Synthesis and biological evaluation of 4-arylcoumarin analogues of combretastatins. Part 2

Article abstract of DOI:10.1021/jm901826e

Figure Presented. A series of A-ring variously methoxylated 4-(3-hydroxy-4-methoxyphenyl)coumarins related to combretastatin A-4 was prepared by cross-coupling reactions. Cytotoxicity studies indicated a potent activity against HBL100 cell line. Substitut

Full text of DOI:10.1021/jm901826e

ARTICLE
pubs.acs.org/jmc
Synthesis and Biological Evaluation of 4-Arylcoumarin Analogues of
Combretastatins. Part 2
Sꢀebastien Combes,*,‡ Pascale Barbier,§ Soazig Douillard,§ Anne McLeer-Florin,|| Vꢀeronique Bourgarel-Rey,§
Jean-Thomas Pierson,‡ Alexey Yu. Fedorov,^ Jean-Pierre Finet,‡ Jean Boutonnat,|| and Vincent Peyrot*,§
‡Laboratoire Chimie Provence, UMR-CNRS 6264, Universitꢀe d’Aix-Marseille, Facultꢀe des Sciences Saint-Jꢀer^ome, case 521, 13397
Marseille Cedex 20, France
§Centre de Recherche en Oncologie Biologique et en Oncopharmacologie, CRO2 INSERM UMR 911, Universitꢀe d’Aix-Marseille,
Facultꢀe de Pharmacie, 27 Boulevard Jean Moulin, Marseille 13005, France
Dꢀepartement de Biologie et Pathologie de la Cellule et Dꢀepartement d’Anatomie et Cytologie Pathologique, UMR-CNRS 5525,
CHU Grenoble, H^opital Michallon, UJF, Grenoble, F-38000 France
^Department of Chemistry, Nizhny Novgorod State University, 23 Gagarin Avenue, 603950 Nizhny Novgorod, Russia
ABSTRACT: A series of A-ring variously methoxylated 4-(3-
hydroxy-4-methoxyphenyl)coumarins related to combretasta-
tin A-4 was prepared by cross-coupling reactions. Cytotoxicity
studies indicated a potent activity against HBL100 cell line.
Substitution patterns on A-ring had only a slight effect on
antiproliferative activity. For most cytotoxic compounds, the
activity as potential modulators of P-gp and BCRP efflux pumps
was evaluated. The results show that compounds 2 and 7 were able to restore mitoxantrone accumulation (BCRP) at concentrations
similar to that of cyclosporine A. Compound 7 was the most efficient to reverse P-gp activity. All compounds were found to potently
inhibit in vitro microtubule formation via a substoichiometric mode of action for the most part. Compounds 1 and 2 were found to
have an apparent affinity binding constant similar to that of combretastatin A-4, i.e., 1 ꢁ 10 6 Mꢀ1. The molecular modeling of
coumarin derivatives was performed on the basis of the molecular structure of 7, as determined by single-crystal X-ray
crystallography. The calculations suggested that the presence of a methoxy group out of the plane of the chromenone moiety is
an important steric hindrance factor embedding the accessibility of those molecules inside the binding pocket on tubulin.
’ INTRODUCTION
and combretastatin AVE 8062A.3ꢀ6 It is too early to predict if any
of these combretastatin derivatives will enter clinical use, and
research to develop new analogues with a more favorable ther-
apeutic window is ongoing. In the authors’ laboratories, the
synthesis of different 4-arylcoumarin analogues of combretastatin
A-4 led to the identification of two new compounds 7 and 8
(Chart 1) with potent cytotoxic activity on a CEM leukemia cell
line.7 It was suggested that the cytotoxicity of compounds 7 and 8
may be related to their interaction with microtubules and tubulin
because these compounds, previously synthesized as potential
topoisomerase inhibitors, inhibit microtubule formation from
purified tubulin.
These compounds disturbed cell survival by depolymerizing the
microtubule network, leading to a mitotic block.8 By isothermal
titration calorimetry (ITC), it was demonstrated using a displace-
ment method that combretastatin A-4 and 4-arylcoumarins share
their binding domain in tubulin. A superimposition of the com-
bretastatin A-4 and these analogues on the tubulin colchicine
binding domain indicated that the hydroxyl group present on the
B-ring of compounds 7, 8, and combretastatin A-4 could form a
The microtubule network is an essential component of the
cytoskeleton of eukaryotic cells. The discovery of the binding of
colchicine to tubulin was a fundamental step in the development
of antimitotic drugs (Chart 1). The development of new anti-
mitotic drugs then led to new cancer chemotherapy approaches
and to a better knowledge of microtubule pharmacology and
biochemistry. Antimitotic drugs that bind tubulin or antitubulin
agents are classified as microtubule-stabilizing agents or micro-
tubule-destabilizing agents. The taxoids belonging to the first
class and the vinca alkaloïds belonging to the second one are
commonly used in cancer therapy.1 Considering the success of
the antitubulin pharmacological class, numerous studies are now
focused on the discovery and clinical trial development of new
derivatives. Among them, combretastatin A-4 (CA-4), isolated
from the bark of the Combretum Caffrum tree, has been shown to
be a potent tubulin assembly inhibitor and vascular disrupting
agent at low concentrations.2
At the present time, due to the relative structural simplicity
of combretastatin A-4, hundreds of combretastatin derivatives
have been synthesized and studied. Among them, several candi-
dates are in advanced clinical trials, such as combretastatin A-4
phosphate (CA-4P), combretastatin A1 diphosphate (OXi 4503),
Received: December 10, 2009
Published: April 13, 2011
r
2011 American Chemical Society
3153
dx.doi.org/10.1021/jm901826e J. Med. Chem. 2011, 54, 3153–3162
|

Journal of Medicinal Chemistry
ARTICLE
Chart 1. Structural Analogy of Compounds That Inhibit Tubulin Assembly
Scheme 1. Synthesis of 3-Hydroxy-4-methoxyphenylboronic
Acid 19a
Table 1. Structures and Yields of the 4-Arylcoumarins 1ꢀ8 of
the Suzuki Coupling Reaction
substituent
product
R1
R2
R3
yield (%)
81
unsubstituted
1
H
H
H
monosubstituted
a (a) n-BuLi, THF, ꢀ78 ꢀC; (b) (i-PrO)3B, ꢀ78 ꢀC; (c) HClaq; (d) H2,
2
OMe
H
H
H
65
97
98
Pd/C, THF.
3
OMe
H
H
4
H
OMe
Scheme 2. Synthesis of a Series of 4-Arylcoumarin Analogues
disubstituted
of Combretastatin A-4, 1ꢀ8a
5
OMe
H
OMe
OMe
H
H
80
74
83
6
OMe
OMe
7
OMe
trisubstituted
8
OMe
OMe
OMe
94
polymethoxylated neoflavones with potential cytotoxic activity
were prepared in high yields under Suzuki reaction conditions
using substituted coumarins and arylboronic acids.7,11ꢀ16 Intro-
duction of hydroxylated phenyl ring using protected arylboronic
acids in the palladium-copper-catalyzed Suzuki coupling reaction
requires a subsequent deprotection step to obtain the expected
hydroxylated 4-arylcoumarins. As a selective deprotection is not
reliable enough due to the presence of sensitive groups like
benzopyrone, wefocusedour effortsonthe possibility of preparing
and using an unprotected arylboronic acid in the coupling reaction
key step. 3-Benzyloxy-4-methoxyphenylboronic acid 18 was pre-
pared by reported procedures involving the reaction of the
appropriate aryllithium with triisopropylborate (Scheme 1).15 Se-
lective removal of the benzyl group was successfully performed by
palladium-catalyzed hydrogenolysis, leading to the expected hydro-
xylated phenylboronic acid 19 in 87% yield.
a (a) (CF3SO2)2O, Et3N, CH2Cl2; (b) ArB(OH)2 19, Pd(PPh3)4, CuI,
Na2CO3, toluene-EtOH.
hydrogen bond with the carboxyl group of Thr179 of R-tubulin.8
Moreover, compound 7 presents a methoxy group in front of the
β-tubulin Cys241, which has largely been described to be involved
in the interaction of colchicine analogues.9,10 Curiously, compound 7
is more active than compound 8 that has a trimethoxyphenyl A-ring.
