Total synthesis and absolute configuration of polycavernoside A [6]

Full text of DOI:10.1021/ja982431b

10770
J. Am. Chem. Soc. 1998, 120, 10770-10771
Scheme 1
Total Synthesis and Absolute Configuration of
Polycavernoside A
Kenshu Fujiwara and Akio Murai*
DiVision of Chemistry, Graduate School of Science
Hokkaido UniVersity, Sapporo 060-0810, Japan
Mari Yotsu-Yamashita and Takeshi Yasumoto*†
Faculty of Agriculture, Tohoku UniVersity
Tsutsumidori-Amamiyamachi Aoba-Ku
Sendai 981-8555, Japan
ReceiVed July 10, 1998
Polycavernoside A (1) was isolated as a causative toxin of
sudden and fatal human intoxication in Guam in 19911 by
Yasumoto’s group from the red alga PolycaVernosa tsudai, which
had been eaten widely without any problems.2 The planar
structure of 1 has been proved to be a novel 13-membered
macrolactone disaccharide possessing a triene moiety. However,
because of the natural scarcity of the compound, stereochemical
information obtained so far has been limited to the partial relative
structures of each sugar component and bottom (C1-C8) and
upper halves (C9-C15) of the macrolactone part. Later, Murai’s
group determined the relative configuration of the sequence of
the fucose-xylose bottom half of the macrolactone synthetically,3a
but the absolute configuration was still unknown. From the
viewpoint of public health at algal feeding areas, clarification of
the absolute structure is important; such information may promote
a solution to the puzzling outbreak of the intoxication and explain
the strong bioactivity. In this paper, we report the first total
synthesis of 1 and determination of the absolute configuration.
Before planning of the synthesis, we deduced the whole
molecular structure of 1 as shown in Scheme 1, based on the
following points: (1) anti conformational relationship between
The diethyldithioacetal mono-S-oxide 7 was readily obtained
in seven steps from alcohol 9 (Scheme 2). Swern oxidation6 and
subsequent Wittig reaction provided a methyl vinyl ether in 90%
yield, which was hydrolyzed under acidic conditions to give
aldehyde 10 (92%). Then, Evans’ thioacetalization,7 reduction
of the methyl ester, protection of the resultant alcohol with TBS,
and selective oxidation of the thioacetal moiety afforded mono-
sulfoxide 7 in 92% overall yield.
Deprotonation of 7 with LDA (1 equiv) in THF at -78 °C
followed by addition of aldehyde 6 (0.5 equiv), derived from 8
by Swern oxidation,6 provided acid-sensitive dithioacetal 128 (a
mixture of diastereomers, 46% based on 6) and vinyl sulfide 138,9
(a single diastereomer, 33% based on 6) together with recovered
7 (57%) and 6 (17%). When the adduct 12 was treated with
p-toluenesulfonic acid in MeOH-(MeO)3CH (10:1), removal of
the ethylthio and ethylsulfinyl groups, desilylation, and cyclic
methyl acetal formation occurred in one pot to produce the desired
bicyclic compounds 1410 (50%) and C9-epimeric 1510 (18%).
Under the same conditions, vinyl sulfide 13 was transformed into
14 in 65% yield. Swern oxidation6 of diol 14 and the following
removal of PMB11 provided keto aldehyde 16 (81%). Secoic acid
17, obtained after NaClO2 oxidation12 of 16, was lactonized
smoothly by modified Yamaguchi’s method13 to give 12-
membered lactone 5 in >75% yield.
(6) (a) Omura, K.; Swern, D. Tetrahedron 1978, 34, 1651-1660. (b)
Mancuso, A. J.; Huang, S.-L.; Swern, D. J. Org. Chem. 1978, 43, 2480-
2482. (c) Mancuso, A. J.; Swern, D. Synthesis 1981, 165-185.
(7) Evans, D. A.; Truesdale, L. K.; Grimm, K. G.; Nesbitt, S. L. J. Am.
Chem. Soc. 1977, 99, 5009-5017.
(8) Dithioacetal 12 and vinyl sulfide 13 were purified by chromatography
using silica gel doped with 4% Et3N.
