Computational and Synthetic approach with Biological Evaluation of Substituted Quinoline derivatives as small molecule L858R/T790M/C797S triple mutant EGFR inhibitors targeting resistance in Non-Small Cell Lung Cancer (NSCLC)

Article abstract of DOI:10.1016/j.bioorg.2020.104612

New substituted quinoline derivatives were designed and synthesized via a five-step modified Suzuki coupling reaction. A comparative molecular docking study was carried out on two different types of EGFR enzymes which include wild-type (PDB: 4I23) and T790M mutated (PDB: 2JIV) respectively. Compounds were also validated upon T790M/C797S mutated (PDB ID: 5D41) EGFR enzyme at the allosteric binding site. All docking studies confirmed high potency and flexibility towards wild type as well as a mutated enzyme. Anticancer activity of the synthesized derivatives was examined against HCC827, H1975 (L858R/T790M/C797S and L858R/T790M), A549, and HT-29 cell lines by standard MTT assay. Most of the quinoline derivatives revealed a significant cytotoxic effect. The IC50 values of 4-(4-methylquinolin-2-yl)phenyl 4-(chloromethyl)benzoate (5j) were found to be 0.0042 μM, 0.02 μM, 1.91 μM, 3.82 μM and 3.67 μM while IC50 values of osimertinib were 0.0040 μM, 0.02 μM, ND, 0.99 μM and 1.22 μM, respectively. Compound 5j has shown excellent inhibitory activities against EGFR kinases triple mutant with IC 50 value 1.91 μM. It was observed that, compared to H1975, A549 and A431 cell lines, synthesized compounds significantly inhibited proliferation of the HCC827 cell line. These data suggested that synthesized compounds showed promising selective anticancer activity against tumor cells harboring EGFR Del E746-A750. The potency of compound 5j was compared through molecular dynamic simulations and an insilico ADMET study. QSAR models were generated and the best model was correctly compared with respect to predicted and observed activity of compounds. The built model will assist to design, refine and construct novel substituted quinoline derivatives as potent EGFR inhibitors in near future.

Full text of DOI:10.1016/j.bioorg.2020.104612

Bioorganic Chemistry 107 (2021) 104612
Contents lists available at ScienceDirect
Bioorganic Chemistry
journal homepage: www.elsevier.com/locate/bioorg
Computational and Synthetic approach with Biological Evaluation of
Substituted Quinoline derivatives as small molecule L858R/T790M/C797S
triple mutant EGFR inhibitors targeting resistance in Non-Small Cell Lung
Cancer (NSCLC)
Kshipra S. Karnika, Aniket P. Sarkatea, Shailee V. Tiwarib, Rajaram Azadc, Prasad V.L.
,
*
S. Burrad, Pravin S. Waktea
a Department of Chemical Technology, Dr. Babasaheb Ambedkar Marathwada University, Aurangabad, MS 431004, India
b Department of Pharmaceutical Chemistry, Durgamata Institute of Pharmacy, Dharmapuri, Parbhani 431401, MS, India
c Department of Animal Biology, University of Hyderabad, Hyderabad 500046, India
d Department of Biotechnology, KLEF University, Vaddeswaram, AP 522502, India
A R T I C L E I N F O
A B S T R A C T
Keywords:
New substituted quinoline derivatives were designed and synthesized via a five-step modified Suzuki coupling
reaction. A comparative molecular docking study was carried out on two different types of EGFR enzymes which
include wild-type (PDB: 4I23) and T790M mutated (PDB: 2JIV) respectively. Compounds were also validated
upon T790M/C797S mutated (PDB ID: 5D41) EGFR enzyme at the allosteric binding site. All docking studies
confirmed high potency and flexibility towards wild type as well as a mutated enzyme. Anticancer activity of the
synthesized derivatives was examined against HCC827, H1975 (L858R/T790M/C797S and L858R/T790M),
A549, and HT-29 cell lines by standard MTT assay. Most of the quinoline derivatives revealed a significant
cytotoxic effect. The IC50 values of 4-(4-methylquinolin-2-yl)phenyl 4-(chloromethyl)benzoate (5j) were found
to be 0.0042 µM, 0.02 µM, 1.91 µM, 3.82 µM and 3.67 µM while IC50 values of osimertinib were 0.0040 µM, 0.02
µM, ND, 0.99 µM and 1.22 µM, respectively. Compound 5j has shown excellent inhibitory activities against EGFR
kinases triple mutant with IC 50 value 1.91 µM. It was observed that, compared to H1975, A549 and A431 cell
lines, synthesized compounds significantly inhibited proliferation of the HCC827 cell line. These data suggested
that synthesized compounds showed promising selective anticancer activity against tumor cells harboring EGFR
Del E746-A750. The potency of compound 5j was compared through molecular dynamic simulations and an
insilico ADMET study. QSAR models were generated and the best model was correctly compared with respect to
predicted and observed activity of compounds. The built model will assist to design, refine and construct novel
substituted quinoline derivatives as potent EGFR inhibitors in near future.
Quinoline
Suzuki coupling
EGFR
Molecular docking
Molecular dynamics simulations
Anticancer activity
QSAR
1. Introduction
Overexpression of EGFR may lead to adverse conditions, mostly in lung
cancer, breast cancer, and glioblastoma [3]. The activation of epidermal
Epidermal Growth Factor Receptor (EGFR) is one of the most enzyme
belonging to the ErbB family of receptor tyrosine kinase which also in-
cludes ErbB2 (HER-2 or Neu), ErbB3 (HER-3), and ErbB4 (HER-4) [1,2].
growth factor pathways results in the initiation of cancer proliferation,
increased metastasis potential, and neoangiogenesis [4]. Overexpression
of EGFR triggers many downstream signaling pathways which cause
Abbreviations: Cys797 to Ser797 (C797S), Cysteine to Serine; T790M, Threonine 790 Methionine; WT, Wild-Type; EGFR, Epidermal Growth Factor Receptor; XP,
Extra precision docking; PDB, Protein Data Bank; ADMET, Absorption, Distribution, Metabolism, Excretion and Toxicity; MD simulations, Molecular Dynamic
Simulation; RMSD, Root Mean Standard Deviation; Mol MW, Molecular Weight; NSCLC, Non-Small Cell Lung Cancer; RMSF, Root Mean Standard Fluctuation; NPT,
Constant Number of Particles, Pressure and Temperature; AMBER99, Assisted Model Building with Energy Refinement; QSAR, Quantitative Structure Activity
Relationship.
* Corresponding author.
E-mail address: pswkshipra16@gmail.com (P.S. Wakte).
https://doi.org/10.1016/j.bioorg.2020.104612
Received 29 September 2020; Received in revised form 13 December 2020; Accepted 27 December 2020
Available online 5 January 2021
0045-2068/© 2021 Elsevier Inc. All rights reserved.

K.S. Karnik et al.
Bioorganic Chemistry 107 (2021) 104612
more aggressive growth and invasiveness characteristics [5]. Activating
mutations in the EGFR gene left a tremendous impact on treatment
procedures in Non-Small Cell Lung Cancer (NSCLC). To overcome such
mutations, a recent discovery of new EGFR inhibitors played an
important role [6]. EGFR enzymes have certainly been undergoing
missense mutations. Such mutations leave a tremendous impact on the
drugs by showing resistance [7]. Changes in functions of the protein
have forced the researchers to discover new drugs to avoid as well as
inhibit the resistance. EGFR mutations are categorized according to their
nucleotide changes. The initial mutation was found in the form of
exon19 deletion [8]. Drugs like erlotinib and gefitinib which were
extensively used for the treatment of lung cancer, showed resistance [9].
The emergence of mutation provided new targets and compounds
named in the form of generations. Such drugs inhibiting the target
enzyme were considered as 1st generations. Continuing similar quina-
zoline scaffold, 2nd generation drugs were designed by changing the
residing R-groups. Drugs like afatinib and neratinib showed potential
inhibition towards the ongoing mutations [10]. It was all good until the
wild-type EGFR enzyme exhibited a new missense mutation in the form
of replacement of specific AUG gene to UAU gene [11]. This resulted in a
change in the amino acid of threonine to methionine at position 790.
This new T790M mutation also changed the binding property functions
which were observed in the form of toxicity [12].
chemists through installing various active groups to quinoline moiety
via different synthetic protocols [30].
The emergence of a new era of computational drug designing has
become an easier pathway to produce high potency drugs within less
span of time. To improve and understand newer techniques of drug
designing, we have come up with a set of highly potent quinoline scaf-
fold molecules. Novel molecules were subjected to molecular docking
and ADMET studies to understand binding pocket and predict their
potency. Compounds were further subjected to in vitro studies over
mutated cell lines from which the most potent compound was subjected
to molecular dynamics simulations. Finally, to determine and analyze
the molecules, they were subjected to QSAR studies, to build a model to
assist in designing, constructing, and refining new quinoline derivatives
in the coming future. Suzuki cross-coupling reaction has been widely
used for the selective construction of carbon-carbon bonds in organic
synthesis. These reactions are palladium-catalyzed and are carried out
under mild reaction conditions using boronic acids. Here, in this work
Tetrakistriphenylphosphopalladium is used as a catalyst under a mild
basic environment.
2. Result and discussion
2.1. Chemistry
To overcome mutations and observed toxicity, a new set of drugs
came into emergence with a change in scaffold from quinazoline to
pyrimidine ring [13]. The most recent development in the form of third
generation molecules was rociletinib and AZD9291 which showed
improved action against mutations such as exon 19 deletion and T790M
mutation [14]. Though effective, the molecules had several drawbacks
with the development of new resistance in the form of change in genetic
code resulting in C797S mutation [15]. Due to the emergence of the
C797S mutation, protein exhibited functional and conformational
changes. These changes developed a new binding site with high pocket
affinities and were termed as the allosteric binding site of the EGFR
enzyme [16]. A study of a new binding pocket was found to be helpful
enough to overcome resistance as well as inhibit the protein. Com-
pounds like EAI001 and EAI045 had shown their potential towards in-
hibition of the EGFR enzyme by binding to the allosteric site [17]. Fig. 1
shows the evolution of 4th generation allosteric binding EGFR inhibitors
due to new mutations.
Overall synthetic route of compounds 5(a-p) is outlined in Scheme 1.
Complete multi-step reaction resulted high yields in each step. Synthe-
sized compounds with their yields are shown in Table 1. Extraction of
product in each step was easily done by means of either precipitation or
extracted using solvents.
As depicted in scheme 1, initially substituted acetanilide (1) was
cyclised using polyphosphoric acid to form hydroxy compound (2). The
substituted compound (2) was chlorinated using POCl3 to form
substituted chloride compound (3). Substituted chloride compound (3)
with the prepared boronic acid derivative undergone Suzuki coupling
reaction to obtain a carbon–carbon coupled reaction intermediate. Ob-
tained Suzuki intermediate was treated with substituted 4-(4,4,5,5-tet-
ramethyl-1,3,2-dioxaborolan-2-yl)phenyl benzoate (4) to obtain final
compounds5(a-p).
2.2. Molecular docking
Nowadays, a variety of synthetic drugs, bioactive natural products,
pharmaceuticals, and agrochemicals contain nitrogen as the main moi-
ety in heterocyclic compounds [18]. Nitrogen-containing heterocyclic
compounds have played a vital role in increasing the potency of drugs.
An active nitrogen atom provides a variety of interactions which in-
creases the activity [19]. Quinoline exhibit wide range of pharmacologic
properties such as antibacterial [20], anti-inflammatory [21], antifungal
[22], antimalarial [23], anticonvulsant [24], antihypertensive [25],
anti-HIV [26], antiviral [27] and anticancer [28] activities which is a
frequently encountered heterocycle in medicinal chemistry literature
[29]. Different biological activities have been shown by medicinal
2.2.1. Docking with wild type and T790M protein
The first and second-generation compounds were found to have
quinazoline scaffolds, different from that of the designed compounds 5
(a-p). Reported studies stated that quinazoline scaffold exhibited tox-
icities during treatment, thus making it more difficult. The known drugs
like gefitinib, erlotinib, and neratinib bearing similar scaffolds showed
toxicity but were equally potent to hold up in the market. Third-
generation drugs were designed by changing the main scaffold from
quinazoline to pyrimidine. Though these compounds seem to be potent
after the emergence of the C797S mutation, it was certain to conclude
Fig. 1. Evolution of Fourth generation allosteric binding EGFR kinase inhibitors and activations of new mutations in enzyme like Del19, L858R and C797S were
observed along with T790M.
2

