An intramolecular case of sharpless kinetic resolution: Total synthesis of laulimalide

Article abstract of DOI:10.1002/1521-3773(20011015)40:20<3842::AID-ANIE3842>3.0.CO;2-R

The microtubule-stabilizing antitumor agent laulimalide (1) has been obtained in by new synthetic route. The carbon skeleton was assembled by means of Julia-Kocienski (C16-C17) and Horner-Wadsworth-Emmons (C21-C22) olefinations. Still-Gennari olefination

Full text of DOI:10.1002/1521-3773(20011015)40:20<3842::AID-ANIE3842>3.0.CO;2-R

COMMUNICATIONS
[3] M. R. Buchmeiser, Chem. Rev. 2000, 100, 1565 ± 1604.
[4] S. Nguyen, R. H. Grubbs, J. Organomet. Chem. 1995, 497, 195 ± 200.
[5] A. G. M. Barrett, S. M. Cramp, R. S. Roberts, Org. Lett. 1999, 1,
1083 ± 1086.
[6] Q. Yao, Angew. Chem. 2000, 112, 4060 ± 4062; Angew. Chem. Int. Ed.
2000, 39, 3896 ± 3898.
[7] J. Dowden, J. Savovic, Chem. Commun. 2001, 37 ± 38.
[8] S. C. Schürer, S. Gessler, N. Buschmann, S. Blechert, Angew. Chem.
2000, 112, 4062 ± 4065; Angew. Chem. Int. Ed. 2000, 39, 3898 ± 3901.
[9] M. Scholl, S. Ding, C. W. Lee, R. H. Grubbs, Org. Lett. 1999, 1, 953 ±
956.
[10] M. Scholl, T. M. Trnka, J. P. Morgan, R. H. Grubbs, Tetrahedron Lett.
1999, 40, 2247 ± 2250.
[11] L. Ackermann, A. Fürstner, T. Weskamp, F. J. Kohl, W. A. Herrmann,
Tetrahedron Lett. 1999, 40, 4787 ± 4790.
Figure 3. Influence of the chain transfer reagent I on the stability of the
immobilized catalyst as determined from RCM experiments carried out
[12] J. Huang, E. D. Stevens, S. P. Nolan, J. L. Petersen, J. Am. Chem. Soc.
1999, 121, 2674 ± 2678.
[13] M. Kubin, P. Spacek, R. Chromecek, Collect. Czech. Chem. Commun.
1967, 32, 3881 ± 3887.
~
*
with diethyldiallylmalonate with ( ) and without ( ) CTA. A ˆ Activity of
the catalyst.
[14] L. C. Hansen, R. E. Sievers, J. Chromatogr. 1974, 99, 123 ± 133.
ture and the lack of basically any microporosity reduce
diffusion to a minimum. This results in turnover frequences
(TOFs) up to 25 min , thus even exceeding homogeneous
Â
[15] S. Hjerten, J.-L. Liao, R. Zhang, J. Chromatogr. 1989, 473, 273 ± 275.
Â
[16] S. Hjerten, Y.-M. Li, J.-L. Liao, J. Mohammad, K. Nakazato, G.
Pettersson, Nature 1992, 356, 810 ± 811.
[17] J. A. Tripp, F. Svec, J. M. J. Frechet, J. Comb. Chem. 2001, 3, 216 ± 223.
[18] J. A. Tripp, J. A. Stein, F. Svec, J. M. J. Frechet, Org. Lett. 2000, 2,
1
Â
analogues. In comparison, the TOF for diethyldiethylmalo-
Â
1
nate using [Cl2Ru(Mes2-NHC)(PCy3)(CHPh)] is 4 min
(458C).[8]
195 ± 198.
[19] F. Sinner, M. R. Buchmeiser, Angew. Chem. 2000, 112, 1491 ± 1494;
Angew. Chem. Int. Ed. 2000, 39, 1433 ± 1436.
[20] M. R. Buchmeiser, F. Sinner, M. Mupa, K. Wurst, Macromolecules
2000, 33, 32 ± 39.
In terms of a most simple handling, the monolithic systems
presented here may be used either as pressure-stable reactors
or (in miniaturized form) as cartridges for applications in
combinatorial chemistry. The use of NHC ligands even in
RCM successfully suppresses any bleeding of the column, thus
allowing the synthesis of virtually ruthenium-free cyclization
products with a ruthenium content ꢀ70 ppm.
Â
[21] I. Halasz, K. Martin, Angew. Chem. 1978, 90, 954 ± 961; Angew. Chem.
Int. Ed. Engl. 1978, 17, 91 ± 909.
[22] P. Schwab, R. H. Grubbs, J. W. Ziller, J. Am. Chem. Soc. 1996, 118,
100 ± 110.
[23] M. H. Hyun, S. C. Min, Y. J. Cho, M. S. Na, J. Liq. Chromatogr. Relat.
