AAA in KAT/DYKAT processes: First- and second-generation asymmetric syntheses of (+)- and (-)-cyclophellitol

Article abstract of DOI:10.1002/1521-3765(20010903)7:17<3768::AID-CHEM3768>3.0.CO;2-C

Kinetic resolutions and kinetic asymmetric transformations (KAT) as well as dynamic kinetic resolutions and dynamic kinetic asymmetric transformations (DYKAT) are important synthetic protocols. The feasibility of KAT and DYKAT processes for asymmetric all

Full text of DOI:10.1002/1521-3765(20010903)7:17<3768::AID-CHEM3768>3.0.CO;2-C

FULL PAPER
AAA in KAT/DYKAT Processes: First- and Second-Generation Asymmetric
Syntheses of (‡)- and ( )-Cyclophellitol
Barry M. Trost,* Daniel E. Patterson, and Erik J. Hembre[a]
Abstract: Kinetic resolutions and kinet-
ic asymmetric transformations (KAT) as
well as dynamic kinetic resolutions and
dynamic kinetic asymmetric transforma-
tions (DYKAT) are important synthetic
protocols. The feasibility of KAT and
DYKAT processes for asymmetric allyl-
ic alkylations (AAA) is explored utiliz-
ing a single substrate ± conduritol B tet-
raesters. Both processes can be per-
formed
resulting
in
excellent
groups is also described. For this pur-
pose, 4-tert-butyldimethylsiloxy-2,2-di-
methylbutyric acid was developed as a
nucleophile. The utility of effecting
KAT/DYKAT processes through the
Pd-catalyzed AAA reaction is demon-
strated by efficient syntheses of both
enantiomers of the potent glycosidase
inhibitor cyclophellitol.
enantioselectivity. The impact of nucle-
ophile and leaving group on the effec-
tiveness of each is outlined. The ability
to differentiate the various hydroxyl
Keywords: allyl complexes ´ asym-
metric catalysis ´ natural products ´
palladium ´ total synthesis
Conduritols, the various stereoisomers of 3,4,5,6-tetrahy-
droxylated cyclohexene, represent multifunctional starting
materials of growing importance.[7] The symmetry of condur-
itol B (2a and 3a) makes it a particularly interesting
asymmetric building block.[8] As a chiral racemic substance,
a kinetic resolution is feasible (see Scheme 1). Since a benefit
Introduction
Glycosidase inhibitors not only aid in the understanding of
glycoprotein processing but also may find applications in
immunology, diabetes, virology, and cancer. Cyclophelli-
tol (1), first isolated in 1990 from the mushroom Phellinus
sp.,[1] has shown activity as an inactivator of b-glucosidase and
an inhibitor of HIV.[2] Because of its potent biological activity,
cyclophellitol has been the focus of extensive synthetic effort.
However, to date most syntheses start with enantiomerically
pure natural products, most notably carbohydrates.[3] Such
strategies frequently engender long synthetic sequences due
to the need to manipulate functionality largely by the use of
protecting groups. In addition, these strategies only provide
access to one enantiomer of the natural product. In contrast,
de novo asymmetric syntheses can provide the flexibility to
access either enantiomer, and often lead to more efficient
strategies. Only one asymmetric synthesis of cyclophellitol
from achiral starting materials employing a chiral auxiliary
has been reported.[4] Herein, we report a first-[5] and a second-
generation asymmetric synthesis of cyclophellitol based on
the palladium-catalyzed asymmetric allylic alkylation (AAA)
of racemic conduritol tetraesters.[6]
DYNAMIC KINETIC ASYMMETRIC
TRANSFORMATION (DYKAT)
KINETIC
RESOLUTION
KINETIC
ASYMMETRIC
TRANSFORMATION
or
OR
OR
OR
OR
RO
Nu
OR
4
OR
2
OR
Nu
L*
a) R = H
Nu
OR
b) R = COR'
Pd0
OR
4
OR
OR
OR
RO
OR
RO
OR
2
OR
3 (ent-2)
a) R = H
b) R = COR'
Scheme 1. Methods for de-racemization of conduritols B.
[a] Prof. B. M. Trost, Dr. D. E. Patterson, Dr. E. J. Hembre
Department of Chemistry
of the AAA process[9] is incorporation of the enantiomeric
discriminating step as one of the other bond forming events,
thereby reducing the overall number of required steps, the
process would be more properly referred to as a kinetic
asymmetric transformation (KAT). Thus, the desired product
Stanford University
Stanford, CA 94305-5080 (USA)
E-mail: bmtrost@stanford.edu
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/chemistry/ or from the author.
3768
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3768 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 17

3768±3775
is not the resolved starting material (e.g. 2) but the enantio-
merically enriched product (e.g. 4) of the chemical trans-
formation. However, since ionization to a p-allylpalladium
intermediate generates a complex bearing a plane of symme-
try, it is possible to also envision both enantiomers of the
substrate reacting to form an enantiomerically pure product
(e.g. 4), a process termed a dynamic kinetic asymmetric
transformation (DYKAT). In realizing this scheme, the less
reactive enantiomer of 2 or 3 must react faster than the
product 4, which still bears an allyl ester, reacts further.[10] The
advantage of this latter process, if achievable, is obviousÐthe
theoretical yield of the desired enantiomer of the product is
now 100% rather than 50% for a KAT. In this paper, we
explore both strategies for the synthesis of cyclophellitol.
edº (see e.g. 2). A pivalate was chosen as the carboxylate
nucleophile for the resolution because the resulting allyl
pivalate was anticipated to ionize much more slowly that the
starting material, and the pivalate should provide a means for
easy differentiation of the alcohol protecting groups later on.
A potential problem at this stage is the further reaction of
monopivalate 11 to provide the dipivalate 12. Due to the C2
symmetry of the system, the remaining allyl acetate in 11 is
also ªmatchedº for ionization with respect to ligand 9. The
second ionization requires the catalyst to complex the face of
the double bond cis to the pivalate substituent. The steric bulk
of the pivalate was anticipated to hamper this undesired
ionization.
The tetraacetate 8 was synthesized in three steps from
benzoquinone by a simple modification of the procedure of
Guo et al.[6] In our modification, the diacetate of the
dibromodiol was formed with acetic anhydride and potassium
carbonate to which was added acetic acid and heating to
produce the tetraacetate in a one-pot operation. [Eq. (1)] and
Table 1 summarizes our results.
First generation: Our initial retrosynthesis of (‡)-cyclophel-
litol is shown in Scheme 2.[5] As depicted, cyclophellitol would
be available through a hydroxyl-directed epoxidation of a
HO
HO
OR
OR
OH
OH
OR
OR
Li
OAc
O
O
OR
AcO
OAc
OH
OR
6
5
OAc
(+)-cyclophellitol [ (+)-1]
(9)
O
O
(+)-8
NH HN
OAc
OAc
OR
OR
OAc
OAc
+
O
AcO
PPh2 Ph2P
HO
OAc
O
tC4H9
(1)
OR
(+)-8
[η3-C3H5PdCl]2 (10)
OAc
7
8
11
(nC6H13)4NBr (THAB)
tC4H9CO2H, NaOH
H2O, CH2Cl2
Scheme 2. First-generation retrosynthesis of (‡)-cyclophellitol.
