Proliferation inhibition of novel diphenylamine derivatives

Article abstract of DOI:10.1016/j.bioorg.2018.10.063

Nonsteroidal anti-inflammatory drugs (NSAIDs) are the most widely used drugs in the world but some NSAIDs such as diclofenac and tolfenamic acid display levels of cytotoxicity, an effect which has been attributed to the presence of diphenylamine contained in their structures. A novel series of diphenylamine derivatives were synthetised and evaluated for their cytotoxic activities and proliferation inhibition. The most active compounds in the cytotoxicity tests were derivative 6g with an IC₅₀ value of 2.5 ± 1.1 × 10^?6 M and derivative 6f with an IC₅₀ value of 6.0 ± 3.0 × 10^?6 M (L1210 cell line) after 48 h incubation. The results demonstrate that leukemic L1210 cells were much more sensitive to compounds 6f and 6g than the HEK293T cells (IC₅₀ = 35 × 10^?6 M for 6f and IC₅₀ > 50 × 10^?6 M for 6g) and NIH-3T3 (IC₅₀ > 50 × 10^?6 M for both derivatives). The IC₅₀ values show that these substances may selectively kill leukemic cells over non-cancer cells. Cell cycle analysis revealed that a primary trend of the diphenylamine derivatives was to arrest the cells in the G₁-phase of the cell cycle within the first 24 h. UV–visible, fluorescence spectroscopy and circular dichroism were used in order to study the binding mode of the novel compounds with DNA. The binding constants determined by UV–visible spectroscopy were found to be in the range of 2.1–8.7 × 10⁴ M^?1. We suggest that the observed trend for binding constant K is likely to be a result of different binding thermodynamics accompanying the formation of the complexes.

Full text of DOI:10.1016/j.bioorg.2018.10.063

Bioorganic Chemistry 83 (2019) 487–499
Contents lists available at ScienceDirect
Bioorganic Chemistry
journal homepage: www.elsevier.com/locate/bioorg
Proliferation inhibition of novel diphenylamine derivatives
a,⁎
b,c
a
b
d
Ladislav Janovec , Jana Janočková , Mária Matejová , Eva Konkoľová , Helena Paulíková ,
b,c
b
d
a
a
Daniela Lichancová , Lenka Júnošová , Slávka Hamuľaková , Ján Imrich , Mária Kožurková
b
a
Department of Organic Chemistry, P. J. Safarik University, Faculty of Science, Moyzesova 11, 04001 Kosice, Slovak Republic
Department of Biochemistry, P. J. Safarik University, Faculty of Science, Moyzesova 11, 04001 Kosice, Slovak Republic
c
Biomedical Research Center, University Hospital Hradec Kralove, Sokolovska 581, Hradec Kralove, Czech Republic
d
Department of Biochemistry and Microbiology, Faculty of Chemical and Food Technology, Slovak Technical University, Radlinskeho 9, 81237 Bratislava, Slovak Republic
A R T I C L E I N F O
A B S T R A C T
Keywords:
Nonsteroidal anti-inflammatory drugs (NSAIDs) are the most widely used drugs in the world but some NSAIDs
such as diclofenac and tolfenamic acid display levels of cytotoxicity, an effect which has been attributed to the
presence of diphenylamine contained in their structures. A novel series of diphenylamine derivatives were
synthetised and evaluated for their cytotoxic activities and proliferation inhibition. The most active compounds
Antiproliferative
Diphenylamine
DNA minor groove binder
DNA binding study
−6
in the cytotoxicity tests were derivative 6g with an IC50 value of 2.5 ± 1.1 × 10 M and derivative 6f with an
−6
IC50 value of 6.0 ± 3.0 × 10 M (L1210 cell line) after 48 h incubation. The results demonstrate that leukemic
−
6
L1210 cells were much more sensitive to compounds 6f and 6g than the HEK293T cells (IC50 = 35 × 10 M for
f and IC50 > 50 × 10 M for 6g) and NIH-3T3 (IC50 > 50 × 10 M for both derivatives). The IC50 values
show that these substances may selectively kill leukemic cells over non-cancer cells. Cell cycle analysis revealed
−6
−6
6
that a primary trend of the diphenylamine derivatives was to arrest the cells in the G -phase of the cell cycle
1
within the first 24 h. UV–visible, fluorescence spectroscopy and circular dichroism were used in order to study
the binding mode of the novel compounds with DNA. The binding constants determined by UV–visible spec-
4
−1
troscopy were found to be in the range of 2.1–8.7 × 10 M . We suggest that the observed trend for binding
constant K is likely to be a result of different binding thermodynamics accompanying the formation of the
complexes.
1
. Introduction
Nonsteroidal anti-inflammatory drugs (NSAIDs) are the most widely
through which this interaction occurs include intercalation between
adjacent base pairs, and intrusion into the minor and major grooves.
Intercalation and minor groove binding are the predominant DNA-
binding modes of small ligands [4–6].
used drugs in the world [1], but some NSAIDs such as diclofenac and
tolfenamic acid display levels of cytotoxicity, an effect which has been
attributed to the presence of diphenylamine contained in their struc-
tures (Fig. 1). There is some structural resemblance between the di-
phenylamine core and DNA minor groove binders, and therefore we
synthesized a series of diphenylamine derivatives (6a–6g) in order to
test their cytotoxic activity and their DNA binding ability (Fig. 1) [1].
The minor groove has a helical curvature which can also vary slightly
depending on the sequence, so agents with unfused heterocyclic sys-
tems which bind to the minor groove must therefore have a concave
shape that complements that of the groove [2]. Small molecules which
bind to genomic DNA and which are easily accessible to chromosomal
DNA have proven to be effective anticancer therapeutic agents, and
therefore researchers continue to be interested in designing pharma-
cophores which are capable of interacting with DNA [3]. The modes
In this report, we present the results of our assays showing that
diphenylamino carboxylic acid derivatives 6a–6g demonstrate inter-
esting levels of antiproliferative activity with interaction with DNA
through groove-binding. The IC50 values obtained for the compounds
also show that novel derivatives 6f and 6g may selectively kill leukemic
cells over non-cancer cells.
2. Results and discussions
2.1. Chemistry
Derivatives 6a–6g were synthesized using the following reaction
pathway (Scheme 1) [7]. 4-Acetamido-2-chlorobenzoic acid (3) was
prepared with a yield of 65% through the oxidation of 4-acetamido-2-
⁎
Corresponding author.
E-mail address: ladislav.janovec@upjs.sk (L. Janovec).
https://doi.org/10.1016/j.bioorg.2018.10.063
Received 26 January 2018; Received in revised form 19 October 2018; Accepted 29 October 2018
Available online 02 November 2018
0045-2068/ © 2018 Elsevier Inc. All rights reserved.

L. Janovec et al.
Bioorganic Chemistry 83 (2019) 487–499
Fig. 1. The structures of DNA minor groove binders Hoechst 3325, Berenil, DAPI, the diphenylamine derivative A with antitumor activity, and DNA minor groove
binding ability CD27.
chlorotoluene (2) with KMnO
4
. Compound 2 was obtained by the
elongated alkyl chain. Compound 6f was found to be the most potent
derivative from this set; the high value of log P of 6f (Table 1) enhanced
the permeability of this substance and thereby improved its cytostatic
activity.
acetylation of 4-amino-2-chlorotoluene (1). Substituted acid 5 was
synthesized through the coupling of 4-acetamido-2-chlorobenzoic acid
(
3) with 3-aminoacetanilide (4) in ethoxyethanol in the presence of a
catalytic amount of copper and copper(I) oxide. Amides 6a–6g were
synthetised through the reaction of amines with carboxylic acid 5 in
DMF with the addition of carbonyldiimidazole.
Two of the novel compounds, derivatives 6a and 6f, were found to
possess different levels of cytotoxicity, lipophility and affinity to iso-
lated DNA, therefore the effect of these derivatives on L1210 cells was
examined in more detail. The influence of the two derivatives on cell
proliferation, plasma membrane integrity and cell cycle, and also the
intracellular distribution of the compounds were investigated in order
to determine the mechanism of their cytotoxicity.
3
. Biology
3.1. Cytotoxicity
The proliferation of L1210 cells in the presence of derivatives 6a
and 6f was determined using the trypan blue method in which the dye
is excluded from live cells. As can be seen in Fig. 2, a dose dependent
decrease was observed in the percentage of viable cells treated with
derivatives 6a and 6f, but no increase was discerned in the percentage
of dead cells (not shown). The results of LDH leakage assays based on
the determination of LDH activity in the extracellular medium show
that the addition of derivatives 6a and 6f at the same concentrations
which had inhibited cell proliferation did not induce plasma membrane
damage (Fig. 2). Derivatives 6a and 6f did not induce significant LDH
release (activity of LDHex) from the cells. No significant increase was
observed in LDH (i.e. the activity of LDHex) from the cells exposed to
derivatives 6a and 6f.
The anticancer activity of the novel derivatives was also studied by
evaluating the cytotoxicity of the compounds against leukemic cells and
non-cancer cells in vitro. The results presented in Table 1 demonstrate
that the leukemic L1210 cells were more sensitive to compounds 6a–6g
than the non-cancer HEK293T and NIH-3T3 cells. The immortalized
cells, human embryonic kidney 293 cells (HEK293T) and mouse em-
bryonic fibroblast cells (NIH-3T3) were less sensitive to the tested
substances; indeed, the viability of NIH-3T3 cells decreased only at
relatively high concentrations of the derivative, in excess of
−
6
5
0 × 10 M. The IC50 values of derivatives 6f and 6g show that these
substances may selectively kill leukemic cells over normal cells
(
Table 1). Derivative 6c with fluorine and methyl substituent 6b did not
These two derivatives had also exhibited strong potential for the
inhibition of leukemic cell proliferation without inducing cell death,
and therefore the morphological changes of L1210 cells incubated with
various concentrations of the compounds for 24 h were monitored and
compared with the results for untreated control cells. Fig. 3 shows
images of the cells following the addition of propidium iodide and
display any increased level of biological activity than that displayed by
derivative 6a. However, the replacement of a phenyl with a piperonyl
in derivative 6g led to a dramatic increase in the anticancer potency of
this compound, resulting in a level of toxicity 7-times higher than that
of derivative 6a.
Derivatives 6a and 6d–6f form a set of compounds which feature an
488

