Discovery of novel taspine derivatives as antiangiogenic agents

Article abstract of DOI:10.1016/j.bmcl.2009.11.073

VEGFR-2 plays a critical role in vasculogenesis and inhibitors of VEGFR-2 could be used in the treatment of cancer. Taspine was one of the active ingredients screened by using an endothelial cell membrane chromatography and showed inhibition against VEGFR-2. In our research, we explored how the lactone ring and biphenyl scaffold in taspine influence its potent in vitro anticancer and antiangiogenesis activities. Accordingly, we report the design, synthesis, and preliminary evaluation of four novel taspine derivatives as VEGFR-2 inhibitors. The preliminary biological test showed that one of the compounds showed much better inhibitory activities against CACO-2 (IC₅₀ = 52.5 nM) and ECV304 (IC₅₀ = 2.67 nM) than taspine. This result enlarges the interest in ring-opened taspine derivative skeleton in the search of new antiangiogenesis agents.

Full text of DOI:10.1016/j.bmcl.2009.11.073

Bioorganic & Medicinal Chemistry Letters 20 (2010) 718–721
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of novel taspine derivatives as antiangiogenic agents
*
Jie Zhang, Yanmin Zhang, Sanqi Zhang, Sicen Wang, Langchong He
School of Medicine, Xi’an Jiaotong University, No. 76, Yanta Weststreet, Xi’an, Shaanxi Province 710061, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
VEGFR-2 plays a critical role in vasculogenesis and inhibitors of VEGFR-2 could be used in the treatment
of cancer. Taspine was one of the active ingredients screened by using an endothelial cell membrane
chromatography and showed inhibition against VEGFR-2. In our research, we explored how the lactone
ring and biphenyl scaffold in taspine influence its potent in vitro anticancer and antiangiogenesis activ-
ities. Accordingly, we report the design, synthesis, and preliminary evaluation of four novel taspine deriv-
atives as VEGFR-2 inhibitors. The preliminary biological test showed that one of the compounds showed
much better inhibitory activities against CACO-2 (IC50 = 52.5 nM) and ECV304 (IC50 = 2.67 nM) than tas-
pine. This result enlarges the interest in ring-opened taspine derivative skeleton in the search of new
antiangiogenesis agents.
Received 13 July 2009
Revised 19 October 2009
Accepted 13 November 2009
Keywords:
Taspine derivatives
VEGFR-2 Inhibitors
Tyrosine kinase
Antiangiogenesis
Ó 2009 Elsevier Ltd. All rights reserved.
Angiogenesis is up-regulated in a wide range of disease situa-
tions such as retinopathies, arthritis, endometriosis and cancer.
Blockade of angiogenesis is an attractive approach for the treat-
ment of such diseases especially for cancer treatment.1 Vascular
endothelial growth factor (VEGF) is one of the most important
inducers of angiogenesis and exerts its cellular effects by inter-
acting with VEGF receptors: VEGFR-1 and VEGFR-2. VEGFR-2 is
the major positive signal transducer for endothelial cell prolifer-
ation and differentiation.2 It has been proven that inhibition of
VEGFR-2 activity reduces angiogenesis. For these reasons, VEG-
FR-2 appears to be an interesting target for design of anticancer
agents.