O
tC4H9
O
O
suitably protected homoallylic alcohol 5, which would result
from a 2,3-Wittig rearrangement of 6, available from the
selectively protected conduritol B derivative 7. Enantiomeri-
cally enriched conduritol B derivative 7 would come from the
palladium-catalyzed kinetic resolution of conduritol B tetra-
acetate 8.[6]
A palladium catalyzed kinetic resolution of the racemic C2-
symmetric tetraacetate is possible using the chiral ligand
(R,R)-9 and a Pd source such as 10, since, with respect to the
ligand, one enantiomer of 8 would provide a ªmatchedº
substrate for ionization while the other would be ªmismatch-
O
tC4H9
OAc
OAc
12
Using our standard ligand 9[11] in a two phase water/methylene
chloride solvent system, the initial reaction using 5 mol% 10
led mainly to a mixture of recovered tetraacetate 8, monop-
ivalate 11, and only a small amount of the dipivalate 12
(entry 1). Reducing the catalyst loading in the same period of
time did further minimize the formation of dipivalate
(entries 2 and 3) although, with 1 mol% 10, there was more
Table 1. KAT of racemic conduritol B tetraacetate.
Catalyst 10
9
equiv
equiv
m
Yield [% ee][a]
Mol%
Mol%
tC4H9CO2H
NaOH
conc.
t[h]
(‡)-8
11
12
1
2
3
4
5
2.5
1
15
7.5
3
1.5
1.25
0.65
0.65
0.65
0.4
0.4
0.4
0.5
12
12
12
24
31
39
63
50
(88, 99[b]
53
(68)
74
(30)
54
39
42
27
44
(97)
43
(97)
21
(99)
39
8
22
±
0.75
0.75
0.80
1
3
1
)
5
6
7
0.5
0.5
0.5
3
0.80
0.80
0.80
0.65
0.65
0.65
0.5
0.5
1.0
8
36
24
1.5
1.5
(46)
(98)
[a]% ee determined by chiral HPLC analysis using a Chiracel AD column. [b]% ee after one recrystallization from diethyl ether/petroleum ether.
Chem. Eur. J. 2001, 7, No. 17 ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001 0947-6539/01/0717-3769 $ 17.50+.50/0
3769

B. M. Trost et al.
FULL PAPER
unreacted starting material than would be expected for a
kinetic resolution (entry 3). As a result, the concentrations of
all of the reactants as well as the time was increased. In this
case (entry 4), an excellent result ensued. The reaction
stopped at 50% conversion to produce a quantitative yield
(based upon a 50% theoretical yield) of recovered tetraace-
tate of 88% ee and an 88% yield (based upon a 50%
theoretical yield) of monopivalate 11 of 97% ee. In the case of
the recovered tetraacetate (‡)-8, the ee increased to 99%
after one recrystallization from diethyl ether/petroleum ether.
Decreasing the reaction time to 8 h (entry 5) led to incom-
plete reaction as revealed by recovery of more than the
theoretical amount of tetraacetate for a kinetic resolution and
the diminished ee of the recovered tetraacetate. Further
lowering of the catalyst load (entries 6 and 7) also led to
incomplete reactions. In all of the incomplete reactions, the ee
of the product 11 was excellent (97 ± 99% ee) as was expected.
With the availability of the monopivalate of 97 ± 99% ee
almost quantitatively in a KAT, the stage was set for the first-
generation synthesis of cyclophellitol as outlined in Scheme 3.
depicted. Completion of the synthesis by the published
hydrogenation procedure[3k] gave (‡)-cyclophellitol (1); its
physical and spectral properties are in full agreement with the
natural product.
Second generation: Replacing a KAT by a DYKAT has the
ability to improve the efficiency of any synthesis. Condur-
itol B constitutes a particularly favorable starting material for
such a change in strategy as pointed out in Scheme 1.
Realization of a DYKAT requires 1) the ability of both
enantiomers of the starting material, 2b and 3b, to react and
2) reaction of both to be faster than further reaction of the
allyl ester that remains in the product 4. Since further
ionization of 4 requires coordination of palladium to the face
of the double bond cis to the newly introduced nucleophile;
the steric bulk of the latter may inhibit the further reaction of
4. Adjustment of the leaving group ability of the carboxylate
esters was anticipated to overcome this hurdle. Scheme 4
OH
OH
OH
OH
OCOR
X
O
OCOR
OCOR
HO
O
OR
OBn
a,b,c
d,e
OBn
OBn
OCOR
OCOR
OH
OCOR
PivO
OR
R'O
OBn
OR
(-)-1
19
20
OBn
OBn
17
OCOR
OCOR
11: R = Ac
13: R = H
14: R = Bn
OCOR
OCOR
15: R' = H
16: R' = CH2SnBu3
OCOR
OCOR
OCOR
HO
HO
OCOR
OBn
OBn
g
OH
OH
21
R = CCl3CH2O
ent-21
f
O
O
Scheme 4. Second-generation retrosynthesis of ( )-cyclophellitol.
(+)-1
OBn
18
OH
Scheme 3. First-generation synthesis of (‡)-cyclophellitol. a) NH4OH/
MeOH, room temperature, 95%; b) BnBr, Bu4NI, KH, DME, 79%;
c) DIBAL-H, CH2Cl2, 788C, 90%; d) KH, Bu3SnCH2I, DME, 08C, 88%;
outlines a second-generation retrosynthetic analysis. As
previously, the olefin 19 is anticipated to be the immediate
precursor, but the hydroxylmethyl group is projected to be
introduced in the initial enantiodiscriminating step in the
form of a carboxylic acid equivalent (as in 20) using a DYKAT
of racemic 21. To achieve the DYKAT, the sluggishness of the
mismatched ionization experienced with the tetraacetate had
to be overcome. We, therefore, turned to the trichloroethyl
carbonate (Troc) leaving groups. The success of phenylsulfon-
ylnitromethane[13] as an alkoxycarbonyl anion equivalent[14]
led to its use in this sequence.
The Troc-tetraester 21 was prepared in two steps from the
tetraacetate ± base hydrolysis [(C2H5)3N, CH3OH, H2O, r.t.]
followed by treatement with Troc-chloride. Reaction of
racemic 21 (R ˆ CCl3CH2O) with the sodium salt of (phenyl-
sulfonyl)nitromethane 22, 2.5% [dba3Pd3] ´ CHCl3 and 7.5% 9
in THFat room temperature gave the desired mono-adduct 23
in 81% yield [Eq. (2)]. The presence of a diastereomeric pair
due to the side chain made determination of ee at this stage
difficult; thus, it was postponed. Complete conversion ob-
served in this reaction indicates that the chiral palladium
catalyst can ionize both enantiomers of (‡)-21, and convert
them to a single enantioenriched product.
e) nBuLi, THF,
788C, 72%; f) MCPBA, CH2Cl2, 08C, 78%; g) H2
(1 atm), Pd(OH)2/C, CH3OH, room temperature, 90%.
