464
Chemistry Letters 2000
First Synthesis of Subelliptenone F, an Inhibitor of Topoisomerase II
Susumu Ohira,* Norihiro Fukamachi, Osamu Nakagawa, Masashi Yamada,† Hiroshi Nozaki, and Munekazu Iinuma††
Department of Biological Chemistry, Faculty of Science, Okayama University of Science, 1-1 Ridai-cho, Okayama 700-0005
†Meiji Milk Products Co., LTD., 26-11 Midori 1 chome, Sumida-ku, Tokyo 130-8502
††Gifu Prefectural Institute of Health and Environmental Sciences, 1-1 Naka-fudogaoka, Kakamigahara 504-0838
(Received January 31, 2000; CL-000100)
Subelliptenone F, a potent inhibitor of topoisomerase II,
Given those results, we finally decided to use
methoxymethyl (MOM) group instead of methyl group, and
trimethylsilylethoxymethyl (SEM) group5 for protection of 2-
hydroxyl group of A ring. It was expected that fluoride anion
mediated deprotection of SEM group would produce the phe-
noxy anion that undergoes simultaneous cyclization.
was synthesized for the first time via coupling of A and C ring
components and B ring formation, followed by Claisen
rearrangement of the 3-methyl-2-butenyl ether.
Subelliptenone F (1) was isolated from root bark of the
Guttifereae plant Garcinia subelliptica and characterized by
spectroscopic analysis in 1994.1 Thereafter, 1 was found to
show a potent topoisomerase II inhibitory activity as well as the
other xanthones isolated from Guttiferaeous plants.2 It exhibit-
ed higher activity than etopside that is already used as a clinical
drug. For further study of the biological activity of 1 and the
structure-activity relationship, we planned to develop an effec-
tive approach to 1 and its analogues. Described here is the first
total synthesis of subelliptenone F through coupling of A and C
ring components, and then B-ring formation followed by
Claisen rearrangement to introduce the 1,1-dimethyl-2-propenyl
substituent.
In preliminary studies toward synthesis of 1 and its ana-
logues, we had already gained some insights (Scheme 1). On
treatment of 2 with tetramethylammonium hydroxide in pyri-
dine-water (1:1) at refluxing temperature,3 cyclization on the
desired mode and simultaneous Claisen rearrangement
occurred, giving 3 in 56% yield. Dimethyl ether 4, which arose
from the attack of the phenoxide to the other site, was produced
in 15% yield. Demethylation of 3 was successful using excess
lithium diphenylphosphide,4 although use of BBr3 or BCl3
destroyed the vinyl substituent of 3 with liberated hydrogen
halide. In the synthesis of the xanthone 6, Claisen rearrange-
ment at an earlier stage followed by B ring formation was effec-
tive, since the 3-methyl-2-butenyl group on C ring was totally
eliminated under the acidic condition to liberate two hydroxyl
groups of A ring. Cyclization of 5 and demethylation of the
resultant 6 gave the corresponding phenol in moderate yield.
This methodology, however, did not work at all on the synthe-
sis of 9. Similar treatment of tetrahydroxy benzophenone
derivative 7 gave no desired product 9 but a mixture of
unknown products. On the contrary, suitably protected phenol
8 gave the cyclized product 10 in good yield. In this case, how-
ever, removal of all the methyl protecting groups of 10 was
unsuccessful under various conditions.
Thus, C ring component 13 was synthesized as shown in
Scheme 2. Methoxyhydroquinone 11 was acetylated to diac-
etate, whose less hindered site was selectively hydrolyzed to
give monoacetate 12. The relationship of acetyl group and
methoxy group of 12 was confirmed by NOE experiment.6 The
hydroxyl group of 12 was alkylated with 1-chloro-3-methyl-2-
Copyright © 2000 The Chemical Society of Japan