Acetylene cobalt complex and vinylsilane strategy in the synthesis of ciguatoxin (D)EF analog

Article abstract of DOI:10.1016/S0040-4039(00)00988-6

A synthetic route to the (D)EF analog of ciguatoxin has been explored through acetylene cobalt complex and vinylsilane strategy. The central reactions in the synthesis are: (i) ether ring formation mediated by acetylene cobalt complex and (ii) decomplexation of the endo-acetylene cobalt complexes to vinylsilanes or olefins. Unusual rearrangement of epoxy-silane to allylic alcohol is also described. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00988-6

Tetrahedron Letters 41 (2000) 5951±5955
Acetylene cobalt complex and vinylsilane strategy in the
synthesis of ciguatoxin (D)EF analog
Kazunobu Kira and Minoru Isobe*
Laboratory of Organic Chemistry, Graduate School of Bioagricultural Sciences, Nagoya University, Chikusa,
Nagoya 464-8601, Japan
Received 22 May 2000; revised 8 June 2000; accepted 9 June 2000
Abstract
A synthetic route to the (D)EF analog of ciguatoxin has been explored through acetylene cobalt complex
and vinylsilane strategy. The central reactions in the synthesis are: (i) ether ring formation mediated by
acetylene cobalt complex and (ii) decomplexation of the endo-acetylene cobalt complexes to vinylsilanes or
ole®ns. Unusual rearrangement of epoxy-silane to allylic alcohol is also described. # 2000 Elsevier Science
Ltd. All rights reserved.
Keywords: ciguatoxin; acetylene cobalt complex; hydrosilylation; vinylsilane.
We have been studying various synthetic methodologies applicable to the ciguatoxin (CTX1B,
1)1 class of natural products, that have syn/trans stereochemistry of polycyclic oxy-ring systems.
Our synthetic e€orts have recently culminated in both enantiomeric forms of the left end segment
(ABC)2 or right middle segment, including the HIJK segment.3 These studies base their strategy
on the chemistry of sugar acetylene4 and acetylene biscobalthexacarbonyl complex.5 Interesting
synthetic e€orts have been reported from other research groups lead by Hirama,6 Tachibana7 and
Murai8 based upon various di€erent methodologies.9
* Corresponding author. Tel: +81-52-789-4109; fax: +81-52-789-4111; e-mail: isobem@agr.nagoya-u.ac.jp
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00988-6

5952
This communication includes the synthesis of the left middle segment (D)EF in the form of 2 to
explore our methodologies, which commences from a d-hexopyranose derivative (D0 ring) to extend
the carbon chain toward the E/F ring and subsequent cyclization to the medium-sized ether rings.
Retrosynthetic analysis is illustrated in Scheme 1, where the critical nine-membered ring cycliza-
tion would occur from the precursor propargylic cation 3 that is stabilized by an acetylene cobalt
complex.10 This eneyne could be derived from the allylic alcohol 4. This allylic alcohol should be
transformed from the complex 6 via vinylsilane 5. Similar cyclization of the seven-membered ring
from the precursor propargylic cation 7 again took the form of the acetylene cobalt complex. A
d-pyranose derivative would have access to this starting material, as reported previously.11
Scheme 1. Retrosynthesis of (D)EF analog 2
An optically active compound 8 (preparable from triacetyl-d-glucal in 9 steps11) was selected as
a template to extend the carbon chain as analyzed in retrosynthesis (Scheme 1). The lithium
acetylide of 8 was added to the protected 3-oxy-propanal to give 9, which was converted into
cobalt complex 10. The crucial cyclization happened by treatment with p-toluenesulfonic acid to
give 11 in syn/trans manner as a single isomer.12 Hydrosilylation13 of 11 with triethylsilane in hot
toluene was followed by oxidative deprotection of the p-methoxybenzyl group with DDQ to
a€ord 12. The epoxidation of the silyl ole®n in 12 was not stereoselective with mCPBA, but it
gave a mixture of a- and b-epoxides (5:3) without touching the other ole®ns. Only one of these
epoxides, the b-epoxide, turned out to be convertible to the allylic alcohol by treatment with
14
.
BF3 OEt2 in a separate experiment. Finally, b-epoxide was stereoselectively prepared from 12,
®rstly by oxidation to the carboxylic acid 13, secondly by iodo-lactonization to 14 and its DIBAL
reduction to a€ord the b-epoxide 15. Peterson ole®nation of this aldehyde 15 with 3-lithio-1,3-
bis-(triisopropylsilyl)-propyne15 gave largely the cis ene-yne 16. Treatment of this epoxysilane 16
.
with BF3 OEt2 yielded the allylic alcohol 17 in 80%. Now the con®guration of C-8 with the
hydroxyl group was inverted by Mitsunobu reaction to convert into 18. Addition of the lithium
acetylide of 18 to phenylpropanal provided a propargylic alcohol, which was further converted
.
into cobalt complex 19. Its cyclization was performed by simple treatment with BF3 OEt2 at room
temperature to a€ord 20 in 70% yield as a single stereoisomer. Syn stereochemistry of this
product was shown by NOE experiments.16 This cobalt complex was hydrosilylated into the
vinylsilane 21 by simple heating with triethylsilane13 (Scheme 2).
As shown in Scheme 3, an alternative synthesis of a similar compound was achieved. Addition
of the lithium acetylide of 18 to a protected propanal successively converted it into the
.
corresponding acetylene cobalt complex 22, which was similarly cyclized with BF3 OEt2 at a
room temperature to give 23 in 81% yield as a single stereoisomer. Finally, the cobalt complex
was decomplexed into the corresponding ole®n 24 by simple heating with tributyltin hydride.13

