions with
ASN 1028.
R_obert K, Peter PS, Gyeorgy MK. Recent developments on JAK2
inhibitors:
a patent review. Expert Opin Ther Pat. 2010;20:
471–495.
Domanska D, Brzezianska E. The JAK/STAT protein activation –
role in cancer development and targeted therapy. CST. 2012;7:
187–201.
Caflisch A, Ehrhardt C. Structure-based combinatorial ligand
design, in: P. Veerapandian (Ed). Structure-based drug design.
New York (NY): Marcel Dekker; 1997. p. 541–558.
Cho SJ, Garsia MLS, Bier J, et al. Structure- based alignment and
comparative molecular field analysis of acetylcholinesterase inhib-
itors. J Med Chem. 1996;39:5064–5071.
Sippl W, Contreras JM, Parrot I, et al. Structure-based 3D QSAR
and design of novel acetylcholinesterase inhibitors. J. Comput.
Aided. Mol. Des. 2001;15:395–410.
Sippl W Receptor-based 3D QSAR analysis of estrogen receptor
ligands merging the accuracy of receptor-based alignments with
the computational efficiency of ligand-based methods. J.
Comput. Aided Mol Des. 2000;14:559–572.
Cramer RD, 3rd, Patterson DE, Bunce JD. (Comparative molecular
field analysis (CoMFA). 1. Effect of shape on binding of steroids
to carrier proteins. J Am Chem Soc. 1988;110:5959–5967.
Cramer RD, Bunce JD, Patterson DE, et al. Cross validation, boot-
strapping, and partial least squares compared with multiple
regression in conventional QSAR studies. Quant Struct-Act Relat.
1988;7:18–25.
Acknowledgements
€
We gratefully acknowledge TriposInc, USA and Schrodinger LLC, New
York for providing us the software. We wish to express our gratitude to
Department of Chemistry, Osmania University for providing facilities to
carry out the research work.
[18]
[19]
Disclosure statement
No potential conflict of interest was reported by the authors.
[20]
[21]
[22]
Klebe G, Abraham U, Mietzner T. Molecular similarity indices in a
comparative analysis (CoMSIA) of drug molecules to correlate
and predict their biological activity.
J Med Chem. 1994;37:
Funding
4130–4146.
This research was made possible through grants from DST-SERB (SB/
EMEQ-004/2013), CSIR (01/(2436)/10/EMR-II), DST-PURSE-II/2017 and
UGC/UPE/FAR/OU/2017 New Delhi, India. Dr.Venugopal Akula IICT
Hyderabad for mass spectral studies and author’s RI and SB would like
to acknowledge financial support from UGC for research fellowships.
Liang J, van Abbema A, Balazs M, et al. Lead optimization of a
4-aminopyridine benzamide scaffold to identify potent, selective,
and orally bioavailable TYK2 inhibitors. J Med Chem. 2013;56:
4521–4536.
Schrodinger LLC. New York Glide Version 5.6; 2010.

474
R. ITTEBOINA ET AL.
[23]
Friesner RA, Banks JL, Murphy RB, et al. Glide: a new approach [31]
for rapid, accurate docking and scoring. 1. Method and assess-
ment of docking accuracy. J Med Chem. 2004;47:1739–1749.
Liang J, Vickie T, Anne Van A, et al. Lead identification of novel and
Srilata B, Ramesh I, Sree Kanth S, et al. Structural insights of
Staphylococcus aureus FtsZ inhibitors through molecular docking,
3D-QSAR and molecular dynamics simulations. J Recept Signal
Transduc. 2018;38:61–70.
Ramesh I, Srilata B, Sree Kanth S, et al. Molecular modeling-driven
approach for identification of Janus kinase 1 inhibitors through
3D-QSAR, docking and molecular dynamics simulations. J Recept
Signal Transduc. 2017;37:453–469.
Sree Kanth S, Vijjulatha M. Molecular docking and 3D-QSAR stud-
ies on triazolinone and pyridazinone, non-nucleoside inhibitor of
HIV-1 reverse transcriptase. J Mol Mod. 2010;16:1169–1178.
Kadir OY, Edip T. Synthesis, docking and biological evaluation of
oxamide and fumaramide analogs as potential AChE and BuChE
inhibitors. Med Chem Res. 2015;24:588–602. DOI 10.1007/s00044-
014-1152-4.
[24]
selective TYK2 inhibitors. Eur J Med Chem. 2013;67:175–187. 2013;
Sybyl-X 2.1. St. Louis (MO): Tripos Associates; 2012.
[32]
[25]
[26]
Gasteiger J, Marsili M. Iterative partial equalization of orbital elec-
tro negativity a rapid access to atomic charges. Tetrahedron
1980;36:3219–3228. (80)80168-2.
Ramesh I, Srilata B, Sree Kanth S, et al. Molecular docking, 3D
QSAR and dynamics simulation studies of imidazo-pyrrolopyri-
dines as janus kinase 1 (JAK 1) inhibitors. Comp Bio Chem. 2016;
64:33–46.
[33]
[34]
[27]
[28]
[29]
Shaw DE. 2014. Research, DESMOND (Version 3.8), New York.
Kaminski GA, Friesner RA, Tirado-Rives J, et al. Evaluation and
reparametrization of the OPLS-AA force field for proteins via
comparison with accurate quantum chemical calculations on pep-
tides. J Phys Chem B. 2001;105:6474–6487.
ꢀ
ꢀ
ꢀ
[35]
Falco JL, Pique M, Gonzalez M, et al. Synthesis, pharmacology
and molecular modeling of N-substituted 2-phenyl-indoles and
benzimidazoles as potent GABAA agonists. Eur J Med Chem.
2006;41:8985–8990.
[30]
Berendsen HJC, Grigera JR, Straatsma TP. The missing term in
effective pair potentials. J Phys Chem. 1987;91:6269–6271.