Biocatalytic asymmetric rearrangement of a methylene-interrupted bis-epoxide: Simultaneous control of four asymmetric centers through a biomimetic reaction cascade

Article abstract of DOI:10.1002/chem.200400061

Asymmetric enzyme-catalyzed hydrolysis of methylene-interrupted bis-epoxides 1a and 1b catalyzed by bacterial epoxide hydrolases furnished tetrahydrofuran derivatives 2a and 2b through a hydrolysis-rearrangement cascade. Whereas racemic bis-oxiranes 1b-d underwent kinetic resolution with moderate stereoselectivities to yield products with up to 92% ee and 66% de: meso-bis-oxirane cis,cis-1a was transformed into (6R,7R,9S,10S)-2a in 94% ee and 89% de at high conversion (85%) by Rhodococcus sp. CBS 717.73 as the major product. The reaction sequence resembles a biomimetic reaction cascade and provides an efficient entry into the structural core of annonaceous acetogenins with simultaneous control of four stereocenters.

Full text of DOI:10.1002/chem.200400061

FULL PAPER
Biocatalytic Asymmetric Rearrangement of a Methylene-Interrupted
Bis-epoxide: Simultaneous Control of Four Asymmetric Centers Through
a Biomimetic Reaction Cascade
[a]
Silvia M.Glueck, Walter M.F.Fabian, Kurt Faber,* and Sandra F.Mayer
Abstract: Asymmetric enzyme-cata-
lyzed hydrolysis of methylene-inter-
rupted bis-epoxides 1a and 1b cata-
lyzed by bacterial epoxide hydrolases
furnished tetrahydrofuran derivatives
2a and 2b through a hydrolysis rear-
rangement cascade. Whereas racemic
bis-oxiranes 1b d underwent kinetic
resolution with moderate stereoselec-
tivities to yield products with up to
92% ee and 66% de: meso-bis-oxirane
(6R,7R,9S,10S)-2a in 94% ee and
89% de at high conversion (85%) by
Rhodococcus sp. CBS 717.73 as the
major product. The reaction sequence
resembles a biomimetic reaction cas-
cade and provides an efficient entry
into the structural core of annonaceous
acetogenins with simultaneous control
of four stereocenters.
cis,cis-1a
was
transformed
into
Keywords: annonaceous acetoge-
nins ¥ biomimetic synthesis ¥ cycliza-
tion ¥ epoxide hydrolase ¥ epoxides ¥
rearrangement
Introduction
from double-bond-containing acetogenins.[5,6] Most recently,
the discovery of muridienins (presumed precursors of mono-
THF acetogenins) and chatenaytrienins (precursors for bis-
THF analogues) has provided evidence that acetogenins are
most probably derived from lacceroic (C32) and ghedoic
(C34) fatty acids after enzymatic combination with a three-
carbon unit.[7]
The complex structure and the potent and diverse bioac-
tivity of annonaceous acetogenins has prompted numerous
efforts directed towards their total synthesis.[8] To achieve an
economic sequence involving a minimum of protection
groups, the occurrence of up to three THF units has suggest-
ed the use of reiterative methodology. It is interesting to
note that the majority of synthetic strategies closely fol-
lowed the (presumed) biosynthetic pathways: thus, an open-
chain olefin bearing double bonds in the appropriate posi-
tions and a terminal nucleophile (usually OH) was epoxi-
Annonaceous acetogenins are a subgroup of natural poly-
ether products possessing an (oligo)-THF structure with di-
verse stereochemistry and are derived from the Annonaceae
(custard-apple) plant family.[1] After the discovery of its first
representative–the antileukemic agent uvaricin[2]–in 1982,
they have attracted much interest on account of their di-
verse bioactivities, such as anthelmintic, antitumor, antima-
larial, antimicrobial, antiprotozoal, and pesticidal activities.
These last effects are generally exerted through suppression
of ATP-driven resistance mechanisms, which in turn causes
apoptosis (programmed cell death). For these reasons, anno-
naceous acetogenins became promising new chemotypes for
the development of antitumor and pesticidal agents.
Biogenetically, annonaceous acetogenins are derived from
the polyketide pathway.[3] The hypothesis that they are
formed from (poly)unsaturated fatty acids by enzymatic ep-
oxidation of the olefinic moieties followed by cascade cycli-
zation of the epoxy fatty acids thus obtained[4] is supported
by the identification of matching precursors, the location of
double bonds in the appropriate positions within the fatty
acid chain, and by the semisynthesis of additional THF rings
dized to the corresponding oligo-epoxy derivative, which
15,27]
was subjected to metal-catalyzed,[9] acid-catalyzed,[10
or
base-catalyzed[16] cascade cyclization. This approach proved
to be highly efficient as the stereochemical outcome of the
cyclization cascade could be controlled to a high extent by
the relative (or absolute) stereochemistry of the epoxy
units.[17] Whenever access to nonracemic material was de-
sired, asymmetry was usually introduced at the epoxide-
stage intermediate by means of Sharpless asymmetric epoxi-
dation or -dihydroxylation protocols.[16,14,9] To the best of
our knowledge, no asymmetric catalytic variant of an epox-
ide cyclization cascade has been reported to date.
[a]S. M. Glueck, Prof. W. M. F. Fabian, Prof. K. Faber, Dr. S. F. Mayer
Department of Chemistry, Organic & Bio-Organic Chemistry
University of Graz, Heinrichstrasse 28, 8010 Graz (Austria)
Fax : (+43)316-380-9840
E-mail: Kurt.Faber@uni-graz.at
Chem. Eur. J. 2004, 10, 3467 3478
DOI: 10.1002/chem.200400061
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3467

FULL PAPER
Results and Discussion
The overall cascade is composed of eight possible stereo-
chemical pathways to furnish four (theoretically) possible
stereoisomeric THF derivatives as final products (2a,b).[26]
In order to gain insight into the stereochemistry of the
spontaneous (nonenzymatic) cyclization reaction, the key
parameters, such as the Gibbs free energy (DGreact), deproto-
nation enthalpies (DDG), and
Recently, we discovered an enzyme-triggered cyclization
cascade of haloalkyl-oxiranes catalyzed by bacterial epoxide
hydrolases (Scheme 1).[18,19] Thus, asymmetric enzyme-cata-
lyzed hydrolysis of halomethyl oxiranes (n=1) by bacterial
activation energies (DDG°),
were calculated by density func-
tional
theory
[PB-SCRF-
B3LYP/6-311++g(d,p)//B3LYP/
+6-31 g(d)]on the basis of a
structurally simplified analogue
of epoxydiol 1e in which both
n-pentyl groups were ™reduced∫
to methyl substituents (for com-
putational details, see the Ex-
perimental Section). It was
found that the Gibbs free
energy DGreact for both cycliza-
tion pathways A and
B was
ꢀ17 kcalmolÀ1. However, sig-
nificant energetic discrimina-
tion was found between the de-
protonation energies of the al-
cohol groups (OH at C10
versus OH at C9, D=3 kcal-
molÀ1) in favor of the less hin-
dered OH group at C10. In ad-
Scheme 1. Stereochemical pathways of biocatalytic epoxide hydrolysis cyclization cascade reactions.
epoxide hydrolases furnished the corresponding vic-diols,
which spontaneously underwent an exo-tet-cyclization to
form hydroxyepoxides as the final products. In contrast, hal-
oethyl derivatives (n=2) gave the corresponding THF deriv-
atives. 5-exo-tet-Cyclization is favored over both 6- and 5-
endo-tet-cyclization in nucleophilic substitution reactions in-
volving sp3 centers. The restriction is loosened for ™sp2-like∫
centers; however, it still applies to epoxide ring opening.[20]
Because the enzymatic step proceeded in an enantiocon-
vergent fashion,[21] the final products were predominantly
obtained as single stereoisomers in high enantiomeric excess
(ee) and diastereomeric excess (de) bearing two asymmetric
centers. In order to demonstrate the synthetic elegance of
this strategy, several natural products were synthesized by
means of this strategy.[22]
Prompted by these encouraging results, we envisaged the
application of this methodology to a stereochemically even
more challenging task, that is, a biocatalytic hydrolysis cycli-
zation cascade of methylene-interrupted bis-epoxides. In
this case, the number of possible pathways is even more
complex (Scheme 1). Because both the enzyme-catalyzed
hydrolysis[23] and the spontaneous cyclization[24] follow a
™clean∫ SN2-type mechanism, four positions of enzyme
attack are possible in the first step, thus causing inversion of
the carbon atom attacked by the epoxide hydrolase[25] to fur-
nish four stereoisomeric vic-epoxydiols 1e as intermediates.
Each of these may undergo cyclization to the respective dia-
stereomeric THF product 2a,b following two different path-
ways (A, B as outlined in Scheme 1), again with chiral inver-
sion of the carbon atom involved in the cyclization.
dition, the difference in the Gibbs free energy of activation
of pathways A (DG° =23 kcalmolÀ1) versus (DG° =
B
32 kcalmolÀ1) proved to be high (DDG° ꢀ10 kcalmolÀ1). In
other words, the regioselectivity for pathway A versus B is
virtually ™absolute∫, which is in agreement with Baldwin×s
rules for ring closure.[20,24] As may be deduced from the de
values in Table 3 (see later), the calculations are in good
agreement with the experimental data.
Bis-epoxides cis,cis,meso-1a and (Æ)-cis,cis,anti-1b were
synthesized as follows (Scheme 2): Lindlar hydrogenation of
alkynol 3 gave the allylic alcohol (Z)-4, which was chlorinat-
ed to the corresponding allyl chloride (Z)-5 under Appel
conditions. Compound 5 was then coupled to 1-heptyne to
furnish enyne (Z)-6. Epoxidation of the olefinic bond gave
epoxyalkyne (Æ)-cis-7. Stereoselective Lindlar reduction of
the acetylenic bond furnished epoxyalkene (Æ)-(9Z)-cis-8,
which was epoxidized to a (1:1.2) diastereomeric mixture of
methylene-interrupted bis-epoxides cis,cis,meso-1a and (Æ)-
cis,cis,anti-1b. Both diastereomers were separated by chro-
matography and stereochemically assigned by NMR spec-
troscopy.[27] In order to provide the complete stereochemical
set of substrates, the corresponding diastereomeric trans,cis,
anti- and trans,cis,syn-bis-epoxides (Æ)-trans,cis,anti-1c and
(Æ)-trans,cis,syn-1d were synthesized following the same
general strategy as outlined above, with the exception that
the required trans-olefins were obtained by stereoselective
reduction of the corresponding acetylenic bond of alkynol 3
with LiAlH4 (Scheme 3). Reference material for the expect-
ed diastereomeric biotransformation products (Æ)-2a to (Æ)-
2d was obtained in racemic form by acid-catalyzed hydrol-
3468
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 3467 3478

