molecules
Article
Quinoxalinones as A Novel Inhibitor Scaffold for EGFR
(L858R/T790M/C797S) Tyrosine Kinase: Molecular Docking,
Biological Evaluations, and Computational Insights
Utid Suriya 1 , Panupong Mahalapbutr 2,* , Watchara Wimonsong 3, Sirilata Yotphan 3, Kiattawee Choowongkomon 4,*
and Thanyada Rungrotmongkol 5,6, *
Citation: Suriya, U.; Mahalapbutr, P.;
Wimonsong, W.; Yotphan, S.;
Choowongkomon, K.;
Rungrotmongkol, T. Quinoxalinones
as A Novel Inhibitor Scaffold for
EGFR (L858R/T790M/C797S)
Tyrosine Kinase: Molecular Docking,
Biological Evaluations, and
Computational Insights. Molecules
2022, 27, 8901. https://doi.org/
10.3390/molecules27248901
Academic Editors: Letizia Crocetti
and Maris Cinelli
Received: 6 November 2022
Accepted: 7 December 2022
Published: 14 December 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Molecules 2022, 27, 8901. https://doi.org/10.3390/molecules27248901
1 Program in Biotechnology, Faculty of Science, Chulalongkorn University, Bangkok 10330, Thailand
2 Department of Biochemistry, Center for Translational Medicine, Faculty of Medicine, Khon Kaen University,
Khan Kaen 40002, Thailand
3 Department of Chemistry, Center of Excellence for Innovation in Chemistry, Faculty of Science, Mahidol
University, Rama VI Road, Bangkok 10400, Thailand
4 Department of Biochemistry, Kasetsart University, Bangkok 10900, Thailand
5 Department of Biochemistry, Center of Excellence in Structural and Computational Biology,
Chulalongkorn University, Bangkok 10330, Thailand
6 Program in Bioinformatics and Computational Biology, Graduate School, Chulalongkorn University,
Bangkok 10330, Thailand
* Correspondence: panupma@kku.ac.th (P.M.); kiattawee.c@ku.th (K.C.); thanyada.r@chula.ac.th (T.R.)
Abstract: Combating acquired drug resistance of EGFR tyrosine kinase (TK) is a great challenge
and an urgent necessity in the management of non-small cell lung cancers. The advanced EGFR
(L858R/T790M/C797S) triple mutation has been recently reported, and there have been no specific
drugs approved for this strain. Therefore, our research aimed to search for effective agents that could
impede the function of EGFR (L858R/T790M/C797S) TK by the integration of in silico and in vitro
approaches. Our in-house quinoxalinone-containing compounds were screened through molecular
docking and their biological activity was then verified by enzyme- and cell-based assay. We found
that the four quinoxalinone-containing compounds including CPD4, CPD15, CPD16, and CPD21
were promising to be novel EGFR (L858R/T790M/C797S) TK inhibitors. The IC50 values measured
by the enzyme-based assay were 3.04 ± 1.24 nM; 6.50 ± 3.02 nM,10.50 ± 1.10 nM; and 3.81 ± 1.80 nM,
respectively, which are at a similar level to a reference drug; osimertinib (8.93 ± 3.01 nM). Besides
that, they displayed cytotoxic effects on a lung cancer cell line (H1975) with IC50 values in the range
of 3.47 to 79.43 µM. In this proposed study, we found that all screened compounds could interact
with M793 at the hinge regions and two mutated residues including M790 and S797; which may be
the main reason supporting the inhibitory activity in vitro. The structural dynamics revealed that the
screened compounds have sufficient non-native contacts with surrounding amino acids and could be
well-buried in the binding site’s cleft. In addition, all predicted physicochemical parameters were
favorable to be drug-like based on Lipinski’s rule of five, and no extreme violation of toxicity features
was found. Altogether, this study proposes a novel EGFR (L858R/T790M/C797S) TK inhibitor
scaffold and provides a detailed understanding of compounds’ recognition and susceptibility at the
molecular level.
Keywords: EGFR tyrosine kinase; EGFR (L858R/T790M/C797S) TK; lung cancer; non-small cell lung
cancer; in silico drug discovery and development
1. Introduction
Cancers pose a public health burden and challenge to all nations and remain a ma-
jor cause of mortality and morbidity. Lung cancer is the most common type of cancer
worldwide. It has been reported that more than 2.2 million were new cases; among
these non-small cell lung cancer (NSCLC) accounts for 84% of all newly diagnosed cases
https://www.mdpi.com/journal/molecules
Molecules 2022, 27, 8901 2 of 15

(https://www.cancer.org/, accessed on 30 October 2022). One of the major advancements

in the treatment of NSCLC came with the development of drugs inhibiting the biomarkers
which could be produced by tumors or our body in response to the malignancy. Specifically
targeting the up-regulated kinases such as the epidermal growth factor receptor tyrosine
kinase (EGFR TK) and interrupting the signaling intermediates such as mitogen-activated
protein kinase (MAPK) and Janus kinase (JAK)/STAT have been a promising strategy to
reverse the rate of cancer-related mortality.
EGFR TK, one of the members of the ErbB family, is a transmembrane glycoprotein
that plays a pivotal role in the regulation of cellular proliferation, homeostasis, growth, and
survival responses [1]. Structurally, the EGFR TK domain consists of four main elements [2]
including hinge region (residues 788–797), P loop (residues 712–731), C helix (residues
752–767), and activation loop (855–877) as illustrated in Figure 1. The signal transduction
begins when EGF binds to the ligand binding site and triggers homodimerization or het-
erodimerization with other ERBB members (e.g., HER2). Then, the receptor phosphorylates
and activates downstream effectors such as RAS–RAF–MEK–ERK–MAPK and PI3K–AKT–
mTOR signaling cascade [3,4]. Biological proven shreds of evidence have revealed the
correlation between the overexpression of EGFR and human malignancies leading to the
enhancement of tumor cell proliferation, growth, invasion, metastasis, angiogenesis, and
even apoptosis impairment in several types of cancers such as breast cancer, head and
neck cancer, lung cancer, etc. [5–8]. To this end, several pharmaceutical agents have been
developed for therapeutic purposes and now become a standard treatment for patients with
EGFR-positive non-small-cell lung cancer (NSCLC). Undoubtedly, similar to other proteins,
mutations have been reported during the long-term drug administration [9], which include
a single-point mutation in exon 21 (L858R) and exon-19 deletion, L858R/T790M double
mutation, and L858R/T790M/C797S triple mutation. Accordingly, currently available
drugs have been designed to fight against acquired drug resistance, which can be classified
into three generations: (i) the first generation: erlotinib [10] and gefitinib [11–13], which be-
long to the quinazoline-based scaffold (ii) the second generation: afatinib and dacomitinib
were developed to overcome acquired drug resistance caused by T790M mutation [14–16]
(iii) the third generation: WZ-4002 [17], CO-1686 (reciletinib) [18], and osimertinib [19] has
been developed to potently and specifically inhibit L858R/T790M double mutated EGFR.
Nevertheless, there has been reported that approximately 40% of NSCLC patients have
inevitably developed triple mutation, promoting a loss of inhibitory activity of previously
approved agents [20]. Only osimertinib, the FDA- and EMA-approved third-generation in-
hibitor, remains the first-line therapy and standard treatment for NSCLC patients carrying
EGFR mutations [19,21,22]. A rising incidence of drug resistance caused by the advanced
L858R/T790M/C797S triple mutation in NSCLC escalates the urge for the development
of the fourth generation of EGFR inhibitors, which aim to effectively deliver a significant
clinical response in patients with a triple mutation.
There are several potent compounds capable of inhibiting EGFR TK such as trisub-
stitued imidazoles [23], trisubstituted pyridinylimidazoles [24], 9-heterocyclyl substituted
9H-purine derivatives [25], quinoline derivatives [26], methylpyrimidopyridone scaf-
fold [27], pyrimidine derivatives [28], N2 -(4-(4-methylpiperazin-1-yl) phenyl)-N8 -phenyl-
9H-purine-2, 8-diamine [29], sulfonyl fluoride derivatives [30], anilino-1,4-naphthoquinones [31],
etc. Quinoxalinones have broad-ranging pharmacological activities including anticancer
properties [32]; however, there has been no report of the specific binding of quinoxali-
nones toward EGFR (L858R/T790M/C797S) triple mutant. Herein, we performed in silico
screening of in-house quinoxalinone derivatives which were successfully synthesized by
phenyliodine (III) diacetate (PIDA)-induced oxidative C-N bond coupling. Compounds
exhibiting a comparable level of binding affinity to osimertinib were then selected to be
tested experimentally. The potential compounds were predicted for their drug-likeness
properties and toxicity features. Then, the MD simulations were employed to gain insights
into binding recognition and susceptibility, guiding the mechanism of inhibitory action at
 by phenyliodine (III) diacetate (PIDA)-induced oxidative C-N bond coupling. Com-
pounds exhibiting a comparable level of binding affinity to osimertinib were then selected
to be tested experimentally. The potential compounds were predicted for their drug-like-
ness properties and toxicity features. Then, the MD simulations were employed to gain
Molecules 2022, 27, 8901 3 of 15
insights into binding recognition and susceptibility, guiding the mechanism of inhibitory
action at the atomic level. Altogether, our proposed quinoxalinone derivatives could be-
come viable candidates to be further developed as the effective fourth generation of EGFR
the atomic level. Altogether, our proposed quinoxalinone derivatives could become viable
inhibitors.
candidates to be further developed as the effective fourth generation of EGFR inhibitors.

