Synthesis, anticancer activity, and molecular modeling of 1,4-naphthoquinones that inhibit MKK7 and Cdc25

Article abstract of DOI:10.1016/j.ejmech.2019.111719

Cell division cycle 25 (Cdc25) and mitogen-activated protein kinase kinase 7 (MKK7) are enzymes involved in intracellular signaling but can also contribute to tumorigenesis. We synthesized and characterized the biological activity of 1,4-naphthoquinones structurally similar to reported Cdc25 and(or) MKK7 inhibitors with anticancer activity. Compound 7 (3-[(1,4-dioxonaphthalen-2-yl)sulfanyl]propanoic acid) exhibited high binding affinity for MKK7 (K_d = 230 nM), which was greater than the affinity of NSC 95397 (K_d = 1.1 μM). Although plumbagin had a lower binding affinity for MKK7, this compound and sulfur-containing derivatives 4 and 6–8 were potent inhibitors of Cdc25A and Cdc25B. Derivative 22e containing a phenylamino side chain was selective for MKK7 versus MKK4 and Cdc25 A/B, and its isomer 22f was a selective inhibitor of Cdc25 A/B. Docking studies performed on several naphthoquinones highlighted interesting aspects concerning the molecule orientation and hydrogen bonding interactions, which could help to explain the activity of the compounds toward MKK7 and Cdc25B. The most potent naphthoquinone-based inhibitors of MKK7 and/or Cdc25 A/B were also screened for their cytotoxicity against nine cancer cell lines and primary human mononuclear cells, and a correlation was found between Cdc25 A/B inhibitory activity and cytotoxicity of the compounds. Quantum chemical calculations using BP86 and ωB97X-D3 functionals were performed on 20 naphthoquinone derivatives to obtain a set of molecular electronic properties and to correlate these properties with cytotoxic activities. Systematic theoretical DFT calculations with subsequent correlation analysis indicated that energy of the lowest unoccupied molecular orbital E(LUMO), vertical electron affinity (VEA), and reactivity index ω of these molecules were important characteristics related to their cytotoxicity. The reactivity index ω was also a key characteristic related to Cdc25 A/B phosphatase inhibitory activity. Thus, 1,4-naphthoquinones displaying sulfur-containing and phenylamino side chains with additional polar groups could be successfully utilized for further development of efficacious Cdc25 A/B and MKK7 inhibitors with anticancer activity.

Full text of DOI:10.1016/j.ejmech.2019.111719

Journal Pre-proof
Synthesis, anticancer activity, and molecular modeling of 1,4-naphthoquinones that
inhibit MKK7 and Cdc25
Igor A. Schepetkin, Alexander S. Karpenko, Andrei I. Khlebnikov, Marina O.
Shibinskaya, Igor A. Levandovskiy, Liliya N. Kirpotina, Nadezhda V. Danilenko, Mark
T. Quinn
PII:
S0223-5234(19)30871-2
DOI:
https://doi.org/10.1016/j.ejmech.2019.111719
EJMECH 111719
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 18 June 2019
Revised Date: 9 July 2019
Accepted Date: 17 September 2019
Please cite this article as: I.A. Schepetkin, A.S. Karpenko, A.I. Khlebnikov, M.O. Shibinskaya, I.A.
Levandovskiy, L.N. Kirpotina, N.V. Danilenko, M.T. Quinn, Synthesis, anticancer activity, and molecular
modeling of 1,4-naphthoquinones that inhibit MKK7 and Cdc25, European Journal of Medicinal
Chemistry (2019), doi: https://doi.org/10.1016/j.ejmech.2019.111719.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
©
2019 Published by Elsevier Masson SAS.

O
O
R1
R2
MKK7
Anticancer Activity
DFT study
Cdc25A/B
1
00
0.4
0.3
0.2
0.1
0.0
8
6
4
2
0
0
0
0
Naphthoquinones
Shikonin
Plumbagin
0
0
-0.1
-
.01
0.1
1
10
100
2.5
-2.0
-1.5
-1.0
Cdc25B
Concentration (µM)
LUMO, eV

Synthesis, Anticancer Activity, and Molecular Modeling of 1,4-Naphthoquinones that
Inhibit MKK7 and Cdc25
1
2
3,4
2
Igor A. Schepetkin , Alexander S. Karpenko , Andrei I. Khlebnikov , Marina O. Shibinskaya ,
5
1
3
Igor A. Levandovskiy , Liliya N. Kirpotina , Nadezhda V. Danilenko , and
1
,*
Mark T. Quinn
1
Department of Microbiology and Immunology, Montana State University, Bozeman, MT
59717, USA;
2
A.V. Bogatsky Physico-Chemical Institute, National Academy of Sciences of Ukraine, Odessa
65080, Ukraine;
3
4
5
Kizhner Research Center, Tomsk Polytechnic University, Tomsk 634050, Russia;
Faculty of Chemistry, National Research Tomsk State University, Tomsk 634050, Russia;
Department of Organic Chemistry, Kiev Polytechnic Institute, Kiev 03056, Ukraine
Running title: Naphthoquinones as MKK7 and Cdc25 Inhibitors
*
Address for Correspondence:
Mark T Quinn, Ph.D.
Department of Microbiology and Immunology
Montana State University, Bozeman, MT 59717
Phone: 406-994-4707; Fax: 406-994-4303
Email: mquinn@montana.edu
1

ABSTRACT
Cell division cycle 25 (Cdc25) and mitogen-activated protein kinase kinase 7 (MKK7)
are enzymes involved in intracellular signaling but can also contribute to tumorigenesis. We
synthesized and characterized the biological activity of 1,4-naphthoquinones structurally similar
to reported Cdc25 and(or) MKK7 inhibitors with anticancer activity. Compound 7 (3-[(1,4-
dioxonaphthalen-2-yl)sulfanyl]propanoic acid) exhibited high binding affinity for MKK7 (K =
d
2
30 nM), with greater affinity than NSC 95397 (K = 1.1 µM). Although plumbagin had a lower
d
binding affinity for MKK7, this compound and sulfur-containing derivatives 4 and 6-8 were
potent inhibitors of Cdc25A and Cdc25B. Derivative 22e with a phenylamino side chain was
selective for MKK7 versus MKK4 and Cdc25A/B, and its isomer 22f was a selective inhibitor
of Cdc25A/B. Docking studies performed on several naphthoquinones highlighted interesting
aspects concerning the molecule orientation and hydrogen bonding interactions, which could
help to explain the activity of the compounds toward MKK7 and Cdc25B. The most potent
naphthoquinone-based inhibitors of MKK7 and/or Cdc25A/B were also screened for their
cytotoxicity against nine cancer cell lines and primary human mononuclear cells, and a
correlation was found between Cdc25A/B inhibitory activity and cytotoxicity of the
compounds. Quantum chemical calculations using BP86 and ωB97X-D3 functionals were
performed on 20 naphthoquinone derivatives to obtain a set of molecular electronic properties
and to correlate these properties with cytotoxic activities. Systematic theoretical DFT
calculations with subsequent correlation analysis indicated that energy of the lowest unoccupied
molecular orbital E(LUMO), vertical electron affinity (VEA), and reactivity index ω of these
molecules are important characteristics related to their cytotoxicity. The reactivity index ω was
also a key characteristic related to Cdc25A/B phosphatase inhibitory activity. Thus, 1,4-
naphthoquinones displaying sulfur-containing and phenylamino side chains with additional
polar groups could be successfully utilized for further development of efficacious Cdc25A/B
and MKK7 inhibitors with anticancer activity.
2

Keywords: naphthoquinone, kinase inhibitor, mitogen-activated kinase kinase 7, Cdc25
phosphatase, cytotoxicity, molecular docking, DFT analysis.
3

