Content uploaded by Syeda Tasnim Quayum
Author content
All content in this area was uploaded by Syeda Tasnim Quayum on Nov 24, 2023
Content may be subject to copyright.
RESEARCH ARTICLE
Exploring the potential of fluoro-flavonoid derivatives
as anti-lung cancer agents: DFT, molecular docking,
and molecular dynamics techniques
Nusrat Jahan Ikbal Esha
1
| Syeda Tasnim Quayum
1
| Minhaz Zabin Saif
1
|
Mansour H. Almatarneh
2
| Shofiur Rahman
3
| Abdullah Alodhayb
3,4
|
Raymond A. Poirier
5
| Kabir M. Uddin
1
1
Department of Biochemistry and
Microbiology, North South University, Dhaka,
Bangladesh
2
Department of Chemistry, University of
Jordan, Amman, Jordan
3
College of Science, King Saud University,
Riyadh, Saudi Arabia
4
Department of Physics and Astronomy,
College of Science, King Saud University,
Riyadh, Saudi Arabia
5
Department of Chemistry, Memorial
University, St. John's, Newfoundland, Canada
Correspondence
Raymond A. Poirier, Department of Chemistry,
Memorial University, St. John's,
Newfoundland, Canada.
Email: rpoirier@mun.ca
Kabir M. Uddin, Department of Biochemistry
and Microbiology, North South University,
Dhaka, Bangladesh.
Email: kabirmuddin@gmail.com;mohammed.
uddin11@northsouth.edu
Shofiur Rahman, College of Science, King Saud
University, Riyadh 11451, Saudi Arabia.
Email: mrahman1@ksu.edu.sa
Abstract
The present investigation utilized in silico methodologies to explore the diverse phar-
macological activities, toxicity profiles, and chemical reactivity of a series of fluoro-
flavonoid compounds (1–14), which are secondary metabolites of plants known for
their broad range of biological effects. A comprehensive strategy is utilized, incorpo-
rating methods such as prediction of activity spectra for substances (PASS) predic-
tion, absorption, distribution, metabolism, excretion, and toxicity (ADMET)
assessments, and density functional theory (B3LYP) calculations using three basis
sets: 6-31G(d,p), 6-311G(d,p), and 6-311++G(d,p). Furthermore, the study employed
molecular docking technique to identify target proteins, including HER2 (7JXH),
EGFR (4UV7), FPPS (1YQ7), HPGDS (1V40), DCK (1P60), and KEAP1 on Nrf2 (1X2J),
for the investigated compounds, with cianidanol and genistein serving as
reference drugs for the docking process. The investigated fluoro-flavonoid com-
pounds exhibited significantly greater binding affinities (ranging from 8.3 to
10.6 kcal mol
1
) toward HER2, HPGDS, and KEAP1 compared to the reference
drugs, cianidanol and genistein, which displayed binding affinities ranging from 8.4
to 9.4 kcal mol
1
. Furthermore, molecular dynamics simulations were conducted to
assess the stability of the protein-ligand interaction, using the root-mean-square
deviation (RMSD), root-mean-square fluctuation (RMSF), Radius of gyration
(Rg) parameters and principle component analysis (PCA). Among the tested fluoro-
flavonoid analogs, analog 11 showed a RMSD value of .15 nm with the HER2 protein
target, indicating a stable interaction. Based on in silico results, it appears that the
fluoro-flavonoid compound 11 has the potential to serve as a targeted anti-lung can-
cer drug. However, additional in vivo and in vitro studies are necessary to confirm
this hypothesis.
KEYWORDS
ADMET, DFT, fluoro-flavonoid, MD simulation, molecular docking, PASS
Received: 31 May 2023 Revised: 16 October 2023 Accepted: 19 October 2023
DOI: 10.1002/qua.27274
Int J Quantum Chem. 2023;e27274. http://q-chem.org © 2023 Wiley Periodicals LLC. 1of23
https://doi.org/10.1002/qua.27274
1|INTRODUCTION
Cancer, known as the uncontrolled growth of abnormal cells, poses a significant challenge in the field of healthcare and is currently the second
leading cause of death worldwide [1, 2]. Abnormal signaling within cells leads to the development and progression of cancer, causing detrimental
effects on the body [3]. Lung cancer may arise as a result of the lung's aberrant epithelial cell proliferation. Genetic mutations or exposures to haz-
ardous substances like tobacco smoke, air pollution, asbestos, or radon are common causes of this aberrant development [4]. These mutations
and exposure may eventually interfere with the normal control of cell growth and division, causing unchecked cell proliferation and the develop-
ment of lung tumor [5]. Assessments on new chemical entities found that more than 70% of anti-cancer drugs are from natural products or syn-
thesized compounds structurally related to natural products [6, 7]. Recent studies suggest that consuming different fruits and vegetables
increases the ability to fight against cancer and reduces the risk of cancer by at least 20% [8, 9]. In preclinical and clinical research, a number of
phytoconstituents have shown promise as potential lung cancer treatment agents. For instance, sulforaphane from cruciferous vegetables, resver-
atrol from grapes and berries, quercetin from fruits and vegetables, and epigallocatechin-3-gallate (EGCG) from green tea [10]. Furthermore, in
the prevention and treatment of different types of cancer, naturally occurring compounds have been found to be more beneficial than synthetic
compounds due to lower toxicity, greater accessibility to the target site, and less expensive. Consumption of flavonoids is associated with reduced
risk of some chronic diseases, including cancers, cardiovascular disease, and inflammatory disease [9, 11–13].
Flavonoids are secondary plant metabolites characterized by two or more aromatic rings. They are found in fruits, vegetables, grains, bark,
roots, stems, flowers, tea, beer, cocoa, and wine. Over 4000 different flavonoids have been identified [12], and depending on the structural differ-
ences, flavonoids are divided into eight other groups: flavonols, flavanones, flavones, isoflavones, catechins, anthocyanidins, dihydro flavonols,
and chalcones [12, 14–16]. The carbon atoms in flavonoid molecules are assembled in three aromatic rings, commonly denoted as A, B and C,
where B and C is connected by a single bond, thus forming a diphenyl-propane structure with the central unit being a benzo-pyrone (chromone)
and fluorine substituted flavonoid derivatives as shown in Figures 1and 2. Multiple hydroxyl groups, sugar, oxygen, or methyl groups are attached
to this core structure [12, 14–17].
Flavonoids exert innumerable beneficial effects on human health and are considered a molecular template for designing novel therapeutic
agents for many diseases, including lung cancer [14]. Research has shown that flavonoids have abundant biological effects such as antic-
arcinogenic [18], antiviral [19], anti-inflammatories [20], immune stimulation [18], antiallergics [18], and can reduce the risk of cardiovascular dis-
ease [21, 22]. Flavonoids, one of the active components of Chinese herbal medicine, which is used in treating asthma, have a potential that
cannot be ignored in developing new drugs for treating asthma [22]. With the continuous study of the relationship between “flavonoids”and
“tumors,”it has been found that flavonoids play a better role in treating colon, liver, and breast cancers and other associated conditions [11,
23, 24]. Many of these favorable health effects, including anticancer activity, arise from the antioxidant properties of these polyphenolic com-
pounds [25, 26]. For example, flavonoids behave as antioxidants primarily because of their polyphenolic nature. It is reported that the greater the
number of hydroxyl groups present in the flavonoid structure, the greater the antioxidant capacity of the flavonoid [26]. These results indicate
that flavonoids may become sought-after in the future for the development of new drugs to treat cancer patients [25, 27].
FIGURE 1 Structures of the flavones and flavonols derivatives (1–7).
2of23 ESHA ET AL.
The mechanisms of action mediating the beneficial effects of flavonoids on health have been studied in preclinical models. Based on these
activities, some flavonoids have been used in clinics to treat different diseases. For example, some clinically available flavonoids include
ipriflavone, diosmin, flavonihippopha, efloxate, and silibinin [27]. One medication, cianidanol, has completed a phase 2 clinical study for a
FIGURE 2 Structures of the fluoro-flavonoid derivatives (8–14) and reference drugs.
ESHA ET AL.3of23
malignant lung tumor [28]. It exhibited a suppressive impact on the growth of A549 cancer cells. The expression of p21, cyclin E1, AKT, and
P-AKT proteins appeared to be upregulated, while p21 was downregulated, and these changes all seemed to affect the proliferation of cancer
cells [29]. Another drug called genistein has completed clinical trials for several illnesses, including adenocarcinoma of the prostate, breast cancer,
and bladder cancer [30]. It is an isoflavonoid derived from soy products and functions as an antineoplastic and anticancer drug by inhibiting the
action of topoisomerase-II (DNA topoisomerases, type II) and protein tyrosine kinase. It has been demonstrated through experimentation to cause
G2 phase arrest in human and murine cell lines [31, 32].
The molecular structures of flavone and 2-(4-fluorophenyl)-3-hydroxy-4H-chrome-4-one have been identified using surface-enhanced
Raman spectroscopy and crystallography, respectively [33, 34]. Two sub-units make up the asymmetric unit of flavone. In these sub-units, the
pyrone ring forms a dihedral angle of 1.0 (1)with the 2-phenyl substituent, while in the other, the pyrone ring forms a dihedral angle of 9.8
(1)with the 2-phenyl substituent [33, 35]. Furthermore, 2-(2-fluorophenyl)-3-hydroxy-4H-chrome-4-one compound has the fluorine-
substituted benzene rings attached to the 4H-chrome skeleton at the C8 and C23 sites in this asymmetric unit. For the flavonol structure,
the hydroxyl groups are attached to the 4H-chromenon skeletons at the C9 and C24 positions and tilted away from the ring system by
24.5 (1)[34].
