A "Chiral Aldehyde" Equivalent as a Building Block Towards Biologically Active Targets

Article abstract of DOI:10.1002/chem.200305634

Chiral γ-aryloxybutenolides, readily accessible through dynamic kinetic asymmetric transformation (DYKAT) of racemic acyloxybutenolides, were utilized as "chiral aldehyde" building blocks for intermolecular cycloadditions and Michael reactions. Unprecedented selectivity in trimethylenemethane cycloadditions with this building block allowed an efficient synthesis of a novel metabotropic glutamate receptor 1 antagonist in development by the Bayer corporation. These studies further inspired work that culminated in the total synthesis of (+)-brefeldin A, a natural product with a range of significant biological properties. All of the stereochemistry in this target molecule was derived from two palladium-catalyzed asymmetric allylic alkylation reactions. The trans-alkenes were synthesized by a Julia olefination and a ruthenium-catalyzed trans-hydrosilylation-protodesilylation protocol. The route to (+)-brefeldin A lends itself to analogue syntheses and was completed in 18 steps in 6% overall yield.

Full text of DOI:10.1002/chem.200305634

FULL PAPER
A ™Chiral Aldehyde∫ Equivalent as a Building Block Towards Biologically
Active Targets
Barry M. Trost* and Matthew L. Crawley[a]
Abstract: Chiral g-aryloxybutenolides,
readily accessible through dynamic ki-
mate receptor 1antagonist in develop-
ment by the Bayer corporation. These
studies further inspired work that cul-
minated in the total synthesis of (+)-
brefeldin A, a natural product with a
range of significant biological proper-
ties. All of the stereochemistry in this
target molecule was derived from two
palladium-catalyzed asymmetric allylic
alkylation reactions. The trans-alkenes
were synthesized by a Julia olefination
and a ruthenium-catalyzed trans-hydro-
netic
asymmetric
transformation
(DYKAT) of racemic acyloxybuteno-
lides, were utilized as ™chiral aldehyde∫
building blocks for intermolecular cy-
cloadditions and Michael reactions.
Unprecedented selectivity in trimethyl-
enemethane cycloadditions with this
building block allowed an efficient syn-
thesis of a novel metabotropic gluta-
silylation-protodesilylation
protocol.
The route to (+)-brefeldin A lends
itself to analogue syntheses and was
completed in 18 steps in 6% overall
yield.
Keywords: alkylation ¥ asymmetric
synthesis
¥ brefeldin A ¥ chiral
butenolides ¥ total synthesis
Introduction
unique opportunity to access complex molecules with multi-
ple stereogenic centers.[4b]
The art and genius of synthetic organic chemistry is often
best expressed in the enantioselective total synthesis of bio-
logically significant molecules. In this regard, ™chiral alde-
hyde∫ building blocks offer enormous potential in allowing
access to synthetic targets.[1] Previously, allylic acetals de-
rived from chiral diols and enals have been used in a
number of cases to direct stereocontrolled reactions, includ-
ing cuprate additions, cyclopropanations, and additions of
electrophiles to the adjacent olefin.[2] While a variety of ap-
proaches for setting stereogenic centers have been establish-
ed, development of catalytic enantioselective methods to
achieve this goal has been a challenging undertaking.[3] The
ability to design a synthetic strategy without the restrictions
of starting materials from a chiral pool or of the cost and
often poor availability of chiral auxiliaries has differentiated
syntheses using asymmetric catalytic processes from other
approaches.[4a] The use of ™chiral aldehyde∫ building blocks
derived from asymmetric allylic alkylation (AAA) reactions
for stereospecific reaction of double bonds provides a
One such building block, g-hydroxybutenolide derivatives,
serve as particularly useful ones. To the extent that the ster-
eochemistry of C-4 can be propagated by directing either
stepwise [Eq. (1), path a] or concerted [Eq. (1), path b] pro-
cesses, the resulting adducts are functionally equivalent to a
vicinal aldehyde carboxylic acid existing in a protected
form. The potential value of such a building block led to an
elegant series of studies by Feringa et al.[5] In the first itera-
tion, a chiral auxiliary approach with menthol provided a di-
astereomerically pure g-menthoxybutenolide.[5a] A more ap-
pealing strategy effected an enantioselective enzymatic acyl-
ation to form a g-acetoxybutenolide.[5b] However, as the au-
thors subsequently note, ™the high sensitivity of the 5-acyl
[a] Prof. B. M. Trost, Dr. M. L. Crawley
Department of Chemistry, Stanford University
Stanford, CA 94305-5080 (USA)
Fax : (+1)650-725-0002
À
substituent in subsequent C C bond formation is an impor-
tant incentive for the alkoxy furanone.∫[5c]
E-mail: bmtrost@stanford.edu
The development of dynamic kinetic asymmetric transfor-
mations (DYKATs) in AAA reactions, through which race-
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
2237
Chem. Eur. J. 2004, 10, 2237 2252
DOI: 10.1002/chem.200305634
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPER
mic material can be completely converted into enantiopure
adducts through interconversion of diastereomeric p-allyl in-
termediates, has significantly expanded the scope and utility
of the AAA process.[6] DYKATs offer a significant advant-
age over the traditional kinetic resolutions in terms of yield
and the avoidance of costly chiral auxiliary reagents. Entry
into chiral g-aryloxybutenolides was achieved by employ-
ment of a palladium-catalyzed AAA in a DYKAT with race-
mic butenolide 1, (R,R)-ligand 3, tetrabutylammonium chlo-
ride, and 2-naphthol (2) as the nucleophile to afford adduct
4 in 84% yield and 96% ee [Eq. (2)]. This approach pro-
enantioselective total synthesis of (+)-brefeldin A (6) was
completed over two decades ago,[8] most of the efforts since
have either required a strategy with over 20 steps for the
longest linear sequence or have been completed with under
2% overall yield.[9,10] Among the shortest syntheses are
those of Corey,[9e] Haynes,[9m] Suh,[9p] Romo,[9s] and Kobaya-
shi.[9k]
Results
Michael additions to 2-(2-naphthoxy)-5-oxo-2,5-dihydrofur-
an: A large variety of substituents, including ethers, amines,
and alkyl groups, have been stereoselectively introduced
through Michael additions of stabilized nucleophiles.[11] In
order to establish benchmark comparisons with other sys-
tems, nucleophiles known to add to g-substituted buteno-
lides with facial selectivity were examined. A good example
was the addition of dimethyl malonate to ent-4 [Eq. (3)].
vides access to either enantiomer in high ee simply by
switching the ligand and yields a derivative that is quite suit-
À
able for subsequent C C bond-forming events.
In this paper we explore the utility of this readily availa-
À
ble chiral building block for subsequent C C bond-forming
reactions. Special attention was given to cycloadditions,
most notably 1,3-dipolar cycloadditions including that of tri-
methylenemethane. Indeed, very simple strategies emerge
for the syntheses of the metabotropic glutamate receptor an-
tagonist represented by BAY 36-7620 and the fungal metab-
olite (+)-brefeldin A, which, among its myriad biological
functions, disrupts the Golgi complex and induces apoptosis
(see Scheme 1).
This conjugate addition afforded butyrolactone 7 in 77%
yield as a single diastereomer, comparable to the reported
addition of dimethyl malonate to (R)-(À)-5-[(1R,2S,5R)-
menthyloxy]-2(5H)-furanone (8).[12]
The trans configuration of lactone 7 was assigned, in part,
on the basis of comparisons with Feringa×s systems and, in
part, from macromodel calculations of the minimum-energy
conformation of the trans product 6, by means of an exten-
sive conformation search (5000 iterations) with an MM2
force-field in solution phase with chloroform as solvent
(Figure 1). The minimized conformation displayed a dihe-
Scheme 1. Retrosynthetic analysis of BAY 36 7620 and (+)-brefeldin A.
The biological significance of (+)-brefeldin A (6) has
been heavily studied. The unique mode of action in a variety
of therapeutic areas combined with the potential for ana-
logues makes it a particularly attractive synthetic target.[7]
The five stereogenic centers and the two trans-olefins in the
bicyclic macrolactone arrangement provide a range of op-
portunities and a host of challenges. The numerous synthetic
efforts emphasize this point. Despite the fact that the first
Figure 1. Macromodel-predicted minimum-energy conformation of lac-
tone 7.
dral angle of 82.18 that was consistent with the small cou-
pling constant of 2.4 Hz observed between the b- and g-pro-
tons of the lactone. Other low-energy conformations within
2238
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2237 2252

A ™Chiral Aldehyde∫ Equivalent
2237 2252
0.2 kcalmolÀ1 of the predicted ground state had similar dihe-
dral angles (80.2 83.58) between the adjacent protons. The
syn conformation of adduct 7 had a dihedral angle of 228
that was inconsistent with the observed small coupling con-
stant. Further details of the stereochemical assignments for
lactone 7 and other Michael addition adducts are discussed
later in this section.
Since preliminary investigations indicated that aryloxybu-
tenolide 4 could function in Michael additions as well as or
better than other chiral butenolides, we decided to investi-
gate whether other stabilized nucleophiles not previously in-
vestigated for the other butenolides would add selectively.
The addition of nitroalkanes was an evident choice. Treat-
ment of 2-nitropropane with butenolide ent-4 in the pres-
ence of one equivalent of DBU afforded lactone 9 in 93%
yield as a single diastereomer [Eq. (4)]. As in the case of
the malonate, the 2.9 Hz coupling of the vicinal hydrogens is
in accord with the depicted trans geometry.
Figure 2. Energy-minimized structure of Diels Alder adduct 11.
The Diels Alder reaction between 2,3-dimethylbutadiene
12 and butenolide ent-4 afforded anti cycloadduct 13 as a
95:5 diastereomeric mixture of the anti and syn adducts
1
[Eq. (6)]. The ratio was assigned by H NMR spectroscopy,
by comparison of the integration of the g-lactone proton sig-
nals (d=5.79 ppm singlet for anti vs d=6.08 ppm doublet,
J=2.2 Hz for syn). The observation of no coupling constant
between the b- and g-protons for adduct 13 was consistent
with Feringa×s Diels Alder adduct for the menthyloxy
system. In that case Feringa confirmed the relative stereo-
chemistry through X-ray crystallography.
Since the [4+2] cycloaddition reactions proceeded with
excellent selectivities and yields, the next logical step was to
examine [3+2] dipolar cycloaddition reactions. It was desira-
ble to ascertain whether butenolide 4 could provide results
equally impressive as observed in other butenolide systems
for these reactions, and also to see whether previously un-
tried cycloaddition reactions with butenolides could proceed
diastereoselectively.
Diastereoselective cycloaddition reactions with the buteno-
lide: Diels Alder reactions of both cyclic and acyclic dienes
with 5-menthyloxy-2(5H)-furanone have been well docu-
mented by Feringa and co-workers.[13] The diastereoselectivi-
ties were outstanding in most of the reported examples,
though the yields were sometimes moderate (44 to 77%).
To benchmark chiral butenolide 4, two Diels Alder reac-
tions were examined.
The first cycloaddition was with 1,3-cyclohexadiene 10
and butenolide ent-4 [Eq. (5)]. The reaction was carried out
One straightforward comparison was pursued in the cyclo-
addition between ethyl diazoacetate 14 and butenolide ent-4
to give cycloadduct 15 as a single diastereomer in 94% yield
[Eq. (7)]. The assignment of stereochemistry was possible
neat in a microwave and gave quantitative conversion (96%
yield after silica-gel chromatography) into a single diaster-
eomer 11. The stereochemistry of endo-adduct 11 was as-
signed by analogy from Feringa×s[12] examples and by evalua-
tion of the coupling constants of the g- and b-protons in
comparison with a macromodel MM2 force-field solution-
phase minimum-energy structure (Figure 2). The lactone g-
proton displayed a doublet at d=5.70 ppm, with a coupling
constant of 1.7 Hz consistent with the dihedral angle of 1048
predicted by the modeling calculations. This was in agree-
ment with the coupling constant of 1.5 Hz observed by Fer-
inga for the g-lactone proton in the menthyloxy adduct.
by comparison to the known menthyloxy adduct described
by Feringa,[14] who obtained the adduct in 65% yield and in
an anti/syn ratio of 91:9. The anti adduct had no coupling
between the b- and g-protons, while the syn adduct had a
7.0 Hz coupling constant. In the case of adduct 15, the g-
proton displayed a singlet at d=6.49 ppm. This chiral bute-
nolide ent-4 demonstrated both improved yield and selectivi-
ty in this cycloaddition reaction.
2239
Chem. Eur. J. 2004, 10, 2237 2252
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

