Synthesis, antimicrobial activity, molecular docking and ADMET study of a caprolactam-glycine cluster

Article abstract of DOI:10.1080/07391102.2020.1748112

Density functional theory calculations were performed with DFT method using both b3lyp/6-311++G(d,p) and wb97xd/6-311++G(d,p) levels of theory to predict the molecular geometry, to evaluate the molecular electrostatic potential and frontier molecular orbitals of synthesized a new compound: caprolactam-glysine cluster (CL-Gly). Molecular docking study of the CL-Gly was carried out to clarify the interaction and the probable binding modes, between the title compound and DNA. The antibacterial activities of CL-Gly cluster against Gram-positive and Gram-negative bacteria was determined. In silico ADMET study was also performed for predicting pharmacokinetic and toxicity profile of the synthesized cluster which expressed good drug-like behavior and non-toxic nature. It was revealed that the compound has importance in drug discovery process. Communicated by Ramaswamy H. Sarma.

Full text of DOI:10.1080/07391102.2020.1748112

Journal of Biomolecular Structure and Dynamics
ISSN: 0739-1102 (Print) 1538-0254 (Online) Journal homepage: https://www.tandfonline.com/loi/tbsd20
Synthesis, antimicrobial activity, molecular
docking and ADMET study of a caprolactam-
glycine cluster
Sefa Celik, Ali Tugrul Albayrak, Sevim Akyuz, Aysen E. Ozel & Belgi Diren
Sigirci
To cite this article: Sefa Celik, Ali Tugrul Albayrak, Sevim Akyuz, Aysen E. Ozel & Belgi
Diren Sigirci (2020): Synthesis, antimicrobial activity, molecular docking and ADMET study
of a caprolactam-glycine cluster, Journal of Biomolecular Structure and Dynamics, DOI:
10.1080/07391102.2020.1748112
To link to this article: https://doi.org/10.1080/07391102.2020.1748112
View supplementary material
Accepted author version posted online: 27
Mar 2020.
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=tbsd20

Synthesis, antimicrobial activity, molecular docking and ADMET study of a
caprolactam-glycine cluster
Sefa Celika*, Ali Tugrul Albayrakb, Sevim Akyuzc, Aysen E. Ozela and Belgi Diren Sigircid
aPhysics Department, Science Faculty, Istanbul University, Vezneciler, 34134, Istanbul, Turkey;
bChemical Engineering Department, Engineering Faculty, Istanbul University-Cerrahpasa,
Istanbul, Turkey;
cPhysics Department, Science and Letters Faculty, Istanbul Kultur University, Atakoy Campus,
Istanbul, Turkey;
d
Faculty of Veterinary Medicine, Department of Microbiology, Istanbul University-Cerrahpasa,
Buyukcekmece Campus, Istanbul, Turkey.
*Corresponding author email: scelik@istanbul.edu.tr
ABSTRACT
Density functional theory calculations were performed with DFT method using both b3lyp/ 6-
311++G(d,p) and wb97xd/6-311++G(d,p) levels of theory to predict the molecular geometry, to
evaluate the molecular electrostatic potential and frontier molecular orbitals of synthesized a new
compound: caprolactam-glysine cluster (CL-Gly). Molecular docking study of the CL-Gly was
carried out to clarify the interaction and the probable binding modes, between the title compound
and DNA. The antibacterial activities of CL-Gly cluster against Gram-positive and Gram-
negative bacteria was determined. In silico ADMET study was also performed for predicting
pharmacokinetic and toxicity profile of the synthesized cluster which expressed good drug-like

behavior and non-toxic nature. It was revealed that the compound has importance in drug
discovery process.
Key words: Antimicrobial activity; Caprolactam; DFT calculations; Glycine; Molecular
Docking
1. Introduction
Glycine (Gly) is one of the simplest amino acid in the white crystalline solid form, containing a
carboxyl and an amine group and having a melting point of 290 °C (Oxtoby et al., 2011;
Atkinson et al., 2001). Gly also has antibacterial activity. For example, in a study (Minami et al.,
2004) investigating the effect of glycine on Helicobacter pylori bacteria that cause gastritis and
ulcers in the stomach (Roesler, 2016), glycine has been shown to have a stronger (synergistic)
antibacterial activity on bacteria if used in combination with antibiotics. In another study (Sepahi
et al., 2017), glycine alone or in combination with poly-1-arginine (PLA) has been reported to
exhibit toxicity against E. coli O157: H7 and Staphylococcus aureus bacteria. In a patent
(Bontenbal et al., 2006) on the biological activity of glycine, it is reported that glycine alone,
glycine salts and glycine esters show antibacterial activity against harmful pathogens in foods
and beverages. Mahmoud et al.(2015) prepared a series of ternary complexes composed of
glycine, anti-inflammatory drug lornoxicam and transition metal chlorides and it turned out that
all the complexes except Cr (III) complex showed toxic effects against breast carcinomas (MCF7
cell line). It has been found that glycine-rich proteins available in sea hare eggs have an
inhibitory effect against U937 leukemia cell line (Lee et al., 2016). Besides, studies of biological
activity concerning glycine derivatives have been carried out. In a study (Cosquer et al., 2004) in
which glycine betaine analogs were synthesized, it was found that these analogs inhibited

