Antitumor agents 7. Synthesis, antiproliferative activity and molecular modeling of new L-lysine-conjugated pyridophenoxazinones as potent DNA-binding ligands and topoisomerase IIα inhibitors

Article abstract of DOI:10.1016/j.ejmech.2019.111960

A series of L-lysine-conjugated pyridophenoxazinones 2–5 and 2′-5′ were designed and synthesized for developing compounds with multimodal anticancer potentialities. All compounds inhibited the proliferation of a panel of human liquid and solid neoplastic

Full text of DOI:10.1016/j.ejmech.2019.111960

European Journal of Medicinal Chemistry 187 (2020) 111960
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
Antitumor agents 7. Synthesis, antiproliferative activity and molecular
modeling of new -lysine-conjugated pyridophenoxazinones as potent
DNA-binding ligands and topoisomerase II
L
a
inhibitors
,
Silvana Pedatella a, Carmen Cerchia b, Michele Manfra c **, Anna Cioce d,
,
*
Adele Bolognese a, Antonio Lavecchia b
a Department of Chemical Sciences, University of Naples Federico II, via Cynthia 6, Monte Sant’Angelo, 80126, Naples, Italy
b Department of Pharmacy, “Drug Discovery” Laboratory, University of Naples Federico II, via D. Montesano 49, 80131, Naples, Italy
c Department of Science, University of Basilicata, viale dell’Ateneo Lucano 10, 85100, Potenza, Italy
d
ꢀ
ꢀ
Department of Glycotechnology, CIC biomaGUNE, Paseo Miramon 182, 20009, San Sebastian, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of L-lysine-conjugated pyridophenoxazinones 2e5 and 2′-5′ were designed and synthesized for
Received 16 September 2019
Received in revised form
20 November 2019
Accepted 9 December 2019
Available online 14 December 2019
developing compounds with multimodal anticancer potentialities. All compounds inhibited the prolif-
eration of a panel of human liquid and solid neoplastic cell lines. 2 and 5 were the most active com-
pounds with IC50 values in the submicromolar range. UVevis, 1H NMR, unwinding, and docking
experiments demonstrated that they intercalate between the middle 50-GC-30 base pairs with the car-
boxamide side chain lying into major groove. Charge-transfer contribution to the complex stability,
evaluated by ab initio calculations, was found to correlate with cytotoxicity. Relaxation and cleavage
Keywords:
assays showed that 2 and 5 selectively target Topo II
a over Topo IIb and stimulate the formation of
Pyridophenoxazinones
Antiproliferative activity
DNA damage
covalent Topo IIeDNA complexes, functioning as poisons. Moreover, compound 5 induced DNA damage
and arrested MCF-7 cells at the G2/M phase. Altogether, the work provides interesting structure-activity
relationships in the pyridophenoxazinone-
candidate for further in vivo evaluation.
L-lysine conjugate series and identifies 5 as a promising
Topoisomerase II inhibitors
Docking studies
© 2019 Elsevier Masson SAS. All rights reserved.
1. Introduction
and solid tumor cell lines [2e4]. Molecular modeling calculations
and 1H NMR investigations suggested that the pyridophenox-
azinone skeleton binds to DNA through an intercalation mode by
DNA represents one of the most investigated targets for cancer
treatment. A wide variety of chemotherapeutic drugs belong to the
class of DNA-intercalating agents [1]. Intercalators typically bind to
DNA by inserting a planar aromatic chromophore between adjacent
base pairs, inducing substantial changes in DNA structure and
causing unwinding of the DNA helix. As such, chemotherapeutic
intercalators inhibit DNA replication in rapidly growing cancer
cells.
p p stacking and electrostatic interactions with the DNA base
ꢀ
pairs [2e4]. Furthermore, pyridophenoxazinones can engender free
radical intermediates leading to oxidative-stress-induced cell death
[5]. Therefore, PPH-I and PPH-II possess multimodal cytotoxic ac-
tivity including DNA intercalation, metal-chelating ability and
radical production aptitude.
PPH-I and PPH-II are 5H-pyridophenoxazin-5-one derivatives
endowed with high cytotoxic properties on several lymphoblastoid
* Corresponding author. Department of Pharmacy, “Drug Discovery” Laboratory,
University of Naples Federico II, 80131, Naples, Italy.
** Corresponding Author. Department of Science, University of Basilicata, viale
dell’Ateneo Lucano 10, 85100, Potenza, Italy.
E-mail addresses: michele.manfra@unibas.it (M. Manfra), antonio.lavecchia@
unina.it (A. Lavecchia).
Several DNA-intercalating agents are also able to affect topo-
isomerase catalytic activity, thus producing permanent DNA strand
https://doi.org/10.1016/j.ejmech.2019.111960
0223-5234/© 2019 Elsevier Masson SAS. All rights reserved.

2
S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
Abbreviations
HT-29
LMP2
human colon adenocarcinoma
local Møller-Plesset second-order perturbation
theory
human breast adenocarcinoma
humanCD4þ T cell expressing the TATA gene of
leukemia
3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide
microwave
ACHN
AMD
AMSA
ATCC
BSA
human renal adenocarcinoma
actinomycin D
amsacrine
American type Culture Collection
bovine serum albumin
basis set superposition error
MCF-7
MT-4
MTT
BSSE
CCRF-CEM human acute T-lymphoblastic leukemia
CCRF-SB human acute B-lymphoblastic leukemia; Comet
assay, single cell gel electrophoresis
MW
NBO
PBS
RIPA
PMSF
PVDF
RPMI
SDS
natural bond orbital
phosphate buffered saline
(radio-immunoprecipitation assay)
(phenylmethylsulfonyl fluoride)
poly(vinylidene fluoride
Roswell Park Memorial Institute
sodium dodecyl sulfate
ctDNA
Doxo
DTT
calf thymus DNA
doxorubicin
dithiothreitol
EDC
1-ethyl-3-[3-dimethylaminopropyl]carbodiimide
hydrochloride
EIMS
EtBr
FCS
electronic ionization mass spectra
ethidium bromide
fetal calf serum
SKMEL-28 G-361, human skin melanoma
TAE
Tris-acetate-EDTA
THF
TLC
VP-16
tetrahydrofuran
thin-layer chromatography; Topo, topoisomerase
etoposide.
H-bonds hydrogen bonds
HeLa
HOBt
cervix carcinoma
hydroxybenzotriazole
breaks [1,6e8]. Topoisomerases are enzymes responsible for coor-
dinating DNA topology by catalyzing DNA cleavage and re-ligation.
Thus, the ability to interfere with these enzymes or generate
enzyme-mediated damage is an effective strategy for cancer ther-
apy [9,10]. Depending on their mechanism, topoisomerases are
classified in two main groups: type I topoisomerase (Topo I) and
type II topoisomerase (Topo II). Topo I cleaves one strand of a DNA
double helix to be relaxed and then rejoined. Topo II cleaves both
strands of one DNA duplex to make another non-cleaved duplex
pass through a transient, Topo II-mediated break and then followed
by reannealing the cleaved strands. Human Topo II is categorized
planar chromophore with a flexible cationic side chain capable of
more than one DNA groove or Topo interaction could modulate the
DNA-binding properties and consequently the pattern of Topo
enzyme inhibition. It was indeed recognized that a basic chain,
protonated a physiological pH of 7.4, would be expected to produce
strong electrostatic interactions with the negatively charged
phosphate groups on DNA skeleton, reinforcing the affinity of the
compound for DNA. Also, the attachment of an amino acid or a
peptide to an intercalating drug could effectively increase its
selectivity against tumor cells [23]. For example, conjugates of
doxorubicin (Doxo), daunorubicin, and mitoxantrone with an
amino acid or peptides show more selective antitumor activity than
the drugs themselves [24e33]. Finally, the attachment of an amino
acid or a peptide to the drug could also increase its solubility in
aqueous solutions, overcoming administration problems as well as
multidrug resistance protein-mediated drug efflux and toxicity
[34,35]. Typical examples of this class of substituted intercalators
are amonafide [36], TAS-103 (BMS-247615) [37,38], DACA [39] and
S-16020 [40] (Fig. 1). Based on this rationale, a new series of 5H-
pyrido[3,2-a]phenoxazin-5-one (2e5) and 5H-pyrido[3,2-a]phe-
noxazin-5-one (2′-5′) carboxamides (Fig. 1) bearing a flexible basic
into two subclasses, namely, Topo II
cycle dependent enzyme; it peaks during G2/M phase and de-
creases at the end of mitosis. The overexpression of Topo II is only
observed in proliferating cells, but the expression of Topo II is not
a and Topo IIb. Topo IIa is a cell
a
b
changed during the cell cycle. The majority of Topo poisons are
cytotoxic via stabilization of the normally transient, covalently
bound cleavable complex formed between DNA and enzyme. Drug
stabilized DNA-enzyme complexes (ternary complexes) are
believed to be lethal upon DNA replication when irreversible
double-stranded DNA breaks are formed, causing various chro-
mosomal aberrations culminating in cell death [11,12].
L-lysine side chain at position 9 and 10 of ring A of the pyr-
The inhibition of Topo II
ported as a well-defined approach to develop novel antineoplastic
agents because Topo II is necessary for the hypercompaction of
mitotic chromosomes in human cells, whereas inhibitors targeting
Topo II have been frequently associated with undesirable side
effects, such as drug resistance, cardiotoxicity, or secondary ma-
lignancies [13,14]. The development of resistance to Topo-targeting
drugs, however, presents a major clinical problem; resistance
commonly is manifested as classical multidrug resistance, which
leads to increased efflux of drug from the cell, thereby reducing
cytotoxic efficacy [15] and clinical applications. For these reasons,
medicinal chemistry research efforts have been recently devoted to
a
, rather than Topo II
b
, has been re-
idophenoxazinone framework was synthesized. The N-terminal a-
amino or ε-amino group of
L-lysine was linked to chromophore via
a
an amide linkage.
Here, we report the synthesis of new PPHI- and PPHII-
L-lysine-
b
conjugates (2-5- and 2′-5′) together with a first evaluation of their
biological activities. The DNA binding characteristics of these
molecules were examined using a range of biochemical and bio-
physical methods, and their effects on cytotoxic power, Topo I and II
inhibitory activity and the cell cycle progression were investigated.
The structure of 2e5 complexed with the d(GAAGCTTC)2 octamer
was analyzed by 1H NMR and supported by docking studies.
Moreover, ab initio calculations were carried out on the model
the development of new Topo IIa-specific inhibitors as anticancer
“sandwich” complexes of
L-lysine-conjugated piridophenox-
agents [7,16e22].
azinones with flanking DNA base pairs to establish how the p-p
In the present study, we postulated that combining a tetracyclic
stacking interactions affect the binding of compounds to DNA.

S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
3
Fig. 1. PPH-I, PPH-II, selected DNA-intercalating agents and our designed L-lysine-conjugated pyridophenoxazinones 2e5 and 2′-5’.
2. Results and discussion
hypothesis that the amino group firstly attacks the quinoline-5,8-
dione [2]. Under these circumstances, we considered worthwhile
to employ a microwave (MW) assisted synthetic procedure to
obtain 1-1′ and 100-1000 with better yields, very short reaction times
and, more importantly, with diminished production of black poly-
meric residues. Therefore, the reaction was performed into a MW
CEM Discover apparatus at 50 W for 300 of continuous irradiation,
and was found to yield better results. Compounds 1-1′ and 100-1000
were recovered from the reaction mixture with a yield of 30e65%
and this result seemed to suggest that a dehydration Step, pro-
moted by MW heating, could to play a crucial role in the mecha-
nism of this reaction. Scheme 1 reports the yields obtained using
the conv and MW procedures.
2.1. Chemistry
The synthetic procedure to prepare 2-5 and 2′-5′ required three
main steps: i) preparation of pyridophenoxazinone carboxylic acids
1-1′ and 100-1000, ii) preparation of differently masked
L
-lysines on
-, ε-amino and carboxy groups, and iii) coupling reaction be-
tween the carboxylic acids 1-1′ and 100-1000 and differently protected
-lysines. Scheme 1 reports the reactions between equimolecular
the
a
L
amounts of 3-amino-4-hydroxybenzoic acid or 3-hydroxy-4-
amino-benzoic acid and the quinoline-5,8-dione, in acetic acid at
refluxing temperature, which gave rise to 1-1′ and 100-1000, respec-
tively. This conventional procedure (conv) was described to prepare
similar compounds in our previous papers (Scheme 1) [2,4].
The conventional procedure was very expensive and required
long reaction times, and furnished very poor yields. Several side
products, like phenoxazines, aminophenoxazin-3-ones, tripheno-
dioxazines accompanied the desired pyridophenoxazinone acids.
These by-products together with black, intractable, polymeric
materials decreased significantly the yields of the expected acids
and encumbered the subsequent purification procedures. It is
reasonable to assume that the presence of the withdrawing carboxy
group in para position to the amino group of the ortho amino
phenol decreases its reactivity and minimizes the reaction yields.
The smaller detrimental effects, observed when the carboxy group
lies in para position to the phenol hydroxyl group, suggests the
The second reaction Step consisted in preparation of differently
masked L-lysines on the a-, ε-amino and carboxy groups to use in
the successive copulation reaction with the acids 1-1′ and 100-1000.
Scheme 2 details the synthetic procedures used starting from the
commercial Fmoc-L-lys(Boc)OH (A) and Boc-
L-lys(Cbz)OH (D),
respectively. was esterified at the carboxy group in dry
A
dichloromethane with tert-butyl 2,2,2-trichloroacetamidate in
presence of boron trifluoride diethyl etherate to form the corre-
sponding butyl ester B, which furnished the a-amino L-lysine tert-
butyl ester C by treatment with piperidine in acetonitrile (Scheme
2a). To gain access to the ε-amino function, D was used and con-
verted into intermediate (E). After esterification, the ε-amino group
of E was de-protected yielding F by a reduction reaction with
Palladium/Carbon in methanol (Scheme 2b).

