Total synthesis of the light-harvesting carotenoid peridinin

Article abstract of DOI:10.1002/anie.200600502

(Chemical Equation Presented) Light work: The light-harvesting marine C₃₇ carotenoid peridinin was synthesized from (+)-diethyl tartrate and (-)-actinol. Key transformations were the differential reduction of an ester-containing Weinreb amide,

Full text of DOI:10.1002/anie.200600502

Angewandte
Chemie
2 (“allenylsulfone”), 3 (“alkenylstannane”), and 4 (“oxobro-
moacrylate”) (Scheme 1).Compound 4 was designed to yield,
after acetal cleavage and lactonization, a 3-alkenyl-5-(a-
hydroxyalkyl)butenolide.Subsequent anti-selective dehydra-
Total Synthesis
DOI: 10.1002/anie.200600502
Total Synthesis of the Light-Harvesting
Carotenoid Peridinin**
Thomas Olpp and Reinhard Brückner*
Peridinin (1) is the major light-harvesting pigment of
planktonic dinoflagellates[1] conspicuously present in harmful
algae blooms (“red tides”).[2] Peridinin is distinguished by its
antitumor activity,[3] and it is one of the most abundantly
biosynthesized carotenoids on earth.[4] There are more than
700 naturally occurring carotenoids,[5] but the butenolide ring
is almost unique to peridinin (1).[6] The biosynthesis of ist C37
framework is not yet understood.[7]
Scheme 1. Retrosynthetic analysis of peridinin (1). Bt=benzothiazolyl
(formula in bottom line of Scheme 4).
1’
5
=
tion should result in the Z-configured C
C bond in the
butenolide core of 1.[13,14] Entities like the 3-alkenyl-5-(a-
hydroxyalkyl)butenolide exhibiting the required ul relation-
ship[15] between their stereocenters can be obtained by
appropriate modification of the termini of diethyl tartrate
(6).We showed this in a model study [16] and chose to exploit it
here.Compound 4 was endowed with an oxo function in order
to take up allenylsulfone 2 in a (Sylvestre) Julia olefination
and with a bromoolefin moiety for a Stille coupling with
alkenylstannane 3.Both the allenylsulfone 2 and the alkenyl-
stannane 3 were to be derived from (ꢀ)-actinol (5).[17]
Compound 5 is a common starting material for the synthesis
of 4’’-hydroxylated[18] carotenoids like zeaxanthin[19] and
peridinin.[10–12]
A century after its first isolation[8] and a decade after its
structure elucidation,[9] peridinin (1) became a target of total
[10]
synthesis.Ito et al.
as well as Katsumura et al.[11] have
synthesized the natural product, while de Leraꢀs group,
collaborating with us, prepared its allene epimer.[12] Here we
describe the third total synthesis of 1.It is strategically novel
and promises to lead to other natural[6] or related non-natural
butenolide-containing carotenoids.
With an emphasis on convergency, our retrosynthetic
analysis traced back peridinin (1) to the three building blocks
Starting from (+)-diethyl tartrate (6), the ester-containing
Weinreb amide 7 was prepared in four steps and 58% overall
yield as described earlier[16] (Scheme 2).The selective reduc-
tion both of the ester (!CH2OH) and amide (!CHO)
moieties succeeded only when a distinct excess of NaAlH4
was used.LiAlH 4 and diisobutylaluminum hydride (DIBAL)
were either inert or degraded the substrate, while NaBH4 in
MeOH reduced the amide moiety selectively, yet differently
(!CH2OH) than desired.The resulting hydroxyaldehyde 8
was not purified because of its instability, and thus crude 8 was
subjected to the Ando-type[21] Horner–Wadsworth–Emmons
olefination with a mixture of bromophosphonates 9a and 9b
developed for this purpose.[22] This gave hydroxybromoacry-
late 10[23] with 95% E selectivity.[24] Subsequent oxidation
with MnO2 delivered the desired oxobromoacrylate 4.How-
ever, the lability of this compound forced us to treat it without
[*] Dr. T. Olpp,[+] Prof. Dr. R. Brückner
Institut für Organische Chemie und Biochemie
Universität Freiburg
Albertstrasse 21, 79104 Freiburg (Germany)
Fax: (+49)761-203-6100
E-mail: reinhard.brueckner@organik.chemie.uni-freiburg.de
[+] Current address:
Department of Chemistry, Building 18–344
Massachusetts Institute of Technology
77 Massachusetts Avenue, Cambridge, MA 02139 (USA)
[**] This work was generously supported by the Fonds der Chemischen
Industrie (KekulØ fellowship for T.O.) and by the Deutsche
Forschungsgemeinschaft. We are grateful to Dr. Thomas Netscher
(DSM Nutritional Products, Basel) for a donation of (ꢀ)-actinol.
