Preliminary structure-antiangiogenic activity relationships of 4-senecioyloxymethyl-6,7-dimethoxycoumarin.

Article abstract of DOI:10.1016/S0960-894X(02)00392-X

Through a systematic modification of the novel angiogenesis inhibitor 4-senecioyloxymethyl-6,7-dimethoxycoumarin (1) we found that a 6,7-dimethoxy moiety is important for bioactivity of 1. Replacement of the lactone functionality in coumarin 1 by an amide decreased its activity. By substitution of the senecioyl chain with various cinnamoyl groups we discovered 6d, bearing a 4-methoxycinnamoyl instead of senecioyl side chain, with inhibitory activity in HUVEC tube formation assay enhanced by one order of magnitude compared to 1. We have also synthesized compound 12, an analogue of 6d, with equipotency and improved water solubility.

Full text of DOI:10.1016/S0960-894X(02)00392-X

Bioorganic & Medicinal Chemistry Letters 12 (2002) 2345–2348
Preliminary Structure–Antiangiogenic Activity Relationships
of 4-Senecioyloxymethyl-6,7-dimethoxycoumarin
Nguyen-Hai Nam,a Yong Kim,a Young-Jae You,a Dong-Ho Hong,a,b
Hwan-Mook Kimb and Byung-Zun Ahna,*
aCollege of Pharmacy, Chungnam National University, Taejon 305-764, Republic of Korea
bResearch Institute of Biosciences and Biotechnology, Taejon 305-600, Republic of Korea
Received 22 April 2002; accepted 29 May 2002
Abstract—Through a systematic modification of the novel angiogenesis inhibitor 4-senecioyloxymethyl-6,7-dimethoxycoumarin (1)
we found that a 6,7-dimethoxy moiety is important for bioactivity of 1. Replacement of the lactone functionality in coumarin 1 by
an amide decreased its activity. By substitution of the senecioyl chain with various cinnamoyl groups we discovered 6d, bearing a
4-methoxycinnamoyl instead of senecioyl side chain, with inhibitory activity in HUVEC tube formation assay enhanced by one
order of magnitude compared to 1. We have also synthesized compound 12, an analogue of 6d, with equipotency and improved water
solubility. # 2002 Elsevier Science Ltd. All rights reserved.
Introduction
venous endothelial cells (HUVEC) while manifesting a
marginal effect on tumor cells.27 Thus, 1 possesses a
potential for further development as an antiangiogenic
agent. As already widely known, angiogenesis inhibitors
represent an important new class of anticancer agents.
Coumarins constitute a class of compounds which are
found widely in nature and possess diverse biological
activities, for example, antimicrobial,1ꢀ5 antifungal,6ꢀ8
antiviral9ꢀ14 including human immunodeficiency virus
(HIV).10ꢀ14 In cancer drug development arena, cou-
marin-type compounds have attracted increasing inter-
ests recently. Several studies reported a number of
coumarins, natural15ꢀ18 or synthetic,19ꢀ26 with marked
cytotoxic activities.
From this perspective, we have initiated a study on the
structure–activity relationshipof this molecule. In this
paper, the preliminary results from this study are
presented and discussed.
In our own works, we have previously reported the iso-
lation of a novel metabolite, namely 4-senecioyloxy-
methyl-6,7-dimethoxycoumarin (1, Scheme 1), from the
plant Crinum latifolium. This compound was shown to
exhibit significant cytotoxicity against human umbilical
Chemistry
Several series of simple analogues of 1 were synthesized
as shown in Schemes 1–4. A series of 6- and/or 7- sub-
stituted 4-senecioyloxymethylcoumarins (5) were
Scheme 1. (a) MeOH/H2O2 (1:1), H2SO2 (cat.), rt, 48 h; (b) ClCH2COCH2CO2C2H5, H2SO2 (excess), 0 ꢁC, 2 h; (c) senecioic acid (excess), CH3CN,
rt, 30 min.