4
S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
Scheme 1. Synthetic route of compounds 1-1′ and 100-1000. Reagents and conditions: i. 3-amino-4-hydroxybenzoic acid or 3-hydroxy-4-amino-benzoic acid, quinoline-5,8-dione,
Co(II) acetate, CH3COOH, MW assisted heating 150, 50 W; ii. 3-amino-4-hydroxybenzoic acid or 3-hydroxy-4-amino-benzoic acid, quinoline-5,8-dione MW assisted heating 150,
50 W.
Scheme 2. Synthetic route of compound C and F. Reagents and conditions: a) i. t-Butyl-2,2,2-trichloroacetamidate, boron trifluoride diethyl etherate, CH2Cl2 dry, rt, overnight; ii.
Piperidine, CH3CN, 0 ꢁC / rt, 3.5 h. b) i. t-Butyl-2,2,2-trichloroacetamidate, boron trifluoride diethyl etherate, CH2Cl2 dry, rt, overnight; ii. H2, 5% Pd/C, MeOH, rt, 3.5 h.
In the third reaction Step, the ε- and
a
-protected Boc-
L
-lysine t-
NMR spectra and HPLC data of the final compounds 2e5 and 2′-5′
(Figs. S5eS12).
butyl esters, C and F, were used in the coupling reactions with the
pyridophenoxazinone acids 1-1′ and 100-1000 to form the designed
carboxamides. Scheme 3 details the overall synthetic procedure of
the formation of 2e5 and 2′-5’.
2.2. Antiproliferative activity
To the anhydrous solutions of phenoxazinones 1-1′ and 100-1000,
dissolved in THF at 0 ꢁC, an equimolar quantity of hydroxybenzo-
triazole (HOBt) was added to activate the carboxy group. After
The antiproliferative activity of the L-lysine-conjugated pir-
idophenoxazinones 2e5 and 2′-5′ was evaluated against a panel of
human cancer cell lines representative of liquid and solid tumors
and the results are shown in Table 1. The activities, reported as IC50
30 min, equimolar quantities of masked suitable -lysine C or F and
L
of 1-ethyl-3-[3-dimethylaminopropyl]carbodiimide hydrochloride
(EDC) in anhydrous tetrahydrofuran (THF) were added to the re-
action mixtures. The solutions were stirred at 0 ꢁC for 1 h and then
allowed to worm slowly to room temperature. After 12 h, the sol-
vent was evaporated and the crude products, containing the
appropriate, masked carboxamides, were dissolved in AcOEt and
washed with saturated aqueous solution of NH4Cl. The crude
products 2a-5a and 2′a-5′a were de-protected with formic acid at
room temperature yielding 2e5 and 2′-5′ (Scheme 3). Further pu-
rification was performed with thin-layer chromatography (TLC) on
silica gel plates using chloroform-methanol (80:20; v:v, as eluent).
1H NMR spectra of starting phenoxazinones 1, 1′, 100 and 1000 are
included as Supplementary data (Figs. S1eS4) together with 1H
(mM) values, were compared with those of PPH-I, PPH-II from our
laboratory and commercial Doxo, AMD, and etoposide (VP-16).
The tested compounds showed diverse activity against the
liquid and solid tumor cells, being the last ones more sensitive to
their antiproliferative effect. Only 5 showed comparable activity on
both solid and liquid cancer cells. The IC50 values of derivatives 2e5
were lower than the corresponding IC50 reported for derivatives 2′-
5′. Moreover, 2e5 showed a cytotoxic power on solid cell lines
comparable to that of the reference compound AMD and higher
than that of Doxo, VP-16 as well as PPH-I and PPH-II. The maximal
potency was achieved with 2 and 5 bearing the
amide side chain attached to the 9 position of pyridophenox-
azinone system through - and ε-amino group, respectively. The
L-lysine carbox-
a

S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
5
Scheme 3. Synthetic route of compounds 2e5 and 2′-5′. Reagents and conditions: i. HOBT, THF dry, 0 Co, 30’; C or F, EDC, 1 h 0 Co, 12 h at r.t; ii. AcOEt/NH4Cl; HCOOH, rt.
Table 1
Antiproliferative activity of compounds 2e5 and 2′-5′ on human liquid and solid cancer cell lines.
IC50
M)a
(m
Cpd
CCRF-CEM
CCRF-SB
MT-4
HT-29
MCF-7
HeLa
ACHN
SKMEL-28
G-361
2
2′
3
3′
4
4′
5
5′
PPH-I
PPH-II
Doxob
AMDb
VP-16b
0.02
0.017
0.10
0.16
0.04
0.15
0.003
0.15
0.01
0.09
0.02
0.001
1.20
0.02
0.08
0.12
0.16
0.08
0.10
0.005
0.16
0.01
0.15
0.03
0.002
0.10
0.01
0.20
0.09
0.20
0.09
0.15
0.006
0.20
0.01
0.18
0.01
0.001
1.25
0.003
0.10
0.08
0.20
0.02
0.12
0.004
0.10
0.25
0.38
0.15
0.006
2.50
0.007
0.05
0.10
0.10
0.06
0.12
0.003
0.12
0.23
0.37
0.05
0.006
1.00
0.005
0.05
0.08
0.10
0.05
0.15
0.005
0.10
0.20
0.15
0.20
0.008
3.00
0.008
0.06
0.10
0.08
0.05
0.10
0.005
0.07
0.01
0.23
0.04
0.005
2.15
0.002
0.06
0.12
0.21
0.04
0.10
0.002
0.08
0.21
0.40
0.06
0.002
1.10
0.001
0.05
0.06
0.05
0.02
0.03
0.006
0.09
0.01
0.45
0.09
0.006
2.30
a
Compound concentration required to reduce cell proliferation by 50%, as determined with the MTT method, under conditions allowing untreated controls to undergo at
least three consecutive rounds of multiplication. Data represent mean values for three independent determinations. CCRF-CEM, human acute T-lymphoblastic leukemia;
CCRF-SB, human acute B-lymphoblastic leukemia; MT-4, humanCD4þ T cell expressing the TATA gene of leukemia; HT-29, human colon adenocarcinoma; MCF-7, human
breast adenocarcinoma; HeLa, cervix carcinoma; ACHN, human renal adenocarcinoma; SKMEL-28, G-361, human skin melanoma.
b
Controls.
most sensitive cells to the cytotoxic effect of 2 and 5 were HT-29
(IC50 of 0.003 and 0.004 M), SKMEL-28 (IC50 of 0.002 and
0.002 M), MCF-7 (IC50 of 0.007 and 0.003 M) and G-361 (IC50 of
0.001 and 0.006 M). It is interesting that the antiproliferative ac-
tivity of the two series of -lysine-conjugates, 2e5 and 2′-5′ , was
the two series 2e5 and 2′-5′ probably results from their different
mechanisms of action and suggests that the position of pyridine
nitrogen within the pyridophenoxazinone skeleton plays an
important role in the antiproliferative activity of the tested com-
m
m
m
m
L
pounds. Moreover, the position of
idophenoxazinone ring, as well as the
in the amide bond formation affect the difference in the cytotoxic
L
-lysine side chain on the pyr-
generally higher than the activity of the parent unsubstituted
compounds PPH-I and PPH-II. The difference of activity between
a
- or ε-amino group involved

6
S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
activity between the two series 2e5 and 2′-5′ and within each
2.5. Unwinding assay
series. To better highlight the mechanism of action of the pyr-
idophenoxazinone
ties and Topo I and II inhibition were investigated.
L
-lysine-conjugates, their DNA binding proper-
The most powerful cytotoxic compounds of series 2e5 were
investigated through unwinding assays and 1H NMR spectroscopic
experiments using the double strand octamer [d(GAAGCTTC)]2,
expressly synthesized to intercalate the pyridophenoxazinone
chromophore [2]. The unwinding assay is based on the formation of
constrained negative supercoils and compensatory unconstrained
positive supercoils in covalently closed circular DNA by an inter-
calative agent. When an intercalated plasmid is treated with human
Topo I, the unconstrained positive supercoils are removed. The
extraction of the intercalative agent allows the redistribution of the
constrained negative supercoils throughout the DNA molecule and
the conversion of the relaxed DNA into negatively supercoiled DNA
[52]. Thus, the DNA unwinding assay is designed to directly mea-
sure the extent of the drug-induced negative supercoils.
2.3. DNA binding properties
Compounds 2e5 and 2′-5′ were tested to assess whether they
interacted with DNA. The interaction of L-lysine-conjugates with
calf thymus DNA (ctDNA) was monitored by optical spectrophoto-
metric titrations in phosphate buffer, pH 7. The UVeVis spectra of
compounds showed a significant absorption in the range of
430e469 nm, typical for transitions between
levels. When DNA binding occurs, the compounds are inserted in a
different environment than the uncomplexed form in solution,
p-electron energy
The DNA unwinding assay was performed using the relaxed
plasmid pBR322 in the presence of various concentrations of the
more active derivatives 2e5. Ethidium bromide (EtBr) was used as
control. All compounds significantly changed the topological state
which promote the distortion of the electron distribution upon p-
stacking with the bases. This accounts for the significantly different
compound absorption features in the complexed and uncomplexed
forms [41]. The spectroscopic properties of 2e5 and 2′-5′ in the
presence and absence of ctDNA showed hypochromism and red
shift due to the formation of stable DNA-ligand complexes, sug-
gesting that they act as DNA intercalating compounds. The pres-
ence of isosbestic points suggested that the compounds present a
single DNA binding mode. Table 2 lists the spectroscopic data of the
of plasmid pBR322 at 10
respectively), and 2 and 5 exhibited a more marked unwinding at
the lowest 0.1 M concentration (Fig. 3, lanes B and M, respec-
mM concentration (Fig. 3, lanes C, F, I and N,
m
tively). It should be noted that the cytotoxicity of the four com-
pounds (5 z 2 > 3 z 4) correlated well with the levels of DNA
negative supercoils induced by these compounds. A strand passage
L-lysine-conjugates in the absence and in presence of ctDNA.
assay for human Topo I displayed that the L-lysine-conjugates did
not affect the catalytic activity of Topo I (data not shown).
2.4. UVevisible spectroscopic analysis of 5-DNA complexes
2.6. 1H NMR spectroscopic analysis
Fig. 2A shows the absorption spectra of free ctDNA and its
complexes with varying concentration of compound 5, chosen as
representative of series 2e5. When 5 interacts with DNA, it causes
hyperchromic effect (increase in absorbance of DNA) with
increasing concentration of 5 in solution. This phenomenon is
coupled with the 10 nm shift in the absorbance maxima from
260 nm to 250 nm (hypsochromic or blue shift). Increase in
absorbance along with the hypsochromic shift is attributed to the
intercalation of the chromophore of 5 between the DNA base pair.
The binding constant (K) represents a quantitative measure-
ment of binding of 5 with DNA. In this work, K has been determined
by using the “mole ratio method’’, which involves the preparation
of multiple solutions of drugeDNA complex, keeping the concen-
tration of DNA constant while varying the drug concentration from
1 ꢂ 10ꢀ1 mM to 3 ꢂ 10ꢀ2 mM. It is evaluated by examining the
variations in the optical density at 260 nm for free DNA (repre-
sented by A0) and its complexes with 5 (represented by A) [42,43].
Since the double reciprocal plot of 1/(A - A0) vs. 1/(drug concen-
tration) is linear, K can be calculated from the ratio of the intercept
to the slope [43e48] (Fig. 2B). The binding constant for this inter-
Quantitative titrations of the octamer [d(GAAGCTTC)]2 with
compounds 2e5 were performed into a NMR tube of a 500 MHZ
spectrometer, at probe temperature of 25 ꢁC where the proton
signals can be resolved without compromising the octamer duplex
stability. 1H NMR spectra, before and after addition of pyr-
idophenoxazinone-L-lysine conjugates to [d(GAAGCTTC)]2 (2:1 M
ratio), were executed in 99.99% D2O 10 mM phosphate buffer, pH 7.
In phosphate buffer the resonances of 2e5 all broadened signifi-
cantly and characteristic changes were observed for both octamer
and ligand protons. Table 3 shows the 1H NMR resonance values of
2e5 and their complexes to [d(GAAGCTTC)]2. Fig. 4 illustrates the
spectra of the compound 5, the octamer, and the corresponding
complex, taken as representative of the behavior of whole series.
Spectra of compounds 2e4 are included as Supplementary data
(Figs. S13eS15).
The spectral features of the L-lysine derivative/octamer com-
plexes showed that all the signals of the downfield pyr-
idophenoxazinone moiety shifted to higher field in the area of the
spectrum characteristic of the DNA base pairs resonances (7.74 and
7.48
shielded by the magnetic field produced by ring currents due to the
bases owing to the stacking interactions typical of the inter-
d). This shift indicates that the pyridophenoxazinone moiety is
action is 3.8 ꢂ 105
M
ꢀ1, which demonstrates the high affinity of 5
for DNA double helix. The value of binding constant obtained is
consistent with those reported for the interaction of anthracycline
molecules with DNA [48e51].
p-p
calative binding of a planar polycyclic system into DNA double helix
(Table 3 and Fig. 4). Moreover, the general chemical shift pertur-
bation of non-exchangeable octamer protons respect to those of the
unmodified palindromic duplex structure can be due also to lack of
symmetry of the base sequences related to the presence of the
ligand and to its interactions. The large shift of protons H6 of 2e5
toward higher field in all the complexes (ranging between 0.38 and
0.10 ppm) seems to underline the different involvement of the
Table 2
Isosbestic point (nm), lmax, and % of hypochromic effect of compounds 2e5 and 2′-
5′.
2
2′
3
3′
4
4′
5
5′
lmax
443
451
437
445
436
441
443
459
electron-acceptor iminoquinone system in the
p-p stacking in-
ε
5736 5928 5316 8431 6620 1792 4219 1216
Isosbestic point
Red shift
% hypochromism 28
395
447
297
455
26
343
440
30
280
447
19
310
440
39
275
446
17
340
448
58
320
461
17
teractions with DNA base pairs (Table 3). As regards the protons of
lysine side chains, it is to note that upon complex formation, they
are pushed toward lower field. The magnitude of these shifts can be

S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
7
Fig. 2. (A) UVeVisible spectra of free calf thymus DNA (0.2 mM) in the absence and presence of 5 leading to drug: DNA ratio of (a) 0:1 (free calf thymus DNA) (b) 1:1 (c) 2:1 (d) 3:1
(e) 4:1 (f) 5:1 (g) 6:1. (B) Double reciprocal plot of 5 binding to DNA. A0 is the initial absorption of free DNA and A is the absorption at different concentrations at 260 nm. C is the
analytical concentration of 5 in solution.
Fig. 3. L-Lysine-conjugates 2e5 intercalate into DNA. The DNA unwinding reaction mixtures contained 0.3 mg of relaxed plasmid pBR322 DNA, 2 units of human Topo I, and the
indicated concentrations of the various compounds (lanes A-N). Three independent experiments for each assay exhibited essentially identical results; EtBr (2
control.
mM) was used as
affected, at different extent, by ionic or H-bond interactions with
both backbone and base pairs of octamer according to the relative
positioning of the guest molecule into the intercalation site.
structure. This result also is in line with the observed larger cyto-
toxic effect of 5 on the tumor cell lines.
As showed in Table 3, the
L-lysine side chains significantly
2.7. Computational studies
interact with some DNA bases and their
a
- and ε-amino and/or
carboxy groups seem to be involved in strong interactions. It is to
note that one of the diasterotopic hydrogens belonging to the 30-
CH2 group of the compounds 2 and 3 undergoes a shift larger than
In order to rationalize the observed biological results and gain
more insight into the intercalation modality of 2e5, the supramo-
lecular complexes of titled compounds with DNA were investigated
by molecular docking. Ab initio calculations were also performed on
the other one (
D
Нb ¼ 0.14 and 0.04 ppm, respectively). Compound
5, which shows a more significant shift of the chromophore pro-
the model “sandwich” complexes of pyridophenoxazinones-
lysine conjugates with flanking DNA base pairs to establish how the
entity of the two forces contributing to the stacking (dispersion
and charge-transfer interactions) affects the binding of compounds
L-
0
0
0
tons, also displays a large shift of protons H2 , H3 and H6 of the L-
0
0
0
lysine (
suggesting
D
Н2 ¼ 0.32,
D
Н3 ¼ 0.22 and
D
Н6 ¼ 0.76 ppm, respectively),
p-p
a strong perturbation of the DNA double-helical