L. Janovec et al.
Bioorganic Chemistry 83 (2019) 487–499
Scheme 1. Synthesis of derivatives 6a–6g, amides of substituted diphenylamino carboxylic acid.
Table 1
3.2. Intracellular distribution of derivatives 6a and 6f and nuclear DNA as
IC50 values of derivatives 6a–6g against mouse leukemic cells L1210 and non-
cancer cells (HEK293T and NIH-3T3) determined by MTT- assay.
a direct target
−6
In order to examine the mechanism of the cytotoxicity of the di-
phenylamine derivatives against L1210 cells, the intracellular dis-
tribution of derivatives 6a and 6f was monitored using fluorescence
microscopy. Green fluorescence was applied to visualize the com-
Compounds
IC50 (×10 M)
L1210
HEK293T
NIH-3T3
6
6
6
6
6
6
6
a
b
c
d
e
f
20 ± 5.3
31 ± 3.8
34 ± 4.2
10 ± 2.4
10 ± 3.2
6 ± 3.0
50 ± 3.2
38 ± 2.8
50 ± 5.1
48 ± 2.4
26 ± 3.3
35 ± 4.1
> 50
> 50
> 50
> 50
> 50
> 50
> 50
> 50
−6
pounds and the relatively high concentration of 40 × 10 M was used
because of low fluorescence of the compounds. The colocalizations of
the substances were also examined using the nucleic acid stain SytoRed
(
Syto 62 Red). Fig. 5 shows the microphotographs obtained during the
assay. The cellular sequestration of derivatives 6a and 6f is shown in
the strong fluorescence signal primarily observed inside of the nucleus
and extending also to the cytosol of the L1210 cells (Fig. 5). The co-
localization of Syto Red with compounds 6a and 6f confirmed the lo-
calization of both of the derivatives in the nuclei of L1210.
g
2.5 ± 1.1
The cells were incubated with the substances for 48 h. The tested substances
were used in the concentration range of 0–50 × 10−6 M.
In a further assay, cells were treated with the DNA-selective com-
pound Hoechst 33342 in order to provide more information on the
interaction of derivatives 6a and 6f with nuclear DNA. This dye readily
interacts with DNA and is known to act as a minor groove binding li-
gand [8]. L1210 cells were incubated with different doses of derivatives
6a or 6f for 24 h and the treated cells were subsequently washed with
PBS and dyed with Hoechst 33342. The stained nuclei in the treated
cells revealed that derivatives 6a and 6f had reduced the nuclear
binding of Hoechst 33342 in a dose dependent manner (Fig. 6).
Hoechst 33342 dyes. No significant differences were observed between
the morphological features of the treated and untreated cells at lower
concentrations, but binuclear cells were observed following treatment
−
6
with compound 6a at a concentration of 40 × 10 M. The treated cells
did not display any loss of membrane integrity and no induction of
necrosis (PI-positive cells) was observed (microphotographs are not
shown).
In order to determine the mechanism by which the compounds af-
fect cells, the effect of the two derivatives on cell cycle progression was
compared. Cytometric analysis showed that L1210 cells treated with
derivatives 6a and 6f remained at the G1–S boundary. Both substances
induced the accumulation of the cells in the G1 phase of the cell cycle
even at concentrations equivalent to IC50 (Fig. 4).
3.3. DNA binding studies
The newly synthesized derivatives 6a–6g were investigated for their
ability to interact with calf thymus using spectrophotometric titrations
in a Tris-HCl buffer. UV–visible absorption titrations were performed in
489

L. Janovec et al.
Bioorganic Chemistry 83 (2019) 487–499
Fig. 2. Effect of derivatives 6a and 6f on cell proliferation and plasma membrane integrity. L1210 cells were incubated with compounds 6a and 6f for 24 h. The
viable cells were counted (trypan blue exclusion test) and the extracellular activity of LDH (LDHex) was determined using LDH leakage assay.
order to ascertain the binding strength of the studied compounds with
DNA by monitoring changes in the absorption intensity of the ligand
centered bands at a range of around 200–400 nm (Table 2). The ab-
sorption spectra of derivatives 6a–6g in the presence of ctDNA ex-
hibited broad absorption bands in the region of 240–400 nm. The bands
of the studied samples exhibit a similarly lower level of hypochromism
without any red shift following the incremental addition of ctDNA, a
finding which indicates that the compounds bind to DNA via a non-
intercalative binding mode, possibly through groove-binding. In order
to compare the DNA binding affinities of the compounds, binding
constants K were determined for each compound using McGhee and
von Hippel plots. The binding constants K were estimated to range from
maximum absorptions in the absence of ctDNA, the ctDNA samples
were pretreated with EB in order to make the interactions more visible.
EB forms soluble complexes with DNA and emits intense fluorescence in
the presence of ctDNA due to the intercalation of the planar phenan-
thridinium ring between adjacent base pairs on the double helix [10].
Fluorescence spectra were monitored at a fixed concentration of ctDNA
pretreated with EB during titration with increasing concentrations of
derivatives 6a–6g. Figs. 7 and S1 describe the emission spectra of EB
bound to DNA in both the presence and absence of the new derivatives.
The results of the fluorescence studies demonstrate a quenching of
fluorescence intensity of the EB-DNA complex following the addition of
increasing concentrations of compounds 6a–6g (Fig. 7). The results
indicate that the compounds can intercalate into the base pairs of DNA
and displace EB from the DNA. However, the reduction in the emission
intensities of DNA bound EB may be a result of the activity of either
DNA intercalators or DNA groove-binders; studies have shown that non-
intercalating DNA groove binding agents such as berenil, bisamidines,
spermine and spermidine are also capable of displacing EB from DNA
[11–13].
5
−1
0
.21 to 0.87 × 10 M . In comparison, binding constants for groove
5 9 −1
binders generally range from 10 to 10 M
[9]. The binding para-
meters obtained from spectrophotometric analysis are summarized in
Table 1.
In order to examine the ability of compounds 6a–6g to displace EB
(
ethidium bromide) from its EB-DNA complex, a competitive EB
binding study was performed using fluorescence studies. As derivatives
a–6g were not excited at wavelengths corresponding to their
6
In order to identify the binding method of the derivatives to
Fig. 3. Representative merged micro-
photographs of L1210 cells after treatment
with derivatives 6a and 6f for 24 h. The cells
were treated with dual staining of Hoechst
3
3342/PI to show morphological changes.
−6
A-control, B-20 × 10 M of compound 6a,
−
6
C-40 × 10
M
of compound 6a, D-
−
−
6
2
0 × 10
0 × 10
M
M
of
of
compound
6f, E-
6f.
6
4
compound
Microphotographs of the cells were obtained
using an Axio Imager A1 (Zeiss, Germany)
fluorescence microscope at a magnification
of 40 × 10.
490

L. Janovec et al.
Bioorganic Chemistry 83 (2019) 487–499
ctDNA and phenylethyl derivative 6d displayed changes to both the
positive and negative bands. Representative CD spectra results of
compounds 6a and 6g are shown in Figs. 8 and S2.
3.4. Docking study
Previous experiments had suggested that the interactions between
the novel derivatives and DNA had occurred in a minor groove-binding
mode, therefore docking studies utilizing the X-ray model were per-
formed in order to explore this hypothesis. In addition to docking stu-
dies of the novel derivatives, studies were also carried out using the
well-known DNA minor groove binder Hoechst 33258 (Id PDB: 127D)
in order to verify the accuracy of the docking predictions [19]. The
powerful software package AutodockVina was initially used to perform
the computations, but attempts to predict correct binding modes using
this software were unsuccessful; the AutodockVina predictions of CG
track for the main DNA binding sequence of Hoechst 33258 did not
match with the AT track which had been shown in the X-ray model (not
shown) [20]. On this basis, further studies were performed using Au-
todock software [21,22]. The combinations of charge assignment were
tested in order to ensure higher accuracy in the docking predictions;
Gasteiger and PM7 charge assignments were used for Hoechst 33258 in
combination with AMBER ff14SB and Gasteiger charge assignments for
the oligonucleotide [23,24]. In comparing the RMSD for the X-ray li-
gand pose and the docking ligand pose, the Gasteiger charge was con-
sidered to be the most appropriate assignment for both the receptor and
the ligand itself (Figs. S3, S4 and Table S1).
Fig. 4. Cell cycle analysis of mouse leukemic L1210 cells. The cells were in-
cubated with derivatives 6a and 6f for 24 h at concentrations equivalent to IC50
values or at double this concentration. Histograms were analyzed using BD
Accuri C6 Software to determine the percentage of the cells in each phase of the
cell cycle.
polymeric ctDNA, CD spectral assays were also performed (Figs. 8 and
S2). The DNA region of the CD spectrum shown from 220 to 300 nm is
characteristic of the retention of a B-DNA conformation [14]. The CD
spectrum of ctDNA is characterized by a positive band at 275 nm
(
symptomatic of base stacking) and a negative band at 245 nm (an in-
dication of the helicity of the B form of DNA). Ligands which bind to
DNA through groove-binding or electrostatic interactions exhibit little
or no interference with the helicity and stacking bands at the CD
spectrum of ctDNA (Figs. 8 and S2). Intercalators are known to stabilize
the B-form of DNA leading to significant changes in both bands
The oligonucleotide in the X-ray model (Id PDB:127D) retains an
induction fit in its structure, therefore we designed the docking simu-
lation for Hoechst 33258 with oligonucleotide 5′-d(CGCGAATTCGCG)-
3
′ (OLG) in the canonical B-DNA helix which was included as a building
option in the Chimera software package [25–27]. Gasteiger partial
charges were also added for the ligand and for the oligonucleotide. This
decision was made in order to exclude any possible influence of the
induction fit on the accuracy of the docking study. A visual inspection
of the result shows that the Autodock software was able to correctly
reproduce the preference of Hoechst 33258 for an AATT tract with the
correct orientation (Fig. 9A).
[
7,15–18], therefore it should be possible to ascertain the specific
binding mode which has occurred by observing the changes to the DNA
region of CD spectra following interaction between ctDNA and the
studied compounds. Whilst the B-DNA conformation was clearly re-
tained, small changes in the bands in the DNA region of the spectrum
are difficult to interpret, as induced CD signals from the ligand-based
spectroscopic transitions of compounds 6a–6c, 6f and 6g could also fall
within this region. In contrast, the spectra of the interaction between
The possibility of the sequence specific binding of derivatives 6f and
6
g to DNA was examined using the same details in docking simulations
Fig. 5. Intracellular localization of deriva-
tives 6a and 6f. L1210 cells were incubated
with compound 6a A–C and compound 6f
−6
D–F (40 × 10 M) for 24 h. Intracellular
localization of the substance was in-
vestigated using green fluorescence and the
colocalization of the substance with nucleic
acid stain SytoRed was monitored with an
Axio Imager A1 (Zeiss, Germany) fluores-
cence microscope at
3 × 10.
a magnification of
6
491