Introduction of the last remaining carbon through the
equivalent of an SN2' with retention of configuration envi-
sioned employment of a 2,3-sigmatropic rearrangement.[12]
Cleavage of the acetates of 11 in the presence of the
pivalate was straightforward with ammonium hydroxide in
methanol, whereby triol 13 was obtained in 95% yield.
Protection of the triol as the benzyl ethers as in 14, followed
by pivalate cleavage afforded alcohol 15, which could be
converted to the corresponding tin derivative 16. The stage
was set for the 2,3-sigmatropic rearrangement, which was
accomplished by treatment of 16 with n-butyllithium to put
the hydroxymethyl group in place with correct regio- and
diastereochemistry, to afford 17. Epoxidation with MCPBA,
directed by the homoallylic alcohol, gave a 78% yield of the
desired epoxide 18 with a 7% yield of the diastereomeric
epoxide. Comparison of the spectral data for 18 with that
reported in the literature[3k] confirmed the stereochemistry as
3770
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3770 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 17

KAT/DYKAT processes
3768±3775
O-
enantiomeric excess of the product could be determined.
Alcohol 26 was converted to the corresponding tetrabenzoate
27 which could be separated via chiral HPLC to establish the
enantiomeric excess of the palladium reaction as 88%.
Conversion of 26 into cyclophellitol required epoxidation
directed by the homoallylic alcohol. This olefin was expected
to be fairly unreactive toward electrophilic epoxidation
because of the electron withdrawing allylic carbonate. Treat-
ment with peracids (MCPBA, trifluoroperacetic acid, or 3,5-
dinitroperbenzoic acid[20]) at room temperature confirmed the
unreactive nature of 26, giving no reaction or even traces of
product. Reaction with peracids at higher temperature (with
added radical inhibitor) caused only decomposition. Treat-
ment with vanadium acetylacetonate/hydrogen peroxide
caused decomposition, while titanium isopropoxide/hydrogen
peroxide[21] returned mainly starting material with only traces
of the desired epoxide. Reaction of 26 with dimethyldioxirane
proceeded at room temperature to give the desired epoxides
(28 and 29) in good yield, but with no facial selectivity
[Eq. (5)].
N+
PhO2S
+
(+)-21
O-Na+
R = CCl3CH2O
22
PhO2S
NO2
OTroc
(2)
2.5 % [dba3Pd2]•CHCl3,
7.5 % (R,R)-9, Cs2CO3, THF
81%, 88% ee
OTroc
OTroc
23
Several methods to convert 23 to the corresponding acid or
ester were examined. Treatment of the nitronate salt of 23
with TBA-oxone,[15] or oxone[16] gave mainly decomposition
products, or no reaction. Treatment with CAN,[17] also led to
decomposition, along with traces of the corresponding a,b-
unsaturated acid. These methods may have led to decom-
position through conversion to the desired ester, followed by
deprotonation, elimination of the carbonates, and aromatiza-
tion. Simple hydrolysis of the carbonates may have also
contributed to the decomposition products. Treatment with
titanium trichloride[18] also did not give the desired ester, but,
instead, gave the phenylsulfonyl oxime 24 which didnꢁt react
further [Eq. (3)].
HO
HO
HO
OTroc
OTroc
OTroc
OTroc
OTroc
OTroc
a or b
O
+
O
OTroc
OTroc
26
OTroc
28
(5)
29
OH
PhO2S
N
a: dimethyldioxirane, acetone
room temperature.
30%
65%
30%
OTroc
OTroc
TiCl3, NH4OAc, H2O, THF
70%
(3)
23
b: MTO, H2O2 (30% aqueous),
CH2Cl2,, room temperature
15%
OTroc
24
The known efficiency of rhenium-catalyzed epoxidation of
electron poor olefins,[22] led to its investigation for the
epoxidation of 26. Reaction of 26 with methylrhenium
trioxide (MTO) and urea/hydrogen peroxide addition com-
plex[23] in methylene chloride gave a low yield (< 20%) of the
desired epoxide, but with a reasonable diastereoselectivity of
3:1. Some reports claimed that the rhenium catalyst is more
reactive using hydrogen peroxide as the oxidant in a biphasic
(methylene chloride/water) system with added pyrazole,
pyridine or cyanopyridine as a co-catalyst.[22a, 24] The basic
co-catalyst is thought to both protect the product epoxide
from acid-catalyzed ring opening, and to stabilize the catalyst.
Reaction of 26 under these conditions, however, decreased
the yield of the desired epoxide. When the reaction was
carried out without added base, using 10% rhenium catalyst,
the conversion increased, giving complete conversion to the
desired epoxide as a 3.5:1 diastereomeric mixture favoring the
desired diastereomer 28. The diastereomers could be sepa-
rated to give a 65% isolated yield of the desired epoxide 28
along with 15% of the undesired epoxide 29 [Eq. (5)]. The
major diastereomer derived from hydroxyl-directed epoxida-
tion,[25] which was confirmed by conversion of 28 to the
enantiomer of the natural product.
The mildness of dioxiranes[19] as oxidants led to examina-
tion of this method to convert nitrosulfone 23 to the
corresponding carboxylic acid 25. Treatment of 23 with
N,N,N',N'-tetramethylguanidine (TMG) to form the nitronate
salt, followed by reaction with dimethyldioxirane in acetone
afforded the desired acid 25 in 78% yield. Interestingly, the
double bond of 23 or 25 was unreactive to dimethyldioxirane,
as the corresponding epoxide was never observed. Carboxylic
acid 25 was then reduced in the presence of the carbonates by
treatment with borane/dimethyl sulfide to provide the desired
alcohol 26 in 70% yield [Eq. (4)]. At this point the
HO
O
1. TMG, CH2Cl2, 0° C
2. dimethyldioxirane,
acetone
BH3•Me2S, THF,
0° C, 12 h
OTroc
OTroc
23
70%
78%
OTroc
25
(4)
HO
BzO
1. K2CO3, MeOH
2. Bz-Cl, pyridine,
CH2Cl2
OTroc
OTroc
OBz
OBz
71%
The trichloroethyl carbonates of the major epoxide 28 were
hydrolyzed by treatment with potassium carbonate in meth-
anol to give a 95% yield of ( )-cyclophellitol (1) with a full
OTroc
OBz
27
(88% ee by HPLC)
26
Chem. Eur. J. 2001, 7, No. 17
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3771 $ 17.50+.50/0
3771

B. M. Trost et al.
FULL PAPER
agreement of the physical and spectral properties (except for
the optical rotation) with the natural product [Eq. (6)]. At this
point the enantiomeric excess could be improved to 99% by a
single recrystallization from water.
2-bromoethyl tert-butyldimethylsilyl ether [Eq. (8)]. It partic-
ipates in the exactly analogous fashion in the kinetic
resolution/KATas pivalic acid as summarized in Equation (9).