Thus, methoxy groups on A-ring seem to be important for the
activity of 4-arylcoumarin compounds. The present report
focuses on syntheses of a new series of analogues bearing the
3-hydroxy-4-methoxyphenyl moiety supposed to be crucial for
activity and variously substituted on A-ring by methoxy groups
(Scheme 2). The cytotoxic activities, the binding to tubulin, and
microtubule inhibitory effects were reported for unsubstituted
and mono-, di-, and trimethoxylated derivatives. It was important
to state the P-glycoprotein (P-gp) and breast cancer resistance
protein (BCRP) expression of cancer cell lines because some
compounds need high concentrations to inhibit these pumps in
the case of an association with anthracyclines. Finally, molecular
calculations and modeling of the various compounds instructed
the molecular mechanism of interaction.
The activated 4-trifluorosulfonyloxycoumarins 10ꢀ17 were
easily prepared in high yield by treatment of the appropriate
4-hydroxycoumarins with triflic anhydride as previously de-
scribed.7 Use of 3-hydroxy-4-methoxyphenylboronic acid 19
in coupling reaction led to the 4-(3-hydroxy-4-methoxyphe-
nyl)coumarin derivatives 1ꢀ8 (Scheme 2) in good to high yield
(65ꢀ98%, Table 1).
Biology. Cytotoxicity. Cytotoxicity of the synthesized com-
pounds 1ꢀ8 was investigated in HBL100 human epithelial
mammary cell line. A tetrazolium-based assay was used for the
determination of the drug concentration required to inhibit cell
growth by 50% after incubation in the culture medium for 72 h.
The calculated IC50 values are collated in Table 2.
’ RESULTS AND DISCUSSION
Chemistry. Several methods for preparing polysubstituted
4-phenylcoumarins involve direct arylation of activated couma-
rins using organometallic reagents. During the past five years,
3154
dx.doi.org/10.1021/jm901826e |J. Med. Chem. 2011, 54, 3153–3162

Journal of Medicinal Chemistry
ARTICLE
Table 2. Cytotoxicity of CA-4 and 4-Arylcoumarins 1ꢀ8 towards HBL100 Cell Line
compd
CA-4
1
2
3
4
5
6
7
8
IC50 (nM)a
3 ( 0.3
84 ( 1
182 ( 10
39 ( 2
606 ( 22
439 ( 27
714 ( 29
88 ( 1
433 ( 29
a Drug concentration that inhibits the growth of HBL100 by 50% after incubation in liquid medium for 72 h. Data are the mean ( standard deviation
(SD) of three experiments.
Figure 1. BCRP-overexpressing (R, resistant) HCT116 cells (left) and P-gp-overexpressing (R, resistant) K562 cells (right) were treated with variable
doses of the four molecules 1 (—), 2 (---), 3 (ꢀ ꢀ), and 7(ꢀ - ꢀ) (0ꢀ150 μM) and analyzed as described in the Experimental Section. Drug
accumulation results are expressed in mean fluorescence intensity in arbitrary units (AU). Each drug concentration was tested in triplicate. Standard
error (SE) of each point is <10%.
Table 3. Modulation of P-gp and BCRP ABC Efflux Pumps
All compounds were found to exhibit potent cytotoxic activ-
ities in the nanomolar range. In particular, the IC50 value rate at
84, 182, 39, and 88 nM with 1, 2, 3, and 7, respectively, reflecting
a highly pronounced antiproliferative activity. These results are in
accordance with our previous reports dealing with the signifi-
cance of 30-OH and 40-OMe groups on the 4-phenyl B-ring.
Presence of the alkoxybenzene moiety is crucial to obtain
relevant cytotoxic responses. In contrast, the phenyl A-ring can
suffer several modifications and trimethoxy moiety is not as
important as originally postulated. It appears that cytotoxicity
might benefit from a less substituted A-ring. Furthermore an
oxygenated group is not essential as unsubstituted compound 1 is
one of the most cytotoxic compounds.
Interactions with P-gp and BCRP Pumps. On the basis of their
cytotoxic effects, the most active compounds 1ꢀ3 and 7 were
further studied in order to evaluate their activity as potential
modulators of P-gp and BCRP ABC (ATP binding cassette)
efflux pumps, which are involved in chemoresistance in various
cancers.17,18 We also tested if these compounds were substrates
for P-gp and BCRP.
1. Evaluation of the Compounds on P-gp and BCRP
Activity. In K562R and HCT116R cell lines, the four compounds
restored the intracellular accumulation of daunorubicine (P-gp
substrate) or mitoxantrone (BCRP substrate) with a concentra-
tion equivalent or higher to that of the BCRP and P-gp inhibitor
cyclosporine A (Figure 1).
These results outlined in Table 3 show that compounds 2 and
7 were able to restore mitoxantrone accumulation (BCRP
evaluation) at concentrations similar to that of cyclosporine A.
Compound 7 was also the most efficient to reverse P-gp activity
but at a concentration 3-fold higher than that of cyclosporine A.
Compounds 1 and 3 displayed lower modulation activities on
P-gp and BCRP compared to those of compound 7. However,
even if a modulation of P-gp and BCRP could be reached, the
compounds concentrations needed were high. Therefore, in light
of these results, only compound 7 is of interest in a clinical
compd
1
2
3
7
reversion (μM)a
P-gp
80
20
45
10
40
40
30
10
BCRP
a Compound concentration inducing the same reversion as the BCRP
and P-gp inhibitor cyclosporine A (10 μM).
context of ABC pumps modulation as this molecule displayed the
lowest concentration needed to reverse P-gp and BCRP.
2. Evaluation of the Compounds as Potential Substrates
of P-gp and/or BCRP. In the above experiments, when two mol-
ecules (mitoxantrone and one of the studied compounds) were
administered together, a competitive mechanism for binding on
P-gp or BCRP could occur. In the presence of one compound,
the kinetics of binding with P-gp or BCRP could be quite
different. This hypothesis was tested by treating HCT116S and
R and K562S and R cells with 0.1 μM or 10 μM of compound
1ꢀ3 and 7 (Table 4) for 24 h, followed by cell cycle analysis. The
effect of the four compounds on cell viability was also assessed by
MTT after 48 h of treatment. A comparison of the IC50s in the
parental and resistant cell lines provided further evidence of the
ability of these compounds to modulate the sensitivity of the cell
lines in the presence or absence of P-gp or BCRP (Table 5).
At a concentration of 0.1 μM, compounds 1 and 3 were not
able to induce a marked blockade in G2/M compared to the
controls in K562R cells (bold data in Table 4). However, at
10 μM, the effects were the same in sensitive and in resistant cells.