(9) 13 was gradually produced from some of the diastereomers of 12 under
the reaction conditions (checked by TLC analysis). The other diastereomers
of 12 were unchanged for prolonged periods under the same conditions.
(10) The axial orientation of each methoxy group at C9 of 14 and 15 was
confirmed by the existence of NOE between H13 and CH3O at C9 in the
derivative of 14 and the intact 15, respectively: see Supporting Information.
(11) Oikawa, Y.; Yoshioka, T.; Yonemitsu, O. Tetrahedron Lett. 1982, 23,
885-888.
(12) Kraus. G. A.; Taschner, M. J. J. Org. Chem. 1980, 45, 1175-1176.
(13) (a) Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M. Bull.
Chem. Soc. Jpn. 1979, 52, 1989-1993. For the efficiency of excess DMAP
in the method: (b) Hikota, M.; Tone, H.; Horita, K.; Yonemitsu, O. J. Org.
Chem. 1990, 55, 7-9. (c) Hikota, M.; Tone, H.; Horita, K.; Yonemitsu, O.
Tetrahedron 1990, 46, 4613-4628.
3
H7 and H8b suggested by large JH7/H8b value (9 Hz) and no
observation of NOE between H7 and H8b;2c (2) same-side
placement of H8b and H11 on the macrolactone ring suggested
by the presence of NOE between H8b and H11 and the absence
between H8a and H11;2a (3) natural predominance of D-xylose
and L-fucose in marine biota.
Our synthetic plan is outlined in Scheme 1. Introduction of
the triene, considering its potential lability, was scheduled in the
final step after the glycosylation of lactone 3 with 43a,4 derived
from L-fucose and D-xylose. Hydrolytic reconstruction of the
acetal moiety of 5 within the framework of lactone was expected
to set up the stereochemistry at anomeric C10 of 3 in natural
form. Further, this strategy would also rely on the successful
lactonization to 5 from the requisite secoic acid, the stereoselective
formation of six-membered cyclic methyl acetal, and the con-
nection of the acyl anion equivalent5 derived from 7 with aldehyde
6 at the C9-C10 bond. For the preparation of 6 and 7, alcohols
83d and 93c would be available, respectively.
† Present address: Japan Food Research Laboratories, Tama Laboratory,
6-11-10, Nagayama, Tamashi, Tokyo 206-0025, Japan.
(1) Haddock, R. L.; Cruz, O. L. T. Lancet 1991, 338, 195-196.
(2) (a) Y.-Yamashita, M.; Haddock, R. L.; Yasumoto, T. J. Am. Chem.
Soc. 1993, 115, 1147-1148. (b) Y.-Yamashita, M.; Seki, T.; Paul, V. J.; Naoki,
H.; Yasumoto, T. Tetrahedron Lett. 1995, 36, 5563-5566. (c) Y.-Yamashita,
M.; Yasumoto, T.; Haddock, R. L. 34th Symposium on the Chemistry of
Natural Products (Tokyo), Symposium Abstracts, 1992; pp 612-619.
(3) (a) Fujiwara, K.; Amano, S.; Murai, A. Chem. Lett. 1995, 855-856.
(b) Fujiwara, K.; Amano, S.; Murai, A. Chem. Lett. 1995, 191-192. (c)
Fujiwara, K.; Amano, S.; Oka, T.; Murai, A. Chem. Lett. 1994, 2147-2150.
(d) Hayashi, N.; Mine, T.; Fujiwara, K.; Murai, A. Chem. Lett. 1994, 2143-
2146.
(4) For another synthesis of the sugar part of polycavernoside A: Johnston,
J. N.; Paquette, L. A. Tetrahedron Lett. 1995, 36, 4341-4344.
(5) Herrmann, J. L.; Richman, J. E.; Wepplo, P. J.; Schlessinger, R. H.
Tetrahedron Lett. 1973, 4707-4710.