bacterial growth. In another study (Shneine et al., 2017) in which glycine derivatives were
synthesized and their antibacterial effects were examined, glycine derivative (2S)-2-N-[(benzoyl-
pyridin-2-yl-amino)-(4-Chloro-phenyl)-methyl]-L-alanine was found to be the most toxic against
the gram-positive bacteria Staphylococcus aureus and the gram-negative bacteria Klebsiella
pneumonia. In a study (Ajloo et al., 2015) in which palladium complexes of phenanthroline and
glycine derivatives were synthesized, it is revealed that Pd (II) complexes bind to DNA by
intercalation, consequently leading an increase in DNA's thermal stability and that the Pd (II)
complex of methyl glycine exhibits a high toxicity against K562 leukemia cell line.
Though there are few studies on the physical properties (Wu et al., 2011) and use (Wu et al.,
2011; Nimah et al., 2015) of glycine-based ionic liquids, no study on the biological activities of
these ionic liquids is available in the literature.
There are not many studies on the biological activity of caprolactam which is the other starting
material used to synthesize cluster CL-Gly. In a few studies (Iizuka et al., 1967; Baxi, 2013), it
has been notified that growth of some bacteria is inhibited in the presence of ε-caprolactam,
whereas some bacteria use ε-caprolactam as a single source of carbon and nitrogen.
In this study; CL-Gly was synthesized and in order to understand its structure–activity
relationship, the geometry optimization and molecular electrostatic potential and frontier
molecular orbitals calculations were carried out using quantum-chemical calculations with
density functional theory (DFT). Moreover, the antibacterial effects of the starting materials
(caprolactam and glycine) and the cluster were investigated comparatively.
2. Experimental and computational details
2.1. Synthesis

ε-Caprolactam (purity 99 wt%) and glycine (purity 99 wt%) were purchased from Aldrich and
Merck, respectively.
ε-caprolactam (CL) - glycine (Gly) cluster (3) was prepared via a similar method to that (Moriel
et al., 2010) in the literature: 149 mmol (Gly) (2) was added to an aqueous solution containing
12.4% by weight ε-caprolactam (123 mmol CL) (1) and the mixture was stirred at room
temperature for 24 hours.
After removing water at 60 °C under vacuum, 120 ml of acetonitrile and 40 ml of methanol were
added to precipitate the unreacted amino acid, and the mixture was stirred vigorously overnight
and then filtered. The purified cluster (3) was dried overnight at 60 °C and stored under
moisture-free conditions. The formation reaction of the cluster is shown in Figure 1 (Celik et al.,
2019).
Fig. 1. The reaction scheme of formation of CL-Gly cluster
The purity of the compound was checked by IR and Raman analysis. The ATR-FTIR and micro-
Raman spectra of the synthesized cluster CL-Gly is given in comparison to those of solid
caprolactam and Glycine in Figures 2 and 3, respectively. As seen in the figures, although CL-

Gly cluster has CL and Gly vibrational bands, band wavenumber shifts were observed that
indicative of CL and Gly interaction
Fig. 2. The ATR-FTIR spectra of Glycine (a), solid caprolactam (b) and CL-Gly (c).

Fig. 3. Micro-Raman spectra of Glycine (a), solid caprolactam (b) and CL-Gly (c). The enlarged
1700-1550 cm-1 region of the spectra is given in the upper frame.
2.2. Computational methods
The entire calculations in the present work are performed using the Gaussian 03W (Frisch et al.,
2004) program package on personal computer. The geometrical parameters are computed by
optimizing the geometry of the molecule using DFT (Becke, 1993) method both b3lyp and
wb97xd functionals with the 6-311++G(d,p) basis set. The both b3lyp and wb97xd functionals
are used comparatively since the wb97xd (Chai et al., 2008a; 2008b) takes into account both
short-range and long-range interactions, while b3lyp only takes into account short-range

interactions. The optimized structures of CL-CL and CL-Gly and hydrogen bond lengths,
calculated using DFT/b3lyp/6-311++G(d,p) and DFT/wb97xd/6-311++G(d,p) level of theories
are given in Figure 4.
(I)
(II)
(III)
Fig.4. The optimized molecular structures of CL-Gly (I-II) and CL-CL(III) calculated using
DFT/b3lyp/6-311++G(d,p) (I) and DFT/wb97xd/6-311++G(d,p) (II-III) level of theories.
The harmonic force field of the cluster was evaluated via scaled quantum mechanical force field
suggested by Pulay, Fogarasi, Pongor, Boggs, and Vargha (1983). Potential Energy Distributions
of the cluster were computed by MOLVIB program(Sundius, 1990, 2002). Computed harmonic
wavenumbers under 1800 cm−1 were multiplied by 0.98 and wavenumbers over 1800 cm−1 were
multiplied by 0.96 (Balci et al., 2005). Calculated and scaled wavenumbers of CL-CL and CL-