The IC50s of the different compounds on the K562 cell line were
quite similar (Table 5). The same kind of phenomenon was
observed for compounds 1, 2, and 7; at 0.1 μM there was no
increase in G2/M compared to the control cells in HCT116R cell
line. IC50s were similar between HCT116S and R cells for
compounds 1, 2, and 3, whereas compound 7 exhibited a higher
3155
dx.doi.org/10.1021/jm901826e |J. Med. Chem. 2011, 54, 3153–3162

Journal of Medicinal Chemistry
ARTICLE
Table 4. Percentage of the Cells Blocked in G2/M by 4-Arylcoumarins 1ꢀ3 and 7a
cells G2/M (%)
1 (μM)
2 (μM)
3 (μM)
7 (μM)
cell lines
control
0.1
10
0.1
10
0.1
10
0.1
10
K562S
23 ( 2
18 ( 2
32 ( 3
30 ( 4
36 ( 5
24 ( 3
44 ( 2
33 ( 3
43 ( 2
40 ( 3
48 ( 2
49 ( 4
40 ( 3
46 ( 1
44 ( 3
27 ( 1
47 ( 3
42 ( 4
60 ( 5
50 ( 6
38 ( 5
26 ( 6
46 ( 4
39 ( 3
40 ( 4
42 ( 3
56 ( 4
47 ( 4
35 ( 2
35 ( 3
44 ( 2
33 ( 3
43 ( 3
39 ( 3
50 ( 2
43 ( 4
K562R
HCT116S
HCT116R
a Percentage of cells blocked in the G2/M phase of the cell cycle after 24 h incubation of the chemosensitive or chemoresistant cell lines with compounds
1, 2, 3, and 7 at 0.1 mM or 10 mM. Data are the mean ( SD of three experiments. In bold, conditions for which no differences were observed between
control and treated cells.
Table 5. Cytotoxicity of 4-Arylcoumarins 1ꢀ3 and 7 towards
IC50 in the resistant form. This latter compound could thus be a
weak substrate of BCRP.
K562 and HCT116 Cell Lines
The differences observed between cell cycle data and the IC50s
could be due to the fact that (i) IC50s were evaluated after 48 h of
treatment because there was only a moderate apoptosis after
24 h, (ii) the target of these derivatives could not only be tubulin
but also signaling pathways as it was demonstrated for combre-
tastatin A in the literature, (iii) the concentrations needed to
block tubulin polymerization could be higher than those able to
modify cell signaling and induce cell death. These results
demonstrate that the four compounds are either not substrates
(1, 2, and 3) or poor substrates (7) for P-gp or BCRP. As a
blockade in the G2/M phase of the cell cycle was observed at
various concentrations for all the cell lines tested in this study,
irrespective of their tissue of origin (HBL100 is a mammary cell
line, K562 is of hematopoietic origin and HCT116 is a colon cell
line), this pointed toward a major common mechanism related to
the cell cycle. The effects on tubulin assembly by the various
compounds 1ꢀ8 were then investigated.
IC50 (nM)a
cell line
1
2
3
7
K562S
96 ( 10
92 ( 4
86 ( 7
86 ( 11
190 ( 13
200 ( 12
88 ( 8
78 ( 12
150 ( 9
140 ( 9
100 ( 10
88 ( 9
K562R
HCT116S
100 ( 11
120 ( 10
210 ( 8
310 ( 7
HCT116R
a Drug concentration that inhibits the cell growth by 50% after incuba-
tion in liquid medium for 48 h. Data are the mean ( SD of three
experiments.
lower activity, close to a stoichiometric mode of inhibition. For
the disubstituted compounds 5 and 6, it was necessary to add an
excess of ligand to inhibit microtubule formation. In contrast, the
disubstituted compound 7 was found to be a substoichiometric
inhibitor of microtubule formation. Finally, the trisubstituted
compound 8 had a weaker activity against microtubule formation.
Interaction of Compounds with Tubulin: Binding Parameters.
The uncorrected intrinsic fluorescence emission spectra of tubulin
were examined by exciting the tryptophan residues at 295 nm. A
decrease in fluorescence emission intensity at 340 nm was observed
with increasing concentrations of compound 1 (Figure 3). This
quenching of fluorescence was used to perform binding titration
experiments. Quenching of tubulin fluorescence was observed for
all compounds and suggests that compounds bind to tubulin close
to fluorophore, i.e., tryptophan and tyrosine residues. The inset of
Figure 3 shows the titration curve for the association of compound
1 to tubulin. The apparent affinity stoichiometric binding constant
was calculated. We found a Ka value of (1.41 ( 0.52) ꢁ 106 Mꢀ1
for compound 1. Similar experiments and assumptions were used
to determine the binding affinity of the other analogues 2ꢀ8
(Table 6).
Effects on Tubulin Assembly. Figure 2 shows the effects of
compound 1 on the fluorimetry time course of microtubule
assembly from pure tubulin. For monitoring the microtubule
formation, we used 40,6-diamidino-2-phenylindole (DAPI)
fluorescence. This well-known DNA binding agent binds to
microtubules with a greater affinity than to tubulin dimer and
its fluorescence was increased upon microtubule binding.19,20
Therefore, DAPI is an interesting fluorophor to monitor the
microtubule assembly because it does not affect the tubulin
properties to form microtubule and its use is more sensitive than
turbidimetric assay. By using this fluorimetric assay, a clear
inhibition was noted, and the rate of assembly, as well as the
final amount of microtubules, was lower in the presence of 1 than
in the control experiment. A similar experiment (as a positive
control inhibitor, line 6) performed with combretastatin A-4 at 3
μM showed a full inhibition of microtubule formation. The
extent of inhibition by all compounds increased monotonically
with the mole ratio of the total ligand to total tubulin in the
solution (R). Table 6 indicates the half inhibitory molar ratio
(ligand/tubulin) of microtubule formation in vitro. As an exam-
ple for combretastatin A-4 (a substoichiometric inhibitor of
microtubule formation), a mole ratio of 0.07 mol of combretas-
tatin per mol of tubulin was necessary to halve microtubule
formation. The most active compounds (substoichiometric in-
hibitors of the microtubule formation) were the A-ring-unsub-
stituted compound 1 and the monosubstituted compounds 2 and
3. We noted that the third monosubstituted compound 4 had a
The unsubstituted compound 1 and the monosubstituted
compound 2 had the highest apparent affinity binding constant,
in the same range as that of combretastatin A-4, i.e., about 1 ꢁ
6
10
M
ꢀ1. These latter molecules were the most efficient
inhibitors of microtubule formation in vitro. Then 3, 4, and 7
had a less pronounced interaction with protein but still have a
substoichiometric mode of inhibition. Finally, less potent in tubulin
assembly inhibition, with a stoichiometric mode of inhibition, 5, 6,
and 8 have the lowest apparent affinity binding constants. Thus, the
simplest correlation between affinity and inhibitory capacity seems
easily feasible.
