Fluorescence assay and screening of epoxide opening by nucleophiles

Article abstract of DOI:10.1002/ejoc.200400024

Terminal epoxides such as 1 react with nucleophiles (H₂O, Cl^-, Br^-, N₃^-, and CN^-) at the primary oxirane carbon atom to give mostly anti-Markovnikov-type regioisomers 5a-d. The opening products of epoxide (R)-1 with chloride (5a), bromide (5b) and azide (5c) are oxidized by horse liver alcohol dehydrogenase and NAD⁺ to give the corresponding ketones 7a-c and, subsequently, umbelliferone 4 by β-elimination, leading to a fluorescence increase at λ_em = 460 ± 20 nm (λ_ex = 360 ± 20 nm). The epoxide hydrolysis products give no signal. We used this enantio- and chemo-selective fluorogenic assay for epoxide opening to search for catalytic antibodies for nucleophilic epoxide opening that were raised against 1,2-azidoammonium hapten 8, as a mimic for epoxide opening by azide, and against chloromethyl phosphonate hapten 9, as a mimic for the transition state of chlorohydrin formation. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejoc.200400024

FULL PAPER
Fluorescence Assay and Screening of Epoxide Opening by Nucleophiles
[a]
[b]
[a]
Fabrizio Badalassi,[a,b] Gerard Klein, Paolo Crotti, and Jean-Louis Reymond*
´
Keywords: Catalytic antibodies / Enzymes / Epoxides / Fluorescence assay
Terminal epoxides such as 1 react with nucleophiles (H2O,
Cl−, Br−, N3−, and CN−) at the primary oxirane carbon atom to
give mostly anti-Markovnikov-type regioisomers 5a−d. The
We used this enantio- and chemo-selective fluorogenic assay
for epoxide opening to search for catalytic antibodies for nu-
cleophilic epoxide opening that were raised against 1,2-azi-
opening products of epoxide (R)-1 with chloride (5a), brom- doammonium hapten 8, as a mimic for epoxide opening by
ide (5b) and azide (5c) are oxidized by horse liver alcohol
dehydrogenase and NAD+ to give the corresponding ketones
7a−c and, subsequently, umbelliferone 4 by β-elimination,
azide, and against chloromethyl phosphonate hapten 9, as a
mimic for the transition state of chlorohydrin formation.
leading to a fluorescence increase at λem = 460 20 nm (λex
=
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
360 20 nm). The epoxide hydrolysis products give no signal.
Germany, 2004)
Introduction
against two transition state analogs of epoxide opening re-
actions.
Enzymes are versatile catalysts that are used in various
areas, such as industrial processes, diagnostics, biomedical
research and fine chemical synthesis. The development of
new enzymes follows a discovery-oriented approach that
consists of searching for the occurrence of catalysis in pos-
sible sources of a desired enzyme, particularly collections of
microorganisms.[1] The design aspect of such experiments
concentrates on the discovery techniques themselves rather
than on the direct structural design of the catalyst. These
aspects involve molecular biological methods, such as gen-
ome mining and gene shuffling, to define or generate pos-
sible targets of interest,[2] and chemical techniques of en-
zyme assays for detecting the desired catalytic activity as
selectively and sensitively as possible under the operating
conditions that are typical of enzymes, i.e., a buffered aque-
ous environment.[3] While enzyme assays are usually devel-
oped for reactions for which an enzyme already exists, it
may be of interest to also provide assays for reaction of
synthetic interest for which no enzymes are known yet.
Such assays might then provide an opportunity for finding
Epoxides are versatile chiral intermediates in a broad var-
iety of syntheses. Their meaningful utilization depends on
the availability of stereo- and regioselective opening reac-
tions with nucleophiles; proper manipulation of reagents
and reaction conditions allows the regioselective formation
of β-halohydrins and amino- and azido alcohols by means
of metal salt-catalyzed ring opening reactions with appro-
priate nucleophiles.[4] If an enantioselective catalyst is avail-
able, the opening reaction can be utilized for a kinetic reso-
lution of racemic epoxides. This process is possible when
using several metal catalysts having chiral ligands, some of
which have been discovered by combinatorial ligand syn-
thesis.[5] In addition to these chemical catalysts, epoxide hy-
drolases enzymes have been discovered that effect regio- and
enantioselective hydrolyses of epoxides.[6] These enzymes,
however, operate by nucleophilic attack on the epoxide by
a carboxylate side chain, followed by hydrolysis, so that
only diols can be obtained.
Catalytic antibodies can also catalyze enantioselective hy-
such enzymes. Herein, we report a fluorogenic assay that drolysis of epoxides, but the mechanism is different: there
detects the chemoselective opening of epoxides by nucleo- is a direct attack of a water nucleophile on the epoxide.[7]
philes such as halides and azides, a process for which only The mechanism of the reaction catalyzed by limonene epox-
a single enzyme has been described to date. The assay was ide hydrolase from Rhodococcus erythropolis has also been
implemented to search for catalysis in antibodies raised reported to involve direct attack by water as a nucleophile.[8]
This different reaction mechanism suggests the possibility
[a]
Departement für Chemie und Biochemie, Universität Bern,
Freiestrasse 3, 3012 Bern, Switzerland,
of obtaining the products of epoxide opening with nucleo-
philes other than water, such as halides or azides, using a
biocatalyst. Indeed, there have been several reports of bio-
catalytic openings of epoxides with nucleophiles other than
water.[9] Recently, a haloalkane dehalogenase enzyme was
described that catalyzes the nucleophilic opening of epox-
Fax: (internat.) ϩ41-31-6318057
E-mail: jean-louis.reymond@ioc.unibe.ch
[b]
Dipartimento di Chimica Bioorganica
e
Biofarmacia,
`
Universita di Pisa,
56126 Pisa, Italy
Fax: (internat.) ϩ39-050-2219660
Eur. J. Org. Chem. 2004, 2557Ϫ2566
DOI: 10.1002/ejoc.200400024
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2557

F. Badalassi, G. Klein, P. Crotti, J.-L. Reymond
FULL PAPER
ides by azides.[10] We became interested in finding a generic thereby enabling the formation of the fluorescent product
assay for such reactions to facilitate the discovery of simi- umbelliferone 4, via ketones 7aϪd, as the assay signal for
lar catalysts.