K.S. Karnik et al.
Bioorganic Chemistry 107 (2021) 104612
Scheme 1. Synthesis of compounds 5(a-p). Reagents and conditions: (a) Polyphosphoric acid, 130–140 ◦C; (b) POCl3, pyridine, 80–90 ◦C; (c) ACN, Reflux; (d)
Diisopropylamine, Tetrakistriphenylphosphopalladium, 80 ◦C; (e) Sodium methoxide, ACN, r.t.
that these compounds had some defects. The resistance offered by these
compounds made it obvious to change the main scaffold. To overcome
such defects, a similar scaffold family with different atoms were used to
design compounds. Designed quinoline scaffold compounds were set
under-hit and trial method to check their potency against wild type and
known mutated EGFR enzymes.
Thus, to study binding patterns, a mutated EGFR protein (PDB ID: 2JIV)
was considered as the target enzyme. The protein consists of two chains
A and B of similar structures with neratinib as the co-crystalized ligand.
Designed compounds were docked over chain A of the targeted protein.
The results were similar to that of docking results of wild type EGFR
protein. Compounds actively fit into a pocket, but lesser interactions
with nearby residues decreased docking scores as shown in Table 1.
Wild type EGFR protein (PDB ID: 4I23) consisted of CYS797 and
MET793 as the target interaction residues. ATP binding site forms a
pocket involving other residing residues. Each of the docked compounds
perfectly fit into the binding pocket showing interactions with residing
residues as observed in Fig. 2. Though perfectly fitting into the pockets
still compounds didn’t interact potentially to key residues. This resulted
in reduced docking scores and affinities. But an overall view towards
studies showed good potency of compounds.
2.2.2. Docking with T790M and C797S mutated protein
Newly emerged C797S mutation along with known T790M mutation
has changed binding properties of the protein. The emergence of the
C797S mutation decreased the potency of compounds interacting at the
ATP binding site. To overcome such problems, a new binding pocket in
the form of an allosteric binding site came into existence. Allosteric
binding pocket interacting structures like EAI045 showed high potency
T790M was one of the well-known mutations after initial resistance.
3

K.S. Karnik et al.
Bioorganic Chemistry 107 (2021) 104612
Table 1
Structures, combined docking scores and isolated yields of synthesized compounds 5(a-p).
Compound
Structure
Docking scores
Wild type (4I23)
ꢀ 6.81
Yielda (%)
T790M mutated (2JIV)
T790M/C797S mutated (5D41)
5a
ꢀ 6.97
ꢀ 9.4
85%
5b
ꢀ 7.88
ꢀ 7.72
¡10.7
92%
5c
5d
ꢀ 5.91
ꢀ 6.51
ꢀ 6.09
ꢀ 6.37
ꢀ 7.2
ꢀ 8.5
87%
82%
5e
5f
ꢀ 6.02
ꢀ 7.92
ꢀ 6.15
ꢀ 7.51
ꢀ 9.6
95%
93%
¡11.3
5g
5h
ꢀ 6.95
ꢀ 7.83
ꢀ 6.35
ꢀ 7.72
ꢀ 7.5
89%
94%
ꢀ 10.2
5i
5j
ꢀ 7.14
ꢀ 7.92
ꢀ 7.36
ꢀ 8.25
ꢀ 5.0
91%
84%
¡11.5
5k
ꢀ 7.87
ꢀ 7.93
ꢀ 8.2
88%
(continued on next page)
4

K.S. Karnik et al.
Bioorganic Chemistry 107 (2021) 104612
Table 1 (continued)
Compound
Structure
Docking scores
Yielda (%)
Wild type (4I23)
T790M mutated (2JIV)
T790M/C797S mutated (5D41)
5l
ꢀ 7.24
ꢀ 7.65
ꢀ 6.85
ꢀ 7.69
ꢀ 10.2
ꢀ 8.4
96%
91%
5m
5n
ꢀ 8.12
ꢀ 7.90
ꢀ 7.98
ꢀ 7.12
ꢀ 9.8
ꢀ 7.9
89%
83%
5o
5p
ꢀ 8.00
ꢀ 8.24
–
ꢀ 7.59
ꢀ 8.88
–
ꢀ 9.8
98%
Neratinib
EAI0045
–
–
–
ꢀ 10.5
a
Isolated yield.
and acceptability in the market. The ‘Y’ shaped like compound also
determined pocket properties to be more dependent on the shape
configuration.
as shown in Table 1.
2.3. ADMET studies
The newly emerged binding pocket of EGFR protein was an actual
target for inhibition. Key interacting residues GLU762, PHE856 were
found to actively bind the compounds. Molecular docking studies also
showed that compounds at their conformational state were easily able to
fit into the pocket as shown in Fig. 3. The binding affinities of com-
pounds were also proved to be efficient. All visual studies were reflected
into docking scores which determined that designed compounds had
higher docking scores and binding affinities when compared to EAI045
Drug likeliness implies the drug under investigation ought to have
qualities like that of known drugs. For a drug to be harmless and active
in people, it must have an acceptable ADMET profile which implies the
success of a decent drug applicant. Notably, issues related to ADMET
properties make a drug to come up short during the clinical tests [31].
Table 2 shows an analysis of the pharmacokinetic parameters required
for the ADMET study of compounds 5(a-p).
5

K.S. Karnik et al.
Bioorganic Chemistry 107 (2021) 104612
Fig. 2. Docking poses of compound 5a and 5b. The ligand is highlighted in grey colour.
Fig. 3. Allosteric binding of designed compounds 5b, 5f, 5j and standard EAI045 over C797S and T790M mutated EGFR protein (PDB ID: 5D41). Blue lines in the
figures indicate π-π stacking interaction and pink lines indicate hydrogen bonding interaction. (For interpretation of the references to colour in this figure legend, the
reader is referred to the web version of this article.)
Drug-likeness resemblance of compounds was surveyed by observing
the physicochemical properties like molecular weight (MW), predicted
octanol/water partition coefficient (log P), predicted aqueous solubility
(S in mol/L), Polar surface area (PSA), Predicted apparent MDCK cell
permeability (QPPMDCK in nm/sec) and Percent of human oral ab-
sorption respectively. Caco-2 cells are a model for the intestine-blood
barrier [32]. Caco-2 permeability is a great marker of drug absorbance
in the intestine, while MDCK cells are considered to be a good mimic for
the blood brain barrier. The study of insilico ADMET predictions helps in
the drug development process to produce novel molecules that are non-
toxic and possess good oral human absorption. Also, compounds that
pass Lipinski’s rule exhibit good oral absorption in humans [33].
Partition coefficient helps to evaluate absorption and distribution of
drugs within the body (QPlogPo/w), aqueous solubility (QPlogS) in-
dicates the extent to which a molecule is soluble in water. QPlogS was
ranged from ꢀ 3.2 to ꢀ 6.4 and the cell permeability values (QPPMDCK)
ranged from 107 to 1288, indicated that values were within the
acceptable range. The property of percent of human oral absorption was
observed well within the acceptable range, showing that the compounds
were non-toxic. All these ADMET properties were found to be within an
acceptable range defined for human usage, indicating their potential to
act as drug-like molecules. Violations from Lipinski’s rule of five were
also studied. Compounds with fewer (or preferably no violations) of the
rule are most probably orally available [34–36]. PAINS and Brenk alerts
determine the toxic or unwanted moiety in the compound which affects
the biological activity. The bioavailability scores determine that the
6