8
S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
Table 3
Proton chemical shift assignments of unbounded 2e5 and their comparison with the corresponding 2e5/[d(GAAGCTTC)]2 complexes.
Ligand-Unbounded Aromatic Protons (
d)
Ligand/[d(GAAGCTTC)]2 Complex Aromatic Protons (d)
H1
H2
H3
H6
H8
H9/10
H11
H1
H2
H3
H6
H8,9/10,11
2
3
4
5
8.77
8.58
8.65
8.80
7.71
7.54
7.65
7.74
8.77
8.53
8.61
8.83
6.45
6.58
6.36
6.49
7.71
7.11
7.18
7.74
7.59
7.70
7.66
7.59
7.59
7.54
7.65
7.53
8.70
8.49
8.68
8.68
7.55
7.54
7.62
7.52
8.53
8.60
8.68
8.63
6.30
6.24
6.26
6.11
7.44e48
7.44e48
7.44e48
7.44e48
Ligand Unbounded Aliphatic Protons (
d)
Ligand/[d(GAAGCTTC)]2 Complex Aliphatic Protons (d)
H2′
H3′
H4′
H5′
H6′
H2′
H3′
H4′
H5′
H6′
2
3
4.27
1.86a 1.80b
1.44
1.73
2.95
4.38
1.86a
1.94b
1.81a
1.75b
1.82
1.54
1.28
2.92
2.93
4.18
1.81a 1.71b
1.38
1.62
2.91
4.23
1.42
1.65
4
5
3.61
3.59
1.78
1.78
1.36
1.57
1.55
1.64
3.22
3.27
3.65
3.91
1.52
1.57
1.65
1.64
3.26
4.03
1.56
a and b indicate CH2 diastereotopic protons.
to DNA. Autodock 4.2 [53] was used to set up and perform docking
experiments. This software has been reported to accurately
reproduce the experimental binding conformation to nucleic acids
of a series of small ligands (minor groove binders and intercalators)
[54,55]. The d(GAAGCTTC)2 octamer complexed with AMD (PDB
code ¼ 209D) [56] was selected for docking experiments. From a
total of 200 independent docking runs, AutoDock gave well-
clustered results, which were chosen on the basis of the lowest
As regards the carboxamide side chain, which adopts a gauche-
gauche arrangement, the 60- or 20-NHþ3 group of 2e5 makes H-
bonding interactions with the O6 atom of G4, O4 atom of T14 and
both adenine A3 and cytosine C13 NH2 group. The negatively
charged environment of the major groove intercalation site there-
fore indirectly generates a favorable side-chain orientation for H-
bond formation to G4, T14, A3 and C13. In addition, the 10-COO-
group of 4 and 5 engages in a further charge-reinforced H-bond
with the adenine A3 NH2 group. Finally, the hydrogen of the
electron-poor amide 20- or 60-NH group of all ligands is within H-
bonding distance from the NH2 of cytosine C13. Consistently with
the reported intercalation models, when ligand/d(GAAGCTTC)2
complexes take place, most of the protons of carboxamide side
chains shift in the 1H NMR spectra (Table 3). In the 2/d(GAAGCTTC)2
and 5/d(GAAGCTTC)2 complexes, the 20-CH2 protons show a very
large shift (0.11 and 0.32 ppm, respectively) according to the for-
mation of H-bonding interactions between the 20-NH amide and 20-
NHþ3 groups and C13, G4, T14, and A3 DNA bases. Moreover, the
change of chemical shift of one of the diasterotopic 30-CH2 protons
of 2 and 3 in complex with the DNA octamer (Table 3) seems due to
a freezed, stable conformation of the chain where the two protons
experience a different chemical environment, being the proton Hb
facing towards cytosine C13.
docking energy (DGbind) and the highest frequency of occurrence
(focc). Fig. 5 shows the most potent compounds 2e5 intercalated via
the major groove of the 50-GC-30 binding site with their carbox-
amide side chain lying in the major groove. The carboxamide side
chain is oriented perpendicularly to the planar aromatic system and
fits well into the interior of the major groove. The pyridophenox-
azinone system intercalates with a standard stacking distance of
about 3.4 Å to both GꢃC base pairs. It is to note that the chromo-
phore of 3 and 4 lays in a flipped orientation rotated 180ꢁ respect to
that of 2 and 5, whereas the carboxamide side chain of all com-
pounds projects in the same position of DNA. The carboxamide
adopts
a
dihedral angle [C(pyridophenoxazinone)-C(pyr-
idophenoxazinone)-C(C]O)eO(C]O)-)] of 121ꢁ with respect to
the planar pyridophenoxazinone system. The positively charged
pyridine nitrogen of all ligands forms two hydrogen bonds (H-
bonds) with both the O40 atom in the deoxyribose ring of C5 residue
and the O50 atom of the phosphate backbone located between G4
and C5 residues. These H-bonding interactions are in agreement
with the H1, H2 and H3 shifts observed in 1H NMR spectra under the
2e5/[d(GAAGCTTC)]2 complex formation, being the shifts in 2 and
5 larger than 3 and 4. Table S1 of Supplementary data illustrates the
effect of H-bond strength on the H1, H2 and H3 chemical shift dif-
ferences in all ligand bounded and unbounded DNA complexes.
To investigate how the entity of the
p-p stacking interaction
energy, i.e. dispersion forces and charge-transfer interactions, af-
fects the binding of 2e5 to DNA base pairs, the ligand/DNA com-
plexes generated by docking simulations were subjected to
quantum mechanical single point energy calculations. In order to
derive a simplified model for such calculations, compounds 2e5
and their flanking base pairs were extracted from the ligand/DNA
complexes. The deoxyribose rings of the flanking base pairs were

S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
9
Fig. 4. 1H NMR spectrum in phosphate buffer/D2O pH 7 of the (A) octamer d(GAAGCTTC)2, (B) 5, and (C) 5/[d(GAAGCTTC)]2 complex.
replaced by methyl groups, and the carboxamide side chains at 9
and 10 positions of pyridophenoxazinones were substituted by
methylamide groups because the chains are considered to be non-
was performed at the HF/6-31G** level of theory and presented as
ECT in Table 4.
As shown in Table 4, models A and B share comparable disper-
participating groups in
p
ꢀ
p
stacking interactions. The energies of
sion energies, indicating that this kind of interaction is isotropic (i.e.
independent of spatial orientation) and proportional to the
geometrical overlap of the stacked systems. Analysis of the ECT
column, on the other hand, suggests that the charge transfer sta-
bilization energy is greater by 4.10 kcal/mol for pyridophenox-
molecules 2e5, C-G base pair, and G-C base pair were optimized
separately at the HF/6-31G** level of theory in Jaguar 9.0 software
package [57]. The original complexes were then replaced with the
geometry of optimized parts (Fig. 6, model A, representative of
compounds 2 and 5, and model B, representative of compounds 3
and 4). Models A and B were then subjected to a single-point en-
ergy calculation at the LMP2 (local Møller-Plesset second-order
perturbation theory) level of theory by using the 6-31G* basis set.
The LMP2 method [58] is known to minimize basis set super-
position error (BSSE), and therefore counterpoise correction [59]
was not included in the calculations.
azinones represented by model
A
in comparison to
pyridophenoxazinones in model B. These results indicate that the
charge transfer interactions play a much more important role in the
stabilization of the complexes between 2/5 and DNA than between
3/4 and DNA, which might help to explain the superior cytotoxicity
of compounds 2 and 5 versus the corresponding 3 and 4.
The in vacuo
p-p stacking interaction energy was then calcu-
2.8. Inhibition of the catalytic activity of Topo II
a in vitro
lated as Eint,vac ¼ Ecomplex,vac - Eligand,vac - Ebp,vac, where Ecomplex,vac
,
Eligand,vac, and Ebp,vac are the corresponding LMP2/6-31G* calculated
energies of complex, ligand, and DNA base pairs at their normal
distance in the absence of intercalator (Table 4). Similarly, the en-
The relaxation of supercoiled plasmid DNA by human Topo II
was studied in the presence of different concentrations of 2 or 5.
Compound 5 diminished the relaxation reaction by 50% at a con-
a
ergy of the
(Eint,aq) using the polarized continuum model.
As stacking mainly originates from dispersion and charge
p-p staking interaction was also calculated in water
centration of 0.7
m
M and totally inhibited the Topo II-mediated
M (Fig. 7, right). The closely related compound
, but to a lesser
M 2 was required to inhibit the relaxation
reaction by 50%, whereas 2 M was needed to completely inhibit
the catalytic activity (Fig. 7, left).
relaxation at 1.5
m
p-p
2 also inhibited the catalytic activity of Topo II
extent. Approximately 1
a
transfer interactions, we calculated the dispersion and charge
transfer contributions to make a direct quantitative comparison of
their roles in the binding of pyridophenoxazinones to DNA com-
plex. The dispersion (electron correlation) interaction (Ecorr) was
defined as the difference of energy calculated by LMP2 and HF
methods at the 6-31G* level, Ecorr ¼ Eint(LMP2) - Eint(HF) (Table 4). The
energy associated with charge transfer was obtained as result of
natural bond orbital (NBO) analysis in Jaguar [57]. The NBO analysis
m
m
2.9. Induction of Topo IIa and Topo IIb cleavable complexes in vitro
The biologically active 2 and 5 are able to induce cleavable
complexes between Topo II and pBR322 plasmid DNA (Fig. 8) more
a

10
S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
Fig. 5. Pyridophenoxazinones-L-lysine conjugates 2 (A, magenta), 3 (B, orange), 4 (C, white), and 5 (D, yellow) intercalated into 50-GC-30 base pairs of the DNA octamer
d(GAAGCTTC)2. Black dashed lines indicate H-bonds. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
biphasic pattern, which is typical of compounds that are strong
DNA binders [60,61]. Therefore, both 2 and 5 can be considered to
be Topo IIa poisons.
It is notable that Topo II poisons are used widely in the clinic to
treat a variety of cancers; however, their utility is hampered by
severe adverse side effects including cardiac myopathies and sec-
ondary malignancies [9], which appear to be correlated with their
ability to poison Topo II
b
[12,13]. Hence, finding drugs specifically
-mediated toxicity/carcinoge-
targeting Topo II with low Topo II
a
b
nicity is being actively pursued [62]. Thus, DNA cleavage assay
experiments were also carried out to determine whether com-
pound 5 stabilized the cleavable complex induced by Topo II
Results showed that 5 was more selective in inducing linear DNA
formation by acting on Topo II than on Topo II (Fig. S16 in Sup-
plementary data). In order to rationalize the observed inhibitory
effect on the enzymatic activity of Topo II , 5 was docked into the
recently solved X-ray crystal structure of the human Topo II /DNA/
etoposide ternary complex (PDB code ¼ 5GWK) [63]. The overall
results showed that 5 is able to stabilize Topo II /DNA cleavage
b.
Fig. 6. Intercalation complex model containing simplified structures of 2/5 (model A)
and 3/4 (model B) and their flanking base pairs for ab initio calculations.
a
b
effectively than amsacrine (AMSA). The formation of cleavable
a
complex with Topo II
in the presence of 1e5
binding of two carboxamides to DNA generated a characteristic
a
was dose-dependent, reaching a maximum
a
mM 2 and 5. At higher concentrations, the
a

S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
11
Table 4
Stacking interaction energies of models A and B.
p-p
a
a
Model
Eint, vac (kcal/mol)
Eint,aq (kcal/mol)
Ecorrb (kcal/mol)
ECTc (kcal/mol)
A
B
ꢀ6.24
ꢀ6.11
ꢀ12.15
ꢀ10.89
ꢀ28.46
ꢀ27.85
ꢀ16.28
ꢀ12.18
a
Eint is derived from LMP2/6-31G* single point energy calculations.
Ecorr ¼ Eint(LMP2/6-31G*) - Eint(HF/6-31G*).
b
c
The magnitude of the charge transfer as estimated by NBO analysis at HF/6-31G** level.
Fig. 7. Inhibition of the catalytic activity of purified DNA Topo II
purified Topo II in the absence (lane I) or presence of 2 at 0.3, 0.5, 0.7, 1, 1.5, 2 and 5
forms of DNA were separated by gel electrophoreses. S, supercoiled DNA; R, relaxed DNA; T, DNA topoisomers. Data shown are typical of three independent experiments.
a
by L-lysine conjugates 2 and 5 as measured by relaxation. Supercoiled pBR322 plasmid DNA (lane H) was relaxed by
a
m
M (lanes A-G) or with 5 at 0.1, 0.3, 0.5, 0.7, 1, 1.5, and 2 M (lanes L-R). The resulting topological
m
complex (See more details and Fig. S17 in Supplementary data).
Modeling studies also suggested that the enhanced Topo II spec-
ificity of 5 may result from interactions with a serine residue that is
present in Topo II , but not in Topo II (Fig. S18 in Supplementary
2.11. Compound 5 induces G2/M phase cell cycle arrest and
apoptosis in MCF-7 cells
a
a
b
The effect of 5 on the cell cycle progression of MCF-7 cells was
data). Thus, compound 5 was chosen to further investigate its
anticancer mechanism.
evaluated after continuous exposure to 0.01 and 0.1
centrations for 12 h, 24 h and 48 h, respectively. Cultures treated
with 5 at 0.01 M exhibited only a minimal deviation from control
cell cycle distribution (data not shown). Exposure to 5 at 0.1
mM drug con-
m
mM
2.10. DNA damage induced by compound 5
provoked a visible perturbation of cell cycle progression (Fig. 9D). A
noticeable increase in the percentage of cells in G2 phase, and an
important diminution in cell number over control was detected
after 24 and 48 h (control: 17.00% 3.01%; 24 h: 26.00% 7.5%;
48 h: 48.77% 9.55%), outlining a G2 phase accumulation. Mean-
Considering that compound 5 behaves as an intercalative Topo II
inhibitor, we further investigated the action mode of 5 with MCF-
7 cells by observing comet tails in the single cell gel electrophoresis
(comet assay). The assay gives an image of the alterations that have
happened in the chromatin organization in a single cell. It is
considered a more correct fashion of detection of nuclear modifi-
cations in a cell population [64,65]. Compound 5 provoked lesions
in DNA and significantly increased the comet tail lengths with
increasing concentration. Fig. 9A shows an increase in comet for-
mation following treatment with compound 5 at concentrations
while, G0/G1 dropped considerably from 51.70%
6.99%e
42.50% 9.41% and the population in S phase was reduced to
8.73% 0.3% during 48 h incubation. These results suggest that G2/
M phase arrest accounts for the antiproliferative effect of 5
observed in MCF-7 cells.
ranging from 1 to 10
untreated cells. Fig. 9B shows the comet lengths in pixels with
respect to the concentration of 5. In cells treated with 10 M 5, the
m
M, while no comet formation was observed in
3. Conclusions
m
Novel pyridophenoxazinone-L-lysine conjugates (compounds
tail length was ~50% when compared with the tail length of un-
treated MCF-7 cells, which is nearly 8% (p < 0.05). From these data,
it can be concluded that 5 could induce severe DNA damage in
breast cancer cells.
To gain more insight into the induction of DNA damage of
compound 5, we monitored the levels of phosphorylated histone
2e5 and 2′-5′) were designed, synthesized and evaluated for their
anticancer activity against a wide variety of hematological and solid
human tumor cell lines. All the compounds exhibited high anti-
proliferative activity, displaying potency equal or larger of Doxo and
VP-16. 2 and 5 were the most potent compounds of the series with
IC50 values ranging between 0.001 and 0.008
mM against the ma-
H2AX (
biomarker for double strand breaks in DNA [66e68]. As shown in
Fig. 9C, the levels of H2AX in the 5-treated MCF-7 cells were
clearly augmented, indicating that compound 5 was a potent DNA-
damaging agent.
g
H2AX) in MCF-7 cells treated with 5.
g
H2AX is known as a
jority of solid cell lines and 0.01 and 0.15 M against human leu-
m
kemia cell lines. The interaction with DNA, studied by optical
spectrophotometry and 1HNMR spectroscopy, showed that these
compounds are strong DNA intercalators. Docking studies sug-
gested that 2e5 intercalate at the 50-GC-30 binding site with their
g