1. Introduction
Naphthoquinone derivatives, including vitamin K3 analogs and plant-derived
naphthoquinones, have been reported to exhibit a variety of pharmacological properties
involving antimicrobial, antiviral, anti-inflammatory, and antitumor effects [1-4]. Cancer is the
second leading cause of death in the world, and recent advances in the field of novel anticancer
agents has been promising. Naphthoquinone derivatives may be attractive for the development
of novel anticancer agents because they have a broad spectrum of biological activities and could
target multiple signaling pathways in tumor cells. Studies of the antitumor properties and
mechanisms of action of various naphthoquinone derivatives have shown that they can act as
inhibitors of cell division cycle 25 (Cdc25) phosphatases and DNA topoisomerases, disrupt
kinase signaling important for cancer cell proliferation, and inhibit Snail-p53 binding [5-9].
Their cytotoxicity can also be explained by oxidative stress induced via reactive oxygen species
(ROS) generation and glutathione depletion [10-12]. Indeed, the 1,4-naphthoquinone moiety
can form redox isomers, such as napthosemiquinone and catechol, by accepting one or two
electrons, respectively, during metabolism by quinone oxidoreductase, which is overexpressed
in solid tumors from a variety of cancer types [13].
One naphthoquinone with multiple biological mechanisms is NSC 95397 (2,3-bis-[(2-
hydroxyethyl)thio]-1,4-naphthoquinone), an inhibitor of protein kinase B (Akt), IκB kinases α/β
(
(
IKKα/β), TANK-binding kinase (TBK1), and mitogen-activated protein kinase kinase 7
MKK7) [14]. NSC 95397 suppresses proliferation and induces apoptosis in colon cancer cells
through mitogen-activated protein kinase phosphatase-1 (MKP-1) and the extracellular signal-
regulated kinase 1/2 (ERK1/2) pathway [15]. Moreover, NSC 95397 inhibits the C-terminal
binding protein (CtBP1)-protein partner interaction and CtBP1-mediated transcriptional
repression, which are important pathways in cancer cell survival [16]. This compound was first
identified in a high-throughput screening program as a potent irreversible inhibitor of cell
division Cdc25 [17]. Cdc25 dual specificity phosphatases, which include A, B, and C isoforms,
4

serve as important regulators of cell cycle progression by activating the cyclin-dependent
kinases (CDKs) through removal of inhibitory phosphates from tyrosine/threonine residues
[
18]. Cdc25B and Cdc25C play a major role in G /M progression of cell cycle, whereas
2
Cdc25A assists in G /S phase transition [19]. Cdc25 phosphatases seem to have a role in the
1
development of several human malignancies, including acute myeloid leukemia, and Cdc25
inhibition is therefore considered as a possible anticancer strategy [20]. Over half of the 15
different cancers studied had overexpression of both Cdc25A and Cdc25B isoforms; the
remainder had overexpression of either Cdc25A or Cdc25B, whereas there was no significant
overexpression of Cdc25C [21]. Thus, small-molecule inhibitors of these phosphatases may
represent a promising therapeutic anticancer agents [22].
c-Jun N-terminal kinase (JNK) signaling also plays a vital role in metastasis and
malignant transformation in various cancers and is involved in cancer cell apoptosis [23-25].
For example, JNK is highly activated in basal-like 'triple-negative' breast cancer [26]. Reduced
JNK activation is associated with better overall survival of patients with infiltrating ductal
breast carcinoma [27]. Several other publications indicate that targeting JNK signaling has
potential for cancer treatment (e.g., [28]). There are only two direct upstream activators of JNK:
mitogen-activated protein kinase kinase 4 and 7 (MKK4 and MKK7) [29]. MKK4 activates
JNK primarily in response to stress stimuli, whereas MKK7 signaling is triggered by release of
inflammatory cytokines [30]. Thus, by targeting MKK7, the physiological role of JNK
regulated by MKK4 would be preserved. Development of MKK7 inhibitors could represent a
new therapeutic strategy for the treatment of ischemia and other pathologies involving MKK7
and(or) JNK activation [31, 32]. Recently Shraga et al. [33] reported irreversible MKK7
inhibitors, based on indazole acrylamide scaffold. The observation that a naphthoquinone
derivative, NSC 95397, is an inhibitor of MKK7 [14] makes the 1,4-naphthoquinone scaffold
also attractive for development of MKK7 inhibitors.
5

Although several studies have attempted to improve the naphthoquinone scaffold
regarding Cdc25 inhibitory activity [34-38], there are no comprehensive reports on structure-
activity relationship (SAR) analysis and molecular modeling of naphthoquinone derivatives
targeting MKK7 and Cdc25. In the present study, 33 naphthoquinones, including synthetic and
natural occurring compounds were evaluated for their inhibition/binding activity for Cdc25A/B
and MKK7. To provide insights into their possible binding interaction, we also performed
molecular docking studies into MKK7 and Cdc25B proteins. Twenty selected compounds were
evaluated for their in vitro antiproliferative activities against nine human tumor cell lines and
primary human mononuclear cells using sunitinib as a positive control. We report for the first
time that plumbagin, a natural naphthoquinone, has a high inhibitory activity for Cdc25A and B
isoforms and that compounds 7 and 22e have relatively high binding affinities for MKK7 but
not for MKK4.
6

2. Results and discussion
2.1. Screening compounds
Two main classes of Cdc25 inhibitors with naphthoquinone and quinolinedione scaffolds
have been reported, including sulfur-containing analogs and amino derivatives (Figure 1).
Among these compounds, NSC 95397 was previously reported as a potent Cdc25B inhibitor
and a weak MKK7 inhibitor [14, 17]. Based on structures of these compounds, natural and
synthetic naphthoquinones with different substituents mainly at positions 2 and 3 of the 1,4-
naphthoquinone scaffold were selected. These analogs included known naphthoquinones, such
as shikonin (1), plumbagin (2), lapachol (3), Cpd C (5; shown in Scheme 1), vitamin ks-II (9),
GN25 (14), buparvaquone (21), and menadione (18). Twelve compounds with nitrogen in the
R position (22a-g, 23a,b, 24, and 25a,b) were designed as analogs of NSC 663284 (Scheme 1)
2
and synthesized, as described below. All compounds were diluted in dimethyl sulfoxide
(DMSO) at a concentration of 10 mM and stored at -20 °C.
O
Cpd A: R = CH , R = CH COOH,
1
3
2
2
S
S
Cpd B: R = CH COOH, R = CH COOH
R1
R2
1
2
2
2
Cpd C: R = CH , R = (CH ) OH
1
3
2
2 2
NSC 95397: R = (CH ) OH, R = (CH ) OH
1
2 2
2
2 2
O
O
Cl
O
N
X
NH
Cpd D: X = CH
NSC 663284: X = N
O
Scheme 1. Reported naphthoquinone-based sulfur-containing analogs and amino derivatives
with Cdc25 inhibitory activity [14, 17, 38].
2.2. Chemistry
New compounds (22a-g, 23a,b, 24, and 25a,b) were synthesized with high yields via
condensation of 2,3-dichloro-1,4-naphthoquinone with amino-compounds in boiling ethanol
7

(
aqueous) or methanol in presence of base (CH COONa, CaCO or amine excess) [38-41].
3 3
Further transformations of 22g were achieved as described [40] or, in an analogous way, via
condensation with nucleophilic components under basic conditions (Scheme 2). The
compounds were characterized by their physical, analytical, and spectral data [MP, mass-
spectroscopy and NMR.
R
O
O
2
2
2a
2b
22c
2d
2e
22f
2g
a
a
a
a
a
a
b
b
b
c
CH COOH
2
Cl
Cl
NH
Cl
(CH ) COOH
2
2
2
a or b or c
R
+
H2N
R
(CH
)
COOH
3
2
2
(CH ) COOH
2 5
m-C H COOH
6 4
O
2
2a-g, 23a,b, 24
p-C
p-C H COOCH
3
H COOH
O
6
4
2
6
4
23a
(CH ) OH
2 2
(CH ) OH
2 3
morpholin-4-
ylpropyl
23b
2
4
O
O
NH
R
d or e
22g
O
R
2
25b
5a
d
e
piperidin-1-yl
OCH3
O
2
5a,b
Scheme 2. Reagents and conditions for compound synthesis. (a) 22a-f: CaCO , 50% EtOH,
3
boiling, 10 h, 80% yield; (b) 22g, 23a, b: CH COONa, 50% EtOH, boiling, 12 h, 80 % yield;
3
(c) 24: 2 mol. 3-morpholin-4-yl-propylamine, EtOH, boiling, 16 h, 95% yield; (d) 25a: 12 mol.
piperidine, CH OH, boiling, 20 h, 51% yield; (e) 25b: 2 mol. CH ONa, CH OH, boiling, 4 h,
3
3
3
64% yield.
The mass-spectra of most compounds contained molecular ions that had a chlorine
profile (with exception of des-chlorinated 25a,b and 24 with a weak molecular ion). The main
degradation paths under MS/EI conditions were: decarboxylation for carboxylic acids (highest
for aromatic acids) and losing aliphatic sidechains with m/z = 220 splinter ion formation. All of
1
the expected signals according to molecular structure and symmetry were found in the H NMR
spectra. Aromatic signals were at 7-8 ppm, and their quantity, intensity, and multiplicity were
in accordance with calculated results. Common N-H group signals were at 7.2-7.5 ppm, with
the exception of 22b (6.6 ppm) and -NH-Ar compounds (8.4-9.5 ppm). COOH group signals
1
3
were at 12-13 ppm. The most expressed and common feature of C NMR spectra was the
8

