Synthesis of new 2-aminothiazolyl/benzothiazolyl-based 3,4-dihydropyrimidinones and evaluation of their effects on adenocarcinoma gastric cell migration

Article abstract of DOI:10.1007/s11030-021-10229-z

Abstract: Gastric cancer is one of the malignant tumors of the gastrointestinal tract that, despite its decrease in recent years, is still the fourth most common cancer and the second leading cause of cancer-related death. Various strategies including chemotherapy are used to keep cancer cells from spreading and induce apoptotic death in them. Recent studies have shown that dihydropyrimidinones (DHPMs) are privileged structures in medicinal chemistry due to their pharmacological effects. A number of new 2-aminothiazolyl/benzothiazolyl derivatives of 3,4-DHPMs (3–8) were synthesized and structurally identified, and then their effects on the migration behavior of human AGS cells (gastric cancer cells) were investigated. Molecular docking and molecular dynamics (MD) simulations were applied to explore binding potential and realistic binding model of the assessed derivatives through identification of key amino acid residues within L5/α2/α3 allosteric site of kinesin 5 (Eg5) as a validated microtubule-dependent target for monastrol as a privileged DHPM derivative. Graphical abstract: [Figure not available: see fulltext.].

Full text of DOI:10.1007/s11030-021-10229-z

Molecular Diversity
https://doi.org/10.1007/s11030-021-10229-z
ORIGINAL ARTICLE
Synthesis of new 2‑aminothiazolyl/benzothiazolyl‑based
3,4‑dihydropyrimidinones and evaluation of their efects
on adenocarcinoma gastric cell migration
Mohsen Sagha1 · Fatemeh Mousaei2 · Mahtab Salahi2 · Nima Razzaghi‑Asl3
Received: 1 March 2021 / Accepted: 27 April 2021
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2021
Abstract
Gastric cancer is one of the malignant tumors of the gastrointestinal tract that, despite its decrease in recent years, is still the
fourth most common cancer and the second leading cause of cancer-related death. Various strategies including chemotherapy
are used to keep cancer cells from spreading and induce apoptotic death in them. Recent studies have shown that dihydro-
pyrimidinones (DHPMs) are privileged structures in medicinal chemistry due to their pharmacological efects. A number
of new 2-aminothiazolyl/benzothiazolyl derivatives of 3,4-DHPMs (3–8) were synthesized and structurally identifed, and
then their efects on the migration behavior of human AGS cells (gastric cancer cells) were investigated. Molecular dock-
ing and molecular dynamics (MD) simulations were applied to explore binding potential and realistic binding model of the
assessed derivatives through identifcation of key amino acid residues within L5/α2/α3 allosteric site of kinesin 5 (Eg5) as
a validated microtubule-dependent target for monastrol as a privileged DHPM derivative.
Graphical abstract
Keywords Gastric cancer · 3,4-Dihydropyrimidinones · Cell migration
Extended author information available on the last page of the article
Vol.:(0123456789)
1 3

Molecular Diversity
Introduction
Cancer refers to a family of diseases characterized with
uncontrolled growth of cells that are able to spread out
to other parts of the body [1]. Cancers can originate from
environmental and hereditary causes. Eating habits, lack of
physical activity, alcohol consumption, smoking and infec-
tion play an important role in causing cancer [2]. Gastric
cancer is one of the malignant tumors of the gastrointestinal
tract that, despite its decreasing rate in recent years, is still
the ffth incidence and third mortality worldwide [3, 4].
Cell migration plays an important role in the spread of
cancer cells, tissue invasion and metastasis, which is the
leading cause of death in cancer patients. Metastasis is a
complex pathological phenomenon in which the cancer
cells with epithelial origin transit from the epithelial state
to the mesenchymal phenotype, a process that is called
EMT and propagate from the site of the primary tumor to
invade a location other than the site of the primary tumor
[5]. There have been many advances in the treatment of
gastric cancer recently, but the treatment of gastric cancer
still remains an important challenge [6]. Therefore, it is
necessary to seek appropriate and promising strategies for
the treatment of gastric cancer. So, the design and develop-
ment of compounds that are able to prevent cancer cells
from spreading and migrating from one tissue to another
(metastasis) are of particular interest for the treatment of
advanced stages of cancer.
Fig. 1 Chemical structure of monastrol
their derivatives as important cytotoxic and anticancer
agents were reported in the literature to act via allosteric
binding mechanism [14, 15]. In the current study, with the
aim of incorporating thiazole moiety and its more hydro-
phobic analogue, i.e., benzothiazole into the monastrol-
derived 3,5-disubstituted dihydropyrimidine structures
as validated Eg5 inhibitors, a number of new 2-amino-
6-methylbenzothiazolyl 3,4-dihydropyrimidinone (3–5)
and 2-aminothiazolyl 3,4-dihydropyrimidinone (6–8)
derivatives were synthesized and structurally identifed
to evaluate their effects on the migratory behavior of
human gastric cancer cells. Subsequently; binding ability
of assessed compounds within L5/α2/α3 allosteric site of
kinesin 5 (Eg5) as a validated microtubule associated tar-
get for DHPMs was explored through molecular docking
and molecular dynamics simulations. It was interesting to
note that improved polar interactions were probably driv-
ing force for re-orientation of compound 4 in the binding
site of Eg5 to achieve more stable binding mode.
Among diferent chemical compounds indicating the
cytotoxic efects on cancer cells, dihydropyrimidinones
(DHPM) are important heterocyclic compounds in medici-
nal chemistry due to their diverse pharmacological efects
[7, 8]. One of the most important and frst compounds
of dihydropyrimidones that proved to have the anticancer
efects is monastrol (Fig. 1), which specifcally prevents
metastasis through inhibiting the movement of kinesin
spindle protein known as Eg5 protein on microtubule [9].
It has been revealed that monastrol disrupts cell cycle
through allosteric inhibition of microtubule-stimulated
ADP release from Eg5 followed by apoptotic signaling
pathway leading to cell death.
Materials and methods
Eg5 is a mitotic kinesin that provides proper formation
of a bipolar mitotic spindle during mitosis, through ATP
hydrolysis [10], and as mentioned above, dihydropyrimi-
dine derivatives are foremost representative chemotypes for
inhibiting Eg5 that bind to the allosteric L5 binding site in
the motor domain, which is an ATP non-competitive site of
action [11, 12]. It should be also emphasized that one of the
advantageous features of dihydropyrimidinones with regard
to Eg5 is their selective binding pattern to this kinesin [13].
Besides dihydropyrimidinone derivatives, a variety of
structurally diverse Eg5 inhibitors such as thiazoles and
Chemistry
All materials and reagents were purchased from Sigma-
Aldrich (India) company and used without further purifca-
tion. Melting points were determined using an Electro ther-
mal type 9200 MP apparatus (England) and uncorrected.
Infra-Red (IR) spectra were obtained by a Perkin Elmer-
400 FT-IR spectrophotometer (England). Proton nuclear
magnetic resonance (1H-NMR) spectra were recorded on a
Bruker DRX400 spectrometer (400 MHz). Mass (MS) spec-
tra were reported by an Agilent 7890A spectrometer (USA).
1 3