L. Janovec et al.
Bioorganic Chemistry 83 (2019) 487–499
Fig. 6. Dyeing of L1210 cell nuclei with Hoechst 33,342 after incubation with varying concentrations of derivatives 6a and 6f. The cells were incubated with
−
6
−6
−6
−6
−6
−6
derivative 6a (A-control, B-2.5 × 10 M, C-5 × 10 M, D-10 × 10 M) and derivative 6f (E-control, F-2.5 × 10 M, G-5 × 10 M, H-10 × 10 M) for 24 h.
−
6
The cells were then dyed with Hoechst 33,342 (11 × 10 M, 37 °C, 15 min) and monitored with an Axio Imager A1 (Zeiss, Germany) fluorescence microscope at a
magnification of 40 × 10.
Table 2
3.5. Molecular dynamics – A docking like study
Binding constants and log P values of diphenylamine derivatives 6a–6g.
Our closer analysis of the relationship between lipophility and
binding constant K (Table 1) reveals a mutual linear correlation within
derivatives 6a, 6d, 6e. As Fig. 10 shows, the constant K decreases in an
almost linear manner with increasing numbers of methylene groups in
the derivatives.
Compounds
6a
6b
6c
6d
6e
6f
6g
−
1
5
K (M )×10
0.56
0.43
0.21
0.39
0.26
0.87
0.44
ΔG (kJ.mol 1) −39.1 −33.1 −16.8 −30.9 −21.7 −49.1 −33.6
−
*
Log P
2.8
3.3
3.0
3.2
3.8
4.1
2.8
However, the results for derivative 6f did not follow this trend, and
the K value for this derivative was the highest recorded in the assay. In
terms of the correlation observed for the other derivatives, we would
*
Log P values for derivatives 6a-6g were calculated using Molinspiration
software (www.molinspiration.com).
5
−1
predict a binding constant K of value around 0.12 × 10 M
for de-
5 −1
with OLN (canonical B-DNA helix) as a receptor (Fig. 9B and C). While
it might be expected that the derivatives would show an affinity for
AATT sequence over GC sequences, compounds 6f and 6g did not form
any specific highly populated binding cluster within the AATT se-
quence. In contrast, the docking simulation of Hoechst 33258 shows a
huge populated cluster which resembles a binding mode nearly iden-
tical to that shown in the x-ray complex. This outcome might be as-
cribed to the high flexibility of the ligand structure permitting a wide
spectrum of possible binding interactions between ligands and oligo-
nucleotides.
rivative 6f, but the experimentally determined value of 0.87 × 10 M
is more than seven times higher than expected. In order to find an
explanation for such a large discrepancy, we decided to employ
methods of computational chemistry at the molecular mechanic (MD)
level of theory, using an Amber force field to perform a docking-like
study with MD simulations.
A docking pose of the 6f–OLG complex was used to provide a model
for the MD study (Fig. 9B). The phenylbutyl chain was shortened to
build up DNA interaction models for derivatives 6a, 6d & 6e. Each MD
simulation consisted of three steps within a Generalized Born implicit
−
5
−5
Fig. 7. Emission fluorescence spectra EB bound to DNA (1.65 × 10 M) (upper line) in the presence of acridine derivatives 6a and 6g (4.97 × 10 M)
−
4
(
c = 0–4.6 × 10 M) in a 0.01 M Tris-HCl buffer (pH 7.4, 24 °C). The arrow indicates the changes in fluorescence at increasing concentrations of samples.
492

L. Janovec et al.
Bioorganic Chemistry 83 (2019) 487–499
−
4
−6
Fig. 8. Circular dichroism spectra of ctDNA (7.66 × 10 M) in the absence and presence of derivatives 6a and 6g (8.26 × 10 M) in a 0.01 M Tris-HCl buffer (pH
7
.4, 24 °C). DNA free solutions are shown with black lines and DNA solutions in the presence of the studied derivatives are shown in colored lines.
Gobs = Gconf + Gt+r + Ghyd + Gpe + Gmol,
(1)
solvent model using NAMD software. The initial step was the mini-
mization of the ligand–OLG complex, followed by an MD run to elim-
inate any resulting steric distortion. This relaxed structure was then
subjected to minimization and the obtained complex geometry was
used for the ligand–oligonucleotide interaction studies (Fig. 11).
In order to analyze the relationship between the structure of the
derivatives and their Gibbs' free energies, we used Eq. (1) which con-
sists of energy terms associated with the partial energy changes in the
complex formation [28,29]. Definitions of the free energy terms are
provided in Supplementary Materials.
The formation of the ligand–OLG complex was examined in terms of
the binding energy change ΔEbind, the interaction energy change ΔEinter
and the energy change of the hydrophobic transfer ΔGhyd. The approx-
imate value of ΔGhyd was derived from the change in heat capacity ΔC
p
using the calculation of ΔGhyd = 80( ± 10).ΔC . The change in thermal
p
capacity can be estimated from the change in the polar and nonpolar
surfaces of OLG, ligand and complex during interactions based on the
linear correlation ΔC
p
= 0.382( ± 0.026).Anp – 0.121( ± 0.077).A .
p
Definitions of individual components are provided in Supplementary
Materials.
Fig. 9. Docking simulations of putative binding poses of Hoechst 33258 (A), derivative 6f (B) and derivative 6g (C) with 5′-d(CGCGAATTCGCG)-3′ (canonical B-DNA
helix) as the receptor. Simulation performed using Autodock software. Base pairs of the oligonucleotide are depicted as a 1-letter code only and backbones are
presented in a red-blue ribbon style. Nonpolar ligandś hydrogens are omitted for clarity. Colors of structures: hydrogen - white, oxygen - red, nitrogen – blue, carbon –
grey. Image prepared using Chimera software [25].
493

L. Janovec et al.
Bioorganic Chemistry 83 (2019) 487–499
Table 3
Summary of changes of solvent accessible surface areas (SASAs), heat capacity
(
ΔC
p
), hydrophobic free energy (ΔGhyd).
b
b
c
d
Ligand–OLG complex
ΔA
p
ΔAnp
ΔC
p
ΔGhyd
(
Å)
(Å)
(Cal/mol. K)
(Kcal/mol)
a
6
6
6
6
a
−175.8
−185.9
−164.1
−167.1
−269.1
−152.6
−584.4
−642.1
−665.6
−698.2
−568.1
−558.1
−202.0
−222.8
−234.4
−246.5
−184.5
−194.7
−16.2
−17.8
−18.8
−19.7
−14.8
−15.6
a
a
d
e
a
f
*
+
Hoechst 33258
Hoechst 33258
a
b
Putative mode of interaction with 5′-d(CGCGAATTCGCG)-3′ (OLG).
Changes of nonpolar (np) and polar (p) SASAs per intercalation calculated
using equations ΔAnp = Anp(complex) - [Anp(free DNA) + Anp(free ligand)] and
ΔA
p
= A
p
(complex) - [A
p
(free DNA) + A
p
(free ligand)], resp., from data shown
0.382( ± 0.026).ΔAnp -
in Tables S2, S3.
c
Calculated using the equation ΔC
p
=
Fig. 10. Relationship of lipophility (Log P) and ctDNA binding constants for
0.121( ± 0.077).ΔA
p
.
5
−1
d
derivatives 6a & 6d–6f (K/10 M ) obtained from spectrophotometry.
Calculated from ΔGhyd = 80( ± 10).ΔC .
p
*
Coordinates obtained from x-ray structure (PDB ID: 127D).
Coordinates obtained from MD simulation run.
+
Coordinates from MD simulations for ligand–OLG complexes and
their related components were used to calculate ΔCp from changes in
SASA (Table 3). In order to verify the accuracy of this approach, ΔCp
values for the Hoechst 33258–OLG complex were calculated using x-ray
coordinates (PDB ID: 127D) and compared with ΔCp values obtained
from the MD simulation coordinates of the same complex (Table 3). The
change in thermal enthalpy calculated from the MD coordinates is
Table 4
Summary of changes in electrostatic internal energy (ΔEelect), van der Waals
internal energy (ΔEvdw), internal conformation energy (ΔEconf), internal non-
bonded energy (ΔEnb) and binding energy (ΔEbind) for the DNA–ligand complex
obtained from MD simulations.
b
c
d
ΔEnbe
ΔEbindf
Ligand–OLG
ΔEelect
ΔEvdw
ΔEconf
−
1
−1
−1
−
194.7 cal.mol .K
and is in good agreement with the value of
obtained from the x-ray coordinates. It should
a
complex
(Kcal/mol) (Kcal/mol) (Kcal/mol) (Kcal/mol) (Kcal/mol)
−
1
−
184.5 cal.mol .K
a
be noted that ΔCp values of −276 ( ± 38) or −259 ( ± 36) cal.-
6a
−0.3
−0.4
−0.3
−0.4
−2.3
−58.6
−61.9
−62.8
−69.5
−59.5
24.5
24.3
23.3
29.3
24.8
−58.9
−62.3
−63.1
−69.9
−61.8
−34.4
−38.0
−39.8
−40.6
−37.0
a
a
−
1
−1
6d
mol .K
were calculated for the Hoechst 33258–d(CGCA3T3GCG)
2
6
6
e
complex depending on used x-ray structure. The ΔCp value determined
a
f
−
1
−1
using microcalorimetry was -300 ( ± 50) cal.mol .K
.
Hoechst
+
The binding energy ΔEbind of the complex was determined as a gab
33258
of the complex internal energy EComplex and its components, OLG (EOLG
)
a
b
Putative mode of interaction.
and ligand (ELigand) (Table 4). The interaction energy ΔEInter of the
complex was calculated using the NAMD module in VMD software as
the sum of electrostatic and non-binding interactions (Table 5).
From the values shown in Fig. 12, there is a clear decreasing trend of
ΔGhyd that correlates with the increasing proportion of the ligands’ non-
polar SASA which interacts with the DNA upon ligand binding
Obtained using the equation ΔEelect = Eelect (complex) – [Eelect (free OLG) +
E
E
E
elect (free ligand)] from data shown in Table S4.
c
Obtained using the equation ΔEvdw = Evdw (complex) – [Evdw (free OLG) +
vdw (free ligand)] from data shown in Table S4.
d
Obtained using the equation ΔEconf = Econf (complex) – [Econf (free OLG) +
conf (free ligand)] from data shown in Table S4.
e
(
Table 3).
Obtained using the equation ΔE = E (complex) – [Enb (free OLG) + Enb
nb
nb
A similar trend can be observed for interaction energy ΔEinter, pos-
(free ligand)] from data shown in Table S4.
f
sibly arising from the increase in van der Waals non-bonding interac-
tions due to the extension of the phenylalkyl chain (Fig. 13B, Table 5).
An analogous correlation is also evident for the complex binding
Obtained using the equation ΔEbind = Ebind (complex) – [Ebind (free OLG) +
E
bind (free ligand)] from data shown in Table S4.
+
Coordinates obtained from MD simulation run.
Fig. 11. Putative binding mode of derivatives 6a, 6d, 6e & 6f within the complex with 5′-d(CGCGAATTCGCG)-3′ (OLG). The complexes are superimposed on the
structure of the A ring. Color map for structures: red – 6a–OLG complex; green – 6d–OLG complex; magenta – 6e–OLG complex; blue – 6f–OLG complex. Color map
for atoms: red – oxygen; gray – carbon; blue – nitrogen; yellow – phosphorus. Image prepared using Chimera software [25].
494

