Synthesis of phenanthroindolizidine alkaloids and evaluation of their antitumor activities and toxicities

Article abstract of DOI:10.1016/j.bmcl.2011.07.120

We previously reported that phenanthroindolizidine alkaloid 3 and its derivatives had markedly potent in vitro cytotoxicity. However, they had low in vivo antitumor activities and high in vivo toxicities, which was a serious problem. To address this probl

Full text of DOI:10.1016/j.bmcl.2011.07.120

Bioorganic & Medicinal Chemistry Letters 21 (2011) 5978–5981
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of phenanthroindolizidine alkaloids and evaluation of their
antitumor activities and toxicities
⇑
Takashi Ikeda , Takashi Yaegashi, Takeshi Matsuzaki, Ryuta Yamazaki, Syusuke Hashimoto,
Seigo Sawada
Yakult Central Institute for Microbiological Research, 1796 Yaho, Kunitachi-shi, Tokyo 186 8650, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
We previously reported that phenanthroindolizidine alkaloid 3 and its derivatives had markedly potent
in vitro cytotoxicity. However, they had low in vivo antitumor activities and high in vivo toxicities, which
was a serious problem. To address this problem, new phenanthroindolizidine derivatives were synthe-
sized and their antitumor activities and toxicities were evaluated. This study describes the relationship
between the chemical structures, antitumor activities, and toxicities of these phenanthroindolizidine
derivatives. Based on its properties, compound 8 was found to be the most suitable potential antitumor
agent.
Received 21 April 2011
Revised 30 June 2011
Accepted 14 July 2011
Available online 6 August 2011
Keywords:
Phenanthroindolizidine alkaloid
Antitumor activity
Toxicity
Ó 2011 Elsevier Ltd. All rights reserved.
Phenanthroindolizidine alkaloids, which are isolated mainly
from the plants of the Asclepiadaceae family, are known to possess
various biological activities, including anti-arthritis,1 antifungal,2
anti-inflammatory,3 and antitumor4 activities. In particular, their
significant antitumor activities and unique mode of action make
them attractive candidates for use as novel antitumor agents.4,5
Tylophorine (1) and tylocrebrine (2) are two representatives of this
group of alkaloids (Fig. 1).6
these derivatives are a hydroxyl group at the C14 position and S
configuration of the C13a and C14 positions. Accordingly, these
structural features were believed to be related to the observed tox-
icity. Thus, in order to explore the relationship between these fea-
tures and the toxicity, we synthesized derivatives with different
structural features at these positions to that of compound 3 (i.e.,
compounds 6, 8, and their enantiomers) and evaluated their anti-
tumor activities and toxicities.
Compound 3, which is derived from a butterfly (Ideopsis similis),
has been reported to possess strong cytotoxicity.7 Recently, we re-
ported the synthesis of compound 3 and its derivatives, and evalu-
ated their cytotoxicities in vitro and their antitumor activities
in vivo.8 Almost all of the synthesized derivatives of compound 3
exhibited strong in vitro cytotoxicities; however, their in vivo anti-
tumor activities were not comparable to the remarkable cytotoxic-
ities. In addition, it was observed that high-dose treatments with
these compounds frequently caused fatal toxicity, leading to a high
mortality of the test mice. To address these problems, we synthe-
sized new synthetic derivatives of compound 3 and studied them
in order to determine which of its structural features are related
to the antitumor activities and to the toxicities. Here we report
the detailed results of these analyses.
The synthesis of compounds 6 and 8 are illustrated in Scheme 1.
Inversion of stereochemistry at the C14 position of compound 4
under modified Mitsunobu conditions9 produced a lactam, com-
pound 5, which was then reduced to an amine. Following this,
the benzyl ether moiety of the reduced amine was subjected to
hydrogenolysis that resulted in the production of the desired com-
pound 6. Reductive dehydroxylation of the hydroxyl group at the
C14 position of compound 410 produced a lactam, compound 7,
which was then reduced to an amine. Following this, the benzyl
ether moiety of the reduced amine was subjected to hydrogenoly-
sis that resulted in the production of the desired compound 8. The
enantiomers of compounds 6 and 8 were also synthesized using
ent-4 as the starting material.
