Synthesis of derivatives of NK109, 7-OH benzo[c]phenanthridine alkaloid, and evaluation of their cytotoxicities and reduction-resistant properties

Article abstract of DOI:10.1016/S0960-894X(00)00467-4

The N₅-C₆ double bond of NK109 (an antitumor benzo[c]phenanthridine alkaloid) is easily reduced under biological environment. To suppress the inactivation caused by reduction, we synthesized 5-, 6-, and 8-substituted NK109. 5-Substituted derivatives (4a-c) were reduced more easily than NK109. 6-Substituted ones (10a-f) inhibited biological reduction, but showed weak cytotoxic activity. 8-O-Substituted ones (13a-h), especially 8-O-hydroxyethyl NK109 (13d), suppressed biological reduction and exhibited strong cytotoxic activity. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00467-4

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2321±2323
Synthesis of Derivatives of NK109, 7-OH Benzo[c]phenanthridine
Alkaloid, and Evaluation of their Cytotoxicities and
Reduction-Resistant Properties
Takeshi Nakanishi,* Akira Masuda, Masato Suwa, Yuji Akiyama, Naoko Hoshino-Abe
and Masanobu Suzuki
Pharmaceuticals Group, Nippon Kayaku Company, Ltd., 31-12, Shimo 3-chome, Kita-ku, Tokyo 115±8588, Japan
Received 2 June 2000; accepted 4 August 2000
AbstractÐThe N5±C6 double bond of NK109 (an antitumor benzo[c]phenanthridine alkaloid) is easily reduced under biological
environment. To suppress the inactivation caused by reduction, we synthesized 5-, 6-, and 8-substituted NK109. 5-Substituted
derivatives (4a±c) were reduced more easily than NK109. 6-Substituted ones (10a±f) inhibited biological reduction, but showed
weak cytotoxic activity. 8-O-Substituted ones (13a±h), especially 8-O-hydroxyethyl NK109 (13d), suppressed biological reduction
and exhibited strong cytotoxic activity. # 2000 Elsevier Science Ltd. All rights reserved.
Results and Discussion
Introduction
The presence of a hydroxy group at the 7-position of
NK109 (1) is a synthetic benzo[c]phenanthridine alkaloid
(BCPA) with signi®cant antitumor activity. It exhibits
stronger activity against P388 leukemia than the widely
known BCPAs, nitidine and fagaronine.1 NK109 inhibits
topoisomerase II (Topo II) by stabilization of a clea-
vable complex of DNA and Topo II,2 and is e€ective
against several drug-resistant tumor cell lines.3 Thus, we
conducted a clinical trial of NK109 in Japan.4 Although
NK109 possesses sucient antitumor activity for clinical
trial, it is easily metabolized under biological environ-
ment. A metabolic investigation of NK109 in mice
revealed an important metabolic pathway to be reduction
to a 5,6-dihydro NK109 2 (Fig. 1). NK109 strongly
inhibits the growth of tumor cells in vitro. However, a
high dose of NK109 is required to exhibit strong anti-
tumor activity in vivo,1 because the major metabolite (2)
is inactive. In order to improve the metabolic stability of
NK109, we decided to synthesize derivatives of NK109
with reduction-resistant property. In this paper, we
report the synthesis of NK109 derivatives having 5-, 6-,
and 8-O-substituents. We also report their cytotoxicities
against HeLa S3 cells and reduction-resistant properties
using human liver S9 mix.
NK109 is unique among BCPAs and is essential for exhi-
bition of antitumor activity.1,5 We thus modi®ed 5-CH3,
6-H, and 8-OCH3 of NK109 with retaining the 7-OH
group. First, benzo[c]phenanthridine 36 was treated with
several alkylating agents and followed by debenzylation
with acid to yield N-substituted derivatives 4a±d (Table 1).