Taspine was first isolated and characterized from Leontice
ewersmannii Bge in 19563 and recently found in small yield in
blue cohosh, Caulophyllum thalictroides (L.) Michx.4,5 Taspine
was screened for the first time from Radix et Rhizoma Leonticis
(Hong Mao Qi in Chinese, HMQ) using cell membrane chromatog-
raphy by Li and He.6 Taspine has many pharmacological actions
such as bacteriostasis, antibiosis, antivirus treatment of wound
healing, anti-inflammatory, antiulcer, and cytotoxic activity.7 He
anc co-workers screened taspine for the first time using cell
membrane chromatography (CMC), and they found that taspine
had good affinity characteristics in human umbilical vein endo-
thelial cell membrane chromatography model.8 In a further study,
He et al assumed that taspine could inhibit tumor angiogenesis
and one of its mechanisms may be that it inhibited VEGFR-2
and the proliferation of vascular endothelial cells.9 Taspine also
showed a significantly higher effect on acetylcholinesterase and
selectively inhibited the enzyme in a long-lasting and concentra-
tion-dependent fashion.10 However, silico ADME prediction
showed that taspine wasn’t a compound with drug-like proper-
ties. Biological and cellular assays showed that it had very poor
solubility in any solvent. Poor solubility was a major problem in
the biological testing of compounds and it made formulating
the compounds a challenge for in vivo studies, both for efficacy
and pharmacokinetic experiments. Meanwhile, low solubility
can cause low bioavailability or give rise to large fluctuations in
the fraction absorbed in humans. Low solubility was also associ-
ated with stability problems.11
Extensive degradative work established taspine to possess the
dilactonic, tertiary amine structure (Fig. 1). The planar tetracyclic
compound consists of two linked lactone ring; each is condensed
again with an aromatic ring that is substituted with a methoxy
group.12 Taspine is a tetracyclic natural product and may be more
structurally complex than is necessary for optimal pharmacologic
effects. A complex lead compound may have a simpler pharmaco-
phore buried within its structure, and if this pharmacophore can be
clearly defined, the resulting biologically active, simpler molecule
may have improved synthetic tractability and be more useful as
a scaffold for further analogue design. To research the individual
contribution of the lactone ring and biphenyl scaffold of taspine
to the activity against cancer cells, we designed and synthesized
four novel ring-opened target compounds by structure-based drug
design (Fig. 1). The novel ring cleavage compounds were designed
in two pathways: cleavage of the C–C bond of diphenyl and ester
bond of ring B and ring D.
In this Letter, we described the design, synthesis and preliminary
evaluation of four ring-opened derivatives of taspine from commer-
cially available isovanillin (4) using dissection strategies. A molecu-
lar docking study of target compounds with VEGFR-2 was also
performed using SYBYL to identify their binding mode with enzyme.
* Corresponding author. Tel./fax: +86 29 82655451.
E-mail address: helc@mail.xjtu.edu.cn (L. He).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.11.073

J. Zhang et al. / Bioorg. Med. Chem. Lett. 20 (2010) 718–721
719
O
O
R
O
N
O
O
N
A
C
2
4
5
R = NMe2
R = Morpholine
R = 3-Cl-4-F-aniline
O
O
COOMe
OH
HO
MeOOC
B
D
O
O
O
O
O
O
O
3
1
Taspine
Figure 1. Design of ring-opened target compounds 2–5.
COOMe
d
COOMe
c
COOH
b
CHO
COOMe
a
MeO
O
MeO
MeO
MeO
MeO
O
OH
OH
OH
OH
h
9
6
7
8
O
O
O
N
O
O
COOH
e
f
g
2
COOMe
OBn
O
COOMe
OBn
COOMe
OBn
MeO
10+
OBn
O
O
O
11
O
O
O
12
13
14
O
COOMe
e
MeO
BnO
O
MeO
BnO
i
j
HO
MeOOC
+
10
MeO
O
O
COOH
COOMe
OH
13
16
8
17
N
O
O
g
O
O
O
O
h
BnO
MeOOC
HO
MeOOC
f
3
O
O
O
19
18
Scheme 1. Preparation of target compounds. Reagents and conditions: (a) NaOH, KOH, H2O, 84%; (b) CH3OH, H2SO4, 90%; (c) prenyl bromide, K2CO3, acetone, 92%; (d) N,N-
dimethylaniline, N2, reflux, 71%; (e) anhydrous THF, DCC, DMAP, 78%; (f) OsO4, NaIO4, acetone/H2O/t-BuOH, 32%; (g) dimethylamine/Morpholine/3-Cl-4-F-aniline, THF,
CH2Cl2, NaBH(OAc)3, 25–40%; (h) H2, Pd/C, 97%; (i) BnCl, K2CO3, EtOH, 95%; (j) NaOH/H2O, CH3OH, 93%.