The synthesized compounds were evaluated for their cytotoxici-
ties against three human cancer cell lines: A549 (lung), HT-29 (colo-
rectal), and HCT116 (colorectal) (Table 1).11 Compound 8 exhibited a
strong cytotoxicity comparable to compound 3, while compound 6,
ent-6, and ent-8 were less cytotoxic. Based on these results, we then
evaluated the in vivo antitumor activity of compound 8 using a mur-
ine Meth A sarcoma inoculated tumor model (Table 2).12 Compound
8 showed significant antitumor activities (inhibition rate of 70.8%).
In addition, compound 8 did not show toxicity compared to
Compound 3 and its derivatives have been consistently found to
cause toxicity regardless of the differences in the structural fea-
tures of their aromatic moieties.8 In this study, we define toxicity
as the incidence of mortality. The common structural features of
⇑
Corresponding author. Tel.: +81 42 577 8957; fax: +81 42 577 3020.
E-mail address: takashi-ikeda@yakult.co.jp (T. Ikeda).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.07.120

T. Ikeda et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5978–5981
5979
Figure 1. Chemical structures of phenanthroindolizidine alkaloids.
Scheme 1. Reagents and conditions: (a) N,N,N0,N0-tetramethylazodicarboxyamide, PBu3, p-methoxybenzoic acid, benzene (73%); (b) KOH, H2O, MeOH, 1,4-dioxane (55%); (c)
BH3-THF, THF; (d) H2, Pd-C, MeOH (43% for compound 6, 78% for compound 8, performed over two steps); (e) BF3-OEt2, Et3SiH, CH2Cl2 (66%).
Table 1
Table 2
Cytotoxicity of the synthesized compounds
In vivo antitumor efficacy of compound 8
Compound
Inhibition rate (%)
Total dose (mg/kg)
Mortality
8
46.9
70.8
30.6
—
46.2
78.3
25
50
25
50
50
100
0/5
0/5
0/5
5/5
0/5
0/5
3
CPT-11
evaluated their cytotoxicities (Table 3). Both compounds were
found to be cytotoxic; however, their cytotoxicities were not as
high as that of compound 8. This result stimulated us to evaluate
the efficacy of the substituent at R2. Therefore, the hydroxyl group
of compound 8 was changed to an acetoxy group (11), methoxy
group (12), hydrogen atom (13), fluorine atom (14), or an ethyl
group (15), and the cytotoxicities of the resulting compounds were
evaluated. Among these compounds, compound 11 exhibited cyto-
toxicity comparable to that of compound 8. The cytotoxicities of
compounds 12, 13, and 14 were slightly weaker than the cytotox-
icity of compound 8, while the cytotoxicity of compound 15 was
considerably weaker compared with that of compound 8. We then
investigated the effect of different substituents at the positions R4
and R5. Replacing the methoxy groups at R4 and R5 of compound 8
with ethoxy groups (16) led to a slight decrease in the cytotoxicity,
while changing the methoxy group to a methylenedioxy group (17)
led to a remarkable decrease in the cytotoxicity.14
Compound
R
Configuration of (13a, 14)
IC50 (nM)
A549
HT-29
HCT116
3
OH
OH
OH
OH
H
(S, S)
(R, R)
(S, R)
(R, S)
(S, –)
(R, –)
2.7
2.9
1.3
ent-3
6
ent-6
8
30.6
52.3
818.4
0.6
37.0
53.6
935.0
1.8
32.4
44.4
670.8
0.5
ent-8
H
58.2
157.5
59.9
compound 3, which showed high toxicity (mortality: 0/5 for com-
pound 8 vs 5/5 for compound 3). Encouraged by these promising re-
sults, compound 8 was used as the new lead compound, and its
derivatives, compounds 9–17, were synthesized and evaluated for
their antitumor activities.
