New imidazoquinoxaline derivatives: Synthesis, biological evaluation on melanoma, effect on tubulin polymerization and structure-activity relationships

Article abstract of DOI:10.1016/j.bmc.2016.04.004

Microtubules are considered as important targets of anticancer therapy. EAPB0503 and its structural imidazo[1,2-a]quinoxaline derivatives are major microtubule-interfering agents with potent anticancer activity. In this study, the synthesis of several new derivatives of EAPB0503 is described, and the anticancer efficacy of 13 novel derivatives on A375 human melanoma cell line is reported. All new compounds show significant antiproliferative activity with IC₅₀ in the range of 0.077-122 μM against human melanoma cell line (A375). Direct inhibition of tubulin polymerization assay in vitro is also assessed. Results show that compounds 6b, 6e, 6g, and EAPB0503 highly inhibit tubulin polymerization with percentages of inhibition of 99%, 98%, 90%, and 84% respectively. Structure-activity relationship studies within the series are also discussed in line with molecular docking studies into the colchicine-binding site of tubulin.

Full text of DOI:10.1016/j.bmc.2016.04.004

Accepted Manuscript
New imidazoquinoxaline derivatives: Synthesis, biological evaluation on mel-
anoma, effect on tubulin polymerization and structure-activity relationships
Zahraa Zghaib, Jean-François Guichou, Johanna Vappiani, Nicole Bec, Kamel
Hadj-Kaddour, Laure-Anaïs Vincent, Stéphanie Paniagua-Gayraud, Christian
Larroque, Georges Moarbess, Pierre Cuq, Issam Kassab, Carine Deleuze-
Masquéfa, Mona Diab-Assaf, Pierre-Antoine Bonnet
PII:
S0968-0896(16)30226-7
DOI:
http://dx.doi.org/10.1016/j.bmc.2016.04.004
BMC 12918
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
11 December 2015
25 March 2016
1 April 2016
Please cite this article as: Zghaib, Z., Guichou, J-F., Vappiani, J., Bec, N., Hadj-Kaddour, K., Vincent, L-A.,
Paniagua-Gayraud, S., Larroque, C., Moarbess, G., Cuq, P., Kassab, I., Deleuze-Masquéfa, C., Diab-Assaf, M.,
Bonnet, P-A., New imidazoquinoxaline derivatives: Synthesis, biological evaluation on melanoma, effect on tubulin
polymerization and structure-activity relationships, Bioorganic & Medicinal Chemistry (2016), doi: http://
dx.doi.org/10.1016/j.bmc.2016.04.004
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. 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.

New imidazoquinoxaline derivatives: Synthesis, biological evaluation on
melanoma, effect on tubulin polymerization and structure-activity
relationships
Zahraa Zghaib1,2, Jean-François Guichou3,4, Johanna Vappiani1, Nicole Bec5, Kamel
Hadj-Kaddour1, Laure-Anaïs Vincent1, Stéphanie Paniagua-Gayraud1, Christian
Larroque5, Georges Moarbess2, Pierre Cuq1, Issam Kassab2, Carine Deleuze-
Masquéfa1*, Mona Diab-Assaf2, Pierre-Antoine Bonnet1
1 Institut des Biomolécules Max Mousseron (IBMM), UMR 5247, CNRS, Université de
Montpellier, ENSCM, Faculté de Pharmacie, 15, avenue Charles Flahault, BP14491, 34093
Montpellier cedex 5, France
2 Tumorigenèse et Pharmacologie Antitumorale, Lebanese University, BP 90656, Fanar
Jdeideh, Lebanon.
3 CNRS, UMR5048 – Université de Montpellier, Centre de Biochimie Structurale, F-34090
Montpellier, France
4 INSERM, U1054, F-34090 Montpellier, France
5 IRCM, Institut de Recherche en Cancérologie de Montpellier, INSERM U896, 34298
Montpellier, France.
* Corresponding author. Tel.: +33 4 11 75 95 36; fax +33 11 75 95 53
E-mail adress: carine.masquefa@umontpellier.fr
1

Abstract
Microtubules are considered as important targets of anticancer therapy. EAPB0503 and its
structural imidazo[1,2-a]quinoxaline derivatives are major microtubule-interfering agents
with potent anticancer activity. In this study, the synthesis of several new derivatives of
EAPB0503 is described, and the anticancer efficacy of 13 novel derivatives on A375 human
melanoma cell line is reported. All new compounds show significant antiproliferative activity
with IC50 in the range of 0.077-122 µM against human melanoma cell line (A375). Direct
inhibition of tubulin polymerization assay in vitro is also assessed. Results show that
compounds 6b, 6e, 6g, and EAPB0503 highly inhibit tubulin polymerization with percentages
of inhibition of 99%, 98%, 90%, and 84% respectively. Structure-activity relationship studies
within the series are also discussed in line with molecular docking studies into the colchicine-
binding site of tubulin.
Keywords: antiproliferative activity; tubulin depolymerization; human melanoma cancer cell
line (A375); structure-activity relationship; molecular docking; imidazo[1,2-a]quinoxaline
2

1. Introduction
Microtubules are the key components of the cytoskeleton of eukaryotic cells and have an
important role in various cellular functions such as mitosis, exocytosis, and maintenance of
cellular morphology, active transport, cell shape and polarization.1,2 They play a critical role
in cell division by their involvement in the movement and attachment of the chromosomes
during various stages of mitosis. Therefore microtubule dynamics is an important target for
the development of anti-cancer agents.3-5 Microtubules are composed of α/β tubulin
heterodimers, always in a state of equilibrium.3,4 Microtubule targeting agents (MTA), drugs
that interfere with microtubule dynamic stability, are widely employed in the clinic to treat a
variety of cancers or are exploited as probes to gain insights into microtubule structure and
function.4-10 MTAs are antimitotic agents which perturb not only mitosis but also arrest cells
during interphase. MTAs are known to interact with tubulin through at least four binding
sites: the laulimalide, taxane/epothilone, vinca alkaloid, and colchicine sites.1,11 Colchicine
binds to the intradimeric α-β interface of tubulin heterodimers contiguous to the GTP-binding
12-15
domain of the α-tubulin subunit
and the tubulin-colchicine complex prevents further
polymerization of the microtubule. Such complex brings about a conformational change
which blocks the tubulin dimers from further addition and prevents the growth of the
microtubule by producing potent destabilization effects of microtubules, resulting in the
subsequent shutdown of existing tumor vasculature.1,16
Efforts have been increasing for new compounds exhibiting inhibition of tubulin assembly
and disassembly in order to induce anti-proliferative activities and treat a wide variety of
malignancies.17-22 The alteration of microtubule dynamics prevents cells division and
apoptosis of different human cancer lines including Multi Drug Resistant (MDR) cancer cell
lines.23-25
Imidazoquinoxalines derivatives have been designed by our laboratory since 2004 using
different strategies of synthesis. Among these derivatives, EAPB0203 and EAPB0503
(Figure 1) have shown potent antitumor properties in vitro and in vivo against melanoma and
T-cell lymphomas.26-29 Preliminary studies have shown antitubuline activity for those two
derivatives.30
3