Compared to the many experimental studies, only a limited number of in silico studies have been conducted on flavonoid and its derivatives
[35–39]. Dolatkhan et al. [36] studied a series of 4H-chromone-1,2,3,4-tetrahydro pyrimidine-5-carboxylates products, and the in silico results
indicated that these have no toxic effects, are potential antileukemic agents. Recently, Rathod et al. [37] used in silico docking and molecular
dynamics to explore flavonoids for possible anti-cancer activities against cyclin-dependent kinase 8 (CDK8). Their study shows flavonoid com-
plexes possess many dynamic properties, such as high stability, significant binding energy, and fewer conformation fluctuations [38]. Furthermore,
the pharmacokinetics and drug-likeness studies and DFT descriptor values indicated a promising result of the selected natural flavonoids and
significant interactions in the binding pocket of the target protein, and those results can pave the way for drug discovery [38]. Lu et al. [39] con-
ducted a study on 1,3,5-triazine-2,4-diamine combined with 1H-indole-2,3-dione (isatin), (2E)-1,3-diphenylprop-2-en-1-one (chalcone), and 10H-
acridine-9-one (acridone) to determine their molecular and electronic characteristics. The study also evaluated the energy gap, kinetic stability,
and electrophilicity index of delocalized sites of 1,3,5-triazine-2,4-diamine, and investigated the relationship between these characteristics and
their potential biological activity. According to computational analysis, flavonoid compounds exhibit a high affinity for respiratory diseases, such
as SARS-CoV-2, and have the potential to serve as inhibitors. The analysis also suggests that flavonoid compounds have the lowest binding
energy, indicating a strong likelihood of effective binding with the target protein [39]. Moreover, drug-likeness and ADMET studies revealed that
flavonoids are safe and non-toxic [40].
No in silico studies have been performed to calculate the physicochemical, pharmacokinetics, and biological properties of novel fluoro-
flavonoid analogues (1–14), including a different position of the hydroxyl groups. The main objective of this work is to investigate the phar-
macodynamics, toxicity profiles, and biological activities of fluorine-substituted flavonoid analogues (1–14). In addition, density functional the-
ory (DFT) was employed to analyze the thermodynamic stability and molecular orbital properties including HOMO (highest occupied
molecular orbital) LUMO (lowest unoccupied molecular orbital) gap, hardness, softness, chemical potential, electrophilicity index, and dipole
moment of all these derivatives were analyzed to understand their electronic structure and reactivity. We have also performed molecular
docking analysis to calculate the binding affinities against six targeted proteins using cianidanol and genistein, known as catechin and
40,5,7-trihydroxyisoflavone, respectively, as reference drugs, along with the prediction of activity spectra for substances (PASS). Moreover,
MD simulation analyses were performed using the active site of the interaction between the protein and compounds to evaluate the stability
of the protein-ligand complex, thereby constituting an interesting novel class of targeted anti-lung cancer drugs, which may play an essential
role in the biological action.
2|COMPUTATIONAL STUDIES
2.1 |Computational analysis
All calculations were performed with Gaussian 16, Revision C.01 [41]. The geometries were fully optimized at the B3LYP level of theory using the
6-31G(d,p), 6-311G(d,p), and 6–311++G(d,p) basis sets. Frequencies were obtained for all optimized structures to ensure the absence of imagi-
nary frequencies. To identify the most reliable methods for calculating structural data, theoretical values were compared with experimental data
where possible. From our previous work [42], it was found that the activation energies and the thermodynamic properties calculated using
G3MP2, G3MP2B3, G4MP2, G3B3, and CBS-QB3 agree within 10 kJ mol
1
with each other, and within 18 kJ mol
1
with the results of
B3LYP/6-31G(d,p). Accordingly, B3LYP can provide acceptable results and save computation time. The Frontier Molecular Orbital (FMO) in terms
of the energy distribution from HOMO to LUMO and molecular electrostatic potential (MEP) map was calculated using the GaussView 6 software
program [43]. The HOMO, LUMO, and MEP maps are shown in Tables S1 to S16 and Figures S1 to S30. Unless otherwise stated, all values in the
text are for B3LYP/6-31G(d,p).
4of23 ESHA ET AL.
For each compound we explored (1–14) several chemical descriptors, such as Energy gap (E
Gap
), Ionization potential (IP), Electron affinity
(EA), Chemical potential (μ), Electronegativity (χ), Hardness (η), Softness (σ) and Electrophilicity (ω)[
44–46] were calculated using the following
equations:
EGap eVðÞ¼ELUMO EHOMO
ðÞ;IP eVðÞ¼EHOMO;EA eVðÞ¼ELUMO;
μeVðÞ¼IPþEAðÞ=2;χ¼μ;η¼IP EAðÞ=2;σ¼1=η;ω¼μ2=2η:
where μand ηwere used to derive the electrophilicity indexes. Furthermore, to get potential energies and atomic charges, the single point energy
and natural bond orbital (NBO) analysis were performed [47].
2.2 |Evaluation of physicochemical and pharmacokinetic properties
To assess the effectiveness of the fluoro-flavonoids and to determine their physiochemical properties compiled from the PubChem database
(https://pubchem.ncbi.nlm.nih.gov/search/search.cgi), the swissADME server tool (www.swissadme.ch) was used [48]. The server uses robust
predictive results to find physicochemical properties, medicinal chemistry, and drug-likeness [48]. Moreover, the AdmetSAR tool (http://lmmd.
ecust.edu.cn/admetsar2/) and ADMET predictor software version 9.5 were used to determine and analyze the ADMET properties. The resulting
values were determined by using canonical simplified molecular input line entry system (SMILES) of each compound. Several toxicity values, CYP
inhibitors, and hERG pIC50 scores were found. AdmetSAR gives both unregulated and downregulated values for each ADMET property. Addition-
ally, Molinspiration [49] online open-access server (https://www.molinspiration.com/cgi-bin/properties) was used to evaluate the relationship
between the physicochemical properties and molecular activity of the compounds. Drug-likeness of the tested compounds was investigated as
G-protein coupled receptor (GPCR) ligands, ion channel modulators (ICM), kinase inhibitors (KI), nuclear receptor ligands (NRL), protease inhibitors
(PI), and enzyme inhibitors (EI). At first, all the structures (1–14) were drawn in ChemBioDraw Ultra 14.0 to collect MDL Molfile format and were
converted to canonical SMILES.
2.3 |Pharmacological activities
The PASS tool can determine the toxicity, mechanism, receptor regulation, mutagenicity, carcinogenicity, teratogenicity, and many other pharma-
cological effects of any compound [50]. PASS online server tool (http://way2drug.com/passonline) was used to determine the probable
pharmacological activities. This server tool provides access to biochemical information regarding compounds that the United States and Russian
Federation have approved for medicinal use [50].
2.4 |Molecular docking
2.4.1 | Preparation of ligands
The three-dimensional (3D) structure data files (SDF) of the flavonoids and the reference drugs (cianidanol and genistein) were retrieved in SDF
format from the PubChem database. All the flavonoids were drawn using GaussView 6 software, and all the structures were optimized at
B3LYP/6-31G(d,p) using the Gaussian 16 package. The structures of all the ligands selected for docking were subjected to energy minimization
(EM), and then the freely available program OpenBabel plugin of PyRx 0.8 software (available at https://pyrx.sourceforge.io/) was used to convert
to the PDBQT format [51].
2.4.2 | Preparation of target protein
Molecular docking software searches for a suitable protein that can provide a potential binding site to the ligands. The required crystal structures
of proteins for performing molecular docking were retrieved from the RSCB Protein Data Bank tool (https://www.rcsb.org/)[52]. RCBS PDB is
the most extensive archive of more than 505 000 000 atomic coordinate and experimental data files [53]. In total, six different proteins such as
Human Epidermal Growth Factor Receptor 2, HER2 (PDB ID:7JXH); Epidermal Growth Factor Receptor, EGFR (PDB ID:4UV7); Farnesyl Diphos-
phate Synthase, FPPS (PDB ID: 1YQ7); Hematopoietic Prostaglandin D Synthase, HPGDS (PDB ID:1V40); Human Deoxycytidine Kinase, DCK
ESHA ET AL.5of23
(PDB ID:1P60), and Kelch-like ECH-associated protein 1, KEAP1 (PDB ID:1X2J) on Nuclear factor erythroid 2-related factor 2 (Nrf2) were
included in this work. These proteins are viable targets for this study because they participate in pathways that regulate cell growth, survival,
metabolism, inflammation, and oxidative stress [54–58]. Chimera version 1.16 [59] was used for the preparation of the protein structure.
Further preparation of each protein included removing water molecules and ligands, and energy minimization of the macromolecule were car-
ried out for the molecular docking and MD simulation. The protonation state was histidine; the standard residue was kept at default AMBER
ff14SB for all the ligands and proteins. Gasteiger was chosen as the charge calculation method for the other residue of each protein. The energy-
minimized macromolecule was subsequently prepared for docking using Pyrx 0.8 software [52], which employed the UFF force field and a conju-
gate gradient optimization algorithm spanning 200 steps while maintaining an optimization energy threshold of >.1 kcal mol
1
.
2.4.3 | Protein-ligand docking
The protein-ligand docking was done using the AutoDock Vina [60] software. The target protein was used for docking studies with the
ligands. The ligands were docked by selecting a grid box that covers the complete protein at the center. The default exhaustiveness value of
AutoDock Vina, which is set to 8 was utilized. The box parameters were as follows: Center X: 65.45, Y: 5.65, and Z: 76.29, with box dimen-
sions (in Angstroms) of X: 53.09, Y: 59.84, and Z: 60.69. The stability of both proteins and ligands was maintained throughout the docking
experiment. Additionally, the protein-ligand complexes were validated by re-docking. Furthermore, UCSF Chimera version 1.16 [59] was used
to identify the amino acids that interacted with the ligands. Pymol version 2.5 [61] was used to generate three-dimensional structures for
molecular docking images. BIOVIA Discovery Studio [62] visualizer was used to visualize the binding modes of the protein-ligand interaction,
identify the 2D protein interactions between the ligand and protein, and determine the hydrogen density around the residues interacting with
the protein.
2.5 |Molecular dynamics simulations
Molecular dynamics (MD) simulations were carried out with the GROMACS version 2021.6 package [63] using the AMBER99SB force field [64],
which provides a detailed description of the interactions between atoms. MD simulation is a powerful computational technique compared to a
less computationally intensive docking method. It offers high precision and can provide insights into the behavior of complex systems, which may
not be accessible through experimental methods [65]. In this work, we utilized MD simulations to investigate the docked complexes of PDB:
7JXH and compound 11. Furthermore, an additional set of three MD simulations (for compounds 9and 12, as well as the reference drug genistein
was conducted to validate the earlier findings. The topology parameters for the proteins in the system were generated using the Galaxy European
Server [66], a widely-used platform for molecular modeling and simulation. The system was solvated with SPC water molecules in a triclinic box,
and sodium and chloride ions were added to achieve standard salt concentrations and neutralize the system [67].