These derivatives were synthesized by the aforementioned syn-
thetic routes using appropriate starting materials.13 To investigate
the effect of moving a hydroxyl group of compound 8 from R2 to R1
or R3 on cytotoxicity, we synthesized compounds 9 and 10 and
Subsequently, the in vivo antitumor efficacies of these com-
pounds were evaluated using a human colorectal cancer HCT116

5980
T. Ikeda et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5978–5981
Table 3
Cytotoxicities of the derivatives of compound 8
Compound
R1
R2
R3
R4
R5
IC50 (nM)
A549
HT-29
HCT116
9
OH
H
H
H
H
H
H
H
H
H
H
AcO
CH3O
H
H
OH
H
H
H
H
H
H
H
CH3O
CH3O
CH3O
CH3O
CH3O
CH3O
CH3O
CH3CH2O
CH3O
CH3O
CH3O
CH3O
CH3O
CH3O
CH3O
CH3CH2O
21.6
5.5
2.8
43.8
48.9
51.0
440.9
42.2
391.7
32.1
452.1
3.5
45.1
57.1
58.9
484.4
80.5
23.5
4.7
2.0
41.7
49.6
48.1
403.2
38.8
376.8
10
11
12
13
14
15
16
17
F
CH3CH2
OH
OH
OCH2O
878.7
Table 4
In vivo antitumor efficacies of compound 8 and its derivatives
that the hydroxyl group at the C14 position is related to the observed
fatal toxicity and that toxicity could be diminished by removing this
hydroxyl group without affecting the antitumor properties of these
compounds.
Compound
Inhibition rate (%)
Total dose (mg/kg)
Mortality
8
35.7
48.7
34.7
21.0
30.7
—
100
200
200
100
100
200
135
0/5
0/5
0/5
0/5
0/5
5/5
0/5
11
13
14
Acknowledgment
The authors would like to thank Miyuki Uchida and Fukiko
Nishisaka for their excellent assistance.
CPT-11
77.0
References and notes
xenograft model.15 Compounds 8, 11, 13, and 14 were selected as
the testing probes because of their potent in vitro activities (Table
4). Unfortunately, the antitumor activities of these compounds
were low to moderate in this model; this was contrary to the
promising results of the in vitro screening. Among these, com-
pound 8 showed the highest antitumor activity (inhibition rate of
48.7%); however, no toxicity was observed even when it was
administered at a high dose (200 mg/kg). As for compound 14, a
high toxicity with 100% mortality was observed at a dose of
200 mg/kg; however, the C14-hydroxylated derivative of com-
pound 14 showed 100% mortality at a dose of 6.25 mg/kg (data
not shown). These results suggest that fatal toxicity could be con-
siderably diminished by removing the hydroxyl group present at
the C14 position.
In summary, this report describes the relationship between the
chemical structures, antitumor activities, and toxicities of phenan-
throindolizidine derivatives that was determined by investigating
the chemical environment of the C13a and C14 positions. Compound
8, which is the C14-dehydroxylated derivative of compound 3,
showed potent in vitro cytotoxicity and superior in vivo antitumor
activity compared to compound 3, but with a lower level of toxicity
(as evaluated in the murine Meth A sarcoma inoculated tumor mod-
el). Based on these results, derivatives of compound 8 were synthe-
sized and evaluated for their in vitro cytotoxicities. The substituents
of the phenanthrene ring had a strong effect on the observed amount
of cytotoxicity. The in vivo antitumor efficacies of the compounds
that were shown to have potent in vitro cytotoxicities were also
evaluated using the human colon cancer HCT116 xenograft model.
Compound 8 showed the highest, although still relatively moderate,
level of antitumor activity without toxicity. These results suggest
1. You, X.; Pan, M.; Gao, W.; Shiah, H. S.; Tao, J.; Zhang, D.; Koumpouras, F.; Wang,
S.; Zhao, H.; Madri, J. A.; Baker, D.; Cheng, Y. C.; Yin, Z. Arthritis Rheum. 2006, 54,
877.
2. Baumgartner, B.; Erdelmeier, C. A. J.; Wright, A. D.; Rali, T.; Sticher, O.
Phytochemistry 1990, 29, 3327.
3. Yang, C.-W.; Chen, W.-L.; Wu, P.-L.; Tseng, H.-Y.; Lee, S.-J. Mol. Pharmacol. 2006,
69, 749.