Gly were given in Table S1 in comparison with the experimental ATR-FTIR and Raman spectra
of CL-Gly cluster.
2.3. Antibacterial activity
Antibacterial activity of the glycine, caprolactam and caprolactam-glycine were examined by
disc diffusion method and agar dilution method according to clinical and laboratory standards
institute (formerly CLSI) (CLSI, 2012).
The antibacterial activities were evaluated against Gram-positive (Staphylococcus aureus ATCC
29213, Staphylococcus epidermidis ATCC 12228, Enterococcus faecalis ATCC 29212, Bacillus
cereus ATCC 11778, Bacillus subtilis ATCC 6633), Gram-negative (Escherichia coli ATCC
25922, Klebsiella pneumoniae ATCC 4352, Pseudomonas aeruginosa ATCC 27853, Salmonella
enteritidis KUEN 349, Salmonella Typhimurium ATCC 14028) bacteria. The strains were
provided by the Faculty of Veterinary Medicine, Department of Microbiology Culture
Collection, Istanbul University-Cerrahpasa. Mueller–Hinton Agar (Fluka 70191) was used for
the detection of the qualitative and quantitative antibacterial effect and to maintain the strains.
Mueller–Hinton broth (Fluka 90922) (CAMBH) with MgCl2·2H2O (10 mg Mg2+/L) and
CaCl2·6H2O (20 mg Ca2+/L) was used as the medium for dilution. Glycine, caprolactam and
caprolactam-glycine were dissolved in DMSO at a concentration of 300mg/ml, 500mg/ml and
300 mg/ml respectively.

Disc diffusion method was used for screening the samples for the qualitative measurement of
antibacterial activity. The Muller Hinton Agar plates were inoculated with 0.1 mL of 0,5
McFarland dilution of the tested culture. The discs (6 mm in diameter) were impregnated with
50 μl of the Gly, CL, CL-Gly and placed on the inoculated agar. The plates were incubated at
37°C for 24 h. The plates were examined for possible clear zones after incubation. The presence
of any clear zone that formed around the film on the plate medium was recorded as the growth
inhibition zone against the microbial species. Gentamycin standard disc (10µg) was placed onto
agar plates for positive control. The tests were duplicated, and data were averaged.
For the detection of the antibacterial effect of the glycine, caprolactam and caprolactam-glycine
quantitatively, the agar dilution method was performed. The components were prepared for two-
fold step dilution for ten serial dilutions with CAMHB. 1 ml of each inoculum was poured to
each petri dish, and 9 ml Muller-Hinton agar brought to 45-50 ° C was added onto inoculum and
mixed with a circular dial until at the room temperature. A bacterial suspension with 107
CFU/ml final concentration was prepared and was inserted into the microplate wells. The
sterilised replicator with 3-mm pins, which deliver two μl, was placed into the microplate to soak
the pins and transfer it onto the agar plate. The agars were incubated at 37 °C for 24 h. The
minimum inhibitory concentration, MIC, the value was determined beyond the level no
inhibition of growth of test organisms was observed. Furthermore, Gentamicin sulphate (Sigma
G1272) was used as the antibiotic reference standard. The experiments were conducted twice,
and data were averaged.
3. Results and discussion

Glycine, caprolactam and caprolactam-glycine were tested to determine their antibacterial
activity against Gram-positive and negative bacteria through the disc diffusion method.
Caprolactam had the different level of antibacterial activity against four bacterial species,
but it did not affect E.faecalis, B.cereus, E.coli, K.pneumoniae, P.mirabilis, S.typhimirium,and
S.enteritidis species. The inhibition zones of the microorganisms sensitive to Caprolactam were
4-10 mm. The highest inhibition zone (10 mm) was observed against P.aeruginosa. The
inhibition zone against S.aureus and S.epidermidis was 8 mm while the component exhibited
poor antibacterial activity against B. subtilis (4mm) and there was no inhibition against rest of
tested bacteria. The inhibition zones of the microorganisms sensitive to control antibiotic were
16-22 mm. The results obtained with caprolactam were determined to be incompatible with
effects of control antibiotics.
Glycine and caprolactam-glycine did not form a measurable inhibition zone against any
of the tested bacteria. The results of the disk diffusion test of components and positive control
against 11 different bacterial species are shown in Table 1.