To ensure system stability, we conducted an equilibration process, employing a position-restrained dynamics simulation (NVT) at 300 K, last-
ing for 3000 ps, and implemented using the leapfrog algorithm [68, 69]. After this equilibration phase, the entire system underwent a production
run for an additional 3000 ps, with constant temperature and pressure conditions. Finally, the use of MD simulations in the current study allowed
us to analyze the behavior of the docked protein-ligand complexes. This knowledge can help us better understand the physical principles behind
biological macromolecule functional structure. To visualize and analyze MD trajectories that were generated, VMD [70], PyMol [61], and
GROMACS programs were utilized. All the MD simulations were performed for 20 ns at 300 K using GROMACS statistical analysis for the root-
mean-square-deviation (RMSD), radius-of-gyration (Rg), and root-mean-square fluctuation (RMSF) using GROMACS utility gmx rmsd, gmx gyrate,
and gmx rmsf, respectively. The “gmx hbond”utility in GROMACS was used for the analysis of hydrogen bonds, and the “gmx energy”tool was
used to determine the temperature and potential energy. Furthermore, to evaluate the stability of protein complexes, the utilization of principal
component analysis has been included and is found to be highly advantageous [71–73]. The principle component analysis (PCA) of the obtained
MD trajectories was carried out using the Bio3D package via the Galaxy European server [74–76].
3|RESULTS AND DISCUSSION
3.1 |Structural parameters
The flavone (1) and 40-fluoroflavonol (7) structures have been identified using surface-enhanced Raman spectroscopy and x-ray crystallography,
respectively [33, 34]. Structural parameters, including bond lengths and angles, for compounds 1and 7were calculated at the B3LYP level of the-
ory using the (6-31G(d,p), 6-311G(d,p), and 6–311++G(d,p)) basis sets. These parameters were then compared with the experimental data to
6of23 ESHA ET AL.
evaluate the accuracy of the computational method [33, 34]. The structures of compounds 1and 7are shown in Figure 3(see Figures S1 to S2
and Tables S1 to S14), and the bond lengths and angle values of the selected methods are given in Table 1.
In the gas phase, both compounds 1and 7exhibited mean deviations (MDs) for bond lengths ranging from .083 to .102 Å and .080 to .097 Å,
respectively, across various basis sets at the B3LYP level. All basis sets at the B3LYP level of theory exhibited comparable performance in deter-
mining the bond lengths and angles of both compounds. Moreover, the optimized molecular structures showed differences of up to .047 Å for
bond lengths and up to .5for bond angles, for different basis sets. The inclusion of diffuse functions in the basis set has a minimal effect on the
geometries of both compounds 1and 7, as evidenced by the small mean deviations (MDs) obtained at the B3LYP/6–311++G(d,p) level, as shown
in Table 1. B3LYP/6-31G(d,p) provided good agreement with experimental data for the bond lengths and angles of compounds 1and 7,as
shown in Table 1[33, 34]. Based on these findings, the 6-31G(d,p) basis set is an appropriate choice for accurately predicting the molecular geom-
etry of the studied fluoro-flavonoid derivatives.
3.2 |Frontier molecular orbital analysis
The FMO analysis of the fluoro-flavonoid derivatives (1–14) provides valuable information regarding their reactivity and stability. The HOMO-
LUMO gap (Egap) is crucial in determining their chemical reactivity, hardness, softness, chemical potential, and electrophilic index [77]. A large
FIGURE 3 Optimized structures of flavone (1) and 40-fluoroflavonol (7).
TABLE 1 Selected bond distances (Å) and angles (deg) for flavone (1) and 2‐(2‐fluorophenyl)‐3‐hydroxy‐4H‐chromen‐4‐one (7).
a,b
Flavone (1) 4′‐fluoroflavonol (7)
Bond type
B3LYP/6‐31G
(d,p)
B3LYP/6‐311G
(d,p)
B3LYP/6‐311+
+G(d) Exptl
c
B3LYP/6‐31G
(d,p)
B3LYP/
6‐311G(d,p)
B3LYP/
6‐311++G(d,p) Exptl
d
Bond distances (Å)
C
2
O
1
1.364 (.003) 1.362 (.005) 1.361 (.006) 1.367 1.354 (.013) 1.371 (.004) 1.370 (.003) 1.367
C
2
C
1
′1.475 (.000) 1.475 (.000) 1.475 (.000) 1.475 1.474 (.001) 1.474 (.001) 1.475 (.002) 1.473
C
2
C
3
1.358 (.004) 1.355 (.001) 1.355 (.001) 1.354 1.356 (.004) 1.362 (.010) 1.362 (.010) 1.352
C
3
C
4
1.457 (.001) 1.456 (.002) 1.455 (.003) 1.458 1.481 (.037) 1.473 (.029) 1.473 (.029) 1.444
C
1
′C
2
′1.404 (.000) 1.402 (.002) 1.402 (.002) 1.404 1.401 (.032) 1.404 (.035) 1.404 (.035) 1.369
C
2
′C
3
′1.403 (.003) 1.402 (.002) 1.402 (.002) 1.400 1.392 (.047) 1.386 (.041) 1.386 (.041) 1.345
C
4
′C
5
′1.395 (.001) 1.393 (.001) 1.393 (.001) 1.394 1.389 (.044) 1.386 (.041) 1.386 (.041) 1.345
C
4
′H/F 1.084 (.002) 1.083 (.001) 1.084 (.002) 1.082 1.350 (.006) 1.347 (.009) 1.347 (.009) 1.356
Bond angles (deg)
O
1
C
2
C
1
′112.1 (.2) 112.1 (.2) 112.2 (.3) 111.9 111.9 (.4) 111.9 (.4) 112.0 (.3) 112.3
O
1
C
2
C
3
121.9 (.3) 121.8 (.4) 121.9 (.3) 122.2 121.9 (.3) 122.0 (.4) 122.1 (.5) 121.6
C
1
′C2′C3′120.4 (.4) 120.5 (.5) 120.5 (.5) 120.0 120.9 (.0) 121.0 (.1) 120.9 (.0) 120.9
MD .083 .101 .102 .080 .097 .088
a
The values in parenthesis represent the difference between the experimental and calculated values.
b
MD is the mean deviation for the parenthesis value.
c
Reference 33.
d
Reference 34.
ESHA ET AL.7of23
Egap corresponds to high stability and low chemical reactivity, while a narrow energy gap indicates softness, corresponding to high chemical reac-
tivity and low stability. The HOMO and LUMO energies are also crucial in understanding the compound's electron donor and acceptor properties,
respectively. On the other hand, when promoting an electron from the HOMO to the LUMO, compounds with a narrow energy gap are denoted
as soft (σ), indicating high chemical reactivity, and low stability, while those with a large energy gap are referred to as hard (η). Understanding the
electronic properties of drugs is highly valuable for studying their pharmacokinetics [78–80]. All compounds (1–14) had their ionization potential
(IP), electron affinity (EA), chemical potential (μ), electronegativity (χ), global hardness (η), softness (σ), electrophilicity (ω), and dipole moment cal-
culated using the B3LYP method with the 6-31G(d,p) basis set, as presented in Table 2and Figure 4(see Figures S3 S7).
As shown in Table 2, among the compounds 1to 7, compound 7exhibits the largest HOMO-LUMO gap (4.61 eV), likely due to the influence
of the fluorine atom at the 40-position in this series. Similarly, in Table 2, it can be observed that compound 7exhibits the largest hardness
(2.30 eV) and the lowest softness (.43 eV) compared with compounds 1through 7. Additionally, the lower electrophilicity index of compound
7(2.14 eV) provides a foundation for further analysis of its potential biological activity through molecular docking to a suitable protein. For com-
pounds 8to 14, compound 11 has the largest HOMO-LUMO gap (4.51 eV) among the compounds, and like compound 7, it has a fluorine in the
same position. The reference drug for D2, genistein, exhibits similar characteristics to compound 11 when considering their HOMO-LUMO gaps.
However, the gap for reference compound D2 is lower than that of compound 11, indicating that D2 may be less stable and more reactive. The
HOMO- LUMO gaps in this series (8to 14) decrease as follows: 11 (4.54 eV) > D2 (4.23 eV) > 14 (4.21 eV) > 13 (4.20 eV) > 10 (4.15 eV) > 12
(4.11 eV) > 9(4.10 eV) > 8(4.03 eV), except for D1 (cianidanol), which serves as a different structural reference drug (see Figure 1).
The results suggest that compound 11 is more reactive than the reference drug D2, with a higher chemical potential value of 2.04 eV and a
lower electrophilicity index of 2.17 eV compared with the other compounds in the series (8to 14). The electrophilicity index is often used to pre-
dict biological activity and identify reactive sites by measuring the energy lowering resulting from electron transfer between the donor (HOMO)
and acceptor (LUMO). The dipole moment of compound 11 was found to be 4.03 Debye, slightly higher than that of the other compounds (8to
14), which suggests that it may have better binding affinity and interactions with the fluoro group of the receptor protein. Overall, the lower elec-
trophilicity index and higher dipole moment of compound 11 make it a promising candidate for further analysis of its biological activity through
molecular docking studies.
NBO analysis [47] can give insights into the bonding and stability of molecules by examining the interactions between atoms and the distribu-
tion of atomic charges. Figures S8 S9 show how intramolecular charge transfer interactions contribute to stabilization in the studied compounds.
Meanwhile, the MEP maps in Figure 5and Figures S10 S12 provide a visual representation of the electrostatic potential of the molecules and
can help identify regions of electrophilicity and nucleophilicity. The different hues in the MEP maps indicate the strength of the electrical poten-
tial, with blue representing electrophilic regions and red representing nucleophilic regions. Overall, these analyses provide valuable information on
the chemical properties and reactivity of the studied compounds.
TABLE 2 The HOMO and LUMO energies, IP, EA, chemical potential (μ), electronegativity (χ), global hardness (η), softness (σ), electrophilicity
(ω), and dipole moment (Debye) of all compounds (1–14) at 298.15 K calculated using B3LYP/6-31G(d,p).