Supporting information for this article (the NMR data of 1 and
(6’cis)-1) is available on the WWW under http://www.angewand-
te.org or from the author.
Angew. Chem. Int. Ed. 2006, 45, 4023 –4027
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4023

Communications
the previously described precautions were put into effect
(absence of light, degassed solvents, radical scavenger)[16] we
obtained 12 in 51% yield.To our chagrin, 12 was formed in
Z/E mixtures with ratios ranging from 93:7 to 78:22.
The syntheses of allenylsulfone 2 and alkenylstannane 3
proceeded via epoxyalkyne 14 as a common intermediate; the
latter, in turn, was derived from (ꢀ)-actinol (5)[17] (Scheme 3).
Scheme 3. Synthesis of alkenylstannane 3. a) Me2tBuSiCl (1.07 equiv),
NEt3 (1.1 equiv), DMAP (1.05 equiv), CH2Cl2, 08C, 6 h, 97% (Ref. [10b]
93%); b) LDA (1.12 equiv), addition of product from (a), ꢀ788C, 1 h,
addition of PhN(SO2CF3)2 (1.5 equiv), 258C, 2 d, 81% (Ref. [10b]
89%); c) stream of CO, MeOH (30 equiv), [Pd(PPh3)4] (3 mol%), NEt3
(3 equiv), DMF, 808C, 24 h, 99% (Ref. [11]: 97%); d) DIBAL
(2.5 equiv), CH2Cl2, ꢀ788C, 1 h, 81% (Ref. [11] with LiAlH4: 87%);
e) tBuOOH (2.0 equiv), Ti(OiPr)4 (1.5 equiv), (ꢀ)-diisopropyl tartrate
(2.3 equiv), 4 Š molecular sieves, CH2Cl2, ꢀ308C, 30 min, 95%,
de>98% (Ref.[11] 99%, 92% de; Ref. [12] 98%, >98% de); f) Dess–
Martin periodinane,[25] CH2Cl2, 258C, 45 min, 92% (Swern oxidations:
Scheme 2. Synthesis of the butenolide moiety 12 [1H NMR shifts
(500 MHz, CDCl3) supporting the configuration assignment are circled]
of peridinin (1) via oxobromoacrylate 4. a) 2,2-Dimethoxypropane
(1.2 equiv), p-TsOH (cat.), toluene, reflux, 2 h, 87% diethyl (R,R)-2,3-
O,O-isopropylidenetartrate and 11% ethyl methyl (R,R)-2,3-O,O-isopro-
pylidenetartrate (in a mixture; ꢀ=98%; Ref. [20]: ꢀ=96%);
= =
Ref. [11] 100%, Ref. [12] 91%); g) Me3SiCH N N (1.2 equiv), LDA
(1.1 equiv), THF, ꢀ308C, 10 min (Ref. [12]: 92%); h) Bu4NF (3 equiv),
THF, 258C, 100 min, 75% over two steps (Ref. [12]: 78%); i) Bu3SnH
(1.1 equiv), [Pd(PPh3)4] (2 mol%), THF, 08C, 90 min, 53% over two
steps; j) Bu4NF (3 equiv), THF, 258C, 19 h, 63%. DMAP=4-dimethyl-
aminopyridine, DMF=dimethyl formamide, LDA=lithium diisopropyl-
amide.