O
O
1) 2 equiv LDA
2) Br
OTBDMS
HO
HO
HO
(8)
(9)
OH
OTBDMS
K2CO3, MeOH,rt
32
O
(6)
28
95%
OH
OH
OAc
OAc
O
O
(-)-cyclophellitol [ (-)-1]
as in
(+)-8
32
(+)-8
TBDMSO
Equation (1)
OAc
33
Discussion and Conclusion
In this case, recovered tetraacetate of 87% ee was also
obtained quantitatively and the mono-pivalate analogue 33 of
95% ee was obtained in 88% yield. Exposure of 33 to HF/
pyridine only effects desilylation. Acid-catalyzed lactoniza-
tion completes the removal to form the triacetate 34
[Eq. (10)]. Thus, differentiation among all of the hydroxyl
groups of scalemic conduritol B is possible by this approach.
The versatility of the palladium-catalyzed AAA reaction
stems from the myriad of methods for introducing chirality.[9]
Kinetic resolutions and kinetic asymmetric transformations
still constitute an important strategy. On the other hand,
dynamic kinetic resolutions and asymmetric transformations
are intrinsically more efficient. Conduritol B allows exploita-
tion of both strategies and demonstrates the effectiveness of
both with the same substrate.
With the ªstandardº ligand 9, the kinetic resolution/KAT
was nearly perfect. The resolved tetraacetate (‡)-8 was
recovered quantitatively (50% maximum yield) in 83% ee
which was increased to 99% ee with one recrystallization. The
transformed tetraester 11 of 97% ee was formed in 88% yield.
The latter tetraester provides not only a resolution but also a
differentiation among the hydroxyl groups. For example,
removal of the esters proceeds straightforwardly with base to
triol 13 (see Scheme 3). The pivalate provides a steric
differentiation of the triol moiety permitting formation of
the acetonide 30. Alternatively, the two hydroxyl groups anti
to the pivalate may be preferentially protected as silyl ethers
to give a 10:1 mixture of 31 and the 5,6-bis-silyl ether from
which 31 is isolated pure in 72% yield [Eq. (7)]. The position
of the silyl groups, as depicted, is readily discerned in the
1H NMR spectrum. Correlation experiments indicate the
signal at d ˆ 3.55 corresponds to H5. This signal also shows
vicinal coupling to the OH proton.
OAc
O
HF/C5H5N
33
HO
O
OAc
OAc
(10)
OAc
OAc
CSA
HO
OAc
34
The first-generation KAT synthesis provided the natural
(‡)-cyclophellitol in eight steps and 26% overall yield from
racemic conduritol B tetraacetate. The unnatural enantiomer
( )-cyclophellitol is equally accessible, only requiring a
change in the absolute configuration of the ligand in [Eq. (2)].
By modifying the ester of racemic conduritol B, a dynamic
kinetic asymmetric transformation also becomes equally
available. With a carbon nucleophile and a trichloroethyl
carbonate leaving group, the correct balance of the effective-
ness of the leaving group to permit both a ªmatchedº and
ªmismatchedº ionization of the racemic substrate and to
prevent further reaction of the enantiomerically enriched
monoalkylation product is achieved. Kinetic resolutions and
KAT processes are also possible with the Troc-tetraester 11
from which unreacted Troc tetraester of high ee can be
recovered. Thus, DYKAT is not due to equilibration of the
starting tetraester. Its utility is highlighted by the first
asymmetric synthesis of ( )-cyclophellitol in five steps in
27% overall yield from racemic tetracarbonate 21. Obviously,
the (‡)-enantiomer is equally accessible by simply changing
the absolute configuration of the ligand. The unmasking of the
phenylsulfonylnitromethane as a carboxy synthon with di-
methyldioxirane is noteworthy, especially in light of the
failure of our other mild protocols. It is also the first
illustration of the ability of the rhenium-catalyzed epoxida-
tion to be stereochemically directed by a homoallyl alco-
CSA, PhCH3, 100oC
tC4H9
O
(CH3)2C(OCH3)2
89%
O
HO
O
O
tC4H9
O
O
30
HO
OH
(7)
HO
tC4H9
O
13
72%
O
TBDMSO
OTBDMS
TBDMS-Cl
HO
N
NH
31
THF, DMF, 0oC
A novel pivalate-like carboxylic acid 32 was designed to
allow reverse order of ester cleavage. Its synthesis involves
quenching the dianion derived from isobutyric acid with
3772
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3772 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 17

KAT/DYKAT processes
3768±3775
hol.[22±25] These methods should prove useful not only in
providing access to this important class of glycosidase
inhibitors but more generally to highly functionalized chiral
cyclohexane derivatives.
128.4, 128.3, 128.2, 128.1, 127.9, 127.8, 127.7, 127.6, 85.1, 80.8, 78.5, 75.3,
1
75.1, 72.1, 63.3, 45.7; IR (neat): nÄ ˆ 3442, 3030, 2878, 1454, 1359 cm
;
elemental analysis calcd (%) for C28H30O4: C 78.11, H 7.02; found: C 78.33,
H 7.15.
(‡)-2,3,4-Tri-O-benzylcyclophellitol (18): MCPBA (76 mg, 0.44 mmol) was
added to a cooled (08C) solution of the alkene 17 (95 mg, 0.22 mmol) in
CH2Cl2 (2 mL). The mixture was warmed to room temperature. After 16 h,
the mixture was diluted with ethyl acetate (20 mL) and washed with 1m
NaOH (3 Â 10 mL) and brine (10 mL). The aqueous layer was back
extracted with ethyl acetate (10 mL). The combined organic layers were
dried (MgSO4) and concentrated in vacuo to give a crystalline white solid.
Chromatography on silica gel (1:2 petroleum ether/diethyl ether) provided
the desired epoxide 18 (77 mg, 78%) followed by the diastereomeric
epoxide (7 mg, 7%). (Both epoxides can be recrystallized from diethyl
Experimental Section
General methods: See ref. [26].
(
)-(1R,2S,3S,4R)-1-Pivaloxy-2,3,4-triacetoxy-5-cyclohexene (11) and (‡)-
(1S,2R,3R,4S)-1,2,3,4-tetraacetoxy-5-cyclohexene (8): A mixture of (Æ)-
1,2,3,4-tetra-O-acetylconduritol (8, 1.57 g, 5.0 mmol), pivalic acid
B
ether/hexanes.) Data for 18: m.p. 1018C; (lit.: [3k] 92 ± 938C); [a]D23
ˆ
(0.408 g, 4.0 mmol), tetrahexylammonium bromide (0.435 g, 1.0 mmol),
10 (19 mg, 0.05 mmol, 1 mol%), and ligand (R,R)-9 (105 mg, 0.15 mmol,
3 mol%) was degassed by evacuating and purging with argon (3 Â ).