the epoxide opening. Such an assay would be chemoselec-
tive for nucleophilic opening of the epoxide by halide, azide,
and cyanide nucleophiles against the formation of diol 2
because the latter would be oxidized by the alcohol de-
hydrogenases at the primary hydroxyl group, which would
not lead to a fluorescence signal.[17] We decided, therefore,
to explore the kinetics of formation and oxidation of β-
hydroxy derivatives 5aϪd in an aqueous buffered environ-
ment using an alcohol dehydrogenase as a co-oxidant.
Results
Assay Design
We recently reported an assay for epoxide hydrolysis reac-
tions that is applicable to epoxide hydrolase enzymes.[11]
The assay uses optically pure substrates, such as epoxide
1. A fluorescent signal is produced upon formation of the
hydrolysis product diol 2 in the presence of sodium per-
iodate and BSA, which effect the oxidative cleavage of the
diol to form aldehyde 3 and, subsequently, umbelliferone 4
by a β-elimination reaction (Scheme 1). This reaction se-
quence is suitable for a variety of enzyme assays, including
lipases,[12] aldolases,[13] and proteases.[14] The assay was ap-
plied across a broad variety of epoxides in the context of
enzyme fingerprinting experiments.[15] This assay would,
however, not provide any signal if the epoxide was opened
by a nucleophile other than water or ammonia, such as a
halide, azide or cyanide, because the resulting β-hydroxy de-
rivatives 5aϪd would not be oxidized by sodium periodate.
Since we had shown previously that the secondary alcohol
(S)-6 was an excellent substrate for alcohol dehydrogen-
ases,[16] we envisioned that β-hydroxy derivatives 5aϪd
might be suitable substrates for alcohol dehydrogenases,
Substrate Synthesis
Opening of the known racemic epoxide (Ϯ)-1 with HCl/
CHCl3, HBr/CHCl3, NaN3/NH4Cl/MeOH/H2O, and KCN/
LiClO4/MeCN afforded the β-hydroxy derivatives 5aϪd
(Scheme 2). Chlorohydrins (R)-5a and (S)-5a were prepared
by the ring opening reactions of the known epoxides (R)-1
and (S)-1 with NH4Cl/LiClO4/MeCN.
Scheme 2. Synthesis of epoxide opening products; reagents and
conditions: a) CHCl3, aq. HCl, 25 °C, 30 min (77%); b) CHCl3,
aq. HBr, 25 °C, 30 min (70%); c) NaN3, NH4Cl, MeOH/H2O, 80
°C, 18 h (67%); d) CH3CN, KCN, LiClO4, 70 °C, 18 h (50%); e)
CH3CN, NH4Cl, LiClO4, 65 °C, 18 h (66%); f) same as e (62%)
Fluorescence Assay
All epoxides and diols were assayed for their fluorogenic
properties in aqueous buffer at 0.1 m concentration in the
Scheme 1. Principle of fluorescence assay for opening of epoxide 1
with nucleophiles
2558
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 2557Ϫ2566

Fluorescence Assay and Screening of Epoxide Opening by Nucleophiles
FULL PAPER
Table 1. Fluorogenic reactions of epoxides and alcohols with alcohol dehydrogenases
[a]
ADH, µg/mL,Vapp
1
(R)-1
(S)-1
2
5a
(R)-5a
(S)-5a
5b
5c
5d
Horse liver, 100[b]
Yeast,100[b]
2.9
1.9
4.3
3.8
0.00
3.2
1.2
0.24
0.48
16.5
4.3
15
175
2.3
30
502
3.5
77
8.2
2.0
1.1
1090
0.8
39
505
31
39
4.2
1.3
4.7
Therm. brockii, 50[c]
[a]
Apparent rate of release of umbelliferone (pM sϪ1). Conditions: 20 m aq. borate buffer, pH 8.8, 2 mg/mL BSA, 100 µM substrate,
enzyme and: see footnotes[b,c]
.
1 m NADϩ. 1 m NADPϩ, 30 °C. 200 µL assays were monitored in the individual wells of round-
[b]
[c]
bottom polypropylene 96-well plates (Costar) using a Cytofluor II Fluorescence Plate Reader (Perseptive Biosystems, filters λex ϭ 360 Ϯ
20 nm, λem ϭ 460 Ϯ 20 nm). Fluorescence (arbitrary units) was converted to umbelliferone concentration according to a calibration curve
with pure 4 in the same buffer containing BSA.
presence of alcohol dehydrogenases and BSA (Table 1, Fig- fluorogenic reaction that was proportionate to the bromide
ure 1). The epoxides were found to be stable under these concentration was observed with epoxide (R)-1, which indi-
conditions and gave no fluorescence signal. As expected, cates the formation of bromohydrin (R)-5b under the reac-
diol 2 gave only a very weak fluorescence signal. Only horse tion conditions.
liver alcohol dehydrogenase (HLDH) in the presence of
NADϩ proved a useful oxidation system, whereby only the
products of epoxide opening with halide and azide nucleo-
philes (compounds 5aϪc) gave a strong fluorescence signal.