Biocatalytic Asymmetric Rearrangement
3467 3478
Scheme 2. Synthesis of substrates cis,cis,meso-1a and (Æ)-cis,cis,anti-1b.
Scheme 3. Asymmetric synthesis of reference material for 2c,d.
ysis rearrangement of bis-epoxides cis,cis,meso-1a, (Æ)-cis,
cis,anti-1b, (Æ)-trans,cis,anti-1c, and (Æ)-trans,cis,syn-1d
(see Schemes 2 and 3).[27]
ing THF products of type 2a,b, neither the intermediate
epoxy-diols 1e nor any other side products were detected in
measurable amounts.
To obtain rapid insight into the feasibility of the transfor-
mation, a diastereomeric (1:1.2) mixture of cis,cis,meso-1a
and (Æ)-cis,cis,anti-1b was tested in a screening for biohy-
drolytic activity in Tris buffer at pH 8.0 using resting cells of
a variety of 21 Actinomyces spp. known to possess strong
secondary metabolic activity and epoxide hydrolase activity,
in particular. The absence of any undesired spontaneous
nonenzymatic hydrolysis/cyclization reaction was verified
under standard conditions in the absence of biocatalyst
within the anticipated reaction time of ꢀ49 h. We were
pleased to see that the only products formed during
enzyme-catalyzed hydrolysis were the expected correspond-
The results of a careful stereochemical analysis of the
products and the remaining nonconverted oxirane 1b by co-
injection with independently synthesized reference material
of known relative and absolute configuration (see below) by
means of chiral GC are presented in Table 1.
Both diastereomeric substrates were converted by a series
of bacterial cells into the expected THF rearrangement
products through the anticipated cascade cyclization at vari-
ous rates. The stereochemical course of the sequence always
followed a double SN2 mechanism as expected (see above),
since diastereomers 2a and 2b were formed exclusively. No
trace of 2c or 2d was detected, because it would require
Chem. Eur. J. 2004, 10, 3467 3478
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3469

K. Faber et al.
FULL PAPER
Table 1. Enzyme-catalyzed hydrolysis of a mixture of cis,cis,meso-1a and (Æ)-cis,cis,anti-1b.
4) could be verified. Again, all
four possible stereoisomers,
that is, both enantiomers of 2a
[h]
Biocatalyst
c[a]
eeS (1b)
eeP (2a)
eeP (2b)
deS
deP
[%][%]
[%]
[%]
[%] [%]
1
2
3
4
5
6
7
8
9
Rhodococcus ruber DSM 44539
Rhodococcus ruber DSM 44541
Rhodococcus equi IFO 3730
Mycobacterium paraffinicum NCIMB 10420
Rhodococcus ruber DSM 43338
Rhodoccoccus sp. NCIMB 11216
Rhodococcus sp. CBS 717.73
46
14
76
62
15
35
70
8
47[b]
8[b]
77[e]
83[e]
83[d]
53[e]
46[e]
65[d]
73[e]
35[e]
58[d]
71[f]
67[f]
3[g]
52
55
99
91
74
49
84
66
46
24[i] and 2b, were formed in varying
1[j]
amounts. Whereas diastereomer
2a was generally formed with
84[c]
58[c]
48[b]
48[b]
86[b]
13[c]
13[c]
12[i]
56[g]
6[g]
20[i]
reduced enantiomeric excesses
in comparison to the results ob-
tained with the mixture (cf. en-
tries 1 and 3 in Table 1, and en-
tries 2 and 3 Table 2), the enan-
tiomeric composition of isomer
2b was significantly improved.
Thus, diastereomer 2b was
formed in up to 90 92% ee (en-
tries 1 and 5, Table 2), albeit at
moderate de (54 66%). Despite
its remarkable stereochemical
48[i]
12[j]
29[i]
85[i]
86[i]
49[g]
72[f]
79[g]
85[g]
Streptomyces griseus ATCC 10137
Streptomyces griseus DSM 40236
7
[a]Conversion after 49 h, calculated as the sum of the products formed versus the remaining bis-epoxide.
[b] ee of cis,cis,anti-1b: excess of enantiomer eluting first on chiral GC analysis. [c] ee of cis,cis,anti-1b: excess
of enantiomer eluting second on chiral GC analysis. [d] ee of (6S,7S,9R,10R)-2a. [e] ee of (6R,7R,9S,10S)-2a.
[f] ee of (6R,7R,9R,10R)-2b. [g] ee of (6S,7S,9S,10S)-2b. [h] de of (1a/1b), diastereomer cis,cis,anti-1b in excess.
[i] de of (2a/2b): diastereomer 2b in excess. [j] de of (2a/2b): diastereomer 2a in excess.
that one step proceeds with retention to form 2c or 2d from
cis,cis,meso-1a or cis,cis,anti-1b, respectively.
aspects, these results were not considered to be of synthetic
value. The results from an analogous biotransformation of
diastereomeric bis-epoxides (Æ)-trans,cis,anti-1c and (Æ)-
trans,cis,syn-1d (see Scheme 3 and the Experimental Sec-
tion) were unsatisfactory in terms of activities and selectivi-
ties (data not shown).
The racemic cis,cis,anti-1b diastereomer underwent kinet-
ic resolution, as may be deduced by the fact that the remain-
ing (more slowly converted) oxirane enantiomer was detect-
ed with up to 86% ee (entry 7), depending on the strain.
Overall, the enantioselectivities were low-to-moderate (ee<
20), while various biocatalysts exhibited an opposite enan-
tiopreference (cf. entries 1,2,5 7 versus entries 3,4,8,9).
Both diastereomers were converted at significantly different
rates, with the cis,cis,meso-1a diastereomer being faster, as
denoted by the de of the substrate (deS). Remarkably, Rho-
dococcus equi IFO3730 (entry 3) showed absolute selectivity
for the cis,cis,meso-1a (deS 99%).
Overall, all four possible stereoisomeric products (i.e.,
both enantiomers of diastereomers of 2a and 2b) were
formed by the various strains with up to 86% ee.
Since low selectivities are often caused by the presence of
parallel stereochemical pathways competing with each
other, in particular when acting on (dia)stereomeric sub-
strate mixtures, we anticipated that the product purity of the
biotransformation of pure diastereomers would be higher.
Thus, cis,cis,meso-1a and (Æ)-cis,cis,anti-1b were subjected
to biotransformation separately.
In contrast to kinetic resolution with its complex underly-
ing kinetics,[28] which cause the enantiomeric excess of the
substrate (eeS) and the product (eeP) to become a function
of the conversion (c), the stereochemical aspects of the de-
symmetrization of a meso-compound is considerably sim-
pler, as the eeP in this case is independent of c. Thus, better
selectivities were expected from the bioconversion of cis,cis,-
meso-1a. The results, given in Table 3, prove that this as-
sumption was correct: of the two possible diastereomers,
only isomer 2a was formed exclusively by four microorgan-
isms (entries 1 4, de>99%) with up to 59% ee, except for
entry 5, whereby diastereomer 2b was detected in minor
amounts (ꢀ5%). Again, various strains exhibited opposite
stereoselectivities (entries 1, 2, and 5 versus 3 and 4). Most
remarkably, Rhodococcus sp. CBS717.73 (entry 5) selective-
ly formed (6R,7R,9S,10S)-2a in 94% ee in addition to minor
amounts of the three other possible stereoisomers (each
<5%) at high conversion (c=84%; Scheme 4).
Selected results for the separate biotransformation of (Æ)-
cis,cis,anti-1b are gathered in Table 2. As may be deduced
from the ee×s of 1b, kinetic resolution of the racemate with
opposite enantiopreference (entries 1, 2, and 5 versus 3 and
The relative configuration of THF products 2a 2d was
1
determined by a comparison of the H and 13C NMR data
with reference samples obtained from acid-catalyzed hydrol-
ysis/cyclization of diastereomeric mixtures of bis-epoxides
cis,cis,meso-1 a/(Æ)-cis,cis,anti-
1b and (Æ)-trans,cis,anti-1c/(Æ)-
Table 2. Bioconversion of (Æ)-cis,cis,anti-1b.
trans,cis,syn-1d with the litera-
ture data.[27]
Biocatalyst
c[a]
eeS (1b)
eeP (2a)[d]
[%] [%]
eeP (2b)
deP
[%][%]
[%]
The absolute configuration of
the products was proven by co-
injection with independently
synthesized reference com-
pounds 2a 2d on GC with a
chiral stationary phase. The
latter materials were obtained
as shown in Schemes 5 and 6.
1
2
3
4
5
Rhodococcus ruber DSM 44540
Rhodococcus ruber DSM 44539
Rhodococcus equi IFO 3730
Mycobacterium paraffinicum NCIMB 10420
Rhodococcus sp. CBS 717.73
4
77
21
28
59
2[b]
82[b]
37[c]
12[c]
18[b]
17
19
47
28
75
90[e]
56[e]
9[e]
54[g]
22[g]
24[g]
42[h]
66[g]
37[f]
92[e]
[a]Conversion after 91 h, calculated as the sum of the products formed versus the remaining bis-epoxide.
[b] ee of cis,cis,anti-1b: excess of enantiomer eluting first on chiral GC analysis. [c] ee of cis,cis,anti-1b: excess
of enantiomer eluting second on chiral GC analysis. [d] ee of (6S,7S,9R,10R)-2a. [e] ee of 2b: (6R,7R,9R,10R)-
2b. [f] ee of 2b: (6S,7S,9S,10S)-2b. [g] de of (2a/2b): diastereomer 2b in excess. [h] de of (2a/2b): diastereomer
2a in excess.
3470
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 3467 3478