Technol. 1995, 18, 2527 ± 2541.
[24] S.-L. Da, W.-G. Yue, Y.-F. Wen, H.-L. Da, Z.-H. Wang, Anal. Chim.
Acta 1994, 299, 239.
Experimental Section
All experiments were carried out by means of Schlenk techniques using
degassed and dried solvents throughout. Borosilicate columns (3 Â 50 mm,
3 Â 150 mm) were surface-derivatized using bicyclo[2.2.1]hept-2-en-5-yltri-
chlorosilane. Nitrogen and ruthenium contents were determined by
elemental analysis and aqua regia decomposition followed by ICP,
respectively. Molecular weights were determined by GPC (in THF) using
a consecutive UV, refractive index (RI), and light scattering detectors.
An Intramolecular Case of Sharpless Kinetic
Resolution: Total Synthesis of Laulimalide**
Synthesis of monoliths: Solutions of A (NBE/DMN-H6/2-propanol, 25/25/
[22]
ˆ
40 wt%) and B (CH2Cl2/[Cl2Ru( CHPh)(PCy3)2], 9.6/0.4)
were com-
Johann Mulzer* and Elisabeth Öhler*
bined at 08C and the reaction mixture was transferred to a borosilicate
column prechilled to 08C. Polymerization temperature was 08C for 15 min
and room temperature for 1 h. For functionalization, the monolith was
flushed with CH2Cl2 and subsequently fed with 2 mL of a solution of 1
(51.8 mg, 0.09 mmol) and norbornene (47.1 mg, 0.5 mmol) in CH2Cl2.
Columns were closed and kept at 408C overnight. The monolith was
flushed with CH2Cl2 (1 mL), a 10% solution of ethyl vinyl ether in CH2Cl2
(2 mL), and finally CH2Cl2 (2 mL) again. 4-Dimethylaminopyridine
(10.9 mg, 0.09 mmol, dissolved in 1 mL CH2Cl2), CH2Cl2 (1 mL), and
The Sharpless asymmetric epoxidation (SAE) is an efficient
method for resolving racemic mixtures of secondary allylic
alcohols: the matched pair of substrate and reagent generates
an enantiomerically enriched epoxyalcohol, whereas the
mismatched pair remains unreacted (Scheme 1, Eq. (1)).[1]
We decided to change this intermolecular selection to an
intramolecular one when we embarked on a total synthesis of
ˆ
finally [Cl2(PCy3)2Ru( CHPh)] (10.9 mg, 0.09 mmol, dissolved in 1 mL
CH2Cl2) were pumped over the column which was then kept for 1 h. at
408C. Finally, the monolith was flushed with CH2Cl2 for a few hours at a
flow rate of 0.1 mLmin 1. IGPC data (polystyrene, Mp ˆ 274 Da, THF):
specific surface area s ˆ 25 m2 g 1, pore porosity ep ˆ 17%, intergranular
porosity ez ˆ 40%, apparent density 1app ˆ 0.37 gcm 3. Electron micros-
copy: microglobule diameter dp ˆ 1.5 Æ 0.5 mm.
[*] Prof. Dr. J. Mulzer, Dr. E. Öhler
Institut für Organische Chemie der Universität Wien
Währinger Strasse 38, 1090 Wien (Austria)
Fax : (‡43)1-4277-52190
E-mail: johann.mulzer@univie.ac.at
Received: May 22, 2001 [Z17162]
[**] We thank the Austrian Research Foundation (FWF grant P13941-
CHE) for financial support, Ing. Sabine Schneider for HPLC
separations, and Prof. Hermann Kalchhauser and Dr. Hanspeter
Kählig for NMR measurements.
[1] A. Fürstner, Angew. Chem. 2000, 112, 3140 ± 3172; Angew. Chem. Int.
Ed. 2000, 39, 3012 ± 3043.
[2] R. H. Grubbs, S. Chang, Tetrahedron 1998, 54, 4413 ± 4450.
Supporting information for this article is available on the WWW under
http://www.angewandte.com or from the author.
3842
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4020-3842 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 20

COMMUNICATIONS
(Scheme 3) and 27 (Scheme 5) by means of a Julia ± Kocienski
olefination.[12, 13] We decided to use MOM to protect the C20
and C15 hydroxy groups, and orthogonal protecting groups
for the C3 and C19 hydroxy groups (TBS and PMB,
respectively). Furthermore, we planned to synthesize alde-
hyde 16 and sulfone 27 from inexpensive chiral carbon pool
compounds such as d-mannitol,
R2
O
R2
R1
R1
matched
(+)-DIPT
tBuOOH
Ti(OiPr)4
OH
OH
OH
OH
(1)
R1
mismatched
R2
R1
R2
OH
OH
20
H
22
(S)-malic acid, and d-glucose.
H
22
O
15
O
15
17
O
O
Eq. ( 1 )
?
