Synthesis of new camptothecin analogs with improved antitumor activities

Article abstract of DOI:10.1016/j.bmcl.2009.02.031

Novel hexacyclic camptothecin analogs containing cyclic amidine, urea, or thiourea moiety were designed and synthesized based on the proposed 3D-structure of the topoisomerase I (Topo I)/DNA/camptothecin ternary complex. The analogs were prepared from 9-n

Full text of DOI:10.1016/j.bmcl.2009.02.031

Bioorganic & Medicinal Chemistry Letters 19 (2009) 2018–2021
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of new camptothecin analogs with improved antitumor activities
Satoshi Niizuma, Masao Tsukazaki, Hitomi Suda, Takeshi Murata, Jun Ohwada, Sawako Ozawa,
Hiroshi Fukuda, Chikako Murasaki, Masami Kohchi, Kenji Morikami, Kiyoshi Yoshinari, Mika Endo,
Masako Ura, Hiromi Tanimura, Yoko Miyazaki, Tsuyoshi Takasuka, Akira Kawashima, Eitaro Nanba,
*
Kounosuke Nakano, Kotaro Ogawa, Kazuko Kobayashi, Hisafumi Okabe, Isao Umeda, Nobuo Shimma
Kamakura Research Laboratories, Chugai Pharmaceutical Co., Ltd, 200 Kajiwara, Kamakura, Kanagawa 247-8530, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Novel hexacyclic camptothecin analogs containing cyclic amidine, urea, or thiourea moiety were
designed and synthesized based on the proposed 3D-structure of the topoisomerase I (Topo I)/DNA/cam-
ptothecin ternary complex. The analogs were prepared from 9-nitrocamptothecin via 7,9-diaminocam-
ptothecin derivatives as a key intermediate. Among them, 7c exhibited in vivo antitumor activities
superior to CPT-11 in human cancer xenograft models in mice at their maximum tolerated doses though
its in vitro antiproliferative activity was comparable to SN-38 against corresponding cell lines.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 9 December 2008
Revised 4 February 2009
Accepted 7 February 2009
Available online 12 February 2009
Keywords:
Anticancer agent
Camptothecin analog
Topoisomerase I inhibitor
BCRP
Tremendous efforts have been made over the years to discover
more effective agents for the treatment of gastro-intestinal tumors,
but 5-fluorouracil (5-FU)/leucovorine, an oral 5-FU prodrug (cape-
citabine) and a camptothecin derivative (irinotecan [CPT-11]) still
remain the key drugs for the treatment of advanced gastric and/
or colorectal cancer (CRC), in combination with other agents.1,2
CPT-11 is a prodrug of SN-38 with improved water-solubility. It
is used for treatment of CRC in combination with 5-FU/leucovorine
(named FOLFIRI) with or without an additional agent (e.g., bev-
acizumab). The clinical efficacy of CPT-11 is, however, limited
due to the following drawbacks: high inter-patient variability in
pharmacokinetics (from poor bioconversion to the active drug,
SN-38, and SNPs of the metabolizing enzyme, UGT1A1) and severe
toxicity in bone marrow and intestine. The camptothecin analogs
with an amino group in the core structure for increasing solubility,
for example, topotecan and Dx8951 (Fig. 1), showed limited clini-
cal efficacy, likely because of insufficient tissue distribution in hu-
man.3,4 In order to develop a new camptothecin analog with higher
antitumor efficacy and broader spectrum by overcoming the draw-
backs of CPT-11 mentioned above, we designed new lipophilic
camptothecin analogs with high profile tissue distribution and
their water-soluble prodrugs for intravenous application. In this
paper, we describe the design, synthesis, and biological activities
of the parent drug CH0793076 (7c) that exhibited higher antitumor
activity than CPT-11 in various human cancer xenograft models.
We designed new camptothecin analogs (Fig. 2) based on the
ternary complex of DNA-Topo I-camptothecin proposed by Redin-
bo et al.5 In this model, we found a large space around the C-7 po-
sition of camptothecin that allowed the introduction of an
additional F-ring and a hydrophobic side chain (R1) (Fig. 3a). There
7
9
HO
O
O
A
B
N
C
N
N
D
N
E
O
O
camptothecin
SN-83
OH
O
OH
O
O
N
O
O
N
N
N
O
irinotecan (CPT -11)
OH
O
N
NH2
F
HO
O
O
N
N
N
F
N
O
O
O
O
OH
OH
topotecan
Dx8951
* Corresponding author. Tel.: +81 467 47 2280; fax: +81 467 45 6824.
E-mail address: shinmanbo@chugai-pharm.co.jp (N. Shimma).
Figure 1. Camptothecin and representative analogs.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.02.031