We next synthesized 6-substituted NK109 (Scheme 1)
from compound 5, which is a synthetic equivalent of
quaternary BCPA 6 soluble in organic solvent.6 Com-
pound 5 was treated with several Grignard reagents in
THF to give 6-alkyl intermediates 7. They are 5,6-dihydro
BCPA and, therefore, aromatized with MnO2 in toluene.
When R2 is an alkyl group, the oxidation proceeded
with removal of the N-methyl group to yield benzo[c]
phenanthridine 8. After 8 was N-methylated again with
methyl tri¯uoromethane sulfonate, the product was
debenzylated with HCl to give 6-alkyl NK109 10a±d
(Table 2). When R2 is an aryl group, the oxidation pro-
ceeded without removal of the N-methyl group to yield 9.
It was debenzylated with HCl to give 6-aryl NK109 10e±f.
Finally, we synthesized 8-O-substituted NK109 (Scheme
2). A quaternary BCPA 11, synthesized by the synthetic
procedure of NK109,6 was reduced to 7,8-dihydroxy
BCPA 12 by catalytic hydrogenation. It was treated
with several alkyl halides in the presence of potassium
carbonate in acetone.7 The reaction mixture included
*Corresponding author. Tel.: +81-3-3598-5241; fax: +81-3-3598-5422;
e-mail: t_naka@fra.allnet.ne.jp
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00467-4

2322
T. Nakanishi et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2321±2323
Table 2. Yield of 6-alkylation of NK109
Product (yield %)
Alkylating
agent
Substituent (R)
8, 9
10
Me
Et
Pr
c-Pr
Ph
p-(OMe)-Ph
MeMgBr
EtMgBr
PrMgBr
8a (53)
8b (54)
8c (59)
8d (65)
9e (87)
9f (54)
10a (91)
10b (35)
10c (28)
10d (32)
10e (75)
10f (93)
Figure 1. Reduction of NK109 (1).
c-PrMgBr
PhMgBr
p-(OMe)-PhMgBr
Table 1. Synthesis of 5-substituted NK109 (4)
and the reduced product were observed on the HPLC
chromatogram. The synthetic derivatives were also
analyzed in the same way. After calculating the ratio of
the reduced products, we judged the reduction-resistant
property based on comparison of the reduction rate of
NK109 and those of derivatives. These results are
shown in Table 4. Although the reduction rate of
5-substituted derivatives 4 is higher than that of NK109,
that of 8-O-substituted derivatives 13 is lower than that
of NK109. In the case of 6-substituted derivatives 10, no
reduced products were observed. Thus, 8-O-substituents
partially contribute to resistance to the biological
reduction and 6-subsituents completely do. On the other
hand, 5-substituents accelerated the reduction (Fig. 2).
Alkylating agenta
Product (R1)
Yield (%)
EtOTf
ONBSP
c-PrBr-AgOTf
MeOCOCH2OTf
4a (Et)
4b (Pr)
4c (Allylb)
4d (CH2CO2H)
75
14
17
81
aAbbreviations of alkylating agents are as follows: EtOTf, ethyl tri-
¯uoromethanesulfonate; ONBSP, n-propyl 2-nitrobenzenesulfonate
(see ref 6); c-PrBr, c-propyl bromide.
bThe reaction of c-propyl cation and 3 yielded N-allyl derivative, not
N-c-propyl derivative.
mono-O-alkylated and di-O-alkylated products which,
after oxidation with DDQ, were isolated by silica gel
chromatography to give 13±15 (Table 3). Compounds
13 and 14 are easily detected by TLC monitoring,
because they show purple and yellow spots on the TLC
plate, respectively.
We also evaluated the cytotoxicities of NK109 derivatives
against HeLa S3 cells according to the previously reported
procedures.1 These results are also listed in Table 4.
We synthesized several derivatives of NK109 aiming at
those with reduction-resistant property. 6-Substituted
derivatives 10 completely became resistant to the bio-
logical reduction independent from the kind of sub-
stituent; however, they also lost their cytotoxic activity.
