Convergent synthesis of the c18-c30 fragment of amphidinol 3

Article abstract of DOI:10.1055/s-0029-1217754

The C18-C30 fragment of amphidinol 3 has been synthesized in a convergent fashion by employing two asymmetric Sharpless dihydroxylations, a Julia-Kocienski olefination and a Wittig reaction as the key steps.

Full text of DOI:10.1055/s-0029-1217754

2706
LETTER
Convergent Synthesis of the C18–C30 Fragment of Amphidinol 3
S
ynthesis
a
of the
C
18–
n
C30 Fragm
e
i
nt of
A
n
mphidinol
3
e Cossy,*a Tomoki Tsuchiya,a Sébastien Reymond,a Thomas Kreuzer,b Françoise Colobert,b István E. Markóc
a
Laboratoire de Chimie Organique, ESPCI ParisTech, CNRS, 10 Rue Vauquelin, 75231 Paris Cedex 05, France
Fax +33(1)40794660; E-mail: janine.cossy@espci.fr
b
c
Laboratoire de Stéréochimie, Université de Strasbourg, ECPM, CNRS, 25 Rue Becquerel, 67087 Strasbourg Cedex 2, France
Université Catholique de Louvain, Unité de Chimie Organique et Médicinale,
Bâtiment Lavoisier, Place Louis Pasteur 1, 1348 Louvain-la Neuve, Belgium
Received 28 June 2009
ture was first elucidated by Murata et al. in 1999.4 Recent-
ly, the configuration of the C2 stereocenter has been
revised (Scheme 1).5,6 The structural complexity of AM3,
coupled with its interesting biological properties, has
stimulated a wide range of synthetic activity from organic
chemists and an array of fragments have been prepared,7–10
but the total synthesis of AM3 has not been completed yet.
Abstract: The C18–C30 fragment of amphidinol 3 has been syn-
thesized in a convergent fashion by employing two asymmetric
Sharpless dihydroxylations, a Julia–Kocienski olefination and a
Wittig reaction as the key steps.
Key words: amphidinol, Julia–Kocienski olefination, Wittig olefi-
nation, Sharpless dihydroxylation
Previously, we have reported the synthesis of the C1–C13
fragment7a and two syntheses of the C50–C67 fragment of
AM3.7b,c As we plan to construct the C17–C18 and C30–
C31 bonds by using Julia–Kocienski olefinations, we re-
port here the synthesis of the C18–C30 fragment which
possesses six of the 25 stereogenic centers present in
AM3. According to our retrosynthetic analysis, fragment
C18–C30 A could be disconnected into two simplified
Marine dinoflagellates, a type of primitive unicellular al-
gae, are a rich source of natural products such as long-
chain polyketides, amphidinolides, or cyclic polyethers.1,2
Among dinoflagellates, the genus Amphidinum klebsii has
been recognized as a source of novel bioactive secondary
metabolites such as amphidinol 3 (AM3), an antifungal
and haemolytic compound, which was isolated for the
first time by Yasumoto et al. in 1991.3 Amphidinol 3 is a
polyhydroxylated and polyenic compound whose struc-
subunits, phosphonium salt
B
and aldehyde
C
(Scheme 2). These two fragments would be coupled by
1
HO
OH
OH
OH
OH
17
18
OH
OH
OH OH
OH
OH
30
OH
OH
OH
O
OH
HO
OH
OH
OH
67
O
OH
OH
amphidinol 3
PG2O
OPG1
OPG1 OPG1
18
OPG3
30
OPG1
OPG1
A
Scheme 1 Amphidinol 3 revised structure and fragment C18–C30
SYNLETT 2009, No. 16, pp 2706–2710
x
x
.x
x
.2
0
0
9
Advanced online publication: 03.09.2009
DOI: 10.1055/s-0029-1217754; Art ID: G21109ST
© Georg Thieme Verlag Stuttgart · New York

LETTER
Synthesis of the C18–C30 Fragment of Amphidinol 3
2707
employing a Wittig olefination, in order to install a double
bond at C24–C25 possessing a Z-configuration. Indeed,
this geometry is required for the Sharpless asymmetric di-
hydroxylation to provide us with the correct oxygenated
stereogenic centers at C24 and C25. Fragment B would be
easily prepared from hex-5-en-2-one (1) (Scheme 2).