27
20
The synthesis of the C3 ± C16
aldehyde 16 started from the
known a,b-unsaturated lactone 4
(Scheme 2), which was readily
available from tri-O-acetyl-d-glu-
cal in four steps.[14] After conver-
sion into the TBS derivative, addi-
tion of dimethyl copper lithium
smoothly provided the desired
9,11-trans-disubstituted lactone 5
as a single diastereomer.[15] The
formation of epoxide 7 required an
inversion of the configuration at
C9. Therefore, lactone 5 was re-
duced to the 1,5-diol, and the
13
13
O
O
OH
H
Me
O
OH
H
Me
H
H
Me
O
Me
O
9
3
9
3
deoxy-laulimalide 1
laulimalide 2
0.01 N HCl
H
OH
17
O
O
20
16
OH
O
Me
O
Me
H
H
O
9
3
isolaulimalide 3
Scheme 1. The SAE route to laulimalide 2.
SPh
13
O
O
SPh
13
13
a, b
c, d
OH
e
O
O
9
the antitumor macrolide laulimalide (2). The last step of this
synthesis was envisaged as the epoxidation of 16,17-deoxy-
laulimalide (1), in which the C15 ± C17 section represents a
matched and the C20 ± C22 section a mismatched SAE
situation with respect to Equation (1). This implies that 1
would be converted into 2 without the protection of the allylic
alcohols. This strategy would be highly advantageous in view
of the reported tendency of 2 to undergo cyclization to
isolaulimalide (3), even under mildly acidic (and probably
also basic) conditions,[2] which would place high demands on
the removal of a protecting group from the C20 hydroxy
group in the presence of the 16,17-epoxide.
9
9
O
9
Me
Me
Me
OTBS
OTBS
OH
6
7
4
5
SPh
Me
SO2Ph
Me
g, h
13
H
O
O
9
f
H
EtO
O
9
7
8
SO2Ph
Me
i-k
9
l
The interest in a total synthesis of 2 has been kindled by the
striking success of paclitaxel[3] as a novel drug against
previously incurable tumors. However, in view of several
problems associated with the clinical application of paclitaxel,
a great deal of effort has been focused on the search for
potential successors with the same mode of microtubule-
stabilizing antitumor action, but with better bioavailability
and higher activity against multidrug-resistant tumor cells.
Among recent advances have been epothilone B and its
derivatives,[4] discodermolide[5] and eleutherobin.[6] It was
discovered quite recently that 2, a metabolite from various
marine sponges,[2, 7] also shows microtubule stabilization in
eukaryotic cells and is distinguished by an unusually high
antitumor activity against multidrug-resistant cell lines.[8] To
date, only one total synthesis of 2 has been published,[9] along
with several approaches to major fragments.[10]
H
EtO
O
9
10
Scheme 2. Synthesis of 9. Reagents and conditions: a) TBSCl, imidazole,
DMAP, 58C to RT, 16 h, 89%; b) Me2CuLi (2 equiv), Et2O, 0 8C, 1 h, 84%;
c) LiBH4 (2.4 equiv), THF, 08C to RT, 17 h, 98%; d) nBu3P (2 equiv),
(PhS)2 (1.5 equiv), 08C to RT, 17 h, 83%; e) 1. MsCl, Et3N, CH2Cl2, 108C
to RT, 1 h; 2. TBAF, THF, 08C, 1 h, then aq. NaOH (15%), 08C, 1 h, 89%;
f) 1. PhSO2(CH2)2C(OMe)3 (3 equiv), nBuLi, THF, 788C, 30 min, then
BF3 ´ Et2O (3 equiv) and 7, 788C, 1.5 h; 2. TFA/CH2Cl2 (9:1), RT, 3 h;
3) DBU (3.4 equiv), CH2Cl2, 08C, 30 min, 88%; g) MMPP, EtOH, 108C,
1.5 h, 89%; h) 1. DIBAL, (1.5 equiv), CH2Cl2, 788C; 2) EtOH, PPTS, RT,
ˆ
16 h, 93%; i) MMPP, EtOH, RT, 2 h, 96%; j) CH2 CHMgBr (4 equiv), CuI
ˆ
(0.1 equiv), THF,
(20 equiv), PPTS,
408C, 30 min, 96%; k) CH2 CH CH(OEt)2
toluene,
35
to
408C,
80 mbar,
86%;
ˆ
l) 1. [(C6H11)3P]2Cl2Ru CHPh (5 mol%), CH2Cl2, 408C; 2. EtOH, PPTS,
RT, 16 h, 90%. TBS ˆ tert-butyldimethylsilyl, DMAP ˆ 4-dimethylamino-
pyridine, Ms ˆ methanesulfonyl, TBAF ˆ tetrabutylammonium fluoride,
TFA ˆ trifluoroacetic acid, DBU ˆ 1,8-diazabicyclo(5,4,0)undec-7-ene,
MMPP ˆ magnesium salt of monoperoxyphthalic acid, DIBAL ˆ diiso-
butylaluminium hydride, PPTS ˆ pyridinium p-toluenesulfonate.
