Arylthioindoles, potent inhibitors of tubulin polymerization

Article abstract of DOI:10.1021/jm049360d

Several arylthioindoles had excellent activity as inhibitors both of tubulin polymerization and of the growth of MCF-7 human breast carcinoma cells. Methyl 3-[(3,4,5-tri-methoxypheny)thio]-5-methoxy-1H-indole-2-carboxylate (21), the most potent derivative

Full text of DOI:10.1021/jm049360d

6120
J. Med. Chem. 2004, 47, 6120-6123
Chart 1
Arylthioindoles, Potent Inhibitors of
Tubulin Polymerization
Gabriella De Martino,† Giuseppe La Regina,†
Antonio Coluccia,† Michael C. Edler,‡
Maria Chiara Barbera,§ Andrea Brancale,§
Elizabeth Wilcox,‡ Ernest Hamel,‡
Marino Artico,† and Romano Silvestri*,†
Dipartimento di Studi Farmaceutici, Universita` di Roma
“La Sapienza”, Piazzale Aldo Moro 5, I-00185 Roma, Italy,
Welsh School of Pharmacy, Cardiff University,
King Edward VII Avenue, Cardiff, CF10 3XF, U.K., and
Screening Technologies Branch, Developmental Therapeutics
Program, Division of Cancer Treatment and Diagnosis,
National Cancer Institute at Frederick, National Institutes
of Health, Frederick, Maryland 21702
Received August 6, 2004
Abstract: Several arylthioindoles had excellent activity as
inhibitors both of tubulin polymerization and of the growth of
MCF-7 human breast carcinoma cells. Methyl 3-[(3,4,5-tri-
methoxyphenyl)thio]-5-methoxy-1H-indole-2-carboxylate (21),
the most potent derivative, showed IC50 ) 2.0 µM, 1.6 times
more active than colchicine and about as active as combreta-
statin A-4 (CSA4). Compound 21 inhibited the growth of the
MCF-7 cells at IC50 ) 13 nM. Colchicine and CSA4 had
13 nM and 17 nM IC50 values, respectively, with these cells.
Microtubules are involved in a wide number of cel-
lular functions, such as motility, division, shape main-
tenance, and intracellular transport. The major protein
component found in microtubules is tubulin. Interfer-
ence with microtubule assembly, either by inhibition of
tubulin polymerization or by blocking microtubule dis-
assembly, leads to an increase in the number of cells in
metaphase arrest. Inhibition of microtubule function
using tubulin targeting agents is a validated approach
to anticancer therapy.1-5
Many well-described antimitotic agents are derived
from natural sources. Colchicine (1) and Vinca alkaloids,
the first tubulin binding agents discovered, cause
destabilization of microtubules and, as a consequence,
induce the cell to undergo apoptosis.6 On the other hand
paclitaxel, a standard antitumor agent originally ex-
tracted from Taxus brevifolia, binds to a different site
within tubulin leading to the stabilization of microtu-
bules.7 Unfortunately, clinical use of these compounds
is restricted by toxicity, drug resistance, complex for-
mulations, and limited bioavailability.
The indole nucleus is the core structure of a great
number of tubulin polymerization inhibitors.2,8 The
indolyl-3-glyoxamide D-24851 (2) and the 2-aroylindoles
D-64131 (3a) and D-68144 (3b) were discovered by
Baxter Oncology. Compounds 3 are highly active against
various tumors including those resistant to paclitaxel9
(Chart 1).
Several 2-phenylindoles were designed by von
Angerer as simple analogues of 12-formyl-5,6-dihydro-
indolo[2,1-a]isoquinoline. Among them, indole 4 com-
pletely blocked microtubule assemby at a concentration
of 40 µM.10 On the basis of the structure of the natural
product combretastatin A-4 (CSA4, 6), some 2,3-di-
arylindoles, known as heterocombretastatins, were pre-
pared (e.g., 5) by Medarde.11 Flynn et al. reported the
tubulin polymerization inhibitory activity of 2,3-di-
arylindole 7 and 2-aryl-3-arylcarbonylindole 8.12
Sulfur containing compounds, such as sulfonamide
E-7010 (9),13 thiophene 10,14 and benzothiophene12
derivatives, proved effective inhibitors of tubulin po-
lymerization. To our knowledge there have been no
reports on the inhibition of tubulin polymerization by
arylthio/sulfonylindoles.