EAPB0203
EAPB0503
Figure 1: Chemical structure of lead compounds.
Previous structure-activity relationship (SAR) studies with imidazoquinoxalines analogues
were reported.26,27 In continuation of previous work and in order to better study the possible
correlation between the antiproliferative effect of our derivatives on human melanoma cell
line and microtubule disruption, a new series of imidazoquinoxalines derivatives has been
designed. Their biological activities were determined on the human melanoma cancer cell line
A375 and possible correlations with their effect on tubulin polymerization were determined.
Complementary molecular modeling studies on the colchicine binding domains of tubulin
contribute to give new insights on structure activity relationships and possible development of
the series.
2. Results and discussion
2.1.Chemistry
2.1.1. Synthesis of EAPB0503 derivatives
EAPB0503 and its derivatives have attracted our attention since 2004, due to their promising
high potential activity against different types of cancer. They have been synthesized using a
new strategy detailed previously.26,27 Based on these encouraging first data on anticancer
activity, new imidazoquinoxaline derivatives were designed. They have been obtained using
the same strategy of synthesis. The advantage of this new synthetic method remains in the fact
that all considered imidazoquinoxalines can be obtained from the same and common
intermediate, the compound 1 (Scheme 1).The substitution for diversity is then performed at
the last steps of the synthesis. The use of microwave assistance allows to dramatically
4

optimize the time of reactions, purity of the intermediates, and global yields. It is also to be
noted that the synthesis of the parent compound 1 can also be optimized using microwave
assistance by treating the 4,5-lactame derivative precursor with phosphorus oxychloride and
N,N-diethylaniline in a sealed vial using a Biotage® initiator microwave synthesizer.
Nevertheless, the crude compound obtained still requires further purification on silica gel
column chromatography for the next steps of the synthetic scheme.
First diversity on position 4 is obtained by substitution of the chlorine on compound 1 by the
appropriate amine under microwave heating to give compounds 2 and 3. The bromination on
position 1 of compounds 2 and 3 by N-bromosuccinimide under microwave conditions leads
to compound 4 and 5, respectively. In comparison to the initial synthetic procedure,26-28 steps
a and b (Scheme 1) as well as the step leading to the parent compound 1, have been optimized
using microwave irradiation instead of conventional heating.
Scheme 1: Synthesis of EAPB0503 derivatives. a) EtOH, NH2CH3 in ethanol or NH4OH in water,
MW (180°C, 20 min); b) NBS, CHCl3, reflux 1h30; c) Suzuki-cross coupling (R1’–B(OH)2),
Pd(PPh3)4, Na2CO3, DME, MW (140°C, 20 min); d) Iodocyclohexane, DMF, reflux 16h.
A second step for diversity is obtained using different arylboronic acids in Suzuki-cross
coupling reactions. As expected, the palladium catalyzed substitution selectively occurs on the
bromo position to obtain compounds 6a to 6j, and 7a under microwave assistance (140°C, 20
min) with good to acceptable yields. The phenol derivatives 8 and 9 can be obtained by a
treatment of the methoxy compounds EAPB0503 and 7a with iodocyclohexane under
conventional heating. Compounds 2 to 4, EAPB0503, 6a and 6b were described with their
analytical data in previous studies.26-28
2.2. Biological evaluation
5

2.2.1. In vitro cell growth inhibition
All compounds were tested for their cytotoxic effect on human melanoma cell line (A375),
our in vitro model used previously for screening,26-28 using EAPB0503 as reference, which
showed previously an important cytotoxic activity comparing to fotemustine, clinically used
for human melanoma treatment.26-28 The screening results reported in Table 1 demonstrate
that all our new compounds showed significant antiproliferative activity with IC50 in the range
of 0.077-122 µM against human melanoma cell line used. Compounds 6b, 6d, 6e, 6g, 7a and
8 were distinctly more potent from the other compounds. However, 6g was the only
compound presenting the highest significant antiproliferative activity in comparison to
EAPB0503 (P=0.002<0.05). It displayed an IC50 values about 0.077 µM, 3 time-folds more
potent than EAPB0203 (0.2 µM). Compound 6a showed the lowest cytotoxic activity
(P<0.05). Based on the cytotoxicity data, the structure-activity relationship (SAR) for these
new imidazoquinoxaline derivatives was examined. It is interesting to observe that the
compounds with a methoxy or hydroxy substitution at meta-position (7a, 8 and 9), and the
dimethoxy substitutions at 3,4 or 3,5 and 3,6 positions (6d, 6e, and 6f) on the phenyl ring (R1)
showed similar or slightly reduced activity on A375 cancer cell in comparison to EAPB0503.
Moderate decrease in activity than EAPB0503 was observed for the trimethoxy substituted
compounds (6h, 6i and 6j). The ortho-methoxy substitution (6a) showed an important
decrease in the cytotoxic activity. It is important to note that the 2-hydroxy-3-methoxy
compound (6g) presented the most important antiproliferative activity. These findings
conducted us to highlight that at least one methoxy or hydroxy substitution at meta-position
on the phenyl ring is recommended to maintain the cytotoxic activity of these derivatives.
2.2.2. In vitro effect on tubulin polymerization
To investigate whether the antiproliferative activities of these compounds were related or not
to the interaction with tubulin and in order to more investigate the mechanism of action of our
derivatives, we evaluated their capacity for binding and inhibiting tubulin polymerization.
For this issue, tubulin was purified from pig brain tissue according the protocol described in
the experimental part. Colchicine, a known natural antitubuline inhibitor, was used as
reference. The results on tubulin polymerization are illustrated in Table 1.
Table 1: Imidazoquinoxaline derivatives: general formula, IC50 values against A375 cells and effect
on tubulin polymerization inhibition.
6

