A new family of quinoline and quinoxaline analogues of combretastatins

Article abstract of DOI:10.1016/j.bmcl.2004.04.098

The 3-hydroxy-4-methoxyphenyl ring of combretastatin A-4 can be replaced by a 2-naphthyl moiety without significant loss of cytotoxicity and inhibition of tubulin polymerization potency. In this paper we show that the 6- or 7-quinolyl systems can in turn replace both cyclic moieties, keeping in the first case most of the potency as cytotoxic agent and in the second case as inhibitor of tubulin polymerization, related to the activities displayed by model compounds.

Full text of DOI:10.1016/j.bmcl.2004.04.098

Bioorganic & Medicinal Chemistry Letters 14 (2004) 3771–3774
A new family of quinoline and quinoxaline analogues of
combretastatins
Concepci oꢀ n P ꢀe rez-Melero, Ana B. S. Maya, Benedicto del Rey, Rafael Pel ꢀa ez,
Esther Caballero and Manuel Medarde*
Laboratorio de Qu ꢀı mica Org ꢀa nica y Farmac ꢀe utica, Facultad de Farmacia, Universidad de Salamanca, Campus Miguel de Unamuno,
3
7007 Salamanca, Spain
Received 18 November 2003; revised 22 April 2004; accepted 29 April 2004
Available online 25 May 2004
Abstract—The 3-hydroxy-4-methoxyphenyl ring of combretastatin A-4 can be replaced by a 2-naphthyl moiety without significant
loss of cytotoxicity and inhibition of tubulin polymerization potency. In this paper we show that the 6- or 7-quinolyl systems can in
turn replace both cyclic moieties, keeping in the first case most of the potency as cytotoxic agent and in the second case as inhibitor
of tubulin polymerization, related to the activities displayed by model compounds.
Ó 2004 Elsevier Ltd. All rights reserved.
Microtubules generated by polymerization of tubulin
a,b-dimers are the main constituents of the mitotic
spindle, whose formation and activity are required for
chromosome segregation during mitosis and for other
to be active (X ¼ H, OH, NH and their aminoacyl,
2
phosphate or other derivatives for solubilizing pur-
poses). It has been reported that lipophilicity is an
important feature for ligand binding, showing a good
1
8
cell processes. Natural and synthetic compounds of
correlation between log P and binding to tubulin, al-
though there are also contradictory reports. We have
9
varied structures have been shown to inhibit tubulin
polymerization through interaction with the protein at
different binding sites; the colchicine, taxol and Vinca
recently shown that one of the oxygenated A or B
phenyl rings of CS A-4 can be replaced by a naphthalene
2
10
alkaloids sites are the best known of them. Among the
3
moiety keeping high potency, whereas substitution of
ligands of the colchicine binding site, combretastatins
4
both A and B rings leads to an inactive compound.
Following these results, we proceeded further in order to
determine which one of the CS A-4 rings is actually best
replaced by the naphthalene in the CS A-4 analogues.
We then prepared different families of compounds that
were tested for cytotoxicity and inhibition of tubulin
are highlighted as strongly cytotoxic and antiangio-
5
genic agents, combretastatin A-4 (CS A-4) being the
6
most potent member of this family (Fig. 1). The main
problem associated with this class of compounds is their
low aqueous solubility. Therefore, efforts have been
directed at obtaining soluble derivatives, leading to
11
polymerization, leading to the conclusion that the
naphthalene system is a good surrogate for the 3-hy-
droxy-4-methoxyphenyl ring B, producing much less
potent derivatives when the naphthalene replaces the
trimethoxyphenyl ring A (Fig. 1).
7
combretastatin A-4 phosphate prodrug, which cur-
rently is in phase II clinical trials.
8;9
From different SAR studies on combretastatins it was
deduced that a 3,4,5-trimethoxyphenyl ring and a
4-methoxy-3-X-substituted phenyl ring systems, sepa-
rated by a two atoms bridge in a cis disposition, is the
common structural characteristic for these compounds
Even though the introduction of a naphthalene system
keeps most of the potency of combretastatins and
increases their lipophilicity, their low aqueous solubility
is a main drawback in order to assay these compounds
in vivo. To extend our studies to other bicyclic systems
as surrogates of the B ring of CS A-4 and to increase
their poor solubility, we have designed a new family
of analogues replacing the naphthalene moiety by
a quinoline or quinoxaline system and keeping the
Keywords: Combretastatins; Quinolines; Tubulin polymerization
inhibition.
*
Corresponding author. Tel.: +34-923-294528; fax: +34-923-294515;
e-mail: medarde@usal.es
0
960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.04.098