Concerning fragment C, the C20 and C21 oxygenated ste-
reocenters would be installed by applying a Sharpless
asymmetric dihydroxylation to the corresponding E-ole-
fin, whereas the C23 stereogenic center would be generat-
ed by alkylation of Oppolzer’s chiral sultam.
1) NaBH4, MeOH, Et2O, r.t.
30
NH
CCl3
2)
O
OPMB
PMBO
1
2
CSA, C6H12–CH2Cl2, r.t.
85%
1) OsO4 (cat.), NMO
t-BuOH–H2O, r.t.
2) NaIO4, THF–H2O
3) (S,S)-I, Et2O, –78 °C
4) TBSOTf, 2,6-lutidine
CH2Cl2, –20 °C to r.t.
50%
OH OTBS
1) OsO4, NMO
t-BuOH–H2O, r.t.
OTBS
PG2O
27
18
2) NaIO4, THF–H2O
3) NaBH4, EtOH
0 °C to r.t.
OPG1
OPG1 OPG1
25
OPMB
OPMB
30 OPG3
3
4
24
98%
OPG1
OPG1
Wittig / Sharpless
1) I2, PPh3, imidazole
MeCN, r.t.
2) Ph3P, MeCN, Δ
A
78%
Julia / Sharpless
PG2O
18
OPG1
Ph
Ph
Ti
O
OPG1
B
O
O
X–
25
+
Ph3P
O
OPG3
30
+
24
OPG1
H
OTBS
Ph
O
25
Ph
30
C
I–
Ph3P
Oppolzer's
alkylation
OPMB
(S,S)-I
5
30
O
1
Scheme 3 Synthesis of phosphonium salt 5
Scheme 2 Retrosynthetic analysis of fragment C18–C30
the presence of HMPA, and then reacted with allyl bro-
mide. After removal of the chiral auxiliary (lithium pyrro-
lidinoborohydride, THF, 0 °C) and protection of the
hydroxy group (TBSCl, imidazole, DMAP, DMF–THF,
r.t.), 7 was isolated in 74% overall yield (over three steps).
The terminal olefin of the chiral silyl ether 7 was subject-
ed to ozonolysis (O3, CH2Cl2, –78 °C then Et3N)13 to form
the corresponding aldehyde which was condensed with
phenyltetrazolyl sulfone 814 under Julia–Kocienski condi-
tions (KHMDS, THF, –78 °C then r.t.) to generate the E-
alkene 9 in 55% yield over two steps (E/Z > 95:5). In order
to control the absolute and relative configurations of the
stereogenic centers at C20 and C21, E-alkene 9 was sub-
jected to a Sharpless asymmetric dihydroxylation using
AD-mix-a to furnish two diols 10 and 10¢ as a 80:20 mix-
ture. After separation by flash chromatography on silica
gel, the major diol was protected with TBSOTf (2,6-luti-
dine, CH2Cl2, –78 °C then r.t.) to yield 10 in 69% yield
from 9. A selective deprotection of the primary hydroxy
group at C24 present in 10 was achieved by treatment with
ammonium fluoride in methanol at 60 °C (79% yield), as
the reaction was ineffective at room temperature. The re-
sulting primary alcohol was then oxidized using Dess–
Martin periodinane to form aldehyde 11 in quantitative
yield. The key intermediate 11 (fragment C18–C24),
bearing three stereogenic centers, was efficiently prepared
in nine steps and with an overall yield of 22%. With com-
pounds 11 (fragment C18–C24) and 5 (fragment C25–
C30) in hand, we turned our attention towards the crucial
The synthesis of the C18–C30 fragment of AM3 started
with the transformation of unsaturated ketone 1 into a
phosphonium salt of type B (Scheme 3). After reduction
of the ketone function in 1 (NaBH4, MeOH–Et2O, r.t.) and
protection of the obtained alcohol as a p-methoxybenzyl
(PMB) ether [PMBOC(=NH)CCl3, CSA (cat.)], the unsat-
urated ether 2 was isolated (85% overall yield for the two
steps). Oxidative cleavage of the terminal double bond in
2 [OsO4 (cat.), NMO, then NaIO4] was followed by the
treatment of the crude aldehyde with the highly face-
selective Duthaler allyltitanium complex (S,S)-I11 in order
to install the C27 stereogenic center. The obtained alcohol
was then protected as a tert-butyldimethylsilyl (TBS)
ether (TBSOTf, 2,6-lutidine, CH2Cl2), and 3 was isolated
in 50% overall yield (4 steps, starting from 2). In order to
synthesize the phosphonium salt 5 (compound of type B),
3 was transformed into alcohol 4 by oxidative cleavage of
the terminal double bond (OsO4 cat., NMO, then NaIO4)
and reduction of the intermediate aldehyde with sodium
borohydride (98% over three steps). Alcohol 4 was then
iodinated (I2, Ph3P, imidazole) and further reacted with
triphenylphosphine to yield the desired phosphonium salt
5 in 78% overall yield (Scheme 3).