Table 1. Results of the antibacterial activity of Gly, CL, CL-Gly by disc diffusion method
Diameter of inhibition zone (mm)
Bacteria
S.aureus
Gly
CL
8
8
-
CL-Gly
Gentamicin
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
19
20
S.epidermidis
E.faecalis
Gram
positive
bacteria
B.cereus
-
22
18
19
18
16
18
B.subtilis
4
-
E.coli
K.pneumoniae
P.aeruginosa
P.mirabilis
S.typhimirium
S.enteritidis
-
Gram
negative
bacteria
10
-
-
-
Glycine, caprolactam and caprolactam-glycine tested to determine their antibacterial
activity against Gram-positive and negative bacteria through the agar dilution method. It was
decided that the components have different levels of effect against bacterial species.
Glycine was found to affect both gram-positive and gram-negative bacteria. When
evaluated against gram-positive bacteria, it was found to be effective against S.epidermidis and
B. subtilis (MIC=30 mg/ml), but no activity was seen against S.aureus, E.faecalis and B.cereus.
The results obtained were determined to be incompatible with the effects of control antibiotics.
When evaluated against gram-negative bacteria, it was found to be effective against E.coli,

K.pneumoniae, S.typhimirium and S.enteritidis (MIC=30 mg/ml). The component has not shown
any antimicrobial activity against P.aeruginosa and P.mirabilis. The results were incompatible
with results of control antibiotics.
It was determined that caprolactam have an effect on varying rates of bacteria except for
B. subtilis. The caprolactam demonstrated the lowest MIC values against S. aureus and
S.epidermidis (MIC=6.25 mg/ml). It was found to be less effective against E.faecalis and
B.cereus (MIC=50 mg/ml). The highest effect was seen against E.coli and P.aeruginosa
(MIC=25 mg/ml) while it was found to be effective at MIC of 50 mg/ml against S. pneumoniae,
P. mirabilis, S. Typhimurium and S.enteritidis. The results obtained were determined to be
incompatible with the effects of control antibiotics.
CL-Gly has been shown activity in different levels on gram-positive and gram-negative
bacteria. The component demonstrated the lowest MIC values against B.subtilis (MIC=15mg/ml)
and S.epidermidis (MIC=30 mg/ml), while no activity was exhibited against S.aureus, E.faecalis
and B.cereus. When evaluated against gram-negative bacteria, it was found to be effective
against E.coli, K.pneumoniae, S.typhimirium and S.enteritidis at 30 mg/ml MIC value. The
component has not shown any antimicrobial activity against P.aeruginosa and P.mirabilis. The
results were incompatible with results of control antibiotics (Table 2).
Table 2. Results of the antibacterial activity of Glycine, Caprolactam and Caprolactam-Glycine
by agar dilution method
Minimum inhibitory concentrations (MIC) in mg /ml
Bacteria
Gly
CL
CL-Gly
Gentamicin
S.aureus
growth
6.25
growth
0.002
Gram

S.epidermidis
E.faecalis
30
growth
growth
30
6.25
50
30
growth
growth
15
0.002
0.008
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
positive
bacteria
B.cereus
50
B.subtilis
growth
25
E.coli
30
30
K.pneumoniae
P.aeruginosa
P.mirabilis
S.typhimirium
S.enteritidis
30
50
30
Gram
negative
bacteria
growth
growth
30
25
growth
growth
30
50
50
30
50
30
3.1. Structure
The optimization of the new cluster examined reveals two strong hydrogen bonds between CL
and Gly molecules; O1-H27 (1.623 Å) and O24-H9 (1.878 Å). The structural parameters of the
optimized conformation of cluster are given in Table 3.
Dispersion correction contributed significantly to the interaction energy as shown in the study by
Izgorodina et al. (2009). In this study, this correction was made in order to take into account the
weak non-covalent interactions such as charge transfer interactions and van der Waals
interactions. The interaction energy for (II) molecule {ΔE = ECL-Gly - (ECL + EGly)} between CL
and Gly are –20.8 kcal/mol (–0.901 eV), according to DFT/wb97xd functional theory.