K.S. Karnik et al.
Bioorganic Chemistry 107 (2021) 104612
Table 2
Insilico ADMET results of synthesized compounds 5(a-p).
Compd.
No.
MWa
QPlogPo/
QPlogSc
PSAd
QPPMDCKe
Percent Human Oral
Absorptionf
QPPCacog
Violation of
PAINS
Brenk
wb
Lipinski’s Rule
#alertsh
#alerts
EAI045
5a
5b
5c
351.379
407.11
469.45
421.41
441.23
407.11
469.45
421.41
441.23
387.86
493.93
401.88
422.30
382.17
444.18
396.48
416.90
2.811
3.002
2.691
5.515
5.388
6.25
ꢀ 3.523
ꢀ 3.661
ꢀ 3.216
ꢀ 4.366
ꢀ 5.3
166.754
190.052
190.295
195.016
95.472
403.344
602.157
509.433
320.602
389.19
90.648
70.083
88.084
85.595
84.42
436.092
2914.238
283.77
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
1
1278.893
2214.414
1150.59
129.078
1962.004
3157.462
4827.211
411.641
1991.065
1956.587
537.718
126.579
1243.364
3377.713
5d
5e
ꢀ 6.127
ꢀ 6.325
ꢀ 4.581
ꢀ 4.759
ꢀ 3.89
103.093
109.766
103.53
247.856
229.86
83.636
70.972
57.21
5f
6.337
4.908
5.617
3.951
4.088
6.42
5g
5h
5i
508.44
195.002
191.905
100.005
108.093
190.206
190.312
194.89
336.333
272.742
509.455
288.45
85.79
75.608
70.297
75.49
5j
ꢀ 4.456
ꢀ 6.428
ꢀ 3.774
ꢀ 3.584
ꢀ 4.03
5k
5l
2.993
3.927
4.232
3.807
6.153
509.411
107.488
303.376
85.693
91.857
69.462
83.558
68.291
88.99
5m
5n
5o
5p
ꢀ 3.599
ꢀ 5.339
191.139
194.919
1288.775
a
Molecular weight of the molecule.
b
c
d
e
f
Predicted octanol/water partition co-efficient log P (acceptable range: ꢀ 2.0 to 6.5).
Predicted aqueous solubility; S in mol/L (acceptable range: ꢀ 6.5 to 0.5).
PSA (Polar surface area) (acceptable range: 7.0 to 200.0).
Predicted apparent MDCK cell permeability in nm/sec (acceptable range: >500 is high, <25is poor).
Percent of human oral absorption (>80% is high, <25% is poor).
QPPCaco cell permeability.
g
h
Pan-Assay Interference compounds.
potent compound is mostly available at the site of action. As there are no
BBB permeation and violations thus the compound is feasible to be
called a drug target. All 16 compounds showed satisfactory pharmaco-
kinetic parameters range defined for human use along with good XP
docking scores. The most potent compound 5j with the highest docking
score and best ADMET results was subjected to a molecular dynamics
simulation study.
against cancer cell lines with a low level of EGFR expression was eval-
uated to assess the potential non-specific toxicity. The cancer cell lines
A549 expresses the low levels of EGFR (WT EGFR) and therefore are
good counter-screen cell lines for EGFR-targeting inhibitors. These
compounds are also evaluated against HT-29 colon cancer cells, which
expressed a non-special gene type, to test their toxic effects. Synthesized
derivatives were less potent against A549 and HT-29 which shows a high
safety index of synthesized compounds.
Data of Table 3 indicated that, compared to H1975, A549, and A431
cell lines, synthesized compounds significantly inhibited the prolifera-
tion of the HCC827 cell line. These data suggested that compounds
showed promising selective anticancer activity against tumor cells
harboring EGFR Del E746-A750.
2.4. Biological activity
2.4.1. In vitro anticancer activity
To appraise the potency for EGFR inhibitory activity, synthesized
compounds 5(a-p) were tested against HCC827 cell line, harboring an
EGFR-activating mutation (EGFR Del E746-A750), gefitinib-resistant
non–small cell lung cancer NSCLC cell H1975 (possessing EGFR L858R/
T790M double mutations and L858R/T790M/C797S triple mutations),
A549 over-expressing wild-type EGFR (WT-EGFR) and colon cancer cells
HT 29, which expressed a non-special gene type, to test their toxic ef-
fects. The anticancer activity was performed using the MTT method.
Values of results are presented in the form of IC50 values as shown in
Table 3. Osimertinib was used as a standard drug.
SAR of the synthesized compounds can be drawn from anticancer
data reported in Table 3. Quinoline heterocycle is a must for EGFR
inhibitory activity. Compounds 5b, 5f, 5j, and 5n with phenyl ring on
Quinoline heterocycle at the 4th position have shown the most signifi-
cant activity. This means phenyl ring on Quinoline heterocycle at the 4th
position is required for good anticancer activity. Replacement of phenyl
ring on Quinoline heterocycle at the 4th position with other substituents
has resulted in a decrease in activity. But it was also observed that along
with phenyl ring, chloromethyl or trifluoromethyl group on benzoate
ring is essential for activity. Replacement of chloromethyl or tri-
fluoromethyl group on benzoate ring with other groups such as NH2
decreased the activity. Change in the position of the CF3 group from 4th
position to 3rd position also affects the activity profile of the compound.
Substitution of chloro group on quinoline ring also favors good anti-
cancer activity.
Compound 5j 4-(4-phenylquinolin-2-yl)phenyl
4-(chloromethyl)
benzoate showed significant anticancer activity against the selected
cancer cell lines. Compounds 5j and 5f were found to be equipotent to
standard drug Osimertinib. Compound 5j has shown excellent inhibi-
tory activities against EGFR kinases triple mutant with IC 50 value 1.91
µM. Compound 5j has shown IC 50 values of 0.0042 µM, 3.82 µM, and
3.67 µM against HCC827, A549, and HT 29 respectively. Com-
pound 5f has shown good anticancer activity against the EGFR double
and EGFR triple mutant possessing H1975 cell with IC 50 values of 0.09
and 2.24 µM respectively. The other compounds such as 5b, 5h, 5l,
and 5p have also displayed good anticancer activity against selected
cancer cell lines.
Synthesized compounds 5b, 5f, 5h, 5j, and 5l were tested to study
their inhibitory enzymatic activity against EGFR L858R/T790M/
C797S in vitro. Osimertinib was used as a standard drug. The results are
reported in Table 4. Synthesized compound 5j exhibited potent inhibi-
tory activity against triple mutant EGFR L858R/T790M/C797S with an
IC50 value of 113 nM, which was also the most potent compound in anti-
cancer activity against the selected cancer cells in vitro.
Since EGFR is expressing in many tissues for normal cellular pro-
cesses, the potential for toxicity from EGFR inhibitors is a concern. EGFR
inhibitors at concentrations required to inhibit EGFR T790M will also
effectively inhibit WT EGFR, leading to dose-limiting toxicity such as
diarrhea and skin rash. Therefore, growth inhibition by compounds
2.4.2. Western blot assay
To study mechanism of action for anticancer activity, the synthesized
7

K.S. Karnik et al.
Bioorganic Chemistry 107 (2021) 104612
Table 3
In vitro anticancer activity of synthesized compounds against cells harbouring a
different status of the EGFR.
Compound
IC50 µMa
HCC827
(EGFR Del
E746-
NCI-
NCI-H1975
(EGFR
A549/
ATCC
(WT
HT29
(Non-
Special
Gene
H1975
(EGFR
L858R/
T790M)
L858R/
T790M/
C797S)
A750)
EGFR)
Type)
5a
1.8834 ±
0.05
4.99 ±
0.55
ND
14.99
± 0.89
4.18 ±
0.31
15.95 ±
0.64
5b
0.0055 ±
0.25
0.12 ±
0.35
2.83 ± 0.24
10.88 ±
0.79
5c
0.1711 ±
0.09
3.06 ±
0.21
ND
13.65
± 0.89
10.88
± 0.05
14.50
± 0.70
3.48 ±
1.03
15.65 ±
0.16
Fig.4. Effects of compounds 5b, 5f and 5j on EGFR signalling in HCC827 cells.
5d
0.0114 ±
0.03
0.72 ±
0.59
ND
12.99 ±
0.71
and 48 h. The cells then were collected by trypsinization and 0.5 × 106
cells were washed twice with phosphate-buffered saline (PBS) and
stained with 5 µL Annexin V-FITC and 5 µL PI in 1× binding buffer for
15 min at room temperature in the dark. Analyses were performed using
BD LSR Fortessa) with FACS Diva version 6.2 software [37,38].
Apoptosis, autophagy, and necrosis are the major types of cell death.
Apoptosis, or programmed cell death, is a normal physiologic process for
the removal of unwanted cells. One of the earlier events of apoptosis
includes translocation of membrane phosphatidylserine (PS) from the
inner side of the plasma membrane to the surface. Annexin V, a Ca2+
dependent phospholipid-binding protein, has a high affinity for PS. This
assay detects PS expressed on the surface of the apoptotic cells and
fluoresces green after interacting with the labeled Annexin V. During
early apoptosis, membrane asymmetry is lost, and PS translocate from
the cytoplasmic side of the membrane to the external leaflet. Propidium
iodide (PI), the counterstain used in this assay, has the ability to cross
only compromised membranes to intercalate into the DNA. Therefore, PI
is used to detect the late stages of apoptosis and necrosis by the presence
of red fluorescence.
5e
1.1816 ±
0.07
4.48 ±
0.11
ND
16.22 ±
0.02
5f
0.0049 ±
0.23
0.09 ±
0.15
2.24 ± 0.62
7.89 ±
0.92
5g
0.0998 ±
0.16
2.88 ±
0.42
ND
12.96
± 0.89
9.55 ±
0.85
15.33 ±
0.11
5h
0.0068 ±
0.04
0.38 ±
0.25
ND
11.96 ±
0.63
5i
2.5433 ±
0.65
5.62 ±
0.55
ND
15.69
± 0.02
3.82 ±
0.66
18.74 ±
0.74
5j
0.0042 ±
0.11
0.02 ±
0.63
1.91 0.33
3.67 ±
0.31
5k
0.0872 ±
0.63
2.41 ±
1.04
ND
12.22
± 0.72
10.58
± 0.55
19.35
± 1.00
11.44
± 0.80
14.10
± 0.28
11.16
± 0.21
0.99 ±
0.05
14.56 ±
0.64
5l
0.0084 ±
0.16
0.44 ±
0.07
3.09 ± 0.11
12.22 ±
0.05
5m
2.9867 ±
0.17
8.36 ±
0.21
ND
ND
ND
ND
ND
20.34 ±
0.21
5n
0.0262 ±
0.81
2.04 ±
0.70
14.21 ±
1.03
5o
0.9567 ±
0.34
3.96 ±
0.72
15.78 ±
0.89
The ability of synthesized compound 5j to induce apoptosis was
determined using an Annexin V (conjugated to FITC) apoptosis detection
kit. From Fig. 5 it is clear that the compound 5j has induced early
apoptosis and late apoptosis of 16.7% and 4.1% respectively in com-
parison with control with early apoptosis of 3.2% and late apoptosis of
1.1%.
5p
0.0101 ±
0.87
1.02 ±
0.11
12.65 ±
0.05
Osimertinib
0.0040 ±
0.05
0.02 ±
0.25
1.22 ±
0.02
a
The data reported are the mean values from three independent experiments;
ND: Not determined.
2.4.4. Cell cycle arrest
Table 4
Cell cycle progression was evaluated using a flow cytometer, BD LSR
Fortessa) with FACS Diva version 6.2 software. In brief, HCC827 cells
were incubated for 24 h with given compound 5j, each at two concen-
trations closer to IC50 values, obtained during the cell viability assay.
After 24 h of treatment, cells were harvested, washed with PBS, and
fixed in ice-cold 70% ethanol for overnight at 4 ◦C. Next day, all samples
were centrifuged at 3000 RPM for 4 min and stained with propidium
iodide (PI) (5 mg/mL) followed by addition of RNase and analyzed
[39,40].
In vitro enzymatic inhibitory activity of synthesized compounds
against EGFR L858R/T790M/C797S.
Compound
EGFR inhibition IC50 (nM)a
5b
119 ± 0.31
115 ± 0.12
134 ± 0.48
113 ± 0.37
139 ± 0.11
111 ± 0.99
5f
5h
5j
5l
Osimertinib
To understand the mode of action of most active compounds, we had
examined their effects on cell cycle progression by flow cytometry
analysis on HCC827 cells. The blank control group with solvent was
used. The data obtained signposts that the actively synthesized com-
pound 5j can alter cell cycle progression enormously. The percentage
distribution of cells in various phases of the cell cycle is exemplified in
Fig. 8. Fig. 6 revealed that exposure of HCC827 cell to compound 5j at
IC50 concentration for 24 h induced a significant disruption in cell cycle
profile. The compound 5j showed cell cycle arrest at G0/G1 phase [41].
a
The data reported are the mean values from three independent
experiments.
compounds 5b, 5f and 5j were tested using western blot assay on the
phosphorylation of EGFR and the downstream signalling transduction in
HCC827 cells. HCC827 cells were treated with synthesized compounds
5b, 5f and 5j using various concentrations such as 1.00, 0.10 and 0.01
µM for each compound. The results of the activity are as shown in Fig. 4.
2.4.3. Apoptosis
Apoptosis was determined by staining the cells with Annexin V-
Fluorescein isothiocyanate (FITC) and counterstaining with PI using the
Annexin V-FITC/PI apoptosis detection kit (BD Biosciences, San Diego,
CA) according to the manufacturer’s instructions. Briefly, 4 × 106 cell/T
75 flasks were exposed to compound 5j at its IC50 concentration for 24
2.5. Molecular dynamic simulations
Heading towards the final stage of this work, synthesized compounds
were compared based on docking scores, interacting residues, binding
affinities, ADMET, and biological activities. This is done to finally select
8

