Total Synthesis and Absolute Configuration of Otteliones A and B, Novel and Potent Antitumor Agents from a Freshwater Plant

Article abstract of DOI:10.1021/ol035443x

(Matrix presented) The first enantioselective total synthesis of otteliones A and B, biologically important and structurally novel natural products, has been successfully achieved. This total synthesis fully confirms the absolute configuration of these natural products.

Full text of DOI:10.1021/ol035443x

ORGANIC
LETTERS
2003
Vol. 5, No. 21
3903-3906
Total Synthesis and Absolute
Configuration of Otteliones A and B,
Novel and Potent Antitumor Agents
from a Freshwater Plant
,†,‡
Hiroshi Araki,† Munenori Inoue,‡ and Tadashi Katoh*
Department of Electronic Chemistry, Tokyo Institute of Technology,
Nagatsuta, Yokohama 226-8502, Japan, and Sagami Chemical Research Center,
2743-1 Hayakawa, Ayase, Kanagawa 252-1193, Japan
katoh@sagami.or.jp
Received July 31, 2003
ABSTRACT
The first enantioselective total synthesis of otteliones A and B, biologically important and structurally novel natural products, has been
successfully achieved. This total synthesis fully confirms the absolute configuration of these natural products.
Otteliones A (1) and B (2) (Figure 1) were isolated by Ayyad
and Hoye et al. in 1998 from the freshwater plant Ottelia
alismoides collected in the Nile Delta, Egypt.1,2 These sub-
stances were found to exhibit prominent biological properties
such as antitubercular3 and antitumor1,2 activities. Remark-
ably, these small-molecule natural products displayed quite
potent cytotoxicity at nM-pM levels of IC50 values against
a panel of 60 human tumor cell lines at the National Cancer
Institute in the United States.1,4 It has also been reported that
1 inhibits tubulin polymerization into microtubules (IC50
)
Figure 1. Structures of otteliones A (1) and B (2).
1.2 µM) similarly to well-known colchicine, vincristine, and
vinblastine.5 Consequently, otteliones are now serving as new
† Tokyo Institute of Technology.
leads for the development of cancer chemotherapeutic agents.
However, further biological studies are severely limited by
the scarcity of these natural products.6
‡ Sagami Chemical Research Center.
(1) Ayyad, S.-E. N.; Judd, A. S.; Shier, W. T.; Hoye, T. R. J. Org. Chem.
1998, 63, 8102.
(2) Incidentally, ottelione A (1) was identical with the compound named
RPR112378 previously reported in the patent literature by a research group
at Rhoˆne-Poulenc Rorer (now Aventis); see: Leboul, J.; Provost, J. French
Patent WO96/00205 1996;Chem. Abstr. 1996, 124, 242296.
(3) Li, H.; Qu, X.; Shi, Y.; Guo, L.; Yuan, Z. Zhongguo Zhongyao Zazhi
(Chin. J. Chin. Mater. Med.) 1995, 20, 115, 128.
(4) Ottelione A (1) has been also shown to inhibit doxorubicin-resistant
leukemia cells (P388/DOX) with an IC50 value of 1 ng/mL (3 nM); see ref
1.
Structurally, otteliones A and B possess a novel bicyclic
hydrindane skeleton with four contiguous asymmetric car-
(5) Combeau, C.; Provost, J.; Lancelin, F.; Tournoux, Y.; Prod’homme,
F.; Herman, F.; Lavelle, F.; Leboul, J.; Vuilhorgne, M. Mol. Pharm. 2000,
57, 553.
(6) Otteliones A (1) and B (2) had been obtained in 0.0009% yield (dry
weight) from the plant Ottelia alismoides; see ref 1.
10.1021/ol035443x CCC: $25.00 © 2003 American Chemical Society
Published on Web 09/23/2003