It is generally believed that the immonium group of
BCPA is necessary for their antitumor activity,10 and
thus the immonium group would be concerned in the
mode of action.11 In the case of 10, 6-substituents will
a€ect their antitumor activity. 8-O-Substituted derivatives
13 partially inhibit biological reduction. Derivatives with
bulky hydrophobic substituents such as an isopropyl or a
benzyl group (13a and 13h) showed weak activities,
probably because these substituents are not suitable for
the environment of the binding site of Topo II. Those
with hydrophilic substituents such as hydroxyethyl (13c
and 13d) showed similar activity as NK109. 5-Substituted
We evaluated reduction-resistant properties of the
NK109 derivatives 4, 10 and 13, using human liver S-9
mix. First, we detected their reduction product peaks by
HPLC analysis.8 NK109 was reduced quantitatively
with NaBH3CN to 5,6-dihydro NK109,1 and 4, 10, 13
were also converted with NaBH3CN to the correspond-
ing dihydro derivatives almost quantitatively on HPLC
chromatogram. Based on the retention time of each
product, we identi®ed the converted peak to be the cor-
responding reduced product. Subsequently, NK109 and
its derivatives were treated with human liver S-9 mix
and NADPH.9 After incubation at 37 ꢀC for 30 min,
they were analysed on HPLC. Under this condition,
NK109 was reduced partially and peaks of both intact
Scheme 1. Synthesis of 6-substituted NK109 (10). Reagent: (a) R2MgBr, THF; (b) MnO2, toluene; (c) DDQ, CH2Cl2; (d) MeOTf, toluene; (e) HCl,
AcOH.

T. Nakanishi et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2321±2323
2323
Figure 2. Substituent e€ect on activity.
derivatives showed strong activity. However, they were
reduced faster than NK109. In view of the reduction
rate and synthetic diculties,12 5-substituted ones are
not superior to NK109. We also found that derivatives
with a carboxy group (4d and 13g) lost activity. The
reason would be the poor solubility in aqueous media
and poor penetration into the cells.
Scheme 2. Synthesis of 8-O-Substituted NK109 (13). Reagent: (a) H2, Pd-
C, MeOH±H2O; (b) R-Br, K2CO3, acetone; (c) DDQ, CH2Cl2; (d) HCl.
Table 3. Yield of 7- and 8-alkylation of O-demethyl-NK109
Product (yield%)
Alkylating
Agent
C8-OR
C7-OR
C7,8-(OR)2
In conclusion, derivatives with substituents at the
6-position completely became resistant to the biological
reduction. However, they are unsatisfactory because
they showed weak cell growth inhibitory e€ect. The
substituent at the 8-position suppresses biological
reduction. Unless the 8-O-substituent is not bulky and
hydrophobic, they show similar activity as NK109. 8-O-
Hydroxyethyl derivatives of NK109 (13c and 13d) possess
both reduction-resistant properties and strong cytotoxi-
cities, and are therefore expected as new lead com-
pounds to back-up NK109.
Substituent (R)
13
14
15
i-Pr
Allyl
i-PrBr
Allyl-Br
Br(CH2)2OAc
Ð
ClCH2CONH2
BrCH2CO2Me
Ð
13a (14)
13b (6)
13c (9)
13da (5)
13e (4)
13f (4)
13gb (4)
13hc
14a (4)
Ð
14c (3)
Ð
14e (2)
Ð
Ð
Ð
15b (52)
15c (2)
Ð
15e (6)
15f (13)
15gb (1)
Ð
(CH2)2OAc
(CH2)2OH
CH2CONH2
CH2CO2Me
CH2CO2H
Bzl
Ð
Ð
aObtained together with 13c±15c by the reaction of 12 and Br(CH2)2OAc.
bObtained together with 13f and 15f by the reaction of 12 and
BrCH2CO2Me.
cObtained by partial debenzylation of 11: compound 11 and tri¯uoroacetic
acid were warmed at 50 ꢀC for 1 h to give 13h (33%).
References and Notes
1. Nakanishi, T.; Suzuki, M.; Saimoto, A.; Kabasawa, T. J.
Nat. Prod. 1999, 62, 864.