3D structure
Figure 1. (A) 3D structure of EGFR (L858R/T790M/C797S)
(L858R/T790M/C797S) TK TKinincomplex
complexwith
withosimertinib
osimertinib (PDB
(PDB ID:
ID:
6LUD) in which the mutated amino acids were shown in a close-up view as well as a hydrophobic
character of its
character of its binding
bindingcleft.
cleft.(B)
(B)EGFR
EGFRsignaling
signalingpathways
pathways including
including Sos/Ras/Raf/MEK/ERK,
Sos/Ras/Raf/MEK/ERK,
PLC/PKC, and JAK/STAT signaling cascade that control cell differentiation,
PLC/PKC, and JAK/STAT signaling cascade that control cell differentiation, proliferation, and sur-
proliferation, and
vival responses [4,33]. (adapted from Ref. 4,33 and created with BioRender.com, accessed on 30 Oc-
survival responses [4,33]. (adapted from Refs. [4,33] and created with https://BioRender.com,
tober 2022).
accessed on 30 October 2022).

2.
2. Results
Results and
and Discussion
Discussion
2.1. Docking-Based Virtual Screening
To rapidly
To rapidly identify
identifythe
thepotential
potentialcompounds
compoundsfrom from3030in-house
in-housequinoxalinones
quinoxalinones capable
capa-
ble of binding to the TK domain of EGFR (L858R/T790M/C797S), molecular docking was
of binding to the TK domain of EGFR (L858R/T790M/C797S), molecular docking
performed by employing the Autodock Vina XB software package [34,35]. As shown in
Figure 2, the predicted binding energy of quinoxalinone-containing compounds was in a
range − 7.8 to
−7.8 −5.4kcal/mol
to −5.4 kcal/molwhilst
whilstthe
theosimertinib
osimertinibdrug
drug was predicted to be−
to be 7.4 kcal/mol.
−7.4 kcal/mol.
Selection of potential compounds was based on the predicted binding energy, which
Selection of potential compounds was based on the predicted binding energy, which is
is
theoretically proportional to the dissociation constant (kdd) and is widely used in several
theoretically proportional to the dissociation constant (k ) and is widely used in several
computer-aided drug
computer-aided drugdiscovery
discoverycampaigns.
campaigns.ForForthis
thispurpose,
purpose,compounds
compounds showing
showing binding
bind-
energy lower than − 7.0 kcal/mol were selected which resulted in four
ing energy lower than −7.0 kcal/mol were selected which resulted in four quinoxalinone-quinoxalinone-
containing compounds
containing compounds including
including CPD4,
CPD4, CPD15,
CPD15, CPD16,
CPD16, andand CPD21.
CPD21. NoteNote that
that their
their
chemical structures are shown in Figure
chemical structures are shown in Figure 3. 3.
Molecules 2021, 26, x FOR PEER REVIEW 4 of 16
Molecules
Molecules 2021,
2022, 26,
27, x8901
FOR PEER REVIEW 44 of
of 16
15

Figure
Figure2. 2.Docking-based
Docking-basedvirtual
virtualscreening
screeningofof30
30quinoxalinone-containing
quinoxalinone-containingcompounds
compoundsinincompari-
compar-
Figure 2. Docking-based virtual screening of 30 quinoxalinone-containing compounds in compari-
son
isontotothe
theknown
known drug,
drug,osimertinib. (A) (A)
osimertinib. TheThe
fourfour
compounds
compounds possessing a lower
possessing or equal
a lower to −7.0
or equal to
son to the known drug, osimertinib. (A) The four compounds possessing a lower or equal to −7.0
kcal/mol
− 7.0 were then
kcal/mol were selected
then to perform
selected to experiments
perform as highlighted
experiments as in red.
highlighted in (B) Superimposition
red. (B) of
Superimposition
kcal/mol were then selected to perform experiments as highlighted in red. (B) Superimposition of
the docked pose of all screened compounds and osimertinib.
of the
the docked
docked pose
pose of all
of all screened
screened compounds
compounds andand osimertinib.
osimertinib.

Figure 3. The chemical structures of selected compounds including CPD4, CPD15, CPD16, and
CPD213.
Figure asThe
wellchemical
as a reference drug (osimertinib).
structures of selected compounds including CPD4, CPD15, CPD16, and
Figure 3. The chemical structures of selected compounds including CPD4, CPD15, CPD16, and
CPD21 as well as a reference drug (osimertinib).
CPD21 as well as a reference drug (osimertinib).
Molecules 2021, 26, x FOR PEER REVIEW 5 of 16
Molecules 2022, 27, 8901 5 of 15

2.2. In Vitro Assay of EGFR (L858R/T790M/C797S) Inhibition

2.2. InThe
Vitro Assay ofcompounds
screened EGFR (L858R/T790M/C797S) Inhibition
from molecular docking were assayed for kinase inhibitory
The screened
activity, which was compounds
reported as from molecular docking
half-maximal inhibitory were assayed for (IC
concentration kinase
50) ininhibitory
compari-
activity, which
son to the wasdrug,
known reported as half-maximal
osimertinib. As showninhibitory
in Figureconcentration
4, the kinase (ICinhibitory
50 ) in comparison
activity
to the known
against EGFRdrug, osimertinib. As shown
(L858R/T790M/C797S) was in
in Figure 4, the kinase
a nanomolar scale in inhibitory
which the activity against
IC50 values of
EGFR (L858R/T790M/C797S) was in a nanomolar scale in which the
CPD4, CPD15, CPD16, and CPD21 were 3.04 ± 1.24 nM, 6.50 ± 3.02 nM, 10.50 ± 1.10 nM, IC 50 values of CPD4,
CPD15,
and 3.81CPD16, andrespectively,
± 1.80 nM, CPD21 werewhile ± 1.24 nM, showed
3.04 osimertinib 6.50 ± 3.02an IC nM, 10.50of±8.93
50 value 1.10±nM,
3.01 and
nM.
3.81 ± 1.80 nM, respectively, while osimertinib showed
Statistical analysis (Tukey’s test) showed no significant difference an IC value of 8.93 ± 3.01
50 in IC50 value when com- nM.
Statistical analysis (Tukey’s
pared to osimertinib test) showed
at 95% confidence; no CPD4,
thus, significant
CPD15, difference
CPD16, in andICCPD21
50 value when
could be
compared
very potenttoinhibitors
osimertinib at 95%asconfidence;
as similar the approved thus, CPD4,
drug, CPD15,The
osimertinib. CPD16, and activity
inhibitory CPD21
could bescreened
of these very potent inhibitors
compounds wasasfound
similar as similar
to be the approved
to several drug, osimertinib.
previously reported The in-
com-
hibitory activity of these screened compounds was found to be similar to
pounds such as trisubstitued imidazoles (20–300 nM) [23], 9-heterocyclyl substituted 9H-several previously
reported compounds
purine derivatives (18such
nM)as trisubstitued
[25], imidazoles (115–139
quinoline derivatives (20–300 nM) nM)[23],
[26],9-heterocyclyl
methylpyrim-
substituted
idopyridone scaffold (27.5 nM) [27], sulfonyl fluoride derivative (110(115–139
9H-purine derivatives (18 nM) [25], quinoline derivatives nM) [30],nM) [26],
anilino-
methylpyrimidopyridone scaffold (27.5
1,4-naphthoquinones (3.96–18.64 nM) [31], etc.nM) [27], sulfonyl fluoride derivative (110 nM) [30],
anilino-1,4-naphthoquinones (3.96–18.64 nM) [31], etc.