bons, in which the rare and highly sensitive 4-methylene-
2-cyclohexenone substructure is a particularly characteristic
feature. While the relative configurations of otteliones A and
B were deduced by high-field NMR spectroscopy,1,5 their
absolute configurations remain uncertain. The remarkable
biological properties, unique structural features, and limited
availability from natural resources, as well as the need to
confirm the absolute configuration, have made the otteliones
exceptionally intriguing and timely targets for total synthesis.
A number of synthetic approaches toward otteliones A and
B have been reported to date.7 The first elegant total synthesis
of racemic (()-otteliones A (1) and B (2) was recently
achieved by Mehta et al.8 and provided proof of their relative
stereochemistry.
group of 5 would be masked as an internal hemiacetal
moiety, the advanced key intermediate 3 would be elaborated
through the coupling reaction of 5 with aryllithium 4
followed by epimerization at C1 and further functionalization.
The intermediate 3 would be converted to the target mole-
cules 1 and 2 by sequential functional group manipulation
and deprotection, or vice versa. Intermediate 5 could in turn
be accessed from the known tricyclic compound 6,10 which
has previously been prepared in an enantiomerically pure
form in our laboratory.
At first, as shown in Scheme 2, we pursued the synthesis
of the key intermediate 5 starting with the tricyclic compound
Recently, we embarked on a project directed at the total
synthesis of optically active otteliones A and B, as well as
their analogues, with the aim of exploring their structure-
activity relationships. In this communication, we report the
first enantioselective total synthesis of otteliones A (1) and
B (2), which proves their absolute configuration to be as
depicted in Figure 1.9
Scheme 2. Synthesis of Key Intermediate 5a
The key element of our synthetic plan for 1 and 2 outlined
in Scheme 1 is our utilization of the highly and appropriately
Scheme 1. Synthetic Plan for Otteliones A (1) and B (2)
a Reagents and conditions: (a) NaBH4, THF-H2O, 0 °C, 87%.
(b) OsO4, NaIO4, t-BuOH-THF-H2O, 0 °C f rt, 62%.
6.10 Thus, sodium borohydride reduction of 6 provided the
expected product 7 in 87% yield. Subsequent oxidative
cleavage of the olefin moiety in 7 was carried out by
employing the Lemieux-Johnson procedure,11 which led to
the production of the cyclic hemiacetal 5 in 62% yield
through the intermediary dialdehyde 8. The stereochemical
issue with respect to the C8 position in 5 was assigned as
depicted on the basis of NOESY experiment in the 500 MHz
1H NMR spectrum.12
With the key intermediate 5 in hand, we next investigated
the synthesis of the advanced key intermediate 3 having all
four correct stereogenic centers (C1, C3, C3a, and C7a) and
the required functionalities for elaboration to the target
molecules 1 and 2 (Scheme 3). Thus, aryllithium 4 generated
in situ by treatment of 4-bromo-1-methoxy-2-(methoxymeth-
oxy)benzene (9)13 with n-butyllithium in THF at -78 °C
was allowed to react with 5 at the same temperature, pro-
viding the coupling products 10a (80%) and 10b (20%) as
functionalized intermediate 5, which contains not only the
requisite bicyclic hydrindane core structure having the correct
stereogenic centers at C3, C3a, and C7a but also the desirable
functionalities for elaboration of 1 and 2. Since the C1 formyl
(7) (a) Clive, D. L. J.; Fletcher, S. P. Chem. Commun. 2002, 1940. (b)
Mehta, G.; Islam, K. Org. Lett. 2002, 4, 2881. (c) Mehta, G.; Islam, K.
Synlett 2000, 1473. (d) Trembleau, L.; Patiny, L.; Ghosez, L. Tetrahedron
Lett. 2000, 41, 6377. (e) Mehta, G.; Reddy, D. S. Chem. Commun. 1999,
2193.
(10) (a) Izuhara, T.; Katoh, T. Tetrahedron Lett. 2000, 41, 7651. (b)
Izuhara, T.; Katoh, T. Org Lett. 2001, 3, 1653.
(8) Mehta, G.; Islam, K. Angew. Chem., Int. Ed. 2002, 41, 2396.
(9) After submission of this manuscript, we learned that Mehta and Islam
had completed an enantioselective total synthesis of both enantiomers of
otteliones A (1) and B (2) and elucidated the absolute configurations of
naturally occurring 1 and 2; see: Mehta, G.; Islam, K. Tetrahedron Lett.
2003, 44, 6733. The absolute configuration assignments independently
achieved by the Mehta group are compatible with our own results described
in this communication.
(11) Pappo, R.; Allen, D. S.; Lemieux, R. U.; Johnson, W. S. J. Org.
Chem. 1956, 21, 478.
(12) NOE between C8-H and C2-HR was clearly observed.
(13) Compound 9 was prepared from commercially available 2-meth-
oxyphenol via a four-step sequence of reactions [(a) BzCl, pyridine, rt, 98%;
(b) Br2, AcOH, rt; (c) K2CO3, MeOH, rt, 62% (two steps); (d) MOMCl,
i-Pr2NEt, CH2Cl2, rt, 88%] (see Supporting Information for experimental
details).
3904
Org. Lett., Vol. 5, No. 21, 2003