*Corresponding author. Tel.: +82-42-821-5923; fax: +82-42-823-6566; e-mail: ahnbj@cnu.ac.kr
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00392-X

2346
N.-H. Nam et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2345–2348
Scheme 2. (a) Carboxylic acid (excess), CH3CN, TEA, rt, 30 min. (b) (i) NaH, NaI, DMF, rt, 3 h; (ii) 4 N HCl/dioxane, 1 h.
Scheme 3. (a) CH3COCH2CO2C2H5, H2SO2 (excess), 0 ꢁC, 2 h; (b) N-bromosuccinimide (1 mol equiv), benzene, reflux, 6 h; (c) 4-methoxycinnamic
acid (excess), CH3CN, TEA, rt, 30 min.
Scheme 4. (a) CH3CN, TEA, rt, 30 min; (b) 40% HBr/AcOH.
obtained simply by reaction of senecioic acid with 4
which were reached from appropriate phenols 3 and
ethyl 4-chloroacetoacetate by using Pechman condensa-
tion conditions, thus involving stirring of phenol and
ethyl 4-chloroacetoacetate in neat sulfuric acid at room
temperature for 12–16 h (Scheme 1). The phenols 3 were
achieved by acid-catalyzed oxidation of the aldehydes 2
with hydrogen peroxide using a protocol developed by
Matsumoto and co-workers.28 This reaction proved to
be superior to Bayer–Villiger reaction, as separation
from by-products was much more convenient.
The final analogue 12 was synthesized by reacting 4f with
11, which was synthesized as described in the litera-
ture.29 All compounds were confirmed straightforwardly
1
by spectroscopic methods (MS, IR and H NMR).30
Biological Results and Discussion
Attempts to discern structural features essential for the
antiangiogenic activity of compound 1 were first direc-
ted on the substituent pattern on ring B of the coumarin
skeleton. Accordingly, a series of compounds 5 were
synthesized. The synthesized compounds 5a–e and the
parent compound 1 were examined for cytotoxicity in
two tumor cell lines, namely B16 and HCT116.31 It
was revealed that all compounds assayed showed no
significant cytotoxic activity in both cell lines tested
(Table 1). To evaluate the antiangiogenic property of
the compounds in this series, all compounds were
subjected to in vitro HUVEC assay31 at 3 mg/mL. The
results summarized as inhibition percentages in Table 1
revealed that removal of a methoxy groupat positions
6 and 7 was detrimental for antiangiogenic activity.
Compound 5a bearing a 6-chloro substituent was also
inactive in the HUVEC assay model. Contraction of the
6,7-dimethoxy groupto the 6,7-methylenedioxy sub-
stituent (5d) led to a decrease in bioactivity. Meanwhile,
extension to one additional benzene moiety at 6,7-posi-
tion brought about a similar result with compound 5e
being less potent than 1. Though other substituents
have not yet been examined, at this stage, it could be
suggested that the 6,7-dimethoxy substituent is impor-
tant for the antiangiogenic property of compound 1.
Therefore, it was maintained throughout the next phase
of this work.
Analogues with the common structure 6 were obtained
simply by reaction of various carboxylic acids, for
example, acrylic acid, tigloic acid, substituted cinnamic
acids, with 4-chloromethyl-6,7-dimethoxycoumarin (4f)
synthesized above. Nucleophilic displacement of the
chlorine in 2-(dimethylamino)ethyl chloride with 6g led
to 6j (Scheme 2).
The quinoline analogue 10 of 6g was explored through a
sequence shown in Scheme 3. The Pechman condensa-
tion of 4-aminoveratrole (7) with ethyl acetoacetate
afforded the intermediate 8 (6,7-dimethoxy-4-methyl-
quinolin-2-one). When ethyl 4-chloroacetoacetate was
used, a complex mixture of products was formed, prob-
ably due to the reaction of the amino groupwith the
chloro moiety also occurring. The intermediate 8 was
transferred into 9 in 67% yield by a reaction with
N-bromosuccinimide in a molar ratio of 1:1 in anhydrous
benzene at reflux temperature for 6 h. If excess N-bromo-
succinimide was used, 4-dibromomethyl-6,7-dimeth-
oxyquinolin-2-one was also obtained as a side product.
The target compound 10 was furnished simply by a
reaction of 9 with 4-methoxycinnamic acid as described
above.