L. Janovec et al.
Bioorganic Chemistry 83 (2019) 487–499
Table 5
proximity of values (Table 4, Table 5) and thus it can be said that
Summary of changes in electrostatic interaction energy (ΔEelect), van der Waals
interaction energy (ΔEvdw) and total interaction energy (ΔEInter.) for the complex
DNA–ligand obtained from MD simulations.
ΔGmol ≅ ΔEnb ≅ ΔEvdw.
For derivatives 6a, 6d and 6e the decrease in ΔGhyd and ΔGmol is in
contrast to the increase in their ΔGobs values. As is shown in Table 4, the
ΔEconf values (formally identified with ΔGconf) are not sufficient to
overcome the decrease of energy terms ΔGhyd and ΔGmol that contribute
to the overall stability of the complex. A possible explanation of the
b
Ligand–OLG Complex
ΔEelect
ΔEvdw
ΔEinter
(
Kcal/mol)
(Kcal/mol)
(Kcal/mol)
a
6
6
6
6
a
−0,1
0,0
−55,0
−58,0
−61,4
−66,5
−68,6
−55,1
−58,1
−61,4
−66,5
−70,6
a
a
decrease in complex stability is provided by the energy term ΔGt+r,
d
e
0,0
which is directly linked with the loss of the degree of freedom accom-
a
f
0,0
panying complex formation. Since the ΔGt+r value can be calculated as
+
Hoechst 33258
−1,9
−
TΔS, the decrease of the complex stability in the order 6a, 6d, 6e can
be attributed to the increase of a loss of entropy ΔS owing to the
elongation of the derivatives’ alkyl chains.
a
Putative mode of interaction.
Obtained using an equation ΔEinter (complex) = Eelect (complex) + Evdw
complex).
Coordinates obtained from MD simulation run.
b
It is possible to conclude that the increase of hydrophobic surface
area and non-bonding interactions due to the elongation of alkyl chains
leads to decrease in ΔGhyd and ΔGmol. These energetic terms are over-
come by the increase of the loss of entropy ΔS as the consequence of the
loss of the degree of freedom upon complex formation.
(
+
Although derivative 6f is an extension of the series of derivatives 6a,
6
d & 6e, the conclusion discussed above does not explain the increase
in its K value. According to the proposed interaction model, a binding
constant
K
for the 6f–OLG complex would have a value of
5
−1
0
0
.12 × 10 M , but the constant K obtained in experiments is
5 −1
.87 × 10 M . Thus, the increase in the complex stability of deriva-
tive 6f cannot be linked to the structural similarity within the bound
ligands. One possible explanation could be found in the general ther-
modynamic effect during the formation of the complex. Therefore the
binding of derivative 6f into the DNA minor groove might lead to the
expulsion of site-specifically bound water or cations accompanied by a
large favorable entropy term. This may be the main driving force in the
reaction and could explain why the binding constant K for derivative 6f-
OLG complex differs from the trend observed for derivatives 6a, 6d &
6
e.
This explanation is in accordance with the Hoechst 33258–DNA
complex formation where the main driving force is the gain of entropy
as the consequence of water expulsion from a minor groove into the
bulk solvent. The explanation may also point to the fact that iso-
structural does not mean isoenergetic, and therefore a small change in
the alkyl chain might lead to a significant change in the binding ther-
modynamics.
Fig. 12. Relationship of ΔGhyd and number of methylene groups within the
structure of derivatives 6a 6d, 6e & 6f.
energy ΔEbind, the value of which decreases with increasing numbers of
methylene groups in derivatives 6a, 6d, 6e & 6f. The decrease of ΔEbind
is a result of the increase in non-bonding interactions (a decrease in
ΔEnb values) which are the driving force of the complex formation
(
Fig. 13A, Table 4). The main component of the non-bonding interac-
4. Conclusion
tion ΔEnb is the van der Waals interaction, ΔEvdw. When comparing the
energy term ΔEvdw with terms ΔEinter and ΔEbind, there is a clear
A series of new derivatives with a diphenylamino carboxylic acid
framework were synthetised and evaluated for their cytotoxic activities
Fig. 13. A: Relationship of ΔEbind and number of methylene groups within the structure of derivatives 6a, 6d, 6e, 6f. B: Relationship of ΔEinter and number of
methylene groups within the structure of derivatives 6a, 6d, 6e & 6f.
495