Freshly distilled CH2Cl2 (10 mL) was added through an air-tight syringe,
followed by a 0.5m NaOH solution (6.5 mL), which had been previously
degassed by purging with argon. The resulting biphasic mixture was stirred
vigorously at room temperature. After 16 h, the mixture was washed with,
20% aq. K2CO3 (10 mL) and the aqueous layer extracted with diethyl ether
(2 Â 10 mL), the combined organic layers dried (MgSO4), and then filtered
through a short plug of silica gel to remove the catalyst. The silica gel was
rinsed with diethyl ether (20 mL) and the filtrate was concentrated in
vacuo. Chromatography on silica gel (3:1 to 1:1 petroleum ether/diethyl
ether gradient) provided the dipivalate (19 mg, 1%), followed by
monopivalate 11 (776 mg, 44%), followed by tetraacetate (‡)-8 (781 mg,
‡71.78 (c ˆ 1.62, CHCl3), {lit.:[3k] ‡71.08 (c ˆ 0.9, CHCl3)}; 1H NMR
(CDCl3, 300 MHz): d ˆ 7.4 ± 7.2 (m, 15H), 4.93 (d, J ˆ 10.7 Hz, 1H), 4.86
(ABq, J ˆ 13.1 Hz, DnÄ ˆ 15.6 Hz, 2H), 4.77 (ABq, J ˆ 11.2 Hz, DnÄ ˆ
23.2 Hz, 2H), 4.54 (d, J ˆ 10.9 Hz, 1H), 3.96 (dt, J ˆ 11.0, 3.8 Hz, 1H),
3.85 (m, 2H), 3.59 (dd, J ˆ 10.0, 8.1 Hz, 1H), 3.47 (t, J ˆ 10.0 Hz, 1H), 3.34
(d, J ˆ 3.4 Hz, 1H), 3.17 (d, J ˆ 3.9 Hz, 1H), 2.18 (m, 1H), 1.94 (dd, J ˆ 8.1,
3.2 Hz, 1H); 13C NMR (CDCl3, 75 MHz): d ˆ 138.5, 138.0, 137.6, 128.5,
128.3, 128.1, 127.9, 127.8, 127.6, 84.9, 79.8, 75.5, 75.3, 73.1, 62.7, 55.9, 2.9, 43.9;
IR (neat): nÄ ˆ 3314, 3030, 2891, 1454, 1359, 1064 cm 1. These data are
consistent with those reported in the literature.[21] Data for the diastereo-
meric epoxide: m.p. 135 ± 1378C; [a]2D3 ˆ ‡75.78 (c ˆ 0.17, CHCl3); 1H NMR
(CDCl3, 300 MHz): d ˆ 7.4 ± 7.2 (m, 15H), 4.94 ± 4.78 (m, 5H), 4.58 (d,
J ˆ 11.5 Hz, 1H), 3.88 (dd, J ˆ 8.5, 1.7 Hz, 1H), 3.8 ± 3.6 (m, 3H), 3.40 (t,
J ˆ 10.0 Hz, 1H), 3.31 (m, 1H), 3.13 (d, J ˆ 3.9 Hz, 1H), 2.13 (dt, J ˆ 4.1,
10.0 Hz, 1H), 1.11 (t, J ˆ 4.9 Hz, 1H); 13C NMR (CDCl3, 75 MHz):
d ˆ 138.6, 138.3, 138.2, 128.5, 128.4, 128.3, 128.2, 128.0, 127.9, 127.8, 127.7,
50%, 83% ee). Data for 11: m.p. 1108C; [a]D23
ˆ
163.08 (c ˆ 1.11, CHCl3);
1H NMR (CDCl3, 300 MHz): d ˆ 5.68 (s, 2H), 5.57 (m, 2H), 5.36 (m, 2H),
2.05 (s, 3H), 2.03 (s, 3H), 2.01 (s, 3H), 1.16 (s, 9H); 13C NMR (CDCl3,
75 MHz): d ˆ 177.7, 170.2, 169.9, 169.6, 127.7, 127.2, 71.6, 71.4, 71.2, 71.0, 38.7,
26.9, 20.8, 20.6, 20.5; IR (neat): nÄ ˆ 2975, 1758, 1368, 1221 cm 1; elemental
analysis calcd (%) for C17H24O8: C 57.30, H 6.79; found: C 57.40, H 7.00;
enantiomeric excess determined by HPLC (Chiralcel AD column, eluting
with 95:5 heptane/iPrOH, 1 mLmin 1: (‡)-enantiomer tR ˆ 7.69 min, ( )-
enantiomer tR ˆ 8.48 min).
127.6, 82.3, 79.8, 75.8, 75.0, 72.8, 61.7, 54.9, 54.4, 44.1; IR (neat): nÄ ˆ 3356,
1
2923, 1072 cm
.
(1S,2R,3S,4R,5R,6R)-5-Hydroxymethyl-7-oxa-bicyclo[4.1.0]heptane-2,3,4-
triol: (‡)-Cyclophellitol (1): Benzyl ether cleavage was carried out
according to the method of Ziegler and Wang.[3k] The epoxide 18 (50 mg,
0.11 mmol) was dissolved in methanol (3.3 mL) and palladium hydroxide
on carbon (Pearlmanꢁs catalyst, 25 mg) was added. The flask was evacuated
and purged with H2 three times, and then the mixture was stirred under an
atmosphere of H2 (balloon). After 12 h, the hydrogen was removed and the
solution was filtered through a plug of celite and rinsed with methanol
(10 mL). The filtrate was concentrated in vacuo to give a colorless oil.
Chromatography on silica gel (10:1 to 3:1 CH2Cl2/MeOH gradient)
provided (‡)-cyclophellitol (1, 18 mg, 91%) as a white crystalline solid.
M.p. 1458C (lit.:[1, 3b] 149 ± 1518C); [a]D23 ˆ ‡1028 (c ˆ 0.70, H2O), {lit.:
(
)-(1R,2S,3S,4R)-2,3,4-Tribenzyloxy-1-(tributylstannyl)methoxy-5-cyclo-
hexene (16): Potassium hydride (120 mg, 30% suspension in mineral oil,
0.89 mmol followed by removal of mineral oil) was suspended in
dimethoxyethane (DME) (2.3 mL) at 08C. The alcohol 15 (190 mg,
0.456 mmol) was added slowly as a solution in DME (1 mL) and the
mixture was warmed to room temperature. After 30 min, the mixture was
cooled back to 08C and iodomethyltributyltin (287 mg, 0.67 mmol) was
added. The mixture was warmed to room temperature. After 45 min, aq.
ammonium chloride (10 mL) was added and the mixture was extracted with
diethyl ether (2 Â 25 mL). The organic layers were washed with brine
(10 mL), dried (MgSO4) and concentrated in vacuo. Chromatography on
silica gel (20:1 to 10:1 petroleum ether/diethyl ether gradient) provided the
[a]2D7 ˆ ‡1038 (c ˆ 0.5, H2O),[3k] [a]D24 ˆ ‡1038 (c ˆ 0.40, H2O),[3b] [a]2D0
ˆ
‡978 (c ˆ 0.35, H2O)[3k]}; 1H NMR (CD3OD, 300 MHz): d ˆ 4.00 (dd, J ˆ
10.7, 4.4 Hz, 1H), 3.67 (m, 2H), 3.41 (m, 1H), 3.20 (dd, J ˆ 10.5, 8.3 Hz,
1H), 3.06 (m, 2H), 1.97 (m, 1H); 13C NMR (CD3OD, 75 MHz): d ˆ 78.5,
1
72.8, 68.8, 62.4, 57.4, 56.0, 45.9; IR (neat): nÄ ˆ 3360, 2895, 1641, 1081
.