Retrosynthetically, we envisaged a Still ± Gennari macro-
cyclization[11] of phosphonate aldehyde 29 (Scheme 6), which
was expected to give 2,3-olefin with high Z selectivity. The E
16,17 olefin bond could be generated from fragments 16
Angew. Chem. Int. Ed. 2001, 40, No. 20
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4020-3843 $ 17.50+.50/0
3843

COMMUNICATIONS
primary alcohol at C13 was selective-
ly transformed into the correspond-
ing phenyl sulfide 6.[16] Mesylation of
the C9 hydroxy group, desilylation,
and ring closure with sodium hydrox-
ide delivered epoxide 7. Elaboration
into the dihydropyran 9 was accom-
plished by means of ring-closing
metathesis (RCM)[17] of diolefin
10[18] as shown. A slightly higher
overall yield was achieved by using
Ghosezꢁs one-pot lactonization[19] of
7 to give 9 via lactone 8 in 82%
overall yield. The C3 ± C4 unit
was stereoselectively introduced
(Scheme 3) with vinyl tert-butyldi-
methylsilyl ether[20] to form aldehyde
11, which was reduced with sodium
borohydride and the resulting alco-
hol protected with a TBS group to
give the C3 ± C13 fragment 12. Com-
pound 12 was converted into a 1:1
mixture of the C13-epimeric sulfones
13 by deprotonation and subsequent
treatment with (S)-p-methoxybenzyl
glycidyl ether.[21] The methylene
SO2Ph
Me
15
13
H
SO2Ph
Me
SO2Ph
Me
13
13
PMBO
OH
H
H
d
a
b, c
H
H
H
O
3
O
O
9
3
3
9
9
9
O
TBSO
TBSO
13
11
12
H
15
13
15
15
13
H
SO2Ph
13
O
PMBO
PMBO
MOMO
H
Me
f
MOMO
g, h
MOMO
e
Me
Me
H
H
H
H
O
O
O
3
3
3
9
9
9
TBSO
TBSO
TBSO
14
15
16
ˆ
Scheme 3. Synthesis of 16. Reagents and conditions: a) 1. CH2 CHOTBS, (2 equiv), montmorillonite K10
clay, CH2Cl2, 0 ± 58C, 45 min; 2. montmorillonite K10 clay, wet CH2Cl2, RT, 1.5 h; b) NaBH4 (1.2 equiv),
MeOH, 08C, 20 min, 86% over two steps; c) TBSCl, imidazole, DMAP, CH2Cl2, 08C to RT, 16 h, 98%;
d) 12 (1.25 equiv), nBuLi, THF, 788C, 25 min, then BF3 ´ Et2O, then (S)-O-PMB-glycidol, 788C, 2.5 h,
86%; e) MOMCl (10 equiv), EtNiPr2 (20 equiv), CH2Cl2, 08C to RT, 24 h, 94%; f) nBuLi, THF/HMPA
(10:1), 788C, 30 min, then ICH2MgCl (5 equiv), 788C to 258C, 4 h, 75%; g) DDQ (1.5 equiv),
CH2Cl2/pH 7 buffer (20:1), RT, 2.5 h, 81%; h) (COCl)2 (1.5 equiv), DMSO (2.6 equiv), CH2Cl2, 788C,
3 min, then add alcohol, 788C, 30 min, EtNiPr2 (5.2 equiv), 788C to RT, 45 min, 93%. MOMˆ
methoxymethyl, HMPA ˆ hexamethyl phosphoramide, PMB ˆ para-methoxybenzyl, DDQ ˆ 2,3-di-
chloro-5,6-dicyano-1,4-benzoquinone.
group was introduced at C13 of MOM ether 14 by means of
a Julia methylenation[22] to give 15, which was transformed
into aldehyde 16 by means of a two-step operation.