Having prepared 5 (C25–C30 fragment), the synthesis of
the C18–C24 fragment (compound of type C) was per-
formed (Scheme 4). The optically active protected alco-
hol 7 was prepared by alkylation of Oppolzer’s chiral
sultam.12 Compound 6 was first treated with LiHMDS, in
Synlett 2009, No. 16, 2706–2710 © Thieme Stuttgart · New York

2708
J. Cossy et al.
LETTER
final coupling reaction. The Wittig olefination of 11 by a of similar compounds,17 strongly indicated that the major
phosphonium ylide was essential as an asymmetric dihy- isolated diastereomer had the correct relative and absolute
droxylation of a Z-double bond will generate a diol with configuration at the C23 and C24 stereogenic centers.18
the anti relative stereochemistry. When the Wittig reac-
tion of aldehyde 11 with the phosphonium ylide, generat-
In order to complete the synthesis of alcohol 16 (C18–C30
fragment of AM3), selective cleavage of the benzyl ether
ed from 5 with KHMDS, was performed in toluene, the
at C18 over the p-methoxybenzyl ether at C30 was
Z-olefin was isolated in moderate yield (48%) but a good
achieved upon treatment of 15 with Raney nickel W-4.19
stereocontrol of the configuration of the double bond was
Alcohol 16 was obtained in 70% yield20 (Scheme 5).
observed (Z/E > 95:5).15 Although asymmetric dihydrox-
In summary, a convergent approach to the C18–C30 frag-
ment of amphidinol 3 was achieved in 15 steps and in
6.3% overall yield for the longest linear sequence (the to-
tal number of steps is 24). The use of a diastereoselective
ylation of Z-olefins utilizing AD-mix are generally less
stereoselective than those of E-olefins, the asymmetric di-
hydroxylation of the cis-olefin 12 provided two diastereo-
mers 13 (62% isolated yield) and 14 in a ratio of 4:1.
alkylation of Oppolzer’s chiral sultam, two Sharpless
These two diastereomers were separated by flash chroma-
asymmetric dihydroxylations and an enantioselective al-
tography on silica gel. We have to point out that the reac-
lyltitanation enabled us to control the six stereogenic cen-
ters present in this fragment.
tion was sluggish with the commercially available
AD-mix-b (1.4 mg/mmol) at 0 °C, and additional OsO4
and chiral ligand (DHQD)2PHAL (4 mol% each) were
necessary for the reaction to go to completion.16 The con-
Acknowledgment
figuration of the two newly created stereogenic centers at
C24 and C25 were determined after transformation of
We are grateful to Prof. J.-Y. Lallemand (ICSN/CNRS) for financi-
al support. We also thank M.-T. Martin (NMR department of ICSN)
both diols 13 and 14 into their corresponding acetonides
for the NMR studies performed on compounds 15 and 15¢.
15 (89%) and 15¢18 [2,2-dimethoxypropane, CSA (cat.)].
NOE experiments on both acetonides, as well as literature
comparison with chemical shifts and coupling constants
Br
1)
LiHMDS, HMPA
1) O3, Et3N
THF, –78 °C
CH2Cl2, –78 °C
N
OTBS
BnO
OTBS
24
2)
2)
S
O
Li+
N– BH3
O2
BnO
SO2PT
9
6
7
THF, 0 °C
8
3) TBSCl, imidazole
KHMDS, THF, –78 °C to r.t.
55%
DMAP, DMF–THF, r.t.
1) AD-mix-α, MeSO2NH2
t-BuOH–H2O, 0 °C
2) TBSOTf, 2,6-lutidine
CH2Cl2, –78 °C to r.t.