Various synthetic molecules have also been reported
as tubulin inhibitors. The majority of them inhibit the
binding of colchicine to tubulin and thereby act as
destabilizing agents. In recent years about 30 natural,
semisynthetic, or synthetic tubulin binding agents,
acting by either stabilizing or destabilizing mechanisms,
have been under preclinical/clinical investigation.2
* Corresponding author. Phone: +39 06 4991 3800. Fax: +39 06
491 491. E-mail: romano.silvestri@uniroma1.it.
† Universita` di Roma.
‡ National Cancer Insititute.
§ Cardiff University.
10.1021/jm049360d CCC: $27.50 © 2004 American Chemical Society
Published on Web 11/09/2004

Letters
Therefore, here we describe the synthesis of some
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 25 6121
Scheme 1a
novel 3-arylthio/sulfonylindole-2-carboxylates (12-21).
Several were found to be highly potent inhibitors of
tubulin polymerization.
3-Phenylthio-1H-indole (11) was synthesized by re-
acting indole (22) with 1,1′-diphenyldisulfide (29) in the
presence of NaH in anhydrous DMF, following the
procedure reported by Atkinson.15 By this method was
prepared in poor yield methyl 3-phenylthio-1H-indole-
2-carboxylate (12) starting from methyl indole-2-car-
boxylate (23).
To improve the yield, compound 12 was synthesized
by reacting ester 23 with N-phenyl-thiosuccinimide (30)
in the presence of boron trifluoride diethyl etherate, as
we previously reported for compound 15 starting from
24.16
By the same procedure were prepared methyl 3-[(4-
methoxyphenyl)thio]-5-chloro-1H-indole-2-carboxylate
(17) starting from methyl 5-chloro-1H-indole-2-carbox-
ylate (25) and N-[(4-methoxyphenyl)thio]succinimide
(31), and methyl 3-(phenylthio)-5-methoxy-1H-indole-
2-carboxylate (20) starting from methyl 5-methoxyin-
dole-2-carboxylate (26) and 30. Methyl 3-[(3,4,5-trimeth-
oxyphenyl)thio]-1H-indole-2-carboxylate (14) and methyl
3-[(3,4,5-trimethoxyphenyl)thio]-5-chloro-1H-indole-2-
carboxylate (18) were prepared by heating at 50 °C
indole-2-carboxylic acid (27) or its 5-chloro derivative
28 with the 1,1′-(3,4,5-trimethoxyphenyl)disulfide (32)
in the presence of NaH. The crude acids were then
transformed into the corresponding methyl esters by
reaction with trimethylsilyldiazomethane (TMSDM) at
room temperature for 30 min. 5-Methoxyindole 21 was
prepared following a two step procedure involving the
addition of 3,4,5-trimethoxylthiophenol (33) to a solution
of N-chlorosuccinimide (NCS) at -78 °C, and then this
mixture was treated with acid 26 while the reaction
temperature was warmed to 0 °C within 1 h. Oxidation
of sulfur derivatives 12, 15, and 18 with 3-chloroper-
oxybenzoic acid (MCPBA) furnished the corresponding
sulfones 13, 16, and 19, respectively (Scheme 1).
Inhibition of tubulin polymerization, colchicine bind-
ing,17 and the growth of MCF-7 human breast carcinoma
cells18 by the novel indoles 11-21 in comparison with
the effects of the reference compounds colchicine (1) and
CSA4 (6) are shown in Table 2.