L. Janovec et al.
Bioorganic Chemistry 83 (2019) 487–499
and proliferation inhibition. The most active compounds in the cyto-
(416.47): 69.21% C, 5.81% H, 13.45% N; found: 69.40% C, 5.67% H,
toxicity tests were derivative 6g with an IC50 value of
13.60% N.
−
−
6
6
2
6
.5 ± 1.1 × 10
M and derivative 6f with an IC50 value of
.0 ± 3.0 × 10 M (L1210 cell line) after 48 h incubation. The se-
5.1.1.2. 4–(Acetylamino)–2–{[3–(acetylamino)phenyl]
lective cytotoxicity to leukemia L1210 cells was also confirmed, and
was found to be higher than that observed against HEK293T and NIH-
amino}–N–[(4–methylphenyl)methyl] benzamide (6b). Yield = 73%,
1
gray solid, mp = 145–147 °C. H NMR (400 MHz, DMSO–d
6
) σ, ppm
'
''
3
T3 cells. Cell cycle analysis revealed that a primary trend of the di-
10.00 (s, 1H, NH), 9.99 (s, 1H, Ph-NH-Ph), 9.89 (s, 1H, NH ), 8.94 (t,
phenylamine derivatives was to arrest the cells in the G1-phase of the
cell cycle within the first 24 h. UV–visible, fluorescence spectroscopy
and circular dichroism were used in order to study the binding mode of
the novel compounds with DNA. The binding constants determined by
UV–visible spectroscopy were found to be in the range of
1H, NH, J = 6.0 Hz), 7.68 (d, 1H, H-6, J = 8.4 Hz), 7.54 (d, 1H, H-3,
′
'
′' ′'
′'' ′''
J = 1.6 Hz), 7.41 (s, 1H, H-2 ), 7.22 – 7.20 (m, 4H, H-4 ,5 , H-2 ,6 ),
′
'' ′''
′'
7.15 – 7.13 (m, 3H, H-5, H-3 ,5 ), 6.89 – 6.86 (m, 1H, H-6 ), 4.40 (d,
′
''
′'
2H, CH2, J = 6.0 Hz), 2.27 (s, 3H, CH
3
), 2.03 (s, 3H, CH
3
'
), 2.02 (s, 3H,
′
CH
3
).
5
−1
13
0
.21–0.87 × 10 M . The results suggest that the compounds interact
C NMR (100 MHz, DMSO–d
6
) σ, ppm 168.6 (CO), 168.5 (CO),
'
'
′'
′'
with DNA through groove binding. Additional in silico DNA binding
studies with experimental findings revealed differences between iso-
structural and isoenergetic relationships in the series of studied deri-
vatives.
168.3 (CO ), 145.6 (C2), 142.6 (C4), 141.6 (C1 ), 140.4 (C3 ), 136.6
′'' ′'' ′' ′'' ′''
(C1 ), 135.8 (C4 ), 129.5 (C5 ), 129.5 (C6), 128.8 (C3 , C5 ), 127.2
′
''
′''
′'
′'
′'
(C2 , C6 ), 114.3 (C6 ), 112.8 (C4 ), 112.5 (C1), 110.7 (C2 ), 108.7
′
′'
′''
(C5), 104.2 (C3), 42.1 (CH
2
), 24.2 (CH
3
), 24.1 (CH
3
), 20.7 (CH ).
3
Anal. Calcd for C25
H
26
N
4
O (430.49): 69.75% C, 6.09% H, 13.01% N;
3
5
. Experimental
found: 70.01% C, 6.17% H, 12.83% N.
5
.1. Chemistry
5.1.1.3. 4–(Acetylamino)–2–{[3–(acetylamino)phenyl]
amino}–N–[(4–fluorophenyl)metyl] benzamide (6c). Yield = 68%, gray
1
All chemicals and reagents were reagent grade and were used
solid, mp = 148–150 °C. H NMR (400 MHz, DMSO–d
6
) σ, ppm 10.00
1
13
'
''
without further purification.
H
(400 MHz, 600 MHz) and
C
(s, 1H, NH), 9.98 (s, 1H, Ph-NH-Ph), 9.89 (s, 1H, NH ), 8.99 (t, 1H, NH,
(
100 MHz, 150 MHz) NMR spectra were measured on a Varian Mercury
J = 6.0 Hz), 7.68 (d, 1H, H-6, J = 8.4 Hz), 7.55 (d, 1H, H-3,
′
'
′'' ′''
Plus or a Varian VNMRS NMR spectrometers at room temperature in
J = 1.8 Hz), 7.42 (s, 1H, H-2 ), 7.38 – 7.34 (m, 2H, H-2 ,6 ), 7.22 –
′' ′' ′'' ′''
DMSO‑d
6
using TMS as an internal standard (0 ppm for both nuclei).
7.21 (m, 2H, H-4 ,5 ), 7.16 – 7.12 (m, 3H, H-5, H-3 ,5 ), 6.89 – 6.86 (m,
′
'
′'
Melting points were determined with a Koffler hot-stage apparatus and
are uncorrected. Elemental analyses were performed on a Perkin-Elmer
analyzer CHN 2400. Reactions were monitored with thin-layer chro-
matography (TLC) using Silufol plates with detection at 254 nm.
Preparative column chromatography was performed using a Kiesegel
Merck 60 column, type 9385 (grain size 250 nm). Chlorotoluene 2, acid
1H, H-6 ), 4.42 (d, 2H, CH2, J = 6.0 Hz), 2.03 (s, 3H, CH
3
'
), 2.01 (s, 3H,
′
1
CH
3
).
3
C NMR (100 MHz, DMSO–d
6
) σ, ppm 168.6 (CO), 168.5 (CO),
'
'
′''
168.3 (CO ), 161.1 (d, C4 , J = 240.0 Hz), 145.7 (C2), 142.7 (C4),
′
'
′'
′''
′'
141.6 (C1 ), 140.4 (C3 ), 135.8 (d, C1 , J = 3.0 Hz), 129.5 (C6, C5 ),
′
''
′''
′''
′''
129.2 (d, C2 ,C6 , J = 7.5 Hz), 115.0 (d, C3 ,C5 , J = 21.0 Hz), 114.4
′
'
′'
′'
3
, 5 and products 6a–6f were prepared according to the published
(C6 ), 112.8 (C4 ), 112.3 (C1), 110.7 (C2 ), 108.7 (C5), 104.2 (C3), 41.7
′ ′'
procedures [7]. Compounds 2, 3 and 5 are described in Supplementary
(CH
2
), 24.2 (CH
3
), 24.1 (CH
3
). Anal. Calcd for C24
H23FN
4
O (434.46):
3
Material.
66.35% C, 5.34% H, 12.90% N; found: 66.40% C, 5.19% H, 12.69% N.
5
.1.1. General synthesis of 4–(acetylamino)–2–{[3–(acetylamino)phenyl]
5.1.1.4. 4–(Acetylamino)–2–{[3–(acetylamino)phenyl]
amino}–N–substituted benzamide 6a–6 g
amino}–N–[2–(phenyletyl)]benzamide (6d). Yield = 66%, gray solid,
1
A mixture of derivate 5 (1 g, 3.05 mmol) and carbonyldiimidazole
mp = 113–115 °C. H NMR (400 MHz, DMSO–d
6
) σ, ppm 9.98 (s, 1H,
'
''
(
1.5 g, 9.15 mmol) in dimethylformamide (10 mL) was mixed at room
NH), 9.92 (s, 1H, Ph-NH-Ph), 9.89 (s, 1H, NH ), 8.51 (t, 1H, NH,
temperature. After the reaction was finished (approximately 2 h, TLC
methanol - ethyl acetate (6:1 v/v)), appropriate amine (9.51 mmol) was
added. The reaction mixture was stirred at room temperature for an
additional 24 h and subsequently poured into distilled water.
Precipitated crude products 6a–6 g were filtered off and dried over-
night. Derivatives 6a–6 g were purified by column chromatography
using a mobile phase methanol – ethyl acetate (1:6 v/v) and crystallized
from ethanol. Numbering of atoms for 6a–6g in NMR assignments
corresponds to the numbering of the final product 6f and is shown in Fig
S5.
J = 5.4 Hz), 7.55 (d, 1H, H-6, J = 8.4 Hz), 7.52 (s, 1H, H-3), 7.41 (s,
′
'
′'' ′''
′'' ′''
1H, H-2 ), 7.31 – 7.27 (m, 2H, H-3 ,5 ), 7.25 – 7.23 (m, 2H, H-2 ,6 ),
′
'
′'
′''
7.22 – 7.17 (m, 3H, H-4 ,5 , H-4 ), 7.12 (d, 1H, H-5, J = 8.4 Hz), 6.87 –
′
'
6.85 (m, 1H, H-6 ), 3.47 – 3.43 (dd, 2H, CH
2
–1), 2.83 (t, 2H, CH
2
–2
,
′
'
′
J = 7.2 Hz), 2.03 (s, 3H, CH
3
), 2.01 (s, 3H, CH
3
).
1
3
'
C NMR (100 MHz, DMSO–d
6
) σ, ppm 168.6 (CO), 168.5 (CO),
'
'
′' ′'
′''
′'
168.3 (CO ), 145.4 (C2), 142.5 (C4), 141.7 (C1 ), 140.4 (C3 ), 139.5
′
''
′'
′''
′''
′''
(C1 ), 129.5 (C5 ), 129.3 (C6), 128.7 (C2 , C6 ), 128.3 (C3 , C5 ),
′
''
′'
′'
126.1 (C4 ), 114.2 (C6 ), 113.0 (C1), 112.7 (C4 ), 110.6 (C2 ), 108.7
′
′'
(C5), 104.4 (C3), 40.7 (CH
2
–1), 35.1 (CH
2
–2), 24.2 (CH
3
), 24.1 (CH ).
3
Anal. Calcd for C25
H
26
N
4
3
O (430.49): 69.75% C, 6.09% H, 13.01% N;
5
.1.1.1. 4–(Acetylamino)–2–{[3–(acetylamino)phenyl]
found: 69.81% C, 6.19% H, 13.39% N.
amino}–N–benzylbenzamide (6a). Yield: 60%, gray solid, mp:
1
2
14–216 °C. H NMR (400 MHz, DMSO–d
6
) σ, ppm 10.00 (s, 1H,
5.1.1.5. 4–(Acetylamino)–2–{[3–(acetylamino)phenyl]
'
''
NH), 10.00 (s, 1H, Ph-NH-Ph), 9.89 (s, 1H, NH ), 8.98 (t, 1H, NH,
amino}–N–[3–(phenylpropyl)]benzamide (6e). Yield = 71%, gray solid,
1
J = 6.0 Hz), 7.69 (d, 1H, H-6, J = 8.4 Hz), 7.54 (d, 1H, H-3,
mp = 123–125 °C. H NMR (400 MHz, DMSO–d
6
) σ, ppm 9.98 (s, 1H,
′
'
′'' ′''
′'' ′''
'
''
J = 1.6 Hz), 7.41 (s, 1H, H-2 ), 7.33 – 7.32 (m, 4H, H-2 ,6 , H-3 ,5 ),
NH), 9.92 (s, 1H, Ph-NH-Ph), 9.88 (s, 1H, NH ), 8.43 (t, 1H, NH,
′
''
′' ′'
7
.25 – 7.21 (m, 3H, H-4 , H-4 ,5 ), 7.13 (dd, 1H, H-5, J = 8.4 Hz,
J = 6.0 Hz), 7.60 (d, 1H, H-6, J = 8.4 Hz), 7.53 (s, 1H, H-3,
′
'
′'
′'' ′''
J = 1.6 Hz), 6.89 – 6.86 (m, 1H, H-6 ), 4.45 (d, 2H, CH2, J = 6.0 Hz),
J = 1.6 Hz), 7.40 (s, 1H, H-2 ), 7.28 – 7.23 (m, 2H, H-3 ,5 ), 7.23 –
′'' ′'' ′' ′' ′''
′
′'
2
.03 (s, 3H, CH
3
), 2.01 (s, 3H, CH
3
).
7.20 (m, 4H, H-2 ,6 , H-4 ,5 ), 7.18 – 7.16 (m, 1H, H-4 ), 7.13 (d, 1H,
1
3
'
′'
C NMR (100 MHz, DMSO–d ) σ, ppm 168.6 (C]O), 168.5 (C]O),
6
H-5, J = 8.4 Hz), 7.29 – 7.11 (m, 8H), 6.88 – 6.85 (m, 1H, H-6 ), 3.25
'
'
′'
′'
′'
1
68.3 (C]O ), 145.6 (C2), 142.6 (C4), 141.6 (C1 ), 140.4 (C3 ), 139.6
(m, 2H, CH
2
–1), 2.63 (t, 2H, CH
2
–3, J = 7.2 Hz), 2.03 (s, 3H, CH
3
),
′
''
′'
′''
′''
′''
′''
′''
′
(
C1 ), 129.5 (C6, C5 ), 128.3 (C3 , C5 ), 127.2 (C2 , C6 ), 126.7 (C4 ),
2.01 (s, 3H, CH
3
), 1.86 – 1.78 (m, 2H, CH
2
–2).
′
'
′'
′'
13
'
1
14.3 (C6 ), 112.8 (C4 ), 112.4 (C1), 110.7 (C2 ), 108.6 (C5), 104.2
C NMR (100 MHz, DMSO–d
6
) σ, ppm 168.6 (CO), 168.5 (CO),
′
'
′
''
′'
′''
(
C3), 42.3 (CH
2
), 24.1 (CH
3
), 24.1 (CH
3
). Anal. Calcd for C24
H
24
N
4
O
3
168.3 (CO ), 145.3 (C2), 142.4 (C4), 141.8 (C1 ), 141.7 (C1 ), 140.4
496