b) Ref. [16]: HNMe(OMe)·HCl (4 equiv), Me3Al (4 equiv), CH2Cl2,
ꢀ158C, 1 h, 99%; c) Ref. [16]: MeMgBr (1.0 equiv), THF, 08C, 1 h,
ꢀ
=
ꢀ
=
66%; d) Ref. [16]: MeO2C CH CH CH PPh3 (2.0 equiv), toluene,
reflux, 30 h, 90%, E/Z>99:1; e) NaAlH4 (3.0 equiv), THF, ꢀ408C,
15 min; f) 9a/9b mixture[22] (0.83 and 0.46 equiv, respectively,
ꢀ=1.29 equiv), NaH (1.0 equiv), THF, 08C, 10 min, addition of 8, 3 h,
47% (over two steps), 95:5 E/Z mixture; g) MnO2 (40 equiv), CH2Cl2,
258C, 2 h, 84% crude product; h) F3CCO2H/H2O (9:1), 258C, 5 min,
92%; i) 1,1’-thiocarbonyldiimidazole (5 equiv), CH2Cl2 (degassed), di-
tert-butylcresol (250 ppm), exclusion of light, ꢀ788C, 10 min, 51%.
In the introductory steps we followed the literature procedure
through tert-butyldimethylsilylation,[10b] enol triflate forma-
tion,[10b] methoxycarbonylation to give the unsaturated
ester,[11] and DIBAL reduction to give the allyl alcohol.[11]
In the subsequent Sharpless epoxidation conducted with
substoichiometric amounts of Ti(OiPr)4 (0.2 equiv) and (ꢀ)-
diisopropyl tartate (DIPT, 0.3 equiv), we obtained a mixture
of desired and undesired epoxyalcohol.Our diastereoselec-
tivity (36% de) and yield (76%) were distinctly worse than
those reported by Katsumura et al.obtained under identical
conditions (92% de, 99% yield).[11] However, when we
employed 1.5 equiv of Ti(OiPr)4 and 2.3 equiv of (ꢀ)-DIPT,
the epoxidation proceeded with > 98% de and 95% yield to
provide the epoxyalcohol even on a 14-g scale.Oxidation with
the Dess–Martin periodinane[25] delivered epoxyaldehyde 13.
The C1 extension of this compound at ꢀ308C with Shioiriꢀs
lithiodiazomethane[26] followed by desilylation with Bu4NF
led to the desired alkyne 17 selectively (75% yield over two
purification or delay with 90% trifluoroacetic acid.Within
5 min the acetal was removed and the resulting diol lactonized
to provide the 5-(a-hydroxyalkyl)butenolide 11 (77% yield
over two steps).
Surprisingly the ensuing anti elimination 11!12 failed
under established[16] Mitsunobu conditions (PPh3, diethyl
azodicarboxylate, THF, ꢀ308C; Scheme 2).This was also true
for related 5-(a-hydroxyalkyl)butenolides, for example, the
coupling product obtained from 11 and 3, and the diol
obtained from 10 by deprotection/lactonization.Dehydration
of 5-(a-hydroxyalkyl)butenolide 11 to give 5-alkylidene-
butenolide 12 was finally effected at ꢀ788C by reaction
with excess 1,1’-thiocarbonyldiimidazole for 10 min.When
4024 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4023 –4027

Angewandte
Chemie
steps).[27] Alkyne 17 was an intermediate en route to
allenylsulfone 2 (Scheme 4) but was discarded as a precursor
of alkenylstannane 3 (Scheme 3) because its hydrostannyla-
tion was ineffective.When [Pd(PPh 3)4] was used as a catalyst,
catalyzed oxidation with H2O2 carried sulfide 22 on into
key sulfone 2.The yield of 2 and its configurational
homogeneity could not be maximized concomitantly: When
22 was oxidized for 2.5 h, 2 was obtained in 99% yield as an
inseparable[33] 74:26 E/Z mixture; when the reaction time was
limited to 60 min, 2 was isolated in 88% yield as a 90:10 E/Z
mixture.