Table 3. Structural parameters for monomeric form of new cluster obtained by DFT/B3LYP/(6-311++G(d,p)) (I) and
DFT/wb97xd (6-311++G(d,p)) (II), in gas phase
Atoms
R(1,2)
R(1,27)
R(2,3)
R(2,4)
R(3,5)
R(3,6)
R(3,7)
R(4,8)
R(4,9)
R(5,10)
R(5,11)
R(5,29)
R(8,12)
CL-Gly(I)
1.243
1.625
1.517
1.345
1.543
1.090
1.097
1.463
1.023
1.534
1.094
1.096
1.534
CL-Gly(II)
1.238
1.623
1.513
1.341
1.537
1.089
1.097
1.455
1.022
1.529
1.094
1.096
1.529
Atoms
CL-Gly(I)
1.015
1.526
1.095
1.094
1.222
1.321
1.009
127.9
119.8
121.5
118.7
114.3
105.7
CL-Gly(II)
1.013
1.521
1.094
1.094
1.218
1.312
1.003
127.6
119.9
121.6
118.6
113.9
105.7
Atoms
CL-Gly(I)
106.2
114.7
105.7
110.1
109.5
109.9
106.5
170.8
115.5
108.6
108.7
108.8
106.1
CL-Gly(II)
106.3
114.4
105.9
110.1
109.5
109.9
106.6
171.1
115.2
108.6
108.7
108.9
106.2
R(18,25)
R(20,21)
R(20,22)
R(20,23)
R(21,24)
R(21,26)
R(26,27)
A(2,1,27)
A(1,2,3)
A(1,2,4)
A(3,2,4)
A(2,3,5)
A(2,3,6)
A(11,5,29)
A(4,8,12)
A(4,8,13)
A(4,8,14)
A(12,8,13)
A(12,8,14)
A(13,8,14)
A(4,9,24)
A(5,10,12)
A(5,10,15)
A(12,10,15)
A(12,10,16)
A(15,10,16)

R(8,13)
R(8,14)
R(9,24)
1.091
1.097
1.909
1.091
1.097
1.878
1.529
1.098
1.094
1.095
1.096
1.013
1.445
A(2,3,7)
A(5,3,6)
A(5,3,7)
A(6,3,7)
A(2,4,8)
A(2,4,9)
A(8,4,9)
A(3,5,10)
A(3,5,11)
A(3,5,29)
109.3
110.3
109.6
107.4
126.9
115.1
117.9
114.8
107.8
109.2
109.3
110.4
109.6
107.6
126.6
115.2
118.3
114.6
107.9
109.2
A(8,12,10)
A(8,12,17)
A(8,12,28)
A(10,12,17)
A(10,12,28)
A(17,12,28)
A(19,18,20)
A(19,18,25)
A(20,18,25)
A(18,20,21)
114.5
107.8
108.6
108.5
110.4
106.7
110.4
106.1
110.4
116.5
114.2
107.9
108.6
108.5
110.3
106.9
110.3
106.1
110.3
116.2
R(10,12) 1.534
R(10,15) 1.098
R(10,16) 1.095
R(12,17) 1.096
R(12,28) 1.096
R(18,19) 1.015
R(18,20) 1.451
a R and A stand for bond (Ǻ), angle (deg) respectively.

3.2. Molecular electrostatic potential
The separation of the electric charge distribution in a molecule is explained by the
polarity of a molecule. Thus, the centers of positive and negative charge distributions in the
molecule cause a dipole moment. In order to explain the behavior of molecules during reactions
and to understand where they bind to another molecule, MEP, which is calculated from the most
stable (i.e., the lowest-energy) molecular geometries obtained using the DFT method at the
wb97xd /6-311++ G(d,p) level of theory, is needed.
The centers of positive (the most electropositive) and negative (the most electronegative)
charge distributions on the electrostatic potential surface of the cluster are represented by blue
and red color, respectively (Figure 5). Magnitude of electrostatic potential represented by
different colors is in the order: red with color code: −1.340 V<orange<yellow<green<blue with
color code: +1.340 V.
Hydrogen atoms as positive potential regions pertaining to the NH and CH2 groups are
responsible for nucleophilic attack, while oxygen and nitrogen atoms as negative potential
regions are responsible for electrophilic attack. These active sites show whether this cluster with
a dipole moment of 4.79 D has a beneficial effect on the organism.
In a study on dipole moments of some ionic liquids (Studzinska et al., 2011), the dipole
moments were found to be in the range of 1.91-17.31 D and our result is in good agreement with
the previous findings.

Fig.5. Molecular electrostatic potential (MEP) of CL-Gly obtained by DFT/wb97xd/6-
311++G(d,p)
3.3. HOMO and LUMO Energies
The highest molecular orbit occupied by electrons is called HOMO and the lowest molecular
orbit occupied by electrons is called LUMO. HOMO functions as an electron donor because it is
the outermost molecular orbit occupied by electrons, while LUMO functions as an electron
acceptor since it is the first empty molecular orbit unoccupied by electrons. Thus, HOMO
directly represents the ionisation potential of a molecule, while LUMO represents the electron
affinity of the molecule. The energy gap between HOMO and LUMO is an indicator of the
chemical stability of a molecule and is an important parameter used to determine the flow
properties of electrical charge of the molecule (Koopmans, 1934; Fukui, 1982). The LUMO
molecular orbital is localized on caprolactam molecule, whereas the HOMO molecular orbital on
the interaction zone of glycine and caprolactam molecules. The HOMO-LUMO energy range of
the synthesized cluster was calculated and found to be 9.931 eV (Figure 6). In another study