12
S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
Fig. 8. The influence of 2 and 5 on the formation of covalent DNA/Topo II
absence (lane I) or in the presence of 10 M AMSA (lane L) or with 0.5, 1, 1.5, 2, 5, and 10
proteinase K, and the different topological forms of DNA were separated by agarose gel electrophoresis in the presence of EtBr. Lane H, linearized pBR322 DNA. S, supercoiled DNA;
R, relaxed DNA; L, linear DNA; N, nicked circular DNA. Data shown are typical of three independent experiments. (B) Topo II -mediated DNA cleavage in the presence of L-lysine
conjugates 2 and 5. Plasmid DNA and purified Topo II were incubated with different concentrations of 2 or 5, and the formation of linear DNA determined by gel densitometry. Data
a
complexes. (A) Supercoiled pBR322 plasmid DNA (lane G) was incubated with purified Topo II
a in the
m
m
M 2 (lanes A-F) or 5 (lanes M-R). The resulting DNA/Topo II complexes were digested by
a
a
a
shown are the averages of three independent experiments. :, cleavage induced by 2; -, cleavage induced by 5. The dotted line indicates the background levels of cleavable
complexes formed in the absence of drug.
carboxamide side chain lying in the major groove. Ab initio calcu-
lations of the relative contributions of dispersion and charge
transfer demonstrated that the charge transfer interactions play a
much important role in the stabilization of the complex between
compounds 2 and 5 and flanking base pairs, which might help to
explain their high cytotoxicity. DNA unwinding, relaxation and
cleavage assays, performed on 2 and 5, validated their intercalating
properties and showed a selective inhibitory effect on the catalytic
commercial suppliers and were used as received. The purity of
compounds was checked by ascending TLC on Merck’s silica gel
plates 60 F254 (0.25 mm) using UV and fluorescent light or ninhy-
drin (sol. 0.25% in MeOH). Melting points were determined by a
Kofler apparatus and are uncorrected. The purity of the synthesized
compounds was established by combustion analysis with a Thermo
Scientific (Flash 2000) analyzer and was within 0.4%. Mass mea-
surements were performed in positive ion mode using a Micromass
electrospray ionization instrument (LC-T; Waters, UK). UV spectra
were recorded on Varian Cary 5000 UVeViseNIR Spectrophotom-
eter. HPLC analysis was performed by a RP-HPLC on a semi-
preparative C18-bonded silica column (Phenomenex, Jupiter,
250 ꢂ 10 mm) using a Shimadzu SPD 10 A UV/VIS detector. MW
CEM Discover apparatus, 2.45 GHz, CEM Corporation, NC, USA was
used to perform the MW assisted synthesis. NMR measurements
(data reported in ppm) were performed on a Bruker AMX-500
spectrometer equipped with a Bruker X-32 computer, using the
UXNMR software package. NMR spectra were measured at
500 MHz (1H) and 125 MHz (13C). The chemical shifts are refer-
enced to 13CDCl3 and CDCl3 solvent signals at 77.0 and 7.26 ppm,
respectively. Standard pulse sequences were employed for magni-
activity of human Topo II
lized cleavable complexes between Topo II
that they can be considered as Topo II poisons. Further study
a
over Topo II
b
. Moreover, 2 and 5 stabi-
a
and DNA, indicating
a
indicated that compound 5 significantly induced DNA damage and
arrested cancer cells at G2/M phase. All together, these character-
istics indicate that this new series of L-lysine-conjugated pyr-
idophenoxazinones, acting in multimodal way, are endowed with
advantageous antitumor therapeutic potential. The molecule 5
provides interesting opportunities to develop novel DNA-targeted
anticancer agents.
4. Experimental protocols
tude COSY. HMQC and HMBC experiments were optimized for 1JC-H
4.1. Chemistry
2,3
135 Hz and
J
10 Hz, respectively. Me4Si was used as the in-
C-H
ternal reference.
Reagents, starting materials, and solvents were purchased from

S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
13
Fig. 9. Compound 5 induces DNA double strands break in MCF-7 cells. Cells were seeded in 24-well tissue-culture plates and after 70% confluence were treated with 5 for 24 h. After
treatment, cells were processed for comet assay as detailed under Experimental protocols section. (A) Images of MCF-7 cells after treatment with the negative control (nontreated),
VP-16 (positive control, 10
treated cells) in pixels with corresponding to concentration. (C) MCF-7 cells were incubated with DMSO or 5 or 10
harvested and lysed for detection of the expression of H2AX via western blotting. (D) 5-induced cell cycle arrest at the G2/M phase in MCF-7 cells. Cells were treated with vehicle
and 0.1 M 5 for 12, 24, and 48 h, and cell cycle distribution was assessed by flow cytometry.
mM), and compound 5 (1, 5, and 10
mM) showing comet formation. (B) Graphical representation of the selected comet lengths of MCF-7 (control and
mM 5 for the indicated time periods. After treatment, cells were
g
m
4.1.1. Synthesis of 5-oxo-5H-pyridophenoxazine carboxylic acid (1,
10, 100, 1000)
(0.01 mol) and gently warmed until the dark color showed the
complex formation. After 12 h at refluxing temperature, the reac-
tion mixtures were evaporated in vacuo and the solid residues,
dissolved in CHCl3, were washed with acidified water (HCl 6 N) to
break the Co(II)-complex and extracted in CHCl3. After purification
on silica gel column (CHCl3:CH3OH, 8:2 as eluent), the pyr-
idophenoxazinones 1-1′ and 100-1000 were obtained with yields
16e18% and 20e23%, respectively.
4.1.1.1. Conventional procedure. The solutions of 3-amino-4-
hydroxybenzoic acid and 3-hydroxy-4-amino-benzoic acid
(0.01 mol) in methanol-acetic acid (50:50; vv, 10 mL) were added,
respectively, dropwise to an equimolar solution of quinolin-5,8-
dione (0.01 mol) in acetic acid (10 mL). The mixtures were stirred
and refluxed for 12 h and evaporated in vacuo. The solid residues,
extracted in CHCl3, were purified on silica gel column
(CHCl3:CH3OH, 8:2 as eluent). The pyridophenoxazinones 1-1′ and
100-1000 were obtained with yields 10e12% and 20e25%, respectively.
To increase the very poor yields of compounds 1 and 100, the re-
ported reactions were repeated in presence of Co(II) acetate
4.1.1.2. MW assisted procedure. In MW tubs, the solutions of qui-
nolin-5,8-dione (0.1 mmol), cobalt acetate and 3-amino-4-
hydroxybenzoic acid or 3-hydroxy-4-amino-benzoic acid
(0.1 mmol) in acetic acid (5 mL) were added, respectively. The tubes

14
S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
were shook to dissolve the reagents until to obtain dark orange
solutions and were irradiated for 15 min at 50 W of power. The
reaction was stopped and oxygen was bubbled in the tube for 1 min.
After 15 min, the mixture was evaporated under reduced pressure,
washed with THF and filtered on Whatmann paper. Chromatog-
raphy of the products on silica gel columns (CHCl3/MeOH, 8:2 as
eluent) afforded the pure compounds. The yields of 1 and 100 were
35% and 40%, respectively. According to the above reported pro-
cedure compounds 1′ and 1000 were prepared. In this case, no cobalt
acetate was added to reaction mixture allowing the amine groups
of the 3-Amino-4-hydroxybenzoic acid and 3-hydroxy-4-amino-
benzoic acid aminophenols to attack the quinoline quinone 8-
carbonyl group. Under the reported conditions the following
products were obtained whit yields 50%, 67%.
butoxycarbonyl)amino)hexanoate (B) (1.64 g, 3.14 mmol) used
without further purification. Yield 90%; 1H NMR (500 MHz, CDCl3)
d
8.83 (s, 1H), 8.43 (s, 1H), 7.78e7.77 (m, J ¼ 7.5 Hz, 2H), 7.62e7.61
(d, J ¼ 7.3 Hz, 2H), 7.41 (t, J ¼ 7.4 Hz, 2H), 7.32 (t, J ¼ 7.4 Hz, 2H), 5.38
(d, J ¼ 7.1 Hz,1H), 4.56 (bs,1H), 4.35e4.32 (m, 2H), 4.24 (m, 2H), 3.11
(bs, 1H), 1.83 (dd, J ¼ 14.6, 9.5 Hz, 1H), 1.67 (m, 1H), 1.59e1.19 (m,
22H); 13C NMR (50 MHz, CDCl3) 172.13, 164.24, 156.32, 144.04,
141.51, 127.84, 127.21, 125.25, 120.06, 82.09, 79.10, 66.76, 53.96,
46.87, 39.84, 31.95, 29.19, 28.00, 27.59, 21.84; MS (ESI), m/z: 525
(Mþþ1); Anal. Calcd. for C30H40N2O6: C, 68.68%; H, 7.68%; N, 5.34%;
Found: C, 68.70%; H, 7.70%; N, 5.36%.
A solution of compound B (1.64 g, 3.13 mmol) in CH3CN (51 mL)
at 0 ꢁC was added with piperidine (1.12 mL, 3.13 mmol). The
mixture was stirred for 30 min and warmed up to room tempera-
ture. After 3 h, the solvent was evaporated under reduced pressure
and replaced by CHCl3. The solution washed with Brine, dried
(Na2SO4), evaporated under reduced pressure gave a crude residue
that, purified by chromatography on silica gel column
(CHCl3:MeOH, 9:1 as eluent) afforded the oily product C (926 mg,
4.1.2. 5-Oxo-5H-pyrido[3,2-a]phenoxazine-9-carboxylic acid (1)
Yield 38%; 1H NMR (500 MHz, CDCl3)
d
9.11 (dd, J ¼ 4.5, 1.5 Hz,
1H), 9.05 (dd, J ¼ 8.1,1.5 Hz,1H), 7.90 (d, J ¼ 7.9 Hz,1H), 7.86 (m, 2H),
7.74 (dd, J ¼ 8.1, 4.5 Hz,1H) 6.65 (s,1H); 13C NMR (100 MHz, CD3OD)
d
180.53, 163.95, 154.34, 149.93, 148.27, 139.92, 136.55, 129.81,
3.07 mmol). Yield 98%; 1H NMR (500 MHz, CDCl3)
3.26 (dd, J ¼ 7.3, 5.5 Hz, 2H), 3.08 (d, J ¼ 6.2 Hz, 2H), 1.66 (m, 2H),
1.65 (m, 2H), 1.64 (m, 18H); 13C NMR (125 MHz, CDCl3)
176.20,
d 4.65 (s, 1H),
126.12, 123.56, 122.17, 120.13, 118.33,111.45, 105.39; MS (ESI), m/z:
293 (Mþþ1); Anal. Calcd. for C16H8N2O4: C, 65.76%; H, 2.76%; N,
9.59%; Found: C, 65.65%; H, 2.77%; N, 9.63%.
d
156.73, 81.27, 79.34, 55.09, 40.50, 34.67, 29.90, 28.49, 28.13, 22.90;
MS (ESI), m/z: 303 (Mþþ1); Anal. Calcd. for C15H30N2O4: C, 59.57%;
H, 10.00%; N, 9.26%; Found: C, 59.60%; H, 10.01%; N, 9.23%.
4.1.3. 5-Oxo-5H-pyrido[2,3-a]phenoxazine-9-carboxylic acid (10)
Yield 40%; 1H NMR (600 MHz, CDCl3)
d
9.08 (dd, J ¼ 4.5, 1.7 Hz,
1H), 8.64 (dd, J ¼ 8.1, 1.7 Hz, 1H), 8.12e7.96 (m, 2H), 7.92 (s, 1H),
4.1.7. (S)-tert-Butyl 6-amino-2-((tert-butoxycarbonyl)amino)
hexanoate (F)
7.90e7.88 (dd, J ¼ 8.1, 4.5 Hz, 1H), 6.53 (s, 1H); 13C NMR (100 MHz,
CD3OD)
d
182.12, 165.44, 161.03, 154.72, 153.09, 151.05, 147.88,
A solution of commercial Boc-L-lys(Cbz)OH (D) (1 g, 2.63 mmol)
143.86, 138.35, 134.31, 131.35, 128.37,123.75, 122.05, 118.12, 107.71;
MS (ESI), m/z: 293 (Mþþ1); Anal. Calcd. for C16H8N2O4: C, 65.76%;
H, 2.76%; N, 9.59%; Found: C, 65.78%; H, 2.73%; N, 9.64%.
in CH2Cl2 (5.26 mL) and tert-butyl trichloroacetimidate (0.94 mL,
5.26 mmol) in cyclohexane (5.26 mL) was added with boron tri-
fluoride diethyl etherate (52.6 mL, 2.63 mmol). The mixture was
stirred for 24 h at room temperature. The solvent was evaporated
under reduced pressure and replaced by CHCl3. The solution,
washed with Brine, dried (Na2SO4) and evaporated under reduced
pressure, gave a crude residue that, purified chromatographically
on silica gel column (CH2Cl2 as eluent) afforded the oily product (S)-
4.1.4. 5-Oxo-5H-pyrido[3,2-a]phenoxazine-10-carboxylic acid (100)
Yield 50%; 1H NMR (600 MHz, CDCl3)
d
9.10 (m, 1H), 9.03 (m,
1H), 8.36 (m, 1H), 8.14 (m, 1H), 7.98e7.88 (m, 2H), 7.62 (s, 1H), 6.63
(s, 1H); 13C NMR (150 MHz, DMSO‑d6)
182.33, 165.19, 161.23,
d
153.66, 151.93, 147.05,144.04, 138.77, 133.47, 133.14, 130.75, 126.36,
123.99, 123.67, 115.71, 105.53; MS (ESI), m/z: 293 (Mþþ1); Anal.
Calcd. for C16H8N2O4: C, 65.76%; H, 2.76%; N, 9.59%; Found: C,
65.83%; H, 2.71%; N, 9.49%.
tert-butyl
6-(((benzyloxy)carbonyl)amino)-2-((tert-butox-
ycarbonyl)amino)hexanoate (E) (882 mg, 2.03 mmol) used without
further purification. Yield 77%; 1H NMR (500 MHz, CDCl3)
d
7.36e7.33 (m, 5H), 5.09 (s, 2H), 5.06 (bs, 1H), 4.82 (bs, 1H), 4.15
(dd, J ¼ 11.6, 6.4 Hz, 1H), 3.21e3.17 (dd, J ¼ 12.7, 6.4 Hz, 2H), 1.77 (m,
1H), 1.53e1.39 (m, 23H); 13C NMR (100 MHz, CDCl3)
171.68,
4.1.5. 5-Oxo-5H-pyrido[2,3-a]phenoxazine-10-carboxylic acid (1000)
d
Yield 67%; 1H NMR (400 MHz, CDCl3)
d
9.16 (dd, J ¼ 4.6, 1.7 Hz,
163.63, 156.22, 155.19, 136.07, 128.01, 127.57, 81.48, 79.31, 66.11,
53.34, 40.14, 31.87, 28.82, 27.81, 27.47, 21.87; MS (ESI), m/z: 437
(Mþþ1); Anal. Calcd. for C23H36N2O6: C, 63.28%; H, 8.31%; N, 6.42%;
found: C, 63.30%; H, 8.29%; N, 6.39%.
1H), 8.56 (dd, J ¼ 7.9, 1.7 Hz, 1H), 8.40 (d, J ¼ 1.7 Hz, 1H), 8.17 (dd,
J ¼ 8.5, 1.7 Hz, 1H), 7.91 (dd, J ¼ 7.9, 4.6 Hz, 1H), 7.62 (d, J ¼ 8.5 Hz,
1H), 6.57 (s, 1H); 13C NMR (150 MHz, DMSO‑d6)
d 182.74, 165.01,
164.34, 154.57, 150.82, 150.68, 139.54, 138.87, 136.73, 135.91, 129.99,
129.63, 127.72, 126.91, 118.72, 106.90; MS (ESI), m/z: 293 (Mþþ1);
Anal. Calcd. for C16H8N2O4: C16H8N2O4: C, 65.76%; H, 2.76%; N,
9.59%; Found: C, 65.86%; H, 2.81%; N, 9.54%.
A solution of E (882 mg, 2.03 mmol) in MeOH dry (8 mL) was
degassed and added with 10 wt % Pd/C (43 mg, 0.406 mmol). H2
was bubbled, and stirring was performed overnight. The reaction
mixture was filtered through a Celite bed and washed with meth-
anol. The evaporation of the solvent under reduced pressure gave
the oily product F (552 mg, 1.83 mmoli, 90% yield). Yield 90%; 1H
4.1.6. (S)-tert-Butyl 2-amino-6-((tert-butoxycarbonyl)amino)
hexanoate (C)
NMR (500 MHz, CDCl3)
d 5.11 (m, 1H), 4.16 (m, 1H), 3.15 (m, 2H),
A
solution of commercial Fmoc-L-lys(Boc)OH (A) (1.5 g,
2.78 (m, 2H), 1.78 (m, 1H), 1.58 (m, 3H), 1.38 (m, 18H); 13C NMR
3.2 mmol) in CH2Cl2 (6.5 mL) was added with a solution of tert-
butyl trichloroacetimidate (1.15 mL, 6.4 mmol) in cyclohexane
(125 MHz, CDCl3) d 172.33, 155.74, 81.73, 79.52, 53.57, 40.88, 32.31,
31.06, 27.96, 27.63, 22.05; MS (ESI), m/z: 303 (Mþþ1); Anal. Calcd.
for C15H30N2O4: C, 59.57%; H, 10.00%; N, 9.26%; found: C, 59.60%; H,
9.98%; N, 9.25%.
(6.5 mL) and with boron trifluoride diethyl etherate (64
ml,
3.2 mmol). The mixture was stirred for 24 h at room temperature
and, subsequently, the solvent was evaporated under reduced
pressure and replaced by CHCl3. This solution, washed with Brine,
dried (Na2SO4) and evaporated under reduced pressure gave a
crude residue that, purified by chromatography on a silica gel col-
umn (CHCl3 as eluent), afforded the oily product (S)-tert-butyl 2-
((((9H-fluoren-9-yl)methoxy)carbonyl)amino)-6-((tert-
4.1.8. General procedure for the preparation of compounds 2e5 and
20-50
Step 1. The solutions of the carboxylic acids 1-1000 (0.50 mmol) in
anhydrous THF (24 mL), at 0 ꢁC, were added with HOBt