% of tubulin
polymerization
inhibition
IC50a (µM) on A375b
Compounds
R1
R4
/DMSO 5 µM c
0.2 ± 0.02
122 ± 25
0.37 ± 0.4
1.25 ± 0.18
0.20 ± 0.0014
2.4 ± 0.6
84.0 ± 3.2
0
99.0 ± 3.4
0
41.0 ± 7.8
11.33 ± 6.3
0
63.0 ± 8.0
98.5 ± 4.0
3.0 ± 2.4
90.0 ± 7.7
0
EAPB0503
3-OCH3-C6H4-
2-OCH3-C6H4-
4-OCH3-C6H4-
3-OH-C6H4-
CH3-NH-
CH3-NH-
CH3-NH-
CH3-NH-
H2N-
6a
6b
8
7a
3-OCH3-C6H4-
3-OH-C6H4-
9
H2N-
5.9 ± 0.3
6c
2,3-di-OCH3-C6H3-
3,4-di-OCH3-C6H3-
3,5-di-OCH3-C6H3-
2,5-di-OCH3-C6H3-
2-OH-3-OCH3-C6H3-
CH3-NH-
CH3-NH-
CH3-NH-
CH3-NH-
CH3-NH-
0.3 ± 0.003
0.22 ± 0.005
1.20 ± 0.044
0.077 ± 0.005
1.50 ± 0.30
1.80 ± 0.38
2.30 ± 0.09
0.010 ± 0.002
6d
6e
6f
6g
6h
2,3,4-tri-OCH3-C6H3- CH3-NH-
3,4,5-tri-OCH3-C6H2- CH3-NH-
0
0
6i
6j
2,4,6-tri-OCH3-C6H2
CH3-NH-
52 ± 5
Colchicine
a
IC50: compound concentration required for A375 growth inhibition by 50%; Values are derived from three
b
c
independent experiments, the standard deviation are also noted; A375 human melanoma cells. Values are
expressed in percentages, derived from three independent experiments and the SD are also calculated.
Compounds 6b, 6e, 6g, and EAPB0503 (at 5 μM) were highly effective in their ability to
inhibit tubulin assembly with percentages of inhibition values of 99%, 98%, 90% and 84%,
respectively. These compounds exhibited significant high potency on tubulin inhibition in
comparison with colchicine which demonstrated a percentage up to 52% of inhibition (P =
0.01, 0.01, 0.04, and 0.02 respectively < 0.05). The high effect on tubulin inhibition is
correlated with their potent antiproliferative effect on human melanoma cell line. This
reported data showed an important role of theses derivatives as antiproliferative and
antimitotic agents. Based on the structural analogy with colchicine, we decided to introduce
methoxy groups, presents on colchicine, on our compounds to improve their antitubulin
inhibition. Surprisingly, compounds (6a, 6h, 6i, and 6j) bearing three methoxy substitutions
did not show any inhibition of tubulin assembly.
2.2.3. EAPB0503, 6b and 6e causes G2/M cell-cycle arrest
7

Cellular DNA contents were determined by Flow cytometer analysis on cell treatment.
EAPB0503, 6b and 6e induced significant dose-dependent changes in the cell-cycle
distribution (Figure 2). Indeed, all 3 compounds induce a dramatic blockage in G2/M phase.
Those results were agreed to those of antitubulinic agents that usually caused a block of cells
in the G2/M phase of the cell cycle. In fact, their ability to inhibit tubulin assembly will avoid
the mitotic spindle formation and so the mitosis.
Figure 2: EABP0503, 6b and 6e induce G2/M cell arrest on A375 cell line. (A) Effects of
EABP0503, 6b and 6e on the cell-cycle distribution on A375 cell line. Treated cells were stained with
PI (25 µg/ml), and the cell-analysis was performed by a Gallios flow cytometer. (B) The G0/G1, S and
G2/M percentage measured in treated cells.
2.3. Molecular Modeling
In order to better precise and in line with SAR the tubulin affinity and selectivity of our
compounds, molecular docking simulations were determined with this series of compounds.
The 3D structures of all chemical compounds were generated using PRODRG interface.31
Molecular docking studies were performed using the GOLD software 32 to examine a possible
8

binding mode of all compounds in the colchicine binding domains of tubulin. Colchicine
binds between the two chains ( and ) of tubulin making a hydrogen bond with Val181 of
chain 33 and a hydrogen bond with Thr353 of chain  connected with a water molecule
(Figure 3a). The tri-methoxyphenyl ring of colchicine is buried in a hydrophobic pocket of
chain  form by the side chain of residues Cys241, Leu242, Leu248, Ala250, Leu255, Ile318,
Ala354 and Ile378 (Figure 3a). The tropolone ring of colchicine is at the entrance of this
pocket and made Van der Walls interactions with the side chain of residues Leu255, Met259,
Ala316 and Lys352. Docking studies of all compounds into the colchicine site of tubulin
showed a good correlation with the inhibition of tubulin polymerization. As an example,
compounds 6a, 6c, 6h, 6i and 6j could not be docked in this colchicine site and demonstrated
absence of tubulin polymerization inhibition. On the contrary, compounds EAPB0503, 6b,
6e, 6g and 7a could be docked in the colchicine site. Compound 6g with the highest activity
on A375 cells showed a good superimposition of the tropolone ring with the phenyl
substituted (Figure 3b-d) by making a hydrogen bond with Val181 of chain  as the
colchicine. Imidazoquinoxaline ring made a hydrogen bond with residue Asn101 from chain 
and Van der Wall interaction with Leu255, Met259, Ala316 and Lys352 from chain . These
finding suggested that compound 6g does not mimic the tri-methoxyphenyl ring of the
colchicine fitting into the hydrophobic pocket of chain . This result can explain the inactivity
of compounds 6h, 6i and 6j bearing three methoxy groups on the phenyl ring due to clash
with Met259.
9