K.S. Karnik et al.
Bioorganic Chemistry 107 (2021) 104612
Fig.5. Apoptosis study by Annexin V/FITC assay of compound 5j.
Fig.6. Flow cytometry cell cycle analysis of control and of compound 5j treated on HCC827 cells.
one of the most potent molecules for molecular dynamics simulation.
The molecular dynamic simulations give a detailed insight into protein-
ligand interactions. It gives a real-time virtual study of flexible protein
and ligand complex under some defined surrounding such as tempera-
ture, pressure, and water. After comparison, compound 5j was found to
be the most potent compound in terms of both insilico as well as invi-
tro studies. For checking the stability of molecule and protein, we per-
formed molecular dynamic simulations and were interpreted through
RMSD, RMSF, and the protein-ligand contacts. The TIP3P water mole-
cule set to immerse the system provided a real-time virtual system of the
body. The whole system was run for 10 ns at 300 K temperature and
pressure of 1.01325 bar. A total of 1500 frames were set to analyze
deeply into the dynamic simulations. Fig. 7 shows the root mean square
deviation (RMSD) which describes the stability of the protein-ligand
complex during the simulation. After the initial refining, both protein
and ligand were found to be stable till the end of the simulations. The
protein though had some fluctuations still stabilized till the end. This
was due to an increase in the number of frames thus increasing the ac-
curacy to analyze displacement or conformational changes. Changes of
the order 1–3 Å are considered acceptable for EGFR proteins. Any in-
crease in RMSD seen at end of the simulations concludes that the whole
system is not equilibrated. According to the acquired results, the whole
system was found to achieve an equilibrium state during the performed
simulations. The ligand was still inside the system bound to protein
determining the stability and binding affinity.
The ligand RMSF describes conformational changes and molecular
stability of each atom. Ligand RMSF shows ligand’s fluctuations broken
down by atom, corresponding to the 2D structure in the top panel. The
ligand RMSF may give you insights on how ligand fragments interact
with protein and their entropic role in the binding event. The results of
9

K.S. Karnik et al.
Bioorganic Chemistry 107 (2021) 104612
Fig. 7. Time dependent root mean square deviation (RMSD) of ligand and protein.
RMSF concluded in Fig. 10 that compound 5j was found to be stable
during interactions with the protein. The only fluctuation found was at
the 28th position where multiple conformational changes were
observed.
fraction for hydrogen bond for PHE 856 is 0.3. It was also seen that GLY
857 showed more than 30% hydrogen bonding with the compound.
Other than hydrogen bonds, hydrophobic bonds, and water bridges were
also found to be the key interactions during simulations.
Protein-ligand contacts are one of the key results required to be
observed during the simulation. If the interactions match docking re-
sults, it is most likely possible that ligand will undergo the same in-
teractions and show potency. Molecular dynamic simulations after
biological studies are the predictability of compounds to inhibit protein
at the molecular level. The known active residues LYS 745, GLY 857, and
PHE 856 were found to be the most consistent interactions during
simulations. More than 30% of frames visualized were found to show
these interactions. Fig. 9 describes protein-ligand contacts considering
all 1500 frames set for dynamic simulations. Orange colored thin strips
are interactions during each frame. The darker the colour, the more is
the consistency of ligand binding to the residue.
Fig. 11 show a 2-D interaction diagram of compound 5j, binding to
the residues which had more than 30% of interactions. The whole study
and results obtained concluded that compound 5j potentially inhibits
T790M and C797S mutated EGFR enzyme by allosterically binding to
the protein.
2.6. AutoQSAR studies
2.6.1. Generation of QSAR models
To confirm our protein-ligand interaction study and get detailed
information about the type of bonds, Fig. 10 was taken into study. The
bar diagram was plotted with interaction fraction against residue iden-
tifier. The interaction fraction determines the percentage of binding
during simulations. For example, PHE 856 creates hydrogen bonds with
the ligand for more than 30% during simulations. The interaction
Fig.8. Time dependent ligand root mean square fluctuations (RMSF).
10

K.S. Karnik et al.
Bioorganic Chemistry 107 (2021) 104612
Fig. 9. Protein-Ligand Contacts showing good contacts (darker shades) with the amino acid residues over 10 ns time period of simulation.
Fig. 10. Bar diagram of Protein-Ligand contacts (pictograph) of compound 5j indicating the fold of interaction, fraction and contacts. Different colours over bar
determine type of interaction such as hydrogen bond (green), hydrophobic (grey) water bridges (blue) and ionic (pink). (For interpretation of the references to colour
in this figure legend, the reader is referred to the web version of this article.)
Structures of compounds used for the development of common QSAR
studies along with biological activity are shown in Table 5. Most Auto
QSAR models showed a substantial improvement in predictive power
which refers to its ability to generate testable predictions. Table6 de-
scribes the best models when compared to standard values R2 and Q2as
per the literature. All best models obtained were invariably a Kernel-
11

K.S. Karnik et al.
Bioorganic Chemistry 107 (2021) 104612
Fig. 12. Predicted and simulated pIC50 versus experimental pIC50 for the
kpls_radial_29 model. Training and test sets are identified in blue and red colour
respectively. (For interpretation of the references to colour in this figure legend,
the reader is referred to the web version of this article.)
Fig. 11. Simulation interaction diagram (2D) of 5j showing interactions that
occur more than 30.0% of the simulation time in the selected trajectory (0.00
through 10.00 ns).
3. Conclusion
Table 5
In this work, novel substituted quinoline derivatives are designed
and synthesized which can serve as new L858R/T790M/C797S triple
mutant EGFR-TKIs. Among the synthesized compounds, 5b, 5f, 5h,
5j and 5n showed significantly potent anti-proliferation and EGFR
L858R/T790M/C797S enzyme-based inhibition. In particularly, most
potent compound 5j, which inhibited the EGFR L858R/T790M/C797S
with an IC50 value of 113 nM, also inhibited HCC827, H1975 (L858R/
T790M/C797S and L858R/T790M), A549 and HT-29 cell lines. Further,
it was clear that compound 5j have induced early apoptosis and late
apoptosis of 16.7% and 4.1% respectively in comparison with control. It
has also shown cell cycle arrest at the G0/G1 phase. SAR results revealed
that compounds bearing phenyl ring on quinolone heterocycle at 4th
position have shown the most significant activity. Molecular docking
studies showed that compounds at their conformational state were
easily able to fit into the pocket of T790M/C797S mutated (PDB ID:
5D41) EGFR enzyme at the allosteric binding site. The potency of
compound 5j was compared through molecular dynamic simulations
and insilico ADMET study. QSAR models were generated and the best
model was correctly compared to the predicted and observed activity of
compounds. The built model will assist to design, refine and construct
novel substituted quinoline derivatives as potent EGFR inhibitors in near
future.
Structure of compounds used for the development of common QSAR studies
along with biological activity.
Compd. No.
Biological activity
IC50 (µm)
pIC50
5aa
5ba
5c
4.99
0.12
3.06
0.72
4.48
0.09
2.88
0.38
5.62
0.02
2.41
0.44
8.36
2.04
3.96
1.02
5.30
6.92
5.51
6.14
5.34
7.04
5.54
6.42
5.25
7.69
5.61
6.35
5.02
5.69
5.40
5.99
5d
5e
5fa
5g
5h
5i
5j
5k
5l
5m
5n
5o
5pa
a
Compounds in test set.
based partial least squares (KPLS-related solution). Kpls_radial_29 with
the model score of 0.9333, R2 of 0.9322 and Q2 of 0.9631 were
considered as topmost QSAR model of the generated 10 models. Simi-
larly, one more model set was selected with a slightly less score but
optimum R2 and Q2 values as shown in Table6.
4. Experimental section
4.1. Chemistry
QSAR models were internally validated by randomly chosen sets as
30%. The low residual value between experimental and predicted ac-
tivity indicates that the model is of high predictability. Predicted activity
against the experimental activity of training as well as of test set the
selected kpls model is shown in Fig. 12.
Chemicals were purchased from commercial sources, viz. TCI and
Avra Labs, and used without further purification. Progress of the reac-
tion was monitored by thin layer chromatography (TLC) on pre-coated
silica gel 60 F254 aluminium sheets (Merck), and visualization was
done by UV light. 1H NMR (500 MHz) and 13C NMR (125 MHz) spectra
were recorded on Bruker AVANCE II NMR spectrometer in DMSO‑d6
solution. Tetramethylsilane was used as an internal standard. Chemical
shift values are given in ppm relative to TMS as internal reference and
coupling constant (J) in Hertz. The splitting pattern abbreviations are
assigned as singlet (s), doublet (d), triplet (t), broad singlet (brs), double
doublet (dd), and multiplate (m). Mass were recorded on a WATERS, Q-
TOF micro mass equipped with an Electron Spin Impact (ESI) Source.
Elemental analysis was performed on Perkin-Elmer EAL-240 elemental
analyzer.
Table 6
Predicted best QSAR models.
Model Code
Model Score
S.D.
R2
RMSE
Q2
kpls_radial_29
kpls_radial_39
0.9333
0.8898
0.1991
0.2387
0.9322
0.9179
0.1368
0.2278
0.9631
0.8721
12