[34]
The final steps of the synthesis of peridinin (1) consisted
of combining oxobromobutenolide 12 with alkenylstannane 3
and allenylsulfone 2 (Scheme 5).Because of the configura-
Scheme 4. Synthesis of the allenylsulfone 2. a) 19 (1.2 equiv), [Pd-
(OAc)2] (5 mol%), tris(2,6-dimethoxyphenyl)phosphane (5 mol%),
THF, 258C, 44 h, 61%; b) DIBAL (8 equiv), CH2Cl2, 08C, 10 min, 91%;
c) PPh3 (1.0 equiv), 2-mercaptobenzothiazole (1.0 equiv), diethyl azodi-
carboxylate (1.0 equiv), THF, 258C, 10 min, 86%; d) Ac2O (3 equiv),
DMAP (10 mol%), pyridine, 258C, 10 min, 97%, 98:2 E/Z mixture;
e) H2O2 (50 equiv), (NH4)6Mo7O24·4H2O (0.3 equiv), EtOH, 258C,
2.5 h, 99%, 74:26 E/Z mixture.
the yield was 39%, and with [PdCl2(PPh3)2] the yield was
36%; this is in contrast to yields of 69%[12] and > 64%[28]
obtained under the latter conditions elsewhere.Alkenylstan-
nane 3 was prepared more efficiently by hydrostannylating
the silyl-protected alkyne 14, instead of the silyl-free alkyne
17, and desilylating thereafter.This sequence furnished 3 in
33% rather than 29% overall yield in three steps from
aldehyde 13.
The transformation of epoxyalkyne 17 into allenylsulfone
2 began with the palladium-catalyzed addition[29] of the
former to methyl tetrolate (19; Scheme 4).This led to the
unsaturated epoxyester 20 with complete E selectivity.When
20 was treated with 8 equiv of DIBAL, the unsaturated ester
moiety gave an allyl alcohol and the epoxyalkyne moiety an
allenol.The latter proceeded with complete 1,3-chirality
transfer[30] such that triol 23[31] resulted as a single stereoiso-
mer (91%).The OH groups of this triol were perfectly
differentiable in the following derivatizations: Firstly, 2-
mercaptobenzothiazole replaced the allylic OH group in a
Mukaiyama redox condensation in the presence of PPh3 and
diethyl azodicarboxylate[32] affording dihydroxysulfide 21 in
86% yield.Secondly, acetylation with Ac 2O/DMAP occurred
at the secondary rather than tertiary OH group, which
provided acetoxyhydroxysulfide 22 in 97% yield.The product
contained 2% of the Z isomer, which could not be removed
by flash chromatography on silica gel.[33] (NH4)6Mo7O24-
Scheme 5. Total synthesis of peridinin (1[18]). a) 3 (1.2 equiv), [Pd-
(PPh3)4] (10 mol%), CuI (1.65 equiv), di-tert-butylcresol (250 ppm), N-
methylpyrrolidone (degassed), exclusion of light, 258C, 18 h, 83%
yield of the isomeric mixture; b) addition of KHMDS (5.0 equiv) to
solution of 24 (1.0 equiv of indicated mixture) and 2 (1.2 equiv of
indicated mixture) in THF (degassed), di-tert-butylcresol (250 ppm),
exclusion of light, ꢀ788C, 5 min, 61%; c) CH3CN/H2O (70:30), 258C,
exclusion of light, 37 d, HPLC, 57% pure 1 [or 89% 1 taking into
account 37% yield of recovered (6’cis)-1]. KHMDS=potassium hexa-
methyldisilazanide.
tional lability of polyenes 12, 24, and 1 all transformations
were conducted in degassed solvents, in the absence of light,
and in the presence of a radical scavenger.A Stille coupling [35]
of 12 (5Z/5E 93:7) with 3 (E/Z 100:0)—catalyzed by [Pd-
(PPh3)4] and cocatalyzed by CuI[36]—gave the butenolide-
containing pentaene 24 in 83% yield as a 75:22:3 mixture of
isomers.This mixture and allenylsulfone 2 (E/Z 74:26) were
deprotonated with a large excess of KHMDS in the final
(Sylvestre) Julia olefination.[37] Within 5 min this furnished
61% of a mixture of peridinin isomers, the major constituent
Angew. Chem. Int. Ed. 2006, 45, 4023 –4027
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4025

Communications
being (6’cis)-1 (54%).The ratio of isomers (6 ’cis)-1/1—
[1] E.Hofmann, P.M.Wrench, F.P.Sharples, R.G.Hiller, W.Welte,
K.Diederichs, Science 1996, 272, 1788 – 1791.