The compound we initially evaluated was 3-phenyl-
thio-1H-indole (11), which inhibited tubulin polymeri-
zation with an IC50 value of 15 µM, ca. 5 and 7 times
inferior to the reference compounds 1 and 6, respec-
tively.
We therefore attempted the introduction of a meth-
oxycarbonyl function at position 2 of the indole ring.
This chemical modification produced methyl 3-(phen-
ylthio)-1H-indole-2-carboxylate (12), which was about
twice as potent as 11 (IC50 ) 8.2 µM).
A further step was the oxidation of the sulfur atom
of 12 to the sulfone to produce methyl 3-(phenylsul-
fonyl)-1H-indole-2-carboxylate (13). This compound was
found to be inactive at the highest concentration tested.
However, a different behavior was observed for the
analogous pair 15/16. Replacement of the 2-methoxy-
carbonyl group of 12 with an ethoxycarbonyl group (15)
doubled the inhibitory potency of the parent compound,
and the inhibitory activity of 15 was retained by the
a Reagents and reaction conditions: (a, 11) ArSSAr, NaH,
anhydrous DMF, rt, 2 h; (b, 12, 15, 17, 20) N-(ArS)succinimide,
BF3‚Et2O, anhydrous CH2Cl2, rt, 1.5 h, then 45 °C, 2 h; (c, 14, 18)
(i) ArSSAr, NaH, anhydrous DMF, 50 °C, overnight, anhydrous
nitrogen stream; (ii) TMSDM, CH3OH/CH2Cl2, rt, 30 min; (d, 21)
(i) ArSH, NCS, CH2Cl2, -78 °C; (ii) 26, -78 to 0 °C, 1 h; (e, 13,
16, 19) MCPBA (2.5 equiv), CHCl3, rt, 1 h.
Table 1. Structure of Compounds 11-21
compd
R1
R2
R3
R4
R5
S/SO2
11
12
13
14
15
16
17
18
19
20
21
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
S
COOCH3
COOCH3
COOCH3
COOC2H5
COOC2H5
COOCH3
COOCH3
COOCH3
COOCH3
COOCH3
S
SO2
S
S
SO2
S
S
SO2
S
S
OCH3 OCH3 OCH3
H
H
H
H
H
OCH3
H
H
H
Cl
OCH3 OCH3 OCH3 Cl
OCH3 OCH3 OCH3 Cl
H
H
H
OCH3
OCH3 OCH3 OCH3 OCH3
sulfone 16 (compare 12 with 15 and 16). Despite
appreciable inhibition of tubulin polymerization dis-
played by indoles 11, 12, 15, and 16, these compounds
were unable to inhibit the growth of MCF-7 human
breast carcinoma cells.
At this point we decided to introduce a trimethoxy
substitution pattern, a common structural feature of
colchicine, CSA4, and other tubulin inhibitors (see Chart
1). Replacement of the phenylthio of 12 with the 3,4,5-
trimethoxyphenylthio moiety furnished methyl 3-[(3,4,5-
trimethoxyphenyl)thio]-1H-indole-2-carboxylate (14),
which showed IC50 ) 2.9 µM, 2.8 times superior to that
of 12 and comparable with those of colchicine (IC50
)
3.2 µM) and CSA4 (IC50 ) 2.2 µM). Most importantly,
this compound inhibited the growth of the MCF-7 cells
by 50% at 25 nM, a concentration only 2 or 1.4 times
higher than observed with colchicine (IC50 ) 13 nM) or
CSA4 (IC50 ) 17 nM), respectively.