S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
15
(Mþþ1); Anal. Calcd. for C31H36N4O7: C, 64.57%; H, 6.29%; N,
(1.0 mmol) in a portion, respectively. After 30 min, a solution of
-lys(Boc)OtBu (C) (0.70 mmol) in anhydrous THF (24 mL) and
EDC (100 L, 0.56 mmol) was added to the reaction mixtures.
L
9.72%; Found: C, 64.55%; H, 6.30%; N, 9.70%.
m
The mixtures were stirred for 12 h at 0 ꢁC and allowed to worm
slowly to room temperature. The solvent was evaporated under
reduced pressure, replaced with AcOEt and the mixtures were
washed with saturated aqueous solutions of NH4Cl, Brine and
dried (Na2SO4). The evaporation of the solvent under reduced
pressure gave the crude residues that, purified chromato-
graphically on silica gel (CHCl3:MeOH, 95:5 as eluent), afforded
the coupling products 2a-2′a and 3a-3′a. The above reported
4.1.8.5. (S)-tert-butyl 2-((tert-butoxycarbonyl)amino)-6-(5-oxo-5H-
pyrido[3,2-a]phenoxazine-10-carboxamido)hexanoate
(4a).
9.13 (d, J ¼ 3.7 Hz, 1H),
Yield 70%; oil; 1H NMR (500 MHz, CDCl3)
d
9.07e9.05 (d, J ¼ 7.8 Hz, 1H), 8.30 (bs, 1H), 8.06e8.04 (d, J ¼ 8.3 Hz,
1H), 7.77e7.76, 7.77e7.75 (dd, J ¼ 8.1, 4.3 Hz, 1H), 7.42e7.41 (d,
J ¼ 8.3 Hz, 1H), 6.68 (s, 1H), 6.55 (bs, 1H), 5.16e5.14 (d, J ¼ 7.1 Hz,
1H), 4.22 (m, 1H), 3.54 (m, 2H), 1.84 (m, 1H), 1.74 (m, 1H), 1.50e1.44
(m, 22H). 13C NMR (125 MHz, CDCl3)
d 181.09, 172.26, 166.22,
procedure performed by using the acids 1-1000 and Boc-
L-lysOtBu
154.03, 154.01, 150.82, 147.32, 145.98, 145.05, 133.42, 132.55, 131.48,
131.45, 129.01, 128.09, 126.55, 116.45, 109.36, 81.86, 79.83, 53.38,
39.74, 32.68, 29.35, 28.38, 27.95, 27.66; MS (ESI), m/z: 577 (Mþþ1);
Anal. Calcd. for C31H36N4O7: C, 64.57%; H, 6.29%; N, 9.72%; Found:
C, 64.59%; H, 6.27%; N, 9.69%.
(F) yielded the corresponding compounds 4a-4′a and 5a-5′a.
4.1.8.1. (S)-tert-butyl 6-((tert-butoxycarbonyl)amino)-2-(5-oxo-5H-
pyrido[3,2-a]phenoxazine-9-carboxamido)hexanoate
(2a). Yield
9.14 (dd, J ¼ 4.7, 1.6 Hz, 1H), 9.08
90%; 1H NMR (500 MHz, CDCl3)
d
4.1.8.6. (S)-tert-butyl 2-((tert-butoxycarbonyl)amino)-6-(5-oxo-5H-
pyrido[2,3-a]phenoxazine-10-carboxamido)hexanoate
(40a).
(dd, J ¼ 8.0,1.6 Hz,1H), 7.93 (d, J ¼ 8.1 Hz,1H), 7.85 (m, 1H), 7.84 (dd,
J ¼ 8.1 Hz, 1H), 7.76 (dd, J ¼ 8.0, 4.7 Hz, 1H), 6.95 (d, J ¼ 6.0 Hz, 1H),
6.68 (s, 1H), 4.69 (dd, J ¼ 12.1, 6.0 Hz, 1H), 4.60 (bs, 1H), 3.15 (m, 2H),
1.99 (m, 1H), 1.83 (m, 1H), 1.57 (m, 22H). 13C NMR (100 MHz, CDCl3)
Yield 85%; oil; 1H NMR (500 MHz, CDCl3)
d
9.18e9.17, 9.17e9.16 (dd,
J ¼ 4.5, 1.6 Hz, 1H), 8.64e8.63, 8.63e8.62 (dd, J ¼ 7.9, 1.6 Hz, 1H),
8.44 (bs, 1H), 8.11e8.10, 8.10e8.09 (dd, J ¼ 8.5, 2.0 Hz, 1H),
7.76e7.75, 7.75e7.74 (dd, J ¼ 7.9, 4.5 Hz, 1H), 7.42-7-40 (d,
J ¼ 8.5 Hz, 1H), 6.55 (s, 1H), 6.39 (bs, 1H), 5.12e5.10 (d, J ¼ 8.0 Hz,
1H), 4.19 (m, 1H), 3.52 (m, 2H), 1.84 (m, 1H), 1.66 (m, 1H), 1.53e1.50
d
187.39, 182.54, 172.04, 156.98, 154.17, 151.25, 148.30, 144.24,
137.69, 133.61, 131.10, 130.37, 126.46, 124.28, 115.53, 109.33, 82.62,
79.06, 53.04, 41.86, 39.69, 31.73, 29.96, 29.33, 27.98, 27.63; MS (ESI),
m/z: 577 (Mþþ1); Anal. Calcd. for C31H36N4O7: C, 64.57%; H, 6.29%;
N, 9.72%; Found: C, 64.55%; H, 6.30%; N, 9.74%.
(m, 22H). 13C NMR (125 MHz, CDCl3)
d 182.93, 172.27, 165.83,
162.69, 155.83, 154.11, 152.10, 147.78, 147.43, 146.28, 134.64, 132.53,
132.27, 129.26, 128.96, 126.62, 116.40, 107.79, 81.88, 79.58, 53.41,
39.74, 32.49, 29.33, 28.66, 27.94, 27.65; MS (ESI), m/z: 577 (Mþþ1);
Anal. Calcd. C31H36N4O7: C, 64.57%; H, 6.29%; N, 9.72%; Found: C,
64.55%; H, 6.28%; N, 9.70%.
4.1.8.2. (S)-tert-butyl 6-((tert-butoxycarbonyl)amino)-2-(5-oxo-5H-
pyrido[2,3-a]phenoxazine-9-carboxamido)hexanoate
(20a).
Yield 90%; oil; 1H NMR (500 MHz, CDCl3)
d
9.19 (m, 1H), 8.65 (d,
J ¼ 7.9 Hz, 1H), 8.18 (d, J ¼ 8.1 Hz, 1H), 7.86 (s, 1H), 7.83 (dd,
J ¼ 8.1 Hz, 1H), 7.75 (m, 1H), 6.96 (m, 1H), 6.55 (s, 1H), 4.67 (m, 1H),
4.60 (bs, 1H), 3.13 (m, 2H), 1.98 (m, 1H), 1.84 (m, 1H), 1.57 (m, 2H),
4.1.8.7. (S)-tert-butyl 2-((tert-butoxycarbonyl)amino)-6-(5-oxo-5H-
pyrido[3,2-a]phenoxazine-9-carboxamido)hexanoate
(5a). Yield
9.06 (d, J ¼ 3.1 Hz, 1H), 9.00
1.51 (m, 20H). 13C NMR (100 MHz, CDCl3)
d
182.22, 171.51, 170.13,
70%; oil; 1H NMR (400 MHz, CDCl3)
(dd, J ¼ 8.1, 1.5 Hz, 1H), 7.83 (s, J ¼ 8.3 Hz, 1H), 7.78 (m, 2H), 7.72 (m,
1H), 6.87 (bs, 1H), 6.59 (s, 1H), 5.18 (d, J ¼ 8.0 Hz, 1H), 4.18 (m, 1H),
3.51 (m, 2H), 1.80 (m, 1H), 1.72 (m, 1H), 1.50e1.44 (m, 22H). 13C NMR
d
164.73, 164.27, 155.95, 153.71, 152.12, 147.76, 144.12, 134.75, 134.23,
131.37, 128.90, 128.72, 126.41, 123.85, 115.23, 107.40, 82.56, 80.55,
52.82, 39.86, 32.00, 29.59, 28.25, 27.91; MS (ESI), m/z: 577 (Mþþ1);
Anal. Calcd. for C31H36N4O7: C, 64.57%; H, 6.29%; N, 9.72%; Found: C,
64.61%; H, 6.29%; N, 9.73%.
(125 MHz, CDCl3)
d 187.32, 172.10, 168.88, 165.12, 156.09, 153.96,
147.90, 146.99, 135.90, 134.67, 133.82, 132.90, 131.51, 129.14, 128.03
127.55, 117.53, 109.50, 82.99, 79.41, 53.60, 41.15, 32.95, 29.33, 29.30,
28.19, 27.65; MS (ESI), m/z: 577 (Mþþ1); Anal. Calcd. for
C31H36N4O7: C, 64.57%; H, 6.29%; N, 9.72%; Found: C, 64.60%; H,
6.30%; N, 9.69%.
4.1.8.3. (S)-tert-butyl 6-((tert-butoxycarbonyl)amino)-2-(5-oxo-5H-
pyrido[3,2-a]phenoxazine-10-carboxamido)hexanoate
(3a).
Yield 73%; oil; 1H NMR (500 MHz, CDCl3)
d
9.21 (bs, 1H), 9.16e9.15
(dd, J ¼ 8.0, 1.6 Hz, 1H), 8.34 (s, 1H),8.09e8.08 (d, J ¼ 8.2 Hz, 1H),
7.86 (m, 1H), 7.44 (d, J ¼ 8.2 Hz, 1H), 6.99 (d, J ¼ 7.2 Hz, 1H), 6.70 (s,
1H), 4.73e4.69 (m, 1H), 3.14 (m, 2H), 2.03e1.98 (m, 2H), 1.84 (m,
4.1.8.8. (S)-tert-butyl 2-((tert-butoxycarbonyl)amino)-6-(5-oxo-5H-
pyrido[2,3-a]phenoxazine-9-carboxamido)hexanoate
(50a).
2H), 1.52e1.47 (m, 21H). 13C NMR (125 MHz, CDCl3)
d
184.32,
Yield 90%; oil; 1H NMR (400 MHz, CDCl3)
d
9.18 (m, 1H), 8.67 (d,
172.40, 168.61, 165.62, 156.69, 153.86, 147.93, 146.91, 135.30, 134.11,
132.62, 132.11, 131.71, 129.74, 128.13, 127.35, 116.53, 109.63, 82.89,
79.31, 53.20, 41.25, 32.15, 29.56, 29.26, 28.29, 27.95; MS (ESI), m/z:
577 (Mþþ1); Anal. Calcd. for C31H36N4O7: C, 64.57%; H, 6.29%; N,
9.72%; Found: C, 64.60%; H, 6.30%; N, 9.70%.
J ¼ 7.7 Hz, 1H), 8.10 (m, 1H), 7.83 (s, 1H), 7.80e7.78 (m, 2H), 6.78 (bs,
1H), 6.51 (s, 1H), 5.14 (m, 1H), 4.16 (m, 1H), 3.50 (m, 2H), 1.82e1.69
(m, 6H), 1.49e1.43 (m, 18H). 13C NMR (100 MHz, CDCl3)
d 182.24,
171.71, 165.43, 161.12, 155.54, 153.62, 151.92, 146.95, 143.93, 138.36,
134.44, 131.04, 128.82, 126.40, 123.84, 123.47, 115.16, 107.49, 81.92,
79.28, 53.52, 39.92, 32.82, 28.51, 28.20, 27.90, 22.53; MS (ESI), m/z:
577 (Mþþ1); Anal. Calcd. for C31H36N4O7: C, 64.57%; H, 6.29%; N,
9.72%; Found: C, 64.60%; H, 6.31%; N, 9.75%.
4.1.8.4. (S)-tert-butyl 6-((tert-butoxycarbonyl)amino)-2-(5-oxo-5H-
pyrido[2,3-a]phenoxazine-10-carboxamido)hexanoate
(30a).
Yield 88%; oil; 1H NMR (500 MHz, CDCl3)
d
9.14e9.13 (d, J ¼ 2.9 Hz,
1H), 8.59 (d, J ¼ 4.7 Hz, 1H), 8.48 (s, 1H), 8.09e8.08 (d, J ¼ 8.5 Hz,
1H), 7.72e7.71, 7.71e7.70 (dd, J ¼ 7.5, 4.7 Hz, 1H), 7.40e7.38 (d,
J ¼ 8.5 Hz, 1H), 6.94 (bs, 1H), 6.53 (s, 1H), 4.70e4.67, 4.68e4.66 (dd,
J ¼ 12.3, 6.8 Hz,1H), 4.61 (bs, 1H), 3.12 (m, 2H), 1.96 (m, 1H), 1.55 (m,
Step. 2. All the compounds 2a-5a and 2′a-5′a (0.54 mmol) were
dissolved in formic acid (31 mL) and the solutions, stirred for
72 h at room temperature, were evaporated under reduced
pressure. The crude products were purified by Flow Lab semi-
preparative HPLC System on a Phenomenex Axia Packed Luna
1H),1.52e1.45 (m, 22H). 13C NMR (125 MHz, CDCl3)
d 182.33,171.41,
164.76, 155.85, 153.68, 151.61, 147.42, 146.98, 146.05, 134.22, 132.19,
131.96, 131.50, 129.21, 128.59, 126.30, 116.18, 107.64, 82.45, 78.93,
52.96, 40.08, 32.16, 29.51, 28.25, 27.89, 22.19;; MS (ESI), m/z: 577
C18(2) column (50
CH3CN:H2O:TFA (80:20:0.1) as eluent with a flow rate of 12 mL/
min.
ꢂ
21.20 mm, 5 mm, 100 Å) using