B. M. Trost and M. L. Crawley
FULL PAPER
Cycloaddition reactions with diazomethane in ethereal
solution at various temperatures with or without additives
gave a crude diastereoselectivity of 4:1, to afford pyrazoline
16 in 76% yield and as a single diastereomer after chroma-
tography [Eq. (8)]. The corresponding reaction of the men-
palladium p-allyl intermediate 21. The a-silyl group subse-
quently undergoes desilylation, presumably by attack of the
acetate anion on silicon, to give dipole 22 [Eq. (10)].[17]
Dipole 22 can then add to electron-deficient olefins.
thoxy acceptor is reported to provide a selectivity of 3:1in
55% yield.[15] The stereochemistry of addition was assigned
by comparison to similar systems and by use of the Karplus
relationship, which indicates that the coupling constant be-
tween the protons should be less than 2.0 Hz for the anti ad-
dition product while the syn addition product should have a
coupling constant between 8.0 and 11.0 Hz. The observed
signal for the g-lactone proton of the major cycloadduct 16
at d=5.77 ppm appeared as a doublet, J=1.5Hz, consistent
with the assigned stereochemistry.
Treatment of acetate 20a with butenolide ent-4 in the
presence of palladium(ii) acetate with triisopropyl phosphite
as ligand in toluene at reflux afforded cyclopentanoid
adduct 23a in 93% yield and as a single diastereomer
[Eq. (11)]. Lactone 23a was a white solid stable to storage
Access to 3,4-cis-bis-functionalized pyrrolidines is possible
through diastereoselective addition of azomethine ylides to
chiral butenolides. Successful studies on this type of cycload-
dition under ultrasonic conditions have been reported.
Treatment of butenolide 4 with azomethine precursor 17 af-
forded pyrrolidine derivatives 18 and 19 in 92% combined
yield, as a 3.5:1.0 mixture of diastereomers [Eq. (9)]. The di-
at ambient temperature without the need for inert atmos-
phere.
The stereochemistry of addition was assigned as trans by
comparison of the macromodel-generated minimum-energy
conformations of the trans and cis adducts (MM2 force-field
with chloroform as solvent) with the observed coupling con-
stants for lactone 23a (Figure 3). The predicted dihedral
astereomers were separable by silica-gel chromatography
and readily identified, as in the previous cases, by examina-
tion of the coupling constants between the g- and b-lactone
protons. For the major trans adduct 18, the g-proton dis-
played a signal at d=5.92 ppm as a doublet, J=1.5 Hz,
while the in syn adduct 19 the g-proton displayed a signal at
d=6.21ppm as a doublet, J=6.6 Hz.
The success with classical [4+2] and [3+2] cycloaddition
reactions inspired the examination of trimethylenemethane
(TMM) cycloaddition reactions with butenolide 4. The ace-
tate 20 was available in three steps and 48% yield by meth-
ods previously developed in these laboratories starting from
2-methyl-2-propen-1-ol.[16] The TMM precursor 20 generates
a dipole in situ after treatment with palladium(0), to form
Figure 3. Conformations of the trans and cis TMM adducts 23a predicted
by Macromodel.
angle between the b- and g-protons on the lactone was 928
for the trans adduct and 238 for the cis adduct. The observed
signal for the g-proton was a singlet at d=5.90 ppm, consis-
tent with the trans geometry.
2240
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2237 2252

A ™Chiral Aldehyde∫ Equivalent
2237 2252
The success with the unsubstituted TMM cycloaddition
prompted an investigation of the reactions of substituted
TMM dipoles 20a e with the butenolide. These substituted
TMM precursors were synthesized by previously described
methods,[18] and the butenolide was then subjected to the
usual Pd0 catalyst system for such cycloadditions to provide
adducts 23a e (Table 1). The solvent effect on the facial se-
A concise synthesis of Bayer compound 5: The unsubstitut-
ed TMM adduct 23a proved to be a versatile building block
and provided an opportunity for efficient access to a Bayer
drug in development, BAY 36-7620 (5). This compound is
part of a novel class of metabotropic glutamate receptor 1
antagonists invented by Bayer.[19] It is a specific and potent
noncompetitive mGluR1antagonist, structurally not resem-
bling glutamate, which inhibits
>60% of mGluR1constitutive
Table 1. TMM cycloadditions to butenolide ent-4.[a]
activity at slightly under 11nm
Entry
Substrate
Conditions
Product
Yield [%]
dr (epimers)[b]
concentration.[19d]
1
2
3
4
5
6
7
20a
20a
20b
20b
20c
20d
20e
toluene, 1008C, 12 h
THF, 658C, 24 h
THF, 658C, 12 h
toluene, 1008C, 12 h
toluene, 1008C, 12 h
toluene, 1008C, 48 h
toluene, 1008C, 24 h
23a
23a
24
93
93
94
9194:6
>98:2
>98:2
5.5:1
The synthesis of BAY 36-
7620 (5) was a straightforward
process from cyclopentene 23a
(Scheme 2). Reductive removal
of the aryloxy group by use of
24
23d
23e
60
79
>98:2 (1:1)
>98:2 (4:1)
sodium
borohydride
under
[a] All reactions were run as summarized in Equation (11). [b] Diastereomeric ratios are of the crude reaction
mixture.
basic conditions, followed by re-
lectivity is noteworthy; that is, a higher diastereoselectivity
was observed in toluene than in THF even though the
former is at considerably higher temperature (Table 1, en-
tries 3 and 4).
Thus, with the unsubstituted TMM PdL2 species, reaction
proceeds via complex 22b and not 22a. Diacetate 20c did
not react under the reaction conditions, and gave recovered
starting material (Table 1, entry 5). In the case of the cyano-
substituted TMM reaction (Table 1, entries 3 and 4), the
exocyclic double bond of the initial product isomerized to
the endocyclic position under the reaction conditions. While
the epimeric ratio with respect to the methyl group in
entry 6 was 1:1, the ratio increased to 4:1 in the phenyl-sub-
stituted system (Table 1, entry 7).
As in the case of the unfunctionalized TMM adduct 23a,
the g-protons on the lactones produced singlets (the ob-
served chemical shifts were between d=5.80 and 6.20 ppm),
Figure 4. Assignment of relative stereochemistry for cycloadducts 23d
and 23e.
consistent with the trans rather than the cis adducts. The
regio- and stereochemistry of addition for the substituent on
both epimers of the methyl (23d1 and 23d2) and phenyl
(23e1 and 23e2) derivatives were established by nOe experi-
ments (Figure 4). The methyl-substituted derivatives were
readily assigned by observing that, in compound 23d1, pro-
tons Ha and Hb displayed a 2.8% nOe, while there was no
nOe between protons Ha and Hc. Conversely, adduct 23d2
showed protons Ha and Hb with no nOe, while protons Ha
and Hc displayed a 3.0% nOe.
In the case of phenyl adducts 23e1 and 23e2, isolated in a
4:1ratio, the same type of observations allowed the relative
stereochemistry to be assigned. In compound 23e1 the pro-
tons Ha and Hb displayed a 4.2% nOe, while there was no
nOe between protons Hb and Hc. Conversely, adduct 23e2
showed no nOe between protons Ha and Hb, while protons
Hb and Hc displayed a 7.2% nOe.
Scheme 2. Synthesis of Bayer drug 36 7620 (5).
Further studies, including the synthesis of the Bayer
drug 36-7620 and the total synthesis of (+)-brefeldin A (5),
further confirmed the trans nature of the cycloaddition
adduct through the conversion of TMM adducts into known
compounds.
lactonization with aqueous hydrochloric acid, afforded g-di-
hydrolactone 25 in 83% yield. Akylation of adduct 25 with
2-bromomethylnaphthalene afforded adduct 5 in 83% yield.
2241
Chem. Eur. J. 2004, 10, 2237 2252
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

B. M. Trost and M. L. Crawley
FULL PAPER
1
The H NMR data and the sign and magnitude of rotation
matched the Bayer data {[a]D =À30Æ0.1( c=1.60 in
CHCl3) vs. Bayer data [aD]=À33.0 (c=1.0 in CH2Cl2)}.
Thus, a concise, six-step synthesis of Bayer 36-7620 (5) was
completed in 44% yield from commercially available furfu-
ral, compared to the eight-step and 16% yield process re-
ported by Bayer. The use of this methodology should pro-
vide access to both enantiomers of 5 as well as a number of
other alkylation derivatives, though only one was prepared.
Scheme 4. Retrosynthetic analysis of the lower side chain 27.
razole sulfone that has shown exceptional versatility in the
one-pot Julia olefination.
Core aldehyde 26 could be formed from lactone 31 by
opening with N,O-dimethylhydroxylamine (Scheme 5). The
lactone 31 would be synthesized from cyclopentene 32, itself
Total synthesis of (+)-brefeldin A (6): There have been
many synthetic efforts towards (+)-brefeldin A (6) because
of its unique structural features. The majority of these ap-
proaches have used a convergent synthetic approach discon-
necting the molecule into a cyclopentane core fragment and
side chains. These routes have closed the macrolide by lacto-
nization, cycloaddition, or metathesis. This strategy appealed
to us, though it was important to design a synthesis that
could avoid many of the protecting and functional group
manipulations involved that have caused previous syntheses
to be somewhat lengthy.
We envisioned (+)-brefeldin A (6) coming from a core
fragment 26, a six-carbon lower side chain 27, and ethyl pro-
piolate (28) (Scheme 3). The sulfone 27 would be coupled in
Scheme 5. Retrosynthetic analysis of the cyclopentane core 26.
derived from a diastereoselective [3+2] trimethylenemeth-
ane (TMM; generated in situ from allylic acetate 20a) cyclo-
addition reaction with chiral butenolide 4. Butenolide 4 was
derived through an AAA reaction in a DYKAT as previous-
ly described, originating from furfural 33.
Synthesis of the C(11) C(16) lower side chain–a first-gen-
eration approach: The possibility of using a catalytic enan-
tioselective reaction to set the remote C(15) chiral center of
(+)-brefeldin A (5) was appealing. In a continuation of the
research of Trost and Toste on AAA reactions of allylic car-
bonates with phenols, methyl crotyl carbonate 34 was alkyl-
ated with 4-methoxyphenol under a variety of conditions
[Eq. (12), Table 2].[21] The initial results reported by Toste
were encouraging, giving 87:13 branched to linear selectivity,
though only 60% ee. Toste discovered that the enantioselec-
Scheme 3. Retrosynthetic analysis of (+)-brefeldin A (6).
a modified trans-selective Julia olefination[20] with aldehyde
26. The remaining propiolate fragment would be added by
addition to a Weinreb amide, followed by epimerization of
the proton a to the carbonyl group. Subsequent deprotec-
tion, ester saponification, macrolactonization, and global de-
protection would complete the total synthesis of 6.
The six-carbon lower side chain sulfone 27 would be de-
rived from the chiral four-
Table 2. Asymmetric allylic alkylation of 4-methoxyphenol with crotyl carbonate.[a]
carbon allylic ether 29, itself
produced from regio- and
enantioselective allylic alkyla-
a
Entry Pd2dba3
[mol%]
Ligand
[mol%]
Conditions
Yield
[%]
branched/
linear
% ee
tion of the carbonate of (E)-
crotyl alcohol (30) (Scheme 4).
The allylic ether 29 could be
transformed into the Julia pre-
cursor 27 by any number of
two-carbon homologation pro-
tocols. The aryl group of sul-
.0 11
3.0
THF, RT, 0.5
m
85
78
86
87:13
92:8
97:3
60
71
31
2
3
1.0
0.25
3.0
0.75
CH2Cl2, 08C, 0.5m
CH2Cl2, 08C, 0.5m, 30%
Bu4NCl
CH2Cl2, 08C, 0.5m
CH2Cl2, 08C, 0.5m, K2CO3
toluene, 08C, 0.1m
branched SM, as entry 4
4
5
6
7
0.25
0.25
0.25
0.25
0.75
0.75
0.75
0.75
75
75
95
92
95:5
93:7
96:4
96:4
81
81
90
32
fone 27 was envisioned as a tet- [a] All reactions run as summarized in Equation (12).
2242
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2237 2252