We also evaluated the effects of substituents at
position 5 of the indole ring. To this end we chose two
substituents with opposite properties, the electron
withdrawing chlorine atom and the electron donating
methoxy group. Introduction of a chlorine atom at
position 5 of the indole of 14 gave methyl 3-[(3,4,5-
trimethoxyphenyl)thio]-5-chloro-1H-indole-2-carboxyl-
ate (18), which only marginally increased potency as an

6122 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 25
Letters
Table 2. Inhibition of Tubulin Polymerization, Colchicine
Binding and Growth of MCF-7 Human Breast Carcinoma Cells
by Compounds 11-21
tubulina
IC50 ( SD
(µM)
colchicine
bindingb
(% ( SD)
MCF-7c
IC50 ( SD
(nM)
compd
11
12
13
14
15
16
17
18
19
20
21
15 ( 0.7
8.3 ( 0.6
>40
2.9 ( 0.1
4.4 ( 0.3
4.4 ( 0.2
>40
2.5 ( 0.3
>40
6.2 ( 0.3
2.0 ( 0.2
13 ( 5
>2500
>2500
>2500
25 ( 1
>1250
>1000
>2500
42 (10
>2500
>2500
13 ( 3
21( 7
5.1 ( 1
81 ( 1
19 ( 7
33 ( 0.6
3.6 ( 8
76 ( 0.2
1.6 ( 2
31 ( 8
93 ( 0.8
(68 ( 3)d
1 (colchicine)
4e
6 (CSA4)
3.2 ( 0.4
ndg
2.2 ( 0.2
13 ( 3
160
17 ( 10
nd
Figure 1. Proposed binding of 21. DAMA-colchicine is rep-
resented in yellow, compound 21 in green. In red are repre-
sented the residues of the X-ray structure of tubulin, in cyan
the same residues after minimization of tubulin with bound
21. Residue numbers are those used by Ravelli et al.19
100 ( 0.9
(88 ( 0.4)d
28 ( 8
7f
8f
4.1 ( 0.6
1.6
370 ( 2
45 ( 5
54 ( 0.7
a Inhibition of tubulin polymerization. Tubulin was at 10 µM.13
b Inhibition of [3H]colchicine binding. Tubulin was at 1 µM; both
[3H]colchicine and inhibitor were at 5 µM.14 c Inhibition of growth
of MCF-7 human breast carcinoma cells.14 d Inhibition of [3H]colch-
icine binding. [3H]Colchicine was at 5 µM; both tubulin and
inhibitor were at 1 µM. e Reference 6. f Reference 8. g No data.
ref 19), adopting an orientation very similar to that of
the corresponding ring in the DAMA-colchicine of the
crystallized structure (see also Figure 2 in the Support-
ing Information). The methoxy substituent of the indole
is also very close to the corresponding group on ring C
of colchicine, leading to a very similar general binding
of the two inhibitors. Furthermore, the indole moiety
establishes one hydrogen bond between the NH and the
backbone of Thr179 (shown in blue). Several other
hydrophobic contacts (not shown for the sake of clarity)
stabilize further the binding of 21 to the protein. These
observations are consistent with the highly efficient
inhibition of [3H]colchicine binding that occurs with
compound 21 (Table 2).
Further studies are currently underway to verify if
this model can give a rational justification of the SAR
obtained from the biological data.
In conclusion, we synthesized arylthioindoles, a novel
class of inhibitors of tubulin polymerization. Some
compounds had excellent activity in both inhibition of
tubulin polymerization and inhibition of the growth of
MCF-7 human breast carcinoma cells. Methyl 3-[(3,4,5-
trimethoxyphenyl)thio]-5-methoxy-1H-indole-2-carbox-
inhibitor of tubulin polymerization (IC50 ) 2.5 µM),
whereas inhibition of the growth of MCF-7 cells
(IC50 ) 42 nM) was decreased almost 2-fold. In contrast,
the presence of a methoxy group at position 5 of the
indole, giving methyl 3-[(3,4,5-trimethoxyphenyl)thio]-
5-methoxy-1H-indole-2-carboxylate (21), enhanced in-
hibitory potency in both the biochemical and cytological
assays. Compound 21 (IC50 ) 2.0 µM) was 1.6 times
more active than colchicine and about as active as CSA4
as an inhibitor of tubulin polymerization, while as an
inhibitor of the growth of MCF-7 cells (IC50 ) 13 nM) it
was as potent as colchicine and CSA4. Among the
derivatives tested, this compound was the most active.