Figure4.4.Dose–response
Figure Dose–responsecurves
curvesofofkinase
kinaseinhibitory
inhibitoryactivity
activityagainst
againstEGFR
EGFR(L858R/T790M/C797S)
(L858R/T790M/C797S)
TK. Data were represented as mean ± SEM.
TK. Data were represented as mean ± SEM.

2.3.
2.3. Drug-Likeness
Drug-Likeness
The
Thepotent quinoxalinone-containing
potent compounds against
quinoxalinone-containing EGFR (L858R/T790M/C797S)
compounds against EGFR
were predicted the drug-likeness character based on Lipinski’s rule of
(L858R/T790M/C797S) were predicted the drug-likeness character based on Lipinski’s rule five by inspecting
their physicochemical
of five by inspecting their properties (molecular properties
physicochemical weight (MW), the numbers
(molecular weightof (MW),
hydrogenthe bond
num-
donors
bers of(HBD)
hydrogen and acceptors
bond donors (HBA), rotatable
(HBD) bond (RB),(HBA),
and acceptors polar surface area
rotatable (PSA)
bond andpolar
(RB), Log
P). According
surface to Table
area (PSA) and1,Log
theP).
predictive
According results revealed
to Table 1, the that all compounds
predictive could confer
results revealed that all
the drug-likeness
compounds could property
confer theand obey the acceptable
drug-likeness property and value
obeywithin the criteria
the acceptable valueofwithin
the rules
the
as follows: (i) molecular weight ≤ 500 Da, (ii) hydrogen bond donors ≤ 5,
criteria of the rules as follows: (i) molecular weight ≤ 500 Da, (ii) hydrogen bond donors ≤5,and hydrogen
acceptors ≤10, acceptors
and hydrogen (iii) rotatable
≤10,bond ≤10, (iv)bond
(iii) rotatable polar≤10,
surface
(iv) area 140 Å2 and
polar≤surface area (v) lipophilicity
≤140 Å2 and (v)
(expressed as Log P) ≤ 5. Therefore, we believed that these four compounds
lipophilicity (expressed as Log P) ≤5. Therefore, we believed that these four compounds were likely to
were
be developed as promising novel mutated EGFR inhibitors.
likely to be developed as promising novel mutated EGFR inhibitors.
Molecules 2022, 27, 8901 6 of 15

Table 1. Predicted Lipinski’s rule of five for the quinoxalinone-based compounds and osimertinib.
(HBD, hydrogen bond donor; HBA, hydrogen bond accepter; PSA, polar surface area).

Lipinski’s Rule of Five

Compounds M.W. (Da) HBD HBA Rotatable Bonds PSA (Å2 ) logP Drug-Likeness
CPD4 303.32 0 4 3 65.60 2.45 Yes
CPD15 290.32 0 3 2 52.71 2.83 Yes
CPD16 252.27 1 3 3 54.98 2.32 Yes
CPD21 280.32 0 3 4 44.12 3.00 Yes
Osimertinib 499.61 2 5 10 90.78 3.36 Yes

2.4. SASA, Number of H-Bonds, and Contact Atoms

To investigate the buried capacity of each compound within the cleft of the binding
site, the solvent-accessible surface area (SASA) was computed by inspecting the amounts of
water within the 5.0 Å from the ligand. As shown in Figure 5, the averaged SASA in the last
20 ns (considered reaching an equilibrated state, Figure S4) of each complex was in the order
of CPD4 (674 Å2) < CPD15 (730 Å2) < osimertinib (736Å2) < CPD21 (768 Å2) < CPD16 (866 Å2).
The smaller area of occupied solvents could imply the better-fitting ligand conformation
within the binding pocket and the binding events might not be interfered with by the
solvents. Therefore, CPD4 was the most well-buried to binding pocket; however, the
intermolecular interactions of CPD16 and surrounding amino acids might be probably
reduced. Apart from that, the numbers of H-bonds were analyzed during 80–100 ns. It
was found that quinoxalines could form fewer H-bonds when compared to osimertinib
(1–2 bonds), suggesting that this type of interaction was not dominantly responsible for
the quinoxalinone compounds’ recognition since they comprise gradually lower
Molecules 2021, 26, x FOR PEER REVIEW
numbers
7 of 16
of hydrogen bond donors and acceptors when compared to osimertinib (total numbers of
HBD and HBA are 3–4 while there are 7 for osimertinib, Table 1).

Figure 5. Plots of SASA and contact atoms counted within the 5.0 Å from the ligand as well as
Figure 5. Plots of SASA and contact atoms counted within the 5.0 Å from the ligand as well as num-
number
ber of of H-bonds
H-bonds by setting
by setting two criteria
two criteria as follows:
as follows: (1) the between
(1) the distance distancethe
between thebond
hydrogen hydrogen
donor bond
donor
(HD)(HD)
and and hydrogen
hydrogen acceptor
acceptor (HA) of(HA)
≤3.5 Å ≤3.5
of(2) the Å (2) the
angle angle ≥120◦ .
≥120°.

2.5. Hot-Spot Residues

To investigate the key binding residues within the focused site of EGFR tyrosine ki-
nase, the per-residue decomposition energy (ΔG ) was elucidated via the MM/PBSA
approach. The positive and negative ΔG values represent the ligand destabilization
and stabilization, respectively. We noted that only residues exhibiting ΔG value
Molecules 2022, 27, 8901 7 of 15

We also identified the number of contact atoms within the 5.0 Å from the ligand.
As shown in Figure 5, the native contacts of quinoxalinones were less than osimertinib.
Nevertheless, the non-native contacts demonstrated a similar range falling onto 50–80 atoms
(last 20 ns), indicating that the screened compounds could interact with the surrounding
amino acids when the dynamics was taken into account. In the case of osimertinib, due
to its structure being larger than quinoxalinones (1.5 to 2-fold larger molecular weight,
Table 1), it could plausibly expose to a larger number of proximate amino acids when
compared to quinoxalinones.