(Khan et al., 2014), this energy range for choline-based ionic liquid and its complexes was
reported to be about 4 ~ 5.5 eV. The quantum chemical parameters such as ionization potential,
the electron affinity, the electronegativity, the hardness and the electrophilicity, which describes
the biological effect of the molecule, were calculated using Koopmans (1934) theorem and were
found to be 8.979 eV, -0.952 eV, 4.0135 eV, 4.9655 eV and 1.622 eV, respectively.
Fig.6. The atomic orbital HOMO-LUMO composition of the frontier molecular orbital for CL-
Gly

3.4. Docking studies
Molecular docking is a type of modeling that shows a stable three-dimensional structure
consisting of interactions between two or more molecules and which bonds are formed between
them.
Ionic liquids (ILs) have recently been used to dissolve biopolymers such as cellulose, elastin and
chitin (Mantz et al., 2007; Swatloski et al., 2002) due to their high dissolving power (Dadi et al.,
2007; Edgar et al., 2010; Tojo et al., 2014) and also their dissolving degree can be adjusted by
changing the cation and anion in their structure (Bulkowska et al., 2016). It has been reported
that DNA, a natural biopolymer, can be stored at room temperature for a long time by
maintaining its chemical and structural stability in 2-hydroxyethylammonium formate IL and
dissolved at a high concentration in the IL (Singh et al., 2017). It has been demonstrated that this
high dissolution and long-term stability is due to hydrogen bonds between the IL and DNA.
Autodock-Vina program (Trott et al., 2010) was used to show the region in which any binding
between CL-Gly IL and DNA takes place and the interaction between CL and Gly. The three
dimensional molecular structure of DNA was obtained from the protein data bank (PDB ID:
1BNA) (Drew et al., 1981) DNA was made suitable to the docking by removing water molecules
in DNA and adding polar hydrogens in it and the Kollman charges of DNA were determined.
After optimization of CL-Gly molecule in gas phase for adapting the docking, Geistenger

method was used to find the partial charges of the synthesized molecule. The active site of DNA
which was defined within the grid size of 40 x 40 x 40 Å was bound to the CL-Gly molecule by
hydrogen bonds (Figure 7). The stable CL-Gly structure in DFT/wb97xd/6-311++G(d,p) level of
theory is bound to the DT8, DG10, DG16 and DA17 residues of DNA via the hydrogen bonds.
There are H-bonding interactions between the DG16, DA17 residues of DNA and the carbonyl
oxygen of caprolactam molecule and also between the DT8, DG10 residues and
hydrogen/oxygen atom in Gly molecule.
Since a stable complex is formed between the DNA and CL-Gly IL, it can be concluded that the
stability of the DNA is improved. Using UCSF Chimera program (Pettersen et al., 2004) and,
TIP3PBOX solvent model for which simulation time of the IL solvated in a cubic box with 921
water molecules was 1.2 ns (120 103 frames), CL-Gly IL was simulated at the molecular level.
Since the clusters had problems in performing MD simulations, the hydrogen bond lengths
between Gly and CL were kept constant. The stable conformations of ligand can be clearly seen
from RMSD graphs (Figure S1). Following the use of MD simulation results for re-docking of
CL-Gly with DNA, it is remarkable that the lowest-energy conformer (–6.8 kcal/mol) determined
by MD simulations is numerically very close to that obtained by optimized geometries (–6.7
kcal/mol). It follows from the obtained docking model that the nucleobases DC9, DG10, DC11,
DG16, DA17 and DA18 of DNA receptor interacted with CL-Gly (Figure S2). Quantum
mechanical calculation rendered possible to find the optimized geometry. In consequence of
docking analysis using the optimized geometry, the active sites of the DNA receptor to bind were
shown to be DG10, DG16 and DA17, which infers that the results obtained in both calculations
confirmed each other.

In some studies carried out with different ligands over the past decade, it has been shown that
especially the nucleobase DG10 of DNA is active site to bind (Allaka et al., 2018; Cheraghi et
al., 2017; Das et al., 2018; Vijayalakshmi et al., 2014; Vijayalakshmi et al., 2013).
(a)
(b)
Fig. 7. (a) Docking of CL-Gly with DNA. (b) The dotted lines present the interactions (binding
affinity -6.7 kcal/mol)
3.5. Analysis of toxicological and physicochemical properties
Toxic risk assessment is a criterion that indicates whether a chemical has a toxic effect
(ie, a potential health risk) on the organism. Toxic risk assessment comprises the following
factors: hazard identification, dose-response assessment, exposure assessment, and risk
characterization (Pepper et al., 2015).