Molecular Diversity
General procedure for the synthesis of N‑(6‑methylbenzo[d]
thiazole‑2‑yl)‑3‑oxo butane amide (1)
and N‑(thiazole‑2‑yl)‑3‑oxo butane amide (2)
to each well at a fnal concentration of 0.5 mg/mL, and the
plates were incubated for another 4 h at 37 °C in a dark
condition. Then, the resulted formazan crystals were solu-
bilized in 200 μL DMSO and optical density was measured
at 570 nm with background correction at 570 nm using a
Bio-Rad microplate reader (Model 680, USA). Control wells
contained no drugs, and blank wells contained only growth
medium for background correction. The percentage of cell
viability compared to the control wells was calculated for
each concentration of the compounds, and IC50 values were
calculated with SigmaPlot version 12.5. In all groups, the
absorbance of wells containing no cells (Blank) was sub-
tracted from the sample well absorbance.
8 mmol of 2-amino-6-methyl benzothiazole or 2-aminothia-
zole was dissolved in 10 mL xylene (148 °C) and 10 mmol
2,2,6-trimethyl-1,3-dioxin-4-one was added to the mixture.
After 2 h refux, the mixture was cooled to room tempera-
ture. Obtained precipitates were fltered and washed with
petroleum ether (3 × 2 mL). Subsequent recrystallization
from ethanol aforded the pure fnal product.
General procedure for the synthesis of 2‑amino‑6‑methylb‑
enzothiazole derivatives of 3,4‑dihydropyrimidinones (3–5)
and 2‑aminothiazole derivatives of 3,4‑dihydropyrimidi‑
nones (6–8)
Wound healing assay
To evaluate the cell migration property following treatment,
AGS cells (6×104) were seeded onto 24-well plates to grow
in a monolayer for 48 h. Then a sterile 100–200 μL pipette
tip was held vertically to scratch a cross in each well. The
detached cells were removed by washing with 500 μL PBS
and shaking at 500 rpm for 5 min. Then, the IC50 concen-
trations of the compounds were prepared by dilution with
serum-free medium. In the next step, the medium of the
wells was removed and the frst row of the plate was treated
as control and the next rows were treated with IC50 con-
centrations. After 24 h, photographs with Olympus CKX41
inverted microscope were taken using Cellsens software,
standard 1.14 connected to the DP27 camera [17, 18].
For the synthesis of DHPM derivatives, 3.9 mmol N-(6-
methylbenzo[d]thiazole-2-yl)-3-oxo butane amide or
N-(thiazole-2-yl)-3-oxo butane amide, 3.6 mmol urea, 1 mL
HCl, and 3 mmol corresponding aldehyde were mixed in
9 mL ethanol and refuxed for 24 h. After completion of
the reaction (controlled by TLC), a mixture of water and
ice cubes was added to the flask and stirred for 5 min.
Obtained precipitates were fltered and washed with cold
water (3×2 mL). Subsequent recrystallization from ethanol
aforded the pure fnal products (40.6–51.9%).
Biological assessment
Reagents and chemicals
In silico studies
Fetal bovine serum (FBS), RPMI 1640, trypsin, and phos-
phate-bufered saline (PBS) were all obtained from Biosera
(Ringmer, UK). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphe-
nyltetrazolium bromide or MTT was purchased from Sigma
(Saint Louis, MO, USA), and penicillin/streptomycin was
purchased from Invitrogen (San Diego, CA, USA). Cispl-
atin and dimethyl sulfoxide (DMSO) were obtained from
EBEWE Pharma (Unterach, Austria) and Merck (Darmstadt,
Germany), respectively.
Molecular docking
Lamarckian genetic algorithm (LGA) incorporated into
AutoDock4.2 package was used to run ligand fexible dock-
ing simulations on DHPM molecules [19, 20]. All the pre-
processing ligand and receptor steps and docking param-
eters were arranged according to the previous protocols
[21]. Post-docking analysis of ligand-enzyme interactions
was performed by protein–ligand interaction profler (PLIP)
fully automated server [22]. Intended 3D structure of kinesin
5 (Eg5) in complex with fuorastrol (PDB code 2×7e) was
retrieved from Brookhaven Protein Bank (PDB) in the X-ray
crystallographic format [23].
Evaluation of cytotoxicity by MTT assay
Cell viability following exposure to the synthetic compounds
was estimated via MTT reduction assay [16]. AGS cells were
plated in 96-well microplates at a density of 1×104 cells per
well in RPMI 1640 medium supplemented with 10% FBS
and 1% pen/strep for 24 h, and then, the culture medium
was changed to a serum-free medium containing the desired
volume of the synthetic compounds (which were frstly dis-
solved in DMSO and then diluted in medium). After 24 h
of culture, the medium was removed and MTT was added
Molecular dynamics
The all-atom MD simulations were performed using
GROMACS 5.1.1 computational package [24]. Topology
fles and other force feld parameters for DHPM molecules
were generated by PRODRG server [25]. Prior to energy
minimization, water molecules were represented using a
1 3