The product of ring opening with cyanide (5d) gave only a
very weak fluorescence signal. In the optically pure series,
only the chlorohydrin (R)-5a gave a fluorescence signal. The
weak fluorescence signal observed with its enantiomer (S)-
5a is probably caused by the ca. 6% of the R enantiomer
present in the sample.
Figure 2. Fluorogenic reaction of epoxides (R)-1 and (S)-1 with
NaBr, HLDH, and BSA; conditions: 100 µM epoxide (R)-1 or (S)-
1 in 20 m aqueous borate buffer, pH 8.8, 100 µg/mL horse liver
alcohol dehydrogenase, 1 m NADϩ, 2 mg/mL BSA, 30 °C; the
rate of release of umbelliferone V (pM/s) was calculated from the
fluorescence signal obtained; see also table legend
Hapten Design for Catalytic Antibodies
Next, we turned our attention to the application of this
assay for screening antibodies for possible catalysis of the
reaction.[18] We were interested in haptens tailored for epox-
ide opening of substrate 1 by nucleophiles, and aimed at
two complementary designs. Our first attempt was directed
at catalyzing the enantioselective azide opening of epoxide
(R)-1 to azido alcohol (R)-5c, because it would be detect-
able using our ADH-coupled assay (Scheme 3). We envi-
sioned that the naphthalene-derived amino-azide 8 would
be suitable as a transition state analog for this reaction.
This hapten is essentially product-like, with the important
difference that the positive charge in the ammonium group
Figure 1. Fluorescence signals with epoxide opening products and
alcohol dehydrogenase. Conditions: 100 µM substrate 2, 5aϪd in
20 m aqueous borate buffer, pH 8.8, 100 µg/mL horse liver al-
cohol dehydrogenase, 1 m NADϩ, 2 mg/mL BSA, 30 °C; fluor-
escence signals (arbitrary units) were recorded with filters λex
360 Ϯ 20 nm, λem ϭ 460 Ϯ 20 nm; see also table legend
ϭ
The assay was validated by following the spontaneous replaces the hydroxyl group in the product. The positive
formation of bromohydrin 5b from epoxide 1 in the pres- charge was expected to induce a catalytic acidϪbase func-
ence of bromide, the most nucleophilic halide in the series. tional group capable of activating the epoxide substrate by
The reactivities of epoxides (R)-1 and (S)-1 and of bromo- protonation. Furthermore, the positive charge in the hapten
hydrin 5b were assayed in the presence of HLDH and would help to form a combining site having reduced affinity
NADϩ upon increasing the concentration of bromide from for the neutral product, thereby relieving product inhi-
zero to 0.15  (Figure 2). The activity of HLDH, as meas- bition.
ured by a reference substrate, was affected only moderately
Our second hapten was designed for catalysis of the epox-
by the addition of NaBr and decreased linearly by ca. 20% ide opening of (R)-1 by chloride to form the fluorogenic
as the NaBr concentration increased from 0 and 0.15 . A chlorohydrin (R)-5a. We selected the chloromethyl phos-
Eur. J. Org. Chem. 2004, 2557Ϫ2566
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2559

F. Badalassi, G. Klein, P. Crotti, J.-L. Reymond
FULL PAPER
which was converted into the N-hydroxysuccinimide ester
and conjugated to carrier proteins keyhole lympet hemocy-
anin (KLH) and bovine serum albumin (BSA). Both pro-
tein conjugates were finally exposed at pH 14 for 16 h to
afford the corresponding conjugated hapten 8 bearing a pri-
mary amine functionality.
Scheme 3. Transition state analog design of haptens for epoxide
opening reactions
phonate 9 as a suitable anionic, product-like transition state
analog. The chloromethyl group acts as a mimic of the in-
coming chloride ion, while the negatively charged phos-
phonate represents the initially formed alkoxide. The phos-
phonate function would also help induce an acidϪbase
catalytic group in the antibody combining site.