3772
C. P ꢀe rez-Melero et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3771–3774
Figure 1. Structure of combretastatin A-4, naphthylcombretastatin and the new quinoline analogues of combretastatins.
1
3
3
,4,5-trimethoxyphenyl moiety as ring A (Fig. 1). These
methyl analogues) in refluxing ethanol and acetic acid.
The resulting quinoxalinecarboxylic acid was then re-
heterocycles maintain the same size as the original
naphthalene, so steric factors would not affect the
binding to the protein. However, the presence of one or
two nitrogen atoms allows these compounds or their
salts to be more water soluble, so they could be assayed
in aqueous solution. The effect of the position of the
nitrogen atoms on the activity and/or potency of the
analogues would also contribute to enlarge the SAR for
these compounds.
duced to the primary alcohol with LiAlH and subse-
4
quently oxidized to the aldehyde with pyridinium
chlorochromate in CH Cl . The Wittig methodology
2 2
aforementioned was employed for synthesizing the qui-
noxaline analogues of combretastatins.
The aqueous solubility of all the Z-quinolines, 1a–5a, is
3- to 7-fold higher than that of the model naphthyl-
combretastatin, whereas the solubility of the quinoxa-
line derivative 7a is about the same. The most water-
soluble compound 3a has even a higher solubility than
CS A-4.
14
The subfamily of quinoline derivatives (I) was prepared
following the methodology described for the naphthyl-
combretastatins
10;11
(Scheme 1), by Wittig reactions of
the corresponding quinoline carbaldehydes with 3,4,5-
trimethoxybenzyltriphenylphosphorane, obtained in
turn by treatment of the phosphonium bromide with n-
BuLi. The quinoline carbaldehydes were prepared from
the corresponding methylquinolines by oxidation with
The cytotoxic activity of all these compounds was as-
sayed against different tumour cell lines as described.
15
The results are presented in Table 1. The cis isomers are
more potent than the corresponding trans isomers, as
usually observed for all the combretastatins and their
analogues, even though the difference in potency is lower
than in the case of the most potent combretastatins.
12
selenium dioxide in refluxing xylene. The Wittig reac-
tion afforded the desired compounds as a mixture of Z
and E isomers (variable ratios from 1:2 to 2:1), which
were separated by chromatography on silica gel. The
stereochemistry of the isomers was determined as pre-
Among quinoline-substituted compounds, those bearing
the nitrogen atom on the nonattached ring (4 and 5)
show higher cytotoxic potency than the ones with the
nitrogen atom on the ring directly attached to the eth-
ylene (1, 2 and 3). This fact is in agreement with the
requirement of a benzene ring directly attached to the
ethylene bridge for high potency, supporting our previ-
10
viously reported.
For the synthesis of the quinoxaline subfamily (II,
Scheme 2), 3,4-diaminobenzoic acid was used as starting
material. It was condensed with glyoxal (for the plain
quinoxaline derivatives) or butanedione (for the 2,3-di-
Scheme 1. Reagents and conditions: (i) NaBH
4
, MeOH (>95%); (ii) PBr
3
, Et
2
3 2
O, ꢀ40 °C (97%); (iii) PPh , toluene (98%); (iv) SeO , xylene, reflux
(50–80%); (v) n-BuLi, THF, ꢀ78 °C, then aldehyde, THF, ꢀ20 °C (60–85%).