2. (a) Fukuda, M.; Inomata, M.; Nishio, K.; Fukuoka, K.;
Kanzawa, F.; Arioka, H.; Ishida, T.; Fukumoto, H.; Kuro-
kawa, H.; Oka, M.; Saijo, N. Jpn. J. Cancer Res. 1996, 87,
1086. (b) Kabasawa, T.; Kobayashi, F.; Ekimoto, H.; Suzuki,
M.; Hanaoka, M. Proc. Am. Assoc. Cancer Res. 1996, 37, 427.
3. Kanzawa, F.; Nishio, K.; Ishida, T.; Fukuda, M.; Kuro-
kawa, H.; Fukumoto, H.; Nomoto, Y.; Fukuoka, K.; Boja-
nowski, K.; Saijo, N. Br. J. Cancer 1997, 76, 571.
4. Tamura, T.; Yamamoto, N.; Yamamoto, N.; Kamiya, Y.;
Shimada, Y.; Ieda, Y.; Nakagawa, K.; Fukuoka, M.; Saijo, N.
Abstracts of Papers, 58th Annual Meeting of the Japanese
Cancer Assocation: Hiroshima, Japan, 1999; P-13, 2406.
5. Nakanishi, T.; Suzuki, M. Org. Lett. 1999, 1, 985.
6. Nakanishi, T.; Suzuki, M.; Mashiba, A.; Ishikawa, K.;
Yokotsuka, T. J. Org. Chem. 1998, 63, 4235.
Table 4. Evaluation of substituted NK109
Reduction-resistant property
Cytotoxicity
IC50 (mM)
Compound
Relative ratea
Judgementb
NK109
4a
4b
4c
4d
1.0
3.4
1.6
2.6
0.21
0.32
0.40
0.76
0.49
+
>24
10a
10b
10c
10d
10e
10f
n.d.c
n.d.
n.d.
n.d.
n.d.
n.d.
++
++
++
++
++
++
2.9
11
2.9
6.5
17
16
7. In other solvents such as MeOH, CH3CN, and THF, no
alkylating product was obtained.
8. HPLC conditions: Column, Inertsil C8 6.0Â200 mm;
mobile phase, 1% H3PO4+5 mM n-Bu4NBr:CH3OH (65:35);
¯ow rate, 1.5 mL/min; detection, UV (344 nm).
9. (a) Kitamura, S.; Tatsumi, K. Arch. Biochem. Biophys.
1990, 282, 183. (b) Eyles, D. W.; Pond, S. M. Biochem. Phar-
macol. 1992, 44, 867.
13a
13b
13c
13d
13e
13f
13g
13h
0.81
2.4
+
2.4
0.71
0.48
0.52
1.9
15
>24
17
d
n.d.
0.08
0.57
0.96
0.42
0.42
+
+
+
+
+
+
aRelative reduction rate to NK109: (reduction rate of derivative)/(reduc-
tion rate of NK109). The reduction rate is calculated as follows: R/(R+I);
`R' is the peak area of reduced product under the HPLC chromatograph,
and ``I'' is the peak area of intact compound.
10. (a) Stermitz, F. R.; Larson, K. A.; Kim, D. K. J. Med.
Chem. 1973, 16, 939. (b) Zee-Cheng, R. K.-Y.; Yan, S.-J.;
Cheng, C. C. J. Med. Chem. 1978, 21, 199.
11. For example, the interaction with tublin: Wol€, J.; Knip-
ling, L. Biochemistry 1993, 32, 13334.
12. The N-alkylation of BCPA is a€ected by the size of sub-
stituent because of the steric e€ect of 4-H. Therefore, even the rate
of N-ethylation is much slower than that of N-methylation.
bDe®ned as follows: ++, reduced product peak is not detected (relative
rate=0); +, relative rate < 1; , relative rate >1.
cThe reduced product peak was not detected.
dAlthough reduced product peak of 13c was not detected, its metabolites,
13d and reduced product peak of 13d, were observed.