L. Janovec et al.
Bioorganic Chemistry 83 (2019) 487–499
′
′
'
''
′'
′''
′''
′''
′''
(
C3 ), 129.5 (C5 ), 129.4 (C6), 128.3 (C2 ,C6 ), 128.3 (C3 ,C5 ), 125.7
streptomycin (100 µg/mL). The cells were maintained at 37 °C in a
′'
′'
′'
(
C4 ), 114.1 (C6 ), 113.1 (C1), 112.6 (C4 ), 110.5 (C2 ), 108.7 (C5),
humidified 5% CO
2
atmosphere. NIH-3T3 and HEK293 cells were
′
1
2
1
04.4 (C3), 38.7 (CH
2
–1), 32.7 (CH
2
4
–3), 30.7 (CH
2
–2), 24.1 (CH
3
),
harvested using a trypsin solution (0.25% trypsin/EDTA). The L1210
and NIH-3T3 cells were obtained from Dr. Ujhazy, Roswell Park Cancer
Institute, Buffalo and the human embryonic kidney cells HEK293T from
ATCC, Manassas, VA, USA.
′
'
4.1 (CH
3
). Anal. Calcd for C26
H
28
N
O (444.52): 70.25% C, 6.35% H,
3
2.60% N; found: 70.03% C, 6.44% H, 12.25% N.
5
.1.1.6. 4–(Acetylamino)–2–{[3–(acetylamino)phenyl]
amino}–N–[4–(phenylbutyl)]benzamide (6f). Yield = 72%, gray solid,
5.2.2. MTT assay
1
mp = 214–216 °C. H NMR (400 MHz, DMSO–d
6
) σ, ppm 9.98 (s, 1H,
Cell viability was determined using the MTT microculture tetra-
'
''
NH), 9.97 (s, 1H, Ph-NH-Ph), 9.89 (s, 1H, NH ), 8.41 (t, 1H, NH,
zolium assay method which has been described previously [30]. Cells
4
J = 6.0 Hz), 7.58 (d, 1H, H-6, J = 8.4 Hz), 7.53 (d, 1H, H-3,
(25 × 10 /mL)
were
incubated
with
derivatives
6a–6g
′
'
′'' ′''
−6
J = 1.8 Hz), 7.41 (s, 1H, H-2 ), 7.28 – 7.23 (m, 2H, H-3 ,5 ), 7.21 –
(0–50 × 10 M) for 48 h and the viability of cells was determined.
Absorbance was then measured at 570 and 630 nm as a background
using a microplate reader (Synergy H1 Hybrid). The IC50 value was
determined from the dose-response curves of every cell line.
′
'' ′''
′' ′'
′''
7
.17 (m, 4H, H-2 ,6 , H-4 ,5 ), 7.16 – 7.14 (m, 1H, H-4 ), 7.12 (dd, 1H,
′'
H-5, J
1
= 8.4, J
2
= 1.8), 6.87 – 6.85 (m, 1H, H-6 ), 3.25 (dd, 2H,
CH
CH
2
3
–1, J
1
= 7.2, J
2
= 1.2), 2.60 (t, 2H, CH
2
–4, J = 7.2), 2.03 (s, 3H,
′
'
′
), 2.01 (s, 3H, CH
–3).
C NMR (100 MHz, DMSO–d
3
), 1.60 – 1.57 (m, 2H, CH
2
–2), 1.54 – 1.52 (m,
2
1
H, CH
2
5.2.3. Trypan blue exclusion test
1
3
'
6
) σ, ppm 168.6 (CO), 168.5 (CO),
Cell proliferation and the cytotoxic potential of compounds were
determined using a trypan blue dye exclusion test. Cells were seeded
'
'
′''
′'
′''
68.3 (CO ), 145.3 (C2), 142.4 (C4), 142.2 (C1 ), 141.7 (C1 ), 140.4
′
'
′'
′''
′''
′''
6
−6
(
C3 ), 129.5 (C5 ), 129.3 (C6), 128.3 (C2 ,C6 ), 128.2 (C3 ,C5 ), 125.6
(0.15 × 10 cells/mL) in Petri dishes and derivatives 6a and 6f
′
''
′'
′'
′'
(
C4 ), 114.1 (C6 ), 113.1 (C1), 112.6 (C4 ), 110.4 (C2 ), 108.6 (C5),
(0–20 × 10 M) were added after 24 h. Cell proliferation was then
1
04.4 (C3), 38.7 (CH
2
–1), 34.8 (CH
2
–4), 28.7 (CH
2
–2), 28.5 (CH
2
–3),
checked after 24 h.
′
′'
2
4.1 (CH
3
), 24.1 (CH
3
). Anal. Calcd for C27
H
30
N
4 3
O
(458.57): 70.72%
C, 6.59% H, 12.22% N; found: 70.43% C, 6.51% H, 12.17% N.
5.2.4. LDH leakage assay
A lactate dehydrogenase assay (LDH) was performed using the
6
5
.1.1.7. 4–(Acetylamino)–2–{[3–(acetylamino)phenyl]
method described by Grivell and Berry [31]. The cells (0.15 × 10
−
6
amino}–N–piperonylbenzamide
(6 g). Yield = 66%,
gray
solid,
cells/mL) were treated with derivatives 6a–6f (0–20 × 10 M) and
1
−6
mp = 220–222 °C. H NMR (400 MHz, DMSO–d
6
) σ, ppm 9.99 (s, 1H,
after 24 h, 100 × 10 L of the sample from the growth medium of
'
''
−3
−3
NH), 9.98 (s, 1H, Ph-NH-Ph), 9.89 (s, 1H, NH ), 8.91 (t, 1H, NH,
L1210 cells was added to a 1 × 10 L cuvette containing 0.9 × 10
L
−
3
J = 6.0 Hz), 7.66 (d, 1H, H-6, J = 8.4 Hz), 7.54 (d, 1H, H-3,
of a reaction mixture to yield a final concentration of 1 × 10
M
′
'
′'
′'
−3
−3
J = 1.8 Hz), 7.41 (s, 1H, H-2 ), 7.22 – 7.21 (m, 2H, H-4 , 5 ), 7.13
pyruvate, 0.15 × 10 M NADH, and 100 × 10 M phosphate buffer
(pH 7.4). The incubation mixture was mixed thoroughly and the ab-
sorbance was measured at 366 nm for 3 min. The activity of LDH was
also measured in cell lysate.
′
''
(
dd, 1H, H-5, J = 8.4, 1.8 Hz), 6.89 (d, 1H, H-4 , J = 1.8 Hz), 6.88 –
′
'
′''
′''
6
.86 (m, 1H, H-6 ), 6.85 (d, 1H, H-7 , J = 7.8 Hz), 6.80 (dd, 1H, H-6 ,
′''
J
1
= 7.8,
J
2
= 1.8 Hz), 5.97 (s, 2H, CH
2
), 4.34 (d, 2H, CH
2
,
′
'
′
J = 6.0 Hz), 2.03 (s, 3H, CH
3
), 2.01 (s, 3H, CH
3
).
1
3
'
C NMR (100 MHz, DMSO–d
6
) σ, ppm 168.6 (CO), 168.4 (CO),
5.2.5. Cytomorphological changes
'
'
'''
'''
5
1
68.3 (CO ), 147.2 (C3a ), 146.0 (C7a ), 145.6 (C2), 142.7 (C4), 141.6
Cells (1.5 × 10 /mL) were treated with derivatives 6a and 6f
′
'
′'
′''
′'
′''
−6
(
C1 ), 140.4 (C3 ), 133.5 (C5 ), 129.5 (C6,C5 ), 120.5 (C6 ), 114.3
(0–40 × 10 M) and incubated at 37 °C for 24 h. After incubation, the
′
'
′'
′'
′''
(
C6 ), 112.8 (C4 ), 112.4 (C1), 110.7 (C2 ), 108.7 (C5), 108.0 (C7 ),
cells were washed with PBS and stained with Hoechst 33,342 for the
′
''
′''
′
−6
1
07.9 (C4 ), 104.2 (C3), 100.8 (CH
2
5
), 42.2 (CH
2
), 24.2 (CH
3
), 24.1
visualization of the cell nuclei (10 × 10 M, 10 min) and with propi-
′
'
(
CH
3
). Anal. Calcd for C25
H
24
4
N O
(460.48): 65.21% C, 5.25% H,
dium iodide (PI, 1 mg/mL, 10 min) to visualize the damaged membrane
integrity of the cells. Changes in cell morphology and damage to plasma
membrane integrity (PI positive cells) were observed using an Axio
ANO Imager A1 optical and fluorescence microscope (Zeiss, Germany).
1
2.17% N; found: 65.10% C, 5.41% H, 12.10% N.
5.2. Biology
The studied derivatives were dissolved in DMSO to a concentration
5.2.6. Cell cycle analysis
−
3
6
of 4.49 × 10 M. Calf thymus sodium salt was purchased from Sigma
Aldrich chemicals (Sigma Aldrich, USA). Calf thymus DNA (ctDNA)
stock solution was prepared by dissolving 1 mg/mL in a Tris-EDTA
Cells (0.5 × 10 /mL) were washed with PBS after incubation
−6
(6–24 h) with derivative 6a (1, 5 and 10 × 10 M) or derivative 6f (1,
−6
2.5 and 5 × 10 M). The cells were subsequently fixed with ethanol
and stored at -20 °C. Prior to measurements, RNase (0.2 mg/mL) was
added for 20 min at 37 °C. Afterwards, the cells were stained with PI
(0.05 mg/mL) and analyzed using an Accuri C6 flow cytometer.
−
2
−3
buffer (TE buffer contains 1 × 10 M Tris-HCl pH 7.4 and 1 × 10
M
EDTA) at 4 °C overnight. The final concentration of ctDNA was identi-
fied spectrophotometrically at 260 nm. The purity of ctDNA was mea-
sured as a ratio of the absorption at 260/280 nm and was found to be
1
.82, a value which indicates that the ctDNA sample was pure and
5.2.7. Intracellular distribution of derivatives 6a and 6f – Colocalization
devoid of protein contamination. All other chemicals and reagents were
purchased at reagent grade and were used without further purification.
Ethidium bromide, Triton X-100 and dimethyl sulfoxide (DMSO) were
obtained from Sigma-Aldrich Chemie (Germany). EDTA, RNase A and
proteinase K were purchased from Serva (Germany). All other chemi-
cals were purchased from Lachema (Czech Republic).
with SytoRed
6
−6
L1210 cells (0.5 × 10 ) were treated with 40 × 10 M of deriva-
tives 6a and 6f (37 °C) for 2 h and then the samples were labeled with
−6
SytoRed (0.1 × 10 M, 35 min) in order to visualize the cell nuclei.
After incubation with SytoRed, the cells were washed twice with PBS
and immediately visualized using an Axio ANO Imager A1 fluorescence
microscope (Zeiss, Germany).
5
.2.1. Cell culture conditions
Mouse leukemia cell lines L1210 were grown in a RPMI-1640
5.2.8. UV–visible absorption measurements
medium and mouse embryonic fibroblast cell line NIH-3T3 and human
embryonic kidney 293 cells HEK293T in DMEM medium. Both media
were supplemented with 10% FBS, penicillin (100 units/mL), and
The UV–Vis absorption spectra (200 – 450 nm) of a fixed con-
−5
centration of the studied derivatives 6a–6g (4.95 × 10 M) were ob-
tained using a Varian Vary 100 UV–visible spectrophotometer (in
497