presence of three signals at 170-180 ppm, which are from inequivalent C=O groups, with the
exception of 23a,b, 24, which have only two such groups.
2.3. Activity of the naphthoquinones for MKK7
All compounds were evaluated for their ability to bind to MKK7 and compared with
NSC 95397 (compound 4), and the K values are presented in Table 1. The results demonstrated
d
that the naphthoquinone nucleus with a sulfur atom in its side chain is an appropriate scaffold
for MKK7 inhibitor development. Compound 7 was the most potent MKK7 inhibitor, with
binding affinity K ~230 nM. Reference naphthoquinone derivative NSC 95397 has a K in the
d
d
micromolar range, which is in agreement with Yang et al. [14], who reported that the compound
is a weak MKK7 inhibitor. Likewise, natural compound plumbagin (2) and compounds 9 and 11
exhibited binding affinity for MKK7, although in the micromolar range (K ~14-15 µM). In
d
contrast, natural compounds shikonin (1) and lapachol (3), as well as menadione (18) and
buparvaquone (21) did not bind to MKK7. Although no clear SAR emerged from modifications
in this series, several important observations could be useful. As shown in Table 1, elimination
of the methyl group in 9 led to a potent MKK7 inhibitor (compound 7). The same result was
obtained by elongation of the spacer between the sulfur and carboxyl groups by one methylene
group (compound 6). Substitution of both SCH groups in compound 13 (K = 4.6 µM) by
3
d
OCH groups (compound 15) was associated with almost complete loss of binding affinity.
3
Introduction of two OCH groups at positions 5 and 8 of the 1,4-naphthoquinone scaffold in the
3
most potent compound 7 also led to inactive compound 14. Among amino-derivatives at the R2
position, only 22e exhibited good binding affinity toward MKK7 (K = 1.9 µM). Thus, MKK7
d
binding activity may be very sensitive to small modifications of substituents at positions 2 and 3
of the 1,4-naphthoquinone scaffold.
The most potent naphthoquinone derivatives that targeted MKK7 (4, 6-8, and 22e) were
also evaluated for their ability to bind to MKK4, another kinase that can phosphorylate JNK
9

[
42]. No MKK4 binding affinity was found for 4, 7, 8, and 22e, whereas low binding affinity
was found for compound 6 (K = 19.5 ± 0.7 µM) (data not shown).
d
We also tested six 4-oxo-1,4-dihydroquinazoline and quinazolin-4(3H)-one analogs of
compound 7 with similar or related R tail substituents but did not find any binding affinity for
1
MKK7 (Supplementary Table 1S). These results indicate that the presence of the
naphthoquinone moiety is critical for biological activity, possibly due to the hydrophobic
properties of this molecular fragment and/or specific location of the two H-bond acceptors
(carbonyl groups). It should be noted that p-naphthoquinones are prone to reduction, with the
formation of semiquinone anion radicals. In contrast, 4-oxo-1,4-dihydroquinazoline derivatives
have a structure that does not lead to effective resonance stabilization of a radical [43].
2.4. Molecular docking of selected naphthoquinones with MKK7
To provide insights into possible binding interaction of MKK7 inhibitors, we performed
molecular docking studies of selected naphthoquinone derivatives 4 (reference compound NSC
5397), 6, 7, 22e, and 22f into the binding site of human MKK7 (PDB code 2DYL) (see
9
Supplementary Figure S1). Docking calculations gave the best poses of compounds 6 and 7,
where they form H-bonds by their quinone oxygen atoms with the NH group of Met215 and the
SH group of Cys276. The carboxyl group of 6 and 7 formed a H-bond with Lys165 and
Asp277, respectively (Supplementary Figure S2). The presence of an additional methylene
group in 7 compared to 6 leads to some differences in the binding modes for these two
molecules. Indeed, compound 7 is 6-fold more active than 6. Conversely, elongation of the
spacer between the sulfur and carboxyl group by one more methylene group compared to 7 led
to breaking of the H-bond between the naphthoquinone carbonyl and Cys276, while H-bonding
of another naphthoquinone carbonyl to Met215 becomes much weaker. Thus, our modeling
experiments suggest that the length of side chain in 7 could be optimal for supporting MKK7
inhibitory activity. Regarding the compound 4, both oxygen atoms of the quinone moiety in
1
0

molecule are also H-bonded to Cys276 and Met215. In addition, the terminal OH groups in 4
form H-bonds with Glu213 and Asp277 (Supplementary Figure S3).
In the pair of novel molecules 22e and 22f, which contain a carboxyl group in the
benzene ring, the naphthoquinone moieties are displaced in their docking poses relative to
molecules 4, 6, and 7, especially for inactive compound 22f (Figure 1A). Despite the
displacement of 22e relative to 7, the positions of their carboxyl groups almost coincide. The
active compound 22e forms rather strong H-bonds with Asp277 and Met215, as well as
somewhat weaker H-bonds with Cys276 and Lys221 (Figure 1B). Thus, molecule 22e upon
binding interacts with Met215, Cys276, and Asp277, which presumably play a key role in the
binding activity. Unlike 22e, the inactive 22f is not H-bonded to any of these residues (Figure
1
C) but forms only a weak H-bond with Lys221 and a stronger bond with Cys218. Obviously,
the carboxyl group in para position of the benzene ring of molecule 22f cannot stretch to the
important” Cys276 and Asp277. In addition, the para-carboxyl group, due to steric hindrances,
“
causes a noticeable displacement of 22f. As a result of this displacement, H-bonding of the
quinone carbonyl oxygen atom with the key Met215 becomes impossible. Thus, our docking
results satisfactorily explain the observed binding activities of these substituted
naphthoquinones.
Shraga et al. [33] recently reported that the MKK7 binding site is surrounded by Leu266,
Val196, Met212, Ala163, Val150, Cys276, and Cys218. Six of these residues are present in the
nearest vicinity of the docking poses reported here, although they performed covalent docking
of acrylamides, which cannot be directly compared with the results of our non-covalent
docking. Nevertheless, the ligand-receptor complexes modelled by us might be suitable starting
structures for subsequent covalent interaction of the naphthoquinones with MKK7, e.g. with
Cys276. This issue needs further investigation.
2.5. Activity of the naphthoquinones toward Cdc25A and Cdc25B
1
1