3156
dx.doi.org/10.1021/jm901826e |J. Med. Chem. 2011, 54, 3153–3162

Journal of Medicinal Chemistry
ARTICLE
Figure 2. Effect of 1 on in vitro microtubule assembly. Panel shows tubulin alone (line 1) at 15 μM and with various concentrations of compound 1 at
1.5 μM (line 2), 2.7 μM (line 3), 4.4 μM (line 4), 5.5 μM (line 5). Line 6 is a positive control in the presence of combretastatin A-4 at 3 μM.
were performed at the semiempirical level using the Ampac 6.55
Table 6. Inhibitory Effects on Microtubule Assembly at 37 ꢀC
and Binding Parameters of Combretastatin A-4 and
Compounds 1ꢀ8 to Tubulin in PG Buffer at 25 ꢀCa
package22 and atthe DFT level using the Gaussian G03 software.23
All the calculations were carried out on a Silicon Graphics Origin
2000 R12000 workstation. Because the molecules have numerous
possible conformations, we started with the extracted X-ray
structure of combretastatin A-4 and compound 7. After this step,
the built geometries were optimized without constraints at the
DFT B3LYP/6ꢀ31G(d) level of theory. Superimpositions of
B-ring of optimized DFT compounds 1ꢀ8 are shown in Figure 5,
while selected parameters are listed in Table 8.
half inhibitory molar
equilibrium binding
ratio (ligand/tubulin) of
constant to tubulin Ka
compd
microtubule formation in vitro
[Mꢀ1
]
combretastatin
0.07
(3.35 ( 1.46) ꢁ 106b
A-4
1
0.18
0.16
0.23
0.91
2.31
2.68
0.29
1.50
(1.41 ( 0.52) ꢁ 106
(1.64 ( 0.80) ꢁ 106
(7.40 ( 1.10) ꢁ 105
(4.55 ( 1.23) ꢁ 105
(1.34 ( 0.47) ꢁ 105
(2.47 ( 0.70) ꢁ 105
(3.75 ( 1.23) ꢁ 105b
(6.72 ( 0.88) ꢁ 104b
The superimposition of compounds 1ꢀ8 based on B-ring
showed an extensive overlapping of the chromenone core. This
result supports the hypothesis that the spatial arrangement of the
two aromatic rings (A) and (B) plays an essential role in the activity
and binding of compounds that bind to the colchicine domain
(main binding site and additional pockets) on R-andβ-tubulin. The
30-OH and 40-OMe substituents on the 4-phenyl B-ring are crucial
for activity and were identified as pharmacophoric groups. The
slight difference of biological responses observed can be discussed in
terms of the substitution pattern and the number of methoxy groups
on the phenyl A-ring. Vicinal dimethoxy-substituted derivatives
(5, 6, and 8) were 8ꢀ15-fold less active than unsubstituted
compound 1. Therefore, it appears clearly that the presence of a
constraint methoxy group out of the A-ring plane is a detrimental
parameter for the activity against tubulin assembly and binding. This
steric hindrance factor seems to play a pivotal role in accommodat-
ing the A-ring inside the active binding pocket on tubulin. More-
over, the potent inhibitory effect and large binding affinity constant
for compound 1 indicates that 4-arylcoumarin compounds bind to
colchicine domain inside additional pockets buried deeper in the
tubulin heterodimer. This situation is similar to that of the tubulin
vinca alkaloïds domain that is described as consisting of a core
targeted by all ligands that interact with vinblastine for tubulin
binding and of ligand-dependent extensions.24
2
3
4
5
6
7
8
a Values are the mean of three experiments. Ka values are given ( SD of
the mean. b Determined in the author’s laboratory in a previous work.7,8
Molecular Structure and Biological Activity. The molecular
structure of 5,7-dimethoxycoumarin derivative 7, previously stu-
died in our group,7,8,16 was determined by single-crystal X-ray
crystallography and compared to the molecular structure of
combretastatin A-4.21 The ORTEP diagram of compound 7 is
presented in Figure 4, while selected parameters are listed in
Table 7. The tridimensional structure revealed a molecular con-
formation in which the two aromatic rings, A and B, are twisted
and form the same conformation of the diaryl system than
combretastatin A-4, with a very similar torsion angle (48ꢀ com-
pared to 53ꢀ, respectively). These hydrophobic planar groups are
crucial for antiproliferative activity and serve as the rigid moiety of
the molecular scaffold that should satisfy the overall geometric and
steric requirements of binding with tubulin.
’ CONCLUSION
To explain the relative biological activity of the different
compounds (combretastatin A-4, 1ꢀ8) theoretical calculations
It has been long thought that the presence of the trimethoxyben-
zene moiety was crucial to obtain relevant cytotoxic and antitubulin
3157
dx.doi.org/10.1021/jm901826e |J. Med. Chem. 2011, 54, 3153–3162

Journal of Medicinal Chemistry
ARTICLE
Figure 3. Tryptophan fluorescence modifications of tubulin upon binding of compound 1 at 25 ꢀC in PG buffer. The excitation wavelength was 295 nm.
Fluorescence emission of 3.2 μM tubulin (—), with 1 μM ( ), and 12 μM (---) compound 1. Inset: quenching fluorescence titration curve of 3.2 μM
3 3 3
tubulin with various compound 1 concentrations. This curve was analyzed as described in materials and methods, and for this experiment, an affinity
constant of (1.1 ( 0.2) ꢁ 106 Mꢀ1 was obtained.
Figure 5. Superimposition of the B-ring of compounds 1 (red), 2
(yellow), 3 (blue), 4 (green), 5 (orange), 6 (purple), 7 (gray), and 8
(white).
Unsubstituted compound 1 and monosubstituted compound
2 showed a substoichiometric antitubulin activity comparable
to that of combretastatin A-4 with a highly similar apparent
Figure 4. ORTEP drawing of compound 7 (all hydrogen atoms are
omitted for clarity).
affinity binding constant. The slight loss of potency for the
other synthesized compounds was correlated with a decrease
in the values of their affinity binding constants. The interac-
tion with protein surface appeared down regulated by the
presence of a methoxy group directed up to the molecular
plane of ligand. Previous to a use in an in vivo model the P-gp
and BCRP expression of cancer cells was stated. In the case of
a single use in a chemotherapeutic regimen, a concentration of
10 μM of compounds seems to be needed to be insensitive to
the efflux activity. In addition to that sum of results, this work
provide a consistent experimental basis to guide the design of
new potent antitubulin agents that target the whole colchicine
binding domain.
Table 7. X-ray Crystallographic Data of Compound 7
torsion angle
deg
distance
Å
ring (A)ꢀring (B)
ring (A)ꢀ5-OMe
ring (A)ꢀ7-OMe
48.3
5.8
centroid (A)-O40
centroid (A)-Me40
centroid (A)-O30
7.721a
8.786b
6.250c
ꢀ11.6
a CA-4: 7.377 Å. b 8.586 Å. c 5.629 Å.
responses. The modifications of the trimethoxyphenyl A-ring
demonstrated that the methoxy groups are not essential for
the interaction and activity of the 4-arylcoumarin compounds.
3158
dx.doi.org/10.1021/jm901826e |J. Med. Chem. 2011, 54, 3153–3162

Journal of Medicinal Chemistry
ARTICLE
Table 8. DFT Calculation Data of Combretastatin A-4 and Compounds 1ꢀ8
compd
C A-4
1
2
3
4
5
6
7
8
Torsion Angle (deg)
ring (A)ꢀring (B)
Me7O7C7C8
Me6O6C6C7
Me5O5C5C6
53.0
172.2
93.4
8.6
57.9
63.2
57.4
57.7
0.4
63.5
56.5
3.3
63.2
0.5
64.5
3.7
ꢀ0.1
4.0
ꢀ59.5
ꢀ66.9
36.4
0.5
ꢀ63.2
0.7
Distance (Å)
7.610
centroid (A)ꢀO40
centroid (A)ꢀMe40
centroid (A)ꢀO30
7.377
8.587
5.629
7.600
7.818
7.608
7.848
7.600
7.817
7.790
8.833
6.522
8.662
6.301
8.858
6.557
8.673
8.671
6.305
8.869
6.639
8.665
6.286
8.856
6.564
6.304
’ EXPERIMENTAL SECTION
(3H, s), 4.01 (3H, s), 6.35 (1H, s), 6.91 (1H, s), 6.99 (1H, s); δC 56.4,
56.7, 100.3, 102.2, 102.9, 106.1, 118.5, 147.2, 150.0, 154.9, 157.4, 160.4.