Biocatalytic Asymmetric Rearrangement
3467 3478
Table 3. Enzyme-catalyzed hydrolysis of cis,cis,meso-1a.
[e]
Entry
Biocatalyst
c[a]
eeP (2a)
eeP (2b)[d]
deP
[%][%]
[%]
[%]
1
2
3
4
5
Rhodococcus ruber DSM 44540
Rhodococcus ruber DSM 44539
Rhodococcus equi IFO 3730
Mycobacterium paraffinicum NCIMB 10420
Rhodococcus sp. CBS 717.73
87
56
40
38
84
51[b]
24[b]
59[c]
37[c]
94[b]
<1
<1
<1
<1
65
>99
>99
>99
>99
89
[a]Conversion after 49 h, calculated as the sum of the products formed versus the remaining bis-epoxide. [b] ee of 2a: (6R,7R,9S,10S)-2a. [c] ee of 2a:
(6S,7S,9R,10R)-2a. [d] ee of (6R,7R,9R,10R)-2b. [e] de of (2a/2b), diastereomer 2a is in excess.
Scheme 4. Stereochemical course of enzyme-catalyzed hydrolysis cyclization cascade of cis,cis,meso-1a.
centers at C2 and C3 are R,R
and S,S, respectively, as prede-
termined by the Z-configura-
tion of the internal olefin of
(S)-9. The diastereomeric mix-
ture of (2R,3R,5S)-10a and
(2S,3S,5S)-10b was then trans-
formed into the stereoisomeric
products
(6R,7R,9S,10S)-2a,
(6S,7S,9S,10S)-2b, (6S,7S,9S,10R)-
2c, and (6R,7R,9S,10R)-2d, by
means of the following one-pot
sequence without isolation of
intermediates: protection of the
hydroxyl-group in the 3-posi-
tion, followed by a two-step ox-
idative degradation of the vinyl
moiety (OsO4-catalyzed dihy-
droxylation followed by glycol
cleavage with NaIO4) led to the
Scheme 5. Asymmetric synthesis of reference material for 2a d.
formation of THF-carbalde-
hydes, which were trapped in
Allylic alcohol (S)-9, which was obtained from a lipase-
catalyzed kinetic resolution stereoinversion sequence[29] was
halogenated under Appel conditions to furnish the corre-
sponding allylic bromide with inversion of the configuration.
Epoxidation of the Z-configured and more active homoallyl-
ic olefin moiety gave the diastereomeric homoallylic epoxy-
bromides, which were subjected to acid-catalyzed hydroly-
sis cyclization to furnish the diastereomeric THF products
(2R,3R,5S)-10a and (2S,3S,5S)-10b. Because of the double
inversion involved in this last sequence, the absolute config-
uration at C5 is S. The relative configuration of the chiral
situ by a Grignard reagent prepared from n-pentyl bromide.
Finally, FÀ-mediated desilylation of the hydroxyl group gave
reference samples for THF products (6R,7R,9S,10S)-2a,
(6S,7S,9S,10S)-2b, (6S,7S,9S,10R)-2c, and (6R,7R,9S,10R)-
2d.
An analogous sequence leading to the enantiomeric series
(6S,7S,9R,10R)-2a, (6R,7R,9R,10R)-2b, (6R,7R,9R,10S)-2c,
and (6S,7S,9R,10S)-2d was accomplished by means of the
same protocol starting from alcohol (R)-9[29] (Scheme 6).
Chem. Eur. J. 2004, 10, 3467 3478
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3471

K. Faber et al.
FULL PAPER
used for a conformational analysis of
the reactant as well as the products.
Conformations generated thereby
were then first minimized by the semi-
empirical AM1[34] method (program
AMPAC[35]) followed by complete ge-
ometry optimization with Becke×s
three-parameter hybrid HF-DFT func-
tional[36] and the Lee Yang Parr cor-
relation functional[37] (B3LYP), as im-
plemented in Gaussian.[38] The 6-
31G(d) basis set was used throughout
for neutral molecules and 6-31+G(d)
for anionic species. All stationary
points were characterized as true
minima or transition states by frequen-
cy calculations. Transition states were
further characterized by intrinsic reac-
tion co-ordinate calculations (IRC)
along both directions of the normal
mode corresponding to the imaginary
frequency. Zero-point energy correc-
tions (ZPE) are unscaled. Solvent ef-
fects (aqueous solution) were modeled
with the aid of the Poisson Boltzmann
selfconsistent reaction field (PB-SCRF
B3LYP/6-311++G(d,p) single-point calculations) approximation[39,40] as
implemented in the Jaguar program package.[41]
Scheme 6. Asymmetric synthesis of reference material for 2a d.
Conclusion
The following trends for the enzyme-initiated hydrolysis
cyclization cascade of methylene-interrupted bis-epoxides
cis,cis,meso-1a and (Æ)-cis,cis,anti-1b can be summarized as
follows:
The formation of the two possible diastereomeric products
(6R,7R,9S,10S)-2a and (6R,7R,9R,10R)-2b was calculated to be strongly
exothermic and exergonic (with inclusion of bulk solvent effects (PB-
SCRF (H2O) B3LYP/6-311++G(d,p)
+ B3LYP/6-31G(d) thermal DG
corrections, DGreact =17 kcalmolÀ1 with respect to the most stable confor-
mation of the substrate 1e). Notably, both products are predicted to have
an almost equal Gibbs free energy of reaction, DGreact. Formation of tet-
rahydrofuran derivatives (6R,7R,9S,10S)-2a versus (6R,7R,9R,10R)-2b
through pathways A and B requires either intramolecular nucleophilic
attack of the OH at C10 with a concomitant opening of the epoxide ring
at C7 accompanied by proton transfer (pathway A) or, alternatively,
attack of the OH at C9, opening of the epoxide ring at C6 (pathway B).
To locate the two respective transition states, either one of the three dis-
1) The meso-compound cis,cis,meso-1a was converted
faster and more selectively than the diastereomeric bis-
epoxide (Æ)-cis,cis,anti-1b.
2) Bacterial epoxide hydrolases from different origin exhib-
it opposite enantiopreferences for the enantiomeric bis-
epoxide (Æ)-cis,cis,anti-1b.
3) Enzyme-catalyzed hydrolysis of cis,cis,meso-1a with
Rhodoccoccus sp. CBS717.73 predominantly formed a
single enantiomer in 94% ee and 89% de (Scheme 4).
4) The enzyme-triggered hydrolysis cyclization cascade of
methylene-interrupted bis-epoxides allows the simulta-
neous creation of a THF unit containing four stereocen-
ters in a single reaction.
À
À
À
À
tances r(O10 C2), r(O8 C2), and r(O8 H11) for pathway A and r(O9
À
À
C3), r(O8 C3), and r(O8 H12) for pathway B (for atom numbering, see
Figure 1) were used as reaction coordinates. Unfortunately, none of these
reaction path calculations were successful. Direct proton-transfer reac-
tions, for example, additions of neutral nucleophiles to carbonyl com-
pounds, are generally quite high energy processes and require catalysis
(water-assisted hydration or hydrolysis[42]) or the involvement of anionic
nucleophiles. Consequently, calculations were performed for the anionic
species resulting from deprotonation of the OH group at C19 and
C10. Deprotonation of the C9 hydroxy group is less favorable by
ꢀ3 kcalmolÀ1 than deprotonation of the C10 hydroxy group (PB-SCRF
(H2O)B3LYP/6-311++G(d,p) + B3LYP/6-31+G(d) thermal DG correc-
In view of the fact that there is no counterpart in chemo-
catalytic methodology available to date for such an asym-
metric, catalytic cascade reaction, this biotransformation
holds great potential for the short and efficient synthesis of
the THF core to give annonaceous acetogenins. The scaleup,
as well as scope and limitations of this protocol are under
investigation in our laboratories.
À
À
tions). Analogous reaction path calculations (r(O10 C2) and r(O8 C2)
À
À
for pathway A; r(O9 C3) and r(O8 C3) for pathway B) followed by
transition-state optimizations led to the two transitions states TS1 and
TS2 for the formation of furans (6R,7R,9S,10S)-2a and (6R,7R,9R,10R)-
2b from the corresponding deprotonated reactants. With respect to the
most stable deprotonated reactant conformer/isomer, the Gibbs free
energy barriers (PB-SCRF (H2O) B3LYP/6-31++G(d,p)
+ B3LYP/6-
31+G(d) thermal DG corrections) for the formation of (6R,7R,9S,10S)-
2a and (6R,7R,9R,10R)-2b are DG° =23 and 32 kcalmolÀ1, respectively.
Thus, although the Gibbs free energy barrier differences appear some-
what exaggerated, the modeling results clearly support a strong prefer-
ence for formation of the (6R,7R,9S,10S)-furan product 2a according to
pathway A.
Experimental Section
Computational methods: As simplified models for epoxydiol
(6R,7R,9R,10S)-1e and the two diasteromeric products (6R,7R,9S,10S)-
2a (pathway A) and (6R,7R,9R,10R)-2b (pathway B), analogous struc-
tures bearing a methyl group instead of n-pentyl were used. The starting
structures of these models were created by the SYBYL molecular model-
ing package.[30] The random search procedure[31] as implemented in
SYBYL (Merck molecular mechanics force-field MMFF94s[32, 33]) was
General: NMR spectra were recorded in CDCl3 with a Bruker AMX360
spectrometer at 360 MHz(1H) and 90 MHz (13C), and a Bruker DMX
Avance500 at 500 MHz (1H) and 125 MHz(13C), respectively. Chemical
shifts are reported relative to TMS (d=0.00) with CHCl3 as the internal
3472
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 3467 3478