L248
L248
b
a
C241
C241
L242
L242
A354
A250
A250
A354
I318
I318
L255
N101
N101
T353
A316
I378
T353
A316
I378
L255
M259
M259
V181
V181
c
d
Figure 3: Compound 6g in the colchicine binding pocket of tubulin chains  and . a. Structure of
tubulin with colchicine bound PDB 4O2B; b. Hydrogen bonds interaction between compound 6g and
tubulin (view from the top of the ligand); c. Superposition of compound 6g and colchicine (view from
the top). d. Superposition of compound 6g and colchicine (view from the side).; Color coding:
nitrogen atoms blue, oxygen atoms red, carbon atoms pale pink for tubulin chain , carbon atoms
green for tubulin chain , carbon atoms yellow for compound 6g and brown for colchicine; Hydrogen
bonds were showed by black dash-line, the distance was denoted by Å; water molecule is shown as
non-bond sphere.
The results obtained from the docking simulations provided us a general binding mode that
was able to justify the biological activity of the compounds and gave some indication for
further rational improvement of imidazoquinoxaline derivatives.
3. Conclusion
A library of new heterocyclic compounds, imidazo[1,2-a]quinoxaline derivatives were
synthesized under microwave assistance. The total reaction time and the yield of the products
have been optimized comparing to the conventional heating. These new derivatives exhibited
remarkable antiproliferative effects on human melanoma cell lines (A375). Compound 6g
presented the most important antiproliferative activity. SAR studies showed that at least one
methoxy or hydroxy substitution at meta-position on the phenyl ring is mandatory to maintain
10

the cytotoxic activity of these derivatives. Compounds 6b, 6e, 6g, and EAPB0503, showed
potent inhibitions of the tubulin polymerization which are correlated to the antiproliferative
activities. Compound 6g is able to make a hydrogen bond with Val181 of chain  as
colchicine and another hydrogen bond with residue Asn101 from chain  and Van der Wall
interaction with Leu255, Met259, Ala316 and Lys352 from chain . On the basis of these
important properties, these derivatives can be promising leads in the development of
anticancer agents.
4. Experimental
4.1.
Chemistry
All solvents and reagents were obtained from commercial sources and used without further
1
purification unless indicated otherwise. H and 13C NMR spectra were recorded using a
Bruker AC 400 or 300 spectrometer. Chemical shifts are reported in parts per million (ppm)
from the tetramethylsilane resonance in the indicated solvent. Coupling constants are reported
in Hertz (Hz), spectral splitting partners are designed as follow: (s) singlet, (br s) broad
singlet, (d) doublet, (dd) doublet of doublets (t) triplet, (td) triplet of doublets, (m) multiplet.
High Resolution Mass Spectra (HRMS) (ElectroSpray Ionization method (ESI) and Q-Tof
high resolution) were recorded on Synapt G2-S (Waters) by the physical measure department
(IBMM, Montpellier, France). Column chromatography was performed on Aldrich silica gel
60 (230–400 mesh). The microwave organic synthesis was assisted by the Biotage® initiator
(Temperature range: 40 - 300°C; Pressure range: 0 – 30 bar; Power range: 0 - 400 W from
magnetron at 2.45 GHz).
4.1.1. Imidazo[1,2-a]quinoxalin-4-amine (3)
NH4OH of a 33% (w/v) aqueous solution (2.7 mL, 19.7 mmol) was added to a solution of 1
(0.500 g, 2.5 mmol) in acetonitrile (5 mL). The mixture was placed under microwave heating
at 180°C during 20 min. The solvent was removed under reduced pressure, and the residue
was dissolved in CH2Cl2 (20 mL). The organic fraction was successively washed with
saturated NaCl solution (15 mL) and water (15 mL), dried (Na2SO4), and concentrated under
1
reduced pressure. The product was obtained (yield, 94 %) and used without purification. H
NMR (300 MHz, DMSO-d6) : 7.92 (d, 1H), 7.79 (d, 1H), 7.72 (m, 2H), 7.65 (m, 2H), 6.66
11

13
(br s, 2H);
C NMR (300 MHz, DMSO-d6) : 142.49, 131.92, 129.26, 128.88, 127.58,
125.81, 125.26, 124.72, 114.26, 111.75.
4.1.2. 1-bromoimidazo[1,2-a]quinoxalin-4-amine (5)
A solution of 3 (1.25 g, 6.8 mmol) and N-bromosuccinimide (1.3 g, 7.5 mmol) in CHCl3 (200
mL) was heated under reflux for 1h30. The resulting solution was cooled, washed with 5%
sodium hydrogen carbonate (50 mL), dried with Na2SO4 and concentrated under reduced
1
pressure. Yield 80%. H NMR (300 MHz, DMSO-d6) : 7.95 (d, 1H), 7.75 (m, 2H), 7.71 (m,
13
1H), 7.68 (m, 1H), 6.65 (br s, 2H);
C NMR (300 MHz, DMSO-d6) : 142.20, 138.78,
129.66, 128.16, 127.17, 126.95, 126.18, 125.51, 113.98, 99.27. HRMS: m/z calcd for
C11H10N4Br [M]+ 277.0089; Found 277.0088.
General procedure for the Suzuki cross-coupling reaction 21-23
The Suzuki coupling of 4 (300 mg, 1.09 mmol) or 5 with the corresponding aryl boronic acid
in the presence of palladium catalyst Pd(PPh3)4 (60 mg), basic conditions Na2CO3 (250 mg),
DME (10 mL), H2O (5 mL) and under microwave assistance (140°C, 20 min) led to 6c-6j and
7a. These compounds were purified by column chromatography on silica gel, leading to the
pure desired products.
4.1.3. 1-(2,3-dimethoxyphenyl)-N-methylimidazo[1,2-a]quinoxalin-4-amine (6c)
1
2,3-dimethoxyphenyl boronic acid (218 mg, 1.2 mmol). Yellow solid (66%). H NMR (400
MHz, CDCl3) : 7.80 (d, J = 8 Hz, 1H), 7.43 (s, 1H), 7.32-7.10 (m, 4H), 7.02-6.90 (m, 2H),
13
6.28 (br s, 1H), 3.95 (s, 3H), 3.55 (s, 3H), 3.28 (d, 3H). C NMR (400 MHz, CDCl3)  :
153.07, 148.16, 137.72, 133.95, 132.28, 126.91, 126.68, 126.17, 125.23, 124.32, 123.68,
122.68, 115.58, 114.26, 60.72, 55.98, 27.87. HRMS: m/z calcd for C19H19N4O2 [M]+
335.1508; Found 335.1510.
4.1.4. 1-(3,4-dimethoxyphenyl)-N-methylimidazo[1,2-a]quinoxalin-4-amine (6d)
3,4-dimethoxyphenyl boronic acid (218 mg, 1.2 mmol). Off white solid (30%). 1H NMR (400
MHz, CDCl3) : 7.87 (d, J = 8 Hz, 1H), 7.44 (s, 1H), 7.38-7.35 (m, 2H), 7.13-6.97 (m, 4H),
6.50 (br s, 1H), 3.96 (s, 3H), 4.02 (s, 3H), 3.89 (s, 3H), 3.37 (br s, 3H). 13C NMR (400 MHz,
CDCl3) : 149.99, 149.11, 148.28, 138.02, 133.66, 132.06, 130.73, 127.11, 125.82, 123.19,
12