It is interesting to note that, independently of the
presence/nature of the substituent at position 5 of the
indole, the 3,4,5-trimethoxyphenylthio moiety was cru-
cial for effective inhibition of the growth of MCF-7 cells
(compare 12 with 14, 17 with 18, and 20 with 21).
In contrast, inhibition of tubulin polymerization was
less affected by the substituent at position 5 (compare
12 with 14 and 20 with 21). Surprisingly, methyl 3-[(4-
methoxyphenyl)thio]-5-chloro-1H-indole-2-carboxylate
(17) was quite inactive in both assays.
To investigate the possible binding mode of these
novel compounds, we carried out docking studies of the
arylthioindole 21 in the colchicine binding site of tubu-
lin, which was obtained from the recently reported 3D
structure of tubulin cocrystallized with a colchicine
analogue: N-deacetyl-N-(2-mercaptoacetyl)colchicine
(DAMA-colchicine).19
Compound 21 was docked using a modified version
of the docking tool of MOE,20 which implements a
genetic algorithm based search method.21 The resulting
protein/ligand complex was then minimized, revealing
a very interesting and efficient binding model for the
inhibitor to tubulin.
ylate (21), the most potent derivative, showed IC50
)
2.0 µM, 1.6 times more active than colchicine and about
as active as CSA4 as an inhibitor of tubulin assembly.
This compound inhibited the growth of MCF-7 human
breast carcinoma cells with IC50 ) 13 nM, essentially
equivalent to the values obtained with colchicine and
CSA4.
A SAR study led to identification of several crucial
structural requirements needed to enhance the ef-
fectiveness of the arylthioindoles. In particular, we
found required determinants to be (i) the presence of a
methoxy/ethoxycarbonyl group at position 2 of the indole
nucleus; (ii) the 3,4,5-trimethoxyphenylthio group at
position 3 of the indole; (iii) the sulfur atom in the
sulfide oxidation state; (iv) the substituent at position
5 of the indole.
The last seems especially important for inhibition of
the growth of MCF-7 cells. Furthermore, initial molec-
ular modeling studies have revealed a possible binding
mode for these novel inhibitors that could support the
Figure 1 clearly shows how the trimethoxy ring is well
situated in proximity to Cys241 (residue numbers as in

Letters
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 25 6123
(11) Medarde, M.; Ramos, A.; Caballero, E.; Pela´z-Lamamie´ de
Clairac, R.; Lo´pez, J. L.; Garcia´ Gra´valos, D. San Felicia, A.
Synthesis and antineoplastic activity of combretastatin ana-
logue: heterocombretastatin. Eur. J. Med. Chem. 1998, 33, 71-
77.
(12) Flynn, B. L.; Hamel, E.; Jung, M. K. One-pot synthesis of
benzo[b]furan and indole inhibitors of tubulin polymerization.
J. Med. Chem. 2002, 45, 2670-2673.
(13) Yamamoto, K.; Noda, K.; Yoshimura, A.; Kukuoka, M.; Furuse,
K.; Niitani, H. Phase I study of E7010. Cancer Chem. Pharmacol.
1998, 42, 124-134.
(14) Flynn, B. L.; Flynn, G. P.; Hamel, E.; Jung, M. K. The synthesis
and tubulin binding activity of thiophene-based analogues of
combretastatin A-4. Bioorg. Med. Chem. Lett. 2001, 11, 2341-
2343.
(15) Atkinson, J. G.; Hamel, P.; Girard, Y. A. A new synthesis of
3-arylthioindoles. Synthesis 1988, 480-481.
(16) Silvestri, R.; De Martino, G.; La Regina, G.; Artico, M.; Massa,
S.; Vargiu, L.; Mura, M.; Loi, A. G.; Marceddu, T.; La Colla, P.