Biocatalytic Asymmetric Rearrangement
3467 3478
Pb, Aldrich]was used as received. Substrate 3 and 1-heptyne were pur-
chased from Aldrich. m-Chloroperbenzoic acid (m-CPBA, Fluka, 70%)
was used. Bacteria were obtained from culture collections. Growth and
maintenance of bacterial strains, as well as the preparation of lyophilized
cultures was performed as previously described.[43]
General procedure for the enzyme-catalyzed hydrolysis of cis,cis,meso-1a
and (Æ)-cis,cis,anti-1b: Epoxides cis,cis,meso-1a and (Æ)-cis,cis,anti-1b
(5 mL) were hydrolyzed with lyophilized cells (50 mg) rehydrated in Tris
buffer for 0.5 h (1 mL, 0.05m, pH 8.0) by shaking the mixture at 308C at
120 rpm. The reactions were monitored by TLC and GC. After 49 h, the
cells were removed by centrifugation, and the products were extracted
with EtOAc (2î1mL). The combined organic layers were dried
(Na2SO4) and analyzed. The same procedure was also used for the con-
version of the diasteromeric mixture of cis,cis,meso-1a and (Æ)-cis,cis,
anti-1b.
General procedure for the synthesis of (Z)-4, (9Z)-rac,cis-8 and (9Z)-
rac,trans-8: Compounds (Z)-4, (9Z)-rac,cis-8 and (9Z)-rac,trans-8 were
obtained by selective hydrogenation of alkynes 3, rac,cis-7, and rac,trans-
7 with the Lindlar catalyst according to Method A.
Method A: Quinoline (8 mL), KOH (0.8 g, 14.2 mmol), and Lindlar cata-
lyst (1.2 g) were added to a solution of alkyne (10 g, 79.2 mmol) in EtOH
(40 mL). The resulting mixture was vigorously stirred under H2 for 18 h
at atmospheric pressure. The solids were removed by filtration through a
plug of Celite-545, and the solvent was evaporated. Flash chromatogra-
phy afforded stereoisomerically pure alkynes 4 and 8.
(E)-2-Octene-1-ol ((E)-4): A solution of alkyne 3 (30 g, 233 mmol) in an-
hydrous THF (30 mL) at 08C was added dropwise to a stirred solution of
LiAlH4 (9.73 g, 256.3 mmol) in anhydrous THF (100 mL) under an argon
atmosphere. The cooling bath was removed, and the mixture was stirred
for 16 h at room temperature and then cooled to 08C. Aqueous HCl
(5%, 30 mL) was slowly added. After filtration, the product was extract-
ed with Et2O (2î50 mL), and the organic phase was dried and concen-
trated. The residue was purified by flash chromatography (pentane/Et2O,
5:1) to afford (E)-4 (26.86 g, 90%) as a colorless liquid. Rf (petroleum
ether/EtOAc, 1:1)=0.66 (detection II); 1H NMR (360 MHz, CDCl3): d=
0.86 0.90 (m, 3H; CH3), 1.25 1.37 (brm, 6H; 3CH2), 1.69 (brs, 1H;
OH), 2.00 2.04 (m, 2H; CH2), 4.09 (d, J=6.5 Hz, 2H; CH2), 5.63
5.69 ppm (m, 2H; 2CH); 13C NMR (90 MHz, CDCl3): d=14.1, 22.6, 28.9,
31.4, 32.2, 63.8, 128.9, 133.5 ppm.
(Z)-2-Octene-1-ol ((Z)-4): Method A was employed with alkyne 3 (10 g,
79.2 mmol). Flash chromatography (pentane/Et2O, 5:1) gave cis-2-octen-
1-ol (Z)-4 as a colorless liquid (7.1 g, 70%). Rf (petroleum ether/EtOAc,
5:1)=0.25, (detection II); 1H NMR (360.13 MHz, CDCl3): d=0.86 0.91
(m, 3H), 1.27 1.37 (m, 6H), 1.55 (brs, 1H), 2.07 (dt, J=7.4, 6.5 Hz, 2H),
4.19 (d, J=6.0 Hz, 2H;), 5.52 5.61 ppm (m, 2H); 13C NMR (90 MHz,
CDCl3): d=14.1, 22.7, 27.4, 29.3, 31.5, 58.6, 128.4, 133.3 ppm.
Figure 1. Calculated structures of reactants, transition states and products
for cyclization pathways A and B. The structures correspond to com-
pounds depicted in Scheme 1 (R=methyl).
(6R*,7S*,9Z)-6,7-Epoxy-9-pentadecene ((9Z)-rac,cis-8): Method A was
employed with alkyne rac,cis-7 (8.7 g, 39.1 mmol), but without addition
of KOH. Flash chromatography (pentane/Et2O, 10:1) gave alkene (9Z)-
rac,cis-8 as a colorless liquid (7.9 g, 91%). Rf (petroleum ether/EtOAc,
5:1)=0.67, (detection I); 1H NMR (360.13 MHz, CDCl3): d=0.84 0.91
(m, 6H), 1.27 1.65 (brm, 14H), 2.04 (dt, J=7.0, 7.2 Hz, 2H), 2.17 2.19
(m, 1H), 2.37 2.41 (m, 1H), 2.93 (t, J=7.8 Hz, 2H), 5.40 5.46 (m, 1H),
5.50 5.55 ppm (m, 1H); 13C NMR (90 MHz, CDCl3): d=14.1, 22.6, 22.7,
26.3, 26.4, 27.5, 27.8, 29.3, 31.6, 31.9, 56.6, 57.3, 123.9, 132.8 ppm.
standard [d=7.23 (1H) and 76.90 (13C)], coupling constants (J) are given
in Hz.
TLC was performed on silica gel Merck60 (F254), and compounds were
visualized by spraying with Mo reagent [(NH4)6Mo7O24¥4H2O (100 gLÀ1),
Ce(SO4)2¥4H2O(4 gLÀ1) in H2SO4 (10%); detection I]or by dipping into
a KMnO4 solution (2.5 gLÀ1 in H2O; detection II). Compounds were pu-
rified by flash chromatography on silica gel (Merck60, 230–400 mesh).
Petroleum ether had a boiling range of 60 908C. GC analyses were car-
ried out on a Varian3800 gas chromatograph equipped with FID and a
HP1301 (column C) capillary column (30 m, 0.25 mm, 0.25 mm film, N2).
Enantiomeric purities were analyzed with a CP-Chirasil DEXCB column
(column A, 25 m, 0.32 mm, 0.25 mm film) or a Astec Chiraldex B-TA
(column B, 30 m, 0.25 mm). H2 was used as the carrier gas. Reactions
were monitored by GC analysis on an achiral stationary phase, for ana-
lytical data see Table 4. Chiral materials were analyzed either directly or
as the corresponding trifluoroacetate esters by GC on a chiral stationary
phase, for analytical data see Table 5.
(6R*,7R*,9Z)-6,7-Epoxy-9-pentadecene ((9Z)-rac,trans-8): Method A was
employed with alkyne rac,trans-7 (2.6 g, 11.7 mmol), but without addition
of KOH. Flash chromatography (pentane/Et2O, 10:1) gave alkyne (9Z)-
rac,trans-8 as a colorless liquid (2.35 g, 90%). Rf (petroleum ether/
EtOAc, 5:1)=0.70, (detection I); 1H NMR (360 MHz, CDCl3): d=0.87
0.90 (t, J=7.0 Hz, 6H;2CH3), 1.27 1.55 (brm, 14H; 7CH2), 2.03 (dt, J=
6.8, 7.0, Hz, 2H; CH2), 2.24 (m, 1H; CH), 2.41 (m, 1H; CH), 2.68 2.71
(m, 2H; 2CH), 5.38 5.42 (m, 1H; CH), 5.48 5.54 ppm (m, 1H; CH); 13C
NMR (90 MHz, CDCl3): d=14.0, 14.1, 22.6, 22.6, 30.0, 25.7, 27.4, 29.3,
31.5, 31.7, 32.0, 58.1, 58.6, 123.4, 133.0 ppm.
High-resolution mass spectra were recorded on a double-focusing Kratos
Profile Mass Spectrometer with electron impact ionization (EI, +70 eV).
General procedure for the synthesis of allylic chlorides (E)-5 and (Z)-5:
Compounds (Z)-5 and (E)-5 were obtained by chlorination of alcohols
(Z)-4 and (E)-4 according to Method B.
Solvents were dried and freshly distilled. All reactions were performed
under an argon atmosphere, unless otherwise stated. Organic extracts
were dried over Na2SO4, and then the solvent was evaporated under re-
duced pressure. Lindlar catalyst [Pd on CaCO3 (5% w/w) poisoned with
Method B: PPh3 (31.4 g, 120.0 mmol) and allylic alcohol (13.3 g,
103.6 mmol) were dissolved in CCl4 (80 mL). The reaction was complete
Chem. Eur. J. 2004, 10, 3467 3478
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3473

K. Faber et al.
FULL PAPER
Table 4. GC data on an achiral stationary phase.
Table 4. (Continued)
Compound
Compound
Column
Conditions[a]
A
tR [min]
Column
C
Conditions[a]
tR [min]
C
14.11
A
12.69
C
C
C
A
A
A
13.95
13.97
13.90
C
C
A
A
12.57
12.26
[a]Conditions: A) 14.5 psi; held at 140 8C for 8 min; heat rate 158CminÀ1
up to 2008C, then 508CminÀ1 up to 2508C; held at 2508C for 30 min;
B) 14.5 psi; held at 608C for 2 min; heat rate 88CminÀ1 up to 1508C,
then 308CminÀ1 up to 2508C; held at 2508C for 20 min.
C
C
C
A
A
A
15.50
15.84
15.16
after stirring for 12 h at 808C. The solution was concentrated, and pen-
tane (50 mL) was added. The mixture was filtered, and the filtrate was
concentrated in vacuo. Flash chromatography afforded the allylic chlor-
ides (E)-5 and (Z)-5.
(E)-1-Chloro-2-octene ((E)-5): Method B was employed with PPh3
(73.9 g, 281.7 mmol) and alcohol (E)-4 (25.8 g, 201.2 mmol). Flash chro-
matography (pentane/Et2O, 10:1) gave chloroalkene (E)-5 (28.0 g, 95%)
as a colorless liquid. Rf (petroleum ether/EtOAc, 5:1)=0.87, (detec-
tion II); 1H NMR (360 MHz, CDCl3): d=0.85 0.92 (m, 3H; CH3), 1.28
1.42 (brm, 6H; 3CH2), 2.06 (dt, J=7.2, 6.9 Hz, 2H; CH2), 4.02 (d, J=
7.0 Hz, 2H; CH2), 5.60 5.64 (m, 1H; CH), 5.75 5.81 ppm (m, 1H; CH);
13C NMR (90 MHz, CDCl3): d=14.1, 22.5, 28.6, 31.4, 32.1, 45.6, 125.9,
136.3 ppm.
(Z)-1-Chloro-2-octene ((Z)-5): Method B was employed with PPh3
(31.4 g, 120.0 mmol) and alcohol (Z)-4 (13.3 g, 103.6 mmol). Flash chro-
matography (petroleum ether) gave chloroalkene (Z)-5 (10.17 g, 77%) as
a colorless liquid. Rf (petroleum ether/EtOAc, 5:1)=0.84, (detection II);
1H NMR (360 MHz, CDCl3): d=0.90 (t, J=6.8 Hz, 3H), 1.21 1.42 (m,
6H), 2.12 (dt, J=7.2, 6.7 Hz, 2H), 4.11 (d, J=6.6 Hz, 2H), 5.62 5.65 ppm
(m, 2H); 13C NMR (90 MHz, CDCl3): d=14.0, 22.5, 27.1, 29.0, 31.4, 39.6,
125.2, 135.6 ppm.
C
C
A
B
15.31
10.38
General procedure for the synthesis of enynes (E)-6 and (Z)-6: Com-
pounds (Z)-6 and (E)-6 were obtained by coupling 1-heptyne onto allylic
chlorides (E)-5 and (Z)-5 according to Method C.
C
C
C
C
B
B
B
B
9.53
9.44
8.79
8.66
Method C: Bu4N+ClÀ (1 g, 3.56 mmol), K2CO3 (5.17 g, 37.41 mmol), and
CuI (0.25 g, 1.29 mmol) were added to a stirred solution of 1-heptyne
(2.92 g, 30.32 mmol) in anhydrous DMF (25 mL). After the mixture was
stirred for 15 min at room temperature, a solution of allylic chloride (E)-
5 or (Z)-5 (3.87 g, 26.38 mmol) in DMF (8 mL) was added dropwise
within 15 min. The resulting solution was stirred at room temperature for
a further 7 h. After the reaction was complete, the mixture was quenched
by addition of water (15 mL) and Et2O (50 mL). After phase separation,
the organic phase was dried and concentrated. Flash chromatography af-
forded stereoisomerically pure enynes (Z)-6 and (E)-6.
(Z)-6-Pentadecen-9-yne ((Z)-6): Method C was employed with (Z)-5
(3.87 g, 26.38 mmol). Flash chromatography (petroleum ether) gave (Z)-6
(4.25 g, 78%) as a colorless liquid. Rf (petroleum ether/EtOAc, 10:1)=
0.86, (detection II); 1H NMR (360 MHz, CDCl3): d=0.88 0.93 (m, 6H),
1.25 1.55 (m, 12H), 2.03 (dt, J=6.6, 7.6 HZ, 2H), 2.13 2.16 (m, 2H),
2.90 (dt, J=5.1, 2.4 Hz, 2H), 5.40 5.44 ppm (m, 2H); 13C NMR (90 MHz,
CDCl3): d=14.0, 14.1, 17.2, 18.8, 22.3, 22.5, 27.2, 28.7, 29.1, 31.2, 31.5,
78.5, 80.2, 125.0, 131.4 ppm.
C
A
10.92
C
C
A
A
10.47
12.93
(E)-6-Pentadecen-9-yne ((E)-6): Method C was employed with (E)-5
(20.0 g, 136.3 mmol). Flash chromatography (petroleum ether) gave (E)-6
(24.1 g, 86%) as a colorless liquid. Rf (petroleum ether/EtOAc, 10:1)=
0.85, (detection II); 1H NMR (360 MHz, CDCl3): d=0.88 0.93 (m, 6H;
2CH3), 1.27 1.56 (brm, 12H; 6CH2), 2.03 (dt, J=6.8, 6.6 HZ, 2H; CH2),
2.17 2.20 (m, 2H; CH2), 2.89 (brs, 2H; CH2), 5.38 5.45 (m, 1H; CH),
5.64 5.72 ppm (m, 1H; CH); 13C NMR (90 MHz, CDCl3): d=14.0, 14.1,
3474
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 3467 3478