Coupling of 4-Trifluoromethylsulfonyloxycoumarins with
Arylboronic Acids: General Procedure. A mixture of 4-trifl-
uoromethylsulfonyloxycoumarin (1 mmol), tetrakis(triphenylphosph-
ine)palladium (0) (46 mg, 4 mol%), copper(I) iodide (210 mg, 1.1
equiv), sodium carbonate (742 mg, 7 equiv), and arylboronic acid
(252 mg, 1.5 equiv) in dry benzene (10 mL) and absolute ethanol
(3 mL) was refluxed overnight. Then the reaction mixture was diluted
with chloroform (40 mL) and filtered through celite. The filtrate was
washed with a saturated aqueous solution of sodium bicarbonate (3 ꢁ
20 mL), and the combined aqueous layers were extracted with chloro-
form (3 ꢁ 20 mL). The organic phases were combined, washed with
brine, and dried over Na2SO4, and the solvent was distilled under
reduced pressure. The residue was crystallized from ether or purified by
column chromatography to afford the 4-arylcoumarin derivative.
4-(30-Hydroxy-40-methoxyphenyl)coumarin 1. Purified by column
chromatography, eluent Et2Oꢀpentane (7:3), as white needles, 81%,
mp 172 ꢀC; δH 3.99 (3H, s), 5.78 (1H, s), 6.36 (1H, s), 6.97 (1H, dd,
J 8.1 and 1.3), 7.00 (1H, d, J 8.1), 7.06 (1H, d, J 1.3), 7.23 (1H, t, J 7.9),
7.40 (1H,d, J 7.9), 7.54 (1H, td, J 7.9 and 1.5), 7.60 (1H, dd, J 7.9 and
1.5); δC 56.1, 110.9, 114.6, 114.9, 117.3, 119.0, 120.7, 124.2, 127.2,
128.3, 131.8, 145.9, 147.9, 154.2, 155.4, 161.1. Anal. (C16H12O4) C, H.
4-(30-Hydroxy-40-methoxyphenyl)-5-methoxycoumarin 2. Crystal-
lized from Et2O as light-brown crystals, 65%, mp 180 ꢀC; δH 3.53 (3H, s),
3.94 (3H, s), 6.15 (1H, s), 6.67 (1H, dd, J 8.3 and 0.7), 6.79 (1H, dd, J 8.1
and 1.9), 6.86 (1H, d, J 8.1), 6.87 (1H, d, J 1.9), 7.00 (1H, dd, J 8.3 and
0.7) and 7.44 (1H, t, J 8.3); δC 55.6, 55.9, 106.7, 109.2, 109.6, 109.7,
114.1, 115.6, 118.9, 132.2, 132.7, 144.7, 146.7, 155.1, 155.3, 157.3, 160.6.
Anal. (C17H14O5) C, H.
Chemistry. Melting points were taken on a B€uchi capillary appara-
tus and are uncorrected. NMR spectra were obtained on a Bruker AC
300 spectrometer. Chemical shifts (δ) are reported in ppm for a solution
of the compound in CDCl3 with internal reference Me4Si and J values in
hertz. Combustion analyses were performed at the Laboratoire de
Microanalyse of the Centre National de la Recherche Scientifique,
Vernaison, and all final compounds have a purity of >95%. Separation
by column chromatography was performed using Merck Kieselgel 60
(70ꢀ230 mesh). Ether refers to diethyl ether and petroleum spirit refers
to the fraction with distillation range 40ꢀ65 ꢀC. All solvents were
purified by standard techniques. 3-Benzyloxy-4-methoxyphenylboronic
acid 18 was prepared by reported procedures involving reaction of the
appropriate aryllithium with triisopropylborate.15 4-Trifluorometh-
ylsulfonyloxychromen-2-one 10,25 5-methoxy-4-trifluoromethylsulfo-
nyloxychromen-2-one 11,25 6-methoxy-4-trifluoromethylsulfonyloxychro-
men-2-one 12,26 7-methoxy-4-trifluoromethylsulfonyloxychromen-2-one
13,15,27 5,7-dimethoxy-4-trifluoromethylsulfonyloxychromen-2-one 16,25
and 5,6,7-trimethoxy-4-trifluoromethylsulfonyloxychromen-2-one 177 were
prepared as previously reported.
3-Hydroxy-4-methoxyphenylboronic Acid 19. A mixture of
3-benzyloxy-4-methoxyphenylboronic acid 18 (5.16, 20 mmol) and 10%
palladium on graphite (2 w%) in dry tetrahydrofuran (40 mL) was
stirred under hydrogene atmosphere for 24 h. The mixture was diluted
with acetone and filtered through celite, and the filtrate was concen-
trated under reduced pressure. The solid residue was washed with 50%
etherꢀpentane and collected to afford 3-hydroxy-4-methoxyphenyl-
boronic acid 19 as white powder (2.86 g, 85%), mp 221 ꢀC. 1H NMR
(acetone-d6, 300 MHz): δH 3.87 (3H, s), 6.91 (2H, s), 6.94 (1H, d,
J 8.3), 7.35 (1H, s), 7.88 (1H, d, J 1.6), 7.38 (1H, dd, J 8.3).
4-(30-Hydroxy-40-methoxyphenyl)-6-methoxycoumarin 3. Crystal-
lized from Et2O as light-brown crystals, 97%, mp 167 ꢀC; δH 3.76 (3H, s),
3.99 (3H, s), 6.34 (1H, s), 6.96 (1H, dd, J 8.2 et 1.7), 6.99 (1H, d, J 8.2),
7.04 (1H, d, J 2.8), 7.06 (1H, d, J 1.7), 7.12 (1H, dd, J 9.0 and 2.8), 7.33
(1H, d, J 9.0); δC 55.9, 56.1, 110.0, 110.9, 114.9, 115.1, 118.2, 119.0,
119.5, 120.5, 128.4, 146.2, 148.0, 148.6, 155.2, 155.9, 161.3. Anal.
(C17H14O5) C, H.
Preparation of the 4-Trifluoromethylsulfonyloxychro-
men-2-ones: General Procedure. Trifluoromethanesulfonic anhy-
dride (0.66 mL, 1.3 equiv) was added dropwise over 10 min to a mixture
of the appropriate 4-hydroxycoumarin (3 mmol) and triethylamine
(0.55 mL, 1.3 equiv) in dry dichloromethane (20 mL) at 0 ꢀC. Then the
mixture was stirred for 1 h at 0 ꢀC. After dilution with 50% etherꢀ
petroleum spirit (50 mL) and filtration through a short pad of silica, the
solvent was distilled off under reduced pressure to a small volume
(10 mL) and the residue kept at ꢀ15 ꢀC overnight. The precipitate was
collected and washed with petroleum spirit to afford the expected
4-trifluoromethylsulfonyloxychromen-2-one.
4-(30-Hydroxy-40-methoxyphenyl)-7-methoxycoumarin 4. Crystal-
lized from Et2O as light-brown crystals, 98%, mp 173 ꢀC (lit.28
169ꢀ171 ꢀC); δH 3.89 (3H, s), 3.98 (3H, s), 6.19 (1H, s), 6.79 (1H,
dd, J 8.8 and 2.7), 6.88 (1H, d, J 2.7), 6.95 (1H, dd, J 8.1 and 1.5), 6.98
(1H, d, J 8.1), 7.04 (1H, d, J 1.5), 7.49 (1H, d, J 8.8); δC 55.0, 55.2, 100.3,
110.3, 110.8, 111.4, 111.8, 114.6, 119.5, 127.6, 127.7, 146.0, 148.0, 155.0,
155.4, 160.3, 162.3. Anal. (C17H14O5) C, H.
5,6-Dimethoxy-4-trifluoromethylsulfonyloxychromen-2-one 14.