tributylstannylmethylether 16 (288 mg, 88%) as a colorless oil. [a]D23
ˆ
These data are consistent with those reported in the literature.[1, 3k]
83.98 (c ˆ 1.32, CHCl3); 1H NMR (CDCl3, 300 MHz): d ˆ 7.4 ± 7.2 (m,
15H), 5.77 (m, 2H), 5.0 ± 4.8 (m, 4H), 4.71 (ABq, J ˆ 11.5 Hz, DnÄ ˆ 16.4 Hz,
2H), 4.21 (m, 1H), 3.87 (m, 2H), 3.8 ± 3.6 (m, 3H), 1.6 ± 1.4 (m, 6H), 1.4 ±
1.2 (m, 6H), 0.88 (m, 14H); 13C NMR (CDCl3, 75 MHz): d ˆ 138.9, 138.8,
138.4, 128.4, 128.3, 128.3, 128.0, 127.8, 127.7, 127.5, 127.4, 127.3, 85.1, 83.5,
(3R)-(1'-Benzenesulfonyl-1'-nitromethyl)-(4R,5S,6S)-tris-(2,2,2-trichloro-
ethoxycarbonyloxy)-cyclohexene (23): A slurry of the tetracarbonate 21
(4.0 g, 4.72 mmol), the sodium salt of phenylsulfonyl nitromethane 22
(1.13 g, 5.08 mmol), [dba3Pd3] ´ CHCl3 (120 mg, 0.115 mmol) and ligand
(R,R)-9 (240 mg, 0.348 mmol), Cs2CO3 in Eq. (2) in THF (13 mL) was
stirred for 2.5 h. The reaction was concentrated in vacuo and column
chromatography on silica gel (20 ± 50% EtOAc/petroleum ether) gave a
mixture of diastereomers of 23 as a white solid (3.29 mg, 79%). M.p. 103 ±
1058C (EtOAc/petroleum ether); 1H NMR (300 MHz, CDCl3): mixture of
diastereomers d ˆ 8.0 ± 7.4 (m, 5H), 6.45 (d, J ˆ 10.5, 0.5 Hz), 6.33 (d, J ˆ
10.5, 0.5 Hz), 6.1 ± 5.8 (m, 1.5H), 5.7 ± 5.5 (m, 2H), 5.42 (m, 1H), 5.25 (t, J ˆ
9.8, 0.5H), 5.0 ± 4.6 (m, 6H), 4.0 ± 3.8 (m, 1H); 13C NMR (75.5 MHz,
CDCl3): mixture of diastereomers d ˆ 153.5, 153.2, 136.5, 135.9, 135.6,
135.5, 129.9, 129.6, 129.3, 127.3, 126.1, 124.7, 124.0, 97.2, 94.0, 76.2, 75.4, 75.3,
74.5, 73.2, 43.8, 41.3; IR (neat): nÄ ˆ 2965, 1770, 1567, 1450, 1385, 1347,
1238 cm 1; elemental analysis calcd (%) for C22H18Cl9NO13S: C 30.89, H
2.12, N 1.64; found: C 30.63, H 2.33, N 1.74.
83.1, 80.1, 75.6, 75.3, 72.5, 60.0, 29.1, 27.3, 13.7, 8.9; IR (neat): nÄ ˆ 2924, 1454,
1
1069 cm
.
(‡)-(1S,2R,3R,4R)-4-Hydroxymethyl-1,2,3-tribenzyloxy-5-cyclohexene
(17): n-Butyllithium (0.70 mL of a 1.6m solution in hexanes, 1.13 mmol) was
added to a cooled ( 788C) solution of the stannylmethyl ether 16 (270 mg,
0.375 mmol) in THF (7.5 mL). After 2 h, the reaction was quenched with
sat. aq. ammonium chloride (10 mL), warmed to room temperature, and
extracted with diethyl ether (3 Â 20 mL). The organic layers were washed
with brine (10 mL), dried (MgSO4), and concentrated in vacuo. Chroma-
tography on silica gel (2:1 to 1:1 petroleum ether/diethyl ether gradient)
provided the homoallylic alcohol 17 as a colorless oil (117 mg, 72%).
[a]2D3 ˆ ‡104.58 (c ˆ 1.92, CHCl3); 1H NMR (CDCl3, 300 MHz): d ˆ 7.4 ± 7.2
(m, 15H), 5.76 (dt, J ˆ 10.0, 2.5 Hz, 1H), 5.55 (dt, J ˆ 10.0, 1.8 Hz, 1H), 5.01
(d, J ˆ 11.2 Hz, 1H), 4.94 (s, 2H), 4.70 (s, 2H), 4.66 (d, J ˆ 11.2 Hz, 1H),
4.25 (m, 1H), 3.85 (dd, J ˆ 10.0, 7.8 Hz, 1H), 3.66 (m, 3H), 2.49 (m, 1H),
1.47 (t, J ˆ 5.6 Hz, 1H); 13C NMR (CDCl3, 75 MHz): d ˆ 138.7, 138.2, 128.5,
(1R,4R,5S,6S)-4,5,6-Tris-(2,2,2-trichloro-ethoxycarbonyloxy)-cyclohex-2-
enecarboxylic acid (25): Tetramethylguanidine (0.11 mL, 0.702 mmol) was
added at 08C to a solution of nitrosulfone 23 (500 mg, 0.585 mmol) in
Chem. Eur. J. 2001, 7, No. 17
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3773 $ 17.50+.50/0
3773

B. M. Trost et al.
FULL PAPER
methylene chloride (2 mL) and the resulting solution was stirred for 10 min
at 08C. Dimethyldioxirane (ꢀ0.07 ± 0.1m soln. in acetone) was added in
7 mL portions (21 mL total) until the reaction was complete by TLC
analysis. The reaction mixture was concentrated and filtered through a plug
K. S. E. Tanaka, G. C. Winters, R. J. Batchelor, F. W. B. Einstein, A. S.
Bennet, J. Am. Chem. Soc. 2001, 123, 998.
[3] a) T. Akiyama, H. Shima, M. Ohnari, T. Okazaki, S. Ozaki, Bull.