Me
O
Me
Me
O
Me
O
O
ref. [24]
a, b
Me
Fragment 27 was prepared by means of an E-selective D-mannitol
olefination of aldehyde 20 with phosphonate 23 and subse-
quent syn-selective reduction of the C20 carbonyl group
(Scheme 5). The synthesis of 20 from (R)-glycidol by means of
RCM has been reported independently by Ghosh[10c] and our
group.[10g] In a novel approach (Scheme 4) we applied a
23
Me
23
O
O
O
O
17
18
Me
Me
O
O
O
d, e
23
O
Me
bidirectional[23] RCM to tetraolefin 18, which was readily
c
H
23
Me
Me
23
prepared from the known d-mannitol-derived bis-epoxide 17
in two steps.[24] No RCM products were formed across the
central acetonide ring,[25] which served as a barrier against
cross-over metathesis.[26]
O
O
20
19
ˆ
Scheme 4. Synthesis of 20. Reagents and conditions: a) CH2 CMeMgBr
(4 equiv), CuI (0.4 equiv), THF, 408C, 2 h, 92 %; b) 1. KOtBu (3.5 equiv),
THF, 458C, 1.5 h, 2. CH2 CHCH2Br (4 equiv), RT, 17 h, 89%;
c) [(C6H11)3P]2Cl2Ru CHPh (3 mol%), CH2Cl2, RT, 5 h, 83%; d) TFA/
CH2Cl2 (4:1), 08C, 3.5 h, 93%; e) HIO4 (1.5 equiv), Et2O, 0 8C to RT, 2 h,
The synthesis of b-oxophosphonate 23 (Scheme 5) started
from the known butyrolactone 21,[27] which was easily
obtained from natural (S)-malic acid. Treatment of 21 with
the lithium salt of diethyl methanephosphonate and subse-
quent deprotonation with one equivalent of lithium diisopro-
pylamide[28] provided enolate 22, which was silylated to give
23 after hydrolytic workup. Olefination[29] with aldehyde 20
afforded enone 24 stereoselectively and in high yield. Luche
reduction[30] at 958C produced a 7.8:1 C20-epimeric mixture
in favor of the desired epimer syn-25. After separation by
HPLC (high-pressure liquid chromatography), the anti epi-
mer was recycled by Parikh ± Doering oxidation to give 24.[31]
Alcohol syn-25 was converted into alcohol 26, which was
treated with 1-phenyl-1H-tetrazole-5-thiol under Mitsunobu
conditions.[32] Oxidation of the resulting thioether[33] furnished
the crystalline sulfone 27.[34]
ˆ
ˆ
95%.
E/Z mixture from which (E)-28 was isolated by chromatog-
raphy. Deprotection of the C19 hydroxy group and acylation
with bis(2,2,2-trifluoroethyl)phosphonoacetyl chloride,[35] fol-
lowed by 3-O-desilylation and Dess ± Martin oxidation[36] gave
aldehyde 29, which underwent cyclization under Still con-
ditions[11] to furnish a 1.8:1 E/Z mixture[9a] of the macro-
lactones 30. Separation of (E)- and (Z)-30 by chromatogra-
phy, followed by deprotection with dimethylboron bromide[37]
generated the hitherto unknown deoxylaulimalides 1 and 31,
respectively. Exposure of 1 to SAE[38] with (‡)-diisopropyl
tartrate (DIPT) provided laulimalide (2).[39] Careful HPLC
analysis revealed no other products, so that our initial strategy
was successful. Admittedly, the regiocontrol of the epoxida-
For the completion of the synthesis (Schemes 6 and 7),
sulfone 27 and aldehyde 16 were connected by using a one-pot
Julia ± Kocienski olefination[12] to give olefin 28 as an 11.4:1
3844
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4020-3844 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 20

COMMUNICATIONS
tion could also be the result of a
substrate effect. Kinetic resolu-
tion can only be proved unam-
biguously by the ( )-DIPT-
mediated reaction of 1, which
has not yet been carried out
owing to insufficient material.
In conclusion we have pre-
sented a total synthesis of 2 (the
longest linear sequence has 24
steps) and two deoxy derivatives
(1 and 31) which is convergent
and stereocontrolled except for
the ring-closing Still ± Gennari
olefination. Work is underway
in our laboratory to substitute
this step by novel Z-specific
ring-closing C C connections
(for an alternative, see ref. [9b]).
Nevertheless, the modular char-
acter of the approach allows the
design of a variety of suitable
derivatives.
OPMB
OLi
ref. [27]
a
19
OLi
19
O
(S)-malic acid
EtO
EtO
20
O
OPMB
P
O
21
22
OH
20
O
O
H
H
17
19
O
OTES
O
OTES
19
OTES
b
c
23
OPMB
EtO
EtO
OPMB
OPMB
P
O
Me
Me
syn-25
23
24
OMOM
20
O
OMOM
O
N
Ph
H
H
S
O
O
OH
N
27
f, g
d, e
17
N
OPMB
PMBO
N
Me
Me
26
27
Scheme 5. Synthesis of 27. Reagents and conditions: a) MeP(O)(OEt)2 (1.0 equiv), nBuLi (1.0 equiv), THF,
788C, 15 min, then 21, 788C, 30 min, then LDA (1.0 equiv), 788C, 30 min, then TESCl (2.0 equiv), 788C
to RT, 16 h, 83%; b) LiCl (2.5 equiv), Et3N (2.5 equiv), THF, 08C, 10 min, then 20, 08C, 2 h, 86%; c) 1. CeCl3 ´
7H2O (1.5 equiv), NaBH4 (1.2 equiv), MeOH, 958C, 75 min; 2. HPLC separation (syn-25: 85%, anti-25:
11%); d) MOMCl (5 equiv), EtNiPr2 (10 equiv), CH2Cl2, 08C to RT, 24 h; e) TBAF, THF, RT, 15 min, 93%
over two steps; f) PPh3 (1.5 equiv), 1-phenyl-1H-tetrazole-5-thiol (1.5 equiv), DEAD (1.8 equiv), THF, 08C,
30 min, 90%; g) (NH4)6Mo7O24 ´ 4H2O, H2O2 (30%), 08C to RT, 74%. LDA ˆ lithium diisopropylamide,
TES ˆ triethylsilyl, DEAD ˆ diethyl azodicarboxylate.