69%
74%
TBSO
1) NH4F, MeOH, 60 °C
2) Dess–Martin [O]
Ph
TBSO
N
N
N
24
BnO
O
N
BnO
OTBS
PT =
79%
18
24
OTBS
11
H
OTBS
10 + 10' (isomer) (dr = 80:20)
TBSO
OTBS
OH OTBS
30
24
BnO
25
18
OH
+
OPMB
OTBS
13
TBSO
BnO
KHMDS
AD-mix-β
11 + 5
OPMB
25
OSO4 (4 mol%)
(DHQD)2–PHAL (4 mol%)
t-BuOH–H2O, 0 °C
toluene, –78 °C
48%
24
OTBS
12
(Z/E > 95%)
TBSO
OTBS
OH OTBS
78% (13/14 = 4:1)
24
BnO
25
OH
OPMB
14
Scheme 4 Synthesis of aldehyde 11 and synthesis of diol 13
Synlett 2009, No. 16, 2706–2710 © Thieme Stuttgart · New York

LETTER
Synthesis of the C18–C30 Fragment of Amphidinol 3
2709
(11) The allyltitanium complex (S,S)-I was prepared according
to: Hafner, A.; Duthaler, R. O.; Marti, R.; Rihs, G.; Rothe-
Streit, P.; Schwarzenbach, F. J. Am. Chem. Soc. 1992, 114,
2321.
(12) Oppolzer, W.; Moretti, R.; Thomi, S. Tetrahedron Lett.
1989, 41, 5603.
TBSO
OTBS
OH OTBS
30
24
BnO
25
18
OH
OPMB
13
(13) Hon, Y. S.; Lin, S. W.; Lu, L.; Chen, Y. J. Tetrahedron 1995,
51, 5019.
(14) Compound 8 was prepared from 3-bromopropan-1-ol in 3
steps (40% yield).
(15) The starting aldehyde 11 (20%) was recovered as the
corresponding alcohol after the reductive workup and no
improvement in conversion was observed when the
phosphonium bromide was utilized.
89%
MeO
OMe
CSA (cat.)
Me
Me
TBSO
OTBS
O
OTBS
25
30
BnO
24
18
(16) On large scale, longer reaction times of up to one week and
vigorous stirring were required.
(17) (a) Oikawa, H.; Matsuda, I.; Kagawa, T.; Ichihara, A.;
Kohmoto, K. Tetrahedron 1994, 50, 13347. (b) For a
review, see: Bifulco, G.; Dambruso, P.; Gomez-Paloma, L.;
Riccio, R. Chem. Rev. 2007, 107, 3744.
O
OPMB
15
H2, Raney Ni W-4
EtOH, r.t.
70%
(18) NMR study (see ref. 17 for literature data): A syn
relationship between the substituents at C23 and C24 was
observed for compound 15 (Figure 1).
Me
Me
TBSO
OTBS
O
OTBS
30
HO
18
NOE
H25
Me
Me
(CH2)22
H24
O
OPMB
16
Me
O
NOE
R1
R2
24
O
23
25
22
Scheme 5 Synthesis of alcohol 16
Me23
H23
O
O
Me
15
C23(Me)-C24(O) syn
Me
H24
H25
(CH2)22
References and Notes
Me23
1H NMR (ppm)
1.35
1.17
(1) See, for example: Kobayashi, J.; Shimbo, K.; Kubota, T.;
Tsuda, M. Pure Appl. Chem. 2003, 75, 337.
(2) For a review, see: Kobayashi, J.; Kubota, T. J. Nat. Prod.
2007, 70, 451.
15
3.72
3.76
4.04
4.00
1.01
1.03
1.38
1.08
Lit.
(3) Satake, M.; Murata, M.; Yasumoto, M.; Fujita, T.; Naoki, H.
NOE
J. Am. Chem. Soc. 1991, 113, 9859.
NOE
Me23
H24
Me
Me
(4) Murata, M.; Matsuoka, S.; Matsumori, N.; Paul, G. K.;
Tachibana, K. J. Am. Chem. Soc. 1999, 121, 870.
(5) Oishi, T.; Kanemoto, M.; Swasono, R.; Matsumori, N.;
Murata, M. Org. Lett. 2008, 10, 5203.
(6) Recently, the C2 stereocenter of AM2 was also proposed to
be 2R by chemical correlation on a model structure, see:
Kommana, P.; Chung, S. W.; Donaldson, W. A. Tetrahedron
Lett. 2008, 49, 6209.
(7) (a) BouzBouz, S.; Cossy, J. Org. Lett. 2001, 3, 1451.