Biocatalytic Asymmetric Rearrangement
3467 3478
1
Table 5. GC data on a chiral stationary phase.
1d was determined by a comparison of H and 13C NMR spectra with lit-
erature data.[27]
Compound
Column Conditions[a] tR [min]
Method D: Finely powdered Na2HPO4 (ꢀ2.6 equiv) was added to a vigo-
rously stirred solution of alkene (40 mmol) in anhydrous CH2Cl2
(400 mL). The mixture was stirred for 15 min at room temperature and
was then cooled to 08C. m-CPBA (1.05 equiv) was added slowly. The
mixture was allowed to warm to room temperature and was stirred for
an additional 18 h. The white suspension was filtered. The resulting solu-
tion was treated with 10% aqueous Na2S2O5 (100 mL) to destroy excess
peracid. The two-phase system was stirred for 30 min, the layers were
separated, and the organic phase was washed with saturated aqueous
NaHCO3 (50 mL). The organic phase was dried and evaporated. Flash
chromatography afforded oxiranes rac,cis-7, rac,trans-7, cis,cis,meso-1a,
(Æ)-cis,cis,anti-1b, (Æ)-trans,cis,anti-1c, and (Æ)-trans,cis,syn-1d.
(6R*,7S*)-6,7-Epoxypentadec-9-yne (rac,cis-7): Method D was employed
with alkene (Z)-6 (4.0 g, 19.38 mmol). Flash chromatography (pentane/
Et2O, 20:1) gave oxirane rac,cis-7 as a colorless liquid (2.9 g, 73%). Rf
(petroleum ether/EtOAc, 5:1)=0.67, (detection I); 1H NMR (360 MHz,
CDCl3): d=0.81 0.92 (m, 6H), 1.27 1.64 (brm, 14H), 2.13 2.18 (m, 2H),
2.27 (m, 1H), 2.59 2.60 (m, 1H), 2.96 2.98 (m, 1H), 3.09 3.12 ppm (m,
1H); 13C NMR (90 MHz, CDCl3): d=14.0, 18.8, 18.8, 22.3, 22.6, 26.2,
27.5, 28.7, 31.1, 31.7, 55.5, 57.2, 74.9, 82.0 ppm.
A
A
A
A
34.41
29.48/30.31
A
A
A
A
30.95/31.51
30.18/30.50
B
B
A
A
B
C
C
23.51
(6R*,7R*)-6,7-Epoxypentadec-9-yne (rac,trans-7): Method D was em-
ployed with alkene (E)-6 (0.3 g, 1.45 mmol). Flash chromatography (pen-
tane/Et2O, 20:1) gave oxirane rac,trans-7 as a colorless liquid (0.25 g,
83%). Rf (petroleum ether/EtOAc, 5:1)=0.61, (detection I); 1H NMR
(360 MHz, CDCl3): d=0.88 0.92 (t, J=7.0 Hz, 6H; 2CH3), 1.32 1.55
(brm, 14H; 7CH2), 2.13 2.17 (m, 2H; CH2), 2.36 (m, 1H; CH), 2.57 (m,
1H; CH), 2.82 2.85 ppm (m, 2H; 2CH); 13C NMR (90 MHz, CDCl3):
d=14.0, 18.8, 22.4, 22.3, 22.6, 25.7, 28.7, 31.1, 31.6, 31.7, 56.6, 58.5, 74.5,
82.7 ppm.
25.02
A
A
A
15.10/15.50
17.71/18.34
16.47/16.89
(6R*,7S*,9R*,10S*)-6,7:9,10-Bis(epoxy)pentadecane (cis,cis,meso-1a) and
(6R*,7S*,9S*,10R*)-6,7:9,10-bis(epoxy)pentadecane ((Æ)-cis,cis,anti-1b):
Method D was employed with alkene (9Z)-rac,cis-8 (5.72 g, 25.5 mmol).
Because the separation of the diastereomers was incomplete, repeated
flash chromatography (pentane/Et2O, 20:1) with GC analysis of the frac-
tions was required to obtain bis-epoxides cis,cis,meso-1a (1.35 g, 23.7%,
elutes first) and (Æ)-cis,cis,anti-1b (1.5 g, 26.3%, elutes second) as white
crystals.
(6R*,7S*,9R*,10S*)-6,7:9,10-Bis(epoxy)pentadecane (cis,cis,meso-1a): Rf
(petroleum ether/EtOAc, 5:1)=0.28, (detection I); 1H NMR
(500.13 MHz, CDCl3): d=0.84 0.92 (m, 6H), 1.26 1.46 (m, 12H), 1.52
1.55 (m, 4H), 1.72 1.75 (m, 1H), 1.79 1.84 (m, 1H), 2.97 (dt, J=5.5,
6.1 Hz, 2H), 3.06 3.09 ppm (m, 2H); 13C NMR (125 MHz, CDCl3): d=
13.9, 22.5, 26.1, 26.8, 27.7, 31.6, 54.1, 56.6 ppm; HRMS (C15H26O) m/z
calcd: 222.1984 [MÀH2O]+; found: 222.1963 [MÀH2O]+.
(6S,7S,9R,10R)-2a-bis-trifluoroace-
tate
(6R,7R,9S,10S)-2a-bis-trifluoroace-
tate
(6S,7S,9S,10S)-2b-bis-trifluoroace-
tate
(6R,7R,9R,10R)-2b-bis-trifluoroace-
tate
B
B
B
B
D
D
D
D
4.26
4.44
5.32
5.40
(6R*,7S*,9S*,10R*)-6,7:9,10-Bis(epoxy)pentadecane ((Æ)-cis,cis,anti-1b):
Rf (petroleum ether/EtOAc, 5:1)=0.23, (detection I); 1H NMR
(500.13 MHz, CDCl3): d=0.85 0.91 (m, 6H), 1.27 1.57 (m, 16H), 1.74 (t,
J=6.2 Hz, 2H), 2.99 (dt, J=5.1, 6.4 Hz, 2H), 3.13 ppm (t, J=4.3, 6.2 Hz,
2H); 13C NMR (125 MHz, CDCl3): d=14.0, 22.5, 26.1, 27.1, 27.8, 31.6,
54.3, 57.0 ppm; HRMS (C15H26O) m/z calcd: 222.1984 [MÀH2O]+;
found: 222.1959 [MÀH2O]+.
(+)-2c-bis-trifluoroacetate
(6S,7S,9R,10R)-2d-bis-trifluoroace-
B
B
D
D
5.19
3.57
(6R*,7R*,9R*,10S*)-6,7:9,10-Bis(epoxy)pentadecane
((Æ)-trans,cis,anti-
tate
1c) and (6R*,7R*,9S*,10R*)-6,7:9,10-bis(epoxy)pentadecane ((Æ)-trans,-
cis,syn-1d): Method D was employed with alkene (9Z)-rac,trans-8 (0.2 g,
9.05 mmol). Because the separation of the diastereomers was incomplete,
repeated flash chromatography (pentane/Et2O, 20:1) with GC analysis of
the fractions was required to obtain bis-epoxides (Æ)-trans,cis,anti-1c
(0.06 g, 30%, elutes first) and (Æ)-trans,cis,syn-1d (0.07 g, 35%, elutes
second) as colorless liquids.
(6R,7R,9S,10S)-2d-bis-trifluoroace-
tate
[a]Conditions: A) 5.0 psi, held at 140 8C for 30 min, heat rate 208CminÀ1
up to 1708C; held at 1708C for 10 min. B) 6.0 psi, 1708C (isothermal).
C) 7.0 psi, 1708C (isothermal). D) 10.0 psi, 1608C (isothermal).
B
D
3.73
(6R*,7R*,9R*,10S*)-6,7:9,10-Bis(epoxy)pentadecane
((Æ)-trans,cis,anti-
1c): Rf (petroleum ether/EtOAc, 5:1)=0.45, (detection I); 1H NMR
(500 MHz, CDCl3): d=0.87 0.90 (m, 6H; 2CH3), 1.31 1.57 (brm, 16H;
8CH2), 1.82 (t, J=5.5 Hz, 2H; CH2), 2.80 2.85 (m, 2H; 2CH), 2.93 (dt,
J=4.4, 6.0, 1H; CH), 2.99 3.02 ppm (m, 1H; CH); 13C NMR (500 MHz,
CDCl3): d=13.9, 22.5, 25.6, 26.1, 27.8, 31.5, 31.6, 31.8, 30.3, 53.2, 55.7,
56.4, 58.1 ppm.
18.9, 22.1, 22.3, 22.6, 28.9, 29.1, 31.2, 31.5, 32.3, 77.7, 82.2, 124.7,
131.9 ppm.
General procedure for the synthesis of rac,cis-7, rac,trans-7, cis,cis,meso-
1a, (Æ)-cis,cis,anti-1b, (Æ)-trans,cis,anti-1c and (Æ)-trans,cis,syn-1d:
These substrates were prepared by epoxidation of alkenes (Z)-6, (E)-6,
(9Z)-rac,cis-8, and (9Z)-rac,trans-8 with m-chloroperbenzoic acid (m-
CPBA) following Method D. The relative configuration of substrates 1a
(6R*,7R*,9S*,10R*)-6,7:9,10-Bis(epoxy)pentadecane
((Æ)-trans,cis,syn-
1d): Rf (petroleum ether/EtOAc, 5:1)=0.45, (detection I); 1H NMR
Chem. Eur. J. 2004, 10, 3467 3478
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3475