2.5. Hot-Spot Residues

To investigate the key binding residues within the focused site of EGFR tyrosine
kinase, the per-residue decomposition energy (∆Gbind
residue ) was elucidated via the MM/PBSA
bind
approach. The positive and negative ∆Gresidue values represent the ligand destabilization
and stabilization, respectively. We noted that only residues exhibiting ∆Gbind residue value
lower than −0.10 kcal/mol were elucidated and compared with other residues. As seen
in Figure 6, quinoxalinones showed somewhat a similar pattern of binding amino acids
compared to osimertinib. The common amino acids participating in the ligand binding
include L718, F723 (except CPD15), V726 (except CPD21), A743, M790, L792, G796, S797,
L844, and T854 (except CPD21), which agreed well with the previous report [36]. The
amino acids largely contributing to ligand binding consist of both G loop and hinge
region. The L718 and V726 (except CPD21) were found to noticeably influence to the
binding in almost all screened compounds and osimertinib, which agreed well with the
previous findings [36]. In addition, the M793 lining at the hinge region was detected in
all screened compounds’ binding, which has been highlighted as a satisfied condition of
successful EGFR TK inhibitors by controlling the entry of inhibitors to the ATP-binding
pocket [36,37]. Interestingly, all screened compounds could interact with two mutated
residues including M790 and S797, which are lining in the ligand binding site. It is worth
noting that CPD16 showed higher binding energy to M793 than osimertinib, revealing a
satisfied condition of being successful EGFR TK inhibitors. Altogether, this finding could
provide an explanation in terms of structural basis that support the great inhibitory activity
in vitro of all screened compounds.
In addition, the energy contributions to each amino acid were observed. All screened
compounds shared somewhat a similar pattern to osimertinib in which the most amino
nonpolar
acids were recognized through van der Waals interaction energies (∆EvdW + ∆Gsolv )
polar
while the electrostatic term (∆Eelec + ∆Gsolv ) was not preferably dominant toward the
binding (Figure 7). This finding agreed well with the previous study that suggested a
deep hydrophobic pocket within the ATP-binding site of EGFR TK that is controlled by
the residue at 790 position (known as a gatekeeper) [2]. However, CPD16 showed a
higher electrostatic contribution to M793 than other quinoxalinones and even osimertinib,
suggesting this compound could be well recognized during the binding event. Interestingly,
the residue T854 demonstrated a higher contribution of van der Waals interaction energies
to all screened compounds when compared to osimertinib, implying the main distinct
character of inhibitory actions of quinoxalinones.
Molecules 2022, 27, 8901
Molecules 2021, 26, x FOR PEER REVIEW 8 of 16
8 of 15

Figure
Figure 6. Bindingfree
6. Binding free energy
energy decomposed
decomposed to each
to each residue
residue (onlykey
(only some some key influential
influential amino acids
amino acids
were shown).
were shown). (A)(A) The
The level
levelof
of residue
residuecontribution
contributionwaswasshaded
shadedinin
a ared-blue-white
red-blue-white color scale,
color scale, which
which
was was ranging
ranging from higher
from higher to lower
to lower contribution,
contribution, respectively.
respectively. NoteNote
thatthat
thethe mutated
mutated aminoacids were
amino
acids
Molecules 2021, 26, xwere
FOR shown in red word. (B) A close-up view of key influential amino acids contributing to
PEER REVIEW 9 of 16
shown in red word. (B) A close-up view of key influential amino acids contributing to ligand binding
ligand binding in complex with screened compounds and osimertinib.
in complex with screened compounds and osimertinib.
In addition, the energy contributions to each amino acid were observed. All screened
compounds shared somewhat a similar pattern to osimertinib in which the most amino
acids were recognized through van der Waals interaction energies (∆EvdW ∆G )
while the electrostatic term (∆Eelec ∆G ) was not preferably dominant toward the
binding (Figure 7). This finding agreed well with the previous study that suggested a deep
hydrophobic pocket within the ATP-binding site of EGFR TK that is controlled by the
residue at 790 position (known as a gatekeeper) [2]. However, CPD16 showed a higher
electrostatic contribution to M793 than other quinoxalinones and even osimertinib, sug-
gesting this compound could be well recognized during the binding event. Interestingly,
the residue T854 demonstrated a higher contribution of van der Waals interaction energies
to all screened compounds when compared to osimertinib, implying the main distinct
character of inhibitory actions of quinoxalinones.

nonpolar polar
Figure
Figure 7. Calculated 7. Calculated
per-residue vdWper-residue
(∆EvdW +vdW∆Gsol ∆𝐺 electrostatic
(ΔEvdW )+ and ) and electrostatic ∆Gelec
(∆Eelec +(ΔE + ∆𝐺 ) de
sol )
composition energy to each amino acid largely contributed
decomposition energy to each amino acid largely contributed to ligand binding.to ligand binding.

2.6. Toxicity Prediction

Toxicity is one of the crucial factors that has been taken into consideration prior to
continuing the next phase of drug discovery campaigns. Herein, we employed the avail
able databases to predict the main crucial features necessary for inspecting compounds
toxicity. As listed in Table 2, all screened compounds showed no predicted mutagenicity
tumorigenicity, irritant, and reproductive effects (except CPD4; low level) as well as the
Molecules 2022, 27, 8901 9 of 15

2.6. Toxicity Prediction

Toxicity is one of the crucial factors that has been taken into consideration prior
to continuing the next phase of drug discovery campaigns. Herein, we employed the
available databases to predict the main crucial features necessary for inspecting compounds’
toxicity. As listed in Table 2, all screened compounds showed no predicted mutagenicity,
tumorigenicity, irritant, and reproductive effects (except CPD4; low level) as well as the
predicted LD50 values were in the range of slightly toxic (5000 mg/kg > LD50 > 500 mg/kg).
For osimertinib, it was predicted to be highly prone to mutagenicity and was moderately
toxic (500 mg/kg > LD50 > 50 mg/kg). Hence, all screened quinoxalinones were predicted
to be safe and might reduce the rate of failure caused by toxicity.

Table 2. Prediction of the main toxicity features of all screened quinoxalinones in comparison
to osimertinib.

Compounds
Features
Osimertinib CPD4 CPD15 CPD16 CPD21
LD50 (mg/kg) 100 1000 1000 800 1000
Toxicity class 3 4 4 4 4
Mutagenicity High None None None None
Tumorigenicity Low None None None None
Irritant Low None None None None
Reproductive effect Low Low None None None

2.7. Effect of CPD Derivatives on the Cell Viability

The cytotoxic effect of CPD derivatives on H1975 lung cancer cells expressing L858R/
T790M EGFR as well as Vero normal kidney cells was evaluated using MTT assay. Osimer-
tinib, a third-generation EGFR TKi used to treat adenocarcinoma patients carrying EGFR
mutation [22], was used as the positive control. As shown in Table 3, all CPD derivatives
as well as osimertinib inhibited the growth of H1975 cells in a concentration-dependent
manner. Of all CPD derivatives, CPD4 exhibited the most potent cytotoxicity toward H1975
cells (IC50 = 3.47 ± 2.20 µM), followed by osimertinib (IC50 = 18.33 ± 2.02 µM), CPD15 (IC50
= 31.25 ± 3.40 µM), CPD21 (IC50 = 44.67± 2.34 µM), and CPD16 (IC50 = 79.43± 4.35 µM).
Interestingly, all CPD derivatives, especially CPD15, CPD16, and CPD21 exhibited low
cytotoxic effects on Vero normal kidney cells (IC50 > 100 µM). These results suggested that
CPD derivatives are promising anticancer agents with high safety profiles. We noted that
the cell viability of H1975 and Vero was shown in Figure S5 and Figure S6, respectively.

Table 3. Cytotoxic effect of CPD derivatives and osimertinib on the H1975 and Vero cell viability
represented as IC50 values.

IC50 (µM)
Compounds
H1975 Vero
CPD4 3.47 ± 2.20 >20
CPD15 31.25 ± 3.40 >50
CPD16 79.43 ± 4.35 >100
CPD21 44.67 ± 2.34 >100
Osimertinib 18.33 ± 2.02 >100
Molecules 2022, 27, 8901 10 of 15

3. Materials and Methods

3.1. Docking-Based Virtual Screening
In this study, protein-ligand docking was performed using the crystal structure of
EGFR (L858R/T790M/C797S) in a complex with the known drug, osimertinib (PDB ID:
6LUD [38]). The missing residues (No. 748–751 and 991–999) were homologically built
using the SWISS-MODEL server [39]. The newly constructed protein structure was then
validated by plotting a Ramachandran diagram using PROCHECK [40]. In addition to the
protein structure, the quinoxalinones synthesized by Wimonsong et.al., [41] were manually
constructed into 3D structures by using the Gaussian 09 program [42].
For docking validation, the crystalized ligand was defined as a center in the binding
site, and redocked into the same binding site, which was in X, Y, and Z coordinates as
of −48.818, −2.761, and −18.500, respectively, while the box size was set to 40 Å for
every dimension. The crystallized and redocked poses were then aligned to observe the
verification of the docking protocol used as shown in Figure S3. Accordingly, the validated
docking protocol was employed for all ligands during virtual screening. The docking-based
screening was executed by the Autodock Vina XB software package (Sirimulla Researcg
Group@University of Texas at EI Paso, EI Paso, TX, USA).