Light-yellow needles, 97%, mp 95ꢀ98 ꢀC; δH 3.91 (3H, s), 3.97 (3H, s),
6.31 (1H, s), 7.12 (1H, d, J 8.6), 7.24 (1H, d, J 8.6); δC 56.6, 61.6,
108.4, 109.2, 112.2, 118.3, 118.5, 144.7, 147.5, 149.6, 157.1, 159.5.
6,7-Dimethoxy-4-trifluoromethanesulfonyloxycoumarin 15. White
4-(30-Hydroxy-40-methoxyphenyl)-5,6-dimethoxycoumarin 5. Cry-
stallized from Et2O as light-brown crystals, 80%, mp 165ꢀ166 ꢀC; δH
3.24 (3H, s), 3.85 (3H, s), 3.93 (3H, s), 6.20 (1H, s), 6.84 (1H, dd, J 8.3
and 1.5), 6.86 (1H, d, J 8.3), 6.96 (1H, d, J 1.5), 7.11 (1H, d, J 9.2), 7.16
(1H, d, J 9.2); δC 55.9, 56.5, 60.9, 109.7, 111.9, 114.0, 114.3, 116.3,
1
H
needles, 73%, mp 100 ꢀC; H NMR (acetone-d6, 300 MHz): δ 3.95
3159
dx.doi.org/10.1021/jm901826e |J. Med. Chem. 2011, 54, 3153–3162

Journal of Medicinal Chemistry
ARTICLE
117.2, 119.4, 131. 7, 144.7, 146.3, 148.1, 149.6, 154.6, 160.6. Anal.
(C18H16O6) C, H.
chromatographed in drained spin columns (1 cm ꢁ 5 cm) of Sephadex
G25, equilibrated with PG buffer (20 mM sodium phosphate, 0.1 mM
GTP, pH 7), followed by passage through a gravity column of Sephadex
G25 (1 cm ꢁ 10 cm) and equilibrated with the same buffer. Protein
concentration was measured spectrophotometrically with a Perkin-
Elmer spectrophotometer Lambda 800 at 275 nm with an extinction
coefficient of 1.09 L gꢀ1 cmꢀ1 in guanidine hydrochloride or 1.07 L
4-(30-Hydroxy-40-methoxyphenyl)-6,7-dimethoxycoumarin 6. Cry-
stallized from Et2O as light-yellow crystals, 66%, mp 208 ꢀC; δH 3.79
(3H, s), 3.97 (3H, s), 3.99 (3H, s), 5.55 (1H, s), 6.22 (1H, s), 6.91 (1H, s),
6.97ꢀ6.99 (3H, m), 7.06 (1H, s); δC 56.1, 56.4, 56.5, 100.3, 107.6,
110.9, 111.6, 112.0, 114.7, 120.5, 129.0, 146.0, 146.1, 147.8, 150.2, 152.9,
155.4, 161.7. (C18H16O6) C, H.
g
ꢀ1 cmꢀ1 in 0.5% SDS in neutral aqueous buffer.
4-(30-Hydroxy-40-methoxyphenyl)-5,7-dimethoxycoumarin 7. Pur-
ified by column chromatography, eluent Et2O, as white needles, 83%,
mp 152 ꢀC (lit.7 152 ꢀC); δH 3.50 (3H, s), 3.86 (3H, s), 3.94 (3H, s),
5.82 (1H, s), 5.99 (1H, s), 6.24 (1H, d, J 2.3), 6.50 (1H, d, J 2.3), 6.78
(1H, dd, J 8.1 and 1.9), 6.85 (1H, d, J 8.1), 6.86 (1H, d, J 1.9).
(C18H16O6) C, H.
Tubulin Assembly. Microtubule assembly was monitored on a
Fluoroscan Ascent FL spectrofluorometer (Labsystems) using a 96-well
plate.30 The excitation wavelength was set at 355 nm, and the emission
wavelength was set at 460 nm. Experiments were carried out at 37 ꢀC
and performed with 7.5 μM Dapi, 15 μM tubulin in 20 mM sodium
phosphate buffer, 1 mM ethylene glycol tetraacetic acid (EGTA),
10 mM MgCl2, 10ꢀ4 M GTP, and 3.4 M glycerol, pH 6.5. Under these
conditions, the Dapi fluorescence enhancement is directly proportional
to the concentration of polymerized tubulin and was monitored as a
function of time.31 Me2SO concentration was maintained below 4% in
all samples and controls. Experiments were done in triplicate
Binding Measurement by Fluorescence Titration. Fluorescence
measurements were performed with a Perkin-Elmer luminescence
spectrometer 50 with slit widths of 10/10 nm. Uncorrected fluorescence
spectra were obtained in PG buffer, pH 7, by using 0.2 cm (excitation
direction) ꢁ 1 cm cells (Hellma) thermostatted at 25 ꢀC by circulating
water from an external water bath. Determination of the binding
parameters was done by measuring quenching of the intrinsic protein
fluorescence signal by ligands. Tubulin (1ꢀ4 μM) was titrated with
various concentrations of analogues. Fluorescence measurements were
performed with an excitation wavelength of 295 nm in order to
specifically excite the tubulin tryptophanyl residues.
4-(30-Hydroxy-40-methoxyphenyl)-5,6,7-trimethoxycoumarin 8.
Purified by column chromatography, eluent Et2O, as white fine
needles, 94%, mp 157 ꢀC (lit.7 157 ꢀC); δH 3.36 (3H, s), 3.80 (3H, s),
3.94 (3H, s), 3.95 (3H, s), 6.06 (1H, s), 6.71 (1H, s), 6.83 (1H, dd,
J 8.2 and 2.1), 6.99 (1H, d, J 8.2), 6.92 (1H, d, J 2.1); δC 56.0, 56.3,
61.1, 61.2, 96.3, 107.4, 109.8, 114.1, 114.2, 119.3, 132.3, 139.6,
144.7, 146.7, 151.3, 151.8, 155.2, 156.8, 160.8. Anal. (C19H18O7)
C, H.
Biology. Cytotoxicity Assays. Cells were seeded in 96-well plates
to be treated during 72 h. Growth inhibition of HBL100 cell line was
studied after a 72 h treatment with different compounds by using
the colorimetric 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide (MTT; Sigma) cell proliferation assay, and absorbance was
measured at 550 nm with a Dynatech MR 7-000 plate reader. Three
independent experiments were performed, and the IC50 values (i.e.,
concentration half inhibiting cell proliferation) were graphically determined.
Evaluation on P-gp and BCRP Activity. Accumulation studies of
mitoxantrone (BCRP substrate) and daunorubicin (P-gp substrate)
were performed using wild type (sensitive) HCT116S cell line, which
does not express BCRP, and the BCRP-transfected resistant form
HCT116R, and on wild type (sensitive) K562S and the P-gp expressing
(resistant) K562R cell lines.29 Cells grown in RPMI 1640 medium with
10% FCS were trypsinized (HCT116), washed twice, then resuspended
in RPMI 1640 medium with 10% FCS to obtain a density of 106 cells/mL.
Various concentrations (0ꢀ150 μM) of 4-arylcoumarin compounds or
the BCRP and P-gp inhibitor cyclosporine A (10 μM final con-
centration) were added to 1 mL of cells and incubated for 15 min at
37 ꢀC, followed by the addition of mitoxantrone (final concentration
3 μM) or daunorubicin (final concentration 1 μM). After 30 min of
incubation at 37 ꢀC, 4 mL of ice-cold PBS were added to stop drug
accumulation and fluorescence was quantified by flow-cytometry and
analyzed using CELLQUEST PRO software (Becton Dickinson Sciences).