Ligand E
LUMO
(eV) E
HOMO
(eV) E
gap
(eV) IP (eV) EA (eV) μ(eV) χ(eV) η(eV) σ(eV) ω(eV) Dipole (D)
11.801 6.355 4.554 6.355 1.801 4.078 4.078 2.277 .439 3.651 4.183
21.839 6.384 4.545 6.384 1.839 4.111 4.111 2.272 .440 3.719 5.249
31.900 6.490 4.590 6.490 1.900 4.195 4.195 2.295 .435 3.833 3.797
41.503 6.053 4.550 6.053 1.503 3.778 3.778 2.275 .439 3.136 4.037
51.717 5.928 4.211 5.928 1.717 3.822 3.822 2.105 .475 3.469 5.069
62.093 5.943 3.850 5.943 2.093 4.018 4.018 1.925 .519 4.193 4.173
71.503 6.114 4.611 6.114 1.503 3.808 3.808 2.305 .433 3.145 4.037
81.811 5.842 4.013 5.842 1.811 3.827 3.827 2.015 .497 3.634 4.679
92.056 6.164 4.108 6.164 2.056 4.110 4.110 2.054 .486 4.112 2.964
10 2.066 6.221 4.155 6.221 2.066 4.143 4.143 2.077 .481 4.132 2.749
11 1.808 6.357 4.549 6.357 1.808 4.082 4.082 2.274 .439 3.663 4.031
12 1.867 5.977 4.110 5.977 1.867 3.922 3.922 2.055 .486 3.742 3.345
13 1.855 6.062 4.207 6.062 1.855 3.958 3.958 2.103 .475 3.724 2.685
14 1.890 6.104 4.214 6.104 1.890 3.997 3.997 2.107 .474 3.791 4.709
D
1
.235 5.412 5.657 5.412 .235 2.588 2.588 2.823 .354 1.186 2.017
D
2
1.173 5.412 4.239 5.412 1.173 3.292 3.292 2.119 .471 2.557 5.041
Note:D
1
(Cianidanol) and D
2
(Genistein).
8of23 ESHA ET AL.
3.3 |Analysis of physicochemical and pharmacokinetic properties
To determine whether a compound satisfies the Lipinski and Veber guidelines, the physicochemical features of fluoro-flavonoid derivatives (1–
14) must be analyzed. According to Lipinski's guidelines, a substance must fulfill the following five requirements before it can be administered
orally: (a) molecular weight (MW) <500 g/mol or less; (b) octanol–water partition coefficient (logP) below 5; (c) the number of H-bond donors
(HBD) contain no more than 5; (d) the number of H-bond acceptors (HBA) be no more than 10; and (e) the topological polar surface area (TPSA)
must not be larger than 140 Å
2
[81, 82]. Furthermore, Veber specifies two more requirements for drug bioavailability: (a) the number of rotatable
bonds (nrotb) must be less than 10, and (b) TPSA must be less than or equal to 140 Å
2
(Lipinski's guidelines). In this study, SwissADME was used
to determine the compliance of compounds (1–14) with the Lipinski and Veber rules, for those with the most intriguing biological activity. All
chemicals (1–14) were found to comply within the limits specified by Lipinski and Veber guidelines (Table 3and Table S15). Moreover, within the
limits, there was high agreement (less than 6) for compounds with molecular weights of less than 500 g/mol, MLOGP 4.15, and Log S (ESOL). A
drug-like (bioavailability) score of 1 shows that all compounds (1–14) met the requirements for good theoretical evidence in developing novel
medications.
Table 4displays the compliance of all 14 compounds with the six-parameter rule, indicating their potential as G protein-coupled receptors
(GPCR), ion channel modulators (ICM), kinase inhibitors (KI), nuclear receptor ligands (NRL), protease inhibitors (PI), and enzyme inhibitors (EI). As
per the Molinspiration drug-likeness calculation parameters in Table 4(see Figures S13 to S28), compound 11 demonstrated a GPCR value of
.20, highly comparable to the reference drug D2 (genistein).
To develop a novel drug cost-effectively and timely, it is crucial to assess the pharmacokinetic features, including ADMET. In this study,
SwissADME software (http://www.swissadme.ch/index.php) and admetSAR (http://lmmd.ecust.edu.cn/admetsar2/) were utilized to compute the
ADMET characteristics of all active compounds (1–14). Seven fundamental ADMET properties were evaluated for these compounds, as shown in
Table 5. These include human intestinal absorption (HIA), blood–brain barrier (BBB) penetration, plasma protein binding (PPB), cytochrome P450
enzymes (CYP3A4 and CYP2C19) inhibition, hERG inhibition, and synthetic accessibility (SA) score. It is noteworthy that crossing the blood–brain
barrier can be advantageous if the drug is intended to affect the central nervous system (CNS) but can be detrimental otherwise. High blood–
brain barrier penetration is defined as >2, medium absorption as 2 to .1, and low absorption as <.1 [82]. This score indicates drug-likeness (bio-
availability) and suggests that all compounds (1–14) meet the ADMET requirements, which is a promising outcome in the development of novel
drugs (see Table S15).
The findings of our study revealed that the majority of the compounds exhibited BBB penetration values ranging from .5 to .7, implying a
low degree of potential for brain absorption. These results are consistent with the BBB penetration values of reference drugs such as cianidanol
(.67) and genistein (.80). The reference range for potential risk of hERG activity inhibitors (pIC50) is between 5.5 and 6.0 [83]. According to
FIGURE 4 The molecular orbitals of the HOMO and LUMO for compounds 9and 11.
ESHA ET AL.9of23
our findings, compound 11 demonstrated an hERG pIC50 value of approximately .444, closely aligning with the reference drug cianidanol. We
used an approach to determine the synthetic accessibility (SA) score of the drug-like compounds (1–14)[
83], with a range of 1 (extremely simple)
to 10 (very difficult). The SA scores for all the compounds (1–14) fall within the range of 2.75 to 3.88.
3.4 |Analysis of pharmacological activities
In this study, Multilevel Neighborhoods of Atoms (MNA) descriptors were employed in PASS to comprehensively analyze the pharmacological
activities of a series of fluoro-flavonoid derivatives (1–14). The use of MNA descriptors allowed for a unique and detailed definition of the chemi-
cal structures, which in turn helped elucidate the compounds' potential biological activities. The tool PASS can predict multiple biological activities,
such as pharmacological main and side effects, mechanisms of action, mutagenicity, carcinogenicity, teratogenicity, and embryotoxicity. The bio-
logical activity of a compound can be influenced by various factors, including its structural and physicochemical properties, the biological entity
(such as species, gender, age, etc.), and the method of treatment (such as dose, route of administration, etc.). The MNA descriptors are used by
PASS to calculate two probabilities for the expected activity spectrum of a compound: probable activity (Pa) and probable inactivity (Pi). These
probabilities are represented by values ranging from .000 to 1.000 (where Pa +Pi ≠1), indicating the likelihood that a chemical will either be
active or inactive. The predictions from PASS are generally interpreted flexibly. For instance, if Pa >.7, there is a high chance of discovering the
activity experimentally. If .5 < Pa <.7, the compound is less likely to exhibit the activity experimentally, but it is probably dissimilar to known phar-
maceutical agents. If Pa <.5, the chance of discovering the activity experimentally is even less.
All 14 fluoro-flavonoid derivatives (1–14) exhibited Pa values greater than .7, as presented in Table 6. The membrane integrity agonist values
ranged from Pa =.830 to Pa =.957, nearly identical to those of the reference drugs cianidanol (Pa =.983) and genistein (Pa =.913). Compound
11 had a Pa value of .928 and a Pi value of .005 as a membrane integrity agonist, indicating that it was highly potent in promoting membrane
integrity, although slightly less potent than cianidanol. Thus, compound 11 was inferred to be a potent membrane integrity agonist with
FIGURE 5 Maps of electrostatic potential (.02 electrons Bohr
3
surface) (red =electron-rich, blue =electron-deficient) for compounds
9and 11.
10 of 23 ESHA ET AL.
comparable potency to the reference drugs cianidanol and genistein. Moreover, all compounds, except for compounds 10 and 11, showed HIF1A
expression inhibitor values between Pa =.714 to Pa =.952, indicating that these compounds could potentially be therapeutic agents for cancer
treatment. The inhibitory action of these compounds against the HIF pathway, which can be activated by tumor-induced hypoxia, supports their
potential as cancer treatments [84]. The expression of HIF1A is often upregulated in cancer cells and is associated with tumor growth and metas-
tasis. Compound 11 was found to have a Pa value of .781 and a PI value of .31 as a HIF1A expression inhibitor, indicating moderate potency in
TABLE 3 In silico prediction of physicochemical parameters for the fluoro-flavonoid derivatives (1–14).
Ligand MW LogP HBD HBA Nroth TPSA
Lipinski
a
≤500 ≤5≤5≤10
Veber
b
≤10 ≤140
1222.24 2.27 0 2 1 30.21
2240.23 2.67 0 3 1 30.21
3240.23 2.67 0 3 1 30.21
4256.23 2.06 1 4 1 50.44
5256.23 2.06 1 4 1 50.44
6256.23 2.06 1 4 1 50.44
7256.23 2.06 1 4 1 50.44
8364.34 3.83 1 5 3 50.44
9330.35 4.16 0 3 2 26.3
10 378.37 3.61 0 5 4 35.53
11 360.44 3.61 0 5 4 56.51
12 374.36 2.75 0 5 5 56.51
13 346.45 4.45 0 4 4 39.44
14 346.35 4.45 0 4 4 39.44
Cianidanol 209.27 0.24 5 6 1 110.38
Genistein 270.24 0.52 3 5 1 90.9
Abbreviations: HBA, number of hydrogen bond acceptors; HBD, number of hydrogen bond donors; LogP, lipophilicity (O/W); MW, molecular weight;
Nroth, number of rotatable bonds; TPSA, topological polar surface area (Å
2
).
a
Lipinski reference values.
b
Veber reference values.
TABLE 4 Drug-likeness assessment of fluoro-flavonoid derivatives (1–14). by Molinspiration.