Received: June 11, 2001 [Z17267]
OMOM
H
16
O
20
b-e
a
MOMO
H
Me
PMBO
(E)-28
27
H
Scheme 6. Completion of the synthesis of 1 and 2.
TBSO
O
Me
a) KHDMS, DME,
608C, 25 min, then 16
9
3
(1.15 equiv), 608C to RT, 4 h (57% (E)-28 ‡
5% (Z)-28); b) DDQ, CH2Cl2/pH 7 buffer (10:1),
RT, 50 min, 87%; c) (CF3CH2O)2P(O)CH2COCl
(3.5 equiv), DMAP (3.5 equiv), CH2Cl2, 788C,
10 min, then pyridine, 788C to RT, 45 min, 91%;
d) AcOH/H2O/THF (3:1:1), 3.5 h, RT, 89%;
e) Dess ± Martin periodinane (2.5 equiv), wet
CH2Cl2, RT, 30 min, 96%; f) 29 (10 3 m in THF),
28, E/Z = 11.4:1
OR
20
OMOM
16
H
16
H
O
O
20
OR
H
Me
MOMO
H
Me
O
O
f
O
[18]crown-6 (6 equiv),
788C, then KHDMS
O
H
H
O
O
Me
Me
O
(0.95 equiv), 788C, 50 min, 80%; g) 1. Me2BBr
(6 equiv), CH2Cl2, 788C, 25 min; 2. aq. NaH-
CO3/THF (1:2), 80%; h) (‡)-DIPT (1.2 equiv),
Ti(OiPr)4 (1.0 equiv), tBuOOH (1.34 equiv), 4-Š
molecular sieves, 208C, 2 h: 70% (2) and 30%
(1; recovered). KHDMS ˆ potassium 1,1,1,3,3,3-
O
2
9
9
3
3
P
CF3CH2O
OCH2CF3
30, R = MOM, E/Z = 1.8:1
29
g
h
hexamethyldisilazane,
ethane.
DME ˆ 1,2-dimethoxy-
(Z)-30
1
2
[1] B. E. Rossiter in Asymmetric Synthesis, Vol. 5, Chiral Catalysis (Ed.:
J. D. Morrison), Academic Press, Orlando, 1985, pp. 193 ± 246; M. G.
Finn, K. B. Sharpless in Asymmetric Synthesis, Vol. 5, Chiral Catalysis
(Ed.: J. D. Morrison), Academic Press, Orlando, 1985, pp. 247 ± 308.
[2] D. G. Corley, R. Herb, R. E. Moore, P. J. Scheuer, V. J. Paul, J. Org.
Chem. 1988, 53, 3644 ± 3646.
OH
H
16
O
20
a
Me
OH
O
E-30
H
H
Me
O
O
9
[3] For a review, see: K. C. Nicolaou, W.-M. Dai, R. K. Guy, Angew.
Chem. 1994, 106, 38 ± 69; Angew. Chem. Int. Ed. Engl. 1994, 33, 15 ± 44.
[4] For a review, see: K. C. Nicolaou, F. Roschangar, D. Vourloumis,
Angew. Chem. 1998, 110, 2120 ± 2153; Angew. Chem. Int. Ed. 1998, 37,
2014 ± 2045; J. Mulzer, Monatsh. Chem. 2000, 131, 205 ± 238.
31
Scheme 7. Synthesis of 31: a) 1. Me2BBr (6 equiv), CH2Cl2,
788C,
25 min; 2. aq. NaHCO3/THF (1:2), 84%.
Angew. Chem. Int. Ed. 2001, 40, No. 20
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4020-3845 $ 17.50+.50/0
3845

COMMUNICATIONS
[5] S. P. Gunasekera, M. Gunasekera, R. E. Longley, G. K. Schulte, J. Org.
Chem. 1990, 55, 4912 ± 4915. The original structure of discodermolide
was corrected: S. P. Gunasekera, M. Gunasekera, R. E. Longley, G. K.
Schulte, J. Org. Chem. 1991, 56, 1346.
[6] T. Lindel, P. R. Jensen, W. Fenical, B. H. Long, A. M. Casazza, J.
Carboni, C. R. Fairchild, J. Am. Chem. Soc. 1997, 119, 8744 ±
8745.