A ™Chiral Aldehyde∫ Equivalent
2237 2252
tivity plummeted to 31% with the addition of tetrabutylam-
monium chloride, suggesting that p s p interconversion was
unfavorable for this alkylation. Further investigation re-
vealed that toluene was the optimum solvent for this reac-
tion and that dilute conditions at 08C are important to maxi-
mize the selectivity. This reaction is an example of enantio-
selection derived from preferential ionization through enan-
tioselective coordination to one of the prochiral olefin
faces.[22] The optimized conditions gave yields over 95%
with 90% ee and 96:4 branched to linear ratio (entry 6,
Table 2). This reaction was robust and performable on
10 gram scales with no change in selectivity.
optical rotation ([a]D =+6.6Æ0.3 (c=0.70 in CH3OH))
agree well with the reported values ([a]D =++7.0Æ2.0 (c=
0.89 in CH3OH).[25] This unambiguously established that no
epimerization had occurred in the transformation to inter-
mediate 37 and confirmed the results from the AAA reac-
tion.
An alternative approach to the lower side chain: While the
initial route to sulfone 27 gave a high overall yield, it was
also somewhat lengthy. Several other ways to make the
lower side chain 27 were examined, including conversion of
allylic ether 29 into an alkyl iodide (by hydrozirconation-io-
dination of 28) as an alkylation partner. Unfortunately these
efforts were unsuccessful. A revised approach utilized a
cross-metathesis reaction between allylic ether 29 and two
equivalents of sulfone 40 [Eq. (14)]. The latter compound
The chiral allylic ether 29 was transformed into enoate 36
by hydroboration oxidation to alcohol 35 and subsequent
one-pot oxidation to the aldehyde and an in situ Wittig reac-
tion in an overall yield of 64% (Scheme 6). Only one chro-
was prepared in 86% overall yield through a Mitsunobu re-
action between a tetrazolethiol and homoallyl alcohol with
subsequent ammonium molybdate oxidation. Hydrogenation
of the olefin with palladium on carbon cleanly afforded sul-
fone 27 in two steps and 60% yield. This route reduces the
number of linear steps from 29 to only two.
Synthesis of the core–a first-generation approach: The
™chiral aldehyde∫ equivalent 4 was utilized as the key build-
ing block for the core synthesis of (+)-brefeldin A (6). The
previously described cyclopentene 23a was oxidatively
cleaved in a one-pot protocol with catalytic osmium tetrox-
ide and sodium periodate[26] to give a 92% yield of ketone
41 [Eq. (15)]. The crude adduct was sufficiently pure for use
in subsequent transformations without further purification.
Scheme 6. Completion of the C(11) C(16) lower side chain fragment 27.
matographic separation was necessary in this sequence, after
the final step. Enoate 36 was transformed in a sodium boro-
hydride 1,4-reduction catalyzed by nickel dichloride[23] fol-
lowed by reduction with lithium aluminium hydride to
afford the saturated alcohol 37 in 92% yield over two steps.
The primary alcohol 37 was
readily converted into sulfone
27 in 88% yield by Mitsunobu
coupling of 1-phenyl-5-thioltet-
razole (38) followed by oxone
oxidation.[24] The total sequence
afforded sulfone 27 in six steps
and 52% yield from 29.
The absolute stereochemistry
of the adduct 35 derived from the AAA reaction with crotyl
carbonate was established by conversion of intermediate 37
into 39 by use of cerium ammonium nitrate in acetone/water
to afford diol 39 in 87% yield [Eq. (13)]. Diol 39 is known,
and its spectral data as well as the sign and magnitude of its
Ketone 41 was chemo- and diastereoselectively reduced
(96:4 dr) to alcohol 31a by Yamamoto×s method[27]
(Scheme 7). It should be noted that the transformation was
not as trivial as initially expected and that a variety of re-
ducing agents including sodium borohydride, diisobutylalu-
minium hydride, and lithium aluminium hydride led to pref-
erential reduction of the lactone instead of the ketone.
Other reduction methods led either to the wrong chemose-
lectivity or to no diastereoselectivity. Without purification,
2243
Chem. Eur. J. 2004, 10, 2237 2252
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

B. M. Trost and M. L. Crawley
FULL PAPER
Scheme 7. First-generation synthesis of the core 26.
Scheme 8. Further efforts towards the cyclopentane core.
the alcohol 31a was silyl-protected to afford intermediate
31b in 77% yield for the two steps. By a modification of a
Merck protocol, in which twice the normal amounts of base
and amine were utilized,[28] the lactone 31b was then
opened to amide/aldehyde 26 in 84% yield with isopropyl-
magnesium chloride as base. This completed the eight-step
sequence from furfural in 38% overall yield.
borane derivative 43b resulted only in the protodeiodination
product 44. The borane 43b was formed in situ from alkene
43a. Compound 43a was derived from chiral allyl ether 29
in three steps and 61% yield (Scheme 9). The formation of
To validate the presumed stereochemistry of intermediate
26, nOe experiments were conducted. The results confirmed
the predictions. The proton Ha displayed nOes of equal
magnitudes of 3.0% to both Hb and Hc (Figure 5). This veri-
fied that the three protons were in cis positions and that no
epimerization had occurred during the opening of the lac-
tone.
Scheme 9. Synthesis of the C(5) intermediate 43.
only the undesired 44 was surprising, as the same borane
43b reacted under identical conditions with the more hin-
dered vinyl iodide substrate 45 to give an 85% yield of
trans-alkene 46 [Eq. (16)].
Figure 5. Confirmation of the relative stereochemistry for aldehyde 26.
Further elaboration of the core fragment by the planned
Julia olefination with sulfone 27 was unsuccessful. Both lithi-
um and potassium hexamethyldisiliazane were utilized as
bases in dimethoxyethane, tetrahydrofuran, and toluene as
solvents; hexamethylphosphorous amide was also utilized as
an additive. Despite a variety of changes, all efforts caused
decomposition to a complex mixture.
Several explanations for the decomposition were possible.
Firstly, the initial adduct of sulfone addition might lactonize.
Secondly, the initial aldehyde 26 might epimerize under the
basic conditions. Furthermore, the reactivity of the system
was probably altered because of the steric demands of the
system.
However, utilization of the improved cross-coupling pro-
tocol under the conditions described by Fu[31] afforded the
desired extended core fragment 47 in 68% yield [Eq. (17)].
This intermediate was stable and readily purified by silica-
gel chromatography. To confirm that no epimerization had
occurred during the Takai reaction or the Suzuki cross-cou-
pling reaction, nOe experiments were conducted on vinyl
To overcome this obstacle, a modified protocol to attach
the lower side chain was devised. Aldehyde 27 was trans-
formed into the trans-vinyl iodide 42 in 61% yield by the
Takai protocol (Scheme 8).[29] Initial attempts to cross-
couple the vinyl iodide 42 under Suzuki conditions[30] with
2244
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2237 2252

A ™Chiral Aldehyde∫ Equivalent
2237 2252
iodide 42 and amide 47 (Figure 6). For derivative 42, proton
Ha displayed nOes of equal magnitude of 4.1% to protons
Hb and Hc. In the case of amide 47, proton Ha showed a
2.4% nOe to Hb and 2.2% to Hc. These experiments vali-
dated the indicated stereochemistry.
building block is nicely demonstrated by this change of
strategy (see Scheme 10). All of the reactivity problems in
the first route can be traced to the cis configuration of the
three cyclopentane substituents. However, despite the trans-
C(5) C(9) of the natural product, the initial aldehyde could
not be epimerized, because it would give the wrong epimer
of the natural product. Instead, if the opposite enantiomer
of butenolide 4 was prepared by use of the opposite enan-
tiomer of the ligand in the DYKAT reaction, the aldehyde
could then be epimerized and used to install the C(1) C(3)
upper side chain. The amide would be the tether for the
C(11) C(16) fragment. Disconnection of (+)-brefeldin A (6)
to enoate 51 gives a fragment that could be synthesized
from amide 52 by alkyne addition/reduction to the aldehyde,
followed by protection and amide reduction. Amide 52 is
derived from butenolide ent-4.
Figure 6. Confirmation of relative stereochemistry for iodide 42 and
amide 47.
Table 3. Nucleophilic additions to a Weinreb amide.[a]
Unfortunately, after confirm-
ing the stereochemistry and
overcoming the obstacles to
obtain core 47, we discovered
the material was not suitable
for further transformations. The
desired addition of ethyl pro-
piolate resulted only in recov-
ered starting material under a
variety of conditions. Several
dozen sets of reaction condi-
tions were examined, including
reductions of the amide and ep-
Entry
Nucleophile
Additive
T [8C]/t [h]
Product
Yield [%]
84
1
0/1
50a
2
BF3¥OEt2
À78/0.5
50a
78
3
4
5
6
LiCꢀCCO2Et
LiCꢀCCO2Et
BrMgCꢀCH
BrMgCꢀCH
RT/24
À78/1
0/1
À78/0.5
starting material
50b
50c
BF3¥OEt2
BF3¥OEt2
79
85
decomposition
[a] All reactions performed in THF as summarized in Equation (18).
imerizations of the Weinreb amide a-center. Its lack of reac-
tivity, or under some conditions its decomposition, was quite
unexpected.
While the lack of reduction was surprising, it was not
clear whether Weinreb amides were susceptible to nucleo-
philic addition of propiolate anions. While numerous exam-
ples of addition with the lithium salts of alkyl- and aryl-sub-
stituted alkynes are known, there have been no reports of
ester-substituted alkynes adding to the amide. Model sub-
strate 49 was selected because of its ease of preparation and
was treated with various nucleophiles both in the presence
and in the absence of a strong acid catalyst (boron trifluo-
ride etherate) [Eq. (18)]. The results confirmed the reactivi-
ty of a Weinreb amide to the conditions under which sub-
Scheme 10. Revised retrosynthetic analysis.
Efficient synthesis of a functionalized (+)-brefeldin A core:
Starting with butenolide ent-4, the route to core 52 was iden-
tical to that in the first core synthesis, with the exception of
the epimerization in the last step. Some of the yields dif-
fered slightly, but were within 5% of those in the first-gener-
ation approach. The final epimerization step proceeded with
14:1 diastereoselectivity to give intermediate 52
(Scheme 11); the undesired isomer could be recycled. The
sequence from commercial furfural was completed in nine
steps and 32% overall yield and was readily performable on
gram scales.
strate 47 was inert (Table 3). Notably, the reaction between
Weinreb amide 49 and lithium propiolate proceeded at
À788C in the presence of boron trifluoride etherate.
Synthesis of the core–a second-generation approach: Given
this roadblock and the number of difficulties in the original
route to (+)-brefeldin A (6), a new approach was needed.
The flexibility of the butenolide 4 as a ™chiral aldehyde∫
2245
Chem. Eur. J. 2004, 10, 2237 2252
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

B. M. Trost and M. L. Crawley
FULL PAPER
Table 4. Diastereoselective alkylation of the C(4)-aldehyde.[a]
Entry Solvent(s)
1THF/HMPA 5:1
Conditions
4-(S)/4-(R) Yield [%]
À788C, 4 h
6.0:1.0
4.5:1.0
1.0:3.0
88
84
86
80
92
91
2
3
4
5
6
THF/HMPA 9:1 À788C, 4 h
THF
THF
DME
DME
À788C, 4 h
À788C, 4 h, MgBr2 1.0:3.5
À788C, 2 h 1.0:5.0
À788C, 2 h, MgBr2 1.0:6.0
[a] Reaction as depicted in Equation (19).
Scheme 11. Second-generation route to the core 52.
solvent with use of stoichiometric magnesium bromide af-
forded the opposite epimer of alcohol 53. Tentative assign-
ment of the 4S stereochemistry as the major adduct for en-
tries 1and 2 was explained in terms of Felkin Anh control.
The rationale for entries 3 6, Table 4, were explained in
terms of chelate control with the best diastereoselectivity
The relative stereochemistry of aldehyde ent-28, the all-cis
precursor to aldehyde 52, had previously been established
by nOe studies. Aldehydes 52 and ent-27 had the same mo-
lecular weight and connectivity as determined by HRMS,
and 1H NMR and 13C NMR spectroscopy. The 1H NMR
spectra were virtually identical, the chemical shifts differing
by less than 0.1ppm, with the exceptions of those of the two
protons a to the carbonyl groups. Two nOe experiments
showed a 2.7% nOe between Ha and Hb and no nOe be-
tween Ha and Hc. This confirmed the assumption that the al-
dehyde 52 had the desired stereochemistry (Figure 7).
being obtained with the better chelating Mg2+
.
While the recently reported ruthenium-catalyzed trans-hy-
drosilylation conditions had been reported on a variety of
substrates, highly oxygenated substrates had not been exam-
ined.[32] Several model compounds were subjected to the re-
ported conditions (Scheme 12, Table 5). Treatment of sub-
Figure 7. Confirmation of relative stereochemistry for amide 52.
Scheme 12. Model studies for the hydrosilylation protocol.
The introduction of the
upper side chain envisioned the
use of a new protocol for the
formation of trans-alkenes from
alkynes. To obtain the requisite
propargyl alcohol 53, diastereo-
selective addition of ethyl pro-
piolate to aldehyde 52 was ex-
amined [Eq. (19) ,Table 4]. Var-
Table 5. Model studies for the hydrosilylation protocol.[a]
Entry
R
R’
Reactant
Comments
Yield [%] (Product)
1CH
2CH(CH2)8
CH2CH(CH2)8
EtO
C(OCH2)3CMe
CO2Et
CH(OH)CH2CH3
CH(OH)CH(CH3)2
50a
50b
50d
50e
no additive
no additive
1.5 equiv CsF
1.5 equiv CsF
92 (54a)
70 (55b)
77 (55d)
96 (55e)
2
3
4
EtO
[a] Reaction depicted in Scheme 12.
iation of conditions allowed access to both epimers at C(4).
Addition of the lithium anion of ethyl propiolate to the
core 52 in a THF/HMPA (5:1) solvent mixture at À788C
(entry 1) afforded the desired alcohol 53 with 6:1diastereo-
selectivity. A change in conditions to dimethoxyethane as
strate 50a afforded adduct 54a in 92% yield as a 4:1mix-
ture of regioisomers favoring the b-silyl-substituted product
(d=7.10 ppm minor vs d=6.08 ppm major for the olefin
protons). Substrate 50b did not afford adduct 54b, but in-
stead directly formed alkene 55b in 70% yield under the re-
2246
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2237 2252