[2] a) D.M.Anderson, Sci. Am. 1994, 271(2), 62 – 68; b) E.Culotta,
Science 1992, 257, 1476 – 1477; c) http://www.whoi.edu/redtide/.
[3] H.Nishino, Mutat. Res. 1998, 402, 159 – 163.
originating from precursors (5Z,2’’’E)-24 and E-2—was 87:13,
and the ratio of isomers (6’cis,8’Z)-1/(8’Z)-1—originating
from precursors (5Z,2’’’E)-24 and Z-2—was 82:18.These
findings agree well with recent reports on the cis selectivity of
diene- or polyene-delivering olefinations with other allyl-
(benzothiazolyl)sulfones.[38] After isomer separation by prep-
arative reversed-phase HPLC, we obtained our first specimen
of correctly configured synthetic peridinin (1).More interest-
ingly, the purified isomer (6’cis)-1, which was our major
olefination product, gave synthetic peridinin (1) selectively
upon prolonged storage of a solution in 70% aq acetonitrile at
room temperature in the dark.After 37 days we isolated 1 in
57% yield, and unchanged (6’cis)-1 was recovered in 37%
yield; this corresponds to a conversion to 1 in 89% yield.
The 1H NMR shifts and JH,H values determined at
500 MHz in a CD3OD solution of synthetic peridinin (1)
and its 13C NMR shifts recorded at 125.7 MHz agree well with
the data reported for natural peridinin (1) under identical
conditions.[39] The same is true with respect to the eight
1H NMR shift values (500 MHz) for the olefinic carbon atoms
published for natural peridinin (1) in C6D6.[40]
[4] J.A.Haugan, T.Aakermann, S.Liaaen-Jensen,
Methods Enzy-
mol. 1992, 213, 231 – 245.
[5] Carotenoids (Eds.: G. Britton, S. Liaaen-Jensen, H. Pfander),
Carotenoids—Handbook (compiled by A.Z.Mercadante, E.S.
Egeland), Birkhäuser, Basel, 2004.
[6] g-Alkylidenebutenolide rings are also present in the following
carotenoids: a) peridinol: Ref.[6c]; b) anhydroperidinol: D.J.
Repeta, R.B.Gagosien, Geochim. Cosmochim. Acta 1984, 48,
1265 – 1277; c) pyrrhoxanthin: two-dimensional structure: J.E.
Johansen, W.A.Svec, S.Liaaen-Jensen, F.T.Haxo, Phytochem-
istry 1974, 13, 2261 – 2271; three-dimensional structure: T.
Aakermann, S.Liaaen-Jensen, Phytochemistry 1992, 31, 1779 –
1782; d) pyrrhoxanthinol: Ref.[6c]; e) hydratopyrrhoxanthinol:
S.Hertzberg, V.Partali, S.Liaaen-Jensen,
Acta Chem. Scand.
Ser. B 1988, 42, 495 – 503; f) uriolide: P.Foss, R.R.L.Guillard, S.
Liaaen-Jensen, Phytochemistry 1986, 25, 119 – 124; g) deepox-
yuriolide: E.S.Egeland, S.Liaaen-Jensen, Phytochemistry 1995,
40, 515 – 520; h) anhydrouriolide: Ref.[6g]; i) 3 ’-dehydrourio-
lide: Ref.[6g]; j) unnamed carotenoid: T.Maoka, K.Hashimoto,
We also undertook an intensive NMR study of isomer 1
possessing the natural configuration and of its precursor
(6’cis)-1 exhibiting the unnatural configuration of the
N.Akimoto, Y.Fuhiwara,
J. Nat. Prod. 2001, 64, 578 – 581;
k) unnamed carotenoid: M.Suzuki, K.Watanabe, S.Fujiwara, T.
Kurasawa, T.Wakabayashi, M.Tsuzuki, K.Iguchi, T.Yamori,
Chem. Pharm. Bull. 2003, 724 – 727.