L. Janovec et al.
Bioorganic Chemistry 83 (2019) 487–499
a quartz cuvette, 1 cm path length) in a 0.01 M Tris-HCl buffer (pH 7.4)
at laboratory temperature. Gradually increasing concentrations of
ctDNA were added to fixed concentrations of derivatives 6a–6g [16].
Data from the spectrophotometric titrations were used for the calcula-
tion of the binding constants K for derivative-DNA complexes using the
McGhee and von Hippel equation (Fig. S20) [32].
points × 126 points (x,y,z) and a spacing of 0.375 Ǻ.
Docking runs were performed using a Larmarckian genetic algo-
rithm. Docking began with a population of random ligand conforma-
tions in a random orientation and at a random translation. Each
docking experiment was derived from 100 different runs which were set
6
to terminate after a maximum of 5 × 10 energy evaluations or
3
2
7 × 10 generations, yielding 100 docked conformations. The popu-
5
.2.9. Fluorescence measurements
lation size was set to 150. For other parameters, the default values were
used.
Fluorescence spectra were recorded at room temperature at a range
of 500–800 nm with an excitation wavelength at 500 nm. The widths of
both the excitation and emission slits were set at 10 nm. Fluorescence
intensities were expressed in arbitrary units. The interactions of deri-
5.4. Molecular dynamics run – A docking-like study
−
5
vatives 6a, 6c, 6e and 6g (4.97 × 10 M) and 6d and 6f
Molecular models of the ligand–5′-d(CGCGAATTCGCG)-3′ complex
for derivatives 6a, 6d & 6e were prepared using the coordinates of the
6f–5′-d(CGCGAATTCGCG)-3′ complex obtained as a result of the
docking simulation. Chimera software was used to rebuild ligand 6f
into ligands 6a, 6d & 6e by truncating its phenylbutyl chain leaving the
rest of the structure unchanged. All calculations were carried out in
NAMD 2.8 using a generalized born implicit solvent for waters,
parm99.dat parameters set for nucleic acids, and GAFF atom types for
ligands [36–40]. ANTECHAMBER and XLEAP modules as a part of
AMBERTOOLS 1.3 software package were applied to extrapolate
missing ligand force-field parameters and to derive charges using AM1-
BCC method.
−
5
(
3.98 × 10 M) with the ctDNA-EB complex were studied by adding
specific quantities of the compound into the quartz cuvette containing a
fixed concentration of the ctDNA-EB complex solution [13]. The con-
−
5
centration of EB was 6.0 × 10 M and that of the ctDNA was
−
5
1
.65 × 10 M. The influence of the addition of each compound to the
DNA-EB complex solution was observed by recording variations in the
fluorescence emission spectra. All measurements were performed at
2
4 °C.
5
.2.10. Circular dichroism
CD spectra were recorded on a Jasco J-810 spectropolarimeter in
mm quartz cuvettes and are the mean result of three scans from which
1
Each complex and its components were then subjected to 3000 steps
of conjugate gradient minimization followed by 6 ps MD run. Finaly
3000 steps of conjugate gradient minimization were done. Energy in-
formation, averages and coordinates were recorded every 500 steps and
non-bonded list was updated every 10 steps. Coordinates from MD run
were employed as input data for SASAs calculations (see Chapter
the buffer background had been electronically subtracted. All mea-
surements were performed in a 0.01 M Tris-HCl buffer (pH 7.4). The
−
4
concentration of ctDNA was 7.66 × 10 M and the concentration of
−
6
derivatives 6a–6g was 8.26 × 10 M.
5
.4.1.). Final energy terms after the final minimization are shown in
5.3. Docking studies
Table S4.
5
.3.1. A ligand preparation
5
.4.1. Calculation of solvent-accessible surface areas (SASAs).
Molecular models of the Hoechst 33,258 and derivatives 6f and 6g
Solvent-accessible surface areas were determined using the final
were computer-built using the building options in ACD/ChemSketch
coordinates from molecular dynamics simulations for free 5′-d(CGCG
AATTCGCG)-3′, free ligands 6a, 6d, 6e & 6f and for the ligand–5′-d(
CGCGAATTCGCG)-3′ complexes using Tcl module ver. 8.5.6/Tk ver.
[
33]. The models were built as 3D structures and saved as mopac input
files using the ACD/3D Viewer [34]. MOPAC2016 was used to optimize
ligand geometry [23].
8
.5.6 of VMD 1.9.3 (Table S2, S3) [41]. Surfaces of carbon-bound hy-
drogens and all carbons were classified as nonpolar and those of re-
maining hydrophilic atoms were defined as polar for ligands. Surfaces
of carbon-bound hydrogens and all carbons were classified as nonpolar
and those of the remaining hydrophilic atoms were defined as polar for
oligonucleotide. Surfaces were generated from coordinates with 1.4 Å
surrounding radius. The change of the solvent-accessible surface area
on binding, ΔAtotal, is the difference between the area of the complex
and the summed surface areas of the free DNA duplex and the free
ligand:
5
.3.2. A receptor preparation
Chimera software was used to extract the coordinates of a nucleo-
tide from the x-ray of the Hoechst 33258/5′-d(CGCGAATTCGCG)-3′
ternary complex (Id PDB: 127D). The structure of 5′-d(CGCGAATTC
GCG)-3′ in the canonical B-DNA helix was built using the building
options included in Chimera software [25–27].
5
.3.3. A charge assignment
MGL TOOLS 1.5.6 was used to assign Gasteiger partial atomic
charges for ligands and for nuleotides [35]. MOPAC2016 was used to
add PM7 partial atomic charges for ligands [23]. Chimera software was
used to assign AMBER ff14SB partial atomic charges for the oligonu-
cleotide [25–27].
Atotal = Anp + Ap;
Atotal = Anp + Ap,
where Anp and A
p
represent surface contributions from nonpolar and
polar atoms, respectively. The binding-induced alterations in compo-
nent SASAs terms on forming the DNA-ligand complex are given by the
equations:
5
.3.4. The root mean square deviation calculations
Anp = Anp (complex)
[Anp (free DNA) + Anp (free ligand)],
Chimera software was employed to calculate RMSD [25–27].
Ap = Ap (complex)
[Ap (free DNA) + Ap (free ligand)].
5
.3.5. Docking run
Docking simulations were carried out using Autodock ver. 4.2,
while MGL TOOLS 1.5.6 was used to prepare the input files [22–27,35].
United atom representations were used for the ligand and DNA. The
grid for energy for 5′-d(CGCGAATTCGCG)-3′ in the canonical B-DNA
helix was set at the center of macromolecule with dimensions of 80
Acknowledgements
This study was supported by VEGA grant No. 1/0016/18, MH CZ-
DRO (UHHK, 00179906) and the State NMR Program (grant No.
2003SP200280203). Molecular graphics images were produced using
UCSF Chimera package from the Resource for Biocomputing,
Visualization, and Informatics at the University of California, San
points × 120 points × 80 points (x,y,z) and a spacing of 0.375 Ǻ. The
grid for energy for 5′-d(CGCGAATTCGCG)-3′ (Id PDB: 127D) was set at
the center of macromolecule with dimensions of 80 points × 80
498