N.-H. Nam et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2345–2348
Table 1. Cytotoxicity and antiangiogenic activity of synthesized compounds
2347
Compd
R in 5 and 6
Cytotoxicitya (IC50
,
b mg/mL)
Activity in the in vitro HUVEC assay
B16
HCT116
Concentrationc
Inhibitory percentaged
1
6,7-(OCH3)
>10
>10
>10
>10
>10
>10
>10
>10
>10
0.96
>10
>10
>10
>10
>10
>10
>10
>10
>10
1.02
3
3
3
3
3
3
3
3
86.5
e
2
5a
5b
5c
5d
5e
6a
6b
6c
6d
6-Cl
7-OCH3
6-OCH3
—
—
—
43.2
62.5
—
6,7-(OCH2O)—
6,7-CH¼CH–CH¼CH—
–H
–CH3
–C6H5
–(4-OCH3)C6H4
—
3
1 (toxic)
0.3
0.1
3
3
3
3 (toxic)
3
3
3
0.3
12.0
100.0
81.7
47.5
18.7
—
19.7
35.5
—
HCl
6e
6f
6g
6h
6i
6j
10
–(3,4,5-OCH3)3C6H2
–(4-Cl)C6H4
–(4-OH)C6H4
–(3-OH-4-OCH3)C6H3
–(4-NH2)C6H4
6.74
>10
4.15
8.31
>10
5.21
3.41
>10
>10
>10
2.43
#
3.00
>10
>10
>10
1.27
.
–[4-O(CH2)N(CH3)2]C6H4 HCl
36.5
46.4
78.8
12
Suraminf
Adriamycin
g
#
0.09
30 mM
0.11
#
aMeasured in tumor cells: B16, murine melanoma; HCT116, human colon.
bThe concentration (g/mL) that produces a 50% reduction in cell growth.
cNon-cytotoxic concentration in both B16 and HCT116 cell lines, unless otherwise noted.
dCalculated as described in ref 31. The numbers represent the averaged results from triplicate experiments with deviation of less than 10%.
eInactive (inhibition percentage was less than 10%).
fAn established angiogenesis inhibitor.
gNot assayed.
The next phase of structural modifications carried out
on 1 was focused on the senecioyl moiety. We initially
replaced this moiety by acryloyl and tigloyl groups and
found that the corresponding compounds 6a,b were
completely inactive. Compound 6c, resulting from the
replacement of the senecioyl chain by a cinnamoyl, was
also found to be inactive. We continued our attempts
with some substituted cinnamoyl chains and obtained
compound 6d, bearing a 4-methoxycinnamoyl side
chain, with much enhanced bioactivity. This compound
showed cytotoxicity against tumor cell lines (B16 and
HCT116) with IC50 values of 0.96 and 1.02 mg/mL,
respectively. Thus, the cytotoxicity of this compound
was increased by more than one order of magnitude. In
parallel, its inhibitory effects in HUVEC tube formation
assays were also found to be enhanced by nearly 10
times; at 0.3 mg/mL, 6d showed inhibitory percentage of
81.7%, almost comparable to that of 1 at 3 mg/mL. A
higher concentration of this compound (1 mg/mL) com-
pletely inhibited the formation of tube network by
HUVECs; however, this concentration was toxic in
tumor cells. A lower concentration of 6d (0.1 mg/mL)
exhibited considerable activity with an inhibitory per-
centage of 47.5%.
The quinoline 10 was found to be less potent than 1.
Thus, it appears that the lactone functionality in cou-
marin 1 is important for the binding of the molecule to
receptors, probably through hydrogen bonding.
At this stage, we have identified compound 6d as the
most potent one. However, this compound showed very
poor solubility in aqueous systems which rendered the
in vivo evaluation difficult. We further attempted to
look for analogues with improved water solubility.