6’
7’ [18]
=
C
C
bond.We used C 6D6 solutions because in this
solvent the signal (1H) dispersion was greater than in the
previously used[39] CD3OD, and we were able to assign all 1H
and 13C resonances individually.The ROESY spectra of 1 and
(6’cis)-1 proved the two stereostructures based on appropriate
nOe crosspeaks.Specifically, the configurations of the double
bonds were assigned in this manner, the relative configura-
tions of the stereocenters established, and even the 3D
structure of the allene moiety elucidated.
[7] a) I.E. Swift, B.V. Milborrow, Biochem. J. 1981, 199, 69 – 74;
b) I.E.Swift, B.V.Milborrow, Biochem. J. 2005, 389, 919 – 919
(retraction of prior findings).
[8] F.Schütt, Ber. Dtsch. Bot. Ges. 1890, 8, 9 – 32.
[9] Two-dimensional structure: a) H.H. Strain, W.A. Svec, K.
Aitzetmüller, M.C.Grandolfo, J.J.Katz, H.Kjøsen, S.Norgård,
S.Liaaen-Jensen, F.T.Haxo, P.Wegfahrt, H.Rapoport,
Chem. Soc. 1971, 93, 1823 – 1825; three-dimensional structure:
b) HH. .Strain, WA. .Svec, P.Wegfahrt, H.Rapoport, FT. .
Haxo, S.Norgård, H.Kjøsen, S.Liaaen-Jensen,
J. Am.
Acta Chem.
In summary, we have accomplished a highly convergent
total synthesis of peridinin (1).Starting from actinol ( 5) and
diethyl tartrate (6), 1 was synthesized in 15 steps plus two
HPLC separations in the longest linear sequence.A single
HPLC separaration may have sufficed if the isomerization
(6’cis)-1!1 had been effected with the originally obtained
(6’cis)-1/1/(6’cis,8’Z)-1/(8’Z)-1 mixture rather than with pure
(6’cis)-1.The overall yield was 77. % and the total number of
steps[41] 29.[42] Key transformations were the differential
reduction 7!8 of an ester-containing Weinreb amide, the E-
selective olefination 8!10 by the Ando-type bromophosph-
onates 9a/9b, the anti-selective b-elimination 11!12 upon
treatment with 1,1’-thiocarbonyldiimidazole, which estab-
Scand. Ser. B 1976, 30, 109 – 120; c) J.E.Johansen, G.Borch, S.
Liaaen-Jensen, Phytochemistry 1980, 19, 441 – 444.
[10] a) Synthesis of a racemic mixture of diastereomers of 1: M.Ito,
Y.Hirata, Y.Shibata, K.Tsukida, J. Chem. Soc. Perkin Trans. 1
1990, 197 – 199; b) synthesis of enantiomerically and diastereo-
merically pure 1: Y.Yamano, M.Ito, J. Chem. Soc. Perkin Trans.
1 1993, 1599 – 1610.
[11] a) N.Furuichi, H.Hara, T.Osaki, H.Mori, S.Katsumura,
Angew. Chem. 2002, 114, 1065 – 1068; Angew. Chem. Int. Ed.
2002, 41, 1023 – 1026; b) N.Furuichi, H.Hara, T.Osaki, M.
Takano, H.Mori, S.Katsumura, J. Org. Chem. 2004, 69, 7949 –
7959.
[12] B.Vaz, R.Alvarez, R.Brückner, A.R.de Lera,
Org. Lett. 2005,
7, 545 – 548.
1’
5
=
lished the C
C bond Z-selectively in an unprecedented
[13] a) Review: R.Brückner, Chem. Commun. 2001, 141 – 152;
individual reports: b) J.Schmidt-Leithoff, R.Brückner, Helv.
Chim. Acta 2005, 88, 1943 – 1958; c) F.von der Ohe, R.Brück-
ner, New J. Chem. 2000, 24, 659 – 669; d) I.Hanisch, R.
Brückner, Synlett 2000, 374 – 378; e) K.Siegel, R.Brückner,
Chem. Eur. J. 1998, 4, 1116 – 1122; f) F.C.Görth, A.Umland, R.
Brückner, Eur. J. Org. Chem. 1998, 1055 – 1062.
manner, and the cis!trans isomerization (6’cis)-1!1 as the
ultimate step.