12, and treatment of the resulting triacetate 13 with a large
excess of lithium (100 equiv) in liquid ammonia at -78 °C.
Simultaneous oxidation of the primary and secondary hy-
droxy groups in 14 by the use of Dess-Martin periodinane
furnished the corresponding keto-aldehyde 15 (90%), which
was then subjected to twofold Wittig methylenation to
provide the desired key intermediate 3 in 95% yield.
The final route that led to completion of the synthesis of
the targeted otteliones A (1) and B (2) is shown in Scheme
4, which involves the critical elaboration of the highly
Scheme 3. Synthesis of Key Intermediate 3a
Scheme 4. Synthesis of Otteliones A (1) and B (2)a
a Reagents and conditions: (a) 4-bromo-1-methoxy-2-(meth-
oxymethoxy)benzene (9), n-BuLi, THF, -78 °C; at -78 °C add
5, 80% for 10a, 20% for 10b. (b) DBU, toluene, reflux, 30% (60%
recovery of 10a). (c) LiAlH4, THF, 0 °C, 96%. (d) Ac2O, DMAP,
pyridine, rt, 95%. (e) Li, liquid NH3, THF, -78 °C, 98%. (f) Dess-
Martin periodinane, CH2Cl2, rt, 90%. (g) Ph3P+CH3Br-, t-BuOK,
benzene, rt f reflux, 95%.
a Reagents and conditions: (a) TFA, THF-H2O, 0 °C, 86%. (b)
Ac2O, 2 M NaOH, i-PrOH, rt, 91%. (c) CSCl2, DMAP, CH2Cl2,
rt, 93%. (d) (EtO)3P, reflux, 78%. (e) TBAF, THF, rt, 83%. (f)
(CHCl2CO)2O, pyridine, CH2Cl2, rt, 52%. (g) Dess-Martin perio-
dinane, CH2Cl2, rt, 95%. (h) 50% aq NaHCO3-MeCN (1:1), rt,
a mixture of epimeric alcohols that were separated by column
chromatography on silica gel.14 The crucial base-induced
cyclic hemiacetal opening/epimerization of the liberated C1
formyl group in the major product 10a turned out to be
effected by treatment with 1,8-diazabicyclo[5.4.0]undec-7-
ene (DBU) in refluxing toluene for 1.5 h, leading to the
formation of the requisite product 11 (30%) along with the
starting material 10a (60%). Since prolonged reaction time
caused decomposition of 11 and/or 10a, the reaction was
terminated when an approximately 1:2 mixture of 11:10a
was generated. After recycling of unreacted starting material
10a four times, we could obtain ∼65% yield of the desired
compound 11. Since the same treatment of the minor product
10b turned out to be fruitless and resulted in recovery of
the starting material, the projected synthesis was carried
forward using only the major product 10a.15 Compound 11
was further converted to the diol 14 in 89% overall yield
via a three-step sequence involving reduction of the formyl
group, complete acetylation of the three hydoxy groups in
1
90%. (i) t-BuOK, t-BuOH, rt, 79% (1:2 ) 23:77 by H NMR);
isolation of 2 by HPLC, 23%.
sensitive 4-methylene-2-cyclohexenone system. Toward this
end, deprotection of both the acetonide moiety and the MOM
group in 3 by exposure to trifluoroacetic acid (TFA) in the
presence of water followed by chemoselective acetylation
of the phenolic hydroxy group in the resulting triol 16
afforded the acetate 17 in 78% yield for the two steps. To
install the requisite C5-C6 double bond, we decided to
employ a reliable Corey-Winter procedure.16 Thus, treatment
of the diol 17 with thiophosgene in the presence of
4-(dimethylamino)pyridine (DMAP) provided the cyclic
thiocarbonate 18 (93%), which was subsequently subjected
to reductive elimination of the thiocarbonate moiety by
reaction with triethyl phosphite, leading to the formation of
the desired product 19 in 78% yield. Removal of the TBS
and Ac protecting groups from 19 by treatment with
tetrabutylammonium fluoride (TBAF) followed by chemose-
(14) Stereochemistry at the benzylic carbon in the coupling products
10a,b was tentatively assigned on the basis of NOESY experiments in the
1H NMR spectra.
(15) Detailed discussions on this hemiacetal opening/epimerization
reaction will be presented in a full account.
(16) (a) Corey, E. J.; Winter, R. A. E. J. Am. Chem. Soc. 1963, 85, 2677.
(b) Corey, E. J.; Carey, F. A.; Winter, R. A. E. J. Am. Chem. Soc. 1965,
87, 934.
Org. Lett., Vol. 5, No. 21, 2003
3905