Initially we synthesized compound 6j with expectation
that this compound might retain bioactivity of 6d and in
the meantime offer reasonable water solubility. Unfor-
tunately, however, 6j was shown to be much less active
than 6d, probably due the fact that the 4-substituent in
6j is too bulky and not tolerable for bioactivity. Finally,
we came upwith the design of 12. In this compound, the
pyridinyl moiety should much resemble the phenyl moiety
in 6d. Thus, the two compounds are expected to be similar
in conformation. Meanwhile, though we have not eval-
uated in detail, it is obvious that the water solubility of 12
in the hydrobromide salt form should be much enhanced
compared to 6d. Interestingly, compound 12 was found to
be almost equipotent with 6d in HUVEC assay, though it
was slightly less active towards tumor cells.
Replacement of the 4-methoxy group in 6d by either
4-Cl or 4-NH2 substituents (6e,h) was detrimental for its
bioactivity. Introduction of additional substituents into
the 3- and 5-positions (6f,i) led to decreased activity.
Thus, the 4-methoxy groupseems to be the optimal
substituent on the cinnamoyl side chain.
In summary, through this study we have identified that
the 6,7-dimethoxy moiety is important for bioactivity of
compound 1. Replacement of the lactone functionality
in coumarin 1 by an amide decreased its activity. By sub-
stitution of the senecioyl chain with various cinnamoyl

2348
N.-H. Nam et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2345–2348
groups we found compound 6d with inhibitory activity
in HUVEC tube formation assay enhanced by one order
of magnitude. We have also synthesized compound 12,
an analogue of 6d, with equipotency and expected to
have much improved water solubility compared to 6d.
Further in vivo evaluations of 12 and more extended
investigation on this new lead are underway in our lab.
16. Ahsan, M.; Zaman, T. A.; Hasan, C. M.; Ito, C.; Islam,
S. K. N. Fitotherapia 2000, 71, 697.
17. Magiatis, P.; Melliou, E.; Skaltsounis, A. L.; Mitaku, S.;
Leonce, S.; Renard, P.; Pierre, A.; Atassi, G. J. Nat. Prod.
1998, 61, 982.
18. Canedo, L. M.; Fernander Puentes, J. L.; Baz, J. P.
J. Antibiot. 1997, 50, 175.
19. Egan, D.; James, P.; Cooke, D.; O’Kennedy, R. Cancer
Lett. 1997, 118, 201.
20. Valenti, P.; Rampa, A.; Recanatini, M.; Bisi, A.; Belluti,
F.; Da Re, P.; Carrara, M.; Cima, L. Anticancer Drug Des.
1997, 12, 443.
References and Notes
21. Valenti, P.; Rampa, A.; Recanatini, M.; Bisi, A.; Belluti,
F.; Da Re, P.; Carrara, M.; Cima, L. Anticancer Drug Des.
1997, 12, 137.
22. Valenti, P.; Fabbi, G.; Rampa, A.; Bisi, A.; Gobbi, S.; Da
Re, P.; Carrara, M.; Sgevano, A.; Cima, L. Anticancer Drug
Des. 1996, 11, 243.
23. Manfredini, S.; Baraldi, P. G.; Bazzanini, R.; Guarneri,
M.; Simoni, D.; Balzarini, J.; De Clercq, E. J. Med. Chem.
1994, 37, 2401.
1. Eid, A. I.; Ragab, F. A.; el-Ansary, S. L.; Gazayerly, S. M.;
Mourad, F. E. Arch. Pharm. (Weinheim) 1994, 327, 211.
2. Hogberg, T.; Vora, M.; Drake, S. D.; Mitscher, L. A.; Chu,
D. T. Acta Chem. Scand. B 1984, 38, 359.
3. el-Fattah, M. E. A. Arch. Pharm. Res. 1998, 21, 723.
4. Leolardo, B.; Philippo, C. H.; Elkhaili, F. J.; Silvio, M. L.;
Giuseppe, L.; Fedele, M.; Henry, M.; Davide, P.; Daniela, S.
Il Farmaco 1998, 53, 425.
5. Dejardin, J. V.; Stevens, U.; Jakubowski, B.; Lapiere, C. L.
Ann. Pharm. Fr. 1983, 41, 437.
6. Nozawa, K.; Yamada, M.; Tsuda, Y.; Kawai, K.; Naka-
jima, S. Chem. Pharm. Bull. 1981, 29, 3486.