[27] J. Mulzer, A. Mantoulidis, E. Öhler, J. Org. Chem. 2000, 65, 7456 ±
7467; see also D. B. Collum, J. H. McDonald, W. C. Still, J. Am. Chem.
Soc. 1980, 102, 2118 ± 2120.
[28] For a related protocol, see: K. Ditrich, R. W. Hoffmann, Tetrahedron
Lett. 1985, 26, 6325 ± 6328.
[29] M. A. Blanchette, W. Choy, J. T. Davis, A. P. Essenfeld, S. Masamune,
W. R. Roush, T. Sakai, Tetrahedron Lett. 1984, 25, 2183 ± 2186.
[30] J.-L. Luche, J. Am. Chem. Soc. 1978, 100, 2226 ± 2227.
[31] J. R. Parikh, W. von E. Doering, J. Am. Chem. Soc. 1967, 89, 5505 ±
5507.
Á
Ä
[7] a) Ref. [2]; b) E. Quinoa, Y. Kakou, P. Crews, J. Org. Chem. 1988, 53,
3642 ± 3644; c) C. W. Jefford, G. Bernardinelli, J.-I. Tanaka, T. Higa,
Tetrahedron Lett. 1996, 37, 159 ± 162.
[8] S. L. Mooberry, G. Tien, A. H. Hernandez, A. Plubrukarn, B. S.
Davidson, Cancer Res. 1999, 59, 653 ± 660.
[32] O. Mitsunobu, Synthesis 1981, 1 ± 28.
[33] H. S. Schultz, H. B. Freyermuth, S. R. Buc, J. Org. Chem. 1963, 28,
1140 ± 1142.
[9] a) A. K. Ghosh, Y. Wang, J. Am. Chem. Soc. 2000, 122, 11027 ± 11028;
b) for a modification, see: A. K. Ghosh, Y. Wang, Tetrahedron Lett.
2001, 42, 3399 ± 3401; c) another synthesis was recently reported by I.
Paterson, C. De Savi, M. Tudge, Org. Lett., in press. Their manuscript
was submitted one month after ours.
[10] a) A. K. Ghosh, P. Mathivanan, J. Cappiello, Tetrahedron Lett. 1997,
38, 2427 ± 2430; b) A. K Ghosh, Y. Wang, Tetrahedron Lett. 2000, 41,
2319 ± 2322; c) A. K. Ghosh, Y. Wang, Tetrahedron Lett. 2000, 41,
4705 ± 4708; d) A. Shimizu, S. Nishiyama, Tetrahedron Lett. 1997, 38,
6011 ± 6014; e) A. Shimizu, S. Nishiyama, Synlett 1998, 1209 ± 1210;
f) J. Mulzer, M. Hanbauer, Tetrahedron Lett. 2000, 41, 33 ± 36; g) E. K.
Dorling, E. Öhler, J. Mulzer, Tetrahedron Lett. 2000, 41, 6323 ± 6326;
h) E. K. Dorling, E. Öhler, A. Mantoulidis, J. Mulzer, Synlett 2001,
1105 ± 1108; i) G. T. Nadolski, B. S. Davidson, Tetrahedron Lett. 2001,
42, 797 ± 800; j) B. T. Messenger, B. S. Davidson, Tetrahedron Lett.
2001, 42, 801 ± 804; k) I. Paterson, C. De Savi, M. Tudge, Org. Lett.
2001, 3, 213 ± 216.
[34] Prolonged reaction leads to partial epoxidation of the double bonds.
[35] This compound was obtained from commercially available methyl
ester by means of enzymatic hydrolysis (PLE, Fluka 46058) and
subsequent reaction of the acid with oxalyl chloride. We thank Mag.
M. Hanbauer for providing us with the phosphonoacetic acid.
[36] D. B. Dess, J. C. Martin, J. Am. Chem. Soc. 1991, 113, 7277 ± 7287.
[37] Y. Guindon, C. Yoakim, H. E. Morton, J. Org. Chem. 1984, 49, 3912 ±
3920.
[38] Cf. ref. [9c].
1
[39] The H (600 MHz) and 13C (150 MHz) NMR spectra of the synthetic
material are perfectly superimposable with the reported spectra of the
natural compound 2.[7c]
[11] W. C. Still, C. Gennari, Tetrahedron Lett. 1983, 24, 4405 ± 4408.
[12] P. R. Blakemore, W. J. Cole, P. J. Kocienski, A. Morley, Synlett 1998,
26 ± 28.
[13] This strategy has been one of several options that we have pursued so
far.[10f±h] A related strategy has been executed independently and
concurrently by Ghosh.[9a]
[14] E. J. Corey, S. G. Pyne, W.-g. Su, Tetrahedron Lett. 1983, 24, 4883 ±
4886; B. D. Roth, W. H. Roark, Tetrahedron Lett. 1988, 29, 1255 ±
1258; see also F. W. Lichtenthaler, S. Rönninger, P. Jarglis, Liebigs
Ann. Chem. 1988, 1153 ± 1161.