Molecular Diversity
simple point charge (SPC216) model and approximately
29,843 water molecules were added to the system. Overall
charge neutrality of the complex system (Eg5, ligand, and
water molecules) was ensured via insertion of appropriate
amounts of counter-ions (4 Na+ ions).
CH-phenyl), 5.54 (1H, brs, C4H-DHPM), 2.39 (3H, s,
6′-CH3-benzothiazole), 2.17 (3H, s, 6′-CH3-DHPM); MS
m/z (%): 394 (0/9) [M+], 189 (100),164 (23), 111 (64),
77 (9).
GROMOS96 43a1 was used as the force feld for deter-
mination of all bonding and nonbonding interactions. Steep-
est descent followed by conjugate gradient algorithms was
used for energy minimization of the protein–ligand complex.
After energy minimization process, position restraint pro-
cedure was performed in association with NVT and NPT
ensembles. An NVT ensemble was adopted at constant
temperature of 300 K with a coupling constant of 0.1 ps
and time duration of 500 ps. Subsequent to temperature sta-
bilization, NPT ensemble was performed in a way that a
constant pressure of 1 bar was employed with a coupling
constant of 5.0 ps and time duration of 1000 ps. Pressure
was maintained at 1 bar with the Parrinello-Rahman barostat
[26]. The lincs algorithm for covalent bond constraints was
applied [27]. The particle-mesh ewald (PME) and cut-of
methods were used to treat the long-range electrostatic and
van der Waals interactions, respectively [28]. All other bond-
ing parameters were set due to our previous report [29], and
20 ns MD simulation was performed with monitoring of
equilibration by examining the stability of the energy, tem-
perature, and the density of the system as well as the root
mean square deviations (RMSDs) of the backbone atoms.
4‑(3‑Bromophenyl)‑6‑methyl‑N‑(6‑methylbenzo [d]
thiazol‑2‑yl) ‑2‑oxo‑1,2,3,4‑tetrahydropyrimidine‑5‑carbox‑
amide (4)
White crystal, yield 49.6%; mp: 262–264 °C; IR (KBr) νmax
(cm−1): 3434.6 (N–H, amide), 3201.8 (N1-H, urea), 31980.3
(N3-H, urea), 1681.0 (C=O, amide), 1620.9 (C=O, urea),
1
693.4 (C–Br); H NMR (DMSO-d6) δ (ppm) 11.83 (1H,
brs, NH-amide), 9.16 (1H, brs, N1H-urea), 7.86 (1H, brs,
N3H-urea), 7.59 (1H, d, J=8.4 Hz, C4′H- benzothiazole),
7.46–7.48 (2H, m, CH-phenyl & C7′H-benzothiazole),
7.26–7.31 (3H, m, CH-phenyl & C5′H-benzothiazole),
7.01 (1H, d, J = 8.8 Hz, CH-phenyl), 5.56 (1H, brs, C4H-
DHPM), 2.50 (3H, s, 6′-CH3-benzothiazole), 2.20 (3H, s,
6′-CH3-DHPM); MS m/z (%): 250 (28), 206 (52), 165 (46),
135 (21), 111 (100).
4‑Phenyl‑6‑methyl‑N‑(6‑methylbenzo [d] thiazol‑2‑yl)
‑2‑oxo‑1,2,3,4‑tetrahydropyrimidine‑5‑carboxamide (5)
White crystal, yield 51.9%; mp: 268–271 °C; IR (KBr) νmax
(cm−1): 3367.7 (N–H, amide), 3224.5 (N1-H, urea), 3113.9
(N3-H, urea), 1695.8 (C=O, amide), 1605.7 (C=O, urea);
1H NMR (DMSO-d6) δ (ppm) 11.80 (1H, brs, NH-amide),
9.09 (1H, brs, N1H-urea), 7.96 (1H, brs, N3H-urea), 7.58
(1H, d, J = 8.4 Hz, C4′H- benzothiazole), 7.50 (1H, brs,
C7′H-benzothiazole), 7.28–7.35 (5H, m, CH-phenyl), 7.01
(1H, d, J=9.2 Hz, C5′H-benzothiazole), 5.59 (1H, brs, C4H-
DHPM), 2.50 (3H, s, 6′-CH3-benzothiazole), 2.18 (3H, s,
6′-CH3-DHPM); MS m/z (%): 185 (24), 165 (100), 137 (15),
110 (9), 44 (15).
Results and discussion
Chemistry
Chemical structures of DHPM derivatives were confrmed
by spectroscopic methods. Spectroscopic results of synthe-
sized compounds are illustrated below (Actual analytical
data are given in Online Resource through supplementary
material).
4‑(3‑Hydroxyphenyl)‑6‑methyl‑N‑(6‑methylbenzo [d]
thiazol‑2‑yl) ‑2‑oxo‑1,2,3,4‑tetrahydropyrimidine‑5‑carbox‑
amide (3)
4‑(4‑Nitrophenyl)‑6‑methyl‑N‑(thiazol‑2‑yl)‑2‑oxo‑1,2,3,4‑tet‑
rahydropyrimidine‑5‑carboxamide (6)
Pale yellow crystal, Yield 42.13%; mp: 268–270 °C; IR
(KBr) νmax (cm−1): 3343.3 (N–H, amide), 3236.7 (N1-H,
urea), 3127.3 (N3-H, urea), 1715.7 (C=O, amide), 1673.2
(C=O, urea), 1341.5 & 1537.6 (C–NO2 aromatic); 1H NMR
(DMSO-d6) δ (ppm) 11.79 (1H, brs, NH-amide), 9.15 (1H,
brs, N1H-urea), 8.22 (2H, d, J=8.8 Hz, CH-phenyl), 7.91
(1H, brs, N3H-urea), 7.53 (2H, d, J=8.8 Hz, CH-phenyl),
7.43 (1H, d, J=3.2 Hz, C4H-thiazole), 7.14 (1H, brs, C5H-
thiazole), 5.55 (1H, brs, C4H-DHPM), 2.16 (3H, s, 6′-CH3-
DHPM); MS m/z (%): 248 (53), 205 (34), 169 (100), 110
(11), 75 (4).
White crystal, yield 40.6%; mp: 290–292 °C; IR (KBr)
νmax (cm−1): 3357.1 (br, O–H phenol & N–H, amide),
3322.0 (N1-H, urea), 3209.4 (N3-H, urea), 1694.4 (C=O,
1
amide), 1622.9 (C=O, urea); H- NMR (400 MHz,
DMSO-d6) δ (ppm) 11.83 (1H, brs, NH-amide), 9.42 (1H,
brs, N1H-urea), 9.06 (1H, brs, OH-phenol), 7.75 (1H, brs,
N3H-urea), 7.71 (1H, brs, C7′H-benzothiazole), 7.59 (1H,
d, J = 8 Hz, C4′H-benzothiazole), 7.22 (1H, d, J = 7.2 Hz,
C5′H-benzothiazole), 7.09 (1H, t, J = 8 Hz, CH-phenyl),
6.70–6.72 (2H, m, CH-phenyl), 6.62 (1H, d, J = 6.8 Hz,
1 3