7. Nozawa, K.; Yamada, M.; Tsuda, Y.; Kawai, K.; Naka-
jima, S. Chem. Pharm. Bull. 1981, 29, 2491.
24. Baraldi, P. G.; Valenti, P.; Rampa, A.; Recanatini, M.;
Bisi, A.; Belluti, F.; Da Re, P.; Carrara, M.; Cima, L. J. Med.
Chem. 1992, 35, 1877.
25. Simoni, D.; Manfredini, S.; Tabrizi, M. A.; Bazzanini, R.;
Baraldi, P. G.; Balzarini, J.; De Clercq, E. J. Med. Chem.
1991, 34, 3172.
8. Sardari, S.; Mori, Y.; Horita, K.; Micetich, R. G.; Nishibe,
S.; Daneshtalab, M. Bioorg. Med. Chem. 1999, 7, 1933.
9. Ishikawa, T.; Kotake, K.; Ishii, H. Chem. Pharm. Bull.
1995, 43, 1039.
10. Xie, L.; Takeuchi, Y.; Cosentino, L. M.; McPhail, A. T.;
Lee, H. K. J. Med. Chem. 2001, 44, 664.
26. Baraldi, P. G.; Guarneri, M.; Manfredini, S.; Simoni, D.;
Balzarini, J.; De Clercq, E. J. Med. Chem. 1989, 32, 284.
27. Nam, N. H.; Kim, Y.; You, Y. J.; Hong, D. H.; Kim, H.
M.; Ahn, B. Z. Planta. Med. Submitted for publication.
28. Matsumoto, M.; Kobayashi, H.; Hotta, Y. J. J. Org.
Chem. 1984, 49, 4740.
11. Yang, Z. Y.; Xia, Y.; Xia, P.; Brossi, A.; Cosentino, L. M.;
Lee, H. K. Bioorg. Med. Chem. Lett. 2000, 10, 1003.
12. Xie, L.; Takeuchi, Y.; Cosentino, L. M.; Lee, H. K. J.
Med. Chem. 1999, 42, 2662.
29. Ashimiro, N.; Nagamura, S.; Kobayashi, E.; Gomi, K.;
Saito, H. J. Med. Chem. 1999, 42, 669.
30. Illustrated data for compound 6d: Yield 75%. IR (KBr,
n
max cmꢀ1) 1720, 1640; 1H NMR (300 MHz, CDCl3) 7.71 (1H,
13. Huang, L.; Kashiwada, Y.; Cosentino, L. M.; Fan, S.;
Chen, C. H.; McPhail, A. T.; Fujioka, T.; Mihashi, K.; Lee,
H. K. J. Med. Chem. 1994, 37, 3947.
14. Flavin, M. T.; Rizzo, J. D.; Khilevich, A.; Kucherenko,
A.; Sheikman, A. K.; Vilaychack, V.; Lin, L.; Chen, W.;
Greenwood, E. M.; Pengsuparp, T.; Pezzoto, J. M.; Hughes,
S. H.; Flavin, T. M.; Cibulski, M.; Boulanger, W. A.; Shone,
R. L.; Xu, Z. Q. J. Med. Chem. 1996, 39, 1303.
d, J=15.43 Hz), 7.42–7.25 (2H, m), 7.16–6.95 (2H, m), 6.80
(1H, s), 6.78 (1H, s), 6.39 (1H, d, J=15.43 Hz), 6.31 (1H, s),
5.24 (2H, s), 3.81 (3H, s), 3.78 (3H, s). HRMS: [M]+ 396.1573
(C22H20O7), calcd 396.1203.
31. (a) Cytotoxic and in vitro HUVEC tube formation assays
were performed as described in our previous reports: Kim, Y.;
Kim, S. B.; You, Y. J.; Ahn, B. Z. Planta Med. 2002, 68, 271.
(b) Nam, N.H.; Kim, H. M.; Bae, K. H.; Ahn, B. Z. Phy-
tother. Res. In press. The inhibition percentages were calcu-
lated as described therein.
15. Guh, J. M.; Yu, S. M.; Ko, F. N.; Wu, T. S.; Teng, C. M.
Eur. J. Pharmacol. 1996, 298, 19.