Most of the known inhibitors of Cdc25 phosphatases are quinone-based derivatives (e.g.,
5, 6]). Thus, the naphthoquinones under investigation were also examined for their effects on
[
Cdc25A/B activity. Our set of the compounds included shikonin (1), NSC 95397 (4), Cpd C (5),
and menadione (18), which were previously reported as inhibitors of Cdc25 [6, 17, 34, 44-46].
IC values for inhibition of Cdc25A and B isoforms are presented in Table 1. Compounds 6-
5
0
13, 22f, and plumbagin (2), together with the reference compounds 1, 4, 5, and 18 were the
most potent Cdc25A/B inhibitors. As an example, the dose-dependent inhibition of Cdc25B by
plumbagin (2) is shown in Figure 2. Plumbagin (2) was even more active than menadione (18),
which lacks a hydroxyl group. Compounds 6-13 are related sulfur-contained analogs of know
Cdc25 inhibitors 4 and 5. Compound 24 with a high degree of similarity to the structure of NSC
6
63284, a potent Cdc25 inhibitor [44], was inactive. IC50 values for inhibition of Cdc25A
versus inhibition of Cdc25B had a good linear correlation (r=0.83) with one exception
compound 16), suggesting that almost all inhibitors targeted both Cdc25A and Cdc25B.
(
Although there are many publications on the anticancer activity of plumbagin (e.g., [47]), this is
a first report that this natural compound inhibits Cdc25A and B.
To evaluate inhibitory specificity, we analyzed the effects of the naphthoquinones on
activity of human neutrophil elastase (HNE), a member of the chymotrypsin family of serine
proteases. All compounds, except one (compound 8) had no HNE inhibitory activity (Table 1).
2.6. Molecular docking of selected naphthoquinones with Cdc25B
We performed molecular docking of selected naphthoquinones (1, 2, 4, 6, 7, 13, 15, 22e,
and 22f) into Cdc25B (PDB code 1QB0). The binding site is located near Cys473 and Tyr428
and corresponds to one of the three binding sites (also designated as “active sites”) suggested by
docking and molecular dynamics simulations [48]. Previously, it was reported that Cys473 is
catalytically essential for substrate activation, and inhibition of Cdc25B activity can occur by
the oxidation of Cys473 through ROS production [22, 49]. Thus, Cys473 was specified as the
1
2

ligand reactive group (see also [6]), and docking of various ligands to this site was carried out
previously [6, 48].
Our docking studies with Cdc25B were performed with the use of the ROSIE server [50]
and accounted for full flexibility of both ligand and the closest amino acid residues within the
Cdc25B binding site. Supplementary Table S2 shows the obtained distances between the
closest heavy (i.e., non-hydrogen) atoms of Cys473 and of the ligand. Energies for the
interaction of a ligand with Cys473 and Tyr428 (partial docking scores) are also presented.
Tyr428 was mentioned earlier as one of the important residues in a docking study of shikonin
[
6]. Due to the larger size of tyrosine, the partial docking scores related to Tyr428 are higher
than those for Cys473, since they include van der Waals interactions with each of the atoms of
the amino acid residue. In addition, most active compounds form H-bonds or participate in π-π
stacking interactions with Tyr428.
All of the active compounds are located within the binding site near Cys473, which is
favorable for electron transfer from the cysteine sulfur atom to the lowest unoccupied molecular
orbital (LUMO) of the naphthoquinone moiety. It should be noted that the docking pose of
shikonin is in agreement with the results of Zhang et al. [6]. This ligand is connected to the
binding site due to H-bonding of Tyr428 and Arg479 with the OH group and carbonyl oxygen
in the naphthoquinone moiety. In addition, Arg479 forms an H-bond with the hydroxyl group of
the shikonin side chain. The location of plumbagin, another natural naphthoquinone, within the
binding site is shown in Figure 3. Plumbagin forms H-bonds with Glu474 and Arg482 due to
quinone carbonyl groups, as well as an H-bond to Arg479 with the participation of the phenolic
hydroxyl group. The less active compound 15 is anchored at a distance > 7 Å from Cys473 due
to H-bonding of quinone carbonyl groups to Arg482 and Arg544 (Supplementary Table S2).
Naphthoquinone 13, which is a sulfur-containing analogue of 15, also forms an H-bond between
one of the quinone oxygen atoms and Arg482 in its best docking pose (Supplementary Figure
S4). It should be noted that in the docking pose of 13, a cation-π interaction may occur between
1
3

the charged Arg479 and the closely located π-electron system of the ligand. In addition, a π-π
interaction between the naphthoquinone and Tyr428 benzene rings is possible; the distance
between the centers of the rings is 4.00 Å, while the distance between their planes is about 3.2
Å. A similar π-π stacking with Tyr428 was previously considered for quinolinedione NSC
663284 [48]. Arg479 is involved in H-bonding to the quinone oxygen atom of molecule 4
(Supplementary Figure S5). The possibility of Tyr428 π-π interaction with the ligand was also
found for naphthoquinones 6 and 7 (Supplementary Figure S6). Molecule 6 forms H-bonds
with Arg482 and Arg544, with the participation of a quinone oxygen atom, as well as with
Glu478 and Arg482, with the participation of the carboxyl group. Compound 7 has a similar
orientation within the binding site; there is an H-bond between the quinone oxygen atom and
Arg482, while the carboxyl group forms H-bonds with Glu474, Phe475 and Ser477.
According to our docking results, derivatives 22e and 22f, containing a carboxyl-
substituted phenylamine fragment, are differently oriented within the Cdc25B binding site.
Molecule 22e forms strong H-bonds via two oxygen atoms of the carboxyl group with Arg479
(Figure 4). In addition, there is a π-π interaction of the carboxyl-substituted benzene ring in 22e
with Tyr428. As a result, 22e is anchored far from the key Cys473. A carboxyl group in the
para-position of the benzene ring in 22f does not have a possibility for similar effective H-
bonding to Arg479. Compound 22f is located in the binding site such that Arg479 forms H-
bonds with a quinone carbonyl group and simultaneously participates in cation-π interaction
with the benzene ring of the phenylamino substituent. There is also an additional strong H-bond
between Tyr428 and the naphthoquinone carbonyl group. These interactions facilitate location
of 22f in the vicinity of Cys473, which is consistent with its higher biological activity compared
to 22e. It should be noted that an orientation of 22e inside the binding site similar to molecule
22f would be very difficult due to a steric repulsion of the ligand carboxyl group in the meta-
position from Arg482 and Arg544.
2.7. Anticancer cell activity of selected naphthoquinones
1
4

Twenty selected naphthoquinone derivatives were tested for their in vitro anticancer
activity against a panel of nine tumor cell lines and primary peripheral blood mononuclear cells
(
PBMCs) in comparison with sunitinib, a known anticancer drug. The human cancer cell lines
used were lung carcinoma (A549), synovial sarcoma (SW-982), breast adenocarcinoma
MCF7), colorectal adenocarcinoma (LS-174t), monocytic leukemia (THP-1, MonoMac6,
(
U937), promyelocytic leukemia (HL60), and leukemic Jurkat T cells. The cells were treated
with the selected compounds for 24 h at different concentrations (up to 50 µM).
The selected compounds displayed moderate to strong cytotoxic activity against the
tested cancer cell lines (Table 2). For most of the cell lines, the greatest cytotoxic activity (IC50
<
10 µM) was found for compounds 1, 2, 4, 9-13, 18, 22e, and 22f. Furthermore, cytotoxicity of
many compounds was even higher in comparison with reference compound sunitinib. As an
example, the concentration-dependent effect of shikonin (1) and plumbagin (2) on viability of
HL60 cells is shown in Figure 5. Compound 13 was 2- to 8-fold more toxic than 15 in seven of
the cell lines, indicating that substitution of oxygen with sulfur atoms in R /R could increase
1
2
biological activity. The slightly high hydrophobicity of 13 (LogP = 2.83) may contribute to its
enhanced cytotoxicity versus 15 (LogP = 1.29). Although lapachol was previously found to be
active against the Walker-256 carcinoma and Yoshida sarcoma [51], we did not find any
toxicity for this compound against all 9 tumor cell lines tested or the primary PBMCs (Table 2).
Similarly, the 14, a known inhibitor of Snail-p53 binding [9], was weakly toxic or nontoxic for
the tumor cell lines and PBMCs. It should be noted that 22e exhibited good selectivity against
SW-982 synovial sarcoma cells, as compared to PBMCs. Interestingly, there was a definite
linear correlation between activity of the compounds toward Cdc25A/B and anticancer cell
activity for most of the tested cell lines, whereas no correlation was found between cancer cell
line cytotoxicity and binding affinity for MKK7 (Table 3). However, this tendency could also
reflect an ability of naphthoquinones to generate ROS and(or) their redox potential. Indeed,
both inhibitory activity of naphthoquinones toward Cdc25A/B and anticancer cell activity are
1
5