Evaluation as Potential Substrates of P-gp and/or BCRP. Cell Cycle
Analysis. HCT116S and R, K562S and R exponentially growing cells
were incubated for 24 h in the presence of DMSO (vehicle/0.1%) or
4-arylcoumarin compounds at 0.1 and 10 μM. Cell DNA content was
analyzed using the CycleTest PLUS/DNA reagent kit (BD Sciences,
San Jose, CA). Data were collected and analyzed on a FACSCalibur
flow-cytometer (Becton Dickinson, La Jolla, CA) using CELLQUEST
PRO software (BD Sciences, Le Pont de Claix France).
Fluorescence Binding Titration for Compounds. Fluorescence emis-
sion was recorded at 340 nm. The inner filter effects were corrected
according to the procedure as follows:32
Fcorr ¼ Fobs expfðAexc þ AemÞ=2g
Fobs and Fcorr are the observed and corrected fluorescence values at the
emission wavelengths. Aexc and Aem are the absorptions at the excitation
and emission wavelengths, respectively, calculated with Ax = εxlxC, in
which x is the excitation or emission direction, ε is the extinction
coefficient, l is the path length of the cell in the excitation and emission
directions, and C is the ligand concentration. The corrected quenching
fluorescence titration curves were inverted and fitted to the saturation
curve equation by means of nonlinear least-squares regression analysis.
Fmax½Lf ꢂ
Fcorr
¼
Kd þ ½Lf ꢂ
Fmax is the plateau fluorescence value. Concentrations (bound ligand
[Lb] and free ligand [Lf]) and Kd the stoichiometric dissociation binding
constant were determined with the following equation:
1
2
1=2
½Lbꢂ ¼ fð½L0ꢂ þ ½P0ꢂ þ KdÞ ꢀ ðð½L0ꢂ þ ½P0ꢂ þ KdÞ ꢀ 4½P0ꢂ½L0ꢂÞ
g
2
where [L0] and [P0] are the total ligand and protein concentrations,
respectively. The algorithm starts with an arbitrary opening Kd value.
With this value, [Lb] and [Lf] were calculated and then the nonlinear
least-squares regression analyses were executed. The initial set is
corrected in the next step by a NewtonꢀGauss procedure. This iterative
procedure is continued until the minimum sum of squared deviations
between experimental and calculated values of Fcorr is obtained.
The concentrations of compounds were measured spectrophotome-
trically. The extinction coefficients, ε, were determined by dissolving dry
powder of the compounds in Me2SO (dimethylsulfoxide) and diluting in
20 mM NaPi buffer, pH 7, then taking the UV visible spectra. Three
independent determinations gave the values displayed in Table 9.
Cytotoxicity Assays. HCT116S and R, K562S and R were cultured in
96-well plates for 48H. The number of viable cells was determined using
MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide)
assay, as described by the manufacturer (Sigma Aldrich, l’Isle d’Abeau,
France). Three independent experiments were performed and the IC50
values (concentration half inhibiting cell proliferation) were graphically
determined.
Tubulin Purification. Tubulin was extracted from lamb brains by
ammonium sulfate fractionation and ion exchange chromatography
and stored in liquid nitrogen.8 Before use, aliquots of protein were
3160
dx.doi.org/10.1021/jm901826e |J. Med. Chem. 2011, 54, 3153–3162

Journal of Medicinal Chemistry
ARTICLE
Table 9. Extinction Coefficients of 4-Arylcoumarins 1ꢀ6
compd
1
2
3
4
5
6
ε320 nm = 12400 ( 640 Mꢀ1 cmꢀ1
ε330 nm = 12630 ( 800 Mꢀ1 cmꢀ1
ε309 nm = 14500 ( 950 Mꢀ1 cmꢀ1
ε320 nm = 13380 ( 750 Mꢀ1 cmꢀ1
ε295 nm = 14093 ( 650 Mꢀ1 cmꢀ1
ε340 nm = 11140 ( 850 Mꢀ1 cmꢀ1
ε295 nm = 11190 ( 510 Mꢀ1 cmꢀ1
ε295 nm = 10306 ( 550 Mꢀ1 cmꢀ1
ε295 nm = 12614 ( 700 Mꢀ1 cmꢀ1
ε295 nm = 11195 ( 950 Mꢀ1 cmꢀ1
ε295 nm = 14093 ( 500 Mꢀ1 cmꢀ1
ε295 nm = 6911 ( 360 Mꢀ1 cmꢀ1
ε340 nm = 7300 ( 310 Mꢀ1 cmꢀ1
ε340 nm = 7794 ( 420 Mꢀ1 cmꢀ1
ε340 nm = 7930 ( 255 Mꢀ1 cmꢀ1
ε340 nm = 7008 ( 350 Mꢀ1 cmꢀ1
ε340 nm = 6212 ( 290 Mꢀ1 cmꢀ1
ε340 nm = 11140 ( 645 Mꢀ1 cmꢀ1
’ AUTHOR INFORMATION
predominantly with cysteine 239 and secondarily with cysteine 354.
J. Biol. Chem. 2000, 275, 40443–40452.
Corresponding Author
(10) Nguyen, T. L.; McGrath, C.; Hermone, A. R.; Burnett, J. C.;
Zaharevitz, D. W.; Day, B. W.; Wipf, P.; Hamel, E.; Gussio, R. A common
pharmacophore for a diverse set of colchicine site inhibitors using a
structure-based approach. J. Med. Chem. 2005, 48, 6107–6116.
(11) Naumov, M. I.; Nuchev, A. V.; Sitnikov, N. S.; Malysheva, Y. B.;
Shavyrin, A. S.; Beletskaya, I. P.; Gavryushin, A. E.; Combes, S.; Fedorov,
A. Y. 2-(Azidomethyl)arylboronic acids in the synthesis of coumarin-
type compounds. Synthesis 2009, 10, 1673–1682.
(12) Ganina, O. G.; Daras, E.; Bourgarel-Rey, V.; Peyrot, V.;
Andresuk, A. N.; Finet, J.-P.; Fedorov, A. Y.; Beletskaya, I. P.; Combes,
S. Synthesis and biological evaluation of polymethoxylated 4-heteroar-
ylcoumarins as tubulin assembly inhibitor. Bioorg. Med. Chem. 2008,
16, 8806–8812.
*For S.C.: phone, (33) 491-288-291; fax, (33) 491-288-758;
e-mail, sebastien.combes@univ-provence.fr. For V.P.: phone:
(33) 491-835-505; fax, (33) 491-835-505; e-mail, vincent.
peyrot@univmed.fr.
’ ACKNOWLEDGMENT
We thank INTAS for financial support (grant no. 03-514-915)
and Professor Didier Siri and Sandrine Chenesseau for molecular
calculations. We are very grateful to “Alazard et Roux” Slaughter
House (Tarascon, France) for the lamb brain necessary for
tubulin purification.
(13) Luo, Y.; Wu, J. Palladium-catalyzed direct arylation of 4-hydro-
xycoumarins with arylboronic acids via CꢀOH bond activation. Tetra-
hedron Lett. 2009, 50, 2103–2105.
’ ABBREVIATIONS
(14) Wu, J.; Wang, L.; Fathi, R.; Yang, Z. Metal catalysed carbonꢀ
carbon bond-forming reaction. Catalysts Fine Chem. Synth. 2004,
3, 64–67.
(15) Donnelly, D. M. X.; Finet, J.-P.; Guiry, P. J.; Rea, M. D.
Synthesis of C-ring hydroxylated neoflavonoids by ligand coupling
reactions. Synth. Commun. 1999, 29, 2719–2730.