3.2. EGFR Tyrosine Kinase Inhibition

The inhibitory activity of each candidate compound towards EGFR triple mutations
was performed using the luminescent ADP-Glo™ kinase assay (Promega Corporation,
Madison, WI, USA) as previously described [43,44]. According to the reaction recipe, 8 µL
of buffer (40 mM Tris-HCl pH 7.5, 20 mM MgCl2 and 0.1 mg mL−1 bovine serum albumin)
was added to a 384-well plate, followed by 5 µL of enzyme (1.25 ng µL−1 ). Then, 2 µL of
inhibitors and 10 µL of a mixture containing 25 µM ATP and 2.5 µM poly(glu-tyr) were
added and the reaction was incubated for 1 h at room temperature, followed by adding 5 µL
of the ADP-Glo reagent and the reaction was further incubated for 40 min. Subsequently,
10 µL of kinase detection reagent was added and incubated at room temperature for
30 min to convert ADP into ATP. The generated luminescence by ATP in a luciferase
reaction was then measured using a microplate reader (Infinite M200 microplate reader,
Tecan, Männedorf, Switzerland), which corresponds to kinase activity. All assays were
performed in triplicate and the obtained data were represented as the relative inhibition
(%) of inhibitors compared to the control with no inhibitor as shown in Equation (1) [45];

[(positive − negative) − (sample − negative)]

% relative inhibition = × 100 (1)
(positive − negative)

3.3. Molecular Dynamic (MD) Simulations

MD simulations were applied to gain insights into the key binding residues guiding
the mechanism of inhibitory action in the dynamical and physiological system. The initial
structure of each EGFR (L858R/T790M/C797S) TK and focused ligand was retrieved from
the best docking pose (lowest Ebinding ). Then, the constructed complex was subjected to run
under the periodic boundary condition with the isothermal-isobaric (NPT) scheme (310 K,
1 atm), according to the previous studies [35,36,46,47]. The AMBER ff14SB force field and
the generalized AMBER force field version 2 (GAFF2) were applied to treat bonded and
non-bonded interaction parameters of the simulated complex [48] whilst the water model
was used to solvate the system [49]. To neutralize the overall charge of the system, either
sodium or chloride ion was then randomly introduced according to the calculated numbers
of charges. Minimization of the hydrogen atoms and water molecules was carried out
using 500 steps of steepest descent (SD) followed by 1500 steps of conjugated gradient (CG)
methods while the rest of the molecules were fixed. Next, the minimization of the protein-
ligand complex (constrained solvents) and the whole complex system was minimized again
by using the same procedure, according to the previous publications [30,50,51].
Molecules 2022, 27, 8901 11 of 15

The particle mesh Ewald summation approach [52] was used to handle electrostatic
interactions, while all hydrogen atoms were constrained by the SHAKE algorithm [53].
The temperature was heated from 10 to 310 K by using a Langevin thermostat [54] with a
collision frequency of 2 ps−1 whilst the pressure was created by the Berendsen barostat [55].
The MD production lasted to 100 ns by the 2 fs increment of a time step. The MD outputs
were elucidated through the cpptraj module, while the per-residue decomposition energy
(∆Gresidue
binding ) was computed by MM/PBSA.py implemented in AMBER16.

3.4. Pharmacokinetic Properties Prediction

The focused pharmacokinetic properties including physicochemical characteristics
and drug-likeness were predicted by using the web-based program SwissADME (SIB
Swiss Institute of Bioinformatics, Écublens, Switzerland, www.swissadme.ch/, accessed on
30 October 2022) [56]. The screened compounds’ features were calculated compared to the
known drug, osimertinib.

3.5. Synthesis and Characterization of CPD21

Since the CPD21 has not been published yet, the synthesized procedure and character-
ization were described in this study. According to the general procedure [41], 1-Benzyl-3-
ethoxyquinoxalin-2(1H)-one (CPD21) was synthesized from benzylquinoxalin-2(1H)-one
(118.1 mg, 0.50 mmol) in ethanol (2 mL) as shown in Scheme 1. Column chromatography
(silica gel; EtOAc/Hexane, 1:1) gave the title compound 1-benzyl-3-ethoxyquinoxalin-
2(1H)-one as a white solid (251 mg, 90% yield). 1 H NMR (CDCl3 , 400 MHz): δ 7.61 (dd,
J = 7.4, 1.8 Hz, 1H), 7.31–7.17 (m, 8H), 5.51 (s, 2H), 4.57 (q, J = 7.1 Hz, 2H), 1.52 (t, J = 7.1 Hz,
3H) ppm, 13 C NMR (CDCl3 , 100 MHz): δ 154.1, 151.6, 135.3, 131.6, 130.9, 129.0, 127.8, 127.7,
127.1, 127.0, 124.1, 114.6, 63.7, 46.4, 14.3 ppm, and HRMS (ESI): m/z [M + H]+ calcd for
Molecules 2021, 26, x FOR PEER REVIEW 12 of 16
C17 H16 N2 O2 : 281.1285; found: 281.1283 (Figures S1 and S2). For other screened compounds,
their synthesis scheme was described in this paper [37] as listed in Table S1.

Scheme1.1.Synthesis
Scheme SynthesisofofCPD21.
CPD21.

3.6.
3.6.Toxicity
ToxicityPrediction
Prediction
InInsilico
silicoprofiling
profilingofoftoxicity
toxicityfeatured
featured(mutagenicity,
(mutagenicity,tumorigenicity,
tumorigenicity,irritant,
irritant,and
and
reproductive effect) was performed by using the DataWarrior software
reproductive effect) was performed by using the DataWarrior software package (versionpackage (version
5.5.0,
5.5.0,
ThomasThomas Sander,
Sander, https://openmolecules.org,
https://openmolecules.org, accessedaccessed on 30 October
on 30 October 2022,
2022, [57]). The[57]). The
prediction
prediction of lethal
of lethal dose dose(LD
at 50% at 50% (LD
50) and 50 )toxicity
the and theclass
toxicity
wereclass were performed
performed by a web-based
by a web-based Protox-II
Protox-II
server [58]server [58] (Available
(available online, https://tox-new.charite.de/protox_II/,
online, https://tox-new.charite.de/protox_II/, accessed on accessed on
30 October
302022).
October 2022).

3.7. Cell Lines, Culture, and Cytotoxicity Assay

3.7. Cell Lines, Culture, and Cytotoxicity Assay
Human NSCLC cell lines, H1975 (ATCC CRL-5908) and monkey’s normal kidney
Human NSCLC cell lines, H1975 (ATCC CRL-5908) and monkey’s normal kidney
epithelial cell line (Vero, ATCC CCL-81) were purchased from American Type Culture
epithelial cell line (Vero, ATCC CCL-81) were purchased from American Type Culture
Collection (ATCC, Manassas, VA, USA). H1975 cells were grown in RPMI-1640 medium
Collection (ATCC, Manassas, VA, USA). H1975 cells were grown in RPMI-1640 medium
supplemented with 10% FBS, 100 U/mL penicillin, and 100 µg/mL streptomycin. Vero
supplemented with 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin. Vero
cells were grown in Dulbecco’s modified Eagle’s minimal essential medium (DMEM;
cells were grown in Dulbecco’s modified Eagle’s minimal essential medium (DMEM;
Gibco, Grand Island, NY, USA) containing 10% fetal bovine serum (FBS; Gibco), 100 U/mL
penicillin, and 100 μg/mL streptomycin (Gibco). Both cells were maintained to grow at 37
°C in a humidified 5% CO2 atmosphere.
The studied cell lines were seeded into 96-well plates (cell density of 6000 cells/well)
Molecules 2022, 27, 8901 12 of 15

Gibco, Grand Island, NY, USA) containing 10% fetal bovine serum (FBS; Gibco), 100 U/mL
penicillin, and 100 µg/mL streptomycin (Gibco). Both cells were maintained to grow at
37 ◦ C in a humidified 5% CO2 atmosphere.
The studied cell lines were seeded into 96-well plates (cell density of 6000 cells/well)
and were incubated overnight. Then, cells were treated with screened compounds and
the known drug with various concentrations (125, 62.5, 31.25, 15.63, 7.81, 3.91, 1.95, and
0.98 µM) for 72 h. After incubation, the effect of compounds on cell viability was measured
using the MTT assay as follows: 10 µL of MTT solution (5 mg/mL) was added and then
incubated for 3 h. The medium was removed, followed by adding 50 µL of DMSO to
dissolve the crystal formazan product. The purple solution of crystal formazan was then
measured at 570 nm using a microplate spectrophotometer (Infinite M200 microplate reader,
Tecan, Männedorf, Switzerland). The complete medium with 1.0% DMSO (no inhibitor)
was used as a control in accordance with the reaction containing 1.0% DMSO of all wells.