determined by redox potentials of the compounds [52]. Although 6 was weakly active or
inactive in all eight tumor cell lines and primary PBMCs, it had strong cytotoxic activity against
Jurkat T cells (IC = 0.94 µM). Thus, the mechanism of selective cytotoxicity of 6 in Jurkat T
5
0
cells should be investigated in future.
2.8. Density functional theory (DFT) study
We performed DFT analysis of selected naphthoquinone derivatives to identify
characteristics of their electronic structure related to anticancer cell activity/cytotoxicity and
enzyme inhibitory activity. Geometry optimizations were performed with the BP86 functional,
triple-zeta basis set def2-TZVPP. This level of theory provides high-quality results for
molecular geometries and was previously validated on several hundreds of organic molecules
[
53]. Subsequent single-point calculations were then made for each compound with the ωB97X-
D3 functional and the largest Pople basis set 6-311++G(3df,3pd) in order to obtain better results
for orbital energies [54].
We chose the following characteristics of electronic structure for further analysis: energy
of the highest occupied molecular orbital [E(HOMO)], energy of the lowest unoccupied
molecular orbital [E(LUMO)], vertical electron affinity (VEA), absolute hardness η, absolute
electronegativity χ, and reactivity index ω. Values of η, χ, and ω were calculated according to
equations 1-3 [55, 56]:
η = [E(LUMO) – E(HOMO)]/2 (eq. 1);
χ = – [E(LUMO) + E(HOMO)]/2 (eq. 2);
2
ω = χ /2η (eq. 3).
Although the ωB97X-D3 functional presumably leads to more correct electronic
structures, for comparison purposes we also considered the electronic properties calculated with
the BP86 functional, as they were available from the geometry optimization runs. The
1
6

magnitudes obtained at the BP86/def2-TZVPP level of theory are denoted here by subscript
BP86” (e.g. E(LUMO)BP86 χBP86, etc). The electronic characteristics for the investigated
“
,
compounds are presented in Supplementary Tables S2 and S3 together with estimated LogP
values. All of these characteristics were used as variables for pairwise correlation analyses vs
cytotoxic activity, the latter being expressed in a reciprocal scale (1/IC ) where inactive
5
0
compounds were assigned a value of 0. The correlation coefficients obtained are shown in
Table 4 and Supplementary Table 5S. The purpose of the correlation analysis was to establish
how tightly the cytotoxic activity was associated with main characteristics of electronic
structure and LogP of the investigated naphthoquinones.
According to the molecular orbital theory of chemical reactivity, a lower E(LUMO)
means the compound more easily accepts electrons in chemical reactions and thus plays a role
as an electrophilic reagent [57, 58]. We found that all compounds involved in our DFT study
were soft electrophiles, as defined by the level of E(LUMO) lower than 2.5 eV and that the
E(LUMO) affects their cytotoxic activities toward SW-982, Jurkat T cells, and U936 cells. A
negative correlation coefficient for E(LUMO) indicates that decreasing E(LUMO) values can
increase the 1/IC as a measure of cytotoxic activity. Therefore, we can reasonably speculate
5
0
that the naphthoquinones act as electron acceptors upon interaction with a target, and electron
transfer occurs from the target to the compound. It should be noted that an importance of
E(LUMO) for cytotoxicity of another group of compounds, amides of pyrazine-2-carboxylic
acid, was previously shown by Hosseini et al. [57]. However, for a group of celastroid
triterpenoids, it was found that cytotoxicity correlated better with E(HOMO) than with
E(LUMO) [59], although the author postulated that the studied triterpenoids undergo a
nucleophilic attack on their biological action.
Tendency of a molecule to accept an electron and form an anion radical can be directly
characterized by electron affinity. We calculated values of vertical electron affinities (VEA and
VEABP86) as energy differences between an anion radical and a corresponding neutral
1
7

naphthoquinone. Both species were taken at the geometry obtained in an optimization run for a
neutral molecule. Satisfactory and significant positive correlations were obtained between VEA
and 1/IC values for SW-982, Jurkat, and U963 cells, indicating again that higher electron
5
0
acceptor ability of the investigated naphthoquinones favors their biological activity in these cell
lines. As shown in Table 4, Supplementary Table S5, and Supplementary Figures S7 and
S8, tightness of the correlations was in general higher for E(LUMO) and VEA than for the
corresponding values of E(LUMO)BP86 and VEABP86 obtained with the BP86 functional. This
observation agrees with the results obtained by Lin et al. [54], which introduced the ωB97X-D3
functional as highly suitable for calculations of electronic properties, particularly orbital
energies of organic compounds.
Correlations of 1/IC50 values vs E(HOMO) and HOMO-related properties (absolute
hardness η and absolute electronegativity χ) were expectedly weaker, as HOMO does not
participate in one-electron reduction of naphthoquinones. However, the combined characteristic
ω (and also ωBP86) correlates satisfactorily with cytotoxic activity measures for SW-982, Jurkat,
and U963, and MCF7 cell lines. Similar results were obtained earlier for a group of vitamin K2
analogs by Ishihara and Sakagami [55] who also indicated correlation of the reactivity index ω
with cytotoxicity against HSC-2, HSC-3, HSC-4, HepG , and HL60 cells. Log P values did not
2
provide any significant correlations with 1/IC50 values for the cancer cell lines evaluated
(Supplementary Table S5).
Cdc25 phosphatases are primary targets for nucleophile attack by electrophilic
molecules, such as 1,4-naphtoquinones [34-38]. Thus, we have conducted pairwise correlation
analysis for the molecular DFT descriptors and inhibitory activity (1/IC ) of the
5
0
naphthoquinone derivatives for Cdc25A/B. Similar to our correlation analysis regarding
cytotoxicity, we found that the DFT-derived properties, including E(LUMO), VEA, absolute
hardness ηBP86, and reactivity index ω, correlated with Cdc25A inhibitory activity. With
Cdc25B, the correlation was found only for descriptors η, ω, and VEABP86. Negative correlation
1
8

coefficients were found for LUMO energy and absolute hardness, while positive values of r
were obtained for the reactivity index (Table 4 and Supplementary Table S5). These
correlations indicate that high electrophilicity of the compounds would be favorable for
Cdc25A/B inhibitory activity. In other words, our modeling showed that electron transfer
between the naphthoquinone molecules and Cdc25 is the most important factor contributing to
phosphatase inhibition.
Thus, electronic properties associated with the possibility to accept electrons
[
E(LUMO), VEA] and with general chemical reactivity (ω) are the most important variables
related to Cdc25 inhibition and cytotoxic anticancer cell activity of our naphthoquinone
derivatives. Importance of these properties is in accordance with the general concept of donor-
acceptor and nucleophile-electrophile interactions, which are to a great extent defined by the
energies of frontier molecular orbitals [56]. Indeed, the high values of ω for aryl substituted
flavones/flavanones and their analogs were previously found to be associated with high
cytotoxic activity [60].
3. Conclusion
Synthesis and analysis of novel 1,4-naphthoquinone derivatives demonstrated that
several of these compounds had high affinity for MKK7 and high inhibitory activity toward
Cdc25A/B. We report for the first time that four naphthoquinone derivatives 6-8, and 22e and
previously reported compound 4 (NSC 95397) are specific inhibitors of MKK7 vs MKK4, with
the most potent compound being 7. Our modeling experiments suggest that the length of side
chain in 7 could be optimal for supporting MKK7 inhibitory activity. Compounds 1, 2, 4, 6, 7,
1
3, and 22f were potent (IC < 5 µM) inhibitors of Cdc25A/B activity. Although plumbagin (2)
50
had weak affinity for MKK7, this compound and shikonin (1) were potent inhibitors of
Cdc25A/B. Molecular docking studies revealed that all of the naphthoquinone derivatives that
inhibited Cdc25A/B bound within the enzyme binding site near Cys473, which is favorable for
1
9

electron transfer from the cysteine sulfur atom to the LUMO of the naphthoquinone moiety.
Although we did not find a correlation between naphthoquinone binding affinity toward MKK7
and compound cytotoxicity, there was a definite correlation between activity of the compounds
toward Cdc25A/B and cytotoxic activity for eight of the nine cancer cell lines tested. Moreover,
6
and 7 could be prototypic molecules for design of analogs with dual MKK7/Cdc25 inhibitory
activity. Our DFT analysis showed that E(LUMO), VEA, and reactivity index (ω) are important
electronic structure characteristics related to cytotoxicity of the naphthoquinone derivatives.
Some characteristics derived from the DFT calculations also allowed us to suggest a physical
explanation for electronic properties of the molecules that could contribute to Cdc25 inhibition.
Finally, the analogs evaluated here could be successfully utilized for further development of
efficacious Cdc25A/B and MKK7 inhibitors with anticancer activity. The quantum chemistry
modelling of electronic structure by DFT or other computational approaches can be useful for
the estimation of naphthoquinone reactivity towards possible nucleophilic centers in a protein.
On the other hand, docking studies can provide an additional elucidation on reachability of
these centers by a ligand for electron transfer or covalent interaction within a binding site.
2
0