Molecular Diversity
4‑(4‑Fluorophenyl)‑6‑methyl‑N‑(thiazol‑2‑yl)‑2‑oxo‑1,2,3,4‑
tetrahydropyrimidine‑5‑carboxamide (7)
The frst step of the synthetic rout included a nucleophilic
attack of primary aromatic amine into dioxin reactant to
aford 3-oxo butane amide intermediate (1 & 2 in scheme 1).
Subsequent stage involved nucleophilic attack of urea to cor-
responding aldehyde and then reaction with 3-oxo butane
amide to give the fnal derivatives via cyclization and water
removal (Scheme 1).
Pale yellow crystal, Yield 47.5, mp: 170–171 °C; IR (KBr)
νmax (cm−1): 3398.2 (N–H, amide), 3226.7 (N1-H, urea),
3098.3 (N3-H, urea), 1704.0 (C=O, amide), 1667.6 (C=O,
1
urea), 1232.8 (C-F); H NMR (DMSO-d6) δ (ppm) 11.70
(1H, brs, NH-amide), 9.02 (1H, brs, N1H-urea), 7.74 (1H,
brs, N3H-urea), 7.43 (1H, d, J=3.2 Hz, C4H-thiazole), 7.31
(2H, d, J=5.6 Hz, CH-phenyl), 7.28 (2H, d, J=5.6 Hz, CH-
phenyl), 7.13 (1H, brs, C5H-thiazole), 5.54 (1H, brs, C4H-
DHPM), 2.14 (3H, s, 6′-CH3-DHPM); MS m/z (%): 288 (24),
205 (100), 127 (25), 100 (19), 58 (16).
Biological assessment
Cytotoxicity assessment
Prepared DHPM derivatives were assessed for their cyto-
toxic efects on AGS cell line in terms of IC50 values. As
shown in Table 1, compounds 7 with para-fuorophenyl moi-
ety exhibited higher toxicity (IC50 7.06 µg/ml) with regard
to other compounds and the cytotoxic efects of DHPMs
on AGS cell lines could be ordered as 7>6>8>5>4>3.
Results indicated that none of the assessed compounds could
be as cytotoxic as cis-platin against AGS cell lines except
compound 7. Dose-dependent diagrams for compounds 5
and 7 in terms of cell survival are depicted in Fig. 2. We
found that cell survival efect of 4 difered slightly from 3
only in low doses and not in high doses. Similar trend was
observed for compounds 6 and 8, but in the case of com-
pound 7, a curvature with steeper slope could be detected
after 12.5 µg/ml doses (Fig. 2).
4‑(4‑Chlorophenyl)‑6‑methyl‑N‑(thiazol‑2‑yl)‑2‑oxo‑1,2,3,4‑
tetrahydropyrimidine‑5‑carboxamide (8)
Pale yellow crystal, Yield: 51.5%; mp: 260–262 °C; IR
(KBr) νmax (cm−1): 3265.5 (N–H, amide), 3203.2 (N1-H,
urea), 3125.4 (N3-H, urea), 1709.2 (C=O, amide) 1683.9
(C=O, urea), 760.2 (C–Cl); 1H NMR (DMSO-d6) δ (ppm)
11.72 (1H, brs, NH-amide), 9.04 (1H, brs, N1H-urea), 7.77
(1H, brs, N3H-urea), 7.43 (1H, d, J=3.6 Hz, C4H-thiazole),
7.40 (2H, d, J=8.4 Hz, CH-phenyl), 7.28 (2H, d, J=7.4 Hz,
CH-phenyl), 7.14 (1H, brs, C5H-thiazole), 5.53 (1H, brs,
C4H-DHPM), 2.13 (3H, s, 6′-CH3-DHPM); MS m/z (%):
249 (48), 169 (100), 137 (50), 111 (70), 75(15).
Scheme 1 Synthetic route toward dihydropyrimidinones
1 3

Molecular Diversity
Table 1 IC50 values of DHPM derivatives vs assessed AGS cell line
Compound
X
IC50 (µg/ml) SD
Consensus
a
Log Po/w
3
4
5
6
7
8
3-OH
3-Br
H
4-NO2
4-F
4-Cl
–
42.73 1.35
38.34 1.43
21.70 1.19
19.60 0.76
7. 60 0.61
21.50 2.01
11.49 1.35
2.39
3.38
2.79
0.65
1.68
1.90
–
Cis-platin
aData were estimated via SwissADME webserver [31]
4-phenyl and N-phenyl 4-(3-chlorophenyl) thioxo deriva-
tives but weaker than N-phenyl 4-(3-bromophenyl), N-phe-
nyl 4-(3-fuorophenyl) and N-phenyl 4-(3-phenoxyphenyl)
thioxo derivatives against AGS cells [30]. Although not con-
frmed, cytotoxicity results might have some dependence on
lipophilicity and cell penetration. For instance, compounds
4 and 5, with higher lipophilicities (5.1 and 4.23), showed
higher cytotoxicities and fuoro substituent of compound 7
may enhance cellular uptake.
Cell migration inhibitory efect by wound healing assay
As shown in Fig. 3, 24-h post-culturing and treatment of
AGS cells with IC50 dose of compounds 3, 4 and 5, revealed
somewhat inhibitory efects on cell migration compared to
the control group (untreated cells) and this efect was greater
in the case of 4. However, migration in untreated AGS cells
was normal and they were able to largely cover the resulting
cell wound. It was interesting to note that compounds 6–8
exhibited the excitatory efect on AGS cell migration (Fig. 4)
and the efect could be prioritized as 7>6>8. The rate of
cell wound coverage within compounds 6 and 7 was rela-
tively higher than the control and for 8 was relatively simi-
lar to untreated cells even after 48 h. The efect of solvent
(DMSO) on wound healing in AGS cells was explored after
24-h culture, and it was revealed that migration in untreated
AGS cells (control) and DMSO-treated AGS cells was nor-
mal and they were able to largely cover the resulting cell
wound (Fig. 5).
Fig. 2 Percentage of AGS cell viability in diferent doses of com-
pounds 5 & 7
Comparison of the obtained cytotoxicity data with previ-
ous results on similar DHPM derivatives showed that these
compounds were better cytotoxic agents than N-phenyl
1 3