Hapten Synthesis
The azido amine hapten 8 was prepared from 2,7-di-
hydroxynaphthalene 10 (Scheme 4). O-Alkylation with 4-
bromo-1-butene gave mono-ether 11 in low yield (18%), in
spite of the several modifications we attempted. A second
alkylation with tert-butyl 4-bromohexanoate gave the 2,7-
di-O-substituted naphthalene 12. Asymmetric Sharpless di-
hydroxylation with AD mix α afforded the diol (S)-13,
which was transformed into the corresponding epoxide (S)-
15 via the chloroacetate (S)-14.[19] Regioselective opening
of the epoxide at the terminal carbon atom using NaN3 in
DMF gave azido alcohol (S)-16. Reaction of azido alcohol
(S)-16 with PPh3 in MeCN at 80 °C, followed by derivatiz-
ation with ethyl chloroformate, gave the aziridine carbamate
(R)-17. Basic deprotection with tBuOK/tBuOH afforded
pure aziridine (R)-18. Treatment with NaN3/NH4Cl gave
Scheme 4. Synthesis of azidoamine hapten 8 as the trifluoroaceta-
mide 21; reagents and conditions: a) NaH, DMF, BrCH2CH2CHϭ
CH2, 25 °C, 48 h (18%); b) NaH, DMF, Br(CH2)5CO2tBu, 25 °C,
20 h (83%); c) AD mix α, tBuOH/H2O, 0 °C, 48 h (83%, 94% ee);
d) MeC(OMe)3, CH2Cl2, PPTS cat., 25 °C, 1.5 h, then TMSCl, 25
°C, 18 h (99%); e) C6H6, tBuOK, 25 °C, 20 h (81%); f) NaN3,
DMF, 80 °C, 22 h (78%); g) CH3CN, PPh3, 25 °C 1 h, then 80 °C,
15 h, then Et3N, ClCO2Et, Et2O, 0 °C, 3 h (42%); h) tBuOK,
tBuOH, 25 °C, 32 h (98%); i) NaN3, tBuOH/H2O, NH4Cl, 60 °C,
20 h (83%); j) CH2Cl2, TFAA, Et3N, 0 °C, 2 h (83%); k) CF3CO2H/
CH2Cl2, 25 °C, 1.5 h (100%)
Hapten 9 was prepared in three steps from nitrophenethyl
azido amine (R)-19, which was acylated to give the trifluo- alcohol by esterification with chloromethyl phosphonyl di-
roacetamide (R)-20. The tert-butyl ester was removed by chloride, reduction of the nitro group, and acylation with
treatment with trifluoroacetic acid to give acid (R)-21, glutaric anhydride. Details of the synthesis have been re-
2560
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 2557Ϫ2566

Fluorescence Assay and Screening of Epoxide Opening by Nucleophiles
FULL PAPER
Table 2. Data for immunization with haptens 9 and 8 and catalysis HTS data [the following reactions were assayed: (R)-1 with azide
(hapten 8) or chloride (hapten 9) and HLDH/NADϩ/BSA; pivalate ester hydrolysis with 22]
Hapten
Mouse L[a]
Mouse R[a]
Mouse LL[a]
Mouse LR[a]
9
Serum titer[b]
Nr. of binders[c]
Monoclonals[d]
Catalytic[e]
12800
36
0
0
19200
45
0
0
25600
160
16
11
1600
7
25600
50
1
0
800
0
8
Serum titer[b]
Nr. of binders[c]
Monoclonals[d]
Catalytic[e]
1600
1600
6
0
0
0
0
0
0
0
0
0
[a]
[b]
Mouse code according to ear marks.
Dilution factor of blood serum (after immunization with hapten-KLH conjugates) for 50%
reduction of ELISA signal against the hapten-BSA conjugate. A high number indicates strong immune response. [c] Number of hybridoma
cell lines testing positive for binding by ELISA against the hapten-BSA conjugate after fusion and that were grown up to 5 mL for
[d]
catalysis testing.
Number of fully subcloned and stabilized hybridoma. The purified antibodies were also assayed by RP-HPLC for
[e]
hydrolysis and chlorohydrin formation with 24 and epoxide formation with 25 and for hydrolysis of rac-1 with NaIO4/BSA. Number
of fully subcloned hybridoma producing catalytic antibodies. Only activities for the hydrolysis of pivalate esters 22 and 23 were detected.[20]
ported in relation to the esterolytic activity of antibodies
against this hapten.[20]
Antibody Generation and Screening
The KLH-conjugates of the haptens were used for immu-
nization following strandard protocols.[21] Hybridoma cell
lines were generated by fusing spleen cells of the immunized
animals with myeloma. The azido-amine hapten 8 was only
poorly immunogenic. Because of the low serum titers ob-
served, only 13 hapten-binding hybridoma cell lines were
obtained. By contrast, the chloromethyl phosphonate
hapten 9 proved strongly immunogenic and a total of 291
initial clones were obtained after fusion (Table 2). Cell lines
producing hapten-specific antibodies, as judged by ELISA
against the BSA-conjugates, were initially assayed for epox-
ide opening reactions with either chloride (0.15 ) or azide
(0.15 , buffered at pH 8.8) using substrate (R)-1 and the
dehydrogenase-coupled assay described above (15 h incu-
bation at 25 °C). There was no observable catalysis using
any of the samples tested. The cell lines derived from hapten
9 were also tested for hydrolysis of the fluorogenic pivalate
22, which led to the identification and successful subcloning
Scheme 5. Additional substrates tested with antibodies against
of sixteen stable cell lines, nine of which produced catalytic hapten 9; the pivalates 22 and 23 are substrates for various anti-9
antibodies (kcat/kuncat ϭ 103), but 24 and 25 are not
esterolytic antibodies (kcat/kuncat ഠ 103 for the hydrolysis of
22 and 23).[20] Purified antibodies from these sixteen cell
lines were also tested for hydrolysis of the epoxide rac-1
using the periodate-coupled procedure. These antibodies
yond doubt that this substrate is well suitable as a probe of
enzymatic activity. Similarly, the reactivity of the various β-
hydroxy derivatives (R)-5aϪd with horse liver alcohol de-
hydrogenase demonstrate that the primary opening prod-
ucts of epoxide 1 are well suited for enzyme reactions. Al-
though no direct enzyme reaction was observed for forming
a halohydrin from epoxide 1, the chemical reaction ob-
served in the presence of bromide in turn establishes the
validity of the high-throughput screening fluorescence assay
were also tested by HPLC for chlorohydrin formation from
epoxide 24 and epoxide formation from chlorohydrin 25,
because these substrates represent exact structural analogs
of hapten 9 (Scheme 5). None of these reactions were cata-
lyzed by the antibodies.