lective protection of the phenolic hydroxy group in the
resulting diol 20 furnished the dichloroacetate 2117 in 43%
yield for the two steps. Finally, compound 21 was converted
to ottelione A (1) in 86% overall yield via a two-step
sequence involving Dess-Martin oxidation of the C4 hy-
droxy group in 21 and removal of the dichloroacetyl group
in the resulting dienone 22.
was essentially identical to that of natural 2 (contaminated
25
with a small amount of 1, 2:1 ) 85:15) [ [R]D -276°
(c 2.0, CHCl3)18], indicating that natural 2 possesses the
(1S,3S,3aS,7aS) absolute configuration as depicted in the
formulation 2.
In summary, the first enantioselective total synthesis of
otteliones A (1) and B (2) has been achieved starting from
the readily available tricyclic compound 6. The absolute
configuration of 1 and 2 was determined through the present
synthesis. Our approach involves a coupling reaction of the
cyclic hemiacetal 5 with aryllithium 4 (4 + 5 f 10a,b),
base-induced cyclic hemiacetal opening/epimerization at the
C1 position of 10a (10a f 11), and Corey-Winter’s
reductive elimination of the cyclic thiocarbonate 18 (18 f
19) as pivotal steps. The explored synthetic pathway should
hold promise for preparing various ottelione analogues in
enantiomerically pure form due to its generality and flex-
ibility. These efforts are currently underway.
The spectroscopic properties (IR, 1H and 13C NMR, MS)
of the synthetic sample 1 were compatible with those of the
natural product 1.1 Comparison of the optical rotation
25
[synthetic 1, [R]D25 +17.3° (c 0.55, CHCl3); natural 1, [R]D
+14.0° (c 0.87, CHCl3)18] established the absolute config-
uration of ottelione A (1) to be (1S,3S,3aR,7aS) as pictured
in Scheme 4. Epimerization at C3a in ottelione A (1) to
deliver ottelione B (2) was best carried out by exposure to
t-BuOK in t-BuOH at room temperature for 2 h,19 leading
to a 23:77 mixture of 1 and 2. Isolation of 2 from this mixture
was performed by means of HPLC (DAICEL CHIRALPAK
AD-Η) to yield an entirely pure sample of 2, whose
1
spectroscopic properties (IR, H and 13C NMR, MS) were
Acknowledgment. We are especially grateful to Professor
Thomas R. Hoye (Department of Chemistry and Medicinal
Chemistry, University of Minnesota) for providing us with
copies of the IR, 1H and 13C NMR, and MS spectra of natural
otteliones A (1) and B (2) and for kindly informing us about
the optical rotation of 1 and 2. We also thank Dr. Naoki
Sugimoto (National Institutes of Health Sciences) for mea-
surements of HRMS.
identical with those of natural 2. The optical rotation of a
pure synthetic sample of 2 [[R]D25 -333.0° (c 0.18, CHCl3)]
(17) Selection of this dichloroacetyl protecting group was critical because
this protecting group could be smoothly and cleanly removed without
epimerization at the C3a position at the later stage (see 22 f 1, Scheme
4). In our preliminary studies, we had prepared an O-acetyl variant of 22
(R ) Ac); unfortunately, all attempts to remove the O-acetyl group from
this compound resulted in partial epimerization at the C3a position. For
accurately measuring the optical rotation of ottelione A (1), contamination
with the C3a epimerized product [ottelione B (2)] should be prevented.
(18) Private communication from Professor Thomas R. Hoye of Uni-
versity of Minnesota. The value of optical rotation for natural otteliones A
(1) and B (2) was not recorded in the literature cited in ref 1.
(19) Although efficient conversion of ottelione A (1) to ottelione B (2)
has been previously demonstrated by Mehta (DBU, benzene, 65 °C, 83%),8
our initial attempts to follow the Mehta conditions for this conversion
unfortunately resulted in the formation of an equilibrium mixture of 1 and
2 in a ratio of ca. 1:1.
Supporting Information Available: Detailed experi-
mental procedures and full characterization data for all new
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL035443X
3906
Org. Lett., Vol. 5, No. 21, 2003