Ligand GPCR ICM KI NRL PI EI
1.30 .21 .12 .18 .52 .03
2.17 .17 .00 .05 .43 .05
3.11 .13 .03 .14 .44 .07
4.30 .66 .17 .06 .87 .01
5.24 .34 .07 .01 .56 .13
6.17 .29 .10 .09 .44 .14
7.20 .30 .07 .03 .47 .12
8.05 .23 .04 .25 .16 .24
9.06 .34 .27 .07 .33 .04
10 .19 .45 .46 .08 .71 .05
11 .20 .26 .06 .02 .30 .03
12 .22 .49 .16 .18 .33 .08
13 .09 .23 .06 .02 .18 .04
14 .05 .19 .07 .05 .16 .07
Cianidanol .41 .14 .09 .60 .26 .47
Genistein .22 .54 .06 .23 .68 .13
ESHA ET AL.11 of 23
inhibiting the expression of HIF1A. Additionally, some compounds showed stronger activity as kinase inhibitors than the reference drugs that
have no activity, based on calculations of histidine kinase inhibitor and kinase inhibitor characteristics. Kinases are enzymes that play a crucial role
in numerous cellular processes, including cell growth, differentiation, and signaling, and dysregulated kinase activity is frequently linked with vari-
ous diseases, including cancer.
Compound 11 has been found to have a PA value of .823 and a PI value of .005 as a kinase inhibitor, which suggests that it is highly likely to
have activity as a kinase inhibitor and highly unlikely to be inactive as a kinase inhibitor. Since the reference drugs have no activity as a kinase
inhibitor, compound 11 has shown a higher probability of activity as a kinase inhibitor than the reference drugs. However, more studies are
needed to evaluate its efficacy and safety as a kinase inhibitor and determine its potential as a therapeutic agent in treating diseases associated
with dysregulated kinase activity, including cancer.
3.5 |In silico molecular docking
The in silico molecular docking study is one of the most essential tools of computational approaches. It was performed on fluoro-flavonoid com-
pounds (1–14) and reference standards of cianidanol and genistein for the inhibition of targeted proteins HER2 (7JXH), EGFR (4UV7), FPPS
(1YQ7), HPGDS (1V40), human DCK (1P60), and KEAP1 on Nrf2 (1X2J), as shown in Table 7. The molecular docking study revealed that com-
pound 11 had a strong binding affinity for HER2 (10.6 kcal mol
1
) and KEAP1 (9.6 kcal mol
1
). As shown in Table 7, all the docking values
obtained were more effective than the reference drugs cianidanol (HER2: 9.1 kcal mol
1
, KEAP1: 9.4 kcal mol
1
, and HPGDS:
8.4 kcal mol
1
) and genistein (HER2: 9.4 kcal mol
1
, KEAP1: 9.0 kcal mol
1
, and HPGDS: 8.6 kcal mol
1
) in terms of their potential for
cancer treatment and inflammation reduction.
The molecular docking study revealed that while some of the compounds (1–14) showed a strong binding affinity for multiple targeted pro-
teins specifically HER2 (Tables 7and S16), their binding affinity for human DCK, EGFR, and FPPS was relatively lower compared to their affinity
for HER2, HPGDS, and KEAP1. This suggests that these compounds may have potential as therapeutic agents for diseases associated with the
dysregulation of HER2, HPGDS, and KEAP1 on Nfr2. In drug development, it is important to consider the binding affinity of compounds and their
pharmacokinetics, toxicity, and metabolism. In vitro and in vivo studies are necessary to understand the behavior of the compounds in living
TABLE 5 In silico prediction of selected ADMET parameters for the fluoro-flavonoid derivatives (1–14).
Ligand HIA
a
BBB
a
PPB
a
CYP3A4 inhibition
a
CYP2C19 inhibition
a
hERG_pIC50
a
Synthetic accessibility score
b
1+(1.000) (.525) .927 +(.632) +(.899) (.708) 2.88
2+(1.000) +(.550) 1.034 +(.575) +(.768) +(.588) 2.81
3+(1.000) +(.550) .855 +(.575) +(.768) (.562) 2.79
4+(.997) (.675) 1.128 (.752) +(.664) (.757) 2.76
5+(.995) (.725) 1.073 (.749) +(.558) (.827) 2.96
6+(.995) (.725) 1.052 (.749) +(.558) (.773) 2.90
7+(.995) (.725) 1.062 (.749) +(.558) (.681) 2.94
8+(.996) (.575) .893 (.800) +(.762) (.623) 3.53
9+(1.000) +(.650) 1.24 (.685) +(.890) +(.715) 3.57
10 +(1.000) (.500) 1.205 (.744) +(.952) +(.680) 3.88
11 +(.985) (.575) 1.101 (.789) +(.636) (.444) 3.34
12 +(.989) (.550) 1.13 (.646) +(.885) +(.721) 3.40
13 +(.996) (.525) 1.133 (.763) +(.926) +(.809) 3.37
14 +(.996) (.525) 1.138 (.763) +(.926) +(.765) 3.38
D1 +(.892) (.675) 1.014 (.831) (.904) (.468) 3.50
D2 +(.967) (.800) 1.094 +(.796) +(.888) (.857) 2.87
Note: D1 (Cianidanol) and D2 (Genistein).
Abbreviations: BBB: blood–brain barrier penetration; CYP2C19: cytochrome P4502C19; CYP3A4: cytochrome P4503A4; hERG: human ether-a-go-go-
related gene, hERG inhibition potential (pIC
50
), the potential risk for inhibitors ranges 5.5–6; HIA: human intestinal absorption (%); PPB: plasma protein
binding.
a
The values are using admetSAR.
b
The values are using swissADME.
12 of 23 ESHA ET AL.
organisms and evaluate their safety and efficacy. Researchers can optimize the compounds and develop safe and effective therapeutic agents by
combining computational and experimental methods.
Additionally, the interactions between the ligand and protein help to describe how the ligand affects the active parts of the disease-causing
pathogens. The bond types and residue numbers are mentioned (Table S16), and 2D diagrams for (a) ligand in protein pocket, (b) hydrogen bond-
ing, and (c) ligand-protein interaction for compound 11 in HER2 and KEAP1 are shown in Figures 6, 7.
For the ligand-protein interaction through the amino acid residues of HER2 for compound 11 (Figure 6and Table S16), around eight different
interactions were observed. Four of them are hydrophobic, which include Pi-Alkyl on the residue VAL A: 734, ALA A: 751, LEU A: 796, and Pi-
sigma on the residue LEU A: 785. Two interactions are conventional hydrogen bonds at residue LYS A: 753 and ASP A: 863. Among the other
three interactions, two are halogen interactions on the residue GLU A: 770 and GLU A: 776. The protein-ligand interaction for compound 11 in
KEAP1 was illustrated, showing one conventional hydrogen interaction at residue VAL A:465, one halogen interaction with GLY A: 367, and four
interactions of pi-Alkyl type with residues ALA A:556, ARG A:415, ALA A:366, and VAL A: 606, as shown in Figure 7.
It was also observed that compound 9exhibited a strong binding affinity for HPGDS (9.6 kcal mol
1
) with only six interactions, including a
conventional hydrogen bond (TYR B: 208), Pi-Sulfur (MET B: 211), Pi-Alkyl (LEU B: 399), Pi-Pi Stacked (PHE B: 209), and amide-Pi Stacked (TRP
B: 304, GLY B: 213) (Figure S29). The investigation yielded a significant finding, highlighting compound 11 as an exceptionally promising candidate
for drug development due to its interaction with the HER2 protein. Compared to all other compounds (1–10,12–14) and the reference drugs
(cianidanol and genistein), compound 11 established a significantly higher interactive bond with the HER2 protein, involving interactions with
eight distinct amino acid residues (Table 8and Table S16). Compound 11 forms a more extensive network of interactive bonds with more amino
acid residues within the target protein, contributing significantly to its enhanced binding affinity. This extensive and specific binding profile under-
scores compound 11's remarkable affinity for the target protein.
Based on the results of molecular docking and interaction analysis, it has been concluded that compound 11 exhibits a stronger binding affin-
ity toward the target protein than other compounds. Additionally, compound 11 has a stronger interaction with the target protein, indicating that
it may be a more effective inhibitor of the proteins HER2 and KEAP1, which are associated with cancer. Thus, the in silico study suggests that
compound 11 has promising potential as a cancer-related protein inhibitor.
3.6 |In silico molecular dynamics
The results of the MD simulations were analyzed to gain insights into the structural stability and interactions of the protein-ligand complex of
HER2 with compound 11 over 20 ns. The docking analysis indicated compound 11 exhibited the strongest binding affinity (10.6 kcal mol
1
) with
TABLE 6 Predicted biological activity of the fluoro-flavonoid derivatives (1–14) using PASS.
Membrane integrity agonist HIF1A expression inhibitor Kinase inhibitor
Ligand Pa Pi Pa Pi Pa Pi
1.947 .004 .952 .003 .938 .002
2.928 .005 .886 .006 .869 .004
3.910 .009 .924 .004 .884 .003
4.830 .029 .890 .006 .556 .025
5.949 .004 .931 .004 .905 .003
6.941 .004 .943 .004 .913 .003
7.948 .004 .944 .004 .919 .002
8.830 .029 .714 .019 .576 .023
9.936 .004 .827 .010 .428 .055
10 .853 .023 .613 .031 .383 .075
11 .928 .005 .781 .031 .823 .005
12 .957 .003 .691 .021 .790 .006
13 .934 .005 .746 .016 .815 .005
14 .934 .005 .774 .014 .815 .005
Cianidanol .983 .001 .883 .007 - -
Genistein .913 .008 .939 .004 - -
ESHA ET AL.13 of 23
TABLE 7 Molecular docking simulation results for fluoro-flavonoid derivatives (1–14) against six targets.
Binding affinity (kcal mol
1
)
Ligand HER2 (7JXH) HPGDS (1V40) KEAP1 (1X2J) DCK (1P60) EGFR (4UV7) FPPS (1QY7)
19.1 8.8 8.2 8.5 7.3 7.9
29.2 9.0 8.6 6.8 7.7 7.5
39.4 8.3 8.5 6.5 7.4 7.9
48.8 8.7 8.6 7.5 7.4 8.7
59.4 9.4 8.8 8.1 7.5 8.1
69.4 9.4 8.8 8.8 7.5 8.4
78.8 8.3 8.6 7.1 7.4 8.9
89.8 8.9 9.5 8.1 7.6 7.3
99.8 9.6 8.7 8.8 8.3 7.4
10 8.4 8.5 8.3 8.6 7.4 6.7
11 10.6 8.7 9.6 7.6 8.5 6.9
12 10.1 8.6 9.1 8.8 7.8 6.7
13 9.7 8.6 8.4 7.5 7.8 6.7
14 9.5 8.5 8.7 7.4 7.8 6.9
Cianidanol 9.1 8.4 9.4 7.4 7.3 9.7
Genistein 9.4 8.6 9.0 7.7 7.4 8.4
FIGURE 6 Molecular docking poses: (A) Ligand in protein pocket; (B) Active site; (C) Hydrogen bonding in solid for compound 11; (D) Ligand-
protein interaction for 2D diagram of compound 11 in HER2.
