Design, synthesis and antiproliferative activity of tripentones: A new series of antitubulin agents

Article abstract of DOI:10.1016/S0960-894X(01)00403-6

Structure-activity relationship studies of a new series of tripentones (thieno[2,3-b]pyrrolizin-8-ones), led us to prepare several derivatives with antiproliferative activities. The most promising 3-(3-hydroxy-4-methoxyphenyl)thieno[2,3-b]pyrrolizin-8-one 20 (leukemia L1210, IC₅₀ = 15 nM) was shown to be a potent inhibitor of tubulin polymerization.

Full text of DOI:10.1016/S0960-894X(01)00403-6

Bioorganic & Medicinal Chemistry Letters 11 (2001) 2205–2208
Design, Synthesis and Antiproliferative Activity of Tripentones:
A New Series of Antitubulin Agents
Vincent Lisowski,a Cecile Enguehard,b Jean-Charles Lancelot,a
Daniel-Henri Caignard,c Stephanie Lambel,d Stephane Leonce,d
Alain Pierre,d Ghanem Atassi,d Pierre Renardc and Sylvain Raulta,*
a
´
Centre d’Etudes et de Recherche sur le Medicament de Normandie, UFR des Sciences Pharmaceutiques,
´
1, rue Vaubenard, 14032 Caen Cedex, France
bUFR des Sciences Pharmaceutiques, 31 avenue Monge 37200 Tours, France
c
´
ADIR, 1, rue Carle Hebert 92415 Courbevoie Cedex, France
dInstitut de Recherche Servier, 11 rue des Moulineaux, 92150Suresnes, France
Received 4 April 2001; revised 28 May 2001; accepted 7 June 2001
Abstract—Structure–activity relationship studies of a new series of tripentones (thieno[2,3-b]pyrrolizin-8-ones), led us to prepare
several derivatives with antiproliferative activities. The most promising 3-(3-hydroxy-4-methoxyphenyl)thieno[2,3-b]pyrrolizin-8-
one 20 (leukemia L1210, IC50=15 nM) was shown to be a potent inhibitor of tubulin polymerization. # 2001 Elsevier Science Ltd.
All rights reserved.
Introduction
of these anti-tubulin agents2 are natural products such
as paclitaxel,3 vinca alkaloids,4 colchicine5 and com-
Chemotherapy combined with surgery and/or radio-
therapy still has an important role to play in the ther-
apeutic approach of cancer. Cancer cells differ from
their normal counterparts in a number of biochemical
processes, particularly during the control of cell growth
and division. So the mitotic spindle represents an
attractive target for cancer therapy. Many natural and
synthetic substances are known to interfere with the
dynamic assembly of tubulin and prevent the formation
of microtubules which are essential for cellular integrity
and cell division.1 Currently, the most useful members
bretastatin A-46 (Fig. 1).
During our investigation towards the synthesis of new
potential antitumor agents, we synthesized a series of
substituted heterocyclic ketones based on the ‘tri-
pentone’ skeleton of thieno[2,3-b]pyrrolizin-8-ones I.7
The first tests of the National Cancer Institute led to the
discovery of four antiproliferative compounds 1–4 with
an aryl substituent in position 3. Among them, MR
16924 (NSC 676693) 4 (Fig. 2) displays a remarkable
cytostatic activity over all the tested lines with highest
Figure 1. Representative antimitotic compounds which interact with tubulin.
*Corresponding author. Tel.: +33-2-3193-6458; fax: +33-2-3193-1188; e-mail: rault@pharmacie.unicaen.fr
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00403-6

2206
V. Lisowski et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2205–2208
effect against leukemia, central nervous system, ovarian
and breast cancers. A selection of these results is pre-
sented in Table 1.
To improve this result and better characterize the phar-
macological profile of the series, a first structure–activ-
ity relationship studying the influence of the nature and
the position of substituents on the phenyl group in position
3 of the tricyclic system was performed. We focused on the
synthesis of several hydroxy and alkyloxy derivatives 4–21.
Chemistry
All these compounds were prepared according to the
general route described in Scheme 1, starting from sui-
table arylaldehydes. As an example, the product 20 was
synthesized from isovanillin in 12 steps.8
After a reaction of benzylation, the isovanillin gave the
arylacetonitrile compound 26 in good yields by,
Figure 2.
Table 1. Compounds 1–4: selected values from the NCI 60 tumoral cell lines panel (cytostatic log10 GI50 effects in M)
Cell type
Cell line
Leukemia
MOLT4
NSCL*
EKVX
Colon
CNS*
Melanoma
SK-MEL2
Ovarian
Renal
A-498
Prostate
DU-145
Breast
HCC2998
SF-295
OVCAR3
HS-578T
1 (NSC 683507)
2 (NSC 683509)
3 (NSC 683508)
4 (NSC 673693)
À4.9
À4.1
À4.6
À7.4
À5.3
À5.6
À4.6
À6.1
À4.7
À4.0
À4.6
À6.5
À4.7
À4.0
À4.6
À7.4
À4.9
À4.0
À4.5
À6.4
À4.8
À4.0
À4.3
À7.4
À5.8
À6.0
À5.8
À6.2
À4.6
À4.0
À4.1
À6.8
À5.3
À5.7
À4.7
À7.3
NSCL, non small cell lung; CNS, central nervous system.
Scheme 1. Reaction conditions and yields: (a) MeOH, K2CO3, benzyl bromide, reflux, 5 h, 98%; (b) MeOH, NaBH4, rt, 1 h, 90%; (c) dioxane,
SOCl2, rt, 1 h, 85%; (d) CH3CN, (C2H5)4NCN, rt, 24 h, 70%; (e) dry THF, ethylformate, NaH, reflux, 1 h, 90%; (f) DMF, sulfonylbenzene chloride,
rt, 2 h, 50%; (g) MeOH, methyl thioglycolate, NaH, reflux, 2 h, 65%; (h) dioxane, dimethoxyTHF, 4-chloropyridinium hydrochloride, reflux, 2 h,
85%; (i) pyrrolidine, reflux, 2 h 30 min, 80%; (j) (1) POCl3, 80 ꢀC, 1 h 30 min; (2) 10% NaOH, 50 ꢀC, 30 min, 60%; (k) HBr/AcOH 33%, rt, 30 min, 65%.