Four of the title compounds were synthesized by the route out-
lined in Scheme 1. We used commercially available isovanillin as
the starting material. At first, Isovanillin (6) was oxidized to give
isovanillic acid (7) with fused sodium hydroxide and potassium
hydroxide. The methyl ester (8) was prepared to avoid side-reac-
tions of the carboxylate group. Then refluxing of (8) with prenyl
bromide in the presence of K2CO3 in anhydrous acetone afforded
compound (9).13 A solution of (9) in N,N-dimethylaniline was
heated to reflux for 8 h to give compound (10).14 Prenyl group
Table 2
Antiproliferative activities of target compounds on human cancer and normal cells
Compound
Cell lines (IC50, nM)
A549b
Table 1
CACO-2a
ECV304c
Prediction of physicochemical properties of compounds with ACD/Labs
Taspine
153
4.18
453
111
158
180
22.3
ND
ND
2.67
61
c Log P Log D (pH
pKa
Solubility (g/L) (pH
7.4)
PSAa
2
3
4
5
11003
3100
52.5
ND
7.4)
Taspine 1.93
2.92
1.72
1.97
2.65
6.56
8.21
0.021
74.3
94.53
94.53
103.76
103.32
2
3
4
5
3.18
3.18
3.33
6.06
9.73; 8.75 0.14
9.55; 8.57 0.13
9.02; 7.28 0.27
9.01; 3.65 0.061
ND is not determined.
a
CACO-2, human epithelial colorectal adenocarcinoma cells.
A549, carcinomic human alveolar basal epithelial cells.
ECV304, human umbilical endothelial cell line.
b
c
PSA = Polar Surface Area (Å2).
a

720
J. Zhang et al. / Bioorg. Med. Chem. Lett. 20 (2010) 718–721
Figure 2. FlexX docked conformation of compound (4) in the active site of VEGFR-2 (PDB ID: 3C7Q). (A) Compound (4) reacting with the amino acids residues of the active
site of VEGFR-2, Hydrogen bonds between the inhibitor and residues are shown with yellow dotted lines. (B) Kinase inhibitor–protein interactions are depicted by ribbon
structure.
was moved into the para-position of hydroxyl in this Claisen rear-
rangement process. The next step was the coupling of (10) to the
carboxyl of (11) by an ester bond with DCC and DMAP.15 The oxi-
dation of (12) with OsO4/NaIO4 produced aldehyde (13),16 which
reacted with dimethylamine followed by reduction with NaB-
H(OAc)3 to give (14).17 At last, benzyl deprotection of (14) with
palladium-carbon in MeOH gave the target compound (2).18 The
other target compound (3), (4) and (5) were synthesized in the
same way from isovanillin.19
Since it is not always practical to perform experimental mea-
surements, it is useful to rapidly predict molecular physicochemi-
cal properties such as Log P, Log D, pKa, PSA, and solubility values.20
Before biological evaluation, we evaluated a broad range of
molecular physicochemical properties which impacted drug-like
ADME parameters based on chemical structures with ACD/Labs
(Table 1). From Table 1, we can see that solubility values of target
compounds were threefold to 13-fold higher than taspine which
had very poor solubility in any solvent.
All target compounds were tested in functional cell-based as-
says using a variety of cancer and normal cell lines.21 Table 2 sum-
marized some of our findings. As shown in Table 2, all of the title
compounds moderately inhibited the growth of A549 in nanomolar
range which was less potent than the leading compound taspine.
However, the activity of (4), a D-ring opened compound, was much
better than that of taspine against CACO-2 (threefold) and ECV304
(about 10-fold) with the IC50 values of 52.5 nM and 2.67 nM,
respectively. Compound (4) appears to inhibit proliferation more
potently in ECV304, a human umbilical endothelial cell line, and
this may be an indication that the compound is acting through
inhibition of VEGFR-2. If the antiangiogenic properties of (4) are
confirmed, the compound could represent a novel hit for lead opti-
mization efforts. The results demonstrated that the ester bond of B-
ring was more important than that of D-ring in maintaining the
biological activity of this compounds type.
In an effort to elucidate the binding modes of these compounds
with VEGFR-2, the most active compound (4) was constructed with
Sybyl/Sketch module and optimized using Powell’s method with
the Tripos force field with convergence criterion set at 0.05 kcal/
(Å mol), and assigned with Gasteiger–Huckel method.22 The dock-
ing study performed using Sybyl/FlexX module, the residues in a
radius of 6.5 Å around XIN 1172 in the VEGFR-2 (PDB ID: 3C7Q)
were selected as the active site. Other docking parameters implied
in the program were kept default (Fig. 2). Docking of compound (4)
in the active site of VEGFR-2 showed four H-bond interactions be-
tween oxygen of the inhibitor and amino acid residues of the en-
zyme. We found that one oxygen of carboxylate group in ring A
formed two hydrogen bonds to Asn923 with distance of 1.88 Å
and 2.21 Å. The oxygen of phenolic hydroxyl group in ring B
formed a hydrogen bond to Arg1032 with distance of 1.78 Å while
oxygen of morpholine formed hydrogen bonds to Cys919 with
hydrogen length of 1.90 Å. Morpholine played an important role
in the binding of inhibitor with VEGFR-2. Although the computed
information partially supported our assumption, the ligand bound
structure of these taspine derivatives with VEGFR-2 will be definite
confirmation of the binding mode described and efforts are under-
way. This binding hypothesis could provide valuable information
for structure based design of analogs.