16
S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
4.1.8.9. (S)-6-amino-2-(5-oxo-5H-pyrido[3,2-a]phenoxazine-9-
28.65, 24.75; MS (ESI), m/z: 421 (Mþþ1); Anal. Calcd. for
C22H20N4O5: C, 62.85%; H, 4.79%; N, 13.33%; Found: C, 62.90%; H,
4.81%; N, 13.31%.
carboxamido)hexanoic acid (2). Yield 88%; 1H NMR (400 MHz,
CD3OD)
d
9.29e9.30 (d, J ¼ 7.2 Hz, 1H), 9.09 (bs, 1H), 8.07e8.04 (d,
2H), 7.97 (m, 2H), 6.70 (s, 1H), 4.65 (m, 1H), 2.97 (m, 2H), 2.10 (m,
1H), 1.94 (m, 1H), 1.75 (m, 2H), 1.51 (m, 2H). 13C NMR (100 MHz,
4.1.8.15. (S)-2-amino-6-(5-oxo-5H-pyrido[2,3-a]phenoxazine-10-
CD3OD)
d
187.40, 186.04, 170.84, 156.92, 154.97, 150.25, 148.00,
carboxamido)hexanoic acid (40). Yield 83%; oil; 1H NMR (500 MHz,
146.24, 140.61, 140.10, 132.37, 128.99, 126.16, 124.18, 122.80, 115.63,
108.13, 56.04, 41.66, 37.69, 33.73, 24.63; MS (ESI), m/z: 421 (Mþþ1);
Anal. Calcd. for C22H20N4O5: C, 62.85%; H, 4.79%; N, 13.33%;
Found: C, 62.83%; H, 4.80%; N, 13.28%.
CD3OD)
d
9.08 (bs, 1H), 8.62e8.60 (d, J ¼ 8.0 Hz, 1H), 8.39 (d,
J ¼ 4.0 Hz, 1H), 8.08e8.06 (d, J ¼ 7.1 Hz, 1H), 7.86 (s, 1H), 7.51e7.49
(d, J ¼ 7.9 Hz, 1H), 6.50 (s, 1H), 3.68 (s, 1H), 3.42 (m, 2H), 1.83 (m,
2H),1.70 (m, 2H),1.48 (m, 2H). 13C NMR (125 MHz, CD3OD)
d 186.27,
177.01, 165.07, 155.89, 154.75, 148.20, 146.98, 142.90, 137.99, 137.53,
129.97, 128.87, 124.55, 124.01, 122.35, 119.08, 109.77, 55.61, 38.97,
33.20, 26.52, 22.44; MS (ESI), m/z: 421 (Mþþ1); Anal. Calcd. for
C22H20N4O5: C, 62.85%; H, 4.79%; N, 13.33%; Found: C, 62.90%; H,
4.81%; N, 13.33%.
4.1.8.10. (S)-6-amino-2-(5-oxo-5H-pyrido[3,2-a]phenoxazine-10-
carboxamido)hexanoic acid (3). Yield 85%; oil; 1H NMR (400 MHz,
CD3OD)
d
9.42 (d, J ¼ 7.0 Hz, 1H), 9.10 (m, 1H), 8.46 (s, 1H), 8.15 (m,
2H), 7.59e7.57 (d, J ¼ 8.5 Hz, 1H), 6.71 (s, 1H), 4.67 (m, 1H), 2.92 (m,
2H), 2.04 (m, 1H), 1.85 (m, 1H), 1.75 (m, 2H), 1.68 (m, 2H). 13C NMR
(125 MHz, CD3OD)
d
186.42, 175.00, 165.62, 156.66, 155.86, 147.83,
4.1.8.16. (S)-2-amino-6-(5-oxo-5H-pyrido[2,3-a]phenoxazine-9-
147.01, 139.30, 138.11, 137.12, 132.01, 130.61, 129.44, 125.73, 125.55,
118.13, 109.53, 56.20, 41.35, 32.85, 31.16, 24.85; MS (ESI), m/z: 421
(Mþþ1); Anal. Calcd. for C22H20N4O5: C, 62.85%; H, 4.79%; N,
13.33%; Found: C, 62.91%; H, 4.80%; N, 13.35%.
carboxamido)hexanoic acid (50). Yield 93%; oil; 1H NMR (500 MHz,
CD3OD)
d
9.06e9.05 (dd, J ¼ 4.2, 1.2 Hz, 1H), 8.63 (dd, J ¼ 6.1, 1.2 Hz,
1H), 8.01e8.00 (m, 2H), 7.91e7.87 (m, 2H), 6.51 (s, 1H), 3.52 (m, 1H),
3.39 (m, 2H), 1.83 (m, 2H), 1.56 (m, 1H), 1.36 (m, 3H). 13C NMR
(125 MHz, CD3OD)
d 186.26, 177.40, 165.50, 156.00, 154.72, 148.89,
4.1.8.11. (S)-2-amino-6-(5-oxo-5H-pyrido[3,2-a]phenoxazine-10-
147.50, 141.00, 139.50, 138.20, 133.01, 128.28, 125.11, 124.50, 119.85,
118.00, 108.01, 53.60, 39.45, 33.22, 26.88, 22.03; MS (ESI), m/z: 421
(Mþþ1); Anal. Calcd. for C22H20N4O5: C, 62.85%; H, 4.79%; N,
13.33%; Found: C, 6288%; H, 4,80%; N, 13.35%.
carboxamido)hexanoic acid (4). Yield 94%; oil; 1H NMR (400 MHz,
CD3OD)
d
9.26e9.24 (dd, J ¼ 7.5, 4.0 Hz, 1H), 9.06 (bs, 1H),
8.40e8.39 (d, J ¼ 2.0 Hz, 1H), 8.10e8.08 (dd, J ¼ 8.6, 2.0 Hz, 1H),
7.56e754 (d, J ¼ 8.6 Hz,1H), 6.67 (s,1H), 3.98 (m,1H), 3.46e3.45 (m,
2H), 2.01 (m, 2H), 1.92 (m, 1H), 1.75 (m, 2H), 1.56 (m, 1H). 13C NMR
4.2. Chemicals, DNAs, and proteins
(100 MHz, CD3OD)
d 187.49, 176.90, 165.67, 156.00, 155.86, 147.80,
147.38, 139.99, 138.99, 138.42, 132.71, 131.50, 130.01, 129.04, 127.31,
116.13, 107.10, 52.34, 39.05, 29.44, 28.46, 22.15; MS (ESI), m/z: 421
(Mþþ1); Anal. Calcd. for C22H20N4O5: C, 62.85%; H, 4.79%; N,
13.33%; Found: C, 62.90%; H, 4.81%; N, 13.34%.
AMSA, VP-16, Doxo and AMD were purchased from Sigma-
Aldrich (Milan, Italy). Negatively supercoiled plasmid pBR322
DNA was purchased from New England Biolabs (Beverly, MA, USA).
Relaxed plasmid DNA was prepared by incubating negatively
supercoiled pBR322 DNA with human Topo I, which was purchased
4.1.8.12. (S)-2-amino-6-(5-oxo-5H-pyrido[3,2-a]phenoxazine-9-
from Topogen (Port Orange, FL, USA). The a form of human Topo II
was purchased from both Topogen. Purified human Topo IIa and IIb
carboxamido)hexanoic acid (5). Yield 95%; oil; 1H NMR (400 MHz,
CD3OD)
(m, 2H), 7.90e7.88 (m, 2H), 6.68 (s, 1H), 4.01 (t, J ¼ 5.9 Hz, 1H),
3.47e3.45 (m, 2H), 2.04e1.97 (m, 3H), 1.75 (m, 2H), 1.61 (m, 1H). 13
NMR (125 MHz, CD3OD) 187.42, 176.40, 165.82, 156.66, 155.96,
147.80, 147.31, 139.49, 138.91, 138.12, 133.01, 130.21, 129.04, 125.63,
125.50, 118.14, 109.55, 56.00, 41.45, 32.88, 31.06, 24.95; MS (ESI), m/
z: 421 (Mþþ1); Anal. Calcd. for C22H20N4O5: C, 62.85%; H, 4.79%;
N, 13.33%; Found: C, 62.81%; H, 4.77%; N, 13.31%.
d
9.28e9.26 (dd, J ¼ 8.0 4.0 Hz,1H), 9.07 (m, 1H), 8.03e8.01
were purchased from LAE Biotech International.
C
4.2.1. Cells
d
Test compounds were dissolved in DMSO at an initial concen-
tration of 200 M and were then serially diluted in culture medium.
m
Cell lines were from American type Culture Collection (ATCC).
Leukemia- and lymphoma-derived cells were grown in Roswell
Park Memorial Institute (RPMI) 1640 containing 10% fetal calf
serum (FCS), 100 U/mL penicillin G, and 100 mg/mL streptomycin.
4.1.8.13. (S)-6-amino-2-(5-oxo-5H-pyrido[2,3-a]phenoxazine-9-
Solid-tumor-derived cells were grown in their specific media sup-
plemented with 10% FCS and antibiotics. Cell cultures were incu-
bated at 37 ꢁC in a humidified 5% CO2 atmosphere. Cell cultures
were periodically checked for the absence of mycoplasma
contamination by the Hoechst staining method.
carboxamido)hexanoic acid (20). Yield 85%; 1H NMR (400 MHz,
CD3OD)
d
9.09e9.08 (d, J ¼ 3.1 Hz, 1H), 8.63e8.61 (dd, J ¼ 8.0,
4.0 Hz, 1H), 8.06e8.05 (d, J ¼ 8.0 Hz, 1H), 7.96e7.84 (m, 3H), 6.50 (s,
1H), 4.57 (m, 1H), 2.95 (m, 2H), 2.02 (m, 1H), 1.85 (m, 1H), 1.57 (m,
2H), 1.52 (m, 2H). 13C NMR (100 MHz, CD3OD)
d 185.60, 179.24,
169.03, 156.09, 150.58, 149.78, 140.87, 137.45, 137.17, 133.01, 131.62,
130.71, 129.74, 120.21, 118.04, 113.59, 109.46, 42.04, 34.30, 32.34,
29.64, 25.32; MS (ESI), m/z: 421 (Mþþ1); Anal. Calcd. for
C22H20N4O5: C, 62.85%; H, 4.79%; N, 13.33%; Found: C, 62.90%; H,
4.81%; N, 13.34%.
4.2.2. Antiproliferative assays
Exponentially growing leukemia and lymphoma cells were
resuspended at a density of 1 ꢂ 105 cells/mL in RPMI containing
serial dilutions of the test drugs. Cell viability was determined after
96
h
at 37 ꢁC by the 3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide (MTT) method. The activity against
solid-tumor-derived cells was evaluated in exponentially growing
cultures seeded at 5 ꢂ 104 cells/mL and allowed to adhere for 16 h
to culture plates before the addition of drugs. Cell viability was
determined by the MTT method 4 days later.
4.1.8.14. (S)-6-amino-2-(5-oxo-5H-pyrido[2,3-a]phenoxazine-10-
carboxamido)hexanoic acid (30). Yield 83%; oil; 1H NMR (500 MHz,
CD3OD)
d
9.07 (bs, 1H), 8.59 (d, J ¼ 8.0 Hz, 1H), 8.41 (d, J ¼ 8.0 Hz,
1H), 8.10 (d, J ¼ 8.0 Hz,1H), 7.85 (m, 1H), 7.49 (d, J ¼ 7.9 Hz, 1H), 6.50
(s, 1H), 4.63 (m, 1H), 2.85 (m, 2H), 2.12 (m, 1H), 1.94 (m, 1H), 1.76 (m,
2H), 1.61 (m, 2H). 13C NMR (125 MHz, CD3OD)
d
184.84, 168.93,
4.2.3. Linear regression analysis
155.31, 154.35, 149.31, 148.53, 136.20, 134.58, 134.17, 133.77, 133.31,
132.14, 131.08, 130.60, 128.80, 118.06, 108.65, 54.80, 41.15, 32.65,
Tumor cell growth at each drug concentration was expressed as
the percentage of untreated controls, and the concentration

S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
17
resulting in 50% (IC50) growth inhibition was determined by linear
regression analysis.
The double reciprocal plot of 1/DA vs. 1/[M] is linear and the
binding constant (K) can be estimated by computing the ratio of the
intercept to the slope.
4.3. UVevisible spectral measurements
4.4. DNA unwinding assay
The UVeVisible spectra of free DNA and its complexes with
compound 5 were recorded on PerkinElmer 550S spectrometer.
Quartz cuvette with a path length of 1 cm was employed for the
examination. DNA solution of 0.2 mM concentration was used with
varying concentration of drug (1 ꢂ 10ꢀ1 to 3 ꢂ 0ꢀ2 mM). Binding
constant for the complexation between 5 and DNA was determined
using the method described by Kanakis et al. [69]. Calculated
binding constant depends on the supposition that only one kind of
interaction occurs between DNA (D) and 5 (M) in aqueous solution
forming one kind of complex (DM). It is also supposed that the
substrate and the ligand obey Beer-Lambert’s law for the absor-
bance of light. The absorbance of DNA solution (A0) at its total
concentration (Dt) with a path length (l) of 1 cm is:
A DNA intercalation assay was performed as described previ-
ously [52]. Briefly, reaction mixtures (20
HCl (pH 7.5 at 23 ꢁC), 2.5 mM MgCl2, 0.5 mM dithiothreitol (DTT),
50 g/mL bovine serum albumin (BSA), 300 ng of relaxed plasmid
ml) containing 50 mM Tris-
m
DNA, and 2 units of human Topo I (Topogen, Port Orange, Fl) were
incubated at 37 ꢁC for 10 min in the absence or presence of the
indicated concentrations of the various compounds. Reactions were
terminated by extraction with phenol/chloroform/isoamyl alcohol
(25:24:1, v/v) and the DNA products were analyzed by electro-
phoresis through vertical 1.2% SeaKem ME agarose (Lonza, Rock-
land, ME) gels (14 ꢂ 10 ꢂ 0.3 cm) at 2 V/cm for 15 h in TAE buffer
[50 mM Tris-HCl (pH 7.9 at 23 ꢁC), 40 mM sodium acetate and 1 mM
EDTA]. Gels were stained with EtBr and photographed using a
Stratagene Eagle Eye gel documentation system (Agilent Technol-
ogies, Santa Clara, CA).
A0 ¼ εDlDt
(1)
where εD is the molar absorptivity of free DNA. The absorbance of
solution (AM), including the total concentration of DNA (Dt) along
with the total concentration of 5 (Mt), is:
4.5. Computational chemistry
4.5.1. Ligand, DNA and protein setup
The 3D structures of 2e5 were generated with the Maestro Build
Panel implemented in Maestro 11.1 (Schrodinger, LLC, New York,
A
M / εDl½Dꢄ þ εMl½Mꢄ þ εDMl½DMꢄ
(2)
€
where [D] is the concentration of uncomplexed DNA, [M] is the
concentration of uncomplexed 5, [DM] is the concentration of
5eDNA complex, εM is the molar absorptivity of 5 and εDM is the
molar absorptivity of 5eDNA complex. After combining with the
mass balance of DNA and 5, the absorbance equation can be written
as:
NY, 2018) software package, and their protonation state at physi-
€
ological pH 7.4 was assigned using Epik 4.0 (Schrodinger, LLC, New
York, NY, 2018). The X-ray coordinates of d(GAAGCTTC)2 octamer
complexed with AMD (PDB code ¼ 209D) [56] and human Topo II
a
in complex with DNA and etoposide (PDB code ¼ 5GWK) [63] were
extracted from the Protein Data Bank [70]. The structures were then
€
prepared using the Protein Preparation Wizard of the Schrodinger
AM / εDlDt þ εMlMt þ
DM / εDM ꢀ εD ꢀ εM
The absorbance of solution (A) measured against the total con-
D
εDMl½DMꢄ
(3)
graphical interface Maestro. Water molecules and co-crystallized
ligands were removed, hydrogen atoms were added and the cen-
tral intercalation site was utilized for docking experiments.
Dε
4.5.2. Molecular docking
centration of 5 as reference is
AutoDock 4.2 software package [53] was employed for docking
simulations. Ligand, DNA and protein structures were converted to
AD4 format files using the graphical user interface AutoDockTools
(ADT); the GesteigereMarsili partial charges and Kollman charges
were then assigned to the ligand and nucleic acid atoms, respec-
tively. The grid box was centered on the intercalation site in case of
A / εDlDt þ εDMl½DMꢄ
The stability constant (KDM) for the formation of the complex
(DM) can be given as:
D
(4)
K
DM/½DMꢄ
(5)
=
½Dꢄ½Mꢄ
DNA and on the etoposide binding site in case of Topo IIa, allowing
the ligand to explore the whole conformational space. Thus, grid
points of 60 ꢂ 60 ꢂ 60 with a 0.375 Å spacing were calculated
around the docking area for all the ligand atom types using Auto-
Grid 4.2. A total of 200 separate docking calculations were executed
for each ligand. Each docking run consisted of 25 million energy
evaluations using the Lamarckian genetic algorithm local search
method. Otherwise default docking parameters were applied.
Docking conformations were clustered on the basis of the root
mean square deviation (rmsd) between the Cartesian coordinates
of the ligand atoms (cutoff ¼ 0.8 Å) and were ranked based on the
AutoDock scoring function [71].
Combining Eqs. (4) and (5)
D
A/KDMDεDMl½Dꢄ½Mꢄ
(6)
D
A ¼ A ꢀ AO
From the mass balance equation Dt ¼ [D] þ [DM], we get
[D] ¼ Dt/(1 þ KDM[M]), which gives
D
A
DtKDMDεDM½Mꢄ
1 þ KDM½Mꢄ
/
(7)
l
With the purpose to check the reliability of the above reported
docking protocol, docking validation was performed using a
simplified model of AMD, consisting of the phenoxazone ring with
the two cyclic depsipeptides replaced by methylamide groups. We
obtained an almost perfect match between the simplified AMD
model and the crystallized conformation of the phenoxazone ring
of AMD, with a RMSD value of 0.80 Å (Fig. S18 in Supplementary
data). Therefore, this methodology was consistently used to
There is a hyperbolic relation between the free drug molecule
concentration and its interaction with DNA. Linear transformation
of Eq. (6) is obtained by taking the reciprocal of both side of Eq. (7)
that gives:
l
D
1
1
/
þ
(8)
A
DtKDMDεDM½Mꢄ
D
tDεDM