[15] cf. S. Takano, Y. Shimazaki, Y. Iwabucki, K. Ogasawara, Tetrahedron
Lett. 1990, 31, 3619 ± 3622.
2
Methylenetriimidosulfate H2CS(NtBu)3
The First Dianionic Sulfur(vi) Ylide**
Ð
Bernhard Walfort and Dietmar Stalke*
[16] I. Nakagawa, K. Abi, T. Hata, J. Chem. Soc. Perkin Trans. 1 1983,
1315 ± 1318.
[17] For recent reviews on RCM, see: A. Fürstner, Angew. Chem. 2000,
112, 3140 ± 3172; Angew. Chem. Int. Ed. 2000, 39, 3012 ± 3043; R. H.
Grubbs, S. Chang, Tetrahedron 1998, 54, 4413 ± 4450.
[18] cf. M. T. Crimmins, B. W. King, J. Am. Chem. Soc. 1998, 120, 9084 ±
9085, and ref. [10b,f,i].
[19] J. C. Carretero, L. Ghosez, Tetrahedron Lett. 1988, 29, 2059 ± 2062.
[20] K. Toshima, N. Miyamoto, G. Matsuo, M. Nakata, S. Matsumura,
Chem. Commun. 1996, 1379 ± 1380, and ref. [10f].
[21] A. B. Smith III, L. Zhuang, C. S. Brook, A. M. Boldi, M. D. McBriar,
W. H. Moser, N. Murase, K. Nakayama, P. R. Verhoest, Q. Lin,
Tetrahedron Lett. 1997, 38, 8667 ± 8670; K. C. Nicolaou, Y. Li, B.
Weyershausen, H.-x. Wei, Chem. Commun. 2000, 307 ± 308.
[22] C. De Lima, M. Julia, J.-N. Verpeaux, Synlett 1992, 133 ± 134. For a
recent application in the synthesis of spongistatin, see: A. B. Smith III,
V. A. Doughty, Q. Lin, L. Zhuang, M. D. McBriar, A. M. Boldi, W. H.
Moser, N. Murase, K. Nakayama, M. Sobukawa, Angew. Chem. 2001,
113, 197 ± 201; Angew. Chem. Int. Ed. 2001, 40, 191 ± 195; A. B.
Smith III, Q. Lin, V. A. Doughty, L. Zhuang, M. D. McBriar, Y. K.
Kerns, C. S. Brook, N. Murase, K. Nakayama, Angew. Chem. 2001,
113, 202 ± 205; Angew. Chem. Int. Ed. 2001, 40, 196 ± 199. We thank
Prof. A. B. Smith III for sending us an excellent procedure.
[23] C. S. Poss, S. L. Schreiber, Acc. Chem. Res. 1994, 27, 9. Review: S. R.
Magnusson, Tetrahedron 1995, 51, 2167 ± 2213.
Isoelectronic replacement of the oxygen atoms in simple
p-block element oxoanions by an NR imido group is currently
a flourishing area of main group chemistry.[1] These new
species are soluble in nonpolar organic solvents because they
form contact-ion pairs, whose periphery consists of lipophilic
substituents. In contrast, the simple oxoanions form infinite
solid-state lattices as a result of the multiple oxygen ± metal
cation contacts. We were particularly interested in the
polyimidosulfur anions because of the rich redox chemistry.[2]
2
[3]
Triimidosulfite S(NR)3
,
which is analoguous to sulfite
2
SO3 , can be radically oxidized to S(NR)3 (Scheme 1).[4] The
2
cap-shaped S(NR)3 ion is the first tripodal coordinating
2
dianion.[5] Tetraimidosulfate S(NR)4 , which is analogous to
2
[6]
sulfate SO4
,
gives soluble monomeric metal complexes.[7]
[*] Prof. Dr. D. Stalke, Dipl.-Chem. B. Walfort
Institut für Anorganische Chemie der Universität Würzburg
Am Hubland, 97074 Würzburg (Germany)
Fax : (‡49)931-888-4619
Â
[24] Y. Le Merrer, A. Dureault, C. Greck, D. Micas-Languin, C. Gravier,
E-mail: dstalke@chemie.uni-wuerzburg.de
J.-C. Depezay, Heterocycles 1987, 25, 541 ± 548.
[25] A. Ahmed, E. Öhler, J. Mulzer, Synthesis, in press.
[26] The facile formation of medium sized rings by RCM is known; for a
review, see: M. E. Maier, Angew. Chem. 2000, 112, 2153 ± 2157;
Angew. Chem. Int. Ed. 2000, 39, 2073 ± 2077.
[**] This work was supported by the Deutsche Forschungsgemeinschaft
and the Fonds der Chemischen Industrie. Support from Bruker
Nonius, Karlsruhe, and Chemetall, Frankfurt am Main, is kindly
acknowledged.
3846
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4020-3846 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 20