First total synthesis of crassostreaxanthin B

Article abstract of DOI:10.1039/a901116a

The first total synthesis of crassostreaxanthin B 1 was accomplished via the tetrasubstituted olefinic compound 5, prepared by stereoselective rearrangement of epoxides 3a,b using a Lewis acid.

Full text of DOI:10.1039/a901116a

View Article Online / Journal Homepage / Table of Contents for this issue
First total synthesis of crassostreaxanthin B
Chisato Tode, Yumiko Yamano and Masayoshi Ito
Kobe Pharmaceutical University, Motoyamakita-machi, Higashinada-ku, Kobe 658-8558,
Japan
Received (in Cambridge) 9th February 1999, Accepted 13th May 1999
The first total synthesis of crassostreaxanthin B 1 was
O
OR
accomplished via the tetrasubstituted olefinic compound 5,
OAc
O
prepared by stereoselective rearrangement of epoxides 3a,b
using a Lewis acid.
OAc
RO
4 R=Ac
5 R=TBS
2a R=Ac
3a R=TBS
Crassostreaxanthin B 1 having a novel acyclic tetrasubstituted
olefinic end group was isolated from Crassostrea gigas1 and its
OAc
stereostructure including C-3 chirality (3R) was determined by
Matsuno’s group in 1992. However, the absolute configuration
at C-3Ј has remained undetermined. We assumed that this
end group was formed in nature from the epoxide end group of
5,6-epoxy carotenoids† by opening of the C-6–oxygen bond
of the oxirane ring (Scheme 1, route a). Thus, the absolute
O
Crassostreaxanthin B 1
RO
O
2b R=Ac
3b R=TBS
OAc
6
Scheme 2
O
OAc
R
R
route a
route b
as the protecting group should be appropriate synthons for the
biomimetic synthesis of 1.
Epoxides 3a,b were prepared in 11 steps from the known
optically active ketone 74 in good overall yield (43%) as shown
in Scheme 3. Introduction of an ethynyl group to the silyloxy
1
a
6
5
O
AcO
b
R
O
AcO
R = Me, Et, (CH2)3OAc, etc.
O
CHO
ii - iv
Scheme 1
RO
i
TBSO
configuration at C-3Ј in crassostreaxanthin B 1 is considered to
be S, since chiralities at C-3 in most of the known natural
epoxy carotenoids are R. In previous papers,2,3 we reported that
the novel acyclic tetrasubstituted olefinic end group and
cyclopentyl end group of carotenoids were obtained by Lewis
acid-promoted stereoselective rearrangement of the epoxy end
group of 5,6-epoxy carotenoids as shown in Scheme 1. This
is the first example of biomimetic formation of end groups
possessing an acyclic tetrasubstituted olefin. Herein, we wish
to describe the first total synthesis of crassostreaxanthin B 1
applying the stereoselective rearrangement of epoxides with a
Lewis acid and the determination of the absolute configuration
at C-3Ј in native samples.
9
7 R=H
8 R=TBS
v - viii, v, ix
x
OAc
3a,b
43%(11steps)
TBSO
10
Scheme 3 Reagents and conditions: i, TBSCl, Et3N, DMAP; ii,
LiC᎐CTMS; iii, 10% aq. KOH; iv, (Ph3SiO)3VO, PhCO2H, xylene,
reflux; v, NaBH4; vi, MsCl, Py; vii, KCN, 18-crown-6, DMSO, 120 ЊC;
viii, DIBAL-H; ix, Ac2O, Py; x, MCPBA.
᎐
᎐
Toward the biomimetic synthesis of crassostreaxanthin B 1,
we previously investigated3 the reaction of epoxides having
several functional groups at the C-6 position with BF3ؒOEt2.
As a result, a promising rearrangement reaction was found;
epoxides 2a,b possessing the acetoxy propane group provided
the tetrasubstituted olefinic compound 4 in reasonable yield,
however, accompanied by the elimination product 6 (Scheme 2).
Thus, epoxides 3a,b having a TBS (tert-butyldimethylsilyl) ether
ketone 85 and subsequent rearrangement of the resulting
product using tris(triphenylsilyl)vanadate6 as the catalyst
yielded the aldehyde 9, whose carbon chain extension was
accomplished by the usual method to give compound 10.
Epoxidation of 10 with MCPBA led to both isomers
(anti:syn = 2:3) of 3 which, without separation were treated
with SnCl4 to provide the stereoselective rearranged product,
E-tetrasubstituted olefin 5, in reasonable yield (60%). Ketaliz-
OH
3
O
O
OH
3'
Crassostreaxanthin B 1
J. Chem. Soc., Perkin Trans. 1, 1999, 1625–1626
1625