K. Faber et al.
FULL PAPER
(500 MHz, CDCl3) d=0.88 0.91 (m, 6H; 2CH3), 1.26 1.62 (brm, 16H;
8CH2), 1.71 1.76 (m, 2H; CH2), 2.74 (dt, J=2.2, 5.6 Hz, 1H; CH), 2.84
2.86 (m, 1H; CH), 2.96 2.97 (m, 1H; CH), 3.08 3.10 ppm (m, 1H; CH);
13C NMR (500 MHz, CDCl3) d=13.9, 22.5, 25.5, 26.0, 27.7, 31.5, 31.5,
31.8, 31.2, 53.9, 55.9, 56.8, 58.7 ppm.
stereomeric mixture of the epoxy bromide was hydrolyzed rearranged in
a mixture of water (6 mL) and THF (4 mL) under acidic conditions
(conc. H2SO4, 10 drops) at room temperature overnight. The solution was
extracted with EtOAc (2î10 mL). The combined organic layers were
dried and evaporated. The residue was purified by flash chromatography
(petroleum ether/EtOAc, 10:1) to afford the diasteromeric mixture of
the cyclization products (2R,3R,5S)-10a and (2S,3S,5S)-10b (85 mg). Rf1,
Rf2 (petroleum ether/EtOAc, 1:1)=0.66, 0.61 (detection I).
First diastereomer: 1H NMR (360 MHz,CDCl3): d=0.89 0.91 (t, J=
6.7 Hz, 3H; CH3), 1.25 1.34 (m, 4H; CH2), 1.41 1.47 (m, 2H; CH2),
1.64 1.68 (m, 2H; CH2), 1.74 1.80 (m, 1H; CH2), 2.39 2.45 (m, 1H;
CH2), 3.62 3.67 (m, 1H; CH), 4.18 (m, 1H; CH), 4.30 4.36 (m, 1H; CH),
5.10 5.13 (dd, J=10.4, 1.5 Hz, 1H; CH2), 5.28 5.33 (dd, J=17.2, 1.7 Hz,
1H; CH2), 5.92 6.02 ppm (ddd, J=17.2, 10.7, 6.5 Hz, 1H; CH); 13C NMR
(90 MHz, CDCl3): d=14.0, 22.5, 25.9, 28.8, 32.0, 41.7, 73.1, 77.9, 83.6,
115.1, 140.1 ppm.
Second diastereomer: 1H NMR (360 MHz,CDCl3): d=0.89 0.90 (t, J=
6.4 Hz, 3H; CH3), 1.32 1.33 (m, 4H; CH2), 1.54 1.66 (m, 4H; CH2),
1.88 1.91 (m, 1H; CH2), 2.14 2.20 (m, 1H; CH2), 3.82 3.85 (m, 1H;
CH), 4.26 (m, 1H; CH), 4.60 4.66 (m, 1H; CH), 5.08 5.11 (dd, J=9.3,
1.0 Hz, 1H; CH2), 5.22 5.27 (dd, J=16.0, 1.1 Hz, 1H; CH2), 5.79
5.89 ppm (ddd, J=15.2, 8.7, 3.7 Hz, 1H; CH); 13C NMR (90 MHz,
CDCl3): d=14.0, 22.5, 26.0, 29.0, 31.8, 41.9, 73.3, 77.6, 82.4, 114.9,
139.1 ppm.
General procedure for the synthesis of reference material for (Æ)-2a,
(Æ)-2b, (Æ)-2c, and (Æ)-2d: Substrates (Æ)-2a d were prepared by
acid-catalyzed hydrolysis followed by rearrangement of mixtures of the
corresponding diastereomeric bis-epoxides cis,cis,meso-1a/(Æ)-cis,cis,anti-
1b, and (Æ)-trans,cis,anti-1c/(Æ)-trans,cis,syn-1d following a previously
reported procedure[27] (Method E, see Schemes 2 and 3).
Method E: A mixture of bis-epoxides (0.2 g, 0.83 mmol) was hydrolyzed
in a mixture of water (5 mL) and THF (5 mL) under acidic conditions
(6n H2SO4, 20 drops) without an argon atmosphere. The reaction was
complete after 24 h. The solution was extracted with EtOAc (2î20 mL).
The combined organic layers were dried and evaporated to afford (Æ)-
2a and (Æ)-2b, or (Æ)-2c and (Æ)-2d, respectively, which were separat-
ed by flash chromatography (petroleum ether/ethyl acetate 10:1). Details
and spectroscopic data are given below.
(6R*,7R*,9S*,10S*)-6,9-Epoxypentadecane-7,10-diol
((Æ)-2a)
and
(6R*,7R*,9R*,10R*)-6,9-epoxypentadecane-7,10-diol ((Æ)-2b): Method E
was employed with a mixture of bis-epoxides cis,cis,meso-1a and (Æ)-cis,
cis,anti-1b (0.2 g, 0.83 mmol). Flash chromatography (petroleum ether/
EtOAc, 10:1) afforded (Æ)-2a (0.07 g, 31%) and (Æ)-2b (0.13 g, 59%).
(6R*,7R*,9S*,10S*)-6,9-Epoxypentadecane-7,10-diol ((Æ)-2a): Rf (petro-
leum ether/EtOAc, 1:1)=0.46, (detection I); 1H NMR (500.13 MHz,
CDCl3): d=0.85 0.90 (m, 6H), 1.29 1.65 (m, 16H), 1.85 1.90 (m, 1H),
2.00 2.04 (m, 1H), 3.36 3.40 (m, 1H), 3.73 3.77 (m, 1H), 4.00 4.04 (m,
1H), 4.25 ppm (t, J=3.4 Hz, 1H); 13C NMR (125 MHz, CDCl3): d=13.9,
14.0, 22.5, 22.5, 25.2, 25.9, 28.7, 31.0, 31.8, 33.0, 37.8, 73.4, 74.0, 80.1,
82.4 ppm; HRMS (C15H30O3) m/z calcd: 258.2195 [M]+; found: 258.2227
[M]+.
(6R,7R,9S,10S)-6,9-Epoxypentadecane-7,10-diol
(6S,7S,9S,10S)-6,9-epoxypentadecane-7,10-diol
((6R,7R,9S,10S)-2a),
((6S,7S,9S,10S)-2b),
(6S,7S,9S,10R)-6,9-epoxypentadecane-7,10-diol ((6S,7S,9S,10R)-2c), and
(6R,7R,9S,10R)-6,9-epoxypenta-decane-7,10-diol ((6R,7R,9S,10R)-2d): A
solution of the diasteromeric mixture of (2R,3R,5S)-10a and (2S,3S,5S)-
10b (85 mg, 0.46 mmol), TBDMSCl (90.3 mg, 0.60 mmol), and imidazole
(40.4 mg, 0.60 mmol) in CH2Cl2 (10 mL) was stirred at room temperature
overnight and then poured into a mixture of saturated NaHCO3 and
CH2Cl2. The resulting mixture was stirred vigorously for 30 min, the
layers were separated, and the aqueous layer was extracted with CH2Cl2.
The combined organic phases were dried and concentrated. N-Methyl-
morpholine-N-oxide (124.0 mg, 1.1 mmol) and one crystal of OsO4 were
added to a solution of the crude residue (158 mg) in acetone (10 mL).
This mixture was stirred for 1.5 h at room temperature. Sodium sulfite
(196 mg, 3.14 mmol) was added and stirring was continued for 30 min.
The solution was extracted with EtOAc (2î10 mL). The combined or-
ganic phases were washed with water and concentrated under reduced
pressure. Acetone (10 mL) followed by NaIO4 (198.8 mg, 0.93 mmol)
were added to a suspension of the residue (182 mg) in water (4 mL). The
mixture was stirred for 30 min at room temperature. The solution was ex-
tracted with EtOAc (2î10 mL), and the combined organic layers were
dried and evaporated. The residue (141 mg) was dissolved in Et2O
(10 mL). Pentylmagnesium bromide (0.5 mL of a 2m solution in THF,
1 mmol) was added to the vigorously stirred solution and stirring was
continued for 5 h at room temperature. The reaction was quenched by
addition of H2O (5 mL) and Et2O (10 mL). The phases were separated,
and the aqueous layer was extracted with Et2O (2î10 mL). The com-
bined organic phases were dried and evaporated. The residue (130 mg)
was dissolved in THF (10 mL), and Bu4N+FÀ (161 mg, 0.51 mmol) was
added. The reaction was stirred overnight at room temperature. The mix-
ture was quenched by addition of water (5 mL) and Et2O (10 mL). The
phases were separated, and the aqueous layer was extracted with Et2O
(2î10 mL). The combined organic phases were dried and evaporated.
The residue was purified by flash chromatography (petroleum ether/
EtOAc, 10:1) to afford a diasteromeric mixture of the cyclization prod-
ucts (6R,7R,9S,10S)-2a, (6S,7S,9S,10S)-2b, (6S,7S,9S,10R)-2c, and
(6R,7R,9S,10R)-2d (17 mg). Rf (petroleum ether/EtOAc, 1:1; detection I)=
0.46 (2a), 0.51 (2b), 0.54 (2c), 0.34 (2d).
(6R*,7R*,9R*,10R*)-6,9-Epoxypentadecane-7,10-diol ((Æ)-2b): Rf (petro-
leum ether/EtOAc, 1:1)=0.54, (detection I); 1H NMR (500.13 MHz,
CDCl3): d=0.88 0.90 (t, J=6.3 Hz, 6H), 1.31 1.66 (m, 16H), 1.84 (dd,
J=14.1, 3.5 Hz, 1H), 2.36 2.41 (m, 1H), 3.46 3.48 (m, 1H), 3.63 3.68 (m,
1H), 3.94 3.97 (m, 1H), 4.04 ppm (dd, J=5.4 Hz, 1H; 2.7); 13C NMR
(125 MHz, CDCl3): d=14.0, 22.5, 25.6, 25.8, 28.7, 31.6, 32.0, 34.3, 38.7,
71.5, 73.9, 79.0, 84.3 ppm; HRMS (C15H30O3) m/z calcd: 258.2195 [M]+;
found: 258.2198 [M]+.
(6R*,7R*,9R*,10S*)-6,9-Epoxypentadecane-7,10-diol
((Æ)-2c)
and
(6R*,7R*,9S*,10R*)-6,9-epoxypentadecane-7,10-diol ((Æ)-2d): Method E
was employed with a mixture of bis-epoxides (Æ)-trans,cis,anti-1c and
(Æ)-trans,cis,syn-1d (0.2 g, 0.83 mmol). Flash chromatography (petroleum
ether/EtOAc, 10:1) afforded (Æ)-2c (0.08 g, 35%) and (Æ)-2d (0.10 g,
45%).
(6R*,7R*,9R*,10S*)-6,9-Epoxypentadecane-7,10-diol ((Æ)-2c): Rf (petro-
1
leum ether/EtOAc, 1:1)=0.54 (detection I); H NMR (500 MHz, CDCl3):
d=0.87 0.89 (t, J=6.7 Hz, 6H; 2CH3), 1.28 1.68 (brm, 16H; 8CH2),
1.90 (dd, J=14.1, 3.3 Hz, 1H; CH), 2.14 2.19 (m, 1H; CH), 3.56 3.60 (m,
1H; CH), 3.80 3.83 (m, 1H; CH), 3.98 4.01 (m, 1H; CH); 13C NMR
(500 MHz, CDCl3): d=13.9, 13.9, 22.4, 22.5, 25.5, 25.9, 28.7, 31.6, 32.0,
33.3, 34.2, 71.0, 71.9, 79.9, 83.8 ppm; HRMS (C15H30O3) m/z calcd:
258.2195 [M]+; found: 258.2209 [M]+.
(6R*,7R*,9S*,10R*)-6,9-Epoxypentadecane-7,10-diol ((Æ)-2d): Rf (petro-
1
leum ether/EtOAc, 1:1)=0.34 (detection I); H NMR (500 MHz, CDCl3):
d=0.86 0.89 (t, J=5.3 Hz, 6H; 2CH3), 1.29 1.59 (brm, 16H; 8CH2),
1.84 (dd, J=13.2, 6.1 Hz, 1H; CH), 2.08 2.13 (m, 1H; CH), 3.82 3.86 (m,
2H; 2CH), 4.13 4.17 (m, 1H; CH), 4.27 ppm (brs, 1H; CH); 13C NMR
(500 MHz, CDCl3): d=13.9, 22.5, 22.5, 25.5, 25.9, 29.1, 31.8, 32.1, 32.4,
34.1, 72.0, 73.1, 79.9, 83.5 ppm; HRMS (C15H30O3) m/z calcd: 258.2195
[M]+; found: 258.2205 [M]+.
(6R,7R,9S,10S)-2a: 1H NMR (500 MHz, CDCl3): d=0.85 0.90 (m, 6H;
2CH3), 1.29 1.65 (brm, 16H; 8CH2), 1.85 1.90 (m, 1H; CH), 2.00 2.04
(m, 1H; CH), 3.36 3.40 (m, 1H; CH), 3.73 3.77 (m, 1H; CH), 4.0 4.04
(m, 1H; CH), 4.25 ppm (t, J=3.4 Hz, 1H; CH); 13C NMR (500 MHz,
CDCl3): d=13.93, 13.96, 22.47, 22.52, 25.19, 25.89, 28.73, 31.81, 31.89,
33.02, 37.81, 73.35, 74.0, 80.11, 82.36 ppm.
(2R,3R,5S)-2-Pentyl-5-vinyl-tetrahydrofuran-3-ol ((2R,3R,5S)-10a) and
(2S,3S,5S)-2-pentyl-5-vinyl-tetrahydrofuran-3-ol ((2S,3S,5S)-10b): PPh3
(2.13 g, 8.12 mmol) and CBr4 (2.41 g, 7.26 mmol) were dissolved in anhy-
drous CH2Cl2 (30 mL). The stirred solution was cooled to 08C, and (S)-9
(650 mg, 3.86 mmol) was added dropwise. After stirring for 5 h, the solu-
tion was concentrated, and pentane (30 mL) was added. The mixture was
filtered, and the filtrate was concentrated in vacuo. The residue (381 mg)
was epoxidized according to Method D. The residue (122 mg) of the dia-
(6S,7S,9S,10S)-2b: 1H NMR (500 MHz, CDCl3): d=0.88 0.90 (t, J=
6.3 Hz, 6H; 2CH3), 1.31 1.66 (brm, 16H; 8CH2), 1.84 (dd, J=14.1,
3476
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 3467 3478