The partition coefficient (P) is the ratio of the concentration of a compound in octanol as
a non-polar solvent to its concentration in water as a polar solvent, thus being an indication of the
hydrophilicity of a compound and it is represented by clogP as the logarithm of the partition
coefficient. The lower the logP value of a drug, the more hydrophilic it is (Van de Waterbeemd
et al., 2001), thus making it slower to be absorbed orally (Qui et al., 2017).
The solubility of a compound in water is indicated by logS and is a measure of
absorption. the greater the logS, the higher the absorption (Ishikawa et al., 2011). As the
molecular weight of the compounds increases, their polarity decreases further and their solubility
in water decreases (Faust et al., 1998). Water-soluble drugs are less soluble in lipid (Meisenberg
et al., 2016).
The topological polar surface area (TPSA), defined as the surface sum on electronegative
atoms in a molecule, is another absorption criterion for evaluating pharmacokinetics (for
instance, intestinal absorption, Caco-2 monolayer penetration, blood-brain barrier penetration)
(Ertl et al., 2000). The higher the TPSA of a compound, the more difficult it is to penetrate
through the cell membrane (Begley, 2008). When TPSA is greater than 140 Å2 and molecular
weight is greater than 500 g, human intestinal absorption is restricted (Di et al., 2015; Brito,
2011).
Drug score refers to the potential for a molecule to be used as a drug depending on
druglikeness, clogP, logS, molecular weight and toxicity risks (Krishna et al., 2012). A positive
and high drug score toward 1 indicates that molecule has a drug-like structure and usability as a
drug, whereas a drug score approaching 0 indicates a high risk for that molecule to use as a drug
(Guan et al., 2019). However, it is necessary to investigate the effects of the compound in vivo to
confirm its usability as a drug (Hosoya et al., 2016).

The estimated values of the toxicity risks of CL (Celik et al., 2019) was calculated in our
previous study. The estimated values of the toxicity risks of Gly and Cl-Gly IL together with
some important physicochemical properties were determined using OSIRIS Property Explorer
(Osiris, 2010) and Molsoft (https://www.molsoft.com/mprop/). The results are given in Table 4.

Table 4. Prediction of toxicity risks and physicochemical properties by Osiris and Molsoft of title compounds.
Compound
Toxicity risks
Physicochemical properties
Mutagenic
Tumorigenic
Irritant Reproductive
CLogP Solubility
MW
TPSA Drug
likeness
-7.96
Drugscore
Effect
(+)
CL (Celik et al., 2019)
(+)
(+)
(+)
(-)
(+)
(-)
0.52
-1.38
-0.03
-0.97
113
29.1
0.06
0.35
Gly
(-)
-3.35
-0.88
75
63.32
76.55
-1.67
-1.54
CL-Gly
188.12

3.6. Absorption analysis of target cluster
Blood brain barrier (BBB) formed by endothelial cells (Ballabh et al., 2004) protects the
central nervous system by limiting the passage of neurotherapeutics and small-molecule drugs
into the brain (Greene et al., 2016). Drugs with CNS activity can only have an action in brain by
passing through blood-brain barrier (Patrick, 2013). The more hydrophobic the drug is, the
higher the transport to the brain (Mutschler et al., 1995).
BBB permeable (+) is an essential concept that protects peripheral organs (central
nervous system) against neurotoxins and governs and regulates the diffusion of drugs through
blood-brain barrier (Erdo et al., 2017).
Small intestine is an organ with large surface area (Helander et al., 2014) which carries
out the absorption of oral drugs to enable them to enter the bloodstream. Drugs taken orally are
mainly absorbed by the intestines (Murakami, 2017). Intestinal absorption is of great importance
for the rapid absorption of drugs, hence in terms of improving bioavailability (Peterson et al.,
2019). The absorption percentages of the compounds having poor apparent permeability (poor
absorption), medium permeability (mid-level absorption) and high permeability (strong
absorption) coefficients range from 0-20%, 20-70% and 70-100%, respectively (Yee, 1997).
Caco-2 (human colon epithelial cancer cell line, model of the intestinal epithelial barrier)
is an assay dealing with intestinal absorption to evaluate the permeability of drugs (Van Breemen
et al., 2005).
P-glycoprotein (P-gp), which is available in epithelial cells and also the endothelial cells
(You et al., 2014), is an efflux drug transporter that serves as a biological barrier that protects
cells from the harmful effects of drugs by transporting toxins and xenobiotics out of cells