4
4
4
. Experimental section
.1. Chemistry
.1.1. General methods
All reagents were purchased from Sigma-Aldrich or Acros Organics and were used without
further purification. The melting points (in °C) were recorded in open capillaries and are
uncorrected. Mass spectra were recorded at 200 eV collision energy by the direct inlet method
o
on an MX 1321 mass spectrometer at 70 eV collision energy and source T = 200 C in the EI
mode. Mass spectra FAB were recorded on a VG 70-70 EQ mass spectrometer at 10 kV
collision energy, using 3-NO -benzyl alcohol as a solvent. Alugram SIL G/UV pre-coated
2
254
thin layer chromatography (TLC) sheets were used for TLC (Macherey-Nagel) with Eluent A:
1
13
benzene:acetone:acetic acid 10:1:0.1 and Eluent B: benzene:triethylamine 10:1. H- and C-
NMR spectra were recorded on a Bruker spectrometer at 400 MHz and 100 MHz, respectively,
with tetramethylsilane (TMS) as the internal standard (the aromatic protons designations are
presented below). For aromatic protons designations, see Supplementary Figure S9.
Representative NMR spectra for compounds 22a-g, 23a,b, 25a,b are provided in supplementary
material.
4
.1.2. 4-(3-Chloro-1,4-dioxo-1,4-dihydro-naphthalen-2-ylamino)-benzoic acid (22f)
The compound was synthetized, as described [41]. Briefly, a mixture of 2,3-dichloro-
,4-naphthoquinone (1.135 g, 5 mmol), 4-aminobenzoic acid (0.686 g, 5 mmol), and calcium
1
carbonate (0.501 g, 5 mmol) in 50 ml of 50% ethanol was heated under reflux for 10 h
presence of 2,3-dichloro-1,4-naphthoquinone was controlled by TLC in CHCl ). The mixture
(
3
was cooled to room temperature mixture and poured in 200 ml of H O. The precipitate was
2
filtered, washed with 1M HCl (2×20 ml) and H O (2×30 ml) and recrystallized from acetic acid.
2
o
1
Yield = 81% (1.33 g); M.p. = 361-363 C. NMR H (DMSO-d6) δ 7.16 (2H, d, J = 8.4 Hz, Hε),
7.83 – 7.90 (4H, m, Hα-Hδ), 8.06 (2H, br. d, J = 7.2 Hz, Hξ), 9.52 (1H, s, NH), 12.75 (1H, s,
2
1

13
COOH). NMR C (DMSO-d6) δ 118.2, 122.5, 125.7, 126.7, 127.1, 127.6, 129.9, 132.3, 134.0,
135.2, 143.2, 143.84, 167.5, 177.5, 180.5. Mass-spectra EI M/z (I,%): 328(40), 326(100),
2
92(30), 282(20), 248(55). C H ClNO ; M = 327.73 g/mol. Compounds 22a-e were obtained
17
10
4
r
in similar conditions.
4.1.3. (3-Chloro-1,4-dioxo-1,4-dihydro-naphthalen-2-ylamino)-acetic acid (22a)
o
1
Yield = 79%, M.p. = 208-210 C. NMR H (CDCl ): 3.90 (2H, d, J = 5.6 Hz, NHCH ),
3
2
6.57 (1H, br s, NH), 7.08 (1H, dd, J = 7.6 Hz, Hβ), 7.17 (1H, dd, J = 7.2 Hz, Hγ), 7.40 (2H, dd,
J = 7.6 Hz J = 8.0 Hz, Hα+Hδ), 12.05 (1H, br s, COOH). Mass-spectra EI M/z (I,%): 267(10),
1
2
2
65(30), 220(100), 157(20). C H ClNO ; Mr = 265.65 g/mol.
12 8 4
4.1.4. 3-(3-Chloro-1,4-dioxo-1,4-dihydro-naphthalen-2-ylamino)-propionic acid (22b)
o
1
Yield = 75%; M.p. = 178-180 C. NMR H (DMSO-d6): 2.63 (2H, t, J = 6.6 Hz,
CH COOH), 3.95 (2H, td, J = 6.6 Hz, NHCH ), 7.38 (1H, br s, NH), 7.75 (1H, dd, J = 7.8 Hz,
2
2
Hβ), 7.83 (1H, dd, J = 8.4 Hz, Hγ), 7.97 (2H, d, J = 7.8 Hz, Hα+Hδ), 12.39 (1H, br s, COOH).
1
3
NMR C (DMSO-d6) δ 35.6, 40.6, 106.3, 127.0, 127.6 130.5, 131.5, 132.4, 133.2, 135.4,
1
73.1, 175.9, 180.6. Mass-spectra EI M/z (I,%): 280(20), 279(75), 233(20), 222(25), 220(100),
57(35). C H ClNO ; Mr = 279.68 g/mol.
1
1
3
10
4
4.1.5. 4-(3-Chloro-1,4-dioxo-1,4-dihydro-naphthalen-2-ylamino)-butyric acid (22c)
o
1
Yield = 80%; M.p. = 173-174 C. NMR H (DMSO-d6) δ 1.85 (2H, m, CH CH CH ),
2
2
2
2.30 (2H, t, J = 6.6 Hz, CH COOH), 3.75 (2H, td, J = 6.6 Hz, NHCH ), 7.56 (1H, br s, NH),
2 2
7
.75 (1H, td, J = 7.8 Hz J = 1.2 Hz, Hβ), 7.83 (1H, td, J = 8.4 Hz J = 1.2 Hz, Hγ), 7.97 (2H,
1
2
1
2
13
d, J = 8.4 Hz, Hβ), 12.11 (1H, br s, COOH). NMR C (DMSO-d6) δ 26.4, 31.3, 43.8, 126.3,
26.9, 127.6, 130.4, 132.5, 133.1, 135.3, 145.8, 174.6, 176.4, 180.6. Mass-spectra EI M/z
I,%):295(10), 293(40), 240(30), 222(25), 220(100), 212(20). C H ClNO ; Mr = 293.97
1
(
1
4
12
4
g/mol.
2
2