K.S. Karnik et al.
Bioorganic Chemistry 107 (2021) 104612
4.1.1. Synthesis of compounds
4.1.1.3.3. 4-(4,8-dimethylquinolin-2-yl)phenyl3-(trifluoromethyl)ben-
zoate (5c). Light brown solid, 130.63 ◦C, MS (ESI) m/z: for
C25H18F3NO2 [M+H]+: 421.13, 1H NMR (500 MHz, Chloroform-d) δ
8.28 (t, J = 2.2 Hz, 1H, Ar-H), 8.09–8.03 (m, 3H, Ar-H), 7.89–7.84 (m,
1H, Ar-H), 7.74–7.65 (m, 2H, Ar-H), 7.56 (dd, J = 10.7, 7.4 Hz, 1H, Ar-
H), 7.44 (s, 1H, Ar-H), 7.40–7.31 (m, 3H, Ar-H), 2.74–2.69 (m, 6H,
2CH3). 13C NMR (125 MHz, DMSO) δ 165.3, 156.9, 152.4, 148.1, 144.1,
135.3, 131.7, 131.3, 131.0, 130.8, 130.7, 130.4, 129.1, 129.0, 128.9,
128.8, 128.1, 125.4, 125.2, 123.2, 121.9, 120.9, 120.7, 20.9, 18.0; Anal.
Calcd for C25H18F3NO2 (421.13): C, 71.25; H, 4.31; N, 3.32; Found: C,
71.16; H, 4.26; N, 3.37%.
4.1.1.1. Synthesis of 4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)
phenol. To a mixture of (1 g, 0.0091 mol) 4-aminophenol and (1.1 eq,
2.5 gm) B2pin2, (10–15 ml) acetonitrile (ACN) was added as a solvent
and 2 ml t-butyl nitrate was added dropwise in an ice bath which acts as
an activator. After addition, the ice bath was removed and the mixture
was stirred for 15 min at room temperature and then refluxed at 80 ◦C
for 24 h. The reaction was monitored by TLC. After completion of the
reaction, the mixture was set to cool at room temperature. The solvent
was diluted with 20 ml ethyl acetate and washed with 10 ml water. The
organic layer was extracted and treated with anhydrous Na2SO4 to
remove water. The solvent was evaporated using a rotary vacuum
evaporator. The sticky brown compound obtained was purified by col-
umn chromatography using 40% EA:nH mobile phase and was sent for
mass, 1H and 13C NMR analysis.
4.1.1.3.4. 4-(6-chloro-4-methylquinolin-2-yl)phenyl
3-(tri-
fluoromethyl)benzoate (5d). Brownish solid, 109.28 ◦C, MS (ESI) m/z:
for C24H15ClF3NO2 [M+H]+: 441.07, 1H NMR (500 MHz, Chloroform-d)
δ 8.28 (t, J = 2.2 Hz, 1H, Ar-H), 8.22 (d, J = 2.3 Hz, 1H, Ar-H), 8.10–8.04
(m, 3H, Ar-H), 7.92 (d, J = 8.1 Hz, 1H, Ar-H), 7.75–7.65 (m, 2H, Ar-H),
7.56 (dd, J = 10.7, 7.4 Hz, 1H, Ar-H), 7.44 (s, 1H, Ar-H), 7.40–7.34 (m,
2H, Ar-H), 2.71 (s, 3H, CH3). 13C NMR (125 MHz, DMSO) δ 165.3,
156.4, 152.4, 146.6, 141.9, 131.9, 131.7, 131.2, 131.0, 130.8, 130.7,
130.4, 129.8, 129.1, 129.1, 128.9, 128.7, 128.1, 127.3, 124.8, 121.9,
121.9, 120.9, 20.9; Anal. Calcd for C24H15ClF3NO2 (441.07): C, 65.24;
H, 3.42; N, 3.17; Found: C, 65.15; H, 3.47; N, 3.21%.
4.1.1.2. Synthesis of N-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)
phenyl)benzamide. In a RBF, 1.37 g of 4-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)phenol and (0.5gm) sodium methoxide (base) was
mixed with ACN as solvent and pH of the reaction was kept basic. The
substituted benzoylchloride (1.06 ml) was added drop wise. Further
reaction was stirred for 2 h at room temperature and monitored by TLC
making pH of the reaction mixture acidic by Glacial Acetic acid. The
reaction mixture was left overnight and monitored by thin layer chro-
matography (TLC). After completion of reaction, product was isolated in
water-ethyl acetate. The organic layer was extracted and treated with
anhydrous Na2SO4 to remove water. The solvent was evaporated in
rotary vacuum evaporator. The obtained intermediate was purified by
column chromatography using 40% EA:nH mobile phase and was sent
for mass, 1H and 13C NMR analysis.
4.1.1.3.5. 4-(4-methylquinolin-2-yl)phenyl 4-(trifluoromethyl)benzo-
ate (5e). Dark brown solid, 166.84 ◦C, MS (ESI) m/z: for C24H16F3NO2
[M+H]+: 407.11, 1H NMR (500 MHz, Chloroform-d) δ 8.10–8.06 (m,
3H, Ar-H), 8.06–8.01 (m, 2H, Ar-H), 7.97–7.92 (m, 1H, Ar-H), 7.88–7.81
(m, 2H, Ar-H), 7.70–7.63 (m, 1H, Ar-H), 7.54–7.46 (m, 2H, Ar-H),
7.40–7.34 (m, 2H, Ar-H), 2.73 (s, 3H, CH3). 13C NMR (125 MHz,
DMSO) δ 164.7, 156.8, 152.4, 148.4, 141.9, 134.3, 131.9, 131.5, 131.5,
130.2, 129.1, 128.9, 128.5, 128.1, 127.1, 126.5, 125.1, 124.9, 124.5,
122.8, 121.9, 121.6, 19.2;Anal. Calcd for C24H16F3NO2 (407.11): C,
70.76; H, 3.96; N, 3.44; Found: C, 70.85; H, 3.91; N, 3.40%.
4.1.1.3.6. 4-(4-phenylquinolin-2-yl)phenyl 4-(trifluoromethyl)benzo-
ate (5f). Brownish solid, 149.61 ◦C, MS (ESI) m/z: for C29H18F3NO2
[M+H]+: 469.13, 1H NMR (500 MHz, Chloroform-d) δ 8.19 (dd, J = 7.3,
0.8 Hz, 1H, Ar-H), 8.14–8.01 (m, 5H, Ar-H), 7.86 (d, J = 1.7 Hz, 2H, Ar-
H), 7.85–7.75 (m, 2H, Ar-H), 7.69 (ddd, J = 8.3, 7.0, 1.3 Hz, 1H, Ar-H),
7.56–7.43 (m, 4H, Ar-H), 7.41–7.35 (m, 3H, Ar-H). 13C NMR (125 MHz,
DMSO) δ 164.7, 155.9, 152.4, 148.9, 141.0, 138.5, 134.3, 134.1, 133.8,
133.6, 132.2, 131.5, 131.5, 131.5, 129.3, 128.9, 128.7, 128.2, 128.1,
127.8, 127.6, 127.2, 126.6, 126.5, 125.1, 124.9, 122.8, 121.9, 119.4;
Anal. Calcd for C29H18F3NO2 (469.13): C, 74.20; H, 3.86; N, 2.98;
Found: C, 74.11; H, 3.81; N, 2.94%.
4.1.1.3. Synthesis of substituted N-(4-(quinolin-2-yl)phenyl)benzamide.
The
substitutedN-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)
phenyl)benzamide (1.08gm, 0.0033 mol) was mixed with (0.58gm)
Quinoline derivatives anddi-isopropylaminewasused as solvent to
obtain basic conditions. Tetrakistriphenylphosphopalladium (0.2gm, 5
mol %) was used as a catalyst and reaction was refluxed at 80 ◦C. The
reaction was monitored by TLC. After completion of reaction, product
was isolated with layers of water-ethyl acetate. The organic layer was
extracted and treated with anhydrous Na2SO4 to remove water. The
solvent was evaporated in rotary vacuum evaporator. The obtained in-
termediate was purified by column chromatography using 40% EA:nH
mobile phase and was sent for mass, 1H and 13C NMR analysis.
4.1.1.3.7. 4-(4,8-dimethylquinolin-2-yl)phenyl-4-(trifluoromethyl)ben-
zoate (5g). Brown solid, 131.23 ◦C, MS (ESI) m/z: for C25H18F3NO2
[M+H]+: 421.13, 1H NMR (500 MHz, Chloroform-d) δ 8.09–8.01 (m,
4H, Ar-H), 7.89–7.81 (m, 3H, Ar-H), 7.71 (d, J = 8.6 Hz, 1H, Ar-H), 7.44
4.1.1.3.1. 4-(4-methylquinolin-2-yl)phenyl 3-(trifluoromethyl)benzo-
ate (5a). Dark brown solid, 128.21 ◦C, MS (ESI) m/z: for
C
24H16F3NO2[M+H]+: 407.11, 1H NMR (500 MHz, Chloroform-d) δ
(s, 1H, Ar-H), 7.40–7.31 (m, 3H, Ar-H), 2.74–2.69 (m, 6H, 2CH3). 13
C
8.10–8.04 (m, 4H, Ar-H), 7.97–7.92 (m, 1H, Ar-H), 7.72–7.63 (m, 2H,
Ar-H), 7.59–7.46 (m, 3H, Ar-H), 7.40–7.34 (m, 2H, Ar-H), 2.73 (s, 3H,
CH3). 13C NMR (125 MHz, DMSO) δ 165.3, 156.8, 152.4, 148.4, 141.9,
131.9, 131.0, 130.8, 130.4, 130.2, 129.1, 129.0, 128.9, 128.7, 128.5,
128.1, 128.0, 127.3, 125.1, 124.5, 121.9, 121.6, 120.9, 19.2;Anal. Calcd
for C24H16F3NO2 (407.11): C, 70.76; H, 3.96; N, 3.44; Found: C, 70.85;
H, 4.01; N, 3.39%.
NMR (125 MHz, DMSO) δ 164.7, 156.9, 152.4, 148.1, 144.1, 135.3,
134.3, 134.1, 133.8, 131.5, 131.5, 131.3, 130.7, 129.1, 128.9, 128.1,
126.5, 125.4, 123.2, 122.7, 121.9, 120.6, 20.9, 18.0; Anal. Calcd for
C
25H18F3NO2 (421.13): C, 71.25; H, 4.31; N, 3.32; Found: C, 71.14; H,
4.26; N, 3.28%.
4.1.1.3.8. 4-(6-chloro-4-methylquinolin-2-yl)phenyl
4-(tri-
fluoromethyl)benzoate (5h). Light brown solid, 121.83 ◦C, MS (ESI) m/z:
for C24H15ClF3NO2 [M+H]+: 441.07, 1H NMR (500 MHz, Chloroform-d)
δ 8.22 (d, J = 2.3 Hz, 1H, Ar-H), 8.10–8.01 (m, 4H, Ar-H), 7.92 (d, J =
8.1 Hz, 1H, Ar-H), 7.88–7.81 (m, 2H, Ar-H), 7.72 (dd, J = 8.2, 2.6 Hz,
1H, Ar-H), 7.44 (s, 1H, Ar-H), 7.40–7.34 (m, 2H, CH2), 2.71 (s, 3H,
CH3). 13C NMR (125 MHz, DMSO) δ 164.7, 156.4, 152.4, 146.6, 141.8,
134.3, 134.1, 133.8, 133.6, 131.9, 131.5, 131.5, 131.1, 130.7, 129.8,
129.1, 128.9, 128.1, 126.5, 124.8, 121.9, 121.9, 120.6, 20.9; Anal.
Calcd for C24H15ClF3NO2 (441.07): C, 65.24; H, 3.42; N, 3.17; Found: C,
65.33; H, 3.47; N, 3.21%.
4.1.1.3.2. 4-(4-phenylquinolin-2-yl)phenyl 3-(trifluoromethyl)benzo-
ate (5b). Brown solid, 149.61 ◦C, MS (ESI) m/z: for C29H18F3NO2
[M+H]+: 469.13, 1H NMR (500 MHz, Chloroform-d) δ 8.28 (t, J = 2.2
Hz, 1H), 8.19 (dd, J = 7.3, 0.8 Hz, 1H, Ar-H), 8.14–8.04 (m, 4H, Ar-H),
7.86 (s, 1H, Ar-H), 7.82–7.75 (m, 1H, Ar-H), 7.73–7.65 (m, 2H, Ar-H),
7.59–7.43 (m, 5H, Ar-H), 7.42–7.35 (m, 3H, Ar-H). 13C NMR (125
MHz, DMSO) δ 165.3, 155.9, 152.4, 148.9, 141.0, 138.5, 132.2, 131.8,
131.5, 131.2, 131.0, 130.8, 130.4, 129.3, 129.0, 128.9, 128.8, 128.7,
128.2, 128.1, 128.1, 127.8, 127.6, 127.2, 126.6, 125.2, 125.1, 121.9,
120.9, 119.4; Anal. Calcd for C29H18F3NO2 (469.13): C, 74.20; H, 3.86;
N, 2.98; Found: C, 74.09; H, 3.81; N, 2.94%.
4.1.1.3.9. 4-(4-methylquinolin-2-yl)phenyl 4-(chloromethyl)benzoate
(5i). Brownish white solid, 112.57 ◦C, MS (ESI) m/z: for C24H18ClNO2
13