Received: February 7, 2006
Published online: May 9, 2006
[14] Reviews on other approaches to g-alkylidenebutenolides:
a) E.-i. Negishi, M. Kotora, Tetrahedron 1997, 53, 6707 – 6738;
b) R.Brückner, Curr. Org. Chem. 2001, 5, 679 – 718; c) R.Rossi,
F.Bellina in Targets in Heterocyclic Systems: Chemistry and
Properties, Vol. 5 (Eds.: O. A. Attanasi, D. Spinelli), Società
Chimica Italiana, 2002, pp.169 – 198.
Keywords: butenolides · carotenoids · elimination · olefination ·
total synthesis
.
4026 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4023 –4027

Angewandte
Chemie
[15] Nomenclature according to D.Seebach, V.Prelog,
Angew.
[35] Review: V.Farina, V.Krishnamurthy, W.J.Scott,
Org. React.
Chem. 1982, 94, 696 – 702; Angew. Chem. Int. Ed. Engl. 1982, 21,
654 – 660.
[16] T.Olpp, R.Brückner, Angew. Chem. 2005, 117, 1577 – 1581;
Angew. Chem. Int. Ed. 2005, 44, 1553 – 1557.
1997, 50, 1 – 652.
[36] Method: L.S.Liebeskind, R.W.Fengl, J. Org. Chem. 1990, 55,
5359 – 5364.
[37] Method: a) Ref.[32a]; b) Ref.[34]; c) J.B.Baudin, G.Hareau,
S.A.Julia, R. Lorne, O. Ruel, Bull. Soc. Chim. Fr. 1993, 130,
856 – 878.
[17] Preparation of 5: a) HG. W. .Leuenberger, W.Boguth, E.
Widmer, R.Zell, Helv. Chim. Acta 1976, 59, 1832 – 1849; b) Y.
Crameri, K.Puentener, M.Scalone (F.Hoffmann-La Roche AG,
Switzerland), EPXXDW EP 915076A1 19990512, 1999; c) T.
Hoshino, Y.Setoguchi (DSM Ip Assets BV. ,. Netherlands),
PIXXD2 WO 2004029263A2 20040408 PCT Int.Appl., 2004.
[18] The nomenclature in this communication is in compliance with
general IUPAC rules (http://www. chem.qmul.ac.uk/iupac/). The
recommended carotenoid-specific IUPAC nomenclature for
peridinin (1) can be found in Ref.[5] and Ref.[39].
[38] a) A.Sorg, R.Brückner, Synlett 2005, 289 – 293; b) B.Vaz, R.
Alvarez, J.A. Souto, A.R. de Lera,
c) Ref.[11].
[39] J.Krane, T.Aakermann, S.Liaaen-Jensen, Magn. Reson. Chem.
1992, 30, 1169 – 1177.
Synlett 2005, 294 – 298;
[40] Y.Yamano (nꢁe Hirata), S.Sumiya, K.Suzuki, Y.Kurimoto, Y.
Koyama, T.Shimamura, M.Ito,
2991 – 2994.
[41] Counting HPLC separations as steps.
Tetrahedron Lett. 1992, 33,
[19] a) E.Widmer, M.Soukup, R.Zell, E.Broger, H.P.Wagner, M.
Imfeld, Helv. Chim. Acta 1990, 73, 861 – 867; b) M.Soukup, E.
[42] This compares well with previous syntheses: a) Itoꢀs synthesis of
1 from (R)-4-levodione, (E)-3-methyl-2-penten-4-in-1-ol, and
(E,E)-6-hydroxy-2-methyl-2,4-hexadienal (Ref.[10]) comprised
over 35 steps and included 19 steps in the longest linear sequence
(Ref.[41]); b) Katsumuraꢀs access to 1 from (R,R)-actinol and 2-
butyn-1-ol (Ref.[11]) consists of 33 steps, 18 thereof in the
longest linear sequence (Ref.[41]); c) ( R,R)-actinol, crotonic
acid, diethyl methylmalonate, 2-butyn-1-ol, and propynol pro-
vided 6’-epi-1 according to Ref.[12], requiring 30 steps overall
and 14 in the longest linear sequence.