A ™Chiral Aldehyde∫ Equivalent
2237 2252
action conditions. Substrates
50d and 50e, which approxi-
mate alkyne 53a in their substi-
tution patterns (Table 5, en-
tries 3 and 4), gave 77% and
96% yields of products 55d and
55e, respectively, when cesium
fluoride in ethanol was added
after the hydrosilylation. Thus,
ynoates with g-keto substitution
protodesilylate directly, whereas
those with g-alkoxy substitution
allow for isolation of the b-vi-
nylsilanes.
The presumed (4S)-config-
ured alkyne 53a was subjected
to the same ruthenium-cata-
lyzed conditions. The crude
Figure 8. Assignment of stereochemistry through compounds 59 and 60.
mixture was treated with cesium fluoride in ethanol to
afford alkene 56a in 92% yield [Eq. (20)].
chemical shift exhibited by the olefin protons between the
two compounds.
Chemical shift differences between the two spectra fit the
pattern established earlier for the methylmandelate deriva-
tives. Shielding of the a-proton of the olefin in the case of
59 versus 60 indicated a difference of 0.4 ppm. A similar
magnitude but opposite chemical shift difference was ob-
served for the cyclopentane methine proton, whereby a dif-
ference of 0.3 ppm was observed for 59 versus 60 (d=2.71
vs 2.42 ppm). This showed shielding of the protons in 60
(Figure 8). These variations are readily explained by the
model and confirmed the major adduct was correctly as-
signed as 4S.
After protection of the free alcohol 56 as the tert-butyldi-
methylsilyl ether, the amide was chemoselectively reduced
to aldehyde 51 in the presence of the enoate, by use of
DIBAL-H as the reducing agent at À788C [Eq. (21)]. This
completed the synthesis of the highly functionalized cyclo-
pentane core in 13 steps and 18% yield from furfural 33.
Derivative 56a gave an opportunity for confirmation of
the absolute stereochemistry of addition of the ethyl propio-
late. Coupling of 56a with (S)-O-methylmandelic acid (57)
and (R)-O-methylmandelic acid (58) by a method to deter-
mine absolute stereochemistry developed in these laborato-
ries[33] gave derivatives 59 and 60 in 94 95% yield
(Scheme 13). The highlighted segments of each 1H NMR
spectra (Figure 8) illustrate the profound difference in
Completion of the total synthesis: With key aldehyde 61
and sulfone 28 in hand, the critical Julia olefination was ex-
amined. The optimum conditions involved use of potassium
hexamethyldisilazide as the base instead of lithium or
sodium hexamethyldisilazide in dimethoxyethane solvent at
À788C. This gave a 12:1 E/Z selectivity for the formation of
alkene 61 [Eq. (22)]. Use of THF as solvent gave a signifi-
cantly reduced yield (35 vs 81%). This coupling did not
have any of the problems encountered in the first-genera-
tion synthesis. The E/Z ratio was assigned by proton NMR
Scheme 13. Establishment of the relative stereochemistry of the C4 ster-
eogenic center.
2247
Chem. Eur. J. 2004, 10, 2237 2252
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

B. M. Trost and M. L. Crawley
FULL PAPER
building block. The efficacy of this moiety to direct the dia-
stereoselective assembly of substituted furanones was well
demonstrated by its providing equivalent or superior results
to known chiral butenolides under a variety of reactions.
The crystalline natures of the products, amplified by the
aryloxy group, simplified their purification, often requiring
only recrystallization. Conjugate additions to butenolide 4
with stabilized nucleophiles, particularly nitroalkanes, gave
complete diastereofacial control in forming the new stereo-
genic center. Cycloaddition reactions proceeded with high
facial selectivity, often affording only a single diastereomer
with up to four new stereogenic centers in outstanding
yields. The excellent yields in these additions and the ability
to run certain reactions neat with butenolide 4 has further il-
lustrated its synthetic utility. The examples of TMM dipolar
cycloadditions to the chiral butenolide 4 could not have
given better results: complete control of regio- and diaster-
eoselectivity in excellent yield. A short synthesis of Bayer
compound 5 illustrated the utility of these building blocks
for a pharmaceutical application.
The highly versatile synthetic strategy that takes advant-
age of this palladium-catalyzed AAA-derived ™chiral alde-
hyde∫ equivalent has led to a total synthesis of (+)-brefel-
din A (6). The synthesis was highly convergent and assem-
bled the natural product from three components: commer-
cial ethyl propiolate, a chiral six-carbon sulfone-ether frag-
ment, and a highly functionalized cyclopentane core. The
core fragment was readily assembled from a TMM-derived
cyclopentenoid system, formed as a single diastereomer
from a dipolar cycloaddition addition with the ™chiral alde-
hyde∫ building block. After the introduction of the C(1)
C(3) upper side chain, through a diastereoselective addition
reaction that gave access to either epimer at C(4), it is note-
worthy that this was the first total synthesis to employ the
newly developed trans-selective hydrosilylation protocol as a
reduction method. The methodology was extended and used
as a method for trans-selective alkyne reduction in the pres-
ence of other sensitive functional groups in model studies.
The lower side chain 27 was made in several steps from the
adduct of a regio- and enantioselective AAA reaction. This
key reaction was notable for its low catalyst and ligand load-
ing, its scalability, and its outstanding yield. After brief elab-
oration of the AAA-derived piece, it was efficiently intro-
duced to the molecule×s core in a trans-selective Julia olefi-
nation. The effort was completed in 18 linear steps with 6%
overall yield from furfural 33, an inexpensive and readily
available starting material. Four of the stereogenic centers
derive from a palladium-catalyzed DYKAT, while the re-
maining chirality was established by a palladium-catalyzed
AAA reaction. In this synthesis, the potential for the design
and synthesis of analogues is noteworthy, as the developed
methodology allows access to either enantiomer of the core
and lower side chain and allows reaction conditions to con-
trol the relative stereochemistry of the remaining stereogen-
ic centers.
spectroscopy by comparison of the integration of the major
and minor olefin signals. Both the major and minor adducts
had olefins with irregular splitting patterns and coupling
constants: (d=5.38 5.35 and 5.49 5.40 ppm, respectively).
The major adduct was presumed to be the E adduct, as later
confirmed by completion of the total synthesis.
The total synthesis was completed in a four-step sequence.
The 4-methoxyphenol group of 61 was oxidatively cleaved
by use of cerium ammonium nitrate, followed by ester sapo-
nification with sodium hydroxide to afford acid 62 in 76%
yield over two steps (Scheme 14). Macrolactonization by Ya-
Scheme 14. Completion of the total synthesis of (+)-brefeldin A (6).
maguchi×s method[34] to give the bis(silyl)-protected brefel-
din 63 and subsequent global deprotection with tetrabuty-
lammonium fluoride afforded (+)-brefeldin A (6) in two
steps and 51% yield. All of the spectral data, as well as the
sign and magnitude of rotation ([a]D =+89.6Æ0.5 (c=0.40
in MeOH)), matched those of an authentic sample ([a]D =
+91.2Æ0.4 (c=0.50 in MeOH))[35] of (+)-brefeldin A (6).
Conclusion
The enantiopure furanone 4, readily available through a pal-
ladium-catalyzed AAA in a dynamic kinetic asymmetric
transformation, has shown its utility as a ™chiral aldehyde∫
2248
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2237 2252