K.S. Karnik et al.
Bioorganic Chemistry 107 (2021) 104612
[M+H]+: 387.10, 1H NMR (500 MHz, Chloroform-d) δ 8.10–7.99 (m,
5H, Ar-H), 7.97–7.92 (m, 1H, Ar-H), 7.70–7.63 (m, 1H, Ar-H), 7.54–7.46
(m, 2H, Ar-H), 7.42–7.34 (m, 4H, Ar-H), 4.56 (t, J = 1.0 Hz, 2H, CH2),
2.73 (s, 3H, CH3). 13C NMR (125 MHz, DMSO) δ 164.74, 156.79,
152.43, 148.40, 141.89, 141.79, 131.86, 131.53, 130.23, 129.19,
129.07, 128.95, 128.68, 128.48, 126.02, 125.10, 124.48, 123.92,
121.90, 121.61, 45.53, 19.20; Anal. Calcd for C24H18ClNO2 (387.10): C,
74.32; H, 4.68; N, 3.61; Found: C, 74.23; H, 4.73; N, 3.57%.
2CH3). 13C NMR (125 MHz, DMSO) δ 164.7, 156.9, 153.3, 152.4, 148.0,
144.1, 135.3, 132.3, 131.3, 130.7, 129.1, 128.9, 125.4, 123.2, 121.9,
120.9, 120.7, 112.0, 40.3, 20.9, 18.0; Anal. Calcd for C26H24N2O2
(396.18): C, 78.76; H, 6.10; N, 7.07; Found: C, 78.67; H, 6.15; N, 7.11%.
4.1.1.3.16. 4-(6-chloro-4-methylquinolin-2-yl)phenyl
4-(dimethyla-
mino)benzoate (5p). Brown solid, 121.83 ◦C, MS (ESI) m/z: for
C
25H21ClN2O2 [M+H]+: 416.13, 1H NMR (500 MHz, Chloroform-d) δ
8.22 (d, J = 2.3 Hz, 1H, Ar-H), 8.10–8.04 (m, 2H, Ar-H), 7.92 (ddd, J =
8.2, 3.1, 1.7 Hz, 3H, Ar-H), 7.72 (dd, J = 8.2, 2.6 Hz, 1H, Ar-H), 7.44 (s,
1H, Ar-H), 7.40–7.34 (m, 2H, Ar-H), 6.61–6.56 (m, 2H, Ar-H), 2.98 (s,
6H, 2CH3), 2.71 (s, 3H, CH3). 13C NMR (125 MHz, DMSO) δ 164.7,
156.4, 153.3, 152.4, 146.6, 141.9, 132.3, 131.9, 131.2, 130.7, 129.8,
129.1, 128.9, 124.8, 121.9, 121.9, 120.9, 112.0, 40.3, 20.9; Anal. Calcd
for C25H21ClN2O2 (416.13): C, 72.02; H, 5.08; N, 6.72; Found: C, 72.11;
H, 5.03; N, 6.67%.
4.1.1.3.10. 4-(4-phenylquinolin-2-yl)phenyl 4-(chloromethyl)benzoate
(5j). Brownish white solid, 105.34 ◦C, MS (ESI) m/z: for C29H20ClNO2
[M+H]+: 449.12, 1H NMR (500 MHz, Chloroform-d) δ 8.19 (dd, J = 7.3,
0.8 Hz, 1H, Ar-H), 8.14–7.99 (m, 5H, Ar-H), 7.86 (s, 1H, Ar-H),
7.82–7.75 (m, 1H, Ar-H), 7.69 (ddd, J = 8.3, 7.0, 1.3 Hz, 1H, Ar-H),
7.56–7.43 (m, 4H, Ar-H), 7.39 (ddt, J = 9.3, 6.0, 1.3 Hz, 5H, Ar-H),
4.56 (t, J = 1.0 Hz, 2H, CH2). 13C NMR (125 MHz, DMSO) δ 164.74,
155.95, 152.43, 148.86, 141.79, 141.01, 138.47, 132.22, 131.53,
129.27, 129.19, 128.94, 128.68, 128.17, 127.81, 127.65, 127.18,
126.62, 125.09, 121.90, 119.44, 45.53; Anal. Calcd for C29H20ClNO2
(449.12): C, 77.42; H, 4.48; N, 3.11; Found: C, 77.33; H, 4.43; N, 3.16%.
4.1.1.3.11. 4-(4,8-dimethylquinolin-2-yl)phenyl 4-(chloromethyl)ben-
zoate (5k). Brown solid, 116.36 ◦C, MS(ESI) m/z: for C25H20ClNO2
[M+H]+: 401.12, 1H NMR (500 MHz, Chloroform-d) δ 8.09–7.99 (m,
4H, Ar-H), 7.89–7.84 (m, 1H, Ar-H), 7.71 (d, J = 8.6 Hz, 1H, Ar-H),
7.46–7.31 (m, 6H, Ar-H), 4.56 (t, J = 1.0 Hz, 2H, CH2), 2.74–2.69 (m,
6H, 2CH3). 13C NMR (125 MHz, DMSO) δ 164.7, 156.9, 152.4, 148.1,
144.1, 141.8, 135.3, 131.5, 131.3, 130.7, 129.2, 129.1, 128.9, 128.7,
125.4, 123.2, 121.9, 120.7, 45.5, 20.9, 18.0; Anal. Calcd for
4.2. Insilico methodology
All computational studies were performed using Schrodinger Mo-
lecular Modelling software. The molecular docking tool, Glide, was used
for docking studies of all compounds on EGFR enzyme. Qikprop tool was
used to determine ADMET study. The AutoQSAR was used for building
of models. Molecular Dynamic Simulation was carried out using
Desmond.
4.2.1. Molecular docking
To investigate the binding affinity, binding mode and molecular in-
teractions of the synthesised compounds, molecular docking was carried
out using Maestro Glide. The compounds were designed on Marvin
sketch and saved in Triposmol2 format. The compounds were then
subjected to ligand preparation for generating conformations. After the
ligand preparation, the proteins were downloaded from RCSB Protein
Data Bank and were subjected to preparation for optimization and
minimization. Hydrogen was added to heavy atoms, all water molecules
and heteroatoms were then removed keeping native ligands in the active
site. Final structures were optimized and finally minimized using
OPLS3e force field to avoid steric clashes between atoms. The minimized
proteins were then further subjected to receptor grid generation for
determining the docking site. The grid was set to default settings and
defined by catering at the co-crystallised ligand. Finally, docking studies
were performed using glide docking tool by setting docking mode to
extra precision [42].
C
25H20ClNO2 (401.12): C, 74.72; H, 5.02; N, 3.49; Found: C, 74.61; H,
4.97; N, 3.53%.
4.1.1.3.12. 4-(6-chloro-4-methylquinolin-2-yl)phenyl-4-(chlor-
omethyl)benzoate (5l). Light brown solid, 135.71 ◦C, MS (ESI) m/z: for
C24H17Cl2NO2 [M+H]+: 421.06,1H NMR (500 MHz, Chloroform-d) δ
8.22 (d, J = 2.3 Hz, 1H, Ar-H), 8.10–7.99 (m, 4H, Ar-H), 7.92 (d, J = 8.1
Hz, 1H, Ar-H), 7.72 (dd, J = 8.2, 2.6 Hz, 1H, Ar-H), 7.46–7.34 (m, 5H,
Ar-H), 4.56 (t, J = 1.0 Hz, 2H, CH2), 2.71 (s, 3H, CH3). 13C NMR (125
MHz, DMSO) δ 164.7, 156.4, 152.4, 146.6, 141.9, 141.8, 131.9, 131.5,
131.2, 130.7, 129.8, 129.2, 129.1, 128.9, 128.7, 124.8, 121.9, 121.9,
45.5, 20.9; Anal. Calcd for C24H17Cl2NO2 (421.06): C, 68.26; H, 4.06; N,
3.32; Found: C, 68.37; H, 4.01; N, 3.36%.
4.1.1.3.13. 4-(4-methylquinolin-2-yl)phenyl 4-(dimethylamino)benzo-
ate (5m). Dark brown solid, 106.93 ◦C, MS (ESI) m/z: for C25H22N2O2
[M+H]+: 382.17, 1H NMR (500 MHz, Chloroform-d) δ 8.10–8.04 (m,
3H, Ar-H), 7.97–7.90 (m, 3H, Ar-H), 7.70–7.63 (m, 1H, Ar-H), 7.54–7.46
(m, 2H, Ar-H, Ar-H), 7.40–7.34 (m, 2H, Ar-H), 6.61–6.56 (m, 2H, Ar-H),
2.98 (s, 6H, 2CH3), 2.73 (s, 3H, CH3). 13C NMR (125 MHz, DMSO) δ
164.7, 156.8, 153.3, 152.4, 148.4, 141.9, 132.3, 131.9, 130.2, 129.1,
128.9, 128.5, 125.1, 124.5, 121.9, 121.6, 120.9, 112.0, 40.3, 19.2; Anal.
Calcd for C25H22N2O2 (382.17): C, 81.06; H, 5.44; N, 6.30; Found: C,
80.95; H, 5.39; N, 6.34%.
4.2.2. Qikprop
The ADMET results were generated using Qikprop tool.
4.2.3. Molecular dynamic simulations
To understand the stability and interactions affinity of the ligand-
bound protein, the most potent compound 5j, screened based on insili-
co and invitro results were subjected to molecular dynamics simula-
tions. The ligand-bound protein was immersed in a TIP3P orthorhombic
box of 10 Å_10Å_10 Å and system was then neutralized by the system-
built option in the protocol by adding 0.15 M NaCl in a buffer. Then,
steepest-descent and conjugate-gradient minimization methods were
utilized to a satisfactory convergence of 0.1 kcal/mol-Å energy gradient.
NPT ensemble class with a temperature of 300 K and a pressure of
1.01325 bar was used. Using VIPARR facility of DESMOND, the best
suitable AMBER99 force field parameters were generated for duplex
while the OPLS2005 force field was used for ligands [43].
4.1.1.3.14. 4-(4-phenylquinolin-2-yl)phenyl 4-(dimethylamino)benzo-
ate (5n). Brown solid, 129.16 ◦C, MS (ESI) m/z: for C30H24N2O2
[M+H]+: 444.18, 1H NMR (500 MHz, Chloroform-d) δ 8.19 (dd, J = 7.3,
0.8 Hz, 1H, Ar-H), 8.14–8.04 (m, 3H, Ar-H), 7.95–7.90 (m, 2H, Ar-H),
7.86 (s, 1H, Ar-H), 7.82–7.75 (m, 1H, Ar-H), 7.69 (ddd, J = 8.3, 7.0,
1.3 Hz, 1H, Ar-H), 7.56–7.43 (m, 4H, Ar-H), 7.41–7.35 (m, 3H, Ar-H),
6.61–6.56 (m, 2H, Ar-H), 2.98 (s, 6H, 2CH3). 13C NMR (125 MHz,
DMSO) δ 164.7, 155.9, 153.3, 152.4, 148.9, 141.0, 138.5, 132.3, 132.2,
129.3, 128.9, 128.7, 128.2, 127.8, 127.6, 127.2, 126.6, 125.1, 121.9,
120.9, 119.4, 112.0, 40.3; Anal. Calcd for C30H24N2O2 (444.18): C,
81.06; H, 5.44; N, 6.30; Found: C, 80.95; H, 5.39; N, 6.34%.
4.1.1.3.15. 4-(4,8-dimethylquinolin-2-yl)phenyl4(dimethylamino)ben-
zoate (5o). Brown solid, 130.18 ◦C, MS (ESI) m/z: for C26H24N2O2
[M+H]+: 396.18, 1H NMR (500 MHz, Chloroform-d) δ 8.09–8.03 (m,
2H, Ar-H), 7.95–7.90 (m, 2H, Ar-H), 7.89–7.84 (m, 1H, Ar-H), 7.71 (d, J
= 8.6 Hz, 1H, Ar-H), 7.44 (s, 1H, Ar-H), 7.40–7.31 (m, 3H, Ar-H),
6.61–6.56 (m, 2H, Ar-H), 2.98 (s, 6H, 2CH3), 2.74–2.69 (m, 6H,
4.2.4. AutoQSAR
QSAR models have been built using AutoQSAR tool. The panel in-
cludes the type of prediction property using which the models are to be
built, the selection of random training set which were used to build
QSAR models and selection of random validation set used for validating
the built QSAR models. To understand the methodology of the
14