3.8. Statistical Analysis

The data was represented as mean ± standard error of the mean (SEM). Differences
between groups were compared using one-way ANOVA (Tukey’s test) for multiple com-
parisons. The differences in means were determined at the confidence level p ≤ 0.05.

4. Conclusions
Finding pre-clinical drug candidates against acquired drug resistance of EGFR tyrosine
kinase is a great challenge and necessary. Herein, synthesized quinoxalinone-containing
compounds were explored through in silico and in vitro methods. From screening to
biological assay, CPD4, CPD15, CPD16, and CPD21 showed great inhibitory activity in
enzymatic assay with IC50 values down to a nanomolar scale as being a comparable level
when compared to the known drug, osimertinib. They could also inhibit the NSCLC cell
growth in a concentration-dependent manner with safety profiles. Based on computational
modeling, we proposed some common amino-acid residues which are responsible for the
binding of all screened compounds and the drug include L718 and V726 at the G loop
as well as M790, M793, and S797 at the hinge region. Moreover, all screened compounds
could be intermolecularly recognized by two previously mutated amino acids including
M790 and S797, which is the main reason to advocate the inhibitory capability against
EGFR (L858R/T790M/C797S) TK. In addition, all predicted physicochemical parameters
based on Lipinski’s rule of five had met the criteria of drug-likeness and were predicted to
show no highly toxic features. Altogether, we proposed that CPD4, CPD15, CPD16, and
CPD121 could become viable candidates for development as the fourth generation of EGFR
(L858R/T790M/C797S) TK. Animal testing is encouraged to investigate further.

Supplementary Materials: The following are available online at https://www.mdpi.com/article/

10.3390/molecules27248901/s1, Table S1: Chemical structures of all synthesized quinoxalinone-
containing compounds and their original code in the previously published paper, Figure S1: 1H
spectrum of CPD21 (CDCl3, 400 MHz), Figure S2: 13C NMR spectra of CPD21 (CDCl3, 100 MHz),
Figure S3: Alignment of the re-docked pose (green) and available crystallized osimertinib (grey)
showing the validated docking protocol used to perform docking-based virtual screening, Figure S4:
Calculated backbone root-mean-square displacement (RMSD) within 5 Å of all screened compounds
and osimertinib, Figure S5: H1975 cell viability against treated screened compounds and osimertinib
at different doses. Data were represented as mean ± SEM with three replicates, Figure S6: Vero
cell viability against treated screened compounds and osimertinib at different doses. Data were
represented as mean ± SEM with three replicates.
Author Contributions: U.S. carried out the preparation, data collection, virtual screening, molecular
dynamic simulations, enzyme- and cell-based assay, drug-likeness, toxicity prediction, and wrote
the initial version of the manuscript. W.W. synthesized quinoxalinones-containing compounds and
performed structural elucidation by NMR, S.Y. provided laboratory resources to perform organic
synthesis and NMR, K.C. provided laboratory resources to perform enzyme- and cell-based assay,
P.M., K.C., and T.R. conceived this study and are responsible for the overall design, interpretation,
Molecules 2022, 27, 8901 13 of 15

manuscript preparation, and communication. All authors have read and agreed to the published
version of the manuscript.
Funding: This research was supported by the Fundamental Fund of Khon Kaen University. K.C.
thanks the National Research Council of Thailand [grant NRCT5-RSA63002-07] and the Kasetsart Uni-
versity Research and Development Institute (KURDI), Bangkok, Thailand [grant KURDI(FF(KU) 4.65].
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Acknowledgments: U.S. would like to thank the Science Achievement Scholarship (SAST) of
Thailand for the Ph.D. scholarship and the 90th Anniversary of Chulalongkorn University Fund
(Ratchadaphiseksomphot Endowment Fund; GCUGR1125651029D).
Conflicts of Interest: The authors declare no conflict of interest.
Sample Availability: Samples of the compounds are available from the authors but only with the
permission of the Department of Chemistry, Center of Excellence for Innovation in Chemistry, Faculty
of Science, Mahidol University.

References
1. Sequist, L.V.; Lynch, T.J. EGFR Tyrosine Kinase Inhibitors in Lung Cancer: An Evolving Story. Annu. Rev. Med. 2008, 59, 429–442.
[CrossRef] [PubMed]
2. George Priya Doss, C.; Rajith, B.; Chakraborty, C.; NagaSundaram, N.; Ali, S.K.; Zhu, H. Structural signature of the G719S-T790M
double mutation in the EGFR kinase domain and its response to inhibitors. Sci. Rep. 2014, 4, 5868. [CrossRef]
3. Ciardiello, F.; Tortora, G. EGFR antagonists in cancer treatment. N. Engl. J. Med. 2008, 358, 1160–1174. [CrossRef] [PubMed]
4. Chong, C.R.; Jänne, P.A. The quest to overcome resistance to EGFR-targeted therapies in cancer. Nat. Med. 2013, 19, 1389–1400.
[CrossRef] [PubMed]
5. Sigismund, S.; Avanzato, D.; Lanzetti, L. Emerging functions of the EGFR in cancer. Mol. Oncol. 2018, 12, 3–20. [CrossRef]
[PubMed]
6. Bethune, G.; Bethune, D.; Ridgway, N.; Xu, Z. Epidermal growth factor receptor (EGFR) in lung cancer: An overview and update.
J. Thorac. Dis. 2010, 2, 48–51.
7. Salomon, D.S.; Brandt, R.; Ciardiello, F.; Normanno, N. Epidermal growth factor-related peptides and their receptors in human
malignancies. Crit. Rev. Oncol./Hematol. 1995, 19, 183–232. [CrossRef]
8. Gupta, R.; Dastane, A.M.; Forozan, F.; Riley-Portuguez, A.; Chung, F.; Lopategui, J.; Marchevsky, A.M. Evaluation of EGFR
abnormalities in patients with pulmonary adenocarcinoma: The need to test neoplasms with more than one method. Mod. Pathol.
2009, 22, 128–133. [CrossRef]
9. Huang, L.; Fu, L. Mechanisms of resistance to EGFR tyrosine kinase inhibitors. Acta Pharm. Sin. B 2015, 5, 390–401. [CrossRef]
10. Pérez-Soler, R.; Chachoua, A.; Hammond, L.A.; Rowinsky, E.K.; Huberman, M.; Karp, D.; Rigas, J.; Clark, G.M.; Santabárbara,
P.; Bonomi, P. Determinants of tumor response and survival with erlotinib in patients with non—Small-cell lung cancer. J. Clin.
Oncol. 2004, 22, 3238–3247. [CrossRef]
11. Mitsudomi, T.; Morita, S.; Yatabe, Y.; Negoro, S.; Okamoto, I.; Tsurutani, J.; Seto, T.; Satouchi, M.; Tada, H.; Hirashima, T. Gefitinib
versus cisplatin plus docetaxel in patients with non-small-cell lung cancer harbouring mutations of the epidermal growth factor
receptor (WJTOG3405): An open label, randomised phase 3 trial. Lancet Oncol. 2010, 11, 121–128. [CrossRef] [PubMed]
12. Mok, T.S.; Wu, Y.-L.; Thongprasert, S.; Yang, C.-H.; Chu, D.-T.; Saijo, N.; Sunpaweravong, P.; Han, B.; Margono, B.; Ichinose, Y.
Gefitinib or carboplatin–paclitaxel in pulmonary adenocarcinoma. N. Engl. J. Med. 2009, 361, 947–957. [CrossRef]
13. Maemondo, M.; Inoue, A.; Kobayashi, K.; Sugawara, S.; Oizumi, S.; Isobe, H.; Gemma, A.; Harada, M.; Yoshizawa, H.; Kinoshita,
I. Gefitinib or chemotherapy for non–small-cell lung cancer with mutated EGFR. N. Engl. J. Med. 2010, 362, 2380–2388. [CrossRef]
14. Kunimasa, K.; Sugimoto, N.; Tamiya, M.; Inoue, T.; Kawamura, T.; Kanzaki, R.; Okami, J.; Nishino, K. Dacomitinib overcomes
afatinib-refractory carcinomatous meningitis in a lung cancer patient harbouring EGFR Ex.19 deletion and G724S mutation; a
case report. Investig. New Drugs 2022, 40, 1137–1140. [CrossRef] [PubMed]
15. Wecker, H.; Waller, C.F. Afatinib. In Small Molecules in Oncology; Springer: Cham, Switzerland, 2018; pp. 199–215.
16. Dungo, R.T.; Keating, G.M. Afatinib: First global approval. Drugs 2013, 73, 1503–1515. [CrossRef] [PubMed]
17. Sakuma, Y.; Yamazaki, Y.; Nakamura, Y.; Yoshihara, M.; Matsukuma, S.; Nakayama, H.; Yokose, T.; Kameda, Y.; Koizume, S.;
Miyagi, Y. WZ4002, a third-generation EGFR inhibitor, can overcome anoikis resistance in EGFR-mutant lung adenocarcinomas
more efficiently than Src inhibitors. Lab. Investig. 2012, 92, 371–383. [CrossRef]
18. Walter, A.O.; Sjin, R.T.T.; Haringsma, H.J.; Ohashi, K.; Sun, J.; Lee, K.; Dubrovskiy, A.; Labenski, M.; Zhu, Z.; Wang, Z. Discovery of
a Mutant-Selective Covalent Inhibitor of EGFR that Overcomes T790M-Mediated Resistance in NSCLCDevelopment of Covalent
EGFRT790M Inhibitor in NSCLC. Cancer Discov. 2013, 3, 1404–1415. [CrossRef]
Molecules 2022, 27, 8901 14 of 15