Chem. Soc. Jpn. 1993, 66, 3760; b) T. Akiyama, M. Ohnari, H. Shima,
S. Ozaki, Syntlett 1991, 831 ± 83; c) T. K. M. Shing, V. W. F. Tai, J.
Chem. Soc. Chem. Commun. 1993, 995; d) T. K. M. Shing, V. W. F. Tai,
J. Chem. Soc. Perkin Trans. 1 1994, 2017; e) K. I. Sato, M. Bokura, H.
Moriyama, T. Igarashi, Chem. Lett. 1994, 37; f) K. Tatsuta, Y. Niwata,
K. Umezawa, K. Toshima, M. Nakata, Carbohydrate Res. 1991, 222,
189; g) K. Tatsuta, Y. Niwata, K. Umezawa, K. Toshima, M. Nakata,
Tetrahedron Lett. 1990, 31, 1171; h) K. Tatsuta, Y. Niwata, K.
Umezawa, K. Toshima, M. Nakata, Tetrahedron Lett. 1990, 31, 1171;
i) P. Letelier, R. Ralainairina, B. Beaupere, R. Uzan, Synthesis 1997,
925; j) M. E. Jung, S. W. T. Choe, J. Org. Chem. 1995, 60, 3280; k) F. E.
Ziegler, Y. Wang, J. Org. Chem. 1998, 63, 426; F. E. Ziegler, Y. Wang,
J. Org. Chem. 1998, 63, 7920; l) R. E. McDevitt, B. Fraser-Reid, J. Org.
Chem. 1994, 59, 3250; m) H. Takahashi, T. Iimori, S. Ikegami,
Tetrahedron Lett. 2000, 39, 6939.
[4] R. H. Schlessinger, C. P. Bergstrom, J. Org. Chem. 1995, 60, 16.
[5] For a preliminary report of a portion of this work, see: B. M. Trost,
E. J. Hembre, Tetrahedron Lett. 1999, 40, 219.
[6] Z.-X. Guo, A. H. Haines, S. M. Pyke, S. G. Pyke, R. J. K. Taylor,
Carbohydr. Res. 1994, 264, 147.
[7] For reviews, see: M. Balci, Y. Sütbeyaz, H. Secen, Tetrahedron 1990,
46, 3715; T. Suami, S. Ogawa, Adv. Carbohydr. Chem. Biochem. 1990,
48, 21; J. J. Kiddle, Chem. Rev. 1995, 95, 2189.
[8] For a few recent examples, see: O. Arjona, A. de Dios, J. Plument, B.
Saez, Tetrahedron Lett. 1995, 36, 1319; J. E. Innes, P. J. Edwards, S. V.
Ley, J. Chem. Soc. Perkin Trans. 1 1997, 795; C. Sanfilippo, A. Patti, G.
Nicolosi, Tetrahedron: Asymmetry 1999, 10, 3273; M. Honzumi, K.
Hiroya, T. Taniguchi, K. Ogasawara, Chem. Commun. 1999, 1985; O.
Plettenburg, S. Adelt, G. Vogel, H. J. Altenbach, Tetrahedron:
Asymmetry 2000, 11, 1057; K. Sato, S. Sakuma, S. Muramatsu, M.
Bokura, Chem. Lett. 1991, 147; N. Chida, M. Ohtsuka, K. Nakazawa, S.
Ogawa, J. Org. Chem. 1991, 56, 2976; T. K. Park, S. J. Danishefsky,
Tetrahedron Lett. 1994, 35, 2667; S. Amano, N. Ogawa, M. Ohtsuda, N.
Chida, Tetrahedron 1999, 55, 2205.
[9] B. M. Trost, D. L. Van Vranken, Chem. Rev. 1996, 96, 395; B. M. Trost,
C.-B. Lee, in Catalytic Asymmetric Synthesis (Ed.: I. Ojima), Wiley-
VCH, New York, 2000, Chapter 8E, pp. 593 ± 650.
[10] B. M. Trost, D. E. Patterson, E. J. Hembre, J. Am. Chem. Soc. 1999,
121, 10834.
of silica gel to give 25 as a white foam (318 mg, 78%). [a]D25
ˆ
74.98 (c ˆ
1.75, CH2Cl2); 1H NMR (300 MHz, CDCl3): d ˆ 5.97 (m, 1H), 5.87 (m, 1H),
5.56 (m, 2H), 5.36 (m, 1H), 4.77 (m, 6H), 3.70 (m, 1H); 13C NMR
(75.5 MHz, CDCl3): d ˆ 177.1, 173.9, 153.2, 153.0, 125.9, 125.2, 94.1, 94.0,
76.2, 75.6, 72.8, 60.5, 47.3, 20.7; IR (neat): nÄ ˆ 3320, 2962, 1776, 1718, 1381,
1289, 1240 cm 1; elemental analysis calcd (%) for C16H13Cl9O11: C 27.44, H
1.87; found: C 27.58, H 1.79.
(1R,2S,3R,4S,5S,6S)-5-Hydroxymethyl-2,3,4-tris-(2,2,2-trichloro-ethoxy-
carbonyloxy)-7-oxa-bicyclo[4.1.0]heptane (28) and (1S,2S,3R,4S,5S,6R)-5-
Hydroxymethyl-2,3,4-tris-(2,2,2-trichloro-ethoxycarbonyloxy)-7-oxa-bicy-
clo[4.1.0]heptane (29): 30% Aqueous hydrogen peroxide (0.02 mL) was
added to a solution of 26 (93.5 mg, 0.136 mmol) and methyltrioxorhenium
(3.2 mg, 0.0129 mmol) in dichloroethane (0.15 mL). The resulting yellow
solution was stirred for 40 h at room temperature. The reaction was diluted
with methylene chloride (15 mL) and was washed with water (5 mL). The
organic layer was dried (Na2SO4) and concentrated in vacuo to give a
colorless oil (crude 3.5:1 mixture of 28:29 by NMR analysis). Column
chromatography on silica gel (33% EtOAc/petroleum ether) gave 28 as a
white foam (62 mg, 65%). [a]D25
ˆ
36.38 (c ˆ 2.2, CH2Cl2); 1H NMR
(300 MHz, CDCl3): d ˆ 5.24 (td, J ˆ 8.3, 1.7 Hz, 1H), 5.15 (d, J ˆ 8.1 Hz,
1H), 5.02 (t, J ˆ 10.0 Hz, 1H), 4.80 (m, 6H), 3.91 (m, 2H), 3.57 (brs, 1H),
3.30 (m, 1H), 2.49 (t, J ˆ 4.7 Hz, 1H); 13C NMR (75.5 MHz, CDCl3): d ˆ
153.8, 153.4, 153.0, 94.1, 93.5, 77.1, 76.7, 75.2, 71.7, 60.8, 55.4, 52.0, 42.0; IR
(neat): nÄ ˆ 3540, 2961, 1771, 1378, 1290, 1239 cm 1; elemental analysis calcd
(%) for C16H13Cl9O11: C 27.36, H 2.15; found: C 27.54, H 2.34. Further
chromatography gave a 7:1 mixture of 29:28 as a white foam (14 mg, 15%).