4. Gao, W.; Lam, W.; Zhong, S.; Kaczmarek, C.; Baker, D. C.; Cheng, Y. C. Cancer Res.
2004, 64, 678.
5. Gao, W.; Chen, A. P.-C.; Leung, C.-H.; Gullen, E. A.; Furstner, A.; Shi, Q.; Wei, L.;
Lee, K.-H.; Cheng, Y.-C. Bioorg. Med. Chem. Lett. 2008, 18, 704.
6. Gellert, E. Alkaloids: Chemical and Biological Perspectives In Pelletier, S. W.,
Ed.; Academic Press: New York, 1987; p 55.
7. Komatsu, H.; Watanabe, M.; Ohyama, M.; Enya, T.; Koyama, K.; Kanazawa, T.;
Kawahara, N.; Sugimura, T.; Wakabayashi, K. J. Med. Chem. 2001, 44, 1833.
8. Ikeda, T.; Yaegashi, T.; Matsuzaki, T.; Hashimoto, S.; Sawada, S. Bioorg. Med.
Chem. Lett. 2011, 21, 342.
9. Tsunoda, T.; Yamamiya, Y.; Kawamura, Y.; Ito, S. Tetrahedron Lett. 1995, 36,
2529.
10. Nordlander, J. E.; Njoroge, F. G. J. Org. Chem. 1987, 52, 1627.
11. Cell viability was assayed in a 96-well plate using TetraColor ONE (Seikagaku
Corp., Tokyo, Japan), according to the manufacturer’s protocol. Briefly,
exponentially growing A549, HT-29, or HCT116 cells were seeded in 96-well
plates at a density of 103 cells/well. The next day, serially diluted compounds
were added to each plate. After 96 h of incubation, 2-(2-methoxy-4-
nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium
monosodium salt was added to each well and the plates were incubated at
37 °C for 1 h. Absorbance was measured at 450 nm using SPECTRAMax
PLUS384 (Molecular Devices, Sunnyvale, CA, USA). The half maximal
inhibitory concentration (IC50
compound that inhibited cell growth by 50% with the solvent-treated cells as
control.
) was defined as the concentration of a
12. Meth
A
cells (sarcoma, 2.5 ꢀ 105 cells/mouse) were inoculated s.c. into
7 week old male BALB/c mice (5/group), and samples were injected i.v. on
days 1, 5, and 9. Tumors were excised and weighed on day 21 after tumor
inoculation. The inhibition rate (%) was calculated using the following
formula: [1 ꢁ (mean tumor weight of tested mice)/(mean tumor weight of
control mice] ꢀ 100.

T. Ikeda et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5978–5981
5981
13. Compound 6: white powder, 1H NMR (400 MHz, DMSO-d6) d: 1.68–1.88 (3H,
m), 2.22–2.45 (3H, m), 3.21–3.31(1H, m), 3.50 (1H, d, J = 15.0 Hz), 3.93 (3H, s),
3.99 (3H, s), 4.43 (1H, d, J = 15.0 Hz), 4.84–4.95 (1H, m), 5.14 (1H, br s), 7.04
(1H, dd, J = 1.5, 9.0 Hz), 7.19 (1H, s), 7.91 (1H, d, J = 1.5 Hz), 7.91 (1H, s), 8.30
(1H, d, J = 9.0 Hz), 9.60 (1H, br s); HRMS (ESI) calcd for C22H24NO4 [M+H]+,
(3H, s), 4.60 (1H, d, J = 15.5 Hz), 7.24 (1H, s), 7.53–7.64 (2H, m), 7.97–8.04 (1H,
m), 8.16 (1H, s), 8.72–8.78(1H, m); HRMS (ESI) calcd for C20H24NO2 [M+H]+,
334.1807. Found: 334.1816; ½a D27
ꢂ
+109.16 (c = 0.05, CHCl3).