18
S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
perform all the docking experiments within the study. Figures were
were harvested by trypsinization, washed with phosphate buffered
saline (PBS) and resuspended in ice-cold PBS. All steps were per-
formed at room temperature under dimmed or yellow light to
€
generated with Pymol 2.0 (Schrodinger, LLC, New York, NY, 2018).
4.5.3. Quantum mechanics calculations
prevent DNA damage unless otherwise specified. About 25
ml of the
€
All calculations were performed with Jaguar 9.0 (Schrodinger,
resuspended cells was mixed with 75 l 0.5% low-melting agarose
m
LLC, New York, NY, 2017). The structures of pyridophenoxazinones
(where the carboxamide side chains at 9 and 10 positions were
replaced with methylamide groups), and CꢀG and GꢀC base pairs
(where the deoxyribose rings were replaced with methyl groups)
were optimized at the HF/6-31G** level of theory. The single point
energy calculations were performed at the LMP2/6-31G* and HF/6-
31G* levels of theory. The NBO analysis [72] was performed at the
HF/6-31G** level of theory. The interaction between filled orbitals
of one subsystem and vacant orbitals of another subsystem can be
used as a measure of the intermolecular delocalization, as it rep-
resents a deviation of the complex from the Lewis structure. The
hyperconjugative interaction energy can be deduced from the
second-order perturbation approach. This type of interaction is also
called charge transfer energy, though the electron density trans-
ferred between subsystems is relatively small but chemically sig-
nificant. It has been estimated that 0.001e of charge transfer
roughly corresponds to 1 kcal/mol of the stabilization energy [73].
Here, we use
at 37 ꢁC and spread uniformly with the side of a microtip on pre-
heated (37 ꢁC) microscopic slides. The slides were placed at 4 ꢁC in
the dark until gelling occurred and then immersed in prechilled
lysis solution (10 mM Tris, 100 mM EDTA, 2.5 M NaCl, 1% Triton
X100, 10% DMSO and pH 10.0) at 4 ꢁC. After 1 h of incubation, the
buffer was drained from slides and immersed in freshly prepared
alkaline-unwinding solution (300 mM NaOH, 1 mM EDTA and pH
13.0) at room temperature for 30e60 min. After lysis and un-
winding, the slides were placed in a horizontal electrophoresis tank
filled with freshly prepared alkaline electrophoresis buffer, and
electrophoresis was carried out at 30 V for 15 min. After electro-
phoresis, the slides were covered with neutralization buffer (0.4 M
Tris HCl and pH 7.5) for 5 min and washed twice by immersing in
distilled water for 5 min each, then in 70% ethanol for 10 min.
Thereafter, the slides were dried at room temperature for overnight
or at 45 ꢁC for 10e15 min, then stained with PI for 10 min in dark at
4
ꢁC. DNA migration was observed using fluorescence microscope
(Nikon, Japan) at a magnification of 10ꢂ . The comet tail lengths
were analyzed by TriTek CometScore™ software. Data were rep-
resented both by imaging and graphically by randomly selecting
comet lengths of MCF-7 cells.
ECT ¼ SE(2) (1)
where E(2) is all of the individual intermolecular second-order
hyperconjugative interactions between the ligand and the neigh-
boring DNA base pairs that exceed a default threshold of 0.05 kcal/
mol calculated from NBO analysis at the HF/6-31G** level of theory,
reflecting the major contribution of charge transfer to the stabili-
zation of the drug-DNA complex.
4.9. Western blot analysis
MCf-7 cells were collected after treatment with 5 for 0.5, 1 and
3 h and washed three times with ice-cold PBS. For nucleus protein
isolation, cells were lysed with radio-immunoprecipitation assay
(RIPA) lysis buffer (50 mM Tris-HCl at pH 7.5, 150 mM NaCl, 1%
Triton X-100, 10% deoxycholate, 0.1% SDS, 0.1% phenyl-
methylsulfonyl fluoride (PMSF) and proteinase inhibitor cocktails).
Equivalent amounts of proteins were resolved by SDS-PAGE and
transferred to poly(vinylidene fluoride) (PVDF) membranes (Bio-
Rad, CA, USA). Antibodies corresponding to the proteins were used.
4.6. Relaxation assay of human Topo II
a
The reaction mixture contained 20 mM Tris-HCl, pH 7.5, 7.5 mM
MgCl2, 0.5 mM DTT, 150 mM KCl, 1 mM ATP, and 200 ng of pBR322
DNA. The reaction was initiated by the addition of DNA Topo IIa and
allowed to proceed at 30 ꢁC for 15 min. Reactions were terminated
by the addition of loading buffer (0.1% SDS, 0.05% bromphenol blue,
2.5 mM EDTA, and 10% sucrose, final concentrations). The samples
were electrophoresed in 1.2% agarose gels at 0.5 V/cm for 18 h in
TBE buffer (90 mM Tris-borate, and 1 mM EDTA, pH 8). Gels were
b-actin was used as the loading control. Proteins were visualized
with a horse radish peroxidase-conjugated secondary antibody
using an enhanced chemiluminescence kit (Visual Protein, Taiwan)
for detection.
stained with 0.5
mg/mL EtBr to visualize DNA and were photo-
Notes
graphed under UV illumination as described previously [74].
The authors declare no competing financial interest.
Declaration of competing interest
4.7. Formation of Topo IIa and Topo IIb cleavable complexes in vitro
The experimental conditions were the same as for the relaxation
assay except that approximately 50-fold more DNA Topo II and
Topo II was used. The reaction was initiated with the addition of
a
The authors declare that they have no known competing
financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
b
the enzymes. After 10 min at 30 ꢁC, the reactions were terminated
by adding 0.35% SDS and 0.3 mg/mL proteinase K (final concen-
trations) followed by incubation at 56 ꢁC for 1 h. The DNA cleavage
products were separated by electrophoresis in 1.2% agarose gels
Acknowledgments
containing EtBr (0.5
EDTA) buffer and were visualized and photographed under UV
illumination as described previously [74].
m
g/mL) at 1 V/cm for 18 h in TBE (Tris-borate-
This study was financially supported by grants “Combattere la
resistenza tumorale: piattaforma integrata multidisciplinare per un
approccio tecnologico innovativo alle oncoterapie “CAMPANIA
ONCOTERAPIE” (project no. B61G18000470007) of Regione Cam-
pania (Italy) (A. L.). The authors would like to thank the Centro
Interdipartimentale di Metodologie Chimico-Fisiche CIMCF and the
Centro Interdipartimentale di Analisi Strumentale dell’University of
Naples Federico II. We would like to thank Dr. Giorgia Oliviero for
synthesizing the oligonucleotide d[(GAAGCTTC)]2 used in this
work.
4.8. Comet assay
Comet assay was performed according to protocol by Tice et al.
[75]. Approximately 1 ꢂ 105 MCF-7 cells were seeded in 24-well
tissue-culture plates and after 24 h treated with increasing con-
centration of 5 and 10 mM VP-16 (positive control) for 24 h. Cells