Discussion
The fluorescence assay for epoxide hydrolysis based on for epoxide opening based on epoxide (R)-1. Other formats
substrate 1, which has been used by us and others to meas- have been reported for assaying epoxide hydrolysis in
ure the activity of epoxide hydrolases, has established be- HTS.[11,22] In contrast, the enzyme-coupled assay reported
Eur. J. Org. Chem. 2004, 2557Ϫ2566
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2561

F. Badalassi, G. Klein, P. Crotti, J.-L. Reymond
FULL PAPER
here is unprecedented for detecting epoxide opening reac- escent product, and only the products of opening with chlo-
tions with nucleophiles other than water. ride, bromide, and azide (5aϪc) are detected, which, there-
Interestingly, the epoxideϪhalohydrin equilibrium can be fore, allows chemoselective screening. The assay was im-
readily manipulated in aqueous buffer by adjusting the hal- plemented to search for high-throughput screening of hy-
ide concentration. An assay for epoxide closure from the bridoma cell culture supernatant with the goal of
halohydrin might be possible based on the detection of discovering catalytic antibodies, but, unfortunately, without
chloride ions, but such an assay might be difficult to im- success. This assay should be generally useful to assay cata-
plement because chloride ions are present in large concen- lysts capable of opening epoxides with nucleophiles.
trations in most enzyme preparations as buffer or culture
medium components. Here we used a direct HPLC detec-
tion system in the context of analyzing the reactivity of a
few catalytic antibodies against hapten 9 for epoxide forma-
Experimental Section
tion from halohydrin 25. Given that an epoxide hydrolase
from Aspergillus niger is now commerically available — it
General: IR spectra were obtained using a Mattson 3000 FTIR
spectrophotometer. 1H and 13C NMR spectra were determined
converts both enantiomers of epoxide 1 rapidly to the corre-
using a Bruker AC-200 spectrometer from CDCl3 solutions using
Me4Si as the internal standard. Optical rotations were measured
using a PerkinϪElmer 241 digital polarimeter and a 1-dm cell. Di-
rect-phase HPLC analyses were carried out on a Waters 600 E Sys-
sponding diol 2[11,23] — an enzyme/periodate/BSA-coupled
procedure should be possible to test epoxide formation
from both enantiomers of halohydrins 5a/b. A correspond-
ing back-titration protocol should allow us to extend the
assay to any number of halohydrins as long as the corre-
sponding epoxides are hydrolyzed by the epoxide hydrolase.
Our attempts to discover epoxide-reactive catalytic anti-
bodies by immunization against tailored haptens were not
successful. The azido-amine 8 was clearly too weakly
immunogenic to generate a sufficient number of hybridoma,
which precludes any discussion of whether this was an accu-
rate transition state analog of the targeted reaction or not.
The low immunogenicity of the KLH-conjugates of 8 is not
due to the trifluoroacetamide deprotection strategy we
chose, because a similar strategy has been used successfully
in preparing two other unrelated haptens.[24] In the case of
the chloromethyl phosphonate hapten 9, we did isolate sev-
eral antibodies that catalyze the hydrolysis of pivalates, such
as 22 and 23, which clearly establishes that the hapten was
able to induce a catalytic binding pocket.[20] Also, one
might have expected that an epoxide opening reaction could
be triggered by the same factors that are necessary to in-
duce ester hydrolysis. In the present series, the catalytic anti-
bodies against hapten 9 were not irreversibly inhibited by
incubation with epoxide 1 or hapten 9 itself, which implies
that a reactive nucleophilic side-chain was not present in
the antibody combining sites. To date, an intramolecular
hydroxy-epoxide cyclization reported by the Janda group[25]
and an epoxide hydrolysis reaction from our group[7] re-
main the only successful examples of catalytic antibodies
for epoxide opening reactions.
tem Controller equipped with a Waters 990 Photodiode Array De-
tector. Reverse-phase HPLC was performed on a Waters 660 Con-
troller equipped with a Waters 996 Photodiode Array Detector and
using three different solvents: A (0.1% TFA in H2O), B (H2O/
MeCN, 1:1), and C (H2O). The ee of epoxides (R)- and (S)-1 were
determined directly by analysis on a chiral HPLC column OD-H
[Daicel: 25 cm ϫ 0.46 cm (i.d)] using hexane/iPrOH (8:2) as the
eluent: the chlorohydrins (R)- and (S)-5a did not separate under
these conditions. Values of ee for the diol (S)-13 and azido amine
(R)-19 were determined by HPLC on a Chiracel OD-H chiral col-
umn [25 cm ϫ 0.46 cm (i.d.)], using hexane/iPrOH mixtures. All
reactions were followed by TLC on Alugram SIL G/UV254 silica
gel sheets (MachereyϪNagel) with detection by UV and/or spray-
ing with 20% phosphomolybdic acid in EtOH. Benzene and
CH2Cl2 were distilled under nitrogen from sodium/benzophenone
ketyl and CaH2, respectively.