Molecular Diversity
Fig. 3 Efect of synthesized
compounds 3, 4 & 5 on in vitro
cell migration behavior of AGS
cells. After 24-h culture, AGS
cells started to migrate more
slowly than the control group.
The rate of wound healing in
4-treated group indicated to be
much slower than two other
compounds
Fig. 4 The efect of synthesized
compounds 6, 7 & 8 on in vitro
cell migration behavior of AGS
cells. After 24-h culture, in
the case of 6 and 7, AGS cells
started to migrate fast with
regard to the control and migra-
tion rate for 8 was relatively
similar to untreated cells
One possible explanation for observed cell migration
inhibitory efects would be an apparent diference among
the structures of 3–5 with 6–8 in a way that compounds 3–5
included fused benzene ring to the thiazole scafold, while
compounds 6–8 were only composed of thiazole rings. At
the frst glance, such observation directs us toward some-
thing like antagonist or agonist efect on a defnite cellu-
lar site of action for 3–5 and 6–8, respectively. Although
more extensive studies are required to pursue the trend, such
fndings may be useful in further molecular modifcations
with the aim of developing more potent inhibitors of cell
migration.
Molecular docking
Eg5 is known as a microtubule-dependent motor protein
and has been suggested to play an important role in cancer
development due to its critical function in the assembly
and maintenance of bipolar spindle [19]. Eg5 partici-
pates in spindle motions and chromosomes in cell divi-
sion and afects spindle function through cell localization
[32]. Recently, evidence has shown that Eg5 expression
Fig. 5 The efect of DMSO on wound healing in AGS cells in vitro.
After 24-h culture, migration in untreated AGS cells (control) and
DMSO-treated ones was normal and they were able to largely cover
the resulting cell wound
1 3

Molecular Diversity
Table 3 AutoDock 4.2 driven free binding energies for diferent opti-
cal isomers of DHPMs under study with Eg5 enzyme binding site
(PDB code: 2×7e)
is highly associated with diferent types of human can-
cer such as gastric adenocarcinoma [17, 33, 34] and Eg5
inhibitors also show antitumor activity. Studies have
shown that Eg5 protein expression was signifcantly asso-
ciated with metastasis and TNM stage. Thus, Eg5 is a suit-
able target for new anticancer therapies in certain types of
cancer [35].
Ligand
GA Runs
Top-ranked conforma- ΔGb (kcal/mol)
molecule
tional population
R-3
S-3
R-4
S-4
R-5
S-5
R-6
S-6
R-7
S-7
R-8
S-8
50
50
50
50
50
50
50
50
50
50
50
50
8
−8.04
−9.50
−9.14
−9.52
−8.22
−8.78
−6.88
−7.06
−7.19
−7.44
−7.48
−8.12
30
34
35
40
17
21
50
50
50
40
24
In the light of above explanations, we decided to
explore the potential binding interactions of assessed
compounds within L5/α2/α3 allosteric site of Eg5 as a
validated microtubule-dependent target for DHPMs. In
order to explore the structural validation of predicted
docked poses, initial step was dedicated to the estimation
of root mean square deviation (RMSD) of the Cartesian
coordinates of re-docked co-crystallographic ligand atoms
as a validation criterion. On the basis of re-docking results
(Table 2), adaptable predictability levels (≤ 2 Å) could
be achieved with 50 independent GA runs and 2.5 × 107
maximum number of evaluations for PDB accession codes
2X7E and 2IEH, but the former was selected to run the
simulations for lower value of RMSD.
The bold value represents the highest achieved docked Gibbs free
energy
(2) Binding of top-ranked derivatives (3–5) could be char-
acterized by dominant hydrophobic pattern (Fig. 6 and
Table 4). Several hydrophobic interactions exhibited
possible antagonistic behavior in the binding site of
Eg5. The issue was reported previously for dihydropy-
rimidinethione core of monastrol with Gly117, Ile136,
Pro137, Tyr211, Leu214, and Ala218 [36].
A validated simulation protocol was applied to dock
compounds 3–8 into the L5/α2/α3 allosteric site of Eg5
with the aim of selecting top binders. To explain more, L5/
α2/α3 allosteric site is a known Eg5 site targeted by inhibi-
tors [36]. Results showed that R-confgured compound 4
had the highest binding free energy (ΔGb − 9.52 kcal/mol)
(Table 3). To observe the involved Eg5 residues, 3D sche-
matic representation of binding interactions is summarized
in Fig. 6 and Table 4.
(3) Structure binding relationship indicated the importance
of fused benzene ring in tighter binding of compounds
3–5 to Asp896, Tyr211 and Ala133.
Results of docking study with regard to binding of
DHPMs into Eg5 allosteric site may indicate that:
(4) Obtained binding data were relatively correlated with
cell migration inhibitory results. Compounds 6–8
exhibited lower binding energies to Eg5 binding site
(ΔGb − 6.88 to − 8.12 kcal/mol) and fastened the
rate of cell wound coverage with regard to untreated
(1) S-enantiomers of DHPMs showed higher binding afn-
ity to Eg5. Monastrol has also been reported to interact
to Eg5 in the S-confgured form [37].
Table 2 AutoDock4.2 validation results for diferent 3D holo Eg5-ligand complexes retrieved from PDB
No.
PDB code
Resolution (Ǻ)
Co-crystallographic ligand
Top-ranked popula-
tion (out of 50)
ΔGb (kcal/mol)
RMSD from
reference (Å)
1
2X7E
2.40
50
−9.60
1.02
2
2IEH
2.70
49
−10.16
1.79
1 3