4.1.6. 6-(3-Chloro-1,4-dioxo-1,4-dihydro-naphthalen-2-ylamino)-hexanoic acid (22d)
o
1
Yield = 71%; M.p. = 117-120 C. NMR H (DMSO-d6) δ 1.32 (2H, m,
(
CH ) CH (CH ) ), 1.52 (2H, m, HOOCCH CH (CH ) ), 1.61 (2H, m, NHCH CH (CH ) ),
2
2
2
2 2
2
2
2 3
2
2
2 3
2
.18 (2H, t, J = 7.2 Hz, CH COOH), 3.70 (2H, td, J = 7.2 Hz, NHCH )td, 7.53 (1H, br s, NH)br
2
2
s1 H, 7.74 (1H, td, J = 7.2 Hz J = 1.2 Hz, Hβ), 7.83 (1H, td, J = 7.8 Hz J = 1.2 Hz, Hγ), 7.97
1
2
1
2
1
3
(
2H, br d, J = 7.2 Hz, Hα+Hδ), 12.06 (1H, br s, COOH). NMR C (DMSO-d6) δ 24.9, 26.2,
1.1, 34.7, 44.3, 125.93, 126.3, 127.0, 130.4, 132.6, 133.1, 135.4, 145.6, 172.2, 175.8, 180.7.
Mass-spectra FAB M/z (I,%): 323(15). 321(40), 222(35), 220(100). C H ClNO ; Mr = 321.76
3
1
6
16
4
g/mol.
4.1.7. 3-(3-Chloro-1,4-dioxo-1,4-dihydro-naphthalen-2-ylamino)-benzoic acid (22e)
o
1
Yield = 83%; M.p. = 313-316 C. NMR H (DMSO-d6) δ 7.34 (1H, d, J = 8.4 Hz, Hξ),
7
.42 (1H, dd, J = 7.8 Hz J = 1.2 Hz, Hη), 7.68 - 7.70 (2H, m, Hξ+Hθ), 7.82 (1H, td, J = 7.2
1
2
1
Hz J = 1.2 Hz, Hβ), 7.88 (1H, td, J = 7.2 Hz J = 0.6 Hz, Hγ), 8.05 (2H, d, J = 7.8 Hz,
2
1
2
1
1
3
Hα+Hδ), 9.43 (1H, s, NH ), 13.145 (1H, br s, COOH); NMR C (DMSO-d6) δ 115.4, 124.8,
25.4, 126.6, 127.0, 128.0, 128.4, 130.8, 132.4, 133.7, 135.2, 139.5, 143.6, 167.9, 177.3, 180.4.
Mass-spectra EI M/z (I,%): 329(40), 327(100), 292(35), 274(90), 248(30). C H ClNO ; Mr =
1
1
7
10
4
327.73 g/mol.
4.1.8. 4-(3-Chloro-1,4-dioxo-1,4-dihydro-naphthalen-2-ylamino)-benzoic acid methyl ester
(22g)
A mixture of 2,3-dichloro-1,4-naphthoquinone (1.135 g, 5 mmol), 4-amino-benzoic acid
methyl ester (3.023 g, 20 mmol), and sodium acetate trihydrate (0.68 g, 5 mmol) in 50 ml of 50
ethanol was heated under reflux for 12 h (presence of 2,3-dichloro-1,4-naphthoquinone was
%
controlled by TLC in CHCl ). The mixture was cooled to room temperature and poured in 200
3
ml of water. The precipitate was filtered, washed with 1M HCl (3×20 ml) and H O (3×30 ml)
2
o
1
and recrystallized from ethanol. Yield = 76%; M.p. = 281-283 C. NMR H (DMSO-d6) δ 3.84
2
3

(
3H, s, COOCH ), 7.18 (2H, d, J = 9.0 Hz, Hε), 7.84 (1H, dt, J = 7.8 Hz, J = 1.8, Hβ), 7.87-
3
1
2
13
7
.91 (3H, m, Hα+Hδ+Hγ), 8.06 (2H, br t, Hξ), 9.56 (1H, s, NH)s 1H NH; NMR C (DMSO-
d6) δ 52.4; 119.0; 122.5; 124.4; 126.7; 127.1; 129.7; 131.0; 132.3; 134.0; 135.2; 143.1; 144.3;
166.4; 177.5; 180.4. Mass-spectra EI M/z (I,%): 343(33), 341(100), 310(60), 282(22), 262(20),
2
47(20), 138(20). C H ClNO ; Mr = 341.75 g/mol. Compounds 23a,b were obtained under
1
8
12
4
similar conditions.
4.1.9. 2-Chloro-3-(2-hydroxy-ethylamino)-1,4-naphthoquinone (23a)
o
1
Yield = 82%; M.p. = 154-156 C. NMR H (DMSO-d6) δ 3.60 (2H, dt, J = 5.4 Hz,
NHCH ), 3.83 (6H, dt, J = 6.0 Hz, CH OH), 4.93 (1H, t, J = 5.4 Hz, CH OH), 7.24 (1H, br s,
2
2
2
NH), 7.75 (1H, dd, J = 7.8 Hz, Hγ), 7.83 (1H, dd, J = 7.2 Hz, Hβ), 7.97 (2H, dd, J = 7.8 Hz J2
1
1
3
=
3 Hz, Hα+Hδ); NMR C (DMSO-d6) δ 46.8, 59.6, 100.4, 126.3, 126.9, 131.9, 132.5, 133.2,
1
35.4, 140.1, 173.9, 180.6. Mass-spectra EI M/z (I,%): 253(10), 251(400), 235(20), 233(55)
22(50), 220(100). C H ClNO ; Mr = 251.67 g/mol.
2
1
2
10
3
4.1.10. 2-Chloro-3-(3-hydroxy-propylamino)-1,4-naphthoquinone (23b)
o
1
Yield = 77%; M.p. = 120-121 C. NMR H (DMSO-d6) δ 1.775 (2H, m, CH CH CH )
2 ,
2
2
3
.51 (2H, dt, J = 5.4 Hz, NHCH ), 3.82 (2H, dt, J = 6.6 Hz, CH OH), 4.66 (1H, t, J = 4.8 Hz,
2
2
CH OH),7.54 (1H, br s, NH), 7.74 (1H, dd, J = 7.8 Hz, Hγ), 7.83 (1H, dd, J = 7.2 Hz, Hβ), 7.97
2
13
(
2H, dd, J = 7.8 Hz J = 2.4 Hz, Hα+Hδ); NMR C (DMSO-d6) δ 33.7, 42.7, 59.1, 108.5,
1 2
1
26.3, 126.9, 130.4, 132.6, 133.1, 135.4, 145.8, 175.8, 180.6. Mass-spectra EI M/z (I,%):
67(10), 265(50), 222(30), 220(100). C H ClNO ; Mr = 265.70 g/mol.
2
1
3
12
3
4.1.11. 2-Chloro-3-(3-morpholin-4-yl-propylamino)-1,4-naphthoquinone (24)
A mixture of 2,3-dichloro-1,4-naphthoquinone (2.27 g, 10 mmol), 3-morpholin-4-yl-
propylamine (1.586 g, 11 mmol), and sodium acetate trihydrate (2.722 g, 20 mmol) in 100 ml of
ethanol was heated under reflux for 16 h (presence of 2,3-dichloro-1,4-naphthoquinone was
2
4

o
controlled by TLC in CHCl ). The mixture was cooled to -18 C, filtered, washed with cooled
3
ethanol (2×5 ml) and H O (2×10 ml) and recrystallized from ethanol. Yield = 95%; M.p. = 175-
2
o
1
1
76 C. NMR H (DMSO-d6) δ 1.77 (2H, m, CH CH CH ), 2.35 – 2.39 (6H, m, CH NH(CH
2 2 2 2 2
CH ) O), 3.58 (4H, br s, NH(CH CH ) O), 3.83 (2H, dt, J = 7.2 Hz, NHCH ), 7.74 (1H, dd, J =
2
2
2
2 2
2
1
3
7
.2 Hz, Hβ), 7.83 (1H, dd, J = 7.8 Hz, Hγ), 7.97 (2H, d, J = 7.8 Hz, Hα+Hδ); NMR C
(DMSO-d6) δ 26.8, 44.1, 53.9, 56.8, 66.5, 126.3, 126.9, 130.4, 132.6, 133.1, 135.4, 146.1,
1
75.7, 180.7. Mass-spectra EI M/z (I,%): 334(10), 220(40), 114(55), 100(100). C H ClN O ,
17 19 2 3
Mr = 334.81 g/mol.
4.1.12. 4-(1,4-Dioxo-3-piperidin-1-yl-1,4-dihydro-naphthalen-2-ylamino)-benzoic acid methyl
ester (25a)
The compound was synthetized, as described previously [39], using 22g (see Scheme 1).
To a solution of 22g (144 mg, 0.42 mmol) in 2 ml of ethanol, 0.5 ml (426 mg, 5 mmol) of
piperidine was added. The mixture was heated under reflux for 20 h. The mixture was cooled to
o
-
18 C, filtered, and washed with cooled ethanol (2×1 ml) and H O (3×1 ml). Yield = 51%; M.p.
2
o
1
=
186-188 C. NMR H (DMSO-d6) δ 1.36 (6H, br s, NH(CH CH ) CH ), 3.14 (4H, br s,
2 2 2 2
NH(CH CH ) CH ), 3.79 (3H, s, COOCH ), 6.92 (2H, d, J = 8.4 Hz, Hε), 7.75 – 7.79 (4H, m,
2
2 2
2
3
13
Hα-Hδ), 7.93-7.97 (2H, m, Hξ)m 2 H, 8.43 (1H, s, NH); NMR C (DMSO-d6) δ 24.2, 26.3,
50.1, 52.0, 116.8, 120.0, 125.5, 126.5, 126.5, 130.3, 131.1, 132.5, 133.8, 140.9, 147.2, 166.6,
+
1
81.4, 182.2. Mass-spectra FAB M/z (I,%): 391(100) [M+H] . C H N O ; Mr = 390.44 g/mol.
23
22
2
4
4.1.13. 4-(3-Methoxy-1,4-dioxo-1,4-dihydro-naphthalen-2-ylamino)-benzoic acid methyl ester
(25b)
Compound 22g (144 mg, 0.42 mmol) was added to a solution of sodium methylate (46
mg, 0.84 mmol) in 2 ml of methanol. The mixture was heated under reflux for 4 h. The mixture
o
was cooled to -18 C, filtered, and washed with cooled methanol (2×1 ml) and H O (3×1 ml).
2
o
1
Yield = 64%; M.p. 199-201 C. NMR H (DMSO-d6) δ 3.66 (3H,s, OCH ), 3.82 (3H, s,
3
2
5