(؎)-7-(4-Chloro-3-hydroxybutyloxy)-2H-1-benzopyran-2-one (5a): A
solution of epoxide (Ϯ)-1[15] (0.10 g, 0.43 mmol) in CHCl3 (3.0 mL)
was treated with 36% aqueous HCl (3.0 mL) and the reaction mix-
ture was vigorously stirred for 30 min at room temperature. Evap-
oration of the washed (saturated aqueous NaHCO3 and water) or-
ganic solution afforded a crude reaction product that was subjected
to flash chromatography (hexane/EtOAc, 6:4; Rf ϭ 0.19) to give
pure (Ϯ)-5a (0.089 g, 77% yield) as a solid. M.p. 59Ϫ61 °C. IR
(nujol): ν˜ ϭ 3403 (br. s, OH), 1692 (s, CϭO) cmϪ1. 1H NMR: δ ϭ
7.63 (d, J ϭ 9.8 Hz, 1 H), 7.36 (d, J ϭ 8.8 Hz, 1 H), 6.86Ϫ6.79
(m, 2 H), 6.24 (d, J ϭ 9.3 Hz, 1 H), 4.31Ϫ4.07 (m, 3 H), 3.72 (dd,
J ϭ 11.2, 3.9 Hz, 1 H), 3.60 (dd, J ϭ 11.2, 6.8 Hz, 1 H), 2.21Ϫ1.95
(m, 2 H) ppm. 13C NMR: δ ϭ 162.50, 161.98, 156.35, 144.19,
129.46, 113.67, 113.44, 113.26, 102.06, 69.02, 65.62, 50.74, 34.14
ppm. C13H13ClO4 (268.7): calcd. C 58.11, H 4.88; found C 58.27,
H 5.19.
Conclusion
(R)-7-(4-Chloro-3-hydroxybutyloxy)-2H-1-benzopyran-2-one (5a): A
solution of epoxide (R)-1 (0.050 g, 0.21 mmol) in CH3CN (0.2 mL)
containing NH4Cl (0.017 g, 0.31 mmol) and LiClO4 (0.033 g,
0.31 mmol) was stirred at 65 °C for 18 h. After cooling, dilution
with diethyl ether, and evaporation of the washed (saturated aque-
ous NaCl) organic solution afforded a crude reaction product that
was purified by flash chromatography (hexane/EtOAc, 6:4; Rf ϭ
0.19) to give pure (R)-5a (0.037 g, 66% yield, 96% ee[26]) as a semi-
solid. [α]2D0 ϭ ϩ8.9 (c ϭ 0.55, MeOH). C13H13ClO4 (268.7): calcd.
In summary, the opening of epoxide 1 with nucleophiles
is a fluorogenic reaction when coupled with ADH-catalyzed
alcohol oxidation and BSA-catalyzed β-elimination. The as-
say is enantioselective in the sense that only opening prod-
ucts from enantiomer epoxide (R)-1 give rise to a fluor-
escence signal, which represents a limitation of the assay.
Most importantly, the assay is chemoselective because the
simple hydrolysis product, the diol 2, does not yield a fluor- C 58.11, H 4.88; found C 58.20, H 5.15.
2562  2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjoc.org
Eur. J. Org. Chem. 2004, 2557Ϫ2566

Fluorescence Assay and Screening of Epoxide Opening by Nucleophiles
(S)-7-(4-Chloro-3-hydroxybutyloxy)-2H-1-benzopyran-2-one (5a): β-chlorohydrins (R)- and (S)-5a have the same absolute configura-
FULL PAPER
Proceeding as described above for the preparation of (R)-5a, epox-
ide (S)-1 (0.050 g, 0.21 mmol) afforded a crude reaction product
that was purified by flash chromatography (hexane/EtOAc, 6:4) to
tion as the starting epoxide.
7-O-(3-Butenyl)-2-hydroxynaphthalene (11): A suspension of NaH
(2.08 g of a 60% dispersion in mineral oil, 52.0 mmol) in anhydrous
DMF (80 mL) was treated with 2,7-dihydroxynaphthalene (10;
6.40 g, 40.0 mmol) and then the reaction mixture was stirred at
room temp. for 1 h. 4-Bromo-1-butene (8.10 g, 60.0 mmol) was ad-
ded and stirring was prolonged for 48 h. Dilution with water, ex-
traction with Et2O, and evaporation of the washed (saturated aque-
ous NaCl) diethyl ether extracts afforded a crude reaction product
that was subjected to flash chromatography. Elution with hexane/
EtOAc (7:3) afforded pure compound 11 (1.53 g, 18% yield) as a
solid. M.p. 54Ϫ56 °C. 1H NMR (CDCl3): δ ϭ 7.66 (d, J ϭ 8.8 Hz,
2 H), 7.16Ϫ6.66 (m, 4 H), 6.11Ϫ5.80 (m, 1 H), 5.30Ϫ5.00 (m, 2
H), 4.11 (t, J ϭ 6.8 Hz, 2 H), 2.61 (q, J ϭ 6.8 Hz, 2 H) ppm. 13C
NMR (CDCl3): δ ϭ 158.2, 154.6, 136.7, 130.2, 130.1, 129.9, 125.1,
117.7, 117.2, 115.9, 109.5, 106.4, 67.9, 34.3 ppm. C14H14O2 (214.3):
calcd. C 78.48, H 6.59; found C 78.21, H 6.69.
give pure (S)-5a (0.035 g, 62% yield, 87% ee) as a semisolid. [α]2D0
ϭ
Ϫ8.6 (c ϭ 0.86, MeOH). C13H13ClO4 (268.7): calcd. C 58.11, H
4.88; found C 58.05, H 4.71.