Molecular Diversity
Fig. 6 3D representation of
chemical interactions for S iso-
mer of compound 4 in binding
to Eg5 allosteric binding site
(PDB code: 2×7e)
Table 4 Hydrophobic/H-bond interactions of S isomer of compound
4 in binding to the Eg5 enzyme bonding site (PDB code: 2×7e)
Molecular dynamics
MD simulation is an efcient strategy to improve docking
models since fexibility of both ligand and macromolecule
are considered during a reasonable run time [38]. To evalu-
ate the stability of predicted Eg5 complex with regard to the
dynamic characteristics of the protein, 50 ns MD simula-
tion of Eg5 was conducted in the presence of top-ranked
docked ligand S-4 in explicit water. The stability of Eg5
during simulation was assessed via monitoring root mean
square deviation (RMSD), root mean square fuctuations
(RMSF) and radius of gyration (Rg). Moreover; to compare
the MD results with the reference compound, MD of the
co-crystallographic R-fuorastrol and Eg5 complex was also
performed.
Index Residue AA Distance Interacted
Interacted protein
ligand atom/ atom
No
Approximate leveling-of in deviation from primary Eg5
structure was gradually achieved after 10 ns of simulation
indicating that the system folded to a more stable conforma-
tion with regard to the starting structure (Fig. 7). The low
RMSD value of the complex revealed the stable binding of
S-4 to Eg5 and indicated the convergence of Eg5 to equi-
librium structure (Average RMSD 2.75 Å). Besides protein
backbone RMSD, the stability of docked complex must also
be inferred in terms of ligand fuctuations. For this purpose,
the RMSD of all atoms of compound S-4 was estimated and
is shown in Fig. 7. It was found that the RMSD profle of the
interacted ligand is probably stable with regard to the formed
complex (Average RMSD 1.59 Å).
1
2
3
4
130A
133A
136A
211A
ASP 3.00
ALA 3.47
ILE 3.04
TYR 3.59
CH3 (19)
C (13)
CH3 (7)
C (16)
Side chain CH2
Side chain CH3
Side chain CH
Side chain Phenol
CH
5
w
218A
239A
ALA 3.17
PHE 3.59
C(2ʹ)
CH3 (7)
Side chain CH3
Side chain Phenol
CH
7
218A
ALA 2.05
N–H (3)
Backbone N
AGS cells. The case for compounds 3–5 (ΔGb − 8.04
to − 9.52 kcal/mol) was vice versa. However, further
biological assessments are required to confrm the pro-
posed relation of treated-cell migration and Eg5 bind-
ing ability.
For comparison purposes, RMSD plot for a co-crystallo-
graphic ligand R-fuorastrol and Eg5 complex during 50 ns
was also obtained (Fig. 8). The good RMSD value of the ref-
erence complex structure (Average RMSD 2.77 Å). revealed
the stable binding of the cognate ligand to Eg5 binding site
for 50 ns, and moreover the average RMSD variations of the
cognate ligand were found to be 1.86 Å. Relatively similar
1 3

Molecular Diversity
Fig. 9 RMSF of the backbone atoms of Eg5 in complex with S-4
(blue) and R-fuorastrol (orange) during MD simulations (0–50 ns)
Fig. 7 RMSDs of the backbone atoms of Eg5 (up) and all atoms of
compound S-4 (down) during MD simulations (0–50 ns)
Fig. 10 Time dependence of the radius of gyration (Rg) graph of Eg5
in complex with S-4 (blue) and R-fuorastrol (orange) during MD
simulations (0–50 ns)
Fig. 8 RMSDs of the backbone atoms of Eg5 (up) and all atoms of
a co-crystallographic ligand R-fuorastrol (down) during MD simula-
tions (0–50 ns)
(Fig. 9). Conformational changes of the interacted resi-
dues were all very small, and main fuctuations (>0.3 nm)
occurred to residues far from ligand binding site. To explain
more, RMSF variations for co-crystallographic and docked
complexes were found to be 0.15 and 0.14 nm, respectively.
Ile250 (RMSF 0.74 nm) of co-crystallographic complex and
Glu253 (RMSF 0.37 nm) of docked complex were the only
amino acid residues with signifcant RMSF values. In the
case of Ser75, Pr088, Ile89, Asp91, Gly96 and Tyr97 the
docked RMDF was signifcantly more than a corresponding
residue in co-crystallographic complex. The narrow range
of estimated RMSFs for binding site residues confrmed the
RMSD results and demonstrated the capability of S-4 to
form stable interactions with Eg5 during MD simulation.
The compactness of Eg5 was evaluated by radius of gyra-
tion (Rg) (Fig. 10). It was found that residual backbones
and folding of Eg5 were consistently stable after binding
to S-4 and R-fuorastrol. The Rg of the backbone atoms of
Eg5 was decreased from about 2.05 nm at the beginning to
trend could be observed for reference macromolecule struc-
ture since at about 10 ns, but after a less steep slope the
equilibrium state could be achieved. Although the whole
RMSD variation for R-fuorastrol was desirable, after about
15 ns a very stable binding trajectories were attained p to the
end of MD simulations which is indicative of a stable bind-
ing mode toward Eg5 allosteric site. In the case of docked
ligand, although average RMSD during 50 ns was even less
than the cognate ligand and very desirable, particularly ini-
tial 10 ns of the MD simulation was accompanied by more
conformational fuctuations. One of the interesting outcomes
of RMSD plots was appropriate compatibility of correspond-
ing conformational fuctuations between ligand and protein
within both RMSD plots.
Average atomic mobility of the Eg5 Cα atoms is indica-
tive of the fexibility of individual residues within the MD
simulations and can be estimated through RMSF calculation
1 3