S. Niizuma et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2018–2021
2019
R2
F
is also a large space around the C-7 position of camptothecin in a
new model based on the crystal structure of DNA-Topo I- topotec-
an ternary complex (Fig. 3b).6
In order to synthesize various heterocyclic rings for the F-ring,
the key intermediates, 7,9-diaminocamptothecins 4a–e, were pre-
pared from 9-nitrocamptothecin 20(S)-O-acetate (1) as illustrated
in Scheme 1.7 1 was oxidized to the N-oxide using a hydrogen per-
oxide-urea complex, followed by treatment with oxalyl chloride to
give a 7-chloro derivative (2).
X
R1
R1
HN
N
N
N
N
N
F
O
O
N
N
O
O
O
O
OH
OH
Figure 2. Newly designed camptothecin analogs.
A substituted amino group was introduced at C-7 by treatment
of 2 with various amines in refluxing dioxane to give 3a–e in mod-
est to good yields. Derivatives 3a–e were converted to the corre-
sponding diamino derivatives 4a–e by hydrogenation.
The resulting diamino derivatives 4a–e were then transformed
into various camptothecin analogs having an F-ring. Treatment of
4a–e with triphosgene followed by the removal of the acetyl group
at C-20 with hydrazine gave urea type analogs 6a–e. In the same
manner and by treating with thiocarbonyl 1,10-diimidazole, 4d
was converted to the thiourea analog 5d.
The syntheses of cyclic amidine analogs were carried out under
two conditions: (1) 7a–e and 8c derivatives were obtained by
treating 4 with methyl orthoformate or methyl orthoacetate in
the presence of an acid catalyst followed by hydrazine treatment,
and (2) the functionalized cyclic amidine analog 9d was prepared
by the reaction of 4d with acetoxyacetyl chloride in the presence
of diisopropylamine, followed by hydrolysis of the ester groups
with hydrazine.
The in vitro antiproliferative activities of the new camptothecin
analogs against human cancer cell lines, CRC (HCT116) and NSCLC
(non-small cell lung carcinoma; QG56, NCI-H460), were evaluated
(Table 1). The cyclic urea and amidine analogs having a morpholi-
Figure 3. Presumed binding conformation of CH0793076 (7c) with DNA-Topo I
complex. The presumed binding conformation (a) was made based on the model
proposed by Redinbo et al. Originally, we designed our compounds based on the
binding conformation (a). The presumed binding conformation (b) was made based
on the crystal structure of DNA-Topo I-topotecan ternary complex (PDB accession
number, 1K4T). Although the binding conformation (b) is different from the binding
conformation (a), the additional F-ring and the hydrophobic side chain occupy the
same binding pocket. DNA structures which intercalate with CH0793076 (7c) are
omitted to be clarified.
NO2
S
O
R1
O
:
N
O
6a R = CH2CH2
·HCl (20% from 3a)
(50% from 3b)
(81% from 3c)
(32% from 3d)
(20% from 3e)
HN
HN
N
N
N
N
N
:
6b R = (CH2)3CH3
O
O
N
N
N
:
:
:
6c
6d
6e
R = (CH2)4CH3
O
AcO
R = (CH2)2CH(CH3)2
R = CH2CH2Ph
O
O
O
1
HO
HO
5d (30% from 3d)
O
O
a
d
NO2 Cl
e
O
N
R1
R1
N
HCl
:
7a R = CH CH
2 N
·HCl (35% from 3a)
(39% from 3b)
(57% from 3c)
(30% from 3d)
(5% from 3e)
O
NH
N
N
NH2
N
O
2
AcO
:
7b R = (CH2)3CH3
f
O
O
O
2 (47%)
N
N
:
:
:
7c
7d
7e
R = (CH2)4CH3
N
R = (CH2)2CH(CH3)2
R = CH2CH2Ph
b
O
O
c
R1
NO2 NH
AcO
HO
O
O
4a-e
O
N
h
g
N
O
HO
AcO
O
N
N
N
N
N
O
O
:
3a R = CH2CH2
N
O
(70%)
(73%)
(57%)
N
N
:
3b R = (CH2)3CH3
N
:
:
:
O
O
3c
3d
3e
R = (CH2)4CH3
R = (CH2)2CH(CH3)2 (46%)
R = CH2CH2Ph (61%)
HO
HO
O
O
8c (55% from 3c)
9d (62% from 3d)
Scheme 1. Reagents and conditions for camptothecin analogs: (a) (i) hydrogen peroxide-urea complex, TFA, rt, 5 h; (ii) oxalyl chloride, DMF, 15 °C, 1.5 h; (b) R1NH2, 1,4-
dioxane, reflux; (c) Pd/C, H2(g), MeOH/HCl, rt; (d) (i) thiocarbonyl 1,1’-diimidazole, diisopropylethylamine, DMAP, DCM, reflux, 6 h; (ii) hydrazine, MeOH, rt; (iii) 4 N HCl/
MeOH; (e) (i) triphosgene, DCM, rt; (ii) hydrazine, MeOH, rt; (iii) 4 N HCl/MeOH; (f) (i) orthoformic acid methyl ester, TsOH; (ii) hydrazine, MeOH, rt; (iii) 4 N HCl/MeOH; (g)
(i) orthoacetic acid methyl ester, TsOH, DCM, reflux, 3 h; (ii) hydrazine, MeOH, rt; (iii) 4 N HCl/MeOH; (h) (i) acetoxyacetyl chloride, diisopropylethylamine, DCM, rt,
overnight; (ii) hydrazine, MeOH, rt; (iii) 4 N HCl/MeOH.