1
Spectral data for 29: H NMR (300 MHz, CDCl3): d ˆ 5.33 (m, 2H), 5.11
(m, 1H), 4.79 (m, 6H), 3.92 (d, J ˆ 11.5 Hz, 1H), 3.81 (dd, J ˆ 3.2, 11.5 Hz,
1H), 3.64 (d, J ˆ 3.9 Hz, 1H), 3.36 (d, J ˆ 3.9 Hz, 1H), 2.49 (dd, J ˆ 2.9,
9.5 Hz, 1H); 13C NMR (75.5 MHz, CDCl3): d ˆ 154.3, 153.4, 153.1, 94.1,
77.1, 74.5, 73.3, 60.3, 54.3, 53.5, 42.9; IR (neat): nÄ ˆ 3530, 2962, 1772, 1376,
1
1289, 1240 cm
.
(1R,2S,3R,4S,5S,6S)-5-Hydroxymethyl-7-oxa-bicyclo[4.1.0]heptane-2,3,4-
triol: ( )-Cyclophellitol (1): K2CO3 (3 mg, 0.02 mmol) was added to a
solution of 28 (42 mg, 0.060 mmol) in methanol (0.3 mL). The reaction was
stirred at room temperature for 3 h and concentrated in vacuo. The crude
residue was diluted with methylene chloride (0.5 mL), and filtered through
silica gel (20% methanol/methylene choride) to give ( )-1 as a white solid
[11] B. M. Trost, Acc. Chem. Res. 1996, 29, 355.
(10.1 mg, 95%). [a]D25
water: m.p. 142 ± 1448C; [a]D23
ˆ
89.88 (c ˆ 0.5, H2O). After recrystallization from
[12] W. C. Still, A. Mitra, J. Am. Chem. Soc. 1978, 100, 1927.
[13] For the synthesis of (phenylsulfonyl)nitromethane see, P. A. Wade,
H. R. Hinney, N. V. Amin, P. D. Vail, S. D. Morrow, S. A. Hardinger,
M. S. Saft, J. Org. Chem. 1981, 46, 765.
[14] a) B. M. Trost, R. Madsen, S. D. Guile, B. Brown, J. Am. Chem. Soc.
2000, 122, 5947; b) B. M. Trost, D. Stenkamp, S. R. Pulley, Chem. Eur.
J. 1995, 1, 568; c) B. M. Trost, R. Madsen, S. G. Guile, A. E. H. Elia,
Angew. Chem. 1996, 108, 1666; Angew. Chem. Int. Ed. Engl. 1996, 35,
1569; d) B. M. Trost, R. Madsen, S. G. Guile, Tetrahedron Lett. 1997,
38, 1707.
ˆ
100.78 (c ˆ 0.32, H2O); 1H NMR
(300 MHz, CDCl3): d ˆ 3.90 (dd, J ˆ 10.4, 4.2 Hz, 1H), 3.54 (m, 2H), 3.30
(d, J ˆ 3.5 Hz, 1H), 3.09 (m, 1H), 2.94 (m, 2H), 1.85 (tdd, J ˆ 9.5, 4.6,
1.8 Hz, 1H); 13C NMR (75.5 MHz, CDCl3): d ˆ 78.5, 72.8, 68.8, 62.5, 57.4,
56.0, 45.9; IR (neat): nÄ ˆ 3362, 2921, 1642, 1426 cm 1; HRMS: calcd for
‡
C7H11O4 [M OH] : 159.0657; found: 159.0665.
[15] B. M. Trost, R. Brauslau, J. Org. Chem. 1988, 53, 532.
[16] P. Ceccherelli, M. Curini, M. C. Marcotullio, F. Epifano, O. Rosati,
Synth. Commun. 1998, 28, 3057.
Acknowledgement
We thank the National Science Foundation and the National Institutes of
Health for their generous support of our programs. D.E.P. was supported in
part by a fellowship from Pharmacia and Upjohn. E.J.H. was supported by
a NIH postdoctoral fellowship. Mass spectra were provided by the Mass
Spectrometry Facility at the University of California-San Francisco
supported by the NIH Division of Research Resources.
[17] G. A. Olah, B. G. B. Gupta, Synthesis 1980, 44.
[18] J. E. McMurry, J. Melton, J. Org. Chem. 1973, 38, 4367.
[19] W. Adam, M. Makosza, C. R. Saha-Moller, C. G. Zhao, Synlett 1998,
1335.
[20] a) W. H. Rastetter, T. J. Richard, M. D. Lewis, J. Org. Chem. 1978, 43,
3163; b) P. A. Bartlett, L. A. McQuaid, J. Am. Chem. Soc. 1984, 106,
7854.
[21] a) K. B. Sharpless, R. C.Michaelson, J. Am. Chem. Soc. 1973, 95, 6136;
b) M. Bailey, I. Staton, P. R. Ashton, I. E. Marko, W. D. Ollis,
Tetrahedron: Asymmetry 1991, 2, 495.
[1] S. Atsumi, K. Umezawa, H. Iinuma, H. Naganawa, H. Nakamura, Y.
Iitaka, T. Takeuchi, J. Antibiot. 1990, 43, 49.
[22] For reviews on the MTOR-catalyzed epoxidation, see a) W. A.
Herrmann, R. M. Kratzer, H. Ding, W. R. Theil, H. Glas, J. Organo-
met. Chem. 1998, 555, 293; b) A. Gansäuer, Angew. Chem. 1997, 109,
2701; Angew. Chem. Int. Ed. Engl. 1997, 36, 2591.
[23] W. Adam, C. M. Mitchell, Angew. Chem. 1996, 108, 578; Angew.
Chem. Int. Ed. Engl. 1996, 35, 533.
[2] S. Atsumi, H. Iinuma, C. Nosaka, K. Umezawa, J. Antibiot 1990, 43,
1579. See also: Y. Nishimura, Stud. Nat. Prod. Chem. 1997, 19
(Structure and Chemistry, Part E), 351. For a lead reference in
glycosidase inhibitors, see: N. Asano, R. J. Nash, R. S. Molyneux,
G. W. J. Fleet, Tetrahedron: Asymmetry 2000, 11, 1645. For a recent
paper on novel design of carbocyclic glycosidase inhibitors, see:
3774
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3774 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 17

KAT/DYKAT processes
3768±3775
[24] a) J. Rudolph, K. L. Reddy, J. P. Chiang, K. B. Sharpless, J. Am. Chem.
Soc. 1997, 119, 6189; b) C. Coperet, H. Adolfson, K. B. Sharpless,
Chem. Commun. 1997, 1566; c) M. L. A. van Vliet, I. W. C. E. Arends,
R. A. Sheldon, Chem. Commun. 1999, 821; d) H. Adolfsson, A.
Converso, K. B. Sharpless, Tetrahedron Lett. 1999, 41, 3991.
[25] For the case of an allyl alcohol, see: T. R. Boehlow, C. D. Spilling,
Tetrahedron Lett. 1996, 37, 2717.
[26] B. M. Trost, J. Dudash, Jr, E. J. Hembre, Chem. Eur. J. 2001, 7, 1619.
Received: January 17, 2001 [F3005]
Chem. Eur. J. 2001, 7, No. 17
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3775 $ 17.50+.50/0
3775