Novel indolyl aryl sulfones active against HIV-1 carrying NNRTI
resistance mutations: synthesis and SAR studies. J. Med. Chem.
2003, 46, 2482-2493.
(17) Hamel, E. Evaluation of antimitotic agents by quantitative
comparisons of their effects on the polymerization of purified
tubulin. Cell Biochem. Biophys. 2003, 38, 1-21.
(18) Verdier-Pinard, P.; Lai, J.-Y.; Yoo, H.-D.; Yu, J.; Marquez, B.;
Nagle, D. G.; Nambu, M.; White, J. D.; Falck, J. R.; Gerwick,
W. H.; Day, B. W.; Hamel, E. Structure-activity analysis of the
interaction of curacin A, the potent colchicine site antimitotic
agent, with tubulin and effects of analogs on the growth of
MCF-7 breast cancer cells. Mol. Pharmacol. 1998, 53, 62-76.
(19) Ravelli, R. B. G.; Gigant, B.; Curmi, P. A.; Jourdain, I.; Lachkar,
S.; Sobel, A.; Knossow, M. Insight into tubulin regulation from
a complex with colchicine and a stathmin-like domain. Nature
2004, 428, 198-202.
(20) Molecular Operating Environment (MOE 2004.03). Chemical
Computing Group, Inc., Montreal, Quebec, Canada. http://
www.chemcomp.com.
(21) Code “more_dock.svl” obtained from SLV Exchange website
http://svl.chemcomp.com., Chemical Computing Group, Inc.,
Montreal, Canada.
observed biological data. These findings have led us to
continue our studies on novel arylthioindoles in order
to investigate further SAR rules and mechanism of
action and to evaluate their activity against a wider
range of tumor cell lines.
Supporting Information Available: Experimental pro-
cedures for synthesis and biological evaluation of 11-21 and
for molecular modeling. This material is available free of
charge via the Internet at http://pubs.acs.org.
References
(1) Lin, M. C.; Ho, H. H.; Pettit, G. R.; Hamel, E. Antimitotic natural
products combretastatin A-4 and combretastatin A-2: studies
on the mechanism of their inhibition of the binding of colchicine
to tubulin. Biochemistry 1989, 28, 6984-6991.
(2) Beckers, T.; Mahboobi, S. Natural, semisynthetic and synthetic
microtubule inhibitors for cancer therapy. Drugs Future 2003,
28, 767-785 and references therein.
(3) Li, Q.; Sham, H. L. Discovery and development of antimitotic
agents that inhibit tubulin polymerization for treatment of
cancer. Expert Opin. Ther. Pat. 2002, 12, 1663-1702.
(4) Prinz, H. Recent advances in the field of tubulin polymerization
inhibitors. Expert Rev. Anticancer Ther. 2002, 2, 695-708.
(5) Checchi, P. M.; Netlles, J. H.; Zhou, J.; Snyder, P.; Joshi, H. C.
Microtubule-interacting drugs for cancer treatment. Trends
Pharmacol. Sci. 2003, 24, 361-365.
(6) Pohle, K.; Matthies, E.; Peters, J. E. On the tumor growth-
inhibiting action of colchicine. Arch. Geschwulstforsh. 1965, 25,
17-20.
(7) Schiff, P. B.; Fant J.; Horwitz, S. B. Promotion of microtubule
assembly in vitro by Taxol. Nature 1979, 277, 665-667.
(8) Bacher, G.; Beckers, T.; Emig, P.; Klenner, T.; Kutshert, B.;
Nickel, B. New small-molecule tubulin inhibitors. Pure Appl.
Chem. 2001, 73, 1459-1464.
(9) http://www.baxter-oncology.com/english/projects/index.html.
(10) Gastpar, R.; Goldbrunner, M.; Marko, D.; von Angere, E.
Methoxy-substituted 3-formyl-2-phenylindoles inhibit tubulin
polymerizazion. J. Med. Chem. 1998, 41, 4965-4972.
JM049360D