14 of 23 ESHA ET AL.
the HER2 target protein and had good values for drug-like parameters. Various metrics were used to analyze the MD simulation results, including
RMSD, RMSF, and Rg values, as well as potential energies, temperature, hydrogen bonding and PCA.
During the simulation, we calculated the RMSD values of the ligand, protein backbone, and protein-ligand complex, which are shown in
Figure 8. The figure demonstrates that the RMSD of the protein backbone and protein-ligand complex remained constant within the binding site
throughout the simulation, with values ranging between .1 and .4 nm. The RMSD of the protein (green curve in Figure 8) was slightly higher than
that of the ligand (red curve in Figure 8), and the RMSD of the protein-ligand complex (black curve in Figure 8) was higher than that of the protein
(HER2) and ligand 11 (see Figures S30–S32). These results suggest that the ligand and protein are likely to bind tightly and persistently, and the
complex will be able to maintain its overall shape despite the atoms' natural thermal and dynamic motions.
The RMSF measures the average deviation of individual atoms in a protein from their average positions. Separate RMSF values were calcu-
lated for the protein and ligand (See Figures S31–S32). Figure 9shows the RMSF values for the protein (HER2) and ligand 11 complex, illustrating
the fluctuations in the positions of the atoms throughout the simulation and providing insight into their dynamic behavior and interactions. The
RMSF values for the protein range between .01 and .3 Å, except for a fluctuation at atom 950 that reaches up to .28 Å. The ligand's RMSF values
range between .01 and .03 Å, with fluctuations at atoms 2700 and 4500 that reach up to .45 and .5 Å, respectively. These results indicate that
both the ligand and protein are relatively stable during the simulation, with only minor fluctuations in atom position. Notably, the fluctuations at
specific atoms in the ligand may be significant and potentially related to a specific function or interaction of the protein. However, further investi-
gation is required to determine the fluctuations' significance and potential impact on HER2's biological activity.
The Rg is a parameter that indicates the size and conformational changes of the protein structure during an MD simulation. In this study, Rg
values were used to analyze the interaction between the target protein (HER2) and ligand (11) and their environment. Like the RMSD plots, the
Rg values of the ligand, protein, and protein-ligand complex showed only minor changes during the simulation period, indicating that their struc-
tures were minimally affected (Figure 10 and Figures S31–S32). The Rg values for the protein-ligand complex and protein were 1.97 and 1.81 nm,
respectively, as shown in Figure 10. Furthermore, the similarity in Rg values between the protein and the protein-ligand complex suggests that
FIGURE 7 Molecular docking poses: (A) Ligand in protein pocket; (B) Active site; (C) Hydrogen bonding in solid for compound 11; (D) Ligand-
protein interaction for 2D diagram of compound 11 in KEAP1.
ESHA ET AL.15 of 23
binding did not induce substantial conformational changes within the protein. However, it is essential to note that while Rg values provide insight
into a protein's overall size and shape, they may not reflect the exact structural changes that occur at the molecular level.
The results of the MD simulation show that the number of hydrogen bonds formed between compound 11 and the active site of HER2 varied
between 0 and 4 over the 20 ns simulation period, as shown in Figure 11. The variability in the number of hydrogen bonds formed indicates that
compound 11 has the potential to interact with the active site of HER2 and form stable hydrogen bonds with the protein. However, further analy-
sis is necessary to fully understand the nature and significance of these interactions and their potential impact on the biological activity of HER2.
The temperature of the system, as shown in Figure 12, remained relatively stable throughout the 20 ns simulation, fluctuating between
298 and 303 Kelvin for HER2. This suggests that the system was able to maintain a consistent thermal energy, indicating that the protein-ligand
complex did not significantly affect the overall temperature of the system. However, the potential energy of the system showed fluctuations over
the course of the simulation, ranging from 5.89e+05 to 5.87e+05 kJ mol
1
for HER2 (Figure 12).
The simulated MD trajectories of the protein-ligand complex of HER2 with the compound 11 at 300 K was subjected to the conformational
PCA of Cαatoms. In the current work, PCA was done to determine the variability, collective motions, and changes in protein structural conforma-
tional states in the subsets of the principal components that appeared throughout the MD simulations. The results for the PCA study using MD
TABLE 8 Ligand-protein interaction for Human epidermal growth factor receptor 2 (PDB ID: 7JXH) with compound 9,11,12, and reference
drug Genistein.
Drug Amino acid residue Bond category Bond type 2D diagram of interaction
9ALA A:46
LEU A:21
VAL A:29
LEU A:147
THR A:170
GLU A:169
Hydrophobic
Hydrophobic
Hydrophobic
Hydrophobic
Halogen
Halogen
Pi-Alkyl
Pi-Alkyl
Pi-Alkyl
Pi-Sigma
Fluorine
Fluorine
11 ASP A:863
LYS A:753
ALA A:751
VAL A:734
LEU A:796
LEU A:785
GLU A:770
GLU A:766
Hydrogen Bond
Hydrogen Bond
Hydrophobic
Hydrophobic
Hydrophobic
Hydrophobic
Halogen
Halogen
Conventional HB
Conventional HB
Pi-Alkyl
Pi-Alkyl
Pi-Alkyl
Pi-Sigma
Fluorine
Fluorine
12 LEU A:21
ALA A:46
LYS A:48
LEU A:91
LEU A:147
VAL A:29
THR A:93
Hydrophobic
Hydrophobic
Hydrophobic
Hydrophobic
Hydrophobic
Hydrophobic
Hydrophobic
Pi-Alkyl
Pi-Alkyl
Pi-Alkyl
Pi-Alkyl
Pi-Sigma
Pi-Sigma
Pi-Sigma
Genistein GLU A:65
LEU A:91
ALA A:46
LUS A:48
VAL A:29
LEU A:80
Hydrogen Bond
Hydrogen Bond
Hydrogen Bond
Hydrophobic
Hydrophobic
Hydrophobic
Conventional HB
Conventional HB
Conventional HB
Pi-Alkyl
Pi-Alkyl
Pi-Sigma
16 of 23 ESHA ET AL.
trajectories of the complex of HER2 with compound 11 at 300 K were carried out via the Bio3D package [74–76], and the obtained eigenvalues
versus eigenvector plots are represented in Figure 13. The dominating motion from the trajectory is extracted in the smaller subset and further
compared for the first three eigenvectors (PC1, PC2, and PC3). The color dots were used to represent the captured variance by eigenvectors. The
protein-ligand complex simulated at 300 K showed the highest variability in PC1 (40.76%) with regard to the internal motions of the MD trajec-
tory. While PC2 showed a lower percentage of variance (16.78%) compared with PC1 and PC3 explains 9.26%. The cumulative variance explained
by these three components is 66.8% (Figure 13). In addition, the cosine content value of the eigenvector calculated from the MD trajectory
showed .7 indicating the simulation is converged.
Based on the results of the simulation, it can be inferred that compound 11 has a stronger interaction with HER2, which indicates its potential
to act as an inhibitor of HER2. This finding is particularly promising since HER2 is frequently overexpressed in cancer cells. Therefore, compound
FIGURE 8 RMSD evolution for the protein (HER2, green), ligand (11, red), and protein-ligand complex (black) during 20 ns MD simulation.
FIGURE 9 RMSF evolution for the protein (HER2, black) and protein-ligand complex (red) during 20 ns MD simulation.
ESHA ET AL.17 of 23
11 could have potential as an anti-cancer agent. However, it is crucial to note that additional experimental studies are required to validate the
effectiveness and safety of compound 11 as an HER2 inhibitor and its potential as an anti-cancer drug.
In summary, to validate our identification of compound 11 as a promising candidate for anti-lung cancer drug development, we conducted
molecular dynamics simulations for three additional compounds: compound 9, compound 12, and the reference drug genistein. These compounds
were selected based on their binding energy scores, which closely resembled that of compound 11. The purpose of these simulations was to com-
pare these compounds' stability and structural characteristics with those of compound 11. The analysis yielded compelling evidence supporting
the superior performance of compound 11 (Figure 14).
Regarding RMSD, compound 9, compound 12, and genistein displayed values ranging from .1 to .5 Å, indicating structural fluctuations during
the simulations. In contrast, compound 11 consistently demonstrated exceptional stability, with RMSD values confined to a narrow range of .1 to
.15 Å (Figure S32). This remarkable stability sets compound 11 apart from the other compounds. Furthermore, an examination of Root Mean
Square Fluctuation (RMSF) values revealed that compound 11 exhibited values in the range of .1 to .2 Å (Figure S33), indicating minimal fluctua-
tions in its structural dynamics. Conversely, the other three compounds displayed more significant variability. Moreover, the Rg values, which
reflect the overall compactness of the compounds, further affirmed the stability of compound 11. Compound 9, compound 12, and genistein con-
sistently had Rg values above 1.95 Å, indicative of extended conformations. In contrast, Compound 11 displayed compact Rg values in the 1.8 to
1.81 Å range (Figure S34), reflecting a well-defined and compact structure.
According to the MD simulation results presented in Figure 14 (Figure S35), the number of hydrogen bonds formed between compounds 9,
11,12, and genistein at the active sites of (PDB: 7JXH) ranged from 0 to 4 during the 20 ns simulation period (Figure 14), similar to compound 11
(Figure 11). However, the other ligands exhibited a narrower range compared with Compound 11. This variability suggests that the interaction
between the ligands and the protein is dynamic, allowing the ligands to adopt different conformations and binding modes within the protein-
binding pocket. The formation of stable hydrogen bonds indicates that Compound 11 has the potential to interact effectively with both active
sites of the protein.
Furthermore, PCA demonstrated significantly higher convergence in the simulation of compound 11 with the HER2 protein compared to sim-
ulations involving compound 9, compound 12, and Genistein (Figure S35–S37). This heightened convergence was quantified using cosine content
values, with compound 11 displaying the lowest value of .7, indicating a strong alignment with essential structural dynamics. In contrast, com-
pound 12 exhibited a slightly higher value of .73, genistein registered at .83, and compound 9had the highest value at .92. These results empha-
size the exceptional convergence of compound 11 in its interaction with the HER2 protein.