K.S. Karnik et al.
Bioorganic Chemistry 107 (2021) 104612
AutoQSAR panel, a series of the workflow was followed. First, the
compounds generated from Ligprep were put in the AutoQSAR build
model panel. Then the models were subjected to a numerical prediction
property which determines the basic property of the structures to be
used. In the prediction property, the experimental IC50 values were
converted to pIC50 values to build the models. The training set was
randomly chosen by setting it to 75% of the total. The test set was
automatically set to 25% of the compounds left from the training set.
The validation set was used similar to the training set option and 30% of
the total were chosen to validate models. Thus, of the 24 compounds as
the data set, 4 compounds were used as a training set, which was tested
using the 20 compounds of the test set and further validated using 8
compounds of the validation set. Maximum pair correlation was set to
0.80 while other parameters were kept as default. The execution of the
data input resulted in 10 best QSAR models each consisting of 16
compounds exhibiting predicted activity compared with the experi-
mental activity. The output model sets were ranked according to the
model score, Coefficient of determination (R2) and Q2 explained and
determined using different formulae in the literature. Obtained Q2
values were compared with R2 values [44].
antibodies were used at 1:5000. The results were detected by an
enhanced chemiluminescence system (Millipore) [45].
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgement
The authors K. S. K and P. S. W are thankful to Indian Council of
Medical Research (ICMR) for providing funding in support of Senior
Research Fellowship with Ref No: 3/2/2/33/2018/Online OncoFship/
NCD-III. The authors are thankful to The Head, Department of Chemi-
cal Technology, Dr. Babasaheb Ambedkar Marathwada University,
Aurangabad 431 004 (MS), India, for providing the laboratory facility.
The author P. V. L. S. B is thankful to Indian Council of Medical Research
(ICMR) [ISRM/12(07)/2019] and DST-SERB [CRG/2018/003276] for
providing financial support. The authors are also thankful to The Head,
Department of Biotechnology, KLEF University, Vaddeswaram 522502
(AP), India for the computational support.
4.3. Invitro methodology
4.3.1. MTT assay
Appendix A. Supplementary data
Virtually screened compounds had been screened against HCC827,
H1975, A549, and HT-29 cells by standard MTT assay. These cell lines
were cultured in 10% fetal bovine serum (FBS). In exact 4 × 103 cells
have been suspended in MEM medium and plated into the 96-well plate
in 5% CO2 for 24 h. The synthesized compounds were added to the
culture medium and cells cultured maintained for 72 h. Freshly prepared
MTT was added to each well and incubated with cells at 37 ◦C for 4 h.
MTT reduction was quantified by measurement of absorbance at 492 nm
using a multimode reader (Synergy Mx, BioTek).
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bioorg.2020.104612.
References
[1] B.R. Voldborg, L. Damstrup, M.S. Thomsen, H.S. Poulsen, Epidermal growth factor
receptor (EGFR) and EGFR mutations, function and possible role in clinical trial,
Ann. Oncol. 12 (1997) 1197–1206.
[2] K.S. Kolibaba, B.J. Druker, Protein tyrosine kinases and cancer, BBA 1333 (1997)
F217–F248.
4.3.2. In vitro enzymatic activity assay
[3] J. Stamos, M.X. Sliwkowiski, C. Eigenbrot, Structure of the Epidermal Growth
Factor Receptor Kinase Domain Alone and in Complex with a 4-Anilinoquinazoline
Inhibitor, J. Biol. Chem. 277 (2002) 46265–46272.
The assay was performed as reported by Kashem[45]. All of the
enzymatic reactions were conducted at 30℃ for 40 min. The 50 µL re-
action mixture contains 40 mMTris, pH 7.4, 10 mM MgCl2, 0.1 mg/mL
BSA, 1 mM DTT, 10 µM ATP, 25 ng kinase and the 0.2 mg/ml enzyme-
substrate (Poly (Glu, Tyr)). The compounds were diluted in 10% DMSO
and 5 µL of the dilution was added to a 50 µL reaction so that the final
concentration of DMSO is 1% in all of the reactions. The assay was
performed by using the Kinase-Glo Plus luminescence kinase assay kit. It
measures kinase activity by quantitating the amount of ATP remaining
in solution following a kinase reaction. The luminescent signal from the
assay is correlated with the amount of ATP present and is inversely
correlated with the amount of kinase activity. The IC50 values were
calculated using nonlinear regression with normalized dose–response fit
using GraphPad Prism 5.0 Software.
[4] H. Ogiso, R. Ishitani, O. Neruki, S. Fuki, M. Yamanaka, J.H. Kim, K. Saito,
A. Satamako, M. Inoue, Crystal Structure of the Complex of Human Epidermal
Growth Factor and Receptor Extracellular Domains, Cell 110 (2002) 775–787.
[5] J. Baselga, C.L. Arteaga, Critical Update and Emerging Trends in Epidermal Growth
Factor Receptor Targeting in Cancer, J. Clin. Oncol. 23 (2005) 2445–2459.
[6] M.P. Santhi, G. Bupesh, V. SenthilKumar, K. Meenakumari, K. Prabhu,
S. Sugunthan, E. Manikandan, K. Saravanan, Anticancer activity and drug likeliness
of quinoline through insilico docking against cervical and liver cancer receptors,
Indian J. Med Res. Pharm. Sci. 3 (2016) 2349–5340.
[7] S.V. Sharma, D.W. Bell, J. Settleman, D.A. Haber, Epidermal growth factor receptor
mutations in lung cancer, Nat. Rev. Cancer 7 (2007) 169–181.
[8] K.R.A. Abdellatif, E.K.A. Abdelall, M.A. Abdelgawad, D.M.E. Amin, H.A. Omar,
Design, synthesis and biological evaluation of new 4-(4-substitutedanilino)
quinoline derivatives as anticancer agents, Med. Chem. Res. 26 (2017) 929–939.
[9] Liao, B. Chi, C Chi Lin, Yang, J.C. Hsin, Second and third-generation epidermal
growth factor receptor tyrosine kinase inhibitors in advanced non-small cell lung
cancer, Curr. Opin. Oncol. 27 (2015) 94–101.
[10] D. Westover, J. Zugazagoitia, B.C. Cho, C.M. Lovly, L. Paz-Ares, Mechanisms of
acquired resistance to first- and second-generation EGFR tyrosine kinase inhibitors,
Clin. Cancer Res. 29 (2008) i10–i19.
4.3.3. Western blot assay
HCC827 cells were seeded in 6-well plates at 1 × 106 per well,
incubated at 37 ◦C with 5% CO2 for 24 h before drug exposure. Cells
were treated with different concentrations of selected synthesized
compounds for 8 h, then collected and suspended in lysis buffer
(Beyotime) and centrifuged for 20 min at 12000 rpm, later the insoluble
material was removed. Same amounts of proteins were loaded and
separated by 8% SDS-PAGE and transferred to polyvinylidene fluoride
membranes (Millipore) after that. The anti-EGFR, anti-pEGFR
(Tyr1068), and anti-AKT were diluted at 1:1000, while the anti-pAKT
were diluted at 1:2000. All antibodies above were purchased from
Sigma Aldrich. Anti-β-actin was diluted at 1:5000 and the secondary
[11] L.V. Sequist, Second-Generation Epidermal Growth Factor Receptor Tyrosine
Kinase Inhibitors in Non-small Cell Lung Cancer, Oncologist 12 (2007) 325–330.
[12] K. Politi, D. Ayeni, T. Lynch, The Next Wave of EGFR Tyrosine Kinase Inhibitors
Enter the Clinic, Cancer Cell 27 (2015) 751–753.
[13] C. Tan, N.B. Kumarakulasinghe, Y. Huang, Y. Ang, J.R. Choo, B.C. Goh, R.A. Soo,
Third generation EGFR TKIs: current data and future directions, Mol. Cancer. 17
(2018) 1–14.
[14] A. Russo, T. Franchina, G.R.R. Ricciardi, V. Smiroldo, M. Picciotto, M. Zanghì,
C. Rolfo, V. Adamo, Third generation EGFR TKIs in EGFR-mutated NSCLC: Where
are we now and where are we going, Crit. Rev. Oncol. Hematol. 117 (2017) 38–47.
[15] R. Minari, P. Bordi, M. Tiseo, Third-generation epidermal growth factor receptor-
tyrosine kinase inhibitors in T790M-positive non-small cell lung cancer: review on
emerged mechanisms of resistance, Transl. Lung Cancer Res. 5 (2016) 695–708.
15