In summary, our research has led to further insights in the
structure-activity relationship of taspine analogs. We have demon-
strated that the lactone ring B is important for activity, while the
lactone ring D can be opened retaining and even improving the
antiproliferative properties of taspine. The data suggest that these
compounds could also exhibit an antiangiogenic mechanism of ac-
tion, similarly to taspine, and could therefore be a promising start-
ing point for further medicinal chemistry efforts. Our efforts in
pursuing the optimization of the physicochemical and biological
properties of taspine analogs will be reported in due course.
Acknowledgments
This work was supported by the National Natural Science Foun-
dation of China (Grant Nos. 30730110 and 30901839). We would
like to thank Professor Sicen Wang and Dr. Miao Li for MS measure-
ments. We also thank Professor Wenfang Xu and Huawei Zhu for
the docking study. We are grateful to Professor Zongru Guo for
practical guidance during the course of the project.
References and notes
1. Saaristo, A.; Karpanen, T.; Alitalo, L. Oncogene 2000, 19, 6122.
2. Gasparrini, G.; Longo, R.; Toi, M.; Ferrara, N. Nat. Clin. Pract. Oncol. 2005, 2, 562.
3. Platanova, T. F.; Kuzovkov, A. D.; Sheinker, Y. N. J. Gen. Chem. 1956, 26, 2957.

J. Zhang et al. / Bioorg. Med. Chem. Lett. 20 (2010) 718–721
721
4. Bettolo, R. M.; Scarpati, M. L. Phytochemistry 1979, 18, 520.
5. Kennelly, E. J.; Flynn, T. J.; Mazzola, E. P.; Roach, J. A.; McCloud, T. G.; Danford,
D. E.; Betz, J. M. J. Nat. Prod. 1999, 62, 1385.
6. Li, Y. P.; He, L. C. Chin. Sci. Bull. 2007, 52, 410.
7. Perdue, G. P.; Blomster, R. N.; Blake, D. A.; Farnsworth, N. R. J. Pharm. Sci. 1979,
68, 124.
8. Li, Y. P.; Yang, G. D.; He, L. C. Zhong Yao Cai 2007, 30, 220.
9. Zhang, Y. M.; He, L. C.; Meng, L.; Luo, W. J. Vasc. Pharmacol. 2008, 48, 129.
10. Rollinger, J. M.; Schuster, D.; Baier, E.; Ellmerer, E. P.; Langer, T.; Stuppner, H. J.
Nat. Prod. 2006, 69, 1341.
11. Trapani, A.; Sitterberg, J.; Bakowsky, U.; Kissel, T. Int. J. Pharm. 2009, 375, 97.
12. Scarpati, M. L.; Bianco, A.; Scalzo, R. L. Synth. Commun. 1991, 21, 849. All key
intermediates provided acceptable MS data and 1H NMR spectra that exhibit
no discernible impurities. Methyl 5-(3-(benzyloxy)-4-methoxybenzoyloxy)-4-
methoxy-2-(3-methylbut-2-enyl)benzoate (13): mp 115.0–116.0 °C. EI-MS(m/
z): 489.9 (M+), 1H NMR (300 MHz, CDCl3) d ppm 1.67 (s, 6H), 3.76 (s, 3H), 3.77
(s, 3H), 3.71 (d, 2H, J = 6.0 Hz), 3.89 (s, 3H), 5.13 (s, 2H), 5.23 (t, J = 6.0 Hz, 1H),
6.79 (s, 1H), 6.89 (d, J = 8.5 Hz, 1H), 7.22–7.33 (m, 5H), 7.39 (s, 1H), 7.69 (s, 1H),
7.79 (d, J = 8.5 Hz, 1 H)..