10.1021/ja982431b CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/03/1998

Communications to the Editor
J. Am. Chem. Soc., Vol. 120, No. 41, 1998 10771
Scheme 2a
a Reagents and conditions: (a) (COCl)2 (1.5 equiv), DMSO (2.4 equiv), CH2Cl2, -78 °C, 20 min, then Et3N (5 equiv), -78 f 0 °C; (b) Ph3PdCHOCH3
(1.5 equiv), THF, -78 °C, 5 min, then 20 °C, 16 h, 90% from 9; (c) THF-H2O-TFA (1:1:0.2), 20 °C, 4.5 h, 92%; (d) TMSSEt (3 equiv), ZnI2 (cat.),
CH2Cl2, 20 °C, 5.5 h; (e) LiAlH4 (2 equiv), THF, 0 °C; (f) TBS-Cl (1.1 equiv), imidazole (2.2 equiv), DMF, 0 °C, 99% from 10; (g) m-CPBA (1 equiv),
CH2Cl2, -60 °C, 93%; (h) LDA (1 equiv), THF, -78 °C, 10 min; then 6 (0.5 equiv), -78 °C, 30 min, 46% of 12 from 6, 33% of 13 from 6, 57%
recovery of 7, 17% recovery of 6; (i) TsOH‚H2O (0.4 equiv), MeOH-(MeO)3CH (10:1), 25 °C, 21 h, 50% of 14, 18% of 15; (j) TsOH‚H2O (0.4 equiv),
MeOH-(MeO)3CH (10:1), 25 °C, 3 days, 65% of 14; (k) (COCl)2 (10 equiv), DMSO (16 equiv), CH2Cl2, -60 °C, 20 min, then Et3N (34 equiv), -60
f 0 °C, 88%; (l) DDQ (1.5 equiv), CH2Cl2, 20 °C, 2.5 h, 92%; (m) NaClO2 (5 equiv), NaH2PO4 (10 equiv), 2-methyl-2-butene (100 equiv), t-BuOH-
H2O (3.5:1), 0 °C, 20 min, 100% (crude); (n) 2,4,6-trichlorobenzoyl chloride (4 equiv), Et3N (8 equiv), THF, 20 °C, 10 h, then DMAP (37 equiv),
toluene, 60 °C, 1 h, >75%; (o) OsO4 (0.2 equiv), NMO (12 eq), 1,4-dioxane-H2O (3:1), 20 °C, 31 h, then NaIO4 (15 equiv), 40 min, >75%; (p) H2,
Pd/C, MeOH, 20 °C, 40 min; (q) TBSOTf (10 equiv), 2,6-lutidine (25 equiv), CH2Cl2, 0 °C, 5 min, >95% (2 steps); (r) CrCl2 (270 equiv), CHI3 (90
equiv), THF, 0 °C, 30 min, >80%; (s) THF-H2O-TFA (2.5:2.5:1), 20 °C, 6.5 h, >95%; (t) 4 (4 equiv), NBS (8 equiv), MS4A, CH3CN, -20 °C, 5
min, >50%; (u) DDQ (30 equiv), CH2Cl2-H2O (25:1), 20 °C, 10 h, >70%; (v) t-BuLi (2 equiv), Et2O-THF (1:1), -78 °C, 30 min, then HgCl2 (0.5
equiv), -78 f 0 °C, 30 min, 100% (crude); (w) Cp2ZrCl2 (1 equiv), LiEt3BH (1 equiv), THF, 20 °C, 1 h, then 23, 30 min, then I2 (1 equiv), 58%; (x)
22 (excess), Pd(PPh3)4 (cat), THF, 0 f 20 °C, >75%.
Toward the final construction of the triene part, introduction
of the trans-iodovinyl group was necessitated at this stage.
Dihydroxylation of the vinyl moiety of 5 followed by oxidative
cleavage and replacement of benzyl to TBS led to aldehyde 18
in >70% overall yield. Chemo- and stereoselective iodovinylation
of the aldehyde was achieved to produce 19 under Takai’s
conditions.14,15 The 6-membered cyclic methyl acetal 19 was
converted to the desired 5-membered cyclic hemiacetal 3 quan-
titatively under mild acidic conditions. Stereochemistry at C10
of 3 was confirmed by the existence of NOE between H11 and
H8b which showed a large 3JH7/H8b (9.2 Hz) as seen in natural 1.