A ™Chiral Aldehyde∫ Equivalent
2237 2252
125.1, 118.3, 111.4, 105.0, 92.8, 84.0, 39.4 ppm; IR (film): n˜ =3055, 2982,
1790, 1631, 1599, 1511, 1350, 1215, 1163, 1076, 964 cmÀ1; elemental analy-
sis calcd (%) for C15H12N2O3: C 67.16, H 4.51; found: C 66.92, H 4.76.
Experimental Section
(4R,5S)-4-(1’,1’-Bismethoxycarbonylmethyl)-5-(2-naphthoxy)dihydrofur-
an-2-one (7): Dimethyl malonate (72.7 mg, 0.550 mmol) and DMF
(1.5 mL) were mixed, and sodium methoxide in methanol (2n, 0.10 mL)
was added. Butenolide ent-4 (113 mg, 0.500 mmol) in DMF (1.0 mL) was
added, and the solution was stirred at RT for 12 h. The mixture was dilut-
ed with methylene chloride (25 mL), washed with a saturated aqueous
solution of ammonium chloride (0.5n 25 mL), water (20 mL), and brine
(20 mL), and after drying over sodium sulfate was concentrated in vacuo.
Flash chromatography (diethyl ether/pet. ether 2:1) afforded lactone 7 as
a white foam/gel (138 mg, 77%). [a]D =+137Æ1 (c=4.00 in CHCl3); 1H
NMR (500 MHz, CHCl3): d=7.81 7.77 (m, 3H), 7.51 7.38 (m, 3H), 7.20
(dd, J=9.0, 2.2 Hz, 1H), 6.14 (d, J=2.4 Hz, 1H), 3.79 (s, 3H), 3.77 (s,
3H), 3.68 (d, J=2.2 Hz, 1H), 3.42 3.36 (m, 1H), 3.11 (dd, J=18.2,
9.5 Hz, 1H), 2.62 ppm (dd, J=18.2, 5.0 Hz, 1H); 13C NMR (75 MHz,
CHCl3): d=173.7, 167.5, 153.8, 134.0, 130.1, 129.7, 127.6, 126.6, 124.8,
118.6, 111.5, 103.1, 53.2, 53.0, 51.9, 41.4, 31.4 ppm; IR (film): n˜ =3023,
(3R,3aS,6aR)-5-Benzyl-3-(2-naphthoxy)hexahydrofuro[3,4-c]pyrrol-1-
one (18/19): TFA (0.5m, 20 mL, 1.02 mg, 0.010 mmol) was added to a sol-
ution of butenolide 4 (22.6 mg, 0.100 mmol) and dipole precursor 17
(47.4 mg, 0.200 mmol) in methylene chloride (0.25 mL). The solution was
stirred at À108C for 4 h. Flash chromatography (pet. ether/diethyl ether
1:1) gave two products with a crude dr>3:1, as determined by 1H NMR
spectroscopy by comparison of the integrals of the major and minor sig-
nals for the anomeric center protons (d=5.92 pm, major, d=6.21ppm,
minor). The products were both white solids; 18 (25.4 mg, 71%); 19
(7.6 mg, 21%). The combined yield was 92% (33.0 mg).
Compound 18: M.p. 50 518C; [a]D =+121Æ0.1( c=1.50 in CHCl3); 1H
NMR (500 MHz, CDCl3): d=7.79 7.76 (m, 3H), 7.48 7.25 (m, 8H), 7.19
(dd, J=8.8, 2.5 Hz, 1H), 5.92 (d, J=1.5 Hz, 1H), 3.60 (abx, J=56.2,
13.0 Hz, 2H), 3.35 3.30 (m, 2H), 3.18 3.08 (m, 2H), 2.51 2.45 ppm (m,
2H); 13C NMR (75 MHz, CDCl3): d=177.9, 153.9, 137.9, 134.2, 130.0,
129.7, 128.5, 128.4, 127.6, 127.3, 126.6, 124.6, 118.6, 111.0, 106.1, 58.6,
57.8, 57.1, 45.5, 43.7 ppm; IR (film): n˜ =3060, 2963, 2806, 1786, 1630,
1600, 1511, 1467, 1254, 1176, 1095, 977, 954 cmÀ1; elemental analysis
calcd (%) for C23H21NO3: C 76.86, H 5.88; found: C 76.92, H 6.11.
2956, 1797, 1734, 1631, 1600, 1511, 1467, 1436, 1254, 1213, 1161 cmÀ1
;
HRMS: m/z calcd for C19H18O7: 358.1052; found: 358.1061 [M]+.
(4R,5S)-5-(Naphthoxy)-4-(1’-methyl-1’-nitroethyl)dihydrofuran-2-one (9):
DBU (15.2 mg, 0.100 mmol) was added to a solution of 4-(2-naphthoxy)-
butenolide ent-4 (226 mg, 1.00 mmol) and 2-nitropropane (134 mg,
1.50 mmol) in methylene chloride (5.00 mL). The solution was stirred at
RT for 1h and was then subjected to flash chromatography (pet. ether/
diethyl ether 1.5:1.0) to afford nitroalkane 9 as a white solid (293 mg,
93%). M.p. 114 1158C; [a]D =+163Æ0.8 (c=1.00 in CHCl3); 1H NMR
(300 MHz, CDCl3): d=7.83 7.79 (m, 3H), 7.52 7.42 (m, 3H), 7.22 (dd,
J=8.8, 2.5 Hz, 1H), 6.07 (d, J=2.9 Hz, 1H), 3.39 3.35 (m, 1H), 3.05 (dd,
J=18.3, 10.0 Hz, 1H), 2.53 (dd, J=18.3, 6.0 Hz, 1H), 1.71 ppm (s, 6H);
13C NMR (75 MHz, CDCl3): d=172.8, 153.6, 134.0, 130.3, 129.9, 127.7,
127.3, 126.8, 125.0, 118.5, 111.7, 102.1, 87.5, 50.0, 29.8, 24.3, 24.0 ppm; IR
(film): n˜ =2994, 1796, 1631, 1600, 1542, 1511, 1468, 1348, 1252, 1213,
1158, 1069 cmÀ1; HRMS: m/z calcd for C17H17NO5: 315.1106; found:
315.1097 [M]+.
Compound 19: M.p. 135 1368C; [a]D =+167Æ0.4 (c=1.50 in CHCl3);
1H NMR (500 MHz, CDCl3): d=7.80 7.76 (m, 3H), 7.48 7.25 (m, 8H),
7.19 (dd, J=8.8, 2.5 Hz, 1H), 6.21 (d, J=6.6 Hz, 1H), 3.80 (d, 13.0 Hz,
1H), 3.70 (dd, J=10.2, 2.4 Hz, 1H), 3.60 (d, J=13.0 Hz, 1H) 3.35 3.25
(m, 3H), 2.57 (dd, J=9.7, 7.0 Hz, 1H), 2.33 ppm (dd, J=9.8, 7.0 Hz,
1H); 13C NMR (75 MHz, CDCl3): d=177.4, 154.4, 138.9, 134.1, 130.0,
129.7, 128.5, 128.4, 127.6, 127.3, 126.6, 124.6, 118.6, 111.1, 100.9, 58.5,
57.2, 53.1, 45.2, 42.3 ppm; IR (film): n˜ =3060, 2963, 2806, 1786, 1630,
1600, 1511, 1467, 1254, 1176, 1095, 977, 954 cmÀ1; elemental analysis
calcd (%) for C23H21NO3: C 76.86, H 5.88; found: C 77.00, H 6.12.
(3aS,6aR)-5-Methylenehexahydrocyclopenta[c]-furan-1-one
(25):
Sodium borohydride (95 mg, 2.50 mmol) was added to a solution of
sodium hydroxide (200 mg, 5.0 mmol) in water (5 mL). After 5 min, ent-4
(140 mg, 1.00 mmol) was added. The solution was stirred for 12 h at RT,
and concentrated aqueous HCl (1mL) was added dropwise (vigorous
bubbling). The solution was stirred for an additional 1h at RT, diluted
with methylene chloride (20 mL), and washed with water (20 mL). The
organic phase was dried over sodium sulfate and concentrated in vacuo.
Flash chromatography (pet. ether/diethyl ether 1:1) afforded product 25
as a clear oil (58 mg, 83%). The data matched the data provided by
{3R-[3a(1R,2S,5R)-3aa,4a,7a,7aa]}3a,4,7,7a-Tetrahydro-3-(2-naph-
thoxy)-4,7-ethanoisobenzofuran-1-(3H)-one (11): Butenolide ent-4
(45.2 mg, 0.200 mmol) and 1,3-cyclohexadiene 10 (0.19 mL, 160 mg,
2.00 mmol) were heated neat at 1508C in the microwave for 1h. The
mixture was concentrated in vacuo and diluted with methylene chloride
(1.00 mL). Flash chromatography (pet. ether/diethyl ether 4:1) afforded
cycloadduct 11 as a white solid as a single diastereomer (58.7 mg, 96%).
M.p. 94.5 95.08C; [a]D =+234Æ1 (c=3.40 in CHCl3); 1H NMR
(500 MHz, CHCl3): d=7.86 7.77 (m, 3H), 7.50 7.47 (m, 1H), 7.44 7.40
(m, 1H), 7.20 (dd, J=9.0, 2.4 Hz), 1H), 6.39 (t, J=7.2 Hz, 1H), 6.34 (t,
J=7.2 Hz, 1H), 5.70 (d, J=1.7 Hz, 1H), 3.20 3.18 (m, 1H), 3.06 (dd, J=
9.6, 3.5 Hz, 1H), 2.98 2.97 (m, 1H), 2.93 2.90 (m, 1H), 1.66 1.63 (m,
2H), 1.41 1.37 ppm (m, 2H); 13C NMR (75 MHz, CHCl3): d=177.5,
154.1, 134.3, 134.1, 132.4, 129.9, 129.6, 127.6, 127.2, 126.6, 124.5, 118.7,
110.9, 104.7, 46.1, 44.9, 31.8, 31.7, 23.5, 23.2 ppm; IR (film): n˜ =3053,
2949, 2870, 1784, 1630, 1600, 1511, 1468, 1391, 1365, 1253, 1213, 1168,
1143 cmÀ1; HRMS: m/z calcd for C20H18O3: 306.1256; found: 306.1254
[M]+.
Bayer company.[19] [a]D =À58Æ0.1( c=3.80 in CHCl3); Bayer data[19]
:
1
[a]D =À58.7 (c=1.00 in CHCl3); H NMR (500 MHz, CDCl3): d=4.93 (s,
2H), 4.45 4.42 (m, 1H), 4.04 (dd, J=9.2, 2.8 Hz, 1H), 3.06 3.01 (m, 2H),
2.72 2.69 (m, 3H), 2.24 2.19 ppm (m, 1H); 13C NMR (75 MHz, CDCl3):
d=180.1, 147.8, 108.2, 72.6, 43.9, 39.1, 38.7, 35.3 ppm; IR (film): n˜ =3077,
2917, 2851, 1766, 1667, 1480, 1432, 1374, 1171, 1125, 1050, 983 cmÀ1
;
HRMS: m/z calcd for C8H10O2: 138.0681; found: 138.0686.
BAY 36-7620 (5): A solution of lithium bis(trimethylsilyl)amide in THF
(1.0m, 0.220 mL, 36.8 mg, 0.220 mmol) was added under argon at À788C
to a solution of substrate 25 (27.6 mg, 0.200 mmol) in toluene (1.00 mL).
After 30 min at RT, a solution of 2-(bromomethyl)naphthalene (48.6 mg,
0.220 mmol) in toluene (1.00 mL) was added; then the bright yellow solu-
tion was stirred for 12 h at RT. The mixture was quenched with water
(1.0 mL), and the organic layer was dried over sodium sulfate. Flash
chromatography (pet. ether/diethyl ether 3:1) afforded product 26 as a
colorless oil/gel (46.2 mg, 83%). The data matched the data provided by
the Bayer company.[19] [a]D =À30Æ0.1( c=1.60 in CHCl3); Bayer data:
[a]D =À33.0 (c=1.0 in CH2Cl2);[19] 1H NMR (500 MHz, CDCl3): d=7.83
7.78 (m, 3H), 7.67 (brs, 1H), 7.49 7.45 (m, 2H), 7.36 (dd, J=8.4, 2.2 Hz,
1H), 4.92 (d, J=14.4 Hz, 2H), 3.75 (dd, J=9.2, 3.9 Hz, 1H), 3.64 (dd,
J=9.2, 7.7 Hz, 1H), 3.46 (d, J=13.7 Hz, 1H), 2.92 (d, J=13.7 Hz, 1H),
2.90 2.82 (m, 2H), 2.76 2.70 (m, 1H), 2.58 (d, J=6.1Hz, 1H), 2.25
2.21ppm (m, 1H); 13C NMR (75 MHz, CDCl3): d=181.6, 147.1, 134.3,
133.4, 132.4, 128.4, 128.3, 127.8, 127.7, 126.6, 126.2, 125.8, 108.4, 72.0,
56.1, 43.6, 42.1, 41.8, 39.1 ppm; IR (film): n˜ =3055, 2971, 2911, 1765,
1664, 1600, 1508, 1431, 1376, 1175, 1138, 1049 cmÀ1; HRMS: m/z calcd for
C19H18O2: 278.1307; found: 278.1302.
(3S,4R,6aS)-4-(2-Naphthoxy)-3,3a,4,6a-tetrahydrofuro[3,4-c]pyrazol-6-
one (16): A solution of diazomethane [ca. 2.00 mmol, prepared by treat-
ment of 1-methyl-3-nitro-1-nitrosoguanidine (294 mg, 2.00 mmol) in di-
ethyl ether (2.00 mL) with 20% aqueous KOH solution (6 mL)] was
added at 08C to a solution of butenolide 4 (45.2 mg, 0.200 mmol) in di-
ethyl ether (2.00 mL). The solution was stirred for 2 h at À108C. The re-
action mixture was quenched with SiO2, diluted with CH2Cl2 (20 mL), fil-
1
tered, and concentrated in vacuo. H NMR spectroscopy (crude material)
indicated
a dr>4:1by comparison of the a-acetoxysulfone protons
(major d=5.88 ppm, minor d=6.14 ppm). Flash chromatography (pet.
ether/diethyl ether 1:3) gave product 16 as a white solid (38.4 mg, 76%).
1H NMR indicated the product was a single diastereomer. M.p. 172
1738C; [a]D =+77.8Æ0.2 (c=1.00 in CHCl3); 1H NMR (300 MHz,
CDCl3): d=7.80 7.2 (m, 3H), 7.49 7.38 (m, 3H), 7.16 (dd, J=19.2,
2.5 Hz, 1H), 5.88 (ddd, J=9.3, 1.5, 1.0 Hz, 1H), 5.77 (d, J=1.5 Hz, 1H),
5.14 5.04 (m, 1H), 4.97 4.89 (m, 1H), 3.27 3.19 ppm (m, 1H); 13C NMR
(75 MHz, CDCl3): d=166.3, 153.4, 134.0, 130.2, 130.0, 127.7, 127.3, 126.9,
2249
Chem. Eur. J. 2004, 10, 2237 2252
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