K.S. Karnik et al.
Bioorganic Chemistry 107 (2021) 104612
[16] H. Patel, R. Pawara, A. Ansari, S. Surana, Recent updates on third generation EGFR
inhibitors and emergence of fourth generation EGFR inhibitors to combat C797S
resistance, Eur. J. Med. Chem. 142 (2017) 32–47.
[31] S. Sivanandan, K. Jain, N. Plakkal, M. Bahl, T. Sahoo, S. Mukherjee, Y.K. Gupta,
R. Agarwal, Issues, challenges, and the way forward in conducting clinical trials
among neonates: investigators’ perspective, J. Perinatol. 3 (2019) 20–30.
[32] I.D. Angelis, L. Turco, Caco- 2 cells as a model for intestinal absorption, Curr.
Protoc. Toxicol. 47 (2011) 20–26.
[17] S. Wang, Y. Song, D. Liu, EAI045: The fourth-generation EGFR inhibitor
overcoming T790M and C797S resistance, Cancer Lett. 385 (2017) 51–54.
[18] J. Zhao, A. Guerrero, K. Kelnar, H.P. Peltire, A.G. Bader, Synergy between next
generation EGFR tyrosine kinase inhibitors and miR–34a in the inhibition of non-
small-cell lung cancer, Lung Cancer. 108 (2017) 96–102.
[33] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental and
computational approaches to estimate solubility and permeability in drug
discovery and development settings, Adv. DrugDeliv. Rev. 64 (2012) 4–17.
[34] D.K. Lokwani, R. Azad, A.P. Sarkate, P. Reddanna, D.B. Shinde, Structure Based
Library Design (SBLD) for new 1,4-dihydropyrimidine scaffold as simultaneous
COX-1/COX-2 and 5-LOX inhibitors, Bioorgan. Med. Chem. 23 (2015) 4533–4543.
[35] M.R. Bhosle, L.D. Khillare, J.R. Mali, A.P. Sarkate, D.K. Lokwani, S.V. Tiwari,
DIPEAc promoted one-pot synthesis of dihydropyrido[2,3-d:6,5-d′]
[19] Y. Zheng, W. Zhang, L. Meng, X. Wu, L. Zhang, L. An, C. Lia, C. Gao, L. Xu, Y. Liu,
C. Li, C. Gao, L. Xu, Y. Liu, Design, Synthesis and Biological Evaluation of 4-Aniline
QuinazolineDerivativesConjugated with Hydrogen Sulfide (H2S) Donors as Potent
EGFR Inhibitors against L858R Resistance Mutation, Eur. J. Med. Chem. 202
(2020), 112555.
[20] L.M. Antypenko, S.I. Kovalenko, O.V. Karpenko, A.M. Katsev, V.P. Novikov, N.
S. Fedyunina, 1-R-2-(1,2,4 Triazolo 1,5-c quinazoline-2-ylthio) etanon(ol) s:
Synthesis, Bioluminescence Inhibition, Molecular Docking Studies, Antibacterial
and Antifungal Activities, Curr. Comput. Aided Drug Des. 12 (2016) 29–41.
[21] M. Yulya, A. Oleksii, N. Inna, B. Galina, K. Sergii, Directed Search of Anti-
inflammatory Agents among (3H-Quinazoline-4-ylidene) hydrazides of N-protected
Amino acids and their Heterocyclization Products, Antiinflamm. Antiallergy
Agents Med. Chem. 19 (2020) 61–73.
dipyrimidinetetraone and pyrimido[4,5-d]pyrimidine derivatives as potent
tyrosinase inhibitors and anticancer agents: in vitro screening, molecular docking
and ADMET predictions, New J. Chem. 42 (2018) 18621–18632.
[36] S.V. Tiwari, J.A. Seijas, M.P. Vazquez-Tato, A.P. Sarkate, K.S. Karnik, A.P.
G. Nikalje, Ionic Liquid-Promoted Synthesis of Novel Chromone-Pyrimidine
Coupled Derivatives, Antimicrobial Analysis, Enzyme Assay, Docking Study and
Toxicity Study, Molecules 23 (2018) 1–23.
[37] A.F. Zaher, S.M. Abuel-Maaty, H.B. El-Nassan, S.A.S. Amer, T.M. Abdelghany,
Synthesis, antitumor screening and cell cycle analysis of novel benzothieno[3,2-b]
pyran derivatives, J. EnzymeInhib. Med. Chem. 31 (2016) 145–153.
[38] E.F. Abdelhaleem, M.K. Abdelhameid, A.B. Kassab, M.M. Kandeel, Design and
synthesis of thienopyrimidine urea derivatives with potential cytotoxic and pro-
apoptotic activity against breast cancer cell line MCF-7, Eur. J. Med. Chem. 143
(2018) 1807–1825.
[22] R. Musiol, M. Serda, S. Hensel-Bielowka, J. Polanski, Quinoline-Based Antifungals,
Curr. Med. 17 (2010) 1960–1973.
[23] R. Neelarapu, J.R. Maignan, C.L. Lichorowic, A. Monastyrskyi, T.S. Mutka, A.
N. LaCrue, L.D. Blake, D. Casandra, S. Mashkouri, J.N. Burrows, P.A. Willis, D.
E. Kyle, R. Manetsch, Design and Synthesis of Orally Bioavailable Piperazine
Substituted 4(1H)-Quinolones with Potent Antimalarial Activity: Structure-Activity
and Structure-Property Relationship Studies, J. Med. Chem. 61 (2018) 1450–1473.
[24] H.S.A. Al-Salem, G.H. Hegazy, K.E.H. El-Taher, S.M. El-Messery, A.M. Al-Obaid, H.
I. El-Subbagh, Synthesis, anticonvulsant activity and molecular modelling study of
some new hydrazinecarbothioamide, benzenesulfonohydrazide, and
phenacylacetohydrazide analogues of 4(3H)-quinazolinone, Bioorg. Med. Chem.
Lett. 25 (2015) 1490–1499.
[39] V.S. Dofe, A.P. Sarkate, R. Azad, C.H. Gill, Novel quinoline-based oxadiazole
derivatives induce G2/M arrest and apoptosis in human breast cancer MCF-7 cell
line, Res. Chem. Intermed. 43 (2017) 7331–7345.
[40] K.T. Chan, F.Y. Meng, Q. Li, C.Y. Cy, T.S. Lam, Y. To, W.H. Lee, M. Li, K.H. Chu,
M. Toh, Cucurbitacin B induces apoptosis and S phase cell cycle arrest in BEL-7402
human hepatocellular carcinoma cells and is effective via oral administration,
Cancer Lett. 294 (2010) 118–124.
[25] R.P. Modh, E. De Clercq, C. Pannecouque, K.H. Chikhalia, Design, synthesis,
antimicrobial activity and anti-HIV activity evaluation of novel hybrid quinazoline-
triazine derivatives, J. EnzymeInhib. Med. Chem. 29 (2014) 100–108.
[26] M.U. Rahman, A. Rathore, A.A. Siddiqui, G. Parveen, M.S. Yar, Synthesis and
characterization of quinazoline derivatives: search for hybrid molecule as diuretic
and antihypertensive agents, J. EnzymeInhib. Med. Chem. 29 (2014) 733–743.
[27] G.I. Solyanik, Quinazoline compounds for antitumor treatment, Exp. Oncol. 41
(2019) 3–6.
[41] H. Lei, S. Fan, H. Zhang, Y. Liu, Y. Hei, J. Zhang, A. Zheng, M. Xin, S. Zhang,
Discovery of novel 9-heterocyclyl substituted 9H-purines as L858R/T790M/C797S
mutant EGFR tyrosine kinase inhibitors, Eur. J. Med. Chem. 186 (2020) 1–43.
[42] K.S. Karnik, A.P. Sarkate, D.K. Lokwani, D.B. Shinde, Novel 2-(nitrooxy)ethyl 2-(4-
(substituted phenyl)-2-((substituted phenyl)amino)thiazol-5-yl)acetate as anti-
inflammatory, analgesic and nitric oxide releasing agents: Synthesis and Molecular
Docking studies, Antiinflamm. Antiallergy Agents Med. Chem. 16 (2017) 153–167.
[43] K.S. Karnik, A.P. Sarkate, D.K. Lokwani, I.S. Narula, P.V.L.S. Burra, Development of
triple mutant T790M/C797S allosteric EGFR inhibitors: a computational approach,
J. Biomol. Struct. Dyn. (2020), https://doi.org/10.1080/07391102.2020.1786460.
[44] K.S. Karnik, I.S. Narula, A.P. Sarkate, P.S. Wakte, Auto QSAR- A Fast Approach for
Creation and Application of QSAR Models through Automation, Chemistry select. 5
(2020) 5756–5762.
[28] B. Ganapathi, Narsaiah, Design and Synthesis of Novel Pyrimidine/
Hexahydroquinazoline-Fused Pyrazolo 3,4-b Pyridine Derivatives, Their Biological
Evaluation and Docking Studies, Chem. Select. 4 (2019) 138–144.
[29] E.-D. Li, Q. Lin, Y.-Q. Meng, L.-Y. Zhang, P.-P. Song, N. Li, J.-C. Xin, P. Yang, C.-
N. Bao, D.-Q. Zhang, Y. Zhang, J.-K. Wang, Q.-R. Zhang, H.-M. Liu, 2,4-
Disubstituted quinazolines targeting breast cancer cells via EGFR-PI3K, Eur. J.
Med. Chem. 172 (2019) 36–47.
[45] M.A. Kashem, R.M. Nelson, J.D. Yingling, S.S. Pullen, A.S. Prokopowicz III, J.
W. Jones, J.P. Wolak, G.R. Rogers, M.M. Morelock, R.J. Snow, C.A. Homon,
S. Jakes, Three mechanistically distinct kinase assays compared: measurement of
intrinsic ATPase activity identified the most comprehensive set of ITK inhibitors,
J. Biomol. Screen. 12 (2007) 70–83.
[30] S.M. Prajapati, K.D. Patel, R.H. Vekariya, S.N. Panchal, H.D. Patel, Recent advances
in the synthesis of quinolines: a review, RSC Adv. 4 (2014) 24463–24476.
16