ˇ
Widmer, T.Lukµc , Helv. Chim. Acta 1990, 73, 868 – 873.
[20] M.Carmack, C.J.Kelley, J. Org. Chem. 1968, 33, 2171 – 2173.
[21] K.Ando, T.Oishi, M.Hirama, H.Ohno, T.Ibuka, J. Org. Chem.
2000, 65, 4745 – 4749, and references therein.
[22] T.Olpp, R.Brückner, Synthesis 2004, 2135 – 2152.
1
[23] All new compounds gave satisfactory H and 13C NMR spectra
and correct combustion analyses—except the unstable hydroxyl-
aldehyde 8, aldehydes 11 and 12, thioethers 21 and 22, and
peridinin (1); each of these compounds, however, provided a
correct high-resolution mass spectrum.
[24] A one-pot reduction/olefination procedure 7!10 failed because
of the incompatibility of unconsumed reducing agent with the
deprotonated phosphonates 9a and 9b.
[25] a) D.B.Dess, J.C.Martin, J. Org. Chem. 1983, 48, 4155 – 4156;
b) D.B.Dess, J.C.Martin, J. Am. Chem. Soc. 1991, 113, 7277 –
7287.
[26] Method: K.Miwa, T.Aoyama, T.Shioiri, Synlett 1994, 107 – 108.
[27] At lower temperatures (e.g., at ꢀ788C for 5 min) and again after
desilylation, we obtained only half the amount of alkyne (in the
cited case: 40% 17) but also the corresponding methylketone 15
(in the cited case: 14%).Probably, 15 results from a ꢁ [1,2]
hydrogen shift in the “Tiffeneau/Demjanov-type substructure”
ꢀ
+
ꢀ
ꢀ
ꢀ
ꢀ ꢀ ꢂ
ꢀ
ꢀ
ꢀ ꢀ
O (H )C C N N
of
a
zwitterion
O (H )C( R) C
+
ꢀ
ꢂ
( SiMe3)ꢀN N , which must have formed initially at least to
some extent.This side reaction was neither observed during the
same transformation earlier[12] nor does it appears to be known
from any other Shiorii C1-extension.
[28] trans-Selective hydrostannylation of 17 with Bu3SnH and cat.
[PdCl2(PPh3)2]: M.Kuba, N.Furuichi, S.Katsumura, Chem. Lett.
2002, 1248 – 1249.
[29] Method: B.M.Trost, M.T.Sorum, C.Chan, A.E.Harms, G.
Rühter, J. Am. Chem. Soc. 1997, 119, 698 – 708.
[30] DIBAL reduction of carotenoid epoxyalkynes giving allenols:
a) original report (to the best of our knowledge): E.Widmer,
Pure Appl. Chem. 1985, 57, 741 – 752; b) specific precedence: M.
Ito, Y.Hirata, K.Tsukida, N.Tanaka, K.Hamada, R.Hino, T.
Fujiwara, Chem. Pharm. Bull. 1988, 36, 3328 – 3340.
[31] Different preparations of 23: a) racemic: Ref.[30b]; b) enantio-
merically pure: A.Baumeler, C.H.Eugster, Helv. Chim. Acta
1991, 74, 469 – 487; c) enantiomerically pure: Ref.[10b]; d) non-
stereoselective: M.Nakano, N.Furuichi, H.Mori, S.Katsumura,
Tetrahedron Lett. 2001, 42, 7307 – 7310.
[32] Method: a) J.B. Baudin, G. Hareau, S.A. Julia, O. Ruel,
Tetrahedron Lett. 1991, 32, 1175 – 1178; b) P.R. Blakemore,
P.J.Kocienski, A.Morley, K.Muir, J. Chem. Soc. Perkin Trans. 1
1999, 955 – 968.
[33] W.C.Still, M.Kahn, A.Mitra, J. Org. Chem. 1978, 43, 2923 –
2925.
[34] Method: J.B.Baudin, G.Hareau, S.A.Julia, O.Ruel,
Bull. Soc.
Chim. Fr. 1993, 130, 336 – 357.
Angew. Chem. Int. Ed. 2006, 45, 4023 –4027
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4027