2020
S. Niizuma et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2018–2021
Table 1
In vitro antiproliferative activities of the derivatives against human cancer cell lines
Core structure
X
R1
R2
Compound
In vitro IC50 (nM)
QG56 (NSCLC)
18
HCT116 (CRC)
3.6
NCI-H460 (NSCLC)
26
O
—
6a
X
CH2CH2 N
O ·HCl
R1
HN
N
N
O
O
O
O
S
(CH2)3CH3
(CH2)4CH3
(CH2)2CH(CH3)2
CH2CH2Ph
(CH2)2CH(CH3)2
—
—
—
—
—
6b
6c
6d
6e
5d
0.40
0.85
0.05
0.25
0.02
1.2
8.5
0.58
1.2
0.16
0.99
8.2
0.72
0.38
0.08
O
N
O
HO
O
—
H
7a
2.2
9.4
7.2
R2
CH2CH2 N
O ·HCl
R1
N
N
N
—
—
—
—
—
—
(CH2)3CH3
(CH2)4CH3
(CH2)2CH(CH3)2
CH2CH2Ph
(CH2)4CH3
H
H
H
H
7b
7c
7d
7e
8c
9d
0.52
0.36
<0.04
0.36
0.12
0.44
2.3
2.3
0.25
2.1
1.8
1.8
1.6
2.3
0.11
0.91
0.66
0.52
O
N
Me
CH2OH
O
HO
(CH2)2CH(CH3)2
O
SN-38
0.55
2.8
3.3
In vitro cytotoxic assay: The cells were exposed to test compounds for 3 days at 37 °C in an incubator containing 5% CO2 in air. The concentration of a test compound
producing 50% inhibition (IC50) of cell growth was calculated.
noethyl group in the side chain (6a, 7a) showed weaker activity
than SN-38, whereas the analogs having hydrophobic side chains
(6b–e, 7b–e) exhibited activities comparable to or stronger than
SN-38. From the introduction of a methyl or hydroxymethyl group
on the amidine carbon (8c, 9d), the antiproliferative activities were
retained. Among the analogs synthesized, 5d, 6d, and 7d with an
isopentyl side chain exhibited most potent activities with IC50 val-
ues against human colon cancer cell line HCT116 of 0.02, 0.05, and
<0.04 nM, respectively.
These analogs were further evaluated for their in vivo antitumor
activity against human cancer xenograft HCT116 (CRC) as well as
the effective dose range, defined as the range from the effective
dose for 50% tumor growth inhibition (ED50) to the maximum tol-
Table 2
Effect of BCRP on antiproliferative activities of SN-38 and 7c in PC-6/BCRP and PC-6/
pRC human small cell lung cancer cell lines
IC50 (nM)[Mean S.E.]
SN-38
7c
PC-6/BCRP
PC-6/pRC
Ratio
5.1 1.8
0.43 0.07
12
0.35 0.04
0.18 0.01
2
The PC-6/BCRP and PC-6/pRC cells were established by transfecting the BCRP gene
and vector, respectively, into PC-6 human small-cell lung cancer cells. The cells
were exposed to test compounds for 6 days. The experiments were performed in
triplicate.
2000
3000
2500
2000
1500
1000
500
NCI-H460
(NSCLC)
BCRP+
HCT116
(CRC)
BCRP-
1500
1000
500
*
*
0
10
0
20
30
40
5
15
25
35
Days after tumor inoculation
Figure 4. Antitumor effect of CH0793076 and CPT-11 in human cancer xenograft models. CH0793076 and CPT-11 were administered at the maximum tolerated dose (MTD)
by bolus intravenous injection once per week for 3 weeks. The MTD was 40 mg/kg for CH0793076 and 100 mg/kg for CPT-11. Each group consisted of 4 to 5 mice. Values
indicate the mean tumor volume with standard deviation. s: vehicle, N: CPT-11, d: CH0793076. BCRP protein was detected by Western blotting. : Statistically significantly
*
differences in mice treated with CH0793076 compared with mice treated with CPT-11 (P < 0.05).