V. Lisowski et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2205–2208
2207
Table 2. In vitro antiproliferative activity and L1210 cell cycle effects of compounds 4–21
R1
IC50 (mM)
Effect on the cell cyclea
5
6
7
8
Hydrogen
Methyl
3,4,5-Trimethoxyphenyl
4-Hydroxyphenyl
4-Butoxyphenyl
> 100
> 100
36.4
30.1
29.6
29
28.1
26.7
18.9
11.7
5.2
NT
NT
NT
NT
NT
NT
NT
NT
9
10
11
12
13
14
15
16
17
18
19
4
3-Hydroxyphenyl
4-Propoxyphenyl
Phenyl
2-Methoxyphenyl
3,4-Dimethoxyphenyl
3-Methoxyphenyl
3,4-Methylenedioxyphenyl
2-Hydroxyphenyl
4-Ethoxyphenyl
3,4-Dihydroxyphenyl
4-Methoxyphenyl
3-Hydroxy-4-methoxyphenyl
4-Hydroxy-3-methoxyphenyl
75% G2+M+8 N at 5 mM
86% G2+M+8 N at 25 mM
94% G2+M+8 N at 25 mM
88% G2+M+8 N at 10 mM
72% G1 at 50 mM
88% G2+M+8 N at 1 mM
86% G2+M+8 N at 0.5 mM
83% G2+M+8 N at 0.5 mM
80% G2+M+8 N at 0.5 mM
86% G2+M+8 N at 25 mM
4.2
3.7
0.43
0.28
0.19
0.015
8
20
21
NT, not tested; IC50, drug concentration that inhibits cell growth by 50%.
aPercent of untreated cell in the phases of the cycle: 41% (G1); 28% (S); 24% (G2+M); 1% (8 N).
Table 3. Antiproliferative activity of 20 on selected tumoral cell lines
Cell line
L1210
0.015
P388
0.038
VCR-20
0.021
B16
DU145
0.043
H-69
0.036
KB 3-1
0.018
KB-A1
0.015
Ovcar-3
0.031
A-2780
0.041
IC50 (mM)
0.023
Results and Discussion
successively, a reduction, a chlorination and then a
cyanation reaction. By treatment with ethylformate and
sodium hydride, 26 formed with 90% yield the corre-
sponding enolate 27, which is protected by a sulfo-
nylbenzene group to give 28.9 This product, by
treatment with methylthioglycolate and sodium hydride
in tetrahydrofuran, afforded, according to Kirsch’s
cyclisation,10 the methyl 3-aminothiophene carboxylate
29. The amino function of the latter was then reacted
with dimethoxytetrahydrofuran in the presence of 4-
chloropyridinium hydrochloride in dioxane to give the
pyrrole 30 whose methyl carboxylate function by
refluxing in excess of pyrrolidine gave the corresponding
carboxamide 31. Cyclization was then performed by
action of phosphorus oxychloride to access to the tri-
cyclic ketone 32 with 60% yield. The last deprotective
step liberated the phenol function by action of a 33%
solution of bromhydric acid in glacial acetic acid to give
the tripentone 20.11
Compounds 4–19 were then evaluated in vitro for their
antiproliferative activity against the L1210 leukemia cell
line. The results expressed as IC50 are reported in Table
2 with their effects on the cell cycle.
Considering these results, it appears that some small
structural modifications are responsible for great varia-
tions of the IC50 values and also that the hydroxy and
methoxy series are not really comparable in terms of
SAR, but the size of substituents seems to be funda-
mental and optimal with methoxy in the para position.
Taking into account the fact that the two best com-
pounds were 4 and 19 derivatives, we investigated the
synthesis of ‘mixed’ compounds bearing in the 3 and 4
positions either hydroxy or methoxy substituents. The
result went beyond our expectations because one of the
isomers, the isovanillic one 20 (IC50=0.015 mM), was
about 1000-fold more active than the vanillic one 21
(IC50=8 mM) and was the best compound of the series,
10-fold more potent than the parent 4-methoxy deriva-
tive 4 (Fig. 3). We actually try to explain this enigmatic
difference in activity between 20 and 21. First X-ray
structures were solved for compounds 20 and 14 and it
seems that the orientation of the aryl ring could play a
key role in the binding interaction with tubulin.
This result for compound 20 was confirmed over a panel
of nine tumoral cell lines (Table 3). Flow cytometric
studies showed that L1210 cells treated with tripentones
Figure 3. Monomethylated analogues.