19. Planchard, D.; Brown, K.H.; Kim, D.W.; Kim, S.W.; Ohe, Y.; Felip, E.; Leese, P.; Cantarini, M.; Vishwanathan, K.; Jänne, P.A.; et al.
Osimertinib Western and Asian clinical pharmacokinetics in patients and healthy volunteers: Implications for formulation, dose,
and dosing frequency in pivotal clinical studies. Cancer Chemother. Pharmacol. 2016, 77, 767–776. [CrossRef]
20. Eberlein, C.A.; Stetson, D.; Markovets, A.A.; Al-Kadhimi, K.J.; Lai, Z.; Fisher, P.R.; Meador, C.B.; Spitzler, P.; Ichihara, E.; Ross,
S.J. Acquired Resistance to the Mutant-Selective EGFR Inhibitor AZD9291 Is Associated with Increased Dependence on RAS
Signaling in Preclinical ModelsRAS Pathway Activation and Resistance to EGFR TKIs. Cancer Res. 2015, 75, 2489–2500. [CrossRef]
21. Ricciuti, B.; Baglivo, S.; Paglialunga, L.; De Giglio, A.; Bellezza, G.; Chiari, R.; Crinò, L.; Metro, G. Osimertinib in patients with
advanced epidermal growth factor receptor T790M mutation-positive non-small cell lung cancer: Rationale, evidence and place
in therapy. Ther. Adv. Med. Oncol. 2017, 9, 387–404. [CrossRef]
22. Zhao, J.; Zou, M.; Lv, J.; Han, Y.; Wang, G.; Wang, G. Effective treatment of pulmonary adenocarcinoma harboring triple EGFR
mutations of L858R, T790M, and cis-C797S by osimertinib, bevacizumab, and brigatinib combination therapy: A case report.
OncoTargets Ther. 2018, 11, 5545–5550. [CrossRef]
23. Günther, M.; Juchum, M.; Kelter, G.; Fiebig, H.; Laufer, S. Lung Cancer: EGFR Inhibitors with Low Nanomolar Activity against a
Therapy-Resistant L858R/T790M/C797S Mutant. Angew. Chem. Int. Ed. 2016, 55, 10890–10894. [CrossRef] [PubMed]
24. Günther, M.; Lategahn, J.; Juchum, M.; Döring, E.; Keul, M.; Engel, J.; Tumbrink, H.L.; Rauh, D.; Laufer, S. Trisubstituted
Pyridinylimidazoles as Potent Inhibitors of the Clinically Resistant L858R/T790M/C797S EGFR Mutant: Targeting of Both
Hydrophobic Regions and the Phosphate Binding Site. J. Med. Chem. 2017, 60, 5613–5637. [CrossRef]
25. Lei, H.; Fan, S.; Zhang, H.; Liu, Y.-J.; Hei, Y.-Y.; Zhang, J.-J.; Zheng, A.Q.; Xin, M.; Zhang, S.-Q. Discovery of novel 9-heterocyclyl
substituted 9H-purines as L858R/T790M/C797S mutant EGFR tyrosine kinase inhibitors. Eur. J. Med. Chem. 2020, 186, 111888.
[CrossRef] [PubMed]
26. Karnik, K.S.; Sarkate, A.P.; Tiwari, S.V.; Azad, R.; Burra, P.V.L.S.; Wakte, P.S. 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). Bioorganic Chem. 2021, 107, 104612. [CrossRef]
27. Shen, J.; Zhang, T.; Zhu, S.-J.; Sun, M.; Tong, L.; Lai, M.; Zhang, R.; Xu, W.; Wu, R.; Ding, J. Structure-based design of 5-
methylpyrimidopyridone derivatives as new wild-type sparing inhibitors of the epidermal growth factor receptor triple mutant
(EGFRL858R/T790M/C797S). J. Med. Chem. 2019, 62, 7302–7308. [CrossRef]
28. Zhang, H.; Wang, J.; Shen, Y.; Wang, H.-Y.; Duan, W.-M.; Zhao, H.-Y.; Hei, Y.-Y.; Xin, M.; Cao, Y.-X.; Zhang, S.-Q. Discovery of 2, 4,
6-trisubstitued pyrido [3, 4-d] pyrimidine derivatives as new EGFR-TKIs. Eur. J. Med. Chem. 2018, 148, 221–237.
29. Yang, J.; Wang, L.-J.; Liu, J.-J.; Zhong, L.; Zheng, R.-L.; Xu, Y.; Ji, P.; Zhang, C.-H.; Wang, W.-J.; Lin, X.-D. Structural optimization
and structure–activity relationships of N 2-(4-(4-Methylpiperazin-1-yl) phenyl)-N 8-phenyl-9 H-purine-2, 8-diamine derivatives,
a new class of reversible kinase inhibitors targeting both EGFR-activating and resistance mutations. J. Med. Chem. 2012, 55,
10685–10699. [CrossRef]
30. Ferlenghi, F.; Scalvini, L.; Vacondio, F.; Castelli, R.; Bozza, N.; Marseglia, G.; Rivara, S.; Lodola, A.; La Monica, S.; Minari, R. A
sulfonyl fluoride derivative inhibits EGFRL858R/T790M/C797S by covalent modification of the catalytic lysine. Eur. J. Med.
Chem. 2021, 225, 113786.
31. Mahalapbutr, P.; Leechaisit, R.; Thongnum, A.; Todsaporn, D.; Prachayasittikul, V.; Rungrotmongkol, T.; Prachayasittikul, S.;
Ruchirawat, S.; Prachayasittikul, V.; Pingaew, R. Discovery of Anilino-1,4-naphthoquinones as Potent EGFR Tyrosine Kinase
Inhibitors: Synthesis, Biological Evaluation, and Comprehensive Molecular Modeling. ACS Omega 2022, 7, 17881–17893.
[CrossRef]
32. Jiang, X.; Wu, K.; Bai, R.; Zhang, P.; Zhang, Y. Functionalized quinoxalinones as privileged structures with broad-ranging
pharmacological activities. Eur. J. Med. Chem. 2022, 229, 114085. [CrossRef] [PubMed]
33. An, Z.; Aksoy, O.; Zheng, T.; Fan, Q.W.; Weiss, W.A. Epidermal growth factor receptor and EGFRvIII in glioblastoma: Signaling
pathways and targeted therapies. Oncogene 2018, 37, 1561–1575. [CrossRef] [PubMed]
34. Koebel, M.R.; Schmadeke, G.; Posner, R.G.; Sirimulla, S. AutoDock VinaXB: Implementation of XBSF, new empirical halogen
bond scoring function, into AutoDock Vina. J. Cheminformatics 2016, 8, 27. [CrossRef] [PubMed]
35. Suriya, U.; Mahalapbutr, P.; Rungrotmongkol, T. Integration of In Silico Strategies for Drug Repositioning towards P38&alpha;
Mitogen-Activated Protein Kinase (MAPK) at the Allosteric Site. Pharmaceutics 2022, 14, 1461. [PubMed]
36. Todsaporn, D.; Mahalapbutr, P.; Poo-arporn, R.P.; Choowongkomon, K.; Rungrotmongkol, T. Structural dynamics and kinase
inhibitory activity of three generations of tyrosine kinase inhibitors against wild-type, L858R/T790M, and L858R/T790M/C797S
forms of EGFR. Comput. Biol. Med. 2022, 147, 105787. [CrossRef] [PubMed]
37. Fabbro, D.; Cowan-Jacob, S.W.; Moebitz, H. Ten things you should know about protein kinases: IUPHAR R eview 14. Br. J.
Pharmacol. 2015, 172, 2675–2700. [CrossRef]
38. Kashima, K.; Kawauchi, H.; Tanimura, H.; Tachibana, Y.; Chiba, T.; Torizawa, T.; Sakamoto, H. CH7233163 Overcomes Osimertinib-
Resistant EGFR-Del19/T790M/C797S Mutation. Mol. Cancer Ther. 2020, 19, 2288–2297. [CrossRef]
39. Waterhouse, A.; Bertoni, M.; Bienert, S.; Studer, G.; Tauriello, G.; Gumienny, R.; Heer, F.T.; de Beer, T.A.P.; Rempfer, C.; Bordoli,
L.; et al. SWISS-MODEL: Homology modelling of protein structures and complexes. Nucleic Acids Res. 2018, 46, W296–W303.
[CrossRef]
40. Laskowski, R.A.; MacArthur, M.W.; Moss, D.S.; Thornton, J.M. PROCHECK: A program to check the stereochemical quality of
protein structures. J. Appl. Crystallogr. 1993, 26, 283–291. [CrossRef]
Molecules 2022, 27, 8901 15 of 15