S. Niizuma et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2018–2021
2021
erated dose (MTD). From the results, 7c (CH0793076) was selected
as a candidate for further development.
tent antitumor activity in human cancer xenograft models than
CPT-11, regardless of the level of BCRP expression. Thus, 7c was se-
lected as a parent drug for further development of the new water-
soluble prodrug which enables intravenous administration.
Wehavedevelopedanewwater-solubleprodrugof7cthatcanbe
activated by a non-enzymatic process, namely pH-dependent acti-
vation. The results will be reported in a separate paper.
7c showed somewhat stronger inhibitory activity (IC50 = 2.3
lM)
against human topoisomerase I than SN-38 (IC50 = 5.5 M). It has
l
been reported that CPT-11 is sensitive to the drug efflux pump BCRP
(breast cancer resistant protein) that is overexpressed in several hu-
man cancer cell lines.8 Although information on expression levels in
the tumors in cancer patients is still limited, BCRP is thought to be
one of the factors that limit the antitumor activity of CPT-11. We
tested the effect of BCRP on the antiproliferative activity of SN-38
and 7c in human small cell lung cancer cell line PC-6 that had been
transfected with a BCRP gene (PC-6/BCRP) and with the vector alone
(PC-6/pRC). The activity of SN-38 showed a decrease in the BCRP-
gene transfected cells, but, in contrast, 7c was only marginally influ-
enced by the presence of BCRP (Table 2).
Acknowledgments
We are grateful to Mr. Toshihiko Fujii and Mr. Kiyoaki Sakata for
conducting the in vitro antiproliferative assays.
References and notes
The antitumor activity of 7c (one a week iv treatment ꢀ 3 at the
MTD) was further evaluated in two cancer xenograft models: hu-
man colon cancer HCT116 (BCRP negative) and human non-small
cell lung cancer NCI-H460 (BCRP positive).
The results are shown in Figure. 4. Compound 7c exhibited sig-
nificantly higher efficacy than CPT-11 in these two human cancer
xenografts regardless of the level of BCRP expression.
In summary, we successfully designed and synthesized new
hexacyclic camptothecin analogs based on the proposed structure
of the Topo I/DNA/camptothecin ternary complex. CH0793076 (7c)
is not a substrate of drug efflux pump BCRP and showed more po-
1. Hecht, J. R. Am. J. Health Syst. Pharm. 2008, 65, S15–S21.
2. Wu, D. Tetrahedron 2003, 59, 8649–8687.
3. Liew, S. T.; Yang, L.-X. Curr. Pharm. Des. 2008, 14, 1078–1097.
4. Mitsui, I.; Kumazawa, E.; Hirota, Y. Jpn. J. Cancer Res. 1995, 86, 776–782.
5. Redinbo, M. R.; Stewart, L.; Kuhn, P.; Champoux, J. J.; Hol, W. G. J. Science 1998,
279, 1504–1513.
6. Staker, B. L.; Hjerrild, K.; Feese, M. D.; Behnke, C. A.; Burgin, A. B., Jr.; Stewart, L.
Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 15387–15392.
7. Cao, Z.; Harris, N.; Kozielski, A.; Cardeman, D.; Stehlin, J. S.; Giovanella, B. J. Med.
Chem. 1998, 41, 31–37.
8. Kawabata, S.; Oka, M.; Shiozawa, K.; Tsukamoto, K.; Nakatomi, K.; Soda, H.;
Fukuda, M.; Ikegami, Y.; Sugahara, K.; Yamada, Y.; Kamihira, S.; Doyle, L. A.;
Ross, D. D.; Kohno, S. Biochem. Biophys. Res. Commun. 2001, 280, 1216–1223.