C. P ꢀe rez-Melero et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3771–3774
3773
Scheme 2. Reagents and conditions: (i) glyoxal, AcOH, EtOH, reflux (31%); (ii) butanedione, AcOH, EtOH, reflux (52%); (iii) LiAlH
4
, THF (75%);
(iv) CCP, DCM (30%); (v) 3,4,5-trimethoxybenzyl bromide, THF, n-BuLi, ꢀ78 °C, then aldehyde, THF, ꢀ20 °C (55–60%).
Table 1. Cytotoxic activity of the compounds assayed expressed as the
compound is not the inhibition of microtubule forma-
tion through binding to tubulin, but a different one
leading nonetheless to a highly cytotoxic effect. Thus,
compound 4a is a good candidate to study the mecha-
nism of action of the analogues of combretastatins that
have shown unexpected low potency as inhibitors of
tubulin polymerization related to their cytotoxicity.
ꢀlog (IC50) M
MeO
MeO
OMe
4a
N
Compound
P-388 A-549 HT-29 MEL-28 H116
However, the second most potent analogue, 7-quinolyl
derivative 5a, inhibits tubulin polymerization with an
IC50 20 lM, close to the 12 lM measured for naphthyl-
combretastatin. In this respect, the decrease in cyto-
toxicity parallels that in polymerization inhibition, as
usually observed for the most potent antimitotic agents
of this type. Compound 5a mimics the structural rigidity
of the naphthalene moiety and also includes a nitrogen
atom at the same relative position as the highly potent
Combretastatin A-4
Naphthylcombretastatin
8.5
7.8
8.5
7.8
7.5
7.8
8.5
7.8
8.5
7.8
I
1a
1b
2a
2b
3a
4a
5a
5b
6.5
6.4
6.4
6.5
6.4
6.4
6.5
6.4
6.4
6.5
6.4
6.5
6.4
6.4
6.4
<5.5 <5.5 <5.5
<5.5
5.1
<5.5
5.1
5.1
7.5
6.8
5.8
5.1
7.5
6.8
5.8
4.8
7.5
6.8
5.8
0
17
3
-amine analogue of CS A-4.
7.5
6.8
7.5
6.8
The other compounds tested for inhibition of tubulin
polymerization did not reveal any significant activity, in
accordance with their lower cytotoxicity. None of them
inhibited more than 20% of the polymerization activity
of the control at 40 lM. Compounds 6a and 7a did not
display any polymerization inhibitory activity at even
5.8
5.8
II
6
6
7
7
a
5.1
5.1
5.1
5.1
<4.8
<4.8
<4.8
5.1
<4.8
<4.8
<4.8
b
a
b
<4.8 <4.8 <4.8
<4.8 <4.8 <4.8
<4.8 <4.8 <4.8
60 lM.
In summary, we have synthesized a new family of ana-
logues of combretastatin A-4 bearing a quinoline or
quinoxaline system replacing the 3-hydroxy-4-meth-
oxyphenyl ring of combretastatin A-4 or the naphthyl
moiety of our previously described naphthylcombre-
tastatin. The quinoline derivatives showed an increased
aqueous solubility related to the parent naphthylcom-
bretastatin. Only the quinoline derivatives in which the
nitrogen atom is on the nonattached ring elicited rele-
vant cytotoxic activity. Such cytotoxicity correlated with
an inhibitory effect on tubulin polymerization only in
the case of 5a. Thus, the introduction of quinoline
moieties bonded through the non-nitrogenated ring
seems to be a convenient structural modification for
antimitotic agents acting at the colchicine binding site,
as it is the case with the related naphthyl system. Com-
pound 4a, highly cytotoxic but very weak inhibitor of
tubulin polymerization, could be very useful for the elu-
cidation of the mechanism of action of combretastatin
ous results with other analogues, for example, 3- and 5-
11
indolyl derivatives. Quinoxaline analogues display
lower potency than their quinoline counterparts, indi-
cating that an additional nitrogen atom is detrimental
for activity.
The more potent cis isomers were also assayed for
inhibition of tubulin polymerization, following the
11;16
methodology described in the literature,
in order to
know the correlation between their cytotoxic activity
and their interaction with the microtubule system. The
results are shown in Figure 2.
The 6-quinolyl derivative 4a, that showed the highest
cytotoxic potency, unexpectedly displays a very weak
potency as inhibitor of tubulin polymerization. This
would mean that the actual mechanism of action for this