(؎)-7-(4-Bromo-3-hydroxybutyloxy)-2H-1-benzopyran-2-one (5b):
Proceeding as described above for the preparation of (Ϯ)-5a, the
reaction of epoxide (Ϯ)-1 (0.116 g, 0.50 mmol) in CHCl3 (3.0 mL)
with 48% aqueous HBr (2.0 mL) afforded a crude reaction product
that was subjected to flash chromatography (hexane/EtOAc, 6:4;
Rf ϭ 0.19) to give pure (Ϯ)-5b (0.109 g, 70% yield) as a solid. M.p.
73Ϫ75 °C. IR (nujol): ν˜ ϭ 3410 (br. s, OH), 1691 (s, CϭO) cmϪ1
.
1H NMR: δ ϭ 7.63 (d, J ϭ 9.4 Hz, 1 H), 7.37 (d, J ϭ 9.3 Hz, 1
H), 6.87Ϫ6.80 (m, 2 H), 6.25 (d, J ϭ 9.5 Hz, 1 H), 4.32Ϫ4.02 (m,
3 H), 3.61 (dd, J ϭ 10.4, 3.7 Hz, 1 H), 3.47 (dd, J ϭ 10.3, 6.8 Hz,
1 H), 2.22Ϫ1.90 (m, 2 H) ppm. 13C NMR: δ ϭ 162.5, 161.9, 156.4,
144.1, 129.5, 113.7, 113.4, 113.3, 102.1, 68.6, 65.7, 40.5, 35.0 ppm.
C13H13BrO4 (313.1): calcd. C 49.86, H 4.18; found C 50.09, H 4.51.
7-O-(3-Butenyl)-2-O-(tert-butoxycarbonylpentyl)naphthalene (12): A
solution of compound 11 (1.20 g, 5.60 mmol) in anhydrous DMF
(5 mL) was added dropwise under argon at 0 °C to a stirred suspen-
sion of NaH (0.224 g of a 60% dispersion in mineral oil,
5.60 mmol) in anhydrous DMF (5.0 mL). After stirring for 1 h at
room temp., tert-butyl 6-bromohexanoate (1.40 g, 5.6 mmol) was
added and the resulting reaction mixture was stirred at the same
temperature for 20 h. Dilution with Et2O (40 mL) and evaporation
of the washed (saturated aqueous NaCl) diethyl ether extracts af-
forded a crude product (1.98 g) that was subjected to flash chroma-
tography. Elution with hexane/EtOAc (8:2; Rf ϭ 0.53) afforded
pure ester 12 (1.79 g, 83% yield) as a yellow liquid. IR: ν˜ ϭ
(؎)-7-(4-Azido-3-hydroxybutyloxy)-2H-1-benzopyran-2-one (5c): A
solution of epoxide (Ϯ)-1 (0.464 g, 2.0 mmol) in a MeOH/H2O
mixture (8:1, 9.0 mL) was treated with NaN3 (0.60 g, 9.2 mmol)
and NH4Cl (0.214 g, 4.0 mmol) and then the reaction mixture was
stirred at 80 °C for 18 h.[19,27] After cooling, dilution with diethyl
ether and evaporation of the washed (water) organic solution af-
forded a solid residue that was purified by flash chromatography
(hexane/EtOAc, 6:4; Rf ϭ 0.17) to give pure (Ϯ)-5c (0.37 g, 67%
yield) as a solid. M.p. 58Ϫ60 °C. IR (nujol): ν˜ ϭ 3297 (br. m, OH),
1
2093 (m, ϪN3), 1697 (s, CϭO) cmϪ1. H NMR: δ ϭ 7.63 (d, J ϭ
9.3 Hz, 1 H), 7.35 (d, J ϭ 8.8 Hz, 1 H), 6.71Ϫ6.89 (m, 2 H), 6.23
(d, J ϭ 9.8 Hz, 1 H), 4.21Ϫ4.00 (m, 3 H), 3.47 (dd, J ϭ 12.4, 4.1
Hz, 1 H), 3.37 (dd, J ϭ 12.2, 6.8 Hz, 1 H), 2.15Ϫ1.86 (m, 2 H)
ppm. 13C NMR: δ ϭ 162.5, 162.0, 156.3, 144.2, 129.5, 113.6, 113.5,
113.2, 102.0, 68.5, 65.7, 57.7, 34.0 ppm. C13H13N3O4 (275.3): calcd.
C 56.73, H 4.76, N 15.27; found C 56.34, H 4.40, N 15.04.
1
1726 cmϪ1 (CϭO). H NMR (CDCl3): δ ϭ 7.64 (d, J ϭ 8.8 Hz, 2
H), 7.05Ϫ6.93 (m, 4 H), 6.06Ϫ5.86 (m, 1 H), 5.30Ϫ5.12 (m, 2 H),
4.16Ϫ4.03 (m, 4 H), 2.61 (q, J ϭ 6.8 Hz, 2 H), 2.28 (t, J ϭ 7.3 Hz,
2 H), 1.97Ϫ1.74 (m, 2 H), 1.74Ϫ1.61 (m, 2 H), 1.61Ϫ1.47 (m, 2
H), 1.46 (s, 9 H) ppm. 13C NMR (CDCl3): δ ϭ 173.7, 158.3, 158.2,
136.7, 135.21, 129.8, 125.0, 117.7, 116.9, 107.0, 106.8, 80.7, 68.4,
67.9, 36.2, 34.3, 29.7, 28.9, 26.4, 25.6 ppm. C24H32O4 (384.5): calcd.