Molecular Diversity
about 1.94 nm after 12.5 ns in the presence of R-fuorastrol
and 1.92 nm after 30 ns in the presence of S-4. Moreover,
Rg plot showed that the decrease in Rg was associated with
steeper slope during the initial 3 ns of simulation. Results
indicated that compound S-4 was able to potentially induce
desirable compact structure of Eg5 (Fig. 10).
to converge, representing that the systems evolved into sta-
ble and equilibrated states. It was interesting to note that
improved polar interactions were probably driving force
for re-orientation of compound 4 in the binding site of Eg5
to achieve more stable binding mode. Results of this study
prompt us to perform further structural modifcation of the
assessed compounds and achieve DHPM derivatives that
possess superior efects in preventing the migration of AGS
cancer cells.
To explore the stability of binding interactions, it was
decided to achieve the complex structure after 50 ns and
as is depicted in Fig. 11, Ala218 was the sole residue with
conserved H-bond interaction via backbone N and N3H
of ligand. Conformational fuctuation of the system dur-
ing 50 ns provided a new hydrogen bond between Leu214
backbone N and N3-H of ligand. In this regard, the HBD
group N3H of the DHPM ring seemed to be an important
site of action for this compound. A few amino acid resi-
dues changed their attitude toward ligand and participated
in hydrophobic contacts to the carbon skeleton of the docked
complex. Besides H-bond interaction, Ala218 also made
a hydrophobic contact via side chain methyl to the ortho
carbon atom of bromophenyl moiety. Tyr211, Phe239 and
Ala133 could not be detected for the stable complex after
50 ns MD simulation.
Supplementary material
The spectra (actual analytical data) of the newly synthesized
compounds are accessible in Supplementary Materials.
Supplementary Information The online version contains supplemen-
tary material available at https://doi.org/10.1007/s11030-021-10229-z.
Acknowledgements The research leading to ofered results received
funding from Ardabil University of Medical Sciences (ARUMS) under
Grant No. IR.ARUMS.REC.1398.259.
Authors’ contributions All authors contributed equally in this work
(concept, design, defnition of intellectual content, literature search,
experimental studies, data acquisition, data analysis, statistical analysis,
manuscript preparation, manuscript editing and manuscript review).
Conclusion
DHPM derivative (4) with 4-(3-bromo phenyl) substituent
showed the greatest inhibitory efect of AGS cell migration,
probably due to its bulkier nature and higher hydrophobicity
(Table 1). In contrast to 4, DHPMs that possessed thiazole
but not benzothiazole ring exhibited no inhibitory efect on
in vitro cell migration of AGS cells indicating a determinant
role of phenyl ring of benzothiazole in detected inhibitory
efects of N-(benzothiazol-2-yl)-5-carboxamide derivatives.
MD simulations of top-ranked docked pose of compound
4 showed that RMSD of ligand-receptor complex inclined
Funding The research leading to ofered results received funding from
Ardabil University of Medical Sciences (ARUMS) under Grant No.
IR.ARUMS.REC.1397.137.
Availability of data and material All the data are available and can be
ofered on request.
Code availability Not applicable.
Declaration
Conflict of interest Authors declare to have no conficts of interests.
References
1. Pecorino L (2016) Molecular biology of cancer: mechanisms,
targets, and therapeutics. Oxford University Press, New York
2. Mons U, Gredner T, Behrens G, Stock C, Brenner H (2018) Can-
cers due to smoking and high alcohol consumption. Dtsch Arztebl
Int 115(35–36):571–577
3. Hartgrink HH, Jansen EP, van Grieken NC, van de Velde CJ
(2009) Gastric cancer. Lancet 374(9688):477–490
4. Jiang F, Shen X (2019) Current prevalence status of gastric cancer
and recent studies on the roles of circular RNAs and methods used
to investigate circular RNAs. Cell Mol Biol Lett 24:53
5. Seyfried TN, Huysentruyt LC (2013) On the origin of cancer
metastasis. Crit Rev Oncog 18(1–2):43–73
Fig. 11 3D representation of chemical interactions for S isomer of
compound 4 in binding to Eg5 allosteric binding site (PDB code:
2×7e) after 50-ns MD simulations
1 3