Biocatalytic Asymmetric Rearrangement
3467 3478
3.5 Hz, 1H; CH), 2.36 2.41 (m, 1H; CH), 3.46 3.48 (m, 1H; CH), 3.63
(m, 1H; CH), 3.94 3.97 (m, 1H; CH), 4.04 ppm (dd, J=5.4, 2.7 Hz, 1H;
CH); 13C NMR (500 MHz, CDCl3): d=13.94, 22.51, 25.59, 25.82, 28.67,
31.64, 31.95, 43.29, 38.68, 71.46, 73.88, 78.96, 84.26 ppm.
0.46 (2a), 0.51 (2b), 0.54 (2c), 0.34 (2d). Spectroscopic data were in ac-
cordance to those described above.
(6S,7S,9S,10R)-2c: 1H NMR (500 MHz, CDCl3): d=0.87 0.89 (t, J=
6.7 Hz, 6H; 2CH3), 1.28 1.68 (brm, 16H; 8CH2), 1.90 (dd, J=14.1,
3.3 Hz, 1H; CH), 2.14 2.19 (m, 1H; CH), 3.56 3.60 (m, 1H; CH), 3.80
3.83 (m, 1H; CH), 3.98 4.01 ppm (m, 1H; CH); 13C NMR (500 MHz,
CDCl3): d=13.90, 13.93, 22.43, 22.52, 25.48, 25.89, 28.69, 31.63, 31.96,
33.29, 34.17, 70.98, 71.86, 79.89, 83.75 ppm.
(6R,7R,9S,10R)-2d: 1H NMR (500 MHz, CDCl3): d=0.86 0.89 (t, 6H;
J=5.3, 2CH3), 1.29.1.59 (brm, 16H; 8CH2), 1.84 (dd, J=13.2, 6.1 Hz,
1H; CH), 2.08 2.13 (m, 1H; CH), 3.82 3.86 (m, 2H; 2CH), 4.13 4.17
(m, 1H; CH), 4.27 ppm (brs, 1H; CH); 13C NMR (500 MHz, CDCl3): d=
13.93, 22.47, 22.50, 25.51, 25.90, 29.10, 31.83, 32.13, 32.35, 34.10, 71.98,
73.08, 79.94, 83.49 ppm.
Acknowledgement
This study was performed within the Spezialforschungsbereich ™Biokata-
lyse∫ (project no. F104). Financial support by the Fonds zur Fˆrderung
der wissenschaftlichen Forschung (Vienna) is gratefully acknowledged.
Special thanks go to H. Sterk and R. Saf (Graz) for their generous assis-
tance in NMR and MS spectroscopy, respectively.
[1]F. A. Alali, X.-X. Liu, J. L. McLaughlin, J. Nat. Prod. 1999, 62, 504
540.
(2R,3R,5R)-2-Pentyl-5-vinyl-tetrahydrofuran-3-ol ((2R,3R,5R)-10b) and
(2S,3S,5R)-2-pentyl-5-vinyl-tetrahydrofuran-3-ol ((2S,3S,5R)-10a): PPh3
(3.0 g, 11.44 mmol) and CBr4 (3.41 g, 10.28 mmol) were dissolved in an-
hydrous CH2Cl2 (30 mL). The stirred solution was cooled to 08C and
(R)-9 (919 mg, 5.46 mmol) was added dropwise. After stirring for 5 h, the
solution was concentrated, and pentane (30 mL) was added. The mixture
was filtered, and the filtrate was concentrated in vacuo. The crude resi-
due (370 mg) was epoxidized according to Method D. The residue
(128 mg) of the diastereomeric mixture of the epoxy-bromide was hydro-
lyzed in a mixture of water (6 mL) and THF (4 mL) under acidic condi-
tions (conc. H2SO4, 10 drops) at room temperature overnight. The solu-
tion was extracted with EtOAc (2î10 mL). The combined organic layers
were dried and evaporated. The residue was purified by flash chromatog-
raphy (petroleum ether/EtOAc, 10:1) to afford the diasteromeric mixture
of the cyclization products (2S,3S,5R)-10a and (2R,3R,5R)-10b (57 mg).
Rf1, Rf2 (petroleum ether/EtOAc, 1:1)=0.66, 0.61 (detection I). Spectro-
scopic data were in accordance to those previously reported.
[2]S. D. Jolad, J. J. Hoffman, K. H. Schram, J. R. Cole, M. S. Tempesta,
G. R. Kriek, R. B. Bates, J. Org. Chem. 1982, 47, 3151 3153.
[3]For a stereochemical model of polyether antibiotics, see: D. E.
Cane, W. D. Celmer, J. W. Westley, J. Am. Chem. Soc. 1983, 105,
3594 3600.
[4]D. O×Hagan, Nat. Prod. Rep. 1989, 6, 205 219.
[5]L. Zeng, Q. Ye, N. H. Oberlies, G. Shi, Z. M. Gu, K. He, J. L.
McLaughlin, Nat. Prod. Rep. 1996, 13, 275 306.
[6]Z. M. Gu, G. X. Zhao, N. H. Oberlies, L. Zeng, J. L. Laughlin, in
Recent Advances in Phytochemistry, Vol. 29, Plenum, New York,
1995, pp. 249 310.
[7]C. Gleye, S. Raynaud, R. Hocquemiller, A. Laurens, C. Fourneau,
L. Serani, O. Laprevote, F. Robot, M. Laboef, A. Fournet, A. R.
De Arias, B. Figadere, A. Cave, Phytochemistry 1998, 47, 749 754.
[8]For recent reviews for the synthesis of acetogenins, see: M. C. El-
liott, E. Williams, J. Chem. Soc. Perkin Trans. 1 2001, 2303 2340;
M. C. Elliott, J. Chem. Soc. Perkin Trans. 1 2000, 1291 1318; G. Ca-
siraghi, F. Zanardi, L. Battistini, G. Rassu, Chemtracts: 1998, 11,
803 827; B. Figadere, Acc. Chem. Res. 1995, 28, 359 365.
[9]S. C. Sinha, A. Sinha, S. C. Sinha, E. Keinan, J. Am. Chem. Soc.
1997, 119, 12014 12015.
[10]R. J. Capon, R. A. Barrow, J. Org. Chem. 1998, 63, 75 83.
[11]W. Ebenezer, G. Pattenden, Tetrahedron Lett. 1992, 33, 4053 4056.
[12]I. Paterson, P. A. Craw, Tetrahedron Lett. 1989, 30, 5799 5802.
[13]I. Paterson, I. Boddy, Tetrahedron Lett. 1988, 29, 5301 5304.
[14]P. Neogi, T. Doundoulakis, A. Yazbak, S. C. Sinha, S. C. Sinha, E.
Kainan, J. Am. Chem. Soc. 1998, 120, 11279 11284.
[15]S. C. Sinha, A. Sinha, S. C. Sinha, E. Keinan, J. Am. Chem. Soc.
1988, 120, 4017 4018.
[16]Y. Morimoto, T. Iwai, Y. Nishikawa, T. Kinoshita, Tetrahedron:
Asymmetry 2002, 13, 2641 2647.
[17]K. C. Nicolaou, T. Montagnon, S. A. Snyder, Chem. Commun. 2003,
551 564.
[18]S. F. Mayer, A. Steinreiber, R. V. A. Orru, K. Faber, Eur. J. Org.
Chem. 2001, 4537 4542.
[19]For a review on enzyme-initiated cascade reactions, see: S. F. Mayer,
W. Kroutil, K. Faber, Chem. Soc. Rev. 2001, 30, 332 339.
[20]K. D. Janda, C. G. Shevlin, R. A. Lerner, Science 1993, 259, 490
493.
[21]W. Kroutil, M. Mischitz, K. Faber, J. Chem. Soc. Perkin Trans. 1
1997, 3629 3636. For a mathematical treatment of enantio-conver-
gent kinetics, see: K. Faber, W. Kroutil, Tetrahedron: Asymmetry
2002, 13, 377 382.
[22]S. F. Mayer, A. Steinreiber, R. V. A. Orru, K. Faber, J. Org. Chem.
2002, 67, 9115 9121; S. F. Mayer, A. Steinreiber, M. Goriup, R. Saf,
K. Faber, Tetrahedron: Asymmetry 2002, 13, 523 528.
[23]M. Mischitz, C. Mirtl, R. Saf, K. Faber, Tetrahedron: Asymmetry,
1996, 7, 2041 2046.
[24]J. E. Baldwin, J. Chem. Soc. Chem. Commun. 1976, 734 738.
[25]M. Nardini, R. Rink, D. B. Janssen, B. W. Dijkstra, J. Mol. Catal. B
2001, 11, 1035 1042; J. Zou, B. M. Hallberg, T. Bergfors, F. Oesch,
M. Arand, S. L. Mowbray, T. A. Jones, Structure 2000, 8, 111 122.
[26]From eight pathways in total, two pathways each stereochemically
™cancel out∫ by leading to the same THF stereoisomer, thus a total
of four stereoisomers is theoretically possible.
(6S,7S,9R,10R)-6,9-Epoxypentadecane-7,10-diol
(6R,7R,9R,10R)-6,9-epoxypentadecane-7,10-diol ((6R,7R,9R,10R)-2b),
(6R,7R,9R,10S)-6,9-epoxypenta-decane-7,10-diol ((6R,7R,9R,10S)-2c),
and (6S,7S,9R,10S)-6,9-epoxypentadecane-7,10-diol ((6S,7S,9R,10S)-2d):
solution of the diasteromeric mixture of (2R,3R,5R)-10b and
((6S,7S,9R,10R)-2a),
A
(2S,3S,5R)-10a (57 mg, 0.31 mmol), TBDMSCl (60.3 mg, 0.40 mmol), and
imidazole (27.2 mg, 0.40 mmol) in CH2Cl2 (10 mL) was stirred at room
temperature overnight and then poured into a mixture of saturated
NaHCO3 and CH2Cl2. The mixture was stirred vigorously for 30 min, the
layers were separated, and the aqueous layer was extracted with CH2Cl2.
The combined organic phases were dried and concentrated. N-Methyl-
morpholine-N-oxide (70.6 mg, 0.60 mmol) and one crystal of OsO4 were
added to a solution of the crude residue (91 mg) in acetone (10 mL). The
mixture was stirred for 1.5 h at room temperature. Sodium sulfite
(230 mg, 1.83 mmol) was added and stirring was continued for 30 min.
The solution was extracted with EtOAc (2î10 mL). The combined or-
ganic layers were washed with water and concentrated under reduced
pressure. The residue (90 mg) was suspended in water (4 mL) to which
was added acetone (10 mL) followed by NaIO4 (98.2 mg, 0.46 mmol).
The mixture was stirred for 30 min at room temperature. Then the solu-
tion was extracted with EtOAc (2î10 mL), and the combined organic
layers were dried and evaporated. The residue (70 mg) was dissolved in
Et2O (10 mL). Pentylmagnesium bromide (0.5 mL of a 2m solution in
THF, 1 mmol) was added to the vigorously stirred solution and stirring
was continued for 5 h at room temperature. The reaction was quenched
by addition of H2O (5 mL) and Et2O (10 mL). The phases were separat-
ed, and the aqueous layer was extracted with Et2O (2î10 mL). The com-
bined organic phases were dried and evaporated. The residue (122 mg)
was dissolved in THF (10 mL), and Bu4N+FÀ (107 mg, 0.34 mmol) was
added. The reaction was stirred overnight at room temperature. The mix-
ture was quenched by addition of water (5 mL) and Et2O (10 mL). The
phases were separated, and the aqueous layer was extracted with Et2O
(2î10 mL). The combined organic phases were dried and evaporated.
The residue was purified by flash chromatography (petroleum ether/
EtOAc, 10:1) to afford the diasteromeric mixture of the cyclization prod-
ucts (6S,7S,9R,10R)-2a, (6R,7R,9R,10R)-2b, (6R,7R,9R,10S)-2c and
(6S,7S,9R,10S)-2d (10 mg). Rf (petroleum ether/EtOAc, 1:1; detection I)=
Chem. Eur. J. 2004, 10, 3467 3478
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3477