B. M. Trost and M. L. Crawley
FULL PAPER
(S)-1-Methylprop-2-enyl 4-methoxyphenyl ether (29): Carbonate 34
(2.15 g, 16.5 mmol) was added under argon at 08C to a degassed solution
of 4-methoxyphenol (1.86 g, 15.0 mmol), [Pd2(dba)3]¥CHCl3 (77.6 mg,
0.075 mmol), and ligand 3 (155.4 mg, 0.225 mmol) in toluene (400.0 mL).
After stirring at 08C under argon for 12 h, the reaction mixture was di-
rectly subjected to chromatography (pet. ether/diethyl ether 10:1) to
afford 29 as a clear oil (2.61g, 95%). The characterization data matched
known values.[36] [a]D =À7.5Æ0.1( c=5.40 in CHCl3); Lit. [23] [a]D =
À1.5Æ0.1( c=1.0 in CH2Cl2); 1H NMR (300 MHz, CDCl3): d=6.87 6.77
(m, 4H), 5.93 5.87 (m, 1H), 5.23 (dd, J=17.3, 1.3 Hz, 1H), 5.14 (dd, J=
10.0, 1.3 Hz, 1H), 4.68 (quintet, J=6.3 Hz, 1H), 3.74 (s, 3H), 1.40 ppm
(d, J=6.3 Hz, 3H); 13C NMR (75 MHz, CDCl3): d=153.9, 152.0, 139.5,
117.4, 115.6, 114.5, 114.4, 75.7, 55.6, 21.3 ppm; IR (film): n˜ =2981, 2933,
1.92 (m, 1H), 1.80 1.74 (m, 1H), 0.86 (s, 9H), 0.04 (s, 3H), 0.03 ppm (s,
3H); 13C NMR (125 MHz, CHCl3): d=202.6, 174.5, 72.2, 61.2, 51.5, 39.1,
38.3, 35.4, 32.4, 25.7, 18.0, À4.8 ppm; IR (film): n˜ =2931, 2857, 2713,
1723, 1665, 1464, 1388, 1253, 1118, 1006, 838 cmÀ1; HRMS: m/z calcd for
C14H26NO4Si: 300.1632; found: 300.1647 [MÀCH3]+.
Ethyl (S)-4-[(1R,2R,4R)-4-(tert-butyldimethylsilyloxy)-2-(methoxymeth-
ylcarbamoyl)cyclopentyl]-4-hydroxybut-2-ynoate (53)
Compound 53a: A n-butyllithium solution (1m in hexanes, 0.13 mL,
1.80 mmol) was added at À788C under argon to a solution of ethyl pro-
piolate (177 mg, 1.80 mmol) in THF/HMPA 6:1 (7 mL). The solution was
stirred for 1h at À788C and became dark orange in color. Aldehyde 52
(378 mg, 1.20 mmol) in THF/HMPA solution (6:1, 7 mL) was added at
À788C, and the solution was stirred for 4 h. The mixture was quenched
with ammonium chloride solution (10 mL) and extracted with diethyl
ether (3î20 mL), and the organic phase was dried over sodium sulfate
and concentrated in vacuo. Crude 1H NMR showed a 6.5:1.0 diastereo-
meric mixture of (4S)/(4R) by comparison of the a-hydroxymethyl
proton signals of the Weinreb amide (major d=3.73 ppm, minor d=
3.70 ppm). Flash chromatography (diethyl ether/pet. ether 3:1) first
eluted ynoate 53a as a colorless oil (364 mg, 73%) and then ynoate 53b
as a colorless oil (75 mg, 15%), contaminated with a small amount of
53a. The combined yield was 88% (439 mg). [a]D =À14.8Æ0.2 (c=3.40
2834, 1506, 1465, 1442, 1228, 1039, 928, 826 cmÀ1
; tr(S)=15.57 min,
tr(R)=17.30 min, 90% ee, (Chiralcel OD, l=254 nm, hept/iPrOH
99.9:0.1).
2-(S)-(2-Naphthoxy)-5-oxo-2,5-dihydrofuran (ent-4): 2-Naphthol (3.00 g,
20.8 mmol) in methylene chloride (50 mL) was added at 08C under argon
to
a
degassed solution of butenolide
1
(4.98 g, 25.0 mmol),
[Pd2(dba)3]¥CHCl3 (538 mg, 0.520 mmol), S,S-ligand ent-3 (1.08 g,
1.56 mmol), and tetrabutylammonium chloride hydrate (1.74 g,
6.25 mmol) in methylene chloride (150 mL) by the procedure reported by
Trost and Toste.[6] The solution was stirred at 0 58C for 14 h. After con-
centration to 50 mL volume, it was subjected to chromatography on silica
gel (pet. ether/diethyl ether 3:1) to afford butenolide ent-4 as a white
solid (2.84 g, 84%). The characterization data were consistent with the
literature data.[6] M.p. 73 748C; lit.[6] m.p. 72 748C; [a]D =+328Æ1.2
(c=1.1 in CH2Cl2); lit.[6] [a]D =+336 (c=1.05 in CH2Cl2); 1H NMR
(300 MHz, CDCl3): d=7.81 7.77 (m, 3H), 7.58 7.42 (m, 4H), 7.23 (dd,
J=9.0, 2.7 Hz, 1H), 6.51 (d, J=1.3 Hz, 1H), 6.36 ppm (dd, J=5.7 Hz,
1.3 Hz, 1H); 13C NMR (75 MHz, CDCl3): d=169.9, 154.1, 149.8, 134.0,
130.2, 129.8, 127.6, 127.3, 126.7, 125.3, 124.9, 118.5, 111.5, 100.7 ppm; IR
(CHCl3): 3112, 3060, 1793, 1761, 1631, 1600, 1466, 1368 cmÀ1; tr(S)=
11.8 min; tr(R)=15.2 min (major for +328), 96% ee (Chiralcel AD, l=
230 nm, hept/iPrOH 9:1).
1
in CHCl3); H NMR (500 MHz, CHCl3): d=4.58 4.56 (m, 1H), 4.32 4.28
(m, 1H), 4.20 (q, J=7.2 Hz, 2H), 3.78 (d, J=5.4 Hz, 1H), 3.73 (s, 3H),
3.19 (s, 3H), 3.09 3.02 (m, 2H), 2.38 2.34 (m, 1H), 1.82 1.64 (m, 3H),
1.28 (t, J=7.1Hz, 3H), 0.85 (s, 9H), 0.03 (s, 3H), 0.01ppm (s, 3H); 13C
NMR (125 MHz, CHCl3): d=175.6, 153.2, 86.4, 77.2, 71.8, 64.3, 62.0, 61.3,
44.6, 41.6, 40.5, 37.5, 32.6, 25.8, 17.9, 13.9, À4.8 ppm; IR (film): n˜ =3400,
2938, 2857, 2235, 1714, 1662, 1462, 1389, 1249, 1117, 1058, 1006 cmÀ1
;
HRMS: m/z calcd for C19H32NO6Si: 398.1999; found: 398.2008
[MÀCH3]+.
Compound 53b: A n-butyllithium solution (1m in hexanes, 0.13 mL,
1.80 mmol) was added at À788C under argon to a solution of ethyl pro-
piolate (177 mg, 1.80 mmol) and magnesium bromide (331 mg,
1.80 mmol) in DME (5.00 mL). The solution was stirred for 1 h at À788C
and became dark orange in color. Aldehyde 52 (378 mg, 1.20 mmol) in
DME (5 mL) was added at À788C, and the solution was stirred for 4 h.
The mixture was quenched with ammonium chloride solution (10 mL)
and extracted with diethyl ether (3î20 mL), and the organic phase was
dried over sodium sulfate and concentrated in vacuo. Crude 1H NMR
showed a 1.0:6.0 diastereomeric mixture of (4S)/(4R) by comparison of
the a-hydroxymethyl proton signals of the Weinreb amide (minor d=
3.73 ppm, major d=3.70 ppm). Flash chromatography (diethyl ether/pet.
ether 3:1) first eluted ynoate 53a as a colorless oil (65 mg, 13%) and
then ynoate 53b as a colorless oil (389 mg, 78%), contaminated with a
trace amount of 53a. The combined yield was 91% (454 mg). [a]D =
À17.4Æ0.3 (c=0.80 in CHCl3); 1H NMR (500 MHz, CHCl3): d=4.34
4.28 (m, 2H), 4.20 (q, J=7.2 Hz, 2H), 3.78 (d, J=5.4 Hz, 1H), 3.70 (s,
3H), 3.19 (s, 3H), 3.09 3.02 (m, 2H), 2.38 2.34 (m, 1H), 1.82 1.64 (m,
3H), 1.28 (t, J=7.1Hz, 3H), 0.85 (s, 9H), 0.03 (s, 3H), 0.01ppm (s, 3H);
13C NMR (125 MHz, CHCl3): d=175.6, 153.2, 86.6, 77.2, 72.0, 65.2, 62.0,
62.0, 44.6, 41.6, 40.5, 37.5, 32.6, 25.8, 17.9, 13.9, À4.8 ppm; IR (film): n˜ =
3400, 2938, 2857, 2235, 1714, 1662, 1462, 1389, 1249, 1117, 1058,
1006 cmÀ1; HRMS: m/z calcd for C19H32NO6Si: 398.1999; found: 398.2006
[MÀCH3]+.
(3S,3aS,6aR)-5-Methylene-3-(2-naphthoxy)tetrahydrocyclopenta[c]furan-
1-one (23a): Triisopropyl phosphite (491 mL, 416 mg, 2.00 mmol) was
added to a solution of ent-4 (2.26 g, 10.0 mmol), alkene 20a (2.79 g,
15.0 mmol), and palladium(ii) acetate (56.1mg, 0.250 mmol) in toluene
(100 mL). The solution was stirred at reflux for 12 h. After concentration
in vacuo to 10 mL, flash chromatography (pet. ether/diethyl ether 3:1)
gave product 23a as a white solid (2.53 g, 90%). 1H NMR indicated the
product was a single diastereomer. M.p. 74 74.58C; [a]D =+235Æ0.3
(c=2.00 in CHCl3); 1H NMR (500 MHz, CDCl3): d=7.79 7.74 (m, 3H),
7.47 7.36 (m, 3H), 7.20 (dd, J=6.6, 2.4 Hz, 1H), 5.90 (s, 1H), 4.97 (s,
2H), 3.39 3.34 (m, 1H), 3.20 (dd, J=9.1, 3.0 Hz, 1H), 2.87 2.78 (m, 3H),
2.36 ppm (dd, J=16.0, 6.0 Hz, 1H); 13C NMR (125 MHz, CDCl3): d=
178.6, 153.9, 146.7, 134.1, 130.0, 129.7, 127.6, 127.3, 126.6, 124.6, 118.6,
111.0, 108.8, 104.7, 46.0, 43.4, 36.2, 35.6 ppm; IR (film): n˜ =2956, 1787,
1631, 1600, 1511, 1467, 1252, 1214, 1158, 1121, 947, 932 cmÀ1; elemental
analysis calcd (%) for C18H16O3: C 77.12, H 5.75; found: C 77.36, H 5.90.
(1R,2R,4R)-4-(tert-Butyldimethylsilyloxy)-2-formylcyclopentane-1-N,O-
dimethylhydroxamine (52): Lactone ent-31b (797 mg, 2.00 mmol) and
N,O-dimethylhydroxylamine hydrochloride (488 mg, 5.00 mmol) made
into a slurry in THF (20 mL) under argon and chilled to À108C. Isopro-
pylmagnesium chloride in THF (4.8 mL, 987 mg, 9.60 mmol) was added
dropwise over 5 min. The solution was stirred for 0.5 h at À108C and was
then quenched with saturated aqueous ammonium chloride (20 mL). The
mixture was extracted with methylene chloride (100 mL). The organic
phase was dried over sodium sulfate, and concentrated in vacuo. The
product was filtered through a plug of silica (diethyl ether/pet. ether 5:1
eluent) to afford a clear oil. The oil was dissolved in methylene chloride
(20 mL) and treated with DBU (304 mg, 2.00 mmol). The solution was
heated at reflux with stirring for 6 h, diluted with methylene chloride to a
total volume of 100 mL, and washed with 1n aqueous ammonium chlo-
ride solution (100 mL). The organic phase was dried over sodium sulfate
and concentrated in vacuo. Flash chromatography (diethyl ether/pet.
ether 3:1) afforded aldehyde 52 as a clear oil (454 mg, 72%, 2 steps).
[a]D =À10.1Æ0.1( c=4.00 in CHCl3); 1H NMR (500 MHz, CHCl3): d=
9.71 (brs, 1H), 4.18 4.13 (m, 1H), 3.71 (s, 3H), 3.56 3.51 (m, 1H), 3.41
3.36 (m, 1H), 3.20 (s, 3H), 2.28 2.24 (m, 1H), 2.05 2.00 (m, 1H), 1.97
Ethyl (E)-(R)-4-[(1R,2R,4R)-4-(tert-butyldimethylsilyloxy)-2-(methoxy-
methylcarbamoyl)cyclopentyl]-4-hydroxybut-2-enoate (56a): The rutheni-
um catalyst (2.10 mg, 0.0040 mmol) was added at 08C under argon to a
methylene chloride solution (10 mL) of alkyne 53a (165 mg, 0.400 mmol)
and triethoxysilane (78.9 mg, 0.480 mmol). The mixture was allowed to
warm to RT with stirring over 2 h. Cesium fluoride (72.9 mg, 0.480 mmol)
and ethanol (1.00 mL) were added, and the solution was stirred at RT for
12 h. The mixture was diluted with methylene chloride (50 mL) and
washed with water (2î30 mL) and brine (30 mL). The organic phase was
dried over sodium sulfate and concentrated in vacuo. Flash chromatogra-
phy (diethyl ether/pet. ether 4:1) afforded enoate 56a as a clear oil
(153 mg, 92%). [a]D =À16.2Æ0.2 (c=1.80 in CHCl3); 1H NMR
(500 MHz, CHCl3): d=6.92 (dd, J=15.6, 5.4 Hz, 1H), 6.05 (d, J=
15.6 Hz, 1H), 4.41 4.39 (m, 1H), 4.27 4.22 (m, 1H), 4.20 (d, J=7.0 Hz,
2H), 3.69 (s, 3H), 3.21 (s, 3H), 3.01 2.95 (m, 1H), 2.93 2.90 (m, 1H),
2.52 (d, J=4.7 Hz, 1H), 2.34 2.30 (m, 1H), 1.69 1.62 (m, 3H), 1.29 (t,
2250
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2237 2252