41. Wimonsong, W.; Yotphan, S. PIDA-induced oxidative C–N bond coupling of quinoxalinones and azoles. Tetrahedron 2021, 81,
131919. [CrossRef]
42. Frisch, M.J.W.C. Gaussian 09 Revision D. 01; Sian Inc.: Wallingford, CT, USA, 2009; Volume 112.
43. Sangpheak, K.; Tabtimmai, L.; Seetaha, S.; Rungnim, C.; Chavasiri, W.; Wolschann, P.; Choowongkomon, K.; Rungrotmongkol, T.
Biological Evaluation and Molecular Dynamics Simulation of Chalcone Derivatives as Epidermal Growth Factor-Tyrosine Kinase
Inhibitors. Molecules 2019, 24, 1092. [CrossRef] [PubMed]
44. Zegzouti, H.; Zdanovskaia, M.; Hsiao, K.; Goueli, S.A. ADP-Glo: A Bioluminescent and Homogeneous ADP Monitoring Assay
for Kinases. ASSAY Drug Dev. Technol. 2009, 7, 560–572. [CrossRef] [PubMed]
45. Sanachai, K.; Aiebchun, T.; Mahalapbutr, P.; Seetaha, S.; Tabtimmai, L.; Maitarad, P.; Xenikakis, I.; Geronikaki, A.;
Choowongkomon, K.; Rungrotmongkol, T. Discovery of novel JAK2 and EGFR inhibitors from a series of thiazole-based chalcone
derivatives. RSC Med. Chem. 2021, 12, 430–438. [CrossRef] [PubMed]
46. Thirunavukkarasu, M.K.; Suriya, U.; Rungrotmongkol, T.; Karuppasamy, R. In Silico Screening of Available Drugs Targeting
Non-Small Cell Lung Cancer Targets: A Drug Repurposing Approach. Pharmaceutics 2022, 14, 59. [CrossRef] [PubMed]
47. Verma, K.; Mahalapbutr, P.; Suriya, U.; Somboon, T.; Aiebchun, T.; Shi, L.; Maitarad, P.; Rungrotmongkol, T. In Silico Screening of
DNA Gyrase B Potent Flavonoids for the Treatment of Clostridium difficile Infection from PhytoHub Database. Braz. Arch. Biol.
Technol. 2021, 64, e21200402. [CrossRef]
48. Wang, J.; Wolf, R.M.; Caldwell, J.W.; Kollman, P.A.; Case, D.A. Development and testing of a general amber force field. J. Comput.
Chem. 2004, 25, 1157–1174. [CrossRef]
49. Jorgensen, W.L.; Chandrasekhar, J.; Madura, J.D.; Impey, R.W.; Klein, M.L. Comparison of simple potential functions for
simulating liquid water. J. Chem. Phys. 1983, 79, 926–935. [CrossRef]
50. Soe, H.M.H.; Chamni, S.; Mahalapbutr, P.; Kongtaworn, N.; Rungrotmongkol, T.; Jansook, P. The investigation of binary and
ternary sulfobutylether-β-cyclodextrin inclusion complexes with asiaticoside in solution and in solid state. Carbohydr. Res. 2020,
498, 108190. [CrossRef]
51. Klaewkla, M.; Charoenwongpaiboon, T.; Mahalapbutr, P. Molecular basis of the new COVID-19 target neuropilin-1 in complex
with SARS-CoV-2 S1 C-end rule peptide and small-molecule antagonists. J. Mol. Liq. 2021, 335, 116537. [CrossRef]
52. Darden, T.; York, D.; Pedersen, L. Particle mesh Ewald: An N· log (N) method for Ewald sums in large systems. J. Chem. Phys.
1993, 98, 10089–10092. [CrossRef]
53. Ryckaert, J.-P.; Ciccotti, G.; Berendsen, H.J.C. Numerical integration of the cartesian equations of motion of a system with
constraints: Molecular dynamics of n-alkanes. J. Comput. Phys. 1977, 23, 327–341. [CrossRef]
54. Uberuaga, B.P.; Anghel, M.; Voter, A.F. Synchronization of trajectories in canonical molecular-dynamics simulations: Observation,
explanation, and exploitation. J. Chem. Phys. 2004, 120, 6363–6374. [CrossRef] [PubMed]
55. Berendsen, H.J.C.; Postma, J.P.M.; Van Gunsteren, W.F.; DiNola, A.; Haak, J.R. Molecular dynamics with coupling to an external
bath. J. Chem. Phys. 1984, 81, 3684–3690. [CrossRef]
56. Daina, A.; Michielin, O.; Zoete, V. SwissADME: A free web tool to evaluate pharmacokinetics, drug-likeness and medicinal
chemistry friendliness of small molecules. Sci. Rep. 2017, 7, 42717. [CrossRef]
57. Sander, T.; Freyss, J.; von Korff, M.; Rufener, C. DataWarrior: An Open-Source Program For Chemistry Aware Data Visualization
And Analysis. J. Chem. Inf. Model. 2015, 55, 460–473. [CrossRef]
58. Banerjee, P.; Eckert, A.O.; Schrey, A.K.; Preissner, R. ProTox-II: A webserver for the prediction of toxicity of chemicals. Nucleic
Acids Res. 2018, 46, W257–W263. [CrossRef]