COOCH ), 7.05 (2H, d, J = 8.4 Hz, Hε), 7.79-7.87 (4H, m, J = 5.4 Hz, Hα+Hδ+Hξ), 7.99 (2H,
3
13
dd, J = 4.2 Hz J = 3.6 Hz, Hβ+Hγ), 8.82 (1H, s, NH); NMR C (DMSO-d6) δ 52.2, 59.9,
1
2
1
1
=
19.8, 122.3, 126., 126.3, 129.8, 130.8, 131.9, 132.9, 133.8, 134.8, 144.2, 146.0, 166.5, 179.8,
82.8. Mass-spectra EI M/z (I,%): 337(10), 294(15), 133(20), 104(20), 76(20). C H NO ; Mr
1
9
15
5
337.34 g/mol.
4.2. Other compounds
Compounds 1-21 (see Table 1) and six 4-oxo-1,4-dihydroquinazoline and quinazolin-
(3H)-one analogs (see Supplementary Table S1) were purchased from Princeton
4
BioMolecular Research (Princeton, NJ, USA), Cayman Chemicals (Ann Arbor, MI, USA),
Sigma-Aldrich (St. Louis, MO, USA), ChemShuttle (Hayward, CA, USA), AlfaAesar
(Haverhill, MA, USA), Totonto Research Chemicals (Toronto, ON, Canada), and Fluka (Buchs,
Switzerland).
4.3. Kinase assay
Dissociation constant (K ) determination was performed by KINOMEscan (Eurofins
d
Pharma Discovery, San Diego, CA, USA), as described previously [61, 62]. In brief, MKK4
and MKK7 were produced and displayed on T7 phage or expressed in HEK-293 cells. Binding
reactions were performed at room temperature for 1 h, and the fraction of kinase not bound to a
test compound was determined by capture with an immobilized affinity ligand and quantified by
quantitative polymerase chain reaction. Primary screening at fixed concentrations of compounds
was performed in duplicate. For K determination, a 12-point half-log dilution series (a
d
maximum concentration of 30 µM) was used. Assays were performed in duplicate, and their
average mean value is displayed. Reference anticancer compound sunitinib was obtained from
Tocris Bioscience (Ellisville, MO, USA).
4.4. Cdc25 inhibition assay
2
6

CycLex® protein phosphatase Cdc25A and B fluorometric assay Kits (Medical &
Biological Laboratories Co., Ltd, Japan) were used to evaluate enzyme inhibition by
naphthoquinone analogues. Cdc25A and B activity was measured in a 96-well half-area
microtiter plate using 3-O-methylfluorescein phosphate (OMFP) as a substrate. Five µL
(0.1ꢀµg/µL) of purified recombinant Cdc25 (Cdc25A or B) was mixed with 40ꢀµL of assay
mixture and incubated with 5ꢀµL of the test compound at various concentrations in a well.
Kinetic measurements were obtained every 30 s for 10 min at 25 °C using a Fluoroskan Ascent
FL fluorescence microplate reader (Thermo Electron, MA) with excitation and emission
wavelengths at 485 and 538 nm, respectively. For all compounds tested, the concentration of
inhibitor that caused 50% inhibition of the enzymatic reaction (IC ) was calculated by plotting
5
0
%
inhibition versus logarithm of inhibitor concentration.
4
.5. HNE inhibitory assay
The HNE inhibition assay was performed in black flat‐bottom 96‐well microtiter plates.
Briefly, a buffer solution containing 200ꢀmM Tris–HCl, pH 7.5, 0.01% bovine serum albumin,
.05% Tween‐20, and 20ꢀmU/mL HNE (Calbiochem, San Diego, CA, USA) was added to wells
0
containing different concentrations of each compound. The reactions were initiated by addition
of 25ꢀµM elastase substrate (N‐methylsuccinyl‐Ala‐Ala‐Pro‐Val‐7‐amino‐4‐methylcoumarin,
Calbiochem) in a final reaction volume of 100ꢀµL/well. Kinetic measurements were obtained
every 30ꢀs for 10 min at 25ꢀ°C using a Fluoroskan Ascent FL fluorescence microplate reader
(Thermo Electron, MA) with excitation and emission wavelengths at 355 and 460ꢀnm,
respectively. The concentration of inhibitor that caused 50% inhibition of the enzymatic
reaction (IC ) was calculated by plotting % inhibition versus logarithm of inhibitor
5
0
concentration.
4.6. Cell culture
2
7

All cells were cultured at 37°C in a humidified atmosphere containing 5% CO . A549
2
carcinoma cells were cultivated in F-12K medium, supplemented with 10% fetal bovine serum
(FBS). SW-982 synovial sarcoma cells were cultivated in Leibovitz's L-15 medium with 10%
FBS. MCF7 breast adenocarcinoma cells were cultivated in Dulbecco's modified Eagle
medium (DMEM), supplemented with 10 µg/ml bovine insulin and 10% FBS. LS-174t
colorectal adenocarcinoma cells were cultivated in Eagle's minimum essential medium with
1
0% FBS. HL60 cells were cultivated in DMEM with 20% FBS. THP-1, U937, and Jurkat T
cells were cultivated in RPMI-1640 medium, supplemented with 10% FBS. MonoMac6 cells
Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Braunschweig, Germany)
(
were grown in RPMI 1640 medium supplemented with 10% FBS and 10 µg/ml bovine insulin.
All cultural media also contained 100 µg/ml streptomycin and 100 U/ml penicillin.
PBMCs were purified from human blood using dextran sedimentation, followed by
Histopaque 1077 gradient separation and hypotonic lysis of erythrocytes. Blood was collected
from healthy donors in accordance with a protocol approved by the Institutional Review Board
at Montana State University, and informed consent was obtained from all donors for the use of
their blood in this study (Protocol #MQ041017). PBMCs were cultured in RPMI 1640 medium
with 10% FBS, 100 µg/ml streptomycin, and 100 U/ml penicillin at 37°C in a humidified
atmosphere containing 5% CO . Cell number and viability were assessed microscopically using
2
trypan blue exclusion.
4.7. Anticancer cell cytotoxicity assay
Cytotoxicity was analyzed with a CellTiter-Glo Luminescent Cell Viability Assay Kit
from Promega (Madison, WI, USA), according to the manufacturer's protocol. Cells were
treated with compound under investigation and incubated for 24 h. After treatment, the cells
were allowed to equilibrate to room temperature for 30 min, substrate was added, and the
samples were analyzed with a Fluoroscan Ascent FL (Thermo Fisher Scientific, Waltham, MA,
2
8