A ™Chiral Aldehyde∫ Equivalent
2237 2252
J=7.1Hz, 3H), 0.87 (s, 9H), 0.04 (s, 3H), 0.03 ppm (s, 3H); 13C NMR
(125 MHz, CHCl3): d=166.4, 164.8, 148.1, 121.2, 71.9, 71.3, 61.3, 60.4,
44.6, 41.0, 40.5, 36.1, 29.7, 25.8, 18.0, 14.2, À4.8 ppm; IR (film): n˜ =3429,
2928, 2856, 1719, 1654, 1464, 1383, 1257, 1175, 1116 cmÀ1; HRMS: m/z
calcd for C20H37NO6Si: 400.2156; found: 400.2176 [MÀCH3]+.
(20 mL). The organic phase was washed with aqueous HCl solution
(0.25n, 20 mL) and saturated aqueous sodium bicarbonate solution
(20 mL) and then dried over sodium sulfate. After concentration in
vacuo, chromatography (pet. ether/diethyl ether 20:1) afforded lactone
63 as a clear oil (16.8 mg, 66%). [a]D =+40.2Æ0.5 (c=0.50 in CHCl3);
1H NMR (500 MHz, CHCl3): d=7.29 (dd, J=15.5, 3.1 Hz, 1H), 5.86 (dd,
J=15.5, 2.0 Hz 1H), 5.63 (ddd, J=15.6, 10.4, 4.5 Hz, 1H), 5.27 (dd, J=
15.6, 5.5 Hz, 1H), 4.92 4.85 (m, 1H), 4.21 4.18 (m, 1H), 4.03 4.00 (m,
1H), 2.28 2.21 (m, 1H), 2.07 1.94 (m, 4H), 1.88 1.79 (m, 2H), 1.75 1.70
(m, 1H), 1.56 1.45 (m, 3H), 1.27 (d, J=6.1 Hz, 3H), 1.26 1.23 (m, 1H)
0.93 (s, 9H), 0.85 (s, 9H), 0.06 (s, 3H), 0.04 (s, 3H), 0.03 (s, 3H),
0.01ppm (s, 3H); 13C NMR (75 MHz, CHCl3): d=166.4, 152.5, 137.3,
129.3, 118.0, 76.3, 72.8, 71.4, 52.8, 43.8, 43.6, 42.0, 34.1, 31.8, 26.7, 25.8,
20.9, 18.1, 18.0, À4.1, À4.8, À4.9 ppm; IR (film): n˜ =2929, 2857, 1716,
1646, 1410, 1255, 1122, 1078, 1004, 967, 837, 774 cmÀ1; HRMS: m/z calcd
for C28H52O4Si2: 508.3404; found: 508.3414 [M]+.
Ethyl (E)-(R)-4-(tert-butyldimethylsilyloxy)-{(1R,2R,4R)-4-(tert-butyldi-
methylsilyloxy)-2-[(E)-(S)-6-(4-methoxyphenoxy)hept-1-enyl]cyclopen-
tyl}but-2-enoate (61): A solution of tetrazole 17 (93.7 mg, 0.225 mmol) in
DME (1.0 mL) was added at À788C under argon to a solution of potassi-
um hexamethyldisilylazide (49.9 mg, 0.250 mmol) in DME (1.00 mL).
After the mixture had been kept for 1h at À788C, a solution of aldehyde
51 (70.6 mg, 49.9 mg) in DME (0.5 mL) was added; the reaction mixture
was stirred for 1h at À788C. The solution was allowed to warm to RT,
stirred for 16 h at RT, poured into brine (5.0 mL), and extracted with di-
ethyl ether (3î5 mL). The organic phase was dried over sodium sulfate
and concentrated in vacuo. Flash chromatography (pet. ether/diethyl
ether 8:1) afforded product 61 as a clear oil (72 mg, 81%) The E/Z ratio
as determined by 1H NMR spectroscopy was 12:1. [a]D =+5.5Æ0.1( c=
1.00 in CHCl3); 1H NMR (300 MHz, CDCl3): d=6.85 (dd, J=15.6,
3.0 Hz, 1H), 6.83 (s, 4H), 5.90 (d, J=15.6 Hz, 1H), 5.38 5.35 (m, 2H),
4.26 4.16 (m, 5H), 3.77 (s, 3H), 2.33 2.28, (m, 1H), 2.03 1.98 (m, 3H),
1.98 1.93 (m, 1H), 1.89 1.82 (m, 1H), 1.75 1.67 (m, 2H), 1.58 1.40 (m,
4H), 1.21 (t, J=6.4 Hz, 3H), 0.92 (s, 9H), 0.87 (s, 9H), 0.05 (s, 3H), 0.03
(s, 6H), 0.00 ppm (s, 3H); 13C NMR (75 MHz, CDCl3): d=166.4, 151.1,
144.7, 134.5, 129.8, 128.4, 125.9, 119.8, 117.4, 114.6, 74.4, 72.8, 71.2, 60.3,
55.7, 49.2, 43.6, 42.6, 36.0, 34.1, 32.4, 29.7, 25.8, 19.8, 18.1, 18.0, À4.0,
À4.3, À4.8, À4.9 ppm; IR (film): n˜ =2929, 2856, 1722, 1655, 1506, 1463,
1370, 1229, 1166, 1106, 1039 cmÀ1; HRMS: m/z calcd for C37H64O6Si2:
660.4241; found: 660.4238 [M]+.
(+)-Brefeldin A (6): A solution of TBAF (1.0m in THF, 100 mL, 26.1mg,
0.100 mmol) was added at 08C to a solution of substrate 63 (16.7 mg,
0.0328 mmol) in THF (1mL). The solution was stirred at 0 8C for 4 h.
The mixture was diluted with ethyl acetate (10 mL) and washed with
water (10 mL), and the organic phase was concentrated in vacuo. Flash
chromatography (ethyl acetate/pet. ether 5:1) afforded (+)-brefeldin A
(6) as a white solid (8.1mg, 77%). The rotation and spectral data match-
ed those observed for natural (+)-brefeldin A. M.p. >2008C; lit.[18] m.p.
202 2038C; [a]D =+89.6Æ0.5 (c=0.40 in MeOH); lit.[18] [a]D =+91.2Æ
0.4 (c=0.50 in MeOH); 1H NMR (500 MHz, CHCl3): d=7.45 (dd, J=
15.5, 3.0 Hz, 1H), 5.80 (dd, J=15.5, 2.0 Hz, 1H), 5.70 (ddd, J=15.0, 10.5,
4.5 Hz, 1H), 5.27 (dd, J=15.6, 5.5 Hz, 1H), 4.79 4.75 (m, 1H), 4.21 4.18
(m, 1H), 4.02 (ddd, J=9.5, 3.1, 2.0 Hz, 1H), 2.37 (quintet, 1H), 2.10
(ddd, J=13.5, 8.4, 5.2 Hz, 1H), 2.02 1.97 (m, 2H), 1.88 1.72 (m, 5H),
1.58 1.50 (m, 1H), 1.46 1.40 (m, 1H), 1.23 (d, J=6.3 Hz, 3H), 0.89
0.85 ppm (m, 1H); 13C NMR (75 MHz, CHCl3): d=168.4, 155.2, 138.1,
131.4, 117.8, 76.6, 73.2, 73.0, 53.2, 45.5, 44.1, 41.8, 35.0, 33.0, 28.0,
21.1 ppm; IR (film): n˜ =3395, 2924, 2852, 1711, 1693, 1638, 1448, 1353,
(E)-(R)-4-(tert-Butyldimethylsilyloxy)-{(1R,2R,4R)-4-(tert-butyldimethyl-
silyloxy)-2-[(E)-(S)-6-hydroxyhept-1-enyl]cyclopentyl}but-2-enoic
acid
(62): Ceric ammonium nitrate (137 mg, 0.250 mmol) in water (0.500 mL)
was added at 08C to a solution of substrate 61 (66.1mg, 0.100 mmol) in
acetone (2.00 mL). The solution was stirred for 30 min at 08C. The mix-
ture was extracted with diethyl ether (2î10 mL), and the organic phase
was dried over sodium sulfate and concentrated in vacuo. The crude mix-
ture was filtered through a thin plug of silica with diethyl ether as eluent.
Free OH: 1H NMR (500 MHz, CHCl3): d=6.92 (dd, J=15.7, 5.3 Hz,
1H), 6.18 (d, J=15.7 Hz, 1H), 5.44 5.40 (m, 2H), 4.24 4.21 (m, 1H),
4.21 4.18 (m, 1H), 4.20 (q, J=7.2 Hz, 2H), 3.86 3.82 (m, 1H), 2.32 2.27
(m, 1H), 2.07 1.96 (m, 4H), 1.82 1.77 (m, 1H), 1.60 1.55 (m, 1H), 1.52
1.40 (m, 5H), 1.24 (d, J=6.1Hz, 3H), 0.98 (s, 9H), 0.92 (t, J=7.2 Hz,
3H), 0.90 (s, 9H), 0.08 (s, 3H), 0.05 (s, 6H), 0.03 ppm (s, 3H); IR (film):
n˜ =3360, 2930, 2858, 1692, 1659, 1472, 1405, 1250 cmÀ1; HRMS: m/z calcd
for C26H49O5Si2: 497.3118; found: 497.3114 [MÀC4H9]+.
1259, 1119 cmÀ1
;
HRMS: m/z calcd for C24H46O4: 280.1675; found:
260.1668 [M]+.
Acknowledgment
We thank the National Institutes of Health (GM-13598) and the National
Science Foundation for their generous support of our programs. Mass
spectra were provided by the Mass Spectrometry Regional Center of the
University of California, San Francisco, supported by the NIH Division
of Research Resources.
The concentrated crude mixture was dissolved in a THF/MeOH mixture
(2:1, 3.00 mL). Aqueous sodium hydroxide (1n, 1mL) was added, and
the mixture was stirred at 608C for 1h. The mixture was acidified with
1n sodium hydrogen sulfate solution (10 mL), extracted with diethyl
ether (3î5 mL), and dried over sodium sulfate. Flash chromatography
(methanol/chloroform 3:97) afforded acid 62 as a clear oil (40.6 mg,
77%, 2 steps). [a]D =À9.6Æ0.5 (c=0.30 in CHCl3); 1H NMR (500 MHz,
CHCl3): d=7.01(dd, J=15.6, 5.1 Hz, 1H), 5.95 (d, J=15.7 Hz, 1H),
5.38 5.32 (m, 2H), 4.25 4.22 (m, 1H), 4.21 4.18 (m, 1H), 3.86 3.82 (m,
1H), 2.32 2.27 (m, 1H), 2.07 1.96 (m, 4H), 1.82 1.77 (m, 1H), 1.59 1.54
(m, 1H), 1.52 1.40 (m, 5H), 1.21 (d, J=6.2 Hz, 3H), 0.95 (s, 9H), 0.90 (s,
9H), 0.08 (s, 3H), 0.05 (s, 6H), 0.03 ppm (s, 3H); 13C NMR (75 MHz,
CHCl3): d=170.2, 153.6, 134.7, 129.8, 119.0, 72.9, 72.2, 68.2, 49.5, 43.6,
42.9, 38.5, 36.0, 32.4, 25.9, 25.8, 25.7, 23.4, 18.1, À3.9, 4.8 ppm; IR (film):
[1] For reviews on the use of chiral acetals in synthesis see: a) B. E.
Rossiter, N. M. Swingle, Chem. Rev. 1992, 92, 566; b) A. Alexakis, P.
Mangeney, Tetrahedron: Asymmetry 1990, 1, 477; c) J. K. Whitesell,
Chem. Rev. 1989, 89, 1581.
[2] For some examples of stereochemical control in addition reactions
to olefins directed by chiral acetals see: a) J. Fujiwara, Y. Fukutani,
K. M. Hasegawa, H. Yamamoto, J. Am. Chem. Soc. 1984, 106, 5004;
b) A. Ghribi, A. Alexakis, J. F. Normant, Tetrahedron Lett. 1984, 25,
3083; c) Y. Fukutani, K. Maroka, H. Yamamoto, Tetrahedron Lett.
1984, 25, 5911; d) E. A. Mash, S. B. Hemperly, K. A. Nelson, P. C.
Heidt, S. Van Deusen, J. Org. Chem. 1990, 55, 2045; e) E. A. Mash,
J. B. Arterburn, J. Org. Chem. 1991, 56, 885.
[3] S. A. Godleski, Comprehensive Organic Synthesis, Vol. 4 (Eds.: B. M.
Trost, I. Fleming, M. F. Semmelhack), Pergamon, Oxford, 1990,
Chapter 3.3.
[4] a) B. M. Trost, Chem. Pharm. Bull. 2002, 50, 1; b) B. M. Trost, M. L.
Crawley, Chem. Rev. 2003, 103, 2921.
n˜ =3360, 2930, 2858, 1699, 1659, 1472, 1410, 1256, 1120, 836, 775 cmÀ1
;
HRMS: m/z calcd for C24H45O5Si2: 469.2805; found: 469.2800
[MÀC4H9]+.
(5E,13E)-(2S,3aR,4R,10S,14aS)-2,4-Bis-(tert-butyldimethylsilyloxy)-9-
methyl-1,2,3,3a,4,9,10,11,12,14a-decahydro-8-oxacyclopentacyclotridecen-
7-one (63): Triethylamine (7.65 mg, 0.075 mmol) and 2,4,6-trichloroben-
zoyl chloride (17.1 mg, 0.070 mg) were added to a solution of substrate
62 (26.4 mg, 0.050 mmol) in THF (0.500 mL). The solution was stirred at
RT for 3 h. That mixture was diluted with toluene (6.0 mL) and added
dropwise over 3 h to a solution of 4-dimethylaminopyridine (30.5 mg,
0.250 mmol) in toluene (10.0 mL) at reflux. The solution was stirred at
reflux for an additional 8 h. The mixture was diluted with diethyl ether
[5] a) B. L. Feringa, B. de Lange, J. C. de Jong, J. Org. Chem. 1989, 54,
2471; b) H. van der Deen, A. D. Cuiper, R. P. Hof, A. van Oeveren,
B. L. Feringa, R. M. Kellogg, J. Am. Chem. Soc. 1996, 118, 3801;
c) H. van der Deen, A. van Oeveren, R. M. Kellogg, B. L. Feringa,
Tetrahedron Lett. 1999, 40, 1755.
2251
Chem. Eur. J. 2004, 10, 2237 2252
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