Molecular Diversity
23. Kaan HYK, Ulaganathan V, Rath O, Prokopcova H, Dallinger
D, Kappe CO, Kozielski F (2010) Structural basis for inhibition
of Eg5 by dihydropyrimidines: stereoselectivity of antimitotic
inhibitors enastron, dimethylenastron and fuorastrol. J Med Chem
53:5676–5683
6. Kobayashi H, Enomoto A, Woods SL, Burt AD, Takahashi M,
Worthley DL (2019) Cancer-associated fbroblasts in gastrointes-
tinal cancer. Nat Rev Gastroenter Hepatol 16(5):282–295
7. Alberts S, Cervantes A, Van de Velde C (2003) Gastric cancer:
epidemiology, pathology and treatment. Ann Onc 14:ii31–ii36
8. Wang F, Li T, Zhang B, Li H, Wu Q, Yang L, Nie Y, Wu K, Shi Y,
Fan D (2013) MicroRNA-19a/b regulates multidrug resistance in
human gastric cancer cells by targeting PTEN. Biochem Biophys
Res Com 434(3):688–694
24. Spoel VDD, Lindahl E, Hess B, Groenhof G, Mark AE, Berendsen
HJ (2005) GROMACS: fast, fexible, and free. J Comput Chem
26:1701–1718
25. Schuttelkopf AW, Aalten DMF (2004) PRODRG: a tool for high-
throughput crystallography of protein-ligand complexes. Acta
Crystallogr D Biol Crystallogr 60(8):1355–1356
9. Bidram Z, Sirous H, Khodarahmi GA, Hassanzadeh F, Dana N,
Hariri AA, Rostami M (2020) Monastrol derivatives: in silico and
in vitro cytotoxicity assessments. Res Pharm Sci 15(3):249–262
10. Sawin KE, LeGuellec K, Philippe M, Mitchison TJ (1992) Mitotic
spindle organization by a plus-end-directed microtubule motor.
Nature 359(6395):540–543
26. Parrinello M, Rahman A (1981) Polymorphic transitions in sin-
gle crystals: A new molecular dynamics method. J Appl Phys
52(12):7182–7190
27. Hess B, Bekker H, Berendsen HJC, Fraaije J (1997) LINCS: a
linear constraint solver for molecular simulations. J Comput Chem
18:1463–1472
11. Pérez-Melero C (2014) KSP inhibitors as antimitotic agents. Curr
Top Med Chem 14(20):2286–2311
28. Darden T, York D, Pedersen L (1993) Particle mesh Ewald: An
N log (N) method for Ewald sums in large systems. J Chem Phys
98(12):10089–10092
12. Luo L, Carson JD, Dhanak D, Jackson JR, Huang PS, Lee Y,
Sakowicz R, Copeland RA (2004) Mechanism of inhibition of
human KSP by monastrol: insights from kinetic analysis and
the effect of ionic strength on KSP inhibition. Biochemistry
43(48):15258–15266
29. Razzaghi-Asl N, Mirzayi S, Mahnam K, Sepehri S (2018) Iden-
tifcation of COX-2 inhibitors via structure-based virtual screen-
ing and molecular dynamics simulation. J Mol Graph Model
83:138–152
13. González-Hernández E, Aparicio R, Garayoa M, Montero MJ,
Sevilla MÁ, Pérez-Melero C (2019) Dihydropyrimidine-2-thiones
as Eg5 inhibitors and L-type calcium channel blockers: potential
antitumour dual agents. MedChemComm 10(9):1589–1598
14. Nakai R, Iida S, Takahashi T, Tsujita T, Okamoto S, Takada C,
Akasaka K, Ichikawa S, Ishida H, Kusaka H, Akinaga S, Murakata
C, Honda S, Nitta M, Saya H, Yamashita Y (2009) K858, a novel
inhibitor of mitotic kinesin Eg5 and antitumor agent, induces cell
death in cancer cells. Cancer Res 69(9):3901–3909
30. Bohlooli S, Nejatkhah N, Sepehri S, Doostkamel D, Razzaghi-Asl
N (2020) Synthesis and cytotoxicity evaluation of novel cyclic/
noncyclic N-heteroaryl enamino amides vs human cancer cell
lines. Res Pharm Sci 15(6):563–570
31. Daina A, Michielin O, Zoete V (2017) SwissADME: a free web
tool to evaluate pharmacokinetics, drug-likeness and medicinal
chemistry friendliness of small molecules. Sci Rep 7:42717
32. Kapitein LC, Peterman EJG, Kwok BH, Kim JH, Kapoor TM,
Schmidt CF (2005) The bipolar mitotic kinesin Eg5 moves on
both microtubules that it crosslinks. Nature 435(7038):114–118
33. Sakowicz R, Finer JT, Beraud C, Crompton A, Lewis E, Fritsch
A, Lee Y, Mak J, Moody R, Turincio R, Chabala JC, Gonzales
P, Roth S, Weitman S, Wood KW (2004) Antitumor activity of a
kinesin inhibitor. Cancer Res 64(9):3276–3280
15. Khathi SP, Chandrasekaran B, Karunanidhi S, Tham CL, Koz-
ielski F, Sayyad N, Karpoormath R (2018) Design and synthe-
sis of novel thiadiazole-thiazolone hybrids as potential inhibi-
tors of the human mitotic kinesin Eg5. Bioorg Med Chem Lett
28(17):2930–2938
16. Van Meerloo J, Kaspers GJ, Cloos J (2011) Cell sensitivity assays:
the MTT assay. In: Cancer cell culture, Humana Press
34. Imai T, Oue N, Nishioka M, Mukai S, Oshima T, Sakamoto N,
Sentani K, Matsusaki K, Yoshida K, Yasui W (2017) Overexpres-
sion of KIF11 in gastric cancer with intestinal mucin phenotype.
Pathobiology 84:16–24
17. Sun XD, Shi XJ, Sun XO, Luo YG, Wu XJ, Yao CF, Yu HY, Li
DW, Liu M, Zhou J (2011) Dimethylenastron suppresses human
pancreatic cancer cell migration and invasion in vitro via allosteric
inhibition of mitotic kinesin Eg5. Acta Pharm Sin 32(12):1543
18. Wang X, Decker CC, Zechner L, Krstin S, Wink M (2019) In vitro
wound healing of tumor cells: inhibition of cell migration by
selected cytotoxic alkaloids. BMC Pharm Toxic 20(1):1–2
19. Morris G, Huey R, Lindstrom W, Sanner MF, Belew RK, Good-
sell DS, Olson AJ (2009) AutoDock4 and AutoDockTools4: auto-
mated docking with selective receptor fexibility. J Comput Chem
30(16):2785–2791
35. Lu M, Zhu H, Wang X, Zhang D, Xiong L, Xu L, You Y (2016)
The prognostic role of Eg5 expression in laryngeal squamous cell
carcinoma. Pathology 48(3):214–218
36. Myers SM, Collins I (2016) Recent fndings and future directions
for interpolar mitotic kinesin inhibitors in cancer therapy. Future
Med Chem 8(4):463–489
37. Garcia-Saez I, DeBonis S, Lopez R, Trucco F, Rousseau B,
Thuery P, Kozielski F (2007) Structure of human Eg5 in complex
with a new monastrol-based inhibitor bound in the R confgura-
tion. J Biol Chem 282:9740–9747
20. Sanner M (1999) Python: a programming language for software
integration and development. J Mol Graph Model 17(1):57–61
21. Razzaghi-Asl N, Ebadi A, Shahabipour S, Gholamin D (2020)
Identifcation of a potential SARS-CoV2 inhibitor via molecular
dynamics simulations and amino acid decomposition analysis. J
Biomol Struct Dyn. https://doi.org/10.1080/07391102.2020.17975
36
38. Durrant JD, McCammon JA (2011) Molecular dynamics simula-
tions and drug discovery. BMC Biol 9:71–79
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional afliations.
22. Salentin S, Schreiber S, Haupt VJ, Adasme MF, Schroeder M
(2015) PLIP: Fully automated protein-ligand interaction profler.
Nucleic Acids Res 43(W1):W443–W447
1 3

Molecular Diversity
Authors and Afliations
Mohsen Sagha1 · Fatemeh Mousaei2 · Mahtab Salahi2 · Nima Razzaghi‑Asl3
2
* Nima Razzaghi-Asl
Student Research Committee, School of Pharmacy, Ardabil
n.razzaghi@pharmacy.arums.ac.ir
University of Medical Sciences, Ardabil, Iran
3
Department of Medicinal Chemistry, School
of Pharmacy, Ardabil University of Medical Sciences,
5618953141 Ardabil, Iran
1
Research Laboratory for Embryology and Stem Cells,
Department of Anatomical Sciences, School of Medicine,
Ardabil University of Medical Sciences, Ardabil, Iran
1 3