122.81, 122.37, 115.98, 113.39, 111.28, 56.08, 56.03, 27.59. HRMS: m/z calcd for
C19H19N4O2 [M]+ 335.1508; Found 335.1510.
4.1.5.1-(3,5-dimethoxyphenyl)-N-methylimidazo[1,2-a]quinoxalin-4-amine (6e)
1
3,5-dimethoxyphenyl boronic acid (218 mg, 1.2 mmol). Yellow solid (40%). H NMR (400
MHz, CDCl3) : 7.79 (d, J = 8 Hz, 1H), 7.37-7.26 (m, 3H), 6.94 (td, 1H), 6.59-6.56 (m, 3H),
6.43 (br s, 1H), 3.75 (s, 6H), 3.29 (br s, 3H). 13C NMR (400 MHz, CDCl3) : 161.05, 132.24,
126.64, 123.02, 116.33, 108.29, 55.58, 28.11. HRMS: m/z calcd for C19H19N4O2 [M]+
335.1508; Found 335.1505.
4.1.6. 1-(2,5-dimethoxyphenyl)-N-methylimidazo[1,2-a]quinoxalin-4-amine (6f)
1
2,5-dimethoxyphenyl boronic acid (218 mg, 1.2 mmol). White solid (83%). H NMR (400
MHz, CDCl3) : 7,86 (d, J = 4 Hz, 1H), 7,44 (s, 1H), 7,38-7,28 (m, 2H), 7,11-6,96 (m, 4H),
13
6,51 (br s, 1H), 3,85 (s, 3H), 3,58 (s, 3H), 3,37 (br s, 3H). C NMR (400 MHz, CDCl3) :
153.55, 152.34, 147.89, 132.25, 126.36, 122.72, 120.15, 117.63, 116.07, 115.54, 112.00,
55.89, 55.84, 28.08. HRMS: m/z calcd for C19H19N4O2 [M]+ 335.1508; Found 335.1509.
4.1.7. 1-(2-hydroxy-3-methoxyphenyl)-N-methylimidazo[1,2-a]quinoxalin-4-amine (6g)
1
2-hydroxy-3-methoxyphenyl boronic acid (201 mg, 1.2 mmol). White solid (9%). H NMR
(300 MHz, CDCl3)  : 7.80 (br s, 1H), 7.41 (s, 1H), 7.36-7.26 (m, 2H), 7.06-6.93 (m, 4H),
6.35 (br s, 1H), 6.05 (br s, 1H), 3.96 (s, 3H), 3.27 (br s, 3H). 13C NMR (300 MHz, CDCl3)  :
146.32, 143.50, 142.84, 139.57, 131.13, 130.76, 129.38, 129.02, 125.91, 124.90, 124.47,
121.72, 121.48, 117.99, 114.94, 108.52, 55.93, 29.30. HRMS: m/z calcd for C18H17N4O2 [M]+
321.1352; Found 321.1354.
4.1.8. 1-(3-methoxyphenyl)imidazo[1,2-a]quinoxalin-4-amine (7a)
1
5 (320 mg, 1.21 mmol), 3-methoxyphenyl boronic acid (370 mg, 2.4 mmol). Yield 79%. H
NMR (300 MHz, CDCl3) : 7.98 (s, 1H), 7.65 (m, 2H), 7.55 (m, 2H), 7.3 (m, 2H), 6.97 (m,
13
2H), 6.2 (br s , 2H), 3.52 (s, 3H). C NMR (300 MHz, CDCl3) : 160.89, 143.10, 131.06,
130.68, 129.88, 129.73, 128.27, 127.39, 126.20, 125.02, 124.86, 119.97, 118.73, 115.08,
111.87, 108.79, 55.30. HRMS: m/z calcd for C17H15N4O [M]+ 291.1246; Found 291.1251.
4.1.9. N-methyl-1-(2,3,4-trimethoxyphenyl)imidazo[1,2-a]quinoxalin-4-amine (6h)
13