(16) Billard, C.; Menasria, F.; Quiney, C.; Faussat, A. M.; Finet, J.-P.;
Combes, S.; Kolb, J.-P. 4-Arylcoumarin analogues of combretastatins
stimulate apoptosis of leukemic cells from chronic lymphocytic leukemia
patients. Exp. Hematol. 2008, 36, 1625–1633.
P-gp, P-glycoprotein; BCRP, breast cancer resistance protein;
CA-4, combretastatin A4; CA-4P, combretastatin A-4 phos-
phate; ITC, isothermal titration calorimetry; THF, tetrahydro-
furan; SE, standard error; SD, standard deviation; AU, arbitrary units;
MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bro-
mide; DAPI, 40,6-diamidino-2-phenylindole; EGTA, ethylene
glycol tetraacetic acid
(17) Lage, H. An overview of cancer multidrug resistance: a still
unsolved problem. Cell. Mol. Life Sci. 2008, 65, 3145–3167.
(18) Wu, C. P.; Calcagno, A. M.; Ambudkar, S. V. Reversal of ABC
drug transporter-mediated multidrug resistance in cancer cells: evalua-
tion of current strategies. Curr. Mol. Pharmacol. 2008, 1, 93–105.
(19) Heusele, C.; Bonne, D. Role of DAPI in microtubule reactions
at steady-state. Biochem. Biophys. Res. Commun. 1985, 133, 662–669.
(20) Bonne, D.; Heusele, C.; Simon, C.; Pantaloni, D. 40,6-Diami-
dino-2-phenylindole, a fluorescent probe for tubulin and microtubules.
J. Biol. Chem. 1985, 260, 2819–2825.
(21) Lara-Ochoa, F.; Espinosa-perez, G. A new synthesis of com-
bretastatins A-4 and AVE-8062A. Tetrahedron Lett. 2007, 48, 7007
7010.
(22) AMPAC 6.55; Semichem, Inc.: P.O. Box 1649, Shawnee, KS
66222, 1998.
’ REFERENCES
(1) Kingston, D. G. I. Tubulin-interactive natural products as antic-
ancer agents. J. Nat. Prod. 2009, 72, 507–515.
(2) Tozer, G. M.; Kanthou, C.; Parkins, C. S.; Hill, S. A. The biology
of the combretastatins as tumour vascular targeting agents. Int. J. Exp.
Pathol. 2002, 83, 21–38.
(3) Hinnen, P.; Eskens, F. A. L. M. Vascular disrupting agents in
clinical development. Br. J. Cancer 2007, 96, 1159–1165.
(4) Kanthou, C.; Tozer, G. M. Tumour targeting by microtubule-
depolymerising vascular disrupting agents. Expert Opin. Ther. Targets
2007, 11, 1443–1457.
(5) Banerjee, S.; Wang, Z.; Mohammad, M.; Sarkar, F. H.; Mohammad,
R. M. Efficacy of selected natural products as therapeutic agents against
cancer. J. Nat. Prod. 2008, 71, 492–496.
(6) Butler, M. S. Natural products to drugs: natural product-derived
compounds in clinical trials. Nat. Prod. Rep. 2008, 25, 475–516.
(7) Bailly, C.; Bal, C.; Barbier, P.; Combes, S.; Finet, J.-P.; Hildebrand,
M. P.; Peyrot, V.; Wattez, N. Synthesis and biological evaluation of
4-arylcoumarin analogues of combretastatins. J. Med. Chem. 2003,
46, 5437–5444.
(8) Rappl, C.; Barbier, P.; Bourgarel-Rey, V.; Gregoire, C.; Gilli, R.;
Carre, M.; Combes, S.; Finet, J.-P.; Peyrot, V. Interaction of 4-arylcou-
marin analogues of combretastatins with microtubule network of
HBL100 cells and binding to tubulin. Biochemistry 2006, 45, 9210–9218.
(9) Bai, R.; Covell, D. G.; Pei, X. F.; Ewell, J. B.; Nguyen, N. Y.;
Brossi, A.; Hamel, E. Mapping the binding site of colchicinoids on
beta-tubulin-2-chloroacetyl-2-demethylthiocolchicine covalently reacts
(23) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A.; Vreven, T.; Kudin,
K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.;
Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.;
Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.;
Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.;
Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels,
A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
3161
dx.doi.org/10.1021/jm901826e |J. Med. Chem. 2011, 54, 3153–3162

Journal of Medicinal Chemistry
ARTICLE
Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C. and Pople, J. A.
Gaussian 03, Revision C.02; Gaussian, Inc.: Wallingford CT, 2004.
(24) Cormier, A.; Marchand, M.; Ravelli, R. B. G.; Knossow, M.;
Gigant, B. Structural insight into the inhibition of tubulin by vinca
domain peptide ligands. EMBO Rep. 2008, 9, 1101–1106.
(25) Boland, G. M.; Donnelly, D. M. X.; Finet, J.-P.; Rea, M. D.
Synthesis of neoflavones by Suzuki arylation of 4-substituted coumarins.
J. Chem. Soc., Perkin Trans. 1 1996, 2591–2597.
(26) Kym, P. R.; Iyengar, R.; Souers, A. J.; Lynch, J. K.; Judd, A. S.;
Gao, J.; Freeman, J.; Mulhern, M.; Zhao, G.; Vasudevan, A.; Wodka, D.;
Blackburn, C.; Brown, J.; Che, J. L.; Cullis, C.; Lai, S. J.; LaMarche, M. J.;
Marsilje, T.; Roses, J.; Sells, T.; Geddes, B.; Govek, E.; Patane, M.; Fry,
D.; Dayton, B. D.; Brodjian, S.; Falls, D.; Brune, M.; Bush, E.; Shapiro,
R.; Knourek-Segel, V.; Fey, T.; McDowell, C.; Reinhart, G. A.; Preusser,
L. C.; Marsh, K.; Hernandez, L.; Sham, H. L.; Collins, C. A. Discovery
and characterization of aminopiperidinecoumarin melanin concentrat-
ing hormone receptor 1 antagonists. J. Med. Chem. 2005, 48, 5888–5891.
(27) Ciattini, P. G.; Morera, E.; Ortar, G. 4-Stannylcoumarins as
useful reagents in a new approach to neoflavonoids. Synth. Commun.
1995, 25, 2883–2894.
(28) Rizzi, E.; Dallavalle, S.; Merlini, L.; Pratesi, G.; Zunino, F. Short
synthesis of cytotoxic 4-arylcoumarins. Synth. Commun. 2006, 36,
1117–1122.
(29) Boumendjel, A.; McLeer-Florin, A.; Champelovier, P.; Allegro,
D.; Muhammad, D.; Souard, F.; Derouazi, M.; Peyrot, V.; Toussaint, B.;
Boutonnat, J. A novel chalcone derivative which acts as a microtubule
depolymerising agent and an inhibitor of P-gp and BCRP in in vitro and
in vivo glioblastoma models. BMC Cancer 2009, 9, 242–253.
(30) Deesamer, S.; Kokpol, U.; Chavasiri, W.; Douillard, S.; Peyrot,
V.; Vidal, N.; Combes, S.; Finet, J.-P. Synthesis and biological evaluation
of isoflavone analogues from dalbergia oliveri. Tetrahedron 2007, 63,
12986–12993.
(31) Heusele, C.; Bonne, D.; Carlier, M. F. Is microtubule assembly a
biphasic process—a flurometric study using 40,6-diamidino-2-phenylin-
dole as a probe. Eur. J. Biochem. 1987, 165, 613–620.
(32) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Plenum
Press: New York, 1983; pp 1ꢀ496.
3162
dx.doi.org/10.1021/jm901826e |J. Med. Chem. 2011, 54, 3153–3162