S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
[23] S.B. Neidle, Ch, J.P.
19
of
Henichart,
Molecular
pharmacology
Appendix A. Supplementary data
IntercalatorꢀGroove binder hybrid molecules, in: Molecular Aspects of Anti-
cancer Drug DNA Interactions, Taylor & Francis Group, Oxon, U.K., 1994.
[24] M. Masquelier, R. Baurain, A. Trouet, Amino acid and dipeptide derivatives of
daunorubicin. 1. Synthesis, physicochemical properties, and lysosomal
digestion, J. Med. Chem. 23 (1980) 1166e1170.
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.ejmech.2019.111960.
[25] K. Breistol, H.R. Hendriks, D.P. Berger, S.P. Langdon, H.H. Fiebig, O. Fodstad,
The antitumour activity of the prodrug N-L-leucyl-doxorubicin and its parent
compound doxorubicin in human tumour xenografts, Eur. J. Cancer 34 (1998)
1602e1606.
References
ꢁ
[1] A. Rescifina, C. Zagni, M.G. Varrica, V. Pistara, A. Corsaro, Recent advances in
[26] Y. van Hensbergen, H.J. Broxterman, Y.W. Elderkamp, J. Lankelma, J.C. Beers,
M. Heijn, E. Boven, K. Hoekman, H.M. Pinedo, A doxorubicin-CNGRC-peptide
conjugate with prodrug properties, Biochem. Pharmacol. 63 (2002) 897e908.
[27] C. Ryppa, H. Mann-Steinberg, M.L. Biniossek, R. Satchi-Fainaro, F. Kratz,
In vitro and in vivo evaluation of a paclitaxel conjugate with the divalent
peptide E-[c(RGDfK)2] that targets integrin alpha v beta 3, Int. J. Pharm. 368
(2009) 89e97.
[28] C.J. Hsiao, T.K. Li, Y.L. Chan, L.W. Hsin, C.H. Liao, C.H. Lee, P.C. Lyu, J.H. Guh,
WRC-213, an l-methionine-conjugated mitoxantrone derivative, displays
anticancer activity with reduced cardiotoxicity and drug resistance: identifi-
cation of topoisomerase II inhibition and apoptotic machinery in prostate
cancers, Biochem. Pharmacol. 75 (2008) 847e856.
small organic molecules as DNA intercalating agents: synthesis, activity, and
modeling, Eur. J. Med. Chem. 74 (2014) 95e115.
[2] A. Bolognese, G. Correale, M. Manfra, A. Lavecchia, O. Mazzoni, E. Novellino,
V. Barone, A. Pani, E. Tramontano, P. La Colla, C. Murgioni, I. Serra, G. Setzu,
R. Loddo, Antitumor agents. 1. Synthesis, biological evaluation, and molecular
modeling of 5H-pyrido[3,2-a]phenoxazin-5-one, a compound with potent
antiproliferative activity, J. Med. Chem. 45 (2002) 5205e5216.
[3] A. Bolognese, G. Correale, M. Manfra, A. Lavecchia, O. Mazzoni, E. Novellino,
V. Barone, P. La Colla, R. Loddo, Antitumor agents. 2. Synthesis, structure-
activity relationships, and biological evaluation of substituted 5H-pyr-
idophenoxazin-5-ones with potent antiproliferative activity, J. Med. Chem. 45
(2002) 5217e5223.
[29] D. Ravel, V. Dubois, J. Quinonero, F. Meyer-Losic, J. Delord, P. Rochaix,
C. Nicolazzi, F. Ribes, C. Mazerolles, E. Assouly, K. Vialatte, I. Hor, J. Kearsey,
A. Trouet, Preclinical toxicity, toxicokinetics, and antitumoral efficacy studies
of DTS-201, a tumor-selective peptidic prodrug of doxorubicin, Clin. Cancer
Res. 14 (2008) 1258e1265.
[30] S. Zitzmann, V. Ehemann, M. Schwab, Arginine-glycine-aspartic acid (RGD)-
peptide binds to both tumor and tumor-endothelial cells in vivo, Cancer Res.
62 (2002) 5139e5143.
[31] W. Arap, R. Pasqualini, E. Ruoslahti, Cancer treatment by targeted drug de-
livery to tumor vasculature in a mouse model, Science 279 (1998) 377e380.
[32] C. Monneret, Recent developments in the field of antitumour anthracyclines,
Eur. J. Med. Chem. 36 (2001) 483e493.
[33] F.M. de Groot, H.J. Broxterman, H.P. Adams, A. van Vliet, G.I. Tesser,
Y.W. Elderkamp, A.J. Schraa, R.J. Kok, G. Molema, H.M. Pinedo, H.W. Scheeren,
Design, synthesis, and biological evaluation of a dual tumor-specific motive
containing integrin-targeted plasmin-cleavable doxorubicin prodrug, Mol.
Cancer Ther. 1 (2002) 901e911.
[34] X. Bu, L.W. Deady, G.J. Finlay, B.C. Baguley, W.A. Denny, Synthesis and cyto-
toxic activity of 7-oxo-7H-dibenz[f,ij]isoquinoline and 7-oxo-7H-benzo[e]
perimidine derivatives, J. Med. Chem. 44 (2001) 2004e2014.
[35] K. Tkaczyk-Gobis, J. Tarasiuk, O. Seksek, B. Stefanska, E. Borowski, A. Garnier-
Suillerot, Transport of new non-cross-resistant antitumor compounds of the
benzoperimidine family in multidrug resistant cells, Eur. J. Pharmacol. 413
(2001) 131e141.
[36] P. De Isabella, F. Zunino, G. Capranico, Base sequence determinants of amo-
nafide stimulation of topoisomerase II DNA cleavage, Nucleic Acids Res. 23
(1995) 223e229.
[37] C. Bailly, Topoisomerase I poisons and suppressors as anticancer drugs, Curr.
Med. Chem. 7 (2000) 39e58.
[38] S. Fujimoto, Promising antitumor activity of a novel quinoline derivative, TAS-
103, against fresh clinical specimens of eight types of tumors measured by
flow cytometric DNA analysis, Biol. Pharm. Bull. 30 (2007) 1923e1929.
[39] L.P. Wakelin, A. Adams, W.A. Denny, Kinetic studies of the binding of acridi-
necarboxamide topoisomerase poisons to DNA: implications for mode of
binding of ligands with uncharged chromophores, J. Med. Chem. 45 (2002)
894e901.
[40] H. Malonne, S. Farinelle, C. Decaestecker, L. Gordower, J. Fontaine,
F. Chaminade, J.M. Saucier, G. Atassi, R. Kiss, In vitro and in vivo pharmaco-
logical characterizations of the antitumor properties of two new olivacine
derivatives, S16020-2 and S30972-1, Clin. Cancer Res. 6 (2000) 3774e3782.
[41] H. Ihmels, D. Otto, Intercalation of organic dye molecules into double-
stranded DNA–general principles and recent developments, in: Super-
molecular Dye Chemistry, Springer, 2005, pp. 161e204.
[4] A. Bolognese, G. Correale, M. Manfra, A. Lavecchia, E. Novellino, S. Pepe,
Antitumor agents. 5. Synthesis, structure-activity relationships, and biological
evaluation of dimethyl-5H-pyridophenoxazin-5-ones, tetrahydro-5h-benzo-
pyridophenoxazin-5-ones, and 5h-benzopyridophenoxazin-5-ones with
potent antiproliferative activity, J. Med. Chem. 49 (2006) 5110e5118.
[5] A. Alberti, A. Bolognese, M. Guerra, A. Lavecchia, D. Macciantelli, M. Marcaccio,
E. Novellino, F. Paolucci, Antitumor agents 4. Characterization of free radicals
produced during reduction of the antitumor drug 5H-pyridophenoxazin-5-
one: an EPR study, Biochemistry 42 (2003) 11924e11931.
[6] C. Bailly, Contemporary challenges in the design of topoisomerase II inhibitors
for cancer chemotherapy, Chem. Rev. 112 (2012) 3611e3640.
[7] J.L. Nitiss, Targeting DNA topoisomerase II in cancer chemotherapy, Nat. Rev.
Cancer 9 (2009) 338e350.
[8] S.M. Vos, E.M. Tretter, B.H. Schmidt, J.M. Berger, All tangled up: how cells
direct, manage and exploit topoisomerase function, Nat. Rev. Mol. Cell Biol. 12
(2011) 827e841.
[9] S.M. Cuya, M.A. Bjornsti, R. van Waardenburg, DNA topoisomerase-targeting
chemotherapeutics: what’s new? Cancer Chemother. Pharmacol. 80 (2017)
1e14.
[10] Y. Pommier, E. Leo, H. Zhang, C. Marchand, DNA topoisomerases and their
poisoning by anticancer and antibacterial drugs, Chem. Biol. 17 (2010)
421e433.
[11] Y. Pommier, P. Pourquier, Y. Fan, D. Strumberg, Mechanism of action of
eukaryotic DNA topoisomerase I and drugs targeted to the enzyme, Biochim.
Biophys. Acta 1400 (1998) 83e105.
[12] D.A. Burden, N. Osheroff, Mechanism of action of eukaryotic topoisomerase II
and drugs targeted to the enzyme, Biochim. Biophys. Acta 1400 (1998)
139e154.
[13] A.M. Azarova, Y.L. Lyu, C.-P. Lin, Y.-C. Tsai, J.Y.-N. Lau, J.C. Wang, L.F. Liu, Roles
of DNA topoisomerase II isozymes in chemotherapy and secondary malig-
nancies, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 11014e11019.
[14] E. Toyoda, S. Kagaya, I.G. Cowell, A. Kurosawa, K. Kamoshita, K. Nishikawa,
S. Iiizumi, H. Koyama, C.A. Austin, N. Adachi, NK314, a topoisomerase II in-
hibitor that specifically targets the
23711e23720.
a isoform, J. Biol. Chem. 283 (2008)
[15] A.K. Larsen, A. Skladanowski, Cellular resistance to topoisomerase-targeted
drugs: from drug uptake to cell death, Biochim. Biophys. Acta 1400 (1998)
257e274.
[16] R. Ishida, T. Miki, T. Narita, R. Yui, M. Sato, K.R. Utsumi, K. Tanabe, T. Andoh,
Inhibition of intracellular topoisomerase II by antitumor bis(2,6-
dioxopiperazine) derivatives: mode of cell growth inhibition distinct from
that of cleavable complex-forming type inhibitors, Cancer Res. 51 (1991)
4909e4916.
[42] Y. Zhu, H. Zeng, J. Xie, L. Ba, X. Gao, Z. Lu, Atomic force microscopy studies on
DNA structural changes induced by vincristine sulfate and aspirin, Microsc.
Microanal. 10 (2004) 286e290.
[17] Y. Onishi, Y. Azuma, H. Kizaki, bis(2,6-dioxopiperaxine) derivatives, topo-
isomerase II inhibitors which do not form a DNA cleavable complex, induce
thymocyte apoptosis, Biochem. Mol. Biol. Int. 32 (1994) 115e122.
[43] G. Tyagi, D.K. Jangir, P. Singh, R. Mehrotra, DNA interaction studies of an
anticancer plant alkaloid, vincristine, using Fourier transform infrared spec-
troscopy, DNA Cell Biol. 29 (2010) 693e699.
[44] E. Froehlich, A. Gupta, J. Provencher-Mandeville, E. Asselin, J. Bariyanga,
G. Berube, H.A. Tajmir-Riahi, Study of DNA interactions with steroidal and
nonsteroidal estrogen-platinum (II)-based anticancer drugs, DNA Cell Biol. 28
(2009) 31e39.
[45] R. Marty, A.A. Ouameur, J.F. Neault, S. Nafisi, H.A. Tajmir-Riahi, AZT-DNA
interaction, DNA Cell Biol. 23 (2004) 135e140.
[46] D.K. Jangir, S. Charak, R. Mehrotra, S. Kundu, FTIR and circular dichroism
spectroscopic study of interaction of 5-fluorouracil with DNA, J. Photochem.
Photobiol., B 105 (2011) 143e148.
[18] B.B. Hasinoff, T.I. Kuschak, J.C. Yalowich, A.M. Creighton,
A QSAR study
comparing the cytotoxicity and DNA topoisomerase II inhibitory effects of
bisdioxopiperazine analogs of ICRF-187 (dexrazoxane), Biochem. Pharmacol.
50 (1995) 953e958.
[19] J.L. Nitiss, DNA topoisomerase II and its growing repertoire of biological
functions, Nat. Rev. Cancer 9 (2009) 327e337.
[20] P. Furet, J. Schoepfer, T. Radimerski, P. Chene, Discovery of a new class of
catalytic topoisomerase II inhibitors targeting the ATP-binding site by struc-
ture based design. Part I, Bioorg. Med. Chem. Lett 19 (2009) 4014e4017.
[21] K.Y. Jun, E.Y. Lee, M.J. Jung, O.H. Lee, E.S. Lee, H.Y. Park Choo, Y. Na, Y. Kwon,
Synthesis, biological evaluation, and molecular docking study of 3-(3’-het-
eroatom substituted-2’-hydroxy-1’-propyloxy) xanthone analogues as novel
topoisomerase IIalpha catalytic inhibitor, Eur. J. Med. Chem. 46 (2011)
1964e1971.
[47] J.H. Choi, A.S. Banks, J.L. Estall, S. Kajimura, P. Bostrom, D. Laznik, J.L. Ruas,
M.J. Chalmers, T.M. Kamenecka, M. Bluher, P.R. Griffin, B.M. Spiegelman, Anti-
diabetic drugs inhibit obesity-linked phosphorylation of PPARgamma by
Cdk5, Nature 466 (2010) 451e456.
[22] W. Hu, X.-S. Huang, J.-F. Wu, L. Yang, Y.-T. Zheng, Y.-M. Shen, Z.-Y. Li, X. Li,
Discovery of novel topoisomerase II inhibitors by medicinal chemistry ap-
proaches, J. Med. Chem. 61 (2018) 8947e8980.
[48] D.K. Jangir, S.K. Dey, S. Kundu, R. Mehrotra, Assessment of amsacrine binding

20
S. Pedatella et al. / European Journal of Medicinal Chemistry 187 (2020) 111960
with DNA using UV-visible, circular dichroism and Raman spectroscopic
techniques, J. Photochem. Photobiol., B 114 (2012) 38e43.
[62] P. Vejpongsa, E.T. Yeh, Topoisomerase 2beta: a promising molecular target for
primary prevention of anthracycline-induced cardiotoxicity, Clin. Pharmacol.
Ther. 95 (2014) 45e52.
[63] Y.-R. Wang, S.-F. Chen, C.-C. Wu, Y.-W. Liao, T.-S. Lin, K.-T. Liu, Y.-S. Chen, T.-
K. Li, T.-C. Chien, N.-L. Chan, Producing irreversible topoisomerase II-mediated
DNA breaks by site-specific Pt (II)-methionine coordination chemistry, Nucleic
Acids Res. 45 (2017) 10861e10871.
[64] J. Kluza, A. Lansiaux, N. Wattez, C. Mahieu, N. Osheroff, C. Bailly, Apoptotic
response of HL-60 human leukemia cells to the antitumor drug TAS-103,
Cancer Res. 60 (2000) 4077e4084.
[65] L.J. Lewis, P. Mistry, P.A. Charlton, H. Thomas, H.M. Coley, Mode of action of
the novel phenazine anticancer agents XR11576 and XR5944, Anti Canccer
Drugs 18 (2007) 139e148.
[66] W.M. Bonner, C.E. Redon, J.S. Dickey, A.J. Nakamura, O.A. Sedelnikova, S. Solier,
Y. Pommier, GammaH2AX and cancer, Nat. Rev. Cancer 8 (2008) 957e967.
[67] A. Ivashkevich, C.E. Redon, A.J. Nakamura, R.F. Martin, O.A. Martin, Use of the
gamma-H2AX assay to monitor DNA damage and repair in translational
cancer research, Cancer Lett. 327 (2012) 123e133.
[68] E.P. Rogakou, W. Nieves-Neira, C. Boon, Y. Pommier, W.M. Bonner, Initiation of
DNA fragmentation during apoptosis induces phosphorylation of H2AX his-
tone at serine 139, J. Biol. Chem. 275 (2000) 9390e9395.
[69] C.D. Kanakis, P.A. Tarantilis, C. Pappas, J. Bariyanga, H.A. Tajmir-Riahi,
M.G. Polissiou, An overview of structural features of DNA and RNA complexes
with saffron compounds: models and antioxidant activity, J. Photochem.
Photobiol., B 95 (2009) 204e212.
[70] H.M. Berman, J. Westbrook, Z. Feng, G. Gilliland, T.N. Bhat, H. Weissig,
I.N. Shindyalov, P.E. Bourne, The protein Data Bank, Nucleic Acids Res. 28
(2000) 235e242.
[49] P. Kolodziejczyk, A. Garnier-Suillerot, Circular dichroism study of the inter-
action of mitoxantrone, ametantrone and their Pd(II) complexes with deox-
yribonucleic acid, Biochim. Biophys. Acta 926 (1987) 249e257.
[50] B. Rafique, A.M. Khalid, K. Akhtar, A. Jabbar, Interaction of anticancer drug
methotrexate with DNA analyzed by electrochemical and spectroscopic
methods, Biosens. Bioelectron. 44 (2013) 21e26.
[51] C. Wan, X. Guo, F. Song, Z. Liu, S. Liu, Interactions of mitoxantrone with duplex
and triplex DNA studied by electrospray ionization mass spectrometry, Rapid
Commun. Mass Spectrom. 22 (2008) 4043e4048.
[52] J.M. Fortune, N. Osheroff, Merbarone inhibits the catalytic activity of human
topoisomerase IIalpha by blocking DNA cleavage, J. Biol. Chem. 273 (1998)
17643e17650.
[53] G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell,
A.J. Olson, AutoDock4 and AutoDockTools4: automated docking with selective
receptor flexibility, J. Comput. Chem. 30 (2009) 2785e2791.
[54] P.A. Holt, J.B. Chaires, J.O. Trent, Molecular docking of intercalators and
groove-binders to nucleic acids using Autodock and Surflex, J. Chem. Inf.
Model. 48 (2008) 1602e1615.
[55] C.G. Ricci, P.A. Netz, Docking studies on DNA-ligand interactions: building and
application of a protocol to identify the binding mode, J. Chem. Inf. Model. 49
(2009) 1925e1935.
[56] M. Shinomiya, W. Chu, R.G. Carlson, R.F. Weaver, F. Takusagawa, Structural,
physical, and biological characteristics of RNA.DNA binding agent N8-
actinomycin D, Biochemistry 34 (1995) 8481e8491.
[57] A.D. Bochevarov, E. Harder, T.F. Hughes, J.R. Greenwood, D.A. Braden,
D.M. Philipp, D. Rinaldo, M.D. Halls, J. Zhang, R.A. Friesner, Jaguar: a high-
performance quantum chemistry software program with strengths in life
and materials sciences, Int. J. Quantum Chem. 113 (2013) 2110e2142.
[58] S. Saebo, W. Tong, P. Pulay, Efficient elimination of basis set superposition
errors by the local correlation method: accurate ab initio studies of the water
dimer, J. Chem. Phys. 98 (1993) 2170e2175.
[71] 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)
1145e1152.
[72] J. Foster, F. Weinhold, Natural hybrid orbitals, J. Am. Chem. Soc. 102 (1980)
7211e7218.
ꢂ
ꢂ
ꢀ
[59] S.F. Boys, F.d. Bernardi, The calculation of small molecular interactions by the
differences of separate total energies. Some procedures with reduced errors,
Mol. Phys. 19 (1970) 553e566.
[60] L.A. Zwelling, J. Mayes, M. Hinds, D. Chan, E. Altschuler, B. Carroll, E. Parker,
K. Deisseroth, A. Radcliffe, M. Seligman, et al., Cross-resistance of an
amsacrine-resistant human leukemia line to topoisomerase II reactive DNA
intercalating agents. Evidence for two topoisomerase II directed drug actions,
Biochemistry 30 (1991) 4048e4055.
[73] J. Chocholousova, V. Spirko, P. Hobza, First local minimum of the formic acid
dimer exhibits simultaneously red-shifted OeHO and improper blue-shifted
CeHO hydrogen bonds, Phys. Chem. Chem. Phys. 6 (2004) 37e41.
[74] K. Bojanowski, S. Lelievre, J. Markovits, J. Couprie, A. Jacquemin-Sablon,
A.K. Larsen, Suramin is an inhibitor of DNA topoisomerase II in vitro and in
Chinese hamster fibrosarcoma cells, Proc. Natl. Acad. Sci. U. S. A. 89 (1992)
3025e3029.
[75] R.R. Tice, E. Agurell, D. Anderson, B. Burlinson, A. Hartmann, H. Kobayashi,
Y. Miyamae, E. Rojas, J.C. Ryu, Y.F. Sasaki, Single cell gel/comet assay: guide-
lines for in vitro and in vivo genetic toxicology testing, Environ. Mol. Mutagen.
35 (2000) 206e221.
[61] A. Skladanowski, S.Y. Plisov, J. Konopa, A.K. Larsen, Inhibition of DNA topo-
isomerase II by imidazoacridinones, new antineoplastic agents with strong
activity against solid tumors, Mol. Pharmacol. 49 (1996) 772e780.