1
2,3,4-trimethoxyphenyl boronic acid (365 mg, 2.17 mmol). Yellow solid (15%). H NMR
(400 MHz, CDCl3) : 7.80 (br s, 1H), 7.35-7.19 (m, 3H), 7.05 (d, J = 8 Hz, 1H), 6.95-6.88 (m,
1H), 6.74 (d, J = 8 Hz, 1H), 3.91 (s, 3H), 3.85 (s, 3H), 3.53 (s, 3H), 3.30 (br s, 3H). 13C NMR
(400 MHz, CDCl3) : 154.61, 151.73, 146.64, 141.37, 131.13, 131.03, 125.31, 121.88,
114.49, 106.14, 60.05, 59.86, 55.12, 28.67. HRMS: m/z calcd for C20H21N4O3 [M]+ 365.1614;
Found 365.1615.
4.1.10 N-methyl-1-(3,4,5-trimethoxyphenyl)imidazo[1,2-a]quinoxalin-4-amine (6i)
1
3,4,5-trimethoxyphenyl boronic acid (365 mg, 2.17 mmol). Yellow solid (28%). H NMR
(400 MHz, CDCl3) : 7.76 (d, J = 8 Hz, 1H), 7.36 (s, 1H), 7.31-7.26 (m, 2H), 6.93 (td, 1H),
6.66 (s, 2H), 6.38 (br s, 1H), 3.90 (s, 3H), 3.78 (s, 6H), 3.26 (br s, 3H). 13C NMR (400 MHz,
CDCl3) : 153.54, 147.92, 138.99, 132.10, 128.34, 125.47, 122.73, 119.10, 116.15, 107.56,
61.12, 56.32, 27.97. HRMS: m/z calcd for C20H21N4O3 [M]+ 365.1614; Found 365.1616.
4.1.11. N-methyl-1-(2,4,6-trimethoxyphenyl)imidazo[1,2-a]quinoxalin-4-amine (6j)
1
2,4,6-trimethoxyphenyl boronic acid (365 mg, 2.17 mmol). Yellow solid (20%). H NMR
(400 MHz, CDCl3) : 7.70 (d, J = 8 Hz, 1H), 7.30-7.21 (m, 3H), 6.88 (td, 1H), 6.41 (br s, 1H),
13
6.18 (s, 2H), 3.86 (s, 3H), 3.56 (s, 6H), 3.22 (br s, 3H). C NMR (400 MHz, CDCl3) :
163.06, 160.18, 148.20, 132.16, 132.06, 126.41, 125.90, 122.62115.04, 114.45, 100.45, 90.78,
55.79, 55.53, 27.77. HRMS: m/z calcd for C20H21N4O3 [M]+ 365.1614; Found 365.1613.
General procedure for the demethylation reaction
To EAPB0503 and 7a under nitrogen atmosphere in anhydrous dimethylformamide,
iodocyclohexane was added to the mixture and the reactions were heated to reflux for 16 h.
Iced water (20 mL) was added to the residues and the products were extracted with CH2Cl2 (3
x 30 mL). The combined organic extracts were washed with water (40 mL), dried (Na2SO4)
and concentrated to dryness under reduced pressure. The crude products were purified by
column chromatography on silica gel using C6H12-EtOAc (50:50 v/v) as eluent, leading to
products, which were then taken into chloroform at 0°C, filtrated and washed with iced
chloroform, yielding to the pure compounds 8 and 9.
4.1.12. 3-(4-methylamino)imidazo[1,2-a]quinoxalin-1-yl)phenol (8)
14

EAPB0503 (200 mg, 0.64 mmol), dimethylformamide (2 mL), iodocyclohexane (3.3 mmol,
0.424 mL). Yield 22% 1H NMR (300 MHz, MeOD)  : 9.84 (br s, 2H), 7.78 (dd, 1H), 7.62
(dd, 1H), 7.4 (m, 2H), 7.35 (m, 1H), 7.29 (m, 2H), 6.85 (m, 2H), 3.25 (s, 3H), 13C NMR (300
MHz, CDCl3)  : 160.07, 142.83, 139.57, 131.46, 131.13, 130.35, 129.59, 129.02, 126.92,
124.90, 124.47, 122.10, 117.66, 114.96, 114.43, 112.32, 29.32. HRMS: m/z calcd for
C17H15N4O [M]+ 291.1246; Found 291.1245.
4.1.13. 3-(4-amino)imidazo[1,2-a]quinoxalin-1-yl)phenol (9)
7a (200 mg, 0.689 mmol), dimethylformamide (5 mL), iodocyclohexane (8.26 mmol, 1.00
mL). Yield 17%. 1H NMR (300 MHz, MeOD)  : 9.88 (br s, 3H), 7.55 (m, 2H), 7.45 (m, 2H),
13
7.32 (m, 2H), 7.20 (s, 1H), 6.98 (m, 2H). C NMR (300 MHz, CDCl3)  : 160.07, 143.10,
131.58, 131.06, 130.35, 129.35, 129.97, 129.97, 129.73, 126.78, 126.20, 125.02, 124.85,
119.97, 118.37, 115.08, 114.43, 113.03. HRMS: m/z calcd for C16H13N4O [M]+ 277.1089;
Found 277.1093.
4.2. Biological evaluation
4.2.1. Materials and reagents.
Compounds and reactants used in this study were bought from Sigma-Aldrich (Saint-Qentin
Fallavier, France). 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT),
isopropyl alcohol and chlorhydric acid were obtained from Merck (Darmstadt, Germany). All
other reagents were of analytical grade and were obtained from commercial sources.
Statistical analysis for the values was performed using t-test and P values are calculated.
Probably values of less than 0.05 were considered statistically significant.
4.2.2. Cell lines and culture techniques.
Melanoma human cancer cell line (A375) was obtained from American Type Culture
Collection (Rockville, Md., USA). Cells were cultured in RPMI medium containing RPMI-
1640 (LONZA, Basel, Switzerland), supplemented with 10% heat-inactived (56°C) foetal
bovine serum (FBS) (Gibco, Waltham, MA, USA), 2 mM L-glutamine, 100 IU/ml penicillin
G sodium, 100 µg/ml streptomycin sulphate and 0.25 µg/ml amphotericin B. Cells were
maintained in a humidified atmosphere of 5% CO2 in air at 37 °C.
15