(b) Cossy, J.; Tsuchiya, T.; Ferrié, L.; Reymond, S.;
Kreuzer, T.; Colobert, F.; Jourdain, P.; Markó, I. E. Synlett
2007, 2286. (c) Colobert, F.; Kreuzer, T.; Cossy, J.;
Reymond, S.; Tsuchiya, T.; Ferrié, L.; Markó, I. E.;
Jourdain, P. Synlett 2007, 2351.
(8) (a) Chang, S.-K.; Paquette, L. A. Synlett 2005, 2915.
(b) Chang, S.-K.; Paquette, L. A. Org. Lett. 2005, 7, 3111.
(c) Bedore, M. W.; Chang, S.-K.; Paquette, L. A. Org. Lett.
2007, 9, 513.
H25
NOE
R2
Me
O
R1
24
O
25
23
22
O
H23
(CH2)22
O
15'
Me
C23(Me)-C24(O) anti
Me
H24
H25
(CH2)22
Me23
1H NMR (ppm)
1.81
1.31
15'
3.76
3.76
4.2
0.83
0.86
1.76
1.20
4.04
Lit.
Figure 1
(19) Other Raney nickel were less selective and both the benzyl
and p-methoxybenzyl groups were cleaved.
(9) (a) Huckins, J. R.; de Vincente, J.; Rychnovsky, S. D. Org.
Lett. 2005, 7, 1853. (b) Huckins, J. R.; de Vincente, J.;
Rychnovsky, S. D. Angew.Chem. Int. Ed. 2006, 7258.
(c) Huckins, J. R.; de Vincente, J.; Rychnovsky, S. D. Org.
Lett. 2007, 9, 4757.
(10) (a) Flamme, E. M.; Roush, W. R. Org. Lett. 2005, 7, 1411.
(b) Flamme, E. M.; Roush, W. R. Org. Lett. 2005, 7, 5509.
(c) Hicks, J. D.; Roush, W. R. Org. Lett. 2008, 10, 681.
(20) Compound 16 (Mixture of Epimers at C30)
Colourless oil; [a]D20 –49.3 (c 0.5, CHCl3). IR (film): 2927,
2855, 1712, 1608, 1513, 1462, 1378, 1248, 1217, 1169,
1094, 1062, 1039, 1006 cm–1. 1H NMR (400 MHz, C6D6):
d = 7.19–7.13 (m, 2 H), 6.79–6.74 (m, 2 H), 4.40–4.27 (m, 2
H), 3.96–3.86 (m, 2 H), 3.70 (s, 3 H), 3.69–3.60 (m, 3 H),
3.57–3.46 (m, 2 H), 3.40–3.31 (m, 1 H), 1.89–1.79 (m, 1 H),
1.78–1.68 (m, 2 H), 1.66–1.55 (m, 1 H), 1.54–1.31 (m, 4 H),
1.38 (s, 3 H), 1.30–1.01 (m, 7 H), 1.24 (s, 3 H), 0.96–0.89
Synlett 2009, No. 16, 2706–2710 © Thieme Stuttgart · New York

2710
J. Cossy et al.
LETTER
(m, 3 H), 0.86–0.71 (m, 27 H), 0.02 to –0.07 (m, 18 H).
13C NMR (100 MHz, C6D6): d = 159.1 (s), 130.9 (s), 129.5
and 129.1 (d), 113.7 (d), 107.7 and 107.5 (s), 82.9 (d), 75.1
(d), 75.1 and 74.9 (d), 72.5 and 72.3 (d), 71.6 (d), 69.8 and
69.7 (t), 68.8 and 68.7 (d), 60.6 and 60.4 (t), 55.2 (q), 36.5
and 36.1 (t), 34.2 (t), 33.5 and 33.5 and 33.5 (t), 31.0 (t), 30.7
and 30.6 (t), 29.7 (t), 28.7 and 28.7 (q), 28.5 and 28.4 (d),
26.1 (2C, q), 26.1–25.8 (q), 22.6 and 22.6 (t), 19.8 and 19.8
(q), 18.0–17.9 (s), 16.5 (q), –4.1 to –4.7 (q). ESI-HRMS: m/
z calcd for C44H86O8Si3 + Na+: 849.5523; found: 849.5506.
Synlett 2009, No. 16, 2706–2710 © Thieme Stuttgart · New York