View Article Online
O
R
OTBS
OTBS
ix, x
O
O
ii,iii
i
O
O
3a,b
5
R
11 R = CH2CH2OAc
12 R = CH2CO2Me
13 R = CHO
15 R = CH2OTBS
16 R = CHO
iv-vi
vii, iv, viii
xi, x
OH
O
O
OH
xii - xiv
O
6
O
OH
4
HO
S
R
(3R, 3'R)-Crassostreaxanthin B 1'
19
3'
(3R, 3'S)-Crassostreaxanthin B 1
CH2P+Ph3Cl –
CH2P+Ph3Cl –
CH2OTBS
MeO
Br
14
OMe
18
HO
17
Scheme 4 Reagents and conditions: i, SnCl4, Ϫ78 ЊC; ii, (CH2OTMS)2, TMSOTf; iii, TBSOTf, 2,6-lutidine(2,6-dimethylpyridine); iv, LAH; v, PDC,
DMF; vi, TMSCHN2; vii, LDA, (ϩ)-camphorylsulfonyloxaziridine; viii, NaIO4; ix, 14, t-BuLi; x, MnO2; xi, TBAF; xii, 17, 1 M NaOMe then Hϩ;
xiii, 18, 1 M NaOMe; xiv, p-TsOH.
ation of this olefin 5 and reprotection of the hydroxy group gave
compound 11, which was converted into the ester 12 by LAH
Acknowledgements
We are indebted to Dr U. Hengartner, Hoffmann-La Roche
Ltd., Basel, Switzerland for his gift of a large amount of (4R,
6R)-4-hydroxy-2,2,6-trimethylcyclohexanone.
reduction followed by PDC oxidation and subsequent methyl-
ation (Scheme 4). Introduction of a hydroxy group into the
ester 12 by use of (ϩ)-camphorylsulfonyloxaziridine7 in the
presence of LDA followed by reduction with LAH afforded
the glycol, which was subjected to glycol cleavage with NaIO4 to
give the aldehyde 13 (23% from 5). Reaction of this aldehyde
13 with vinyl bromide 148 in the presence of t-BuLi and sub-
sequent oxidation of the resulting hydroxy group with MnO2
yielded the ketone 15. Partial deprotection of the allylic TBS
group in 15 with TBAF (tetrabutylammonium fluoride)
afforded the allylic alcohol, which was oxidized with MnO2 to
provide the keto-aldehyde 16 (59%, from 13). Then a double
Wittig condensation of the aldehyde 16 with the Wittig salts
179 and 1810 in the presence of 1 M NaOMe as base, followed
by deprotection of ketal and TBS groups using p-TsOH led
to (3R,3ЈS)-crassostreaxanthin B 1 (6%, from 16). Spectral data
(IR, UV–VIS, 1H-NMR and MS) of synthetic crassostrea-
xanthin B 1 were in good agreement with those of a natural
specimen.11 However, the absolute configuration at C-3Ј in the
native sample could not be confirmed by comparison of
synthetic and natural samples of CD data because these did
not exhibit a clear Cotton effect. Therefore, (3R,3ЈR)-
crassostreaxanthin B 1Ј was independently synthesized from
the (4S,6R)-hydroxy ketone 19.4 HPLC separation of 1 and 1Ј
was clearly accomplished using a chiral column (Chiralcel
OD; Daicel). Synthetic (3R,3ЈS)-crassostreaxanthin B 1 was
confirmed to be identical with a natural specimen by co-
chromatography. Consequently, the absolute configuration at
C-3Ј in the natural specimen was determined to be S.
Notes and references
† We have employed the numbering system used in carotenoids.
1 Y. Fujiwara, T. Maoka, M. Ookubo and T. Matsuno, Tetrahedron
Lett., 1992, 33, 4941.
2 Y. Yamano, C. Tode and M. Ito, J. Chem. Soc., Perkin Trans. 1,
1996, 1337.
3 Y. Yamano, C. Tode and M. Ito, J. Chem. Soc., Perkin Trans. 1,
1998, 2569.
4 H. G. W. Leuenberger, W. Boguth, E. Widmer and R. Zell, Helv.
Chim. Acta, 1976, 59, 1832.
5 Y. Yamano and M. Ito, J. Chem. Soc., Perkin Trans. 1, 1993, 1599.
6 (a) H. Pauling, D. A. Andrew and N. C. Hindley, Helv. Chim. Acta,
1976, 59, 1233; (b) Y. Yamano, C. Tode and M. Ito, J. Chem. Soc.,
Perkin Trans. 1, 1995, 1895.
7 F. A. Davis, G. V. Reddy, B. C. Chen, A. Kumar and M. S. Haque,
J. Org. Chem., 1995, 60, 6148.
8 (a) M. Schlosser and E. Hammer, Helv. Chim. Acta, 1974, 57, 2547;
(b) A. R. de Lera, B. Iglesias, J. Rodríguez, R. Alvarez, S. López,
X. Villanueva and E. Padrós, J. Am. Chem. Soc., 1995, 117, 8220.
9 K. Bernhard, F. Kienzle, H. Mayer and R. K. Müller, Helv. Chim.
Acta, 1980, 63, 1473.
10 E. Widmer, M. Soukup, R. Zell, E. Broger, H. P. Wagner and M.
Imfeld, Helv. Chim. Acta, 1990, 73, 861.
11 An authentic sample of natural crassostreaxanthin B was kindly
furnished by Dr T. Maoka, Kyoto Pharmaceutical University.
This is the first biomimetic total synthesis of the optically
active crassostreaxanthin
rearrangement of epoxides 3a,b with Lewis acid.
B 1 using the stereoselective
Communication 9/01116A
1626
J. Chem. Soc., Perkin Trans. 1, 1999, 1625–1626