4.2.3. Cytotoxicity Assay.
Previously to the experiments, the number of cells by well, the doubling time and the MTT
concentration have been optimized. In all the experiments, A375 were seeded at a final
concentration of 5000 cells/well in 96-well microtiter plates and allowed to attach overnight.
After 20-24 hours incubation, the medium was aspirated carefully from the plates using a
sterile Pasteur pipette, and cells were exposed i) to vehicle controls (1% DMSO/culture
medium and culture medium alone), ii) to EAPB0503, of 10-4 to 10-10 µM, diluted in the
culture medium, and iii) to the synthesized compounds (10-4-10-10 µM) dissolved in a mixture
1% DMSO/culture medium (v/v). After 96 h of incubation, 10 µl of MTT solution in PBS (5
mg/ml, phosphate-buffer saline pH 7.3) were added to each well and the wells were incubated
at 37 °C for 4 h. This colorimetric assay is based on the ability of live and metabolically
unimpaired tumor-cell targets to reduce MTT to a blue formazan product. At the end of the
incubation period, the supernatant was carefully aspirated, then, 100 µl of a mixture of
isopropyl alcohol and 1 M hydrochloric acid (96/4, v/v) were added to each well. After 10
min of incubation and vigorous shaking to solubilize formazan crystals, the optical density
was measured at 570 nm in a microculture plate reader (Dynatech MR 5000, France). For
each assay, at least three experiments were performed in triplicate. The individual cell line
growth curves confirmed all tumor lines in control medium remained in the log phase of cell
growth 96 h after plating. Cell survival was expressed as percent of vehicle control. The IC50
values defined as the concentrations of drugs which produced 50% cell growth inhibition;
50% reduction of absorbance, were estimated from the sigmoidal dose-response curves.
4.2.4. In vitro tubulin polymerization analysis
Tubulin was prepared from pig brain according to the purification procedure described by
Williams and Lee35. To evaluate the effect of the compounds on tubulin assembly in vitro,
tubulin polymerization was monitored turbidimetrically at 350 nm with a MC2
spectrophotometer (Safas, Monaco) equipped with a thermal-jacketed cuvette holder. The
reaction mixture was prepared at 0 °C, and contained PEM (Pipes 0.1M, EGTA 2mM, MgSO4
1mM pH 6.9) buffer, 25% glycerol (v/v), 1 mM GTP, MgSO4 5mM, and 12 µM tubulin. GTP
and tubulin were added at the very last minute. EAPB0503, its derivatives and colchicine
stock solutions were diluted in DMSO to the desired concentration, and 2 µL of the
compound solution was added to the medium (final concentration: 5 μM). The same volume
of DMSO alone was used for negative control. The final volume of the sample was 200 µL.
The reaction was started by placing the cuvette in the spectrophotometer cell compartment
16

thermostated at 37 °C. Ice was added 45 minutes later to initiate depolymerization to check
for signal specificity.
4.2.5. Flow cytometric analysis
The cell cycle distribution was monitored by flow cytometric analysis as previously described
by Brons et al36 . A375 cells were harvested after treatment with EABP0503, 6e or 6b for
24hh, washed with PBS and were spun down (400g, 5 min at 4°C), the supernatant was
discarded, and the pellet placed on ice for 10 min. Subsequently 900 µl of an ice-cold
hypotonic PI solution (221, containing 25 µg/ml Propidium iodure (Sigma), 0.1% w/v tri-
sodium citrate dihydrate (Sigma, St louis), 10% v/v RNA-se A solution (Sigma, St. Louis)
and 0.1% v/v Triton X-100 in distilled water, was added. Cells were kept overnight on ice and
analyzed for DNA content. The stained cells were performed by a Gallios 10
photomultiplicators (Beckman coulter) in Montpellier Rio Imaging facilities (MRI platform,
Montpellier, France). The cell cycle distribution was analyzed using KALUZA flow analysis
software (Beckman coulter). For each experiment, 20 000 events per sample were recorded.
4.2.6. Molecular docking analysis
The molecular modeling studies were performed with GOLD.32 The crystal structure of
tubulin complexed with colchicine (PDB 4O2B) 37 was retrieved from the RCSB Protein Data
Bank. The binding sphere with a radius of 25.0Å was defined with residue Leu255 as the
31
binding site. All compounds were drawn with PRODRG and protonated using BABEL.
32
Finally, they were docked into the binding site using the GOLD protocol with the default
settings. The pictures were generated by the program PYMOL.38
Acknowledgements
This work was supported by the French Infrastructure for Integrated
Structural Biology (FRISBI) ANR-10-INSB-05-01. Zahraa Zghaib thanks Islamic Association
for Guidance and Higher Education for its financial support. Flow cytometry was performed
at MRI-Cytométrie-IRB, Institute For Regenerative Medicine and Biotherapy, Hôpital St Eloi,
CHU de Montpellier.
17