B. M. Trost and M. L. Crawley
FULL PAPER
[6] For a review on dynamic kinetic asymmetric transformations see:
R. S. Ward, Tetrahedron: Asymmetry 1995, 6, 1475; B. M. Trost, F. D.
Toste, J. Am. Chem. Soc. 1999, 121, 3543.
[7] For more information see: a) B. M. Fox, J. A. Vroman, P. E. Fan-
wick, M. Cushman, J. Med. Chem. 2001, 44, 3915; b) A. B. Argade,
R. Devraj, J. A. Vroman, R. D. Haugwitz, M. Hollingshead, M.
Cushman, J. Med. Chem. 1998, 41, 3337, and references therein.
[8] For the first enantioselective synthesis of (+)-brefeldin A see: a) T.
Kitahara, K. Mori, M. Matsui, Tetrahedron Lett. 1979, 20, 3021; b) T.
Kitahara, K. Mori, Tetrahedron 1984, 40, 2935.
[16] B. M. Trost, D. M. T. Chan, T. N. Naninga, Org. Synth. 1984, 62, 58.
[17] B. M. Trost, D. M. T. Chan, J. Am. Chem. Soc. 1983, 105, 2315.
[18] B. M. Trost, T. N. Nanninga, T. Satoh, J. Am. Chem. Soc. 1985, 107,
721. For more details see T. N. Nanninga, Ph.D. Thesis, University of
Wisconsin, 1985, Chapter 3.
[19] We would like to thank Dr. Dirk Heimbach and the scientists at
Bayer for providing us with their data to confirm our results; b) See
also: A. Stolle, H. P. Antonicek, S. Lensky, A. Voerste, G. Mueller,
U. Stropp, E. Hovarth, V. De Vry, F. Mauler, R. Schreiber, German
Patent No. WO9936416, 1999; c) See also: A. Stolle, H. P. Antoni-
cek, S. Lensky, F. Mauler, J. Mittendorf, T. Mueller, A. Voerste,
222nd ACS Meeting, Division of Medicinal Chemistry, Chicago, Illi-
nois 2001, Abstr. 223; d) C. Y. Fiona, A. Stolle, H. P. Antonicek, S.
Lensky, F. Mauler, J. Mittendorf, T. Muller, J. P. Pin, L. Prezeau,
Mol. Pharmacol. 2001, 59, 965.
[20] J. B. Baudin, G. Hareau, S. A. Julia, R. Lorne, O. Ruel, Bull. Soc.
Chim. Fr. 1993, 130, 856 and earlier references; P. J. Kocienski, A.
Bell, P. R. Blakemore, Synlett 2000, 365 and earlier references. For a
review see P. R. Blakemore, J. Chem. Soc. Perkin Trans. 1 2002,
2563.
[21] F. D. Toste, Ph.D. Thesis, Stanford Unversity, 2000, Ch. 6, 494; com-
pare B. M. Trost, F. D. Toste, J. Am. Chem. Soc. 1999, 121, 4545.
[22] B. M. Trost, D. L. Van Vranken, Chem. Rev. 1996, 96, 395.
[23] H. Fukui, F. Yoshinori, K. Fujita, T. Nakagawa, H. Koshino, T.
Nakata, Bioorg. Med. Chem. Lett. 1997, 7, 2081; M. Jung, D. A.
Bustos, H. N. Elsohly, J. D. McChesney, Synlett 1990, 743.
[24] B. M. Trost, D. P. Curran, Tetrahedron Lett. 1981, 22, 1287.
[25] U. Handschin, H. P. Sigg, C. Tamm, Helv. Chim. Acta 1968, 51, 1943.
[26] R. Pappo, D. S. Allen, Jr., R. U. Lemieus, W. S. Johnson, J. Org.
Chem. 1956, 21, 478.
[9] For other enantioselective total syntheses of (+)-brefeldin A see:
a) C. Le Drian, A. E. Greene, J. Am. Chem. Soc. 1982, 104, 5473;
b) H.-J. Gais, K. L. Lukas, Angew. Chem. 1984, 96, 140; Angew.
Chem. Int. Ed. Engl. 1984, 23, 142c) H.-J. Gais, T. Lied, Angew.
Chem. 1984, 96, 143; Angew. Chem. Int. Ed. Engl. 1984, 23, 145;
d) B. M. Trost, J. Lynch, P. Renaut, D. H. Steinman, J. Am. Chem.
Soc. 1986, 108, 284; e) E. J. Corey, P. Carpino Tetrahedron Lett.
1990, 31, 7555; f) J. Nokami, M. Ohkura, Y. Danoh, Y. Sakamoto,
Tetrahedron Lett. 1991, 32, 2409; g) F. Taber, J. Silverberg, E. D.
Robinson, J. Am. Chem. Soc. 1991, 113, 6639; h) G. Solladie, O.
Lohse, J. Org. Chem. 1993, 58, 4555; i) D. Kim, J. I. Lim, Tetrahe-
dron Lett. 1995, 36, 5035; j) V. Bernardes, N. Kann, A. Riera, A.
Moyano, M. A. Pericas, A. E. Greene, J. Org. Chem. 1995, 60, 6670;
k) Y. Kobayashi, K. Watatani, Y. Kikori, R. Mizojiri, Tetrahedron
Lett. 1996, 37, 6125; l) D. Kim, J. Lee, P. J. Shim, J. I. Lim, H. Jo, S.
Kim, J. Org. Chem. 2002, 67, 764; m) R. K. Haynes, W. W. L. Lam,
L-.L. Yeung, I. D. Williams, A. C. Ridley, S. M. Starling, S. C. Von-
willer, T. W. Hambley, P. Lelandais, J. Org. Chem. 1997, 62, 4552;
n) D. Kim, J. Lee, P. J. Shim, J. I. Lim, T. Doi, S. Kim, J. Org. Chem.
2002, 67, 772; o) T.-G. Suh, J.-K. Jung, S.-Y. Seo, K.-H. Min, D.-Y.
Shin, Y.-S. Lee, S.-H. Kim, H.-J. Park, J. Org. Chem. 2002, 67, 4127;
p) Y.-G. Suh, S.-Y. Seo, J.-K. Jung, O.-H. Park, R.-O. Jeon, Tetrahe-
dron Lett. 2001, 42, 1691; q) Y.-G. Suh, J.-K. C. Jung, Y.-C. Lee, S.-
A. Kim, Tetrahedron Lett. 1998, 39, 5377; r) P. Ducray, B. Rousseau,
C. Mioskowski, J. Org. Chem. 1999, 64, 3800; s) D. Wang, D. Romo,
Org. Lett. 2002, 4, 3231.
[27] H. Yamamoto, S. Iguchi, H. Nakai, M. Hayashi, J. Org. Chem. 1979,
44, 1363.
[28] J. M. Williams, R. B. Jobson, N. Yasuda, G. Marchegini, U. H. Dol-
ling, E. J. J. Grabowski, Tetrahedron Lett. 1995, 36, 3461.
[29] T. Takai, K. Nitta, K. Utimoto, J. Am. Chem. Soc. 1986, 108, 7408.
[30] For a review of the Suzuki reaction see: N. Miyaura, A. Suzuki,
Chem. Rev. 1995, 95, 2457.
[31] M. R. Netherton, C. Dai, K. Neuschultz, G. C. Fu, J. Am. Chem.
Soc. 2001, 123, 10099.
[32] B. M. Trost, Z. T. Ball, J. Am. Chem. Soc. 2001, 123, 12726.
[33] B. M. Trost, J. L. Belletire, S. Godleski, P. G. McDougal, J. M. Balko-
vec, J. J. Baldwin, M. E. Christy, G. S. Ponticello, S. L. Varga, J. P.
Springer, J. Org. Chem. 1986, 51, 2370.
[10] For a preliminary account of a portion of this work see B. M. Trost,
M. L. Crawley, J. Am. Chem. Soc. 2002, 124, 9328.
[11] For a discussion of Michael additions to chiral butenolides see:
E. M. Geertsema, C. W. Leung, A. van Oeveren, A. Meetsma, B. L.
Feringa, Tetrahedron Lett. 1995, 36, 7315, and references therein.
[12] F.-A. Kang, Z.-Q. Yu, H.-Y. Yin, C.-L. Yin, Tetrahedron: Asymmetry
1997, 8, 3591.
[13] J. C. De Jong, F. van Bolhuis, B. L. Feringa, Tetrahedron: Asymmetry
1991, 2, 1247.
[34] M. Yamaguchi, J. Innaga, K. Hirata, H. Sacki, T. Katsuki, Bull.
Chem. Soc. Jpn. 1979, 52, 1989.
[14] E. Keller, B. de Lange, M. T. Rispens, B. L. Feringa, Tetrahedron
1993, 49, 8899.
[35] (+)-Brefeldin A was purchased from the Sigma chemical company.
[36] F. Dean Toste, Ph.D. Thesis, Stanford Unversity, 2000, Ch. 5, 406.
[15] M. T. Rispens, E. Keller, B. de Lange, J. Zijlstra, B. L. Feringa, Tet-
rahedron: Asymmetry 1994, 5, 607.
Received: October 17, 2003 [F5634]
2252
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2237 2252