19. All new compounds 2–5 provided acceptable MS data and 1H-NMR spectra that
exhibit no discernible impurities. Methyl 2-(2-(dimethylamino)ethyl)-5-(3-
hydroxy-4-methoxybenzoyloxy)-4-methoxybenzoate (2): 1H NMR,(300 MHz,
CDCl3) d ppm 2.52(s, 6H), 2.71 (t, J = 7.3 Hz, 2H), 3.62 (t, J = 7.0 Hz, 2H), 3.84 (s,
3H), 3.89 (s, 3 H), 3.98 (s, 3H), 6.89 (s, 1H), 7.08 (d, J = 8.7 Hz, 1H), 7.54 (d,
J = 9.0 Hz, 1H), 7.67 (s, 1H), 7.81 (s, 1H); EI-MS(m/z): 403.1 [M+]; mp 112.5–
114.5 °C.
2-Methoxy-5-(methoxycarbonyl)phenyl-2-[2-(dimethylamino)ethyl]
-5-
hydroxy-4-methoxybenzoate (3): mp 107.0–109.5 °C, EI-MS (m/z): 402.9
(M+), 1H NMR (300 MHz, CDCl3) d ppm 2.62 (s, 6H), 3.55 (t, J = 6.6 Hz, 2H),
3.83 (s, 3H), 3.85 (s, 3 H), 3.98 (s, 3H), 4.01 (t, J = 6.9 Hz, 2H), 6.66 (s, 1H), 7.13
(d, J = 9.0 Hz, 1H), 7.72 (s, 1H), 7.86 (d, J = 9.0 Hz, 1H), 7.93 (s, 1H).
Methyl 5-[(3-hydroxy-4-methoxybenzoyl)oxy]-4-methoxy-2-(2-morpholin-4-
ylethyl) benzoate (4): mp 121.0–124.5 °C, EI-MS(m/z): 445.2 (M+), 1H NMR
(300 MHz, CDCl3) d ppm 2.55–2.63 (m, 4H), 2.82 (t, J = 8.0 Hz, 2H), 3.42–3.49
(m, 4H), 3.71 (t, J = 7.8 Hz, 2H), 3.82 (s, 3H), 3.90 (s, 3 H), 3.97 (s, 3H), 6.76 (s,
1H), 7.00 (d, J = 9.0 Hz, 1H), 7.42 (d, J = 8.7 Hz, 1H), 7.73 (s, 1H), 7.87 (s, 1H).
Methyl 2-{2-[(3-chloro-4-fluorophenyl)amino]ethyl}-5-[(3-hydroxy-4-metho-
xybenzoyl) oxy]-4-methoxybenzoate (5): mp 101.0–103.5 °C, EI-MS(m/z):
503.7 (M+), 1H NMR (300 MHz, CDCl3) d ppm 2.49 (t, J = 7.5 Hz, 2H), 3.52 (t,
J = 7.2 Hz, 2H), 3.81 (s, 3H), 3.87 (s, 3 H), 3.95 (s, 3H), 6.88 (d, J = 7.5 Hz, 1H),
6.94 (s, 1H), 6.97–7.10 (m, 2H), 7.19–7.21 (m, 1H), 7.54 (d, J = 9.0 Hz, 1H), 7.68
(s, 1H), 7.77 (s, 1H).
13. Trost, B. M.; Tang, W.; Toste, F. D. J. Am. Chem. Soc. 2005, 127, 14785.
14. Baret, P.; Beguin, C.; Gaude, D.; Gellon, G.; Mourral, C.; Pierre, J. L.; Serratrice,
G.; Favier, A. Tetrahedron 1994, 50, 2077.
15. Burger, E. C.; Tunge, J. A. Org. Lett. 2004, 6, 4113.
16. Kelly, T. R.; Xie, R. L. J. Org. Chem. 1998, 63, 8045.
20. Bhal, S. K.; Kassam, K.; Peirson, I. G.; Pearl, G. M. Mol. Pharm. 2007, 4, 556.
21. Zhao, J.; Zhao, L.; Chen, W.; He, L. C.; Li, X. Biomed. Pharmacother. 2008, 62, 383.
22. Mou, J.; Fang, H.; Jing, F.; Wang, Q.; Liu, Y.; Zhu, H.; Xu, W. Bioorg. Med. Chem.
2009, 17, 4666.
17. Karamali, K.; Mathias, Grober Tetrahedron 2002, 58, 1159.
18. Steâphane, Q.; Laurent, P.; Mayalen, O.; Matthew, A. L. Tetrahedron 2001, 57,
319.