(Brunton et al., 2010). Consequently, it ensures that hazardous substances on human health are
excreted through gastrointestinal tract, bile and urine (Amin, 2013).
Cytochrome P450 is a hemeprotein that metabolizes exogenous and endogenous
compounds that are toxic to cells (Yahia, 2018).
Potassium (K +) channels found in cell membranes allow potassium ions to flow into and
out of the cell (Waxman, 2005), thereby modulating the electrical properties of the cells (Yasuda
et al., 2008).
The various ADMET parameters of the investigated cluster, are characterized using the
silico module admetSAR (Cheng et al., 2012). The predicted values of CL-Gly are given in
Table 5.
Table 5. Prediction of ADMET profiles of the CL-Gly.
ADMET Predicted Profile --- Classification
Model
Result
Probability
Absorption
BBB+
Blood-Brain Barrier
0.8460
0.5000
0.7677
0.5836
0.9595
0.9926
0.8847
Human Intestinal Absorption
Caco-2 Permeability
HIA-
Caco2-
P-glycoprotein Substrate
P-glycoprotein Inhibitor
Substrate
Non-inhibitor
Non-inhibitor
Non-inhibitor
Renal Organic Cation Transporter
Subcellular localization
Distribution
Mitochondria
Metabolism
0.6911

CYP450 2C9 Substrate
CYP450 2D6 Substrate
CYP450 3A4 Substrate
CYP450 1A2 Inhibitor
CYP450 2C9 Inhibitor
CYP450 2D6 Inhibitor
CYP450 2C19 Inhibitor
CYP450 3A4 Inhibitor
CYP Inhibitory Promiscuity
Non-substrate
Non-substrate
Non-substrate
Non-inhibitor
Non-inhibitor
Non-inhibitor
Non-inhibitor
Non-inhibitor
Low CYP Inhibitory
Promiscuity
0.8799
0.7824
0.7221
0.9416
0.9676
0.9613
0.9454
0.9556
1.0000
Excretion
Toxicity
Weak inhibitor
Human Ether-a-go-go-Related Gene
Inhibition
0.9910
0.8988
0.8309
0.9623
0.9416
0.9598
0.8054
0.5000
0.6531
0.6672
Non-inhibitor
Non AMES toxic
Non-carcinogens
Low FHMT
AMES Toxicity
Carcinogens
Fish Toxicity
Tetrahymena Pyriformis Toxicity
Honey Bee Toxicity
Biodegradation
Low TPT
Low HBT
Ready biodegradable
III
Acute Oral Toxicity
Carcinogenicity (Three-class)
Non-required
ADMET Predicted Profile --- Regression
Model
Value
Absorption
-1.8796
Unit
Aqueous solubility
LogS

Caco-2 Permeability
0.1483
Distribution
LogPapp, cm/s
Metabolism
Excretion
Toxicity
Rat Acute Toxicity
1.8099
LD50, mol/kg
pLC50, mg/L
pIGC50, ug/L
Fish Toxicity
2.7847
Tetrahymena Pyriformis Toxicity
-0.7429
The BBB is the microvascular endothelial cell layer of the brain and plays an important
role in separating the brain from the blood. The BBB permeability is required for the central
nervous system drugs, whereas for the non- central nervous system drugs the BBB penetration
should be minimized, to avoid undesired side effect. According to the predicted properties, the
investigated IL is found to be BBB permeable (BBB+). The drug permeability in human
intestinal epithelial (Caco-2) cells is used to control intestinal absorption. It is an in vitro model
used to predict drug absorption in which the drug is administered orally. From Caco2
permeability study it was observed that IL has a negative result in Caco2 permeability, indicating
that it was not permeable. Human intestinal absorption, HIA, as a key procedure of oral
absorption, unfortunately, the investigated IL has a negative results in HIA.
4. Conclusions
In this study molecular structure, electronic properties and the antibacterial activities of new
synthesized cluster CL-Gly were evaluated. It has been shown that CL-Gly has antibacterial
activity in different levels on gram-positive and gram-negative bacteria. The interaction between

DNA and CL-Gly was achieved by molecular docking study. It was found that CL-Gly made a
stable complex with DNA, thus can be used for DNA stabilization. In silico ADMET study was
also performed for predicting pharmacokinetic and toxicity profile of the synthesized cluster
which expressed good drug-like behavior and non-toxic nature. It was revealed that the
compound has a potential to become an important molecule in drug discovery process.
Funding
This study was supported by the Research funds of Istanbul University [ÖNAP-2423], [N-3341],
[N-3875].
REFERENCES
Ajloo, D., Moghadam, M. E., Ghadimi, K., Ghadamgahi, M., Saboury, A. A., Divsalar, A.,
SheikhMohammadi, M. & Yousefi, K. (2015). Synthesis, characterization, spectroscopy,
cytotoxic activity and molecular dynamic study on the interaction of three palladium
complexes of phenanthroline and glycine derivatives with calf thymus DNA. Inorganica
Chimica Acta, 430, 144-160.