L. Janovec et al.
Bioorganic Chemistry 83 (2019) 487–499
Francisco (supported by NIH P41 RR-01081).
Declaration of interest
topoisomerase I and II catalytic inhibitors, Bioorg. Chem. 59 (2015) 168–176,
https://doi.org/10.1016/j.bioorg.2015.03.002.
[
[
16] I. Ahmad, A. Ahmad, M. Ahmad, Binding properties of pendimethalin herbicide to
DNA: multispectroscopic and molecular docking approaches, Chem. Phys. 18
(
2016) 6476–6485, https://doi.org/10.1039/c5cp07351k.
17] J. Janočková, E. Žilecká, J. Kašpárková, V. Brabec, O. Soukup, K. Kuca,
M. Kožurková, Assessment of DNA-binding affinity of cholinesterase reactivators
and electrophoretic determination of their effect on topoisomerase I and II activity,
Mol. Biosyst. 12 (2016) 2910–2920, https://doi.org/10.1039/C6MB00332J.
18] O.M. Salem, M. Vilková, J. Janočková, R. Jendželovský, P. Fedoročko, E. Žilecká,
J. Kašpárková, V. Brabec, J. Imrich, M. Kožurková, New spiro tria(thia)zolidine-
acridines as topoisomerase inhibitors, DNA binders and cytostatic compounds, Int.
J. Biol. Macromol. 86 (2016) 690–700, https://doi.org/10.1016/j.ijbiomac.2016.
The authors have not reported any conflict of interest.
Appendix A. Supplementary data
[
[
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bioorg.2018.10.063.
0
2.018.
19] M. Sriram, G.A. van der Marel, H.L. Roelen, J.H. van Boom, A.H. Wang,
Conformation of B-DNA containing O6-ethyl-G-C base pairs stabilized by minor
groove binding drugs: molecular structure of d(CGC[e6G]AATTCGCG complexed
with Hoechst 33258 or Hoechst 33342, EMBO J. 11 (1992) 225–2327.
References
[
1] (a) Y. Li, X. Qi, X. Xue, X. Wu, Y. Wu, M. Chen, G. Xing, Y. Luan, J. Ren, The
relationship between diphenylamine structure and NSAIDs-induced hepatocytes
injury, Toxicol. Lett. 186 (2009) 111–114, https://doi.org/10.1016/j.toxlet.2009.
[20] O. Trott, A.J. Olson, AutoDock Vina: improving the speed and accuracy of docking
with a new scoring function, efficient optimization, and multithreading, J. Comput.
Chem. 31 (2010) 455–461, https://doi.org/10.1002/jcc.21334.
[21] G.M. Morris, D.S. Goodsell, R.S. Halliday, R. Huey, W.E. Hart, R.K. Belew,
A.J. Olson, Automated docking using a Lamarckian genetic algorithm and and
empirical binding free energy function, J. Comput. Chem. 19 (1998) 1639–1662.
[22] R. Huey, G.M. Morris, A.J. Olson, D.S. Goodsell, A semiempirical free energy force
field with charge-based desolvation, J. Comput. Chem. 28 (2007) 1145–1152,
https://doi.org/10.1002/jcc. 20634.
0
1.005;
(
b) S.M. Abou-Seri, Synthesis and biological evaluation of novel 2,4′-bis
substituted diphenylamines as anticancer agents and potential epidermal growth
factor receptor tyrosine kinase inhibitors, Eur. J. Med. Chem. 45 (2010) 4113–4121,
https://doi.org/10.1016/j.ejmech.2010.05.072;
(
c) L.S. Glass, B. Nguyen, K.D. Goodwin, C. Dardonville, W.D. Wilson, E.C. Long,
M.M. Georgiadis, Crystal structure of a trypanocidal 4,4′-bis(imidazolinylamino)
diphenylamine bound to DNA, Biochemistry 48 (2009) 5943–5952, https://doi.
org/10.1021/bi900204w.
[23] MOPAC2016, Version: 16.146W, James J. P. Stewart, Stewart Computational
Chemistry, web: http://OpenMOPAC.net.
[
2] R. Nanjunda, W.D. Wilson, Binding to the DNA minor groove by heterocyclic di-
cations: from AT-specific monomers to GC recognition with dimers, Curr. Protoc.
Nucleic Acid Chem. (51) (2012), https://doi.org/10.1002/0471142700.nc0808s51
Unit 8.8.
[24] J.A. Maier, C. Martinez, K. Kasavajhala, L. Wickstrom, K.E. Hauser, C. Simmerling,
ff14SB: improving the accuracy of protein side chain and backbone parameters from
ff99SB, J. Chem. Theor. Comput. 11 (2015) 3696–3713, https://doi.org/10.1021/
acs.jctc.5b00255.
[
[
[
[
[
3] R. Simon, S.J. Wang, Use of genomic signatures in therapeutics development,
Pharmacogenom. J. 6 (2006) 166–173, https://doi.org/10.1038/sj.tpj.6500349.
4] G. Bischoff, S. Hoffmann, DNA-binding of drugs used in medicinal therapies, Curr.
Med. Chem. 9 (2002) 312–348.
[25] E.F. Pettersen, T.D. Goddard, C.C. Huang, G.S. Couch, D.M. Greenblatt, E.C. Meng,
T.E. Ferrin, UCSF Chimera–a visualization system for exploratory research and
analysis, J. Comput. Chem. 25 (2004) 1605–1612, https://doi.org/10.1002/jcc.
20084.
5] X. Han, X. Gao, Sequence specific recognition of ligand-DNA complexes studied by
NMR, Curr. Med. Chem. 8 (2001) 551–581.
[26] G.S. Couch, D.K. Hendrix, T.E. Ferrin, Nucleic acid visualization with UCSF
Chimera, Nucleic Acids Res. 34 (2006) e29, https://doi.org/10.1093/nar/gnj031.
[27] Z. Yang, K. Lasker, D. Schneidman-Duhovny, B. Webb, C.C. Huang, E.F. Pettersen,
T.D. Goddard, E.C. Meng, A. Sali, T.E. Ferrin, UCSF Chimera, MODELLER, and IMP:
an integrated modeling system, J. Struct. Biol. 179 (2012) 269–278, https://doi.
org/10.1016/j.jsb.2011.09.006.
6] S. Neidle, C.M. Nunn, Crystal structures of nucleic acids and their drug complexes,
Nat. Prod. Rep. 15 (1998) 1–15.
7] J. Ungvarsky, J. Plsikova, L. Janovec, J. Koval, J. Mikes, L. Mikesova, D. Harvanova,
P. Fedorocko, P. Kristian, J. Kasparkova, V. Brabec, M. Vojtickova, D. Sabolova,
Z. Stramova, J. Rosocha, J. Imrich, M. Kozurkova, Novel trisubstituted acridines as
human telomeric quadruplex binding ligands, Bioorg. Chem. 57 (2014) 13–29,
https://doi.org/10.1016/j.bioorg.2014.07. 010.
[28] M.T. Record Jr., J.-H. Ha, M.A. Fisher, Methods Enzymol. 208 (1991) 291.
[29] J. Ren, T.C. Jenkins, J.B. Chaires, Biochemistry 39 (2000) 8439, https://doi.org/10.
1021/bi000474a.
[
[
8] C.E. Bostock-Smith, M.S. Searle, DNA minor groove recognition by bis-benzimida-
zole analogues of Hoechst 33258: Insights into structure-DNA affinity relationships
assessed by fluorescence titration measurements, Nucleic Acids Res. 27 (1999)
[30] J. Carmichael, W.G. De Graff, A.F. Gazdar, J.D. Minna, J.B. Mitchell, Evaluation of a
tetrazolium-based semiautomated colorimetric assay: assessment of chemosensi-
tivity testing, Cancer Res. 47 (1987) 936–942.
1
619–1624.
[31] A.R. Grivell, M.N. Berry, The effects of phosphate- and substrate-free incubation
conditions on glycolysis in Ehrlich ascites tumour cells, Biochim. Biophys. Acta
1291 (1996) 83–88.
9] H. Ihmels, D. Otto, Intercalation of organic dye molecules into double-stranded
DNA-general principles and recent developments, Top. Curr. Chem. 258 (2005)
1
61–204.
[32] J.D. McGhee, P.H. von Hippel, Theoretical aspects of DNA-protein interactions: co-
operative and non-co-operative binding of large ligands to a one-dimensional
homogeneous lattice, J. Mol. Biol. 86 (1974) 469–489.
[
10] P.O. Vardevanyan, A.P. Antonyan, M.A. Parsadanyan, H.G. Davtyan, Z.R. Boyajyan,
A.T. Karapetyan, Complex-formation of ethidium bromide with poly[d(A-T)]·poly[d
(
A-T)], J. Biomol. Struct. Dyn. 22 (2005) 465–470, https://doi.org/10.1080/
[33] ChemSketch, ACD/Labs Release 11.00 Product version 11.02.
[34] 3D Viewer, ACD/Labs Release 11.00 Product version 11.01.
[35] F. Michel, M.F. Sanner, Python: a programming language for software integration
and development, J. Mol. Graph. Model. 17 (1999) 57–61.
0
7391102.2005.10507017.
[
[
[
11] B.F. Cain, B.C. Baguley, W.A. Denny, Potenial antitumor agents. 28.
Deoxyribonucleic acid polyintercalating agents, J. Med. Chem. 21 (1978) 658–668.
12] D.J. Patel, Nuclear magnetic resonance studies of drug-nucleic acid interactions at
the synthetic DNA level in solution, Acc. Chem. Res. 12 (1979) 118–125.
13] O. Salem, M. Vilkova, J. Plsikova, A. Grolmusova, M. Burikova, M. Prokaiova,
H. Paulikova, J. Imrich, M. Kozurkova, DNA binding, anti-tumour activity and re-
activity toward cell thiols of acridin-9-ylalkenoic derivatives, J. Chem. Sci. 127
[36] J.C. Phillips, R. Braun, W. Wang, J. Gumbart, E. Tajkhorshid, E. Villa, C. Chipot,
R.D. Skeel, L. Kale, K. Schulten, J. Comput. Chem. 26 (2005) 1781–1802, https://
doi.org/10.1002/jcc.20289.
[37] W.L. Jorgensen, J. Chandrasekhar, J. Madura, M.L. Klein, J. Chem. Phys. 79 (1983)
926–935, https://doi.org/10.1063/1.445869.
(
2015) 931–940, https://doi.org/10.1007/s12039-015-0851-9.
[38] J. Wang, P. Cieplak, P.A. Kollman, J. Comput. Chem. 21 (2000) 1049–1074,
https://doi.org/10.1002/1096-987X(200009)21:12<1049::AID-JCC3>3.0.
CO;2-F.
[
[
14] E. Corral, A.C.G. Hotze, H. Dulk, A. Leczkowska, A. Rodger, M.J. Hannon,
J. Reedijk, Ruthenium polypyridyl complexes and their modes of interaction with
DNA: Is there a correlation between these interactions and the antitumor activity of
the compounds? J. Biol. Inorg. Chem. 14 (2009) 439–448, https://doi.org/10.
[39] J. Wang, W. Wang, P.A. Kollman, D.A. Case, J. Mol. Graphics Modell. 25 (2006)
247–260, https://doi.org/10.1016/j.jmgm.2005.12.005.
1
007/s00775-008-0460-x.
[40] J. Wang, R.M. Wolf, J.W. Caldwell, P.A. Kollman, D.A. Case, J. Comput. Chem. 25
(2004) 1157–1174, https://doi.org/10.1002/jcc.20035.
15] J. Janočková, J. Plšíková, J. Kovaľ, R. Jendželovský, J. Mikeš, R. Kašpárková,
V. Brabec, S. Hamuľaková, P. Fedoročko, M. Kožurková, Tacrine derivatives as dual
[41] W.F. Humphrey, A. Dalke, K. Schulten, J. Mol. Graphics 14 (1996) 33–38.
499