References
1. Yaqiong, M.; Senbiao, F.; Li, L.H.; Han, C.; Lu, Y.; Zhao, Y.; Liu, Y.; Zhao, C. Chem.
Biol. Drug. Des. 2013, 82, 12.
2. Janke, C.; Kneusse, M. Neurosci. 2010, 33, 362.
3. Kamal, A.; Reddy, C.R.; Vishnuvardhan, M.V.P.S.; Mahesh, R.; Nayak,V.L.; Prabhakar,
S.; Reddy, C.S. Bioorg. Med. Chem. 2014, 24, 2309.
4. Jordan, A.; Hadfield, J.A.; Lawrence, N.J.; McGown, A.T.. Med. Res. Rev. 1998, 18, 259.
5. Giannakakou, P.; Sackett, D.; Fojo, T.. J. Natl. Cancer Inst. 2000, 92, 182.
6. Mishra, R.C.; Karna, P.; Gundala, S.R.; Pannu, V.; Stanton, R.A.; Gupta, K.K.; Robinson,
M.H.; Lopus, M.; Wilson, L. Biochem. Pharmacol. 2011, 82, 110.
7. Jordan, M.A.; Wilson, L.. Nat. Rev. Cancer 2004, 4, 253.
8. Mollinedo, F.; Gajate, C. Apoptosis 2003, 8, 413.
9. Zhou, J.; Giannakakou, P. Curr. Med. Chem. Anticancer Agents 2005, 5, 65.
10. Pasquier, E.; Honore, S.; Braguer.D. Drug Resist. Updat. 2006, 9, 74.
11. Zheng, C.H.; Chen, J.; Liu, J.; Zhou, X.T.; Liu, N.; Shi, D.; Huang, J.J.; Lv, J.G.; Zhu, J.;
Zhou, Y.J. Arch. Pharm. Chem. Life Sci. 2012, 345, 454.
12. Ravelli, R.B.; Gigant, B. ; Curmi, P.A.; Jourdain, I.; Lachkar, S.; Sobel, A.; Knossow, M.
Nat. 2004, 428, 198.
13. Risinger, A.L. Giles, F.J.; Mooberry, S.L. Cancer Treat. Rev. 2009, 35, 255.
14. Brossi, A.; Yeh, H.J.; Chrzanowska, M.; Wolff, J.; Hamel, E.; Lin, C.M.; Quin, F.;
Suffness, M.; Silverton, J. Med. Res. Rev. 1988, 8, 77.
15. Bhattacharyya, B.; Panda, D.; Gupta, S.; Banerjee, M. Med. Res. Rev. 2008, 28, 155.
16. Paños, J.; Díaz-Oltra, S.; Sánchez-Peris, M.; García-Pla, J.; Murga, J.; Falomir, E.; Carda,
M.; Redondo-Horcajo, M.; Díaz, J.F.; Barasoain, I.; Marco, J.A. Org. Biomol. Chem. 2013,
11, 5809.
2016, 24(3), 444.
25. Jordan, J.A.; Hadfield, J.A.; Lawrence, N.J.; McGown, A.T. Med. Res. Rev. 1998, 18,
259.
26. Moarbess, G.; Deleuze-Masquefa, C. Bonnard, V. Gayraud-Paniagua, S.; Vidal, J.R.;
Bressolle, F.; Pinguet, F.; Bonnet, P.A. Biorg. Med. Chem. 2008, 16, 6601.
27. Masquefa, C.; Moarbess, G.; Bonnet, P.A.; Pinguet, F.; Bazarbachi, A.; Bressolle , F. .
U.S. Patent 8,378,098, 2013.
28. Deleuze-Masquefa, C.; Moarbess, G.; Khier, S.; David, N.; Gayraud-Paniagua, S.;
Bressolle, F.; Pinguet, F.; Bonnet, P.A. Eur. J. Med. Chem. 2009, 44, 3406.
29. Moarbess, G.; El-Hajj, H.; Kfoury, Y.; El-Sabban, M.E.; Lepelletier, Y.; Hermine, O.;
Deleuze-Masquéfa, C.; Bonnet, P.A.; Bazarbachi, A. Blood 2008, 111, 3770.
30. Saliba, J.; Deleuze-Masquéfa, C.; Iskandarani, A.; El Eit, R.; Hmadi, R.;Mahon, F.X.;
Bazarbachi, A.; Bonnet, P.A.; Nasr, R. Anti-Cancer Drugs 2014, 25, 624.
18

31. http://www.davapc1.bioch.dundee.ac.uk/cgi-bin/prodrg
32. PLANTS: Korb, O.; Stützle, T.; Exner, T.E. Lect. Notes Comput. Sci. 2006, 4150, 247.
33. La Regina, G.; Bai, R.; Coluccia, A.; Famiglini, V.; Pelliccia, S.; Passacantilli, S.;
Mazzoccoli, C. ; Ruggieri, V.; Verrico, A.; Miele, A.; Monti, L.; Nalli, M.; Alfonsi, R.; Di
Marcotullio, L.; Gulino, A.; Ricci, B.; Soriani, A.; Santoni, A.; Caraglia, M.; Porto, S.; Da
Pozzo, E.; Martini, C.; Brancale, A.; Marinelli, L.; Novellino, E.; Vultaggio, S.; Varasi, M.;
Mercurio, C.; Bigogno, C.; Dondio, G.; Hamel, E.; Lavia, P.; Silvestri, R. J. Med. Chem.
2015, 58, 5789.
34. Hamel, E.; Lin, C.M. Biochem. 1984, 23 4173.
35. Williams, R.J.; Lee, J. Methods Enzymol., 1982, 85B, 376.
36. Brons, P P T.; Pennings, A H M.; Haanen, C.; Wessels, H M C.; Boezeman, J B M.
Cytometry. 1990, 11 : 837-844.
37. Prota, A.E.; Danel, F. ; Bachmann, F.; Bargsten, K.; Buey, R.M.; Pohlmann, J.; Reinelt,
S.; Lane, H.; Steinmetz, M.O. J. Mol. Biol. 2014, 426, 1848.
38. http://pymol.sourceforge.net
19

List of captions
Figure 1: Chemical structure of lead compounds.
Scheme 1: Synthesis of EAPB0503 derivatives. a) EtOH, NH2CH3 in ethanol or NH4OH in
water, MW (180°C, 20 min); b) NBS, CHCl3, reflux 1h30; c) Suzuki-cross coupling (R1’–
B(OH)2), Pd(PPh3)4, Na2CO3, DME, MW (140°C, 20 min); d) Iodocyclohexane, DMF, reflux
16h.
Table 1: Imidazoquinoxalines derivatives: General formula, IC50 values against A375 cells
and effect on tubulin polymerization inhibition.
Figure 2: EABP0503, 6b and 6e induce G2/M cell arrest on A375 cell line. (A) Effects of
EABP0503, 6b and 6e on the cell-cycle distribution on A375 cell line. Treated cells were
stained with PI (25 µg/ml), and the cell-analysis was performed by a Gallios flow cytometer.
(B) The G0/G1, S and G2/M percentage measured in treated cells.
Figure 3: Compound 6g in the colchicine binding pocket of tubulin chains  and . a.
Structure of tubulin with colchicine bound PDB 4O2B; b. Hydrogen bonds interaction
between compound 6g and tubulin (view from the top of the ligand); c Superposition of
compound 6g and colchicine (view from the top). d. Superposition of compound 6g and
colchicine (view from the side).; Color coding: nitrogen atoms blue, oxygen atoms red,
carbon atoms pale pink for tubulin chain , carbon atoms green for tubulin chain , carbon
atoms yellow for compound 6g and brown for colchicine; Hydrogen bonds were showed by
black dash-line, the distance was denoted by Å; water molecule is shown as non-bond sphere.
20

Graphical abstract
6g (R1: 2-OH-3-OCH3-C6H3-, R4: CH3-NH-) Superposition of 6g and colchicine
IC50 A375 = 0.077 µM
Tubulin inhibition: 90%
21