3774
C. P ꢀe rez-Melero et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3771–3774
Figure 2. Inhibition of tubulin polymerization (as microtubule protein, MTP) by compounds 1a–7a, CS A-4 and naphthylcombretastatin, expressed
as percentage of polymerization for noninhibited protein. MTP concentration is 1.5 mg/mL.
analogues with activities that do not fit into the SAR for
these type of compounds.
8. Nandy, P.; Banerjee, S.; Gao, H.; Hui, M. B.; Lien, E. J.
Pharm. Res. 1991, 8, 776.
9
. Brown, M. L.; Rieger, J. M.; Macdonald, T. L. Bioorg.
Med. Chem. 2000, 8, 1433, and references cited therein.
1
1
1
0. Maya, A. B. S.; del Rey, B.; Pel ꢀa ez Lamami ꢀe de Clairac,
R.; Caballero, E.; Barasoain, I.; Andreu, J. M.; Medarde,
M. Bioorg. Med. Chem. Lett. 2000, 10, 2549.
1. Maya, A. B. S.; P ꢀe rez-Melero, C.; Mateo, C.; Alonso, D.;
Fern ꢀa ndez, J. L.; Pel ꢀa ez, R.; Caballero, E.; Medarde, M.
J. Med. Chem., in press.
Acknowledgements
Financial support came from Spanish DGI-MCyT
PPQ2001-1977), Junta de Castilla y Le oꢀ n (SA073/02)
(
2. Plobek, N.; Delorme, D.; Wei, Z.-Y.; Yang, H.; Zhou, F.;
Schwarz, P.; Gawell, L.; Gagnon, H., et al. J. Med. Chem.
and european FEDER funds. A. B. S. Maya thanks
Universidad de Salamanca for a predoctoral grant.
Cytotoxicity assays were performed by Instituto Biomar
S. A. (Le oꢀ n, Spain). The authors are grateful to CIDTA
for the use of their equipment.
2
000, 43, 3878.
3. Deady, L. W.; Desneves, J.; Ross, A. C. Tetrahedron 1993,
9, 9823.
1
1
4
4. For the determination of aqueous solubility, samples
were shaken in 10 mM phosphate buffer, pH 7.0, at room
temperature for 72 h. Next, they were filtered and their
absorbance in the UV–vis was measured. Solubility: mg/
mL is 1a: 0.04, 3a: 0.07, 4a: 0.03, 5a: 0.04, 7a: 0.01,
naphthylcombretastatin: 0.01 and CS A-4: 0.05. (The
value reported in the literature for CS A-4 is 0.11 mg/mL
in PBS; Ohsumi, K.; Hatanaka, T.; Nakagawa, R.;
Fukuda, Y.; Morinaga, Y.; Suga, Y.; Nihei, Y.; Ohishi,
K.; Akiyama, Y.; Tsuji, T. Anticancer Drug Des. 1999, 14,
539).
15. San Feliciano, A.; Medarde, M.; Pel ꢀa ez Lamami ꢀe de
Clairac, R.; L oꢀ pez, J. L.; Puebla, P.; Gr ꢀa valos, M. D. G.;
Ruiz, P.; Garc ꢀı a de Quesada, M. T. Arch. Pharm.
(Weinheim) 1993, 326, 421.
16. Dumortier, C.; Gorbunoff, M.; Andreu, J. M.; Enge-
lborghs, Y. Biochemistry 1996, 35, 4378–4395.
References and notes
1
2
3
4
5
6
7
. Microtubules; Hyams, J., Lloyd, C. W., Eds.; Wiley-Liss:
New York, 1994.
. Jordan, A.; Hadfield, J. A.; Lawrence, N. J.; McGown, A.
T. Med. Res. Rev. 1998, 18, 259.
. Pettit, G. R.; Cragg, G. M.; Herald, D. L.; Schmidt, J. M.;
Lohavanijaya, P. Can. J. Chem. 1982, 60, 1374.
. Pettit, G. R.; Singh, S. B.; Boyd, M. R.; Hamel, E.; Pettit, R.
K.; Schmidt, J. M.; Hogan, F. J. Med. Chem. 1995, 38, 1666.
. Dark, G. G.; Hill, S. A.; Prise, V. E.; Tozer, G. M.; Pettit,
G. R.; Chaplin, D. J. Cancer Res. 1997, 57, 1829.
. Shi, Q.; Chen, K.; Morris-Natschke, S. L.; Lee, K. H.
Curr. Pharm. Des. 1998, 4, 219.
. Pettit, G. R.; Temple, C., Jr.; Narayanan, V. L.; Varma,
R.; Simpson, M. J.; Boyd, M. R.; Rener, G. A.; Bansal, N.
Anticancer Drug Des. 1995, 10, 299.
17. Ohsumi, K.; Nakagawa, R.; Fukuda, Y.; Hatanaka, T.;
Morinaga, Y.; Nihei, Y.; Ohishi, K.; Suga, Y.; Akiyama,
Y.; Tsuji, T. J. Med. Chem. 1998, 41, 3022.