FIGURE 10 Radius of gyration, Rg (nm) versus time (ps) plots of the target protein, HER2 (Rg
X
), and protein-ligand complex, HER2-11 (Rg
Y
)
during 20 ns MD simulation.
18 of 23 ESHA ET AL.
These comprehensive analyses highlight compound 11 as the most stable and structurally consistent candidate among the compounds exam-
ined. Its low RMSD reduced RMSF fluctuations and compact Rg values as well hydrogen bonding collectively confirm its potential as a robust can-
didate for developing an anti-lung cancer drug. These findings underscore compound 11's promise as a valuable asset in lung cancer therapeutic
research and warrant further experimental validation and exploration of its pharmacological properties.
FIGURE 11 Number of hydrogen bond versus time (ps) plots for the hydrogen bond stabilization during 20 ns MD simulation.
FIGURE 12 Time evolution of (A) temperature and (B) potential energies during 20 ns MD.
ESHA ET AL.19 of 23
4|CONCLUSION
The study aimed to optimize a series of fluoro-flavonoid derivatives (1–14) by utilizing B3LYP in conjunction with three different basis sets:
6-31G(d,p), 6-311G(d,p), and 6–311++G(d,p). The study also involved in silico investigation of these derivatives (1–14) to assess their potential
use against six cancer-associated proteins. This study utilized a range of scientific methods to investigate the properties and potential biological
activity of the analyzed compound. These methods included (DFT) calculations, molecular docking calculations, thermodynamic analysis, investiga-
tion of HOMO and LUMO, drug-likeness assessments, absorption, distribution, metabolism, excretion, and toxicity (ADMET) evaluations, docking,
molecular dynamics (MD) simulations, and examination of the biological activity spectra. The analyzed compounds (1–14) showed promising bind-
ing affinity as potential anti-cancer agents, with computational analysis emphasizing the high efficacy of compound 11. Moreover, these com-
pounds were deemed non-carcinogenic and demonstrated low levels of acute oral toxicity, implying their potential suitability for oral
administration. Crucially, all molecules were in accordance with the Lipinski, Ghose, Veber, Egan, and Muegge rules. Based on our study, it has
FIGURE 13 PCA of the MD trajectory of the complex of HER2 with compound 11 at 300 K. Black dots indicate the energetically unstable
conformational state, and red dots indicate the stable conformational state.
20 of 23 ESHA ET AL.
been found that the investigated compounds have potential as therapeutic agents, with compound 11 showing comparable binding affinity to
commonly used medications, cianidanol and genistein, against three human cancer targets: HER2, KEAP1, and HPGDS. Compound 11 exhibited a
stronger inhibitory effect on HER2 and KEAP1 when compared to the reference drugs, which is significant as HER2 is frequently involved in the
development and progression of various cancers. Additionally, molecular dynamics simulations confirmed its stable conformation and alignment
with the target protein, further solidifying its potential as a candidate for developing anti-lung cancer drugs. These findings suggest that com-
pound 11 could be a promising candidate for the development of new anti-cancer therapies. However, additional experiments involving cancer
cell lines will be necessary to confirm the efficacy of compound 11.
AUTHOR CONTRIBUTIONS
The authors collectively formulated and planned the calculations, conducted data analysis and interpretation, provided materials, analysis tools, or
data and software, and contributed to the writing of the paper. Nusrat Jahan Ikbal Esha: Data curation; formal analysis; investigation; methodol-
ogy; writing—original draft. Syeda Tasnim Quayum and Minhaz Zabin Saif: Resources; visualization. Mansour H. Almatarneh, Shofiur Rahman, and
Abdullah Alodhay: writing—review and editing. Raymond Poirier and Kabir M. Uddin: conceptualization; investigation; project administration;
supervision; resources; software; supervision; validation; visualization; writing—review and editing.
ACKNOWLEDGMENTS
The authors extend their appreciation to the Deputyship for Research and Innovation, “Ministry of Education”in Saudi Arabia for funding this
research (IFKSUOR3–303-2). We would like to express our gratitude to the Digital Research Alliance of Canada for providing us with computer
time. Furthermore, we acknowledge the support provided by NSU Conference and Travel Grant Committee (CTRGC).
FUNDING INFORMATION
The author Abdullah Alodhayb acknowledges Researchers Supporting Project number RSP2023R304, King Saud University, Riyadh, Saudi Arabia.
Furthermore, we acknowledge with thanks the financial backing received from CTRG at NSU.
FIGURE 14 Merged graphical data of (A) RMSD, (B) RMSF, and (C) R
g
values and (D) Number of hydrogen bond for the protein HER2 with
compound 9 (blue), 11 (black), 12 (red), and reference drug genistein (green) over the course of 20 ns MD simulation.
ESHA ET AL.21 of 23
CONFLICT OF INTEREST STATEMENT
The authors have no relevant financial or non-financial interests to disclose.
DATA AVAILABILITY STATEMENT
The list provided includes various software and websites: AdmetSAR, http://lmmd.ecust.edu.cn/admetsar2/; SwissADME, http://www.
swissadme.ch/; Pass prediction http://www.way2drug.com/passonline/; and coordinates of stable local minimum structures can be found in
Data S1.
ORCID
Mansour H. Almatarneh https://orcid.org/0000-0002-2863-6487
Shofiur Rahman https://orcid.org/0000-0003-4219-4758
Abdullah Alodhayb https://orcid.org/0000-0003-0202-8712
Raymond A. Poirier https://orcid.org/0000-0002-8533-7846
Kabir M. Uddin https://orcid.org/0000-0002-5518-2345
REFERENCES
[1] A. D. Westwell, IDrugs 2008,11, 783.
[2] P. Amudha, M. Jayalakshmi, R. Vidya, B. N. Poojitha, Appl. Biochem. Biotechnol. 2022,195, 2597.
[3] G. M. Cooper, The Development and Causes of Cancer, Nih.gov, Washington, DC, 2000,2, 719.
[4] C. S. Dela Cruz, L. T. Tanoue, R. A. Matthay, Clin. Chest Med. 2011,32, 605.
[5] L. K. Robert, Lung Carcinoma –Pulmonary Disorders, MSD Manual Professional Edition, Merck & Co, Inc., Rahway, NJ, USA. 2023.
[6] M. Huang, J.-J. Lu, J. Ding, Nat. Prod. Bioprospect. 2021,11,5.
[7] D. J. Newman, G. M. Cragg, J. Nat. Prod. 2020,83,3.
[8] D. Veeramuthu, W. R. T. Raja, N. A. Al-Dhabi, I. Savarimuthu, Flavonoids –From Biosynthesis to Human Health, IntechOpen, London 2017.
[9] O. Zanoaga, C. Braicu, A. Jurj, A. Rusu, R. Buiga, I. Berindan-Neagoe, Int. J. Mol. Sci. 2019,20, 4291.
[10] P. Pratheeshkumar, C. Sreekala, Z. Zhang, A. Budhraja, S. Ding, Y.-O. Son, X. Wang, A. Hitron, K. Hyun-Jung, L. Wang, J.-C. Lee, X. Shi, Anti Cancer
Agents Med. Chem. 2012,12, 1159.
[11] M. H. Al-Jumaili, M. K.Al Hdeethi, CancerInformatics 2021,2, 11769351211055160.
[12] B. Krusinska, I. Hawrysz, L. Wadolowska, M. Slowinska, M. Biernacki, A. Czerwinska, J. Golota, Nutrients 2018,10, 470.
[13] L. Budisan, D. Gulei, O. M. Zanoaga, A. I. Irimie, C. Sergiu, C. Braicu, C. D. Gherman, I. Berindan-Neagoe, Int. J. Mol. Sci. 2017,18, 1178.
[14] A. N. Panche, A. D. Diwan, S. R. Chandra, J. Nutr. Sci. 2016,5, e47.
[15] P. Venkateswara Rao, S. D. V. S. Kiran, P. Rohini, P. Bhagyasree, J. Pharmacogn. Phytochem 2017,6, 2778.
[16] V. P. Sülsen, E. Lizarraga, N. Z. Mamadalieva, J. H. G. Lago, Evid. Based Complement. Alternat. Med. 2017,2017,1.
[17] M. D. Awouafack, P. Tane, H. Morita, Flavonoids –From Biosynthesis to Human Health, IntechOpen, London 2017.
[18] K. N. Kamlesh, T. Sivakumar, A. Afroze, J. Appl. Pharm. 2017,09, 1000232.
[19] S. Ruiz-Cruz, S. Chaparro-Hernández, K. L. Hernández- Ruiz, L. A. Cira-Chávez, M. I. Estrada-Alvarado, L. E. Gassos Ortega, J. de Jesús Ornelas-Paz,
M. A. L. Mata, Flavonoids –From Biosynthesis to Human Health, IntechOpen, London 2017.
[20] N. Singh, M. Rani, R. Sharmilla, A. K. Yadav, MOJ Food Process. Technol 2017,4, 96.
[21] M. G. Hertog, E. J. Feskens, D. Kromhout, Lancet 1997,349, 699.
[22] J. M. Hodgson, Asia Pac. J. Clin. Nutr. 2008,17, 288.
[23] A. Ekalu, J. D. Habila, Beni-Suef University Journal of Basic andApplied Sciences 2020,9,1.
[24] A. A. Shafei, Int J Pharmacogn Phytochem Res 2016,8, 836.
[25] R. J. Nijveldt, E. van Nood, D. E. van Hoorn, P. G. Boelens, K. van Norren, P. A. van Leeuwen, Am. J. Clin. Nutr. 2001,74, 418.
[26] G. Leonarduzzi, G. Testa, B. Sottero, P. Gamba, G. Poli, Curr. Med. Chem. 2010,17, 74.
[27] H. S. Hafez, D. A. Ghareeb, S. R. Saleh, M. M. Abady, M. A. El Demellawy, H. Hussien, N. Abdel-Monem, Psychopharmacology 2017,234, 3037.
[28] Cianidanol, Uses, Interactions, Mechanism of Action jDrugBank Online.
[29] H. Sun, M. Yin, D. Hao, Y. Shen, Appl. Sci. 2020,10, 2065.
[30] Genistein Uses, Interactions, Mechanism of Action jDrugBank Online.