2208
V. Lisowski et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2205–2208
were arrested in the G2/M phases of the cell cycle, with
a significant percentage of cells having re-initiated a
cycle of DNA synthesis without cell division (8 N DNA
content). Compound 20 at 0.5 mM induced the accumu-
lation of 86% of cells at G2/M stage (66% of cells hav-
ing a DNA content >4 N chromosomes). This effect,
observed with the majority of tubulin-interacting
drugs,12 prompted us to perform an inhibitory poly-
merization tubulin test with deoxypodophyllotoxin as
internal reference (IC50=2.4 mM). The compound 20
altered the assembly reaction of microtubules with an
IC50=2.9 mM and without any depolymerization effect.
the National Cancer Institute, Bethesda, MD, USA for
in vitro studies.
References and Notes
1. Jordan, A.; Hadfield, J. M.; Lawrence, N. J.; McGown,
A. T. Med. Res. Rev. 1998, 18, 259.
2. Hamel, E. Med. Res. Rev. 1996, 16, 207.
3. Swindell, C. S.; Krauss, N. E.; Horwitz, S. B.; Ringel, I. J.
Med. Chem. 1991, 34, 1176.
4. Lavielle, G.; Hautefaye, P.; Schaeffer, C.; Boutin, J. A.;
Cudennec, C. A.; Pierre, A. J. Med. Chem. 1991, 34, 1998.
5. Brossi, A.; Yeh, H. J. C.; Chryanowska, M.; Wolf, J.;
Hamel, E.; Lin, C. H.; Quin, F.; Suffness, M.; Silverton, ?.
Med. Res. Rev. 1988, 8, 77.
6. Hadfield, J. A.; Mc Gown, A. T. Synth. Commun. 1998, 28,
1421.
7. Lancelot, J. C.; Letois, B.; Rault, S.; Robba, M. J. Hetero-
cycl. Chem. 1994, 31, 501.
8. Rault, S.; Enguehard, C.; Lancelot, J. C.; Robba, M.;
Atassi, G.; Pierre, A.; Caignard, D. H.; Renard, P. Fr. Appl.
98.09552, 1998, Eur. Patent 0982308, 2000, Jpn. Kokai Tok-
kyo Koho 044572, 2000
9. Jourdan, F.; Laduree, D.; Robba, M. J. Heterocycl. Chem.
1994, 31, 305.
Conclusion
We have discovered a new type of antimitotic com-
pounds based on the arylthienopyrrolizinone molecular
skeleton. The most promising compound of the series, 3-
(3-hydroxy-4-methoxyphenyl)thieno[2,3-b]pyrrolizin-8-
one 20 has significant human cancer cell growth inhibi-
tory activity in the nanomolar range and interacts with
tubulin in the micromolar range. At present, we are
trying to specify the type of binding site on the protein
with which tripentones interact, and to determine if
this mechanism is responsible for all the anti-
proliferative activity. In vivo studies are also currently
under progress.
10. Kirsch, G.; Cagniant, D.; Cagniant, P. J. Heterocycl.
Chem. 1982, 19, 443.
11. Selected data for compound 20: IR (KBr) n 3419, 1671,
1504, 1210 cmÀ1 1H NMR (CDCl3) 3.95 (s, 3H), 5.87 (bs,
.
1H), 6.01 (m, 1H), 6.67 (m, 1H), 6.81 (m, 1H), 6.93 (d,
J=7.8 Hz), 6.99 (d, J=7.8 Hz), 7.07 (s, 1H), 7.43 (s, 1H). 13C
NMR (CDCl3) 56.1; 110.8; 113.2; 114.1; 115.6; 119.7; 121.0;
125.5; 129.4; 134.1; 135.8; 146.1; 146.9; 149.5; 150.7; 174.4.
Anal. calcd for C16H11NO3S: C, 64.63; H, 3.72; N, 4.71;
found: C, 64.64; H, 4.02; N, 4.48. All compounds were fully
characterized by spectroscopic and elemental analysis.
12. Leoni, M. L.; Hamel, E.; Genini, D.; Shih, H.; Carrera,
C. J.; Cottam, H. B.; Carson, D. A. J. Natl. Cancer Inst. 2000,
92, 217.
Acknowledgements
The authors thank D. Guenard from the Institut des
Substances Naturelles, Gif sur Yvette, for inhibitory
polymerization and depolymerization tubulin tests and