Compound 14: yellow powder, 1H NMR (400 MHz, DMSO-d6) d: 1.53–1.70 (1H,
m), 1.75–1.94 (2H, m), 2.07–2.23 (1H, m), 2.31–2.44 (2H, m), 2.71–2.83 (1H,
m), 3.26–3.42 (2H, m), 3.54 (1H, d, J = 15.4 Hz), 3.94 (3H, s), 4.00 (3H, s), 4.58
(1H, d, J = 15.4 Hz), 7.23 (1H, s), 7.43 (1H, ddd, J = 2.7, 8.8, 11.5 Hz), 8.05 (1H,
dd, J = 6.1, 8.8 Hz), 8.11 (1H, s), 8.59 (1H, dd, J = 2.7, 11.5 Hz); HRMS (ESI) calcd
366.1705. Found: 366.1713; ½a D26
ꢂ
+96.52 (c = 0.05, CHCl3/CH3OH = 1:1).
Compound 8: white powder, 1H NMR (400 MHz, DMSO-d6) d: 1.52–1.68 (1H,
m), 1.75–1.94 (2H, m), 2.07–2.23 (1H, m), 2.26–2.44 (2H, m), 2.71–2.83 (1H,
m), 3.26–3.42 (2H, m), 3.50 (1H, d, J = 15.0 Hz), 3.92(3H, s), 3.98 (3H, s), 4.53
(1H, d, J = 15.0 Hz), 7.09 (1H, dd, J = 2.3, 8.9 Hz), 7.18 (1H, s), 7.83 (1H, d,
J = 8.9 Hz), 7.92(1H, s), 7.94 (1H, d, J = 2.3 Hz); HRMS (ESI) calcd for C22H24NO3
for C22H22NO2F [M+H]+, 352.1713. Found: 352.1645; ½a 2D7
ꢂ
+120.78 (c = 0.10,
CHCl3).
14. Some compounds exhibited selective cytotoxicities against the A549 and
HCT116 cell lines. The reason for this finding remains unknown at this time.
15. After transplanting HCT116 cells (2 ꢀ 106 cells/mouse) subcutaneously into
6 week old male nude mice, the mice were grouped (5 mice/group) on the day
when the estimated tumor volume calculated by the following formula (A)
reached about 100 mm3 (Day 0). Samples were administered intraperitoneally
10 times on days 1–5 & 8–12. Tumors were excised and weighed on day 21
after tumor inoculation. The inhibition rate (%) was calculated using the
following formula: [1 ꢁ (mean tumor weight of tested mice)/(mean tumor
weight of control mice] ꢀ 100.
[M+H]+, 350.1756. Found: 350.1761;
CH3OH = 1:1).
½
a 2D6
ꢂ
+102.99 (c = 0.11, CHCl3/
Compound 11: white powder, 1H NMR (400 MHz, DMSO-d6) d: 1.58–1.70 (1H,
m), 1.75–1.94 (2H, m), 2.07–2.23 (1H, m), 2.31–2.44 (2H, m), 2.35 (3H, s), 2.81
(1H, dd, J = 11.2, 15.4), 3.27–3.42 (2H, m), 3.56 (1H, d, J = 15.3 Hz), 3.94 (3H, s),
4.00 (3H, s), 4.60 (1H, d, J = 15.3 Hz), 7.24 (1H, s), 7.35 (1H, dd, J = 1.0, 9.0 Hz),
8.03 (1H, d, J = 9.0 Hz), 8.07 (1H, s), 8.49 (1H, s); HRMS (ESI) calcd for
C
24H26NO4 [M+H]+, 392.1862. Found: 392.1875; ½a 2D9
ꢂ
+112.42 (c = 0.05, CHCl3).
Compound 13: yellow powder, 1H NMR (400 MHz, DMSO-d6) d: 1.58–1.70 (1H,
m), 1.75–1.94 (2H, m), 2.07–2.23 (1H, m), 2.31–2.44 (2H, m), 2.81 (1H, dd,
J = 10.5, 15.9 Hz), 3.26–3.44 (2H, m), 3.57 (1H, d, J = 15.5 Hz), 3.94 (3H, s), 4.00
Estimated tumor volume = 1/2 ꢀ long diameter ꢀ short diameter ꢀ short
diameter (A).