K. Faber et al.
FULL PAPER
[27]R. J. Capon, R. A. Barrow, J. Org. Chem. 1998, 63, 75 83.
[28]C. J. Sih, S.-H. Wu, Top. Stereochem. 1989, 19, 63 125; C.-S. Chen,
Y. Fujimoto, G. Girdaukas, C. J. Sih, J. Am. Chem. Soc. 1982, 104,
7294 7729; A. J. J. Straathof, J. A. Jongejan, Enzyme Microb. Tech-
nol. 1997, 21, 559 571.
[29]A. Wallner, H. Mang, S. M. Glueck, A. Steinreiber, S. F. Mayer, K.
Faber, Tetrahedron: Asymmetry 2003, 14, 2427 2432.
[30]SYBYL, Version 6.9. St. Louis, MO (USA), Tripos Inc., 2002.
[31]M. Saunders, K. N. Houk, Y. D. Wu, W. C. Still, M. Lipton, G.
Chang, W. C. Guida, J. Am. Chem. Soc. 1990, 112, 1419 1427.
[32]T. A. Halgren, J. Comput. Chem. 1999, 20, 720 729.
[33]T. A. Halgren, J. Comput. Chem. 1999, 20, 730 748.
[34]M. J. S. Dewar, E. G. Zoebisch, E. F. Healy, J. J. P. Stewart, J. Am.
Chem. Soc. 1985, 107, 3902 3909.
[35]AMPAC6.55, 1999 Semichem, Semi/7,7a, Shawnee, USA.
[36]A. D. Becke, J. Chem. Phys. 1993, 98, 5648 5642.
[37]C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785 789.
[38]Gaussian98 (RevisionA.7), M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski,
J. A. Montgomery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich,
J. M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J.
Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli,
C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q.
Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. Cioslowski, J. V. Ortiz, B. B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin,
D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara,
C. Gonzalez, M. Challacombe, P. M. W. Gill, B. G. Johnson, W.
Chen, M. W. Wong, J. L. Andres, M. Head-Gordon, E. S. Replogle,
J. A. Pople, Gaussian, Inc., Pittsburgh, PA, 1998.
[39]D. J. Tannor, B. Marten, R. Murphy, R. A. Friesner, D. Sitkoff, A.
Nicholls, B. Honig, M. Ringnalda, W. A. Goddard, III, J. Am.
Chem. Soc. 1994, 116, 11875 11882.
[40]B. Marten, K. Kim, C. Cortis, R. A. Friesner, R. B. Murphy, M. N.
Ringnalda, D. Sitkoff, B. Honig, J. Phys. Chem. 1996, 100, 11775
11788.
[41]Jaguar 4.1, Portland, Oregon, Schrˆdinger Inc., 2000.
[42]S. Antonczak, M. Ruiz-Lopez, J.-L. Rivail, J. Am. Chem. Soc. 1994,
116, 3912 3921; L. A. M. M. Barbosa, R. A. van Santen, J. Mol.
Struct. 2000, 497, 173 188; F. Haeffner, C.-H. Hu, T. Brinck, T.
Norin, J. Mol. Struct. 1999, 459, 85 93; B. Kallies, R. Mitzner, J.
Mol. Model. 1998, 4, 183 196; M. T. Nguyen, G. Raspoet, L. G.
Vanquickenborn, J. Chem. Soc. Perkin Trans. 2 1999, 813 820; G.
Raspoet, M. T. Nguyen, M. McGarraghy, A. F. Hegarty, J. Org.
Chem. 1998, 63, 6867 6877; G. Schmeer, P. Sturm, Phys. Chem.
Chem. Phys. 1999, 1, 1025 1030; M. Strajbl, J. Florian, A. Warshel,
J. Am. Chem. Soc. 2000, 122, 5354 5366; L. Wang, H. Zipse, Liebigs
Ann. 1996, 1501 1509; I. H. Williams, J. Am. Chem. Soc. 1987, 109,
6299 6307; S. Wolfe, C.-K. Kim, K. Yang, N. Weinberg, Z. Shi, J.
Am. Chem. Soc. 1995, 117, 4240 4260.
[43]W. Kroutil, M. Mischitz, K. Faber, J. Chem. Soc. Perkin Trans. 1
1997, 3629 3636.
Received: January 19, 2004
Published online: May 26, 2004
3478
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 3467 3478