5953
.
Scheme 2. (a) n-BuLi/THF; (b) PPS/MeOH, 80% in two steps; (c) Co2(CO)8/CH2Cl2, 88%; (d) pTsOH H2O/CH2Cl2,
rt, 90%; (e) Et3SiH/toluene, 70ꢀC; (f) DDQ, 80% in two steps; (g) Joneꢀs reagent; (h) I(collidine)2PF6/CH2Cl2, 77% in
.
two steps; (i) DIBAL; (j) DBU, 85% in two steps; (k) n-BuLi/THF, ^78 C to rt, 77%; (l) BF3 OEt2/CH2Cl2, 80%; (m)
TBAF/THF, quant.; (n) PPh3, DEAD, benzoic acid, quant.; (o) K2CO3, MeOH; (p) EVE, PPTS, 76% in two steps; (q)
.
n-BuLi; (r) PPTS, 87% in two steps; (s) Co2(CO)8/CH2Cl2, 78%; (t) BF3 OEt2/CH2Cl2, rt, 70%; (u) Et3SiH/toluene,
70ꢀC, quant.
Although the junction protons of this particular compound have considerable overlap with each
other and no NOE data are available to prove the syn/trans stereochemistry of 24 (Scheme 3).
.
Scheme 3. (a) n-BuLi; (b) PPTS/MeOH, 80% in two steps; (c) Co2(CO)8/CH2Cl2, 87%; (d) BF3 OEt2/CH2Cl2, rt,
81%; (e) n-Bu3SnH/toluene, 62%
Thus, we succeeded in the synthesis of the (D)EF analogs (21 and 24) of ciguatoxin 1 using
acetylene cobalt complex and the unusual rearrangement of epoxysilane 15 to allylalcohol 16.
Further e€orts directed toward the total synthesis of ciguatoxin are in progress in our laboratory.

5954
Acknowledgements
This work was supported by Grant-in-Aid from Ministry of Education, Science, Sports and
Culture of Japan. One of the authors, K.K. thanks JSPS Research Fellowships for Young Scientists.
References
1. Satake, M.; Morohashi, A.; Oguri, H.; Oishi, T.; Hirama, M.; Harada, N.; Yasumoto, T. J. Am. Chem. Soc. 1997,
119, 11325 and references therein.
2. (a) Hosokawa, S.; Isobe, M. Synlett 1995, 1179. (b) Hosokawa, S.; Isobe, M. Synlett 1996, 351. (c) Hosokawa, S.;
Isobe, M. J. Org. Chem. 1999, 64, 37. (d) Saeeng, R.; Isobe, M. Tetrahedron Lett. 1999, 40, 1911.
3. (a) Liu, T.-Z.; Isobe, M. Synlett 2000, 587; (b) Liu, T.-Z.; Isobe, M. Tetrahedron, in press.
4. Isobe, M.; Nishizawa, R.; Hosokawa, S.; Nishikawa, T. J. Chem. Soc., Chem. Commun. 1998, 2665.
5. (a) Tanaka, S.; Tsukiyama, T.; Isobe, M. Tetrahedron Lett. 1993, 34, 5757. (b) Tanaka, S.; Isobe, M. Tetrahedron
1994, 50, 5633. (c) Tanaka, S.; Isobe, M. Tetrahedron Lett. 1994, 35, 7801. (d) Tanaka, S.; Tatsuta, T.; Yamashita,
O.; Isobe, M. Tetrahedron 1994, 50, 12883. (e) Isobe, M.; Yenjai, C.; Tanaka, S. Synlett 1994, 916.
6. (a) Oishi, T.; Nagumo, Y.; Hirama, M. Chem. Commun. 1998, 1041. (b) Maeda, K.; Oishi, T.; Oguri, H.; Hirama,
M. Chem. Commun. 1999, 1063 and references therein.
7. (a) Sasaki, M.; Fuwa, H.; Ishikawa, M.; Tachibana, K. Org. Lett. 1999, 1, 1075. (b) Inoue, M.; Sasaki, M.;
Tachibana, K. Tetrahedron 1999, 55, 10949 and references therein.
8. (a) Oka, T.; Murai, A. Tetrahedron 1998, 54, 1. (b) Oka, T.; Fujiwara, K.; Murai, A. Tetrahedron 1998, 54, 21 and
references therein.
9. (a) Alvarez, E.; Delgado, M.; Diaz, M. T.; Hanxing, L.; Perez, R.; Martin, J. D. Tetrahedron Lett. 1996, 37, 2865.
(b) Leeuwenburgh, M. A.; Kulker, C.; Overkleeft, H. S.; van der Marel, G. A.; van Boom, J. H. Synlett 1999, 12,
1945.
10. Connor, R. E.; Nicholas, K. M. J. Organomet. Chem. 1977, 125, C45.
11. The starting material 8 was prepared in nine steps from triacetyl-d-glucal A.
12. Isobe, M.; Hosokawa, S.; Kira, K. Chem. Lett. 1996, 473.
13. Hosokawa, S.; Isobe, M. Tetrahedron Lett. 1998, 39, 2609.
14. Acid treatment of a- and b-epoxysilane gives ketone and allylic alcohol, respectively.

5955
15. Corey, E. J.; Rucker, C. Tetrahedron Lett. 1982, 23, 719.
16. Stereochemistry of cyclic products is governed by reaction conditions; the anti isomer is the kinetic product, while
the syn isomer is the thermodynamic product.