Glycosylation of 3 with fucosylxylose derivative 43a was carried
out according to Nicolaou’s procedure16 to give the desired
â-glycoside 20 in >50% yield, in which the benzyl group was
removed by DDQ17 to afford 21.
2319 through a one-pot hydrozirconation-iodination sequence20
and sequential lithiation-transmetalation process, was treated with
vinyl iodide 21 and a Pd catalyst to produce 1 in good yield.
The synthetic material displayed physical and spectral proper-
ties including 1H and 13C NMR, IR, UV, and FABMS, as well as
toxicological property to mice,2a,21 identical to the natural product.
The CD spectrum of natural polycavernoside A was identical with
that of synthetic 1 in pattern, sign, and wavelength of each extreme
point.22 Thus, polycavernoside A has the absolute configuration
of 1 shown in Scheme 1.
Acknowledgment. We are grateful to Prof. Y. Fukazawa, Hiroshima
University, for valuable discussions. This work was supported by Grant-
in-Aids from the Ministry of Education, Science, Sports and Culture,
Japan (08245103 and 10308027, A.M.; 07780486 and 09780518, K.F.;
10760043, M.Y.-Y.), a Suntory Institute Bioorganic Research Grant, and
a grant from the Naito Foundation.
We selected the less nucleophilic dienyl mercury 22 as a diene
segment in the final cross-coupling reaction18 with 21 to avoid
any side reactions toward the oxygen functional groups. Although
a homocoupling reaction proceeded as a side reaction in this
system,18 the use of excess reagent could complete the cross-
coupling reaction. Thus, excess 22, prepared from known enyne
Supporting Information Available: Experimental procedures and
spectroscopic data for synthetic intermediates and CD spectra of natural
and synthetic 1 (34 pages, print/PDF). See any current masthead page
for ordering and Internet access instructions. See any current masthead
page for ordering information and Web access instructions.
(14) Takai, K.; Nitta, K.; Utimoto, K. J. Am. Chem. Soc. 1986, 108, 7408-
7410.
(15) The use of large excess reagents accelerated the reaction rate to prevent
the epimerization at C15.
JA982431B
(19) Baldwin, J. E.; Reddy, V. P. J. Am. Chem. Soc. 1988, 110, 8223-
8228.
(16) Nicolaou, K. C.; Seitz, S. P.; Papahatjis, D. P. J. Am. Chem. Soc.
1983, 105, 2430-2434.
(20) Lipshutz, B. H.; Keil, R.; Ellsworth, E. L. Tetrahedron Lett. 1990,
31, 7257-7260.
(17) For the related oxidative cleavage reactions of allyl and benzyl
ethers: Lee-Ruff, E.; Ablenas, F. J. Can. J. Chem. 1989, 67, 699-702.
(18) The cross-coupling reaction between alkadienylmercury and alkenyl
halide for triene synthesis has not been reported. For cross-coupling reactions
using alkenylmercury and arylmercury reagents: (a) Negishi. E.; Takahashi,
T.; Baba, S.; Van Horn, D. E.; Okukado, N. J. Am. Chem. Soc. 1987, 109,
2393-2401. (b) Beletskaya, I. P. J. Organomet. Chem. 1983, 250, 551-564
and references therein.
(21) LD99 in mice (ip) of synthetic 1 was 240-360 µg/kg, while that of
natural 1 was 200-400 µg/kg.
(22) CD spectral data of 1: (natural) CD (CH3CN, 22 °C) λext 210.2 nm
(∆ꢀ 0.71), 227.8 (-0.15), 246.0 (0.21), 258.6 (0.03), 260.8 (0.15), 269.6
(-0.63), 280.4 (-0.65), 299.0 (-0.78); (synthetic) CD (CH3CN, 22 °C) λext
212.0 nm (∆ꢀ 0.32), 228.8 (-0.25), 248.0 (0.19), 260.0 (-0.12), 262.8 (0.15),
268.6 (-0.42), 279.6 (-0.55), 299.8 (-0.73). Optical rotation data of 1:
(natural) [R]22D -59 (c 0.012, CH3CN); (synthetic) [R]22D -66 (c 0.02, CH3CN).