[31] S. Banerjee, Y. Li, Z. Wang, F. H. Sarkar, Cancer Lett. 2008,269, 226.
[32] H. S. Tuli, M. J. Tuorkey, F. Thakral, K. Sak, M. Kumar, A. K. Sharma, U. Sharma, A. Jain, V. Aggarwal, A. Bishayee, Front. Pharmacol. 2019,10, 1336.
[33] T. Teslova, C. Corredor, R. Livingstone, T. Spataru, R. L. Birke, J. R. Lombardi, M. V. Cañamares, M. Leona, J. Raman Spectrosc. 2007,38, 802.
[34] D. Koh, IUCrData 2018,3, x181332.
[35] M. P. Waller, D. E. Hibbs, J. Overgaard, J. R. Hanrahan, T. W. Hambley, Acta Crystallogr 2003,59, o767.
[36] Z. Dolatkhah, S. Javanshir, A. S. Sadr, J. Hosseini, S. Sardari, J. Chem. Inf. Model. 2017,57, 1246.
[37] S. Rathod, K. Shinde, J. Porlekar, P. Choudhari, R. Dhavale, D. Mahuli, Y. Tamboli, M. Bhatia, K. P. Haval, A. G. Al-Sehemi, M. Pannipara, ACS Omega
2023,8, 391.
[38] R. Abdizadeh, F. Hadizadeh, T. Abdizadeh, J. Biomol. Struct. Dyn. 2023,41, 5915.
[39] Y. Lu, B. Zhang, N. Wang, M. Li, N. Xi, Crystals 2022,12, 990.
[40] H. R. A. El-Mageed, D. A. Abdelrheem, M. O. Rafi, M. T. Sarker, K. Al-Khafaji, M. J. Hossain, R. Capasso, T. B. Emran, Biologics 2021,1, 416.
[41] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Petersson, H. Nakatsuji, X. Li, M.
Caricato, A. V. Marenich, J. Bloino, B. G. Janesko, R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, D. Williams-
22 of 23 ESHA ET AL.
Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W.
Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell, J. A.
Montgomery Jr., J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, T. A. Keith, R. Kobayashi, J. Normand,
K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo, R. Cammi, J. W. Ochterski, R. L.
Martin, K. Morokuma, O. Farkas, J. B. Foresman, D. J. Fox, Gaussian 16,Revision C.01, Gaussian, Inc., Wallingford, CT 2019.
[42] (a) K. M. Uddin, A. I. Alrawashdeh, D. J. Henry, P. L. Warburton, R. A. Poirier, Int. J. Quantum Chem. 2019,120, 1. (b) K. M. Uddin, D. J. Henry, A. I.
Alrawashdeh, P. L. Warburton, R. A. Poirier, Struct. Chem. 2017,28, 1467. (c) K. M. Uddin, M. H. Almatarneh, D. M. Shaw, R. A. Poirier, J. Phys. Chem.
A2011,115, 2065. (d) K. M. Uddin, R. A. Poirier, J. Phys. Chem. B 2011,115, 9151. (e) K. M. Uddin, C. G. Flinn, R. A. Poirier, P. L. Warburton, Comput.
Theor. Chem 2014,1027, 91. (f) A. I. Alrawashdeh, M. H. Almatarneh, R. A. Poirier, Can. J. Chem. 2013,91, 518. (g) M. H. Almatarneh, C. G. Flinn, R. A.
Poirier, W. A. Sokalski, J. Phys. Chem. A 2006,110, 8227. (h) K. M. Uddin, M. A. Hosen, M. F. Khan, Y. Ozeki, S. M. A. Kawsar, Philipp. J. Sci. 2022,151,
2215.
[43] C. M. Nisha, A. Kumar, P. Nair, N. Gupta, C. Silakari, T. Tripathi, A. Kumar, Adv. Bioinforma. 2016,2016,1.
[44] J. A. Chamizo, J. Morgado, P. Sosa, Organometallics 1993,12, 5005.
[45] R. A. Alberty, J. Chem. Educ. 1960,37, 660.
[46] R. G. Parr, L. V. Szentpály, S. Liu, J. Am. Chem. Soc. 1999,121, 1922.
[47] R. G. Pearson, J. Am. Chem. Soc. 1985,107, 6801.
[48] A. Daina, O. Michielin, V. Zoete, Sci. Rep. 2017,7, 42717.
[49] A. K. Srivastava, D. Singh, B. K. Roy, Interdiscip. Sci. 2017,9, 116.
[50] D. S. Druzhilovskiy, A. V. Rudik, D. A. Filimonov, T. A. Gloriozova, A. A. Lagunin, A. V. Dmitriev, P. V. Pogodin, V. I. Dubovskaya, S. M. Ivanov, O. A.
Tarasova, V. M. Bezhentsev, K. A. Murtazalieva, M. I. Semin, I. S. Maiorov, A. S. Gaur, G. N. Sastry, V. V. Poroikov, Russ. Chem. Bull. 2017,66, 1832.
[51] S. Dallakyan, A. J. Olson, Methods Mol. Biol. 2014,1263, 243.
[52] H. M. Berman, Nucleic Acids Res. 2000,28, 235.
[53] C. Zardecki, S. Dutta, D. S. Goodsell, M. Voigt, S. K. Burley, J. Chem. Educ. 2016,93, 569.
[54] N. Iqbal, Mol. Biol. Int. 2014,2014,1.
[55] Advances in Lung Cancer Research. National Cancer Institute.
[56] H. S. Budi, F. N. Ahmad, H. Achmad, M. J. Ansari, M. V. Mikhailova, W. Suksatan, S. Chupradit, N. Shomali, F. Marofi, Stem Cell Res Ther 2022,13, 40.
[57] A. A. Thai, B. J. Solomon, L. V. Sequist, J. F. Gainor, R. S. Heist, Lancet 2021,398, 535.
[58] N. Camiña, T. M. Penning, Br. J. Cancer 2022,126, 1244.
[59] T. D. Goddard, C. C. Huang, T. E. Ferrin, J. Struct. Biol. 2007,157, 281.
[60] O. Trott, A. J. Olson, J. Comput. Chem. 2009,31, 455.
[61] S. Yuan, H. C. S. Chan, Z. Hu, Wiley Interdiscip. Rev. Comput. Mol. Sci. 2017,7, e1298.
[62] S. Nawsheen, M. A. Sufian, K. N. Deepa, Health 2019,11, 129.
[63] D. Van Der Spoel, E. Lindahl, B. Hess, G. Groenhof, A. E. Mark, H. J. C. Berendsen, J. Comput. Chem. 2005,26, 1701.
[64] S. A. Showalter, R. Brüschweiler, J. Chem. Theory Comput. 2007,3, 961.
[65] T. Hansson, C. Oostenbrink, W. van Gunsteren, Curr. Opin. Struct. Biol. 2002,12, 190.
[66] S. A. Bray, X. Lucas, A. Kumar, B. A. Grüning, J. Chem. 2020,12, 40.
[67] D. Presti, A. Pedone, G. Mancini, C. Duce, M. R. Tiné, V. Barone, Phys. Chem. Chem. Phys. 2016,18, 2164.
[68] M. A. Cuendet, W. F. van Gunsteren, J. Chem. Phys. 2007,127, 184102.
[69] K. M. Uddin, M. Sakib, S. Siraji, R. Uddin, S. Rahman, A. Alodhayb, K. A. Alibrahim, A. Kumer, M. M. Matin, M. M. H. Bhuiyan, ACS Omega 2023,8,
25817.
[70] W. Humphrey, A. Dalke, K. Schulten, J. Mol. Graph. 1996,14, 33.
[71] S. K. Panda, P. S. S. Gupta, M. K. Rana, Cell Biochem. Funct. 2022,41, 98.
[72] S. Sen, S. K. Panda, A. K. Nayak, M. K. Rana, J. Phys. Chem. Lett. 2023,14, 3230.
[73] S. Biswal, P. Gupta, S. K. Panda, H. R. Bhat, M. K. Rana, Photosynth. Res. 2023,156, 337.
[74] E. Afgan, D. Baker, B. Batut, M. van den Beek, D. Bouvier, M. ˇ
Cech, J. Chilton, D. Clements, N. Coraor, B. A. Grüning, A. Guerler, J. Hillman-Jackson, S.
Hiltemann, V. Jalili, H. Rasche, N. Soranzo, J. Goecks, J. Taylor, A. Nekrutenko, D. Blankenberg, Nucleic Acids Res. 2018,46, W537.
[75] B. J. Grant, A. Lúcia, K. M. J. ElSawy, A. McCammon, L. S. D. Caves, Bioinformatics 2006,22, 2695.
[76] N. Kumar, A. Awasthi, A. Kumari, D. Sood, P. Jain, T. Singh, N. Sharma, A. Grover, R. Chandra, J. Biomol. Struct. Dyn. 2020,40, 101.
[77] N. Cohen, S. W. Benson, Chem. Rev. 1993,93, 2419.
[78] U. Bhattacharya, S. K. Panda, P. S. S. Gupta, M. K. Rana, J. Mol. Struct. 2022,1253, 132258.
[79] P. S. S. Gupta, H. R. Bhat, S. Biswal, M. K. Rana, J. Mol. Liq. 2020,320, 114375.
[80] A. K. Ray, P. S. Sen Gupta, S. K. Panda, S. Biswal, U. Bhattacharya, M. K. Rana, Comput. Biol. Med. 2022,142, 105183.
[81] C. A. Lipinski, Drug Discov. Today Technol. 2004,1, 337.
[82] D. F. Veber, S. R. Johnson, H.-Y. Cheng, B. R. Smith, K. W. Ward, K. D. Kopple, J. Med. Chem. 2002,45, 2615.
[83] D. M. Shadrack, V. M. K. Ndesendo, Comput. Mol. Biol. 2017,07,1.
[84] G. N. Masoud, W. Li, Acta Pharm. Sin. B 2015,5, 378.
SUPPORTING INFORMATION
Additional supporting information can be found online in the Supporting Information section at the end of this article.
How to cite this article: N. J. I. Esha, S. T. Quayum, M. Z. Saif, M. H. Almatarneh, S. Rahman, A. Alodhayb, R. A. Poirier, K. M. Uddin, Int. J.
Quantum Chem. 2023, e27274. https://doi.org/10.1002/qua.27274
ESHA ET AL.23 of 23