C 74.97, H 8.39; found C 74.63; H 8.31.
(؎)-7-(4-Cyano-3-hydroxybutyloxy)-2H-1-benzopyran-2-one (5d): A
solution of epoxide (Ϯ)-1 (0.116 g, 0.50 mmol) in CH3CN (0.5 mL)
was treated with KCN (0.049 g, 0.75 mmol) and LiClO4 (0.080 g,
0.75 mmol) and then the reaction mixture was stirred at 70 °C for
18 h.[28] After cooling, dilution with diethyl ether and evaporation
of the washed (saturated aqueous NaCl) organic solution afforded
a solid residue that was subjected to preparative TLC (EtOAc/hex-
ane, 7:3; Rf ϭ 0.29). Extraction of the most-intense band afforded
pure (Ϯ)-5d (0.064 g, 50% yield) as a solid. M.p. 95Ϫ97 °C. IR
(S)-2-O-(tert-Butoxycarbonylpentyl)-7-O-(3,4-dihydroxybutyl)-
naphthalene (13): AD Mix α (5.46 g) was added at room temp. to
a stirred tBuOH/H2O mixture (1:1, 40 mL). When a clear biphasic
system was obtained, the reaction mixture was cooled to 0 °C and
treated with a solution of olefin 12 (1.50 g, 3.9 mmol) in tBuOH (1
mL). After stirring for 48 h at the same temperature, the starting
material had completely reacted (TLC). Solid Na2S2O5 (5.85 g) was
added in three portions at 0 °C, and stirring was prolonged for
1 h at room temp. Extraction with CH2Cl2 and evaporation of the
collected organic phases afforded a crude solid (1.61 g) that was
subjected to flash chromatography. Elution with CH2Cl2/acetone
(7:3; Rf ϭ 0.46) afforded pure diol (S)-13 (1.35 g, 83% yield, 94%
ee) as a white solid, m.p. 51Ϫ53 °C. [α]2D0 ϭ Ϫ1.86 (c ϭ 1.17,
(nujol): ν˜ ϭ 3435 (br. m, OH), 2251 (w, CN), 1687 (s, CϭO) cmϪ1
.
1H NMR: δ ϭ 7.59 (d, J ϭ 9.5 Hz, 1 H), 7.31 (d, J ϭ 8.5 Hz, 1
H), 6.86Ϫ6.65 (m, 2 H), 6.18 (d, J ϭ 9.4 Hz, 1 H), 4.36Ϫ4.02 (m,
3 H), 2.64 (dd, J ϭ 5.7, 2.2 Hz, 2 H), 2.28Ϫ1.90 (m, 2 H) ppm.
13C NMR: δ ϭ 162.3, 162.0, 156.3, 144.2, 129.5, 118.2, 113.7,
113.5, 113.3, 102.0, 65.5, 65.4, 36.1, 27.0 ppm. C14H13NO4 (259.3):
calcd. C 64.86, H 5.05, N 5.40; found C 64.71, H 5.26, N 5.49.
CHCl3). IR (nujol): ν˜ ϭ 3321 (OH), 1726 cmϪ1 (CϭO). H NMR
1
Determination of Optical Purities: The ee of the diols (R)- and (S)-
2 and epoxides 1 were determined by analyses on a chiral HPLC
(CDCl3): δ ϭ 7.63 (d, J ϭ 8.8 Hz, 2 H), 7.12Ϫ6.93 (m, 4 H),
4.34Ϫ4.14 (m, 2 H), 4.12Ϫ3.94 (m, 3 H), 3.74 (dd, J ϭ 11.0, 3.2
column OD-H (Daicel) [25 cm ϫ 0.46 cm (i.d.)] using hexane/iP- Hz, 1 H), 3.56 (dd, J ϭ 10.5, 6.6 Hz, 1 H), 2.26 (t, J ϭ 6.6 Hz, 2
rOH (8:2): the R isomers were the first-eluting ones in each case.
Chlorohydrins (R)- and (S)-5a did not separate under these con-
ditions.[26] The absolute configurations of the diols (R)- and (S)-2
were assigned on the basis of the Sharpless mnemonic device ap-
plied to the olefin.[19] The configuration of the diols (R)- and (S)-
2 is maintained in the corresponding epoxides (R)- and (S)-1. The
H), 1.99 (q, J ϭ 5.9 Hz, 2 H), 1.91Ϫ1.75 (m, 2 H), 1.75Ϫ1.59 (m,
2 H), 1.59Ϫ1.47 (m, 2 H), 1.44 (s, 9 H) ppm. 13C NMR (CDCl3):
δ ϭ 173.9, 158.4, 157.8, 136.6, 129.8, 125.1, 117.3, 116.7, 107.1,
106.8, 80.8, 70.96, 68.4, 67.5, 65.9, 36.3, 33.3, 29.7, 28.9, 26.4, 25.6
ppm. C24H34O6 (418.5): calcd. C 68.87, H 8.19; found C 69.01,
H 8.39.
Eur. J. Org. Chem. 2004, 2557Ϫ2566
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2563

F. Badalassi, G. Klein, P. Crotti, J.-L. Reymond
FULL PAPER
(S)-2-O-(tert-Butoxycarbonylpentyl)-7-O-(4-chloro-3-methoxy-
carbonyloxybutyl)naphthalene (14): A stirred solution of diol (S)-13
(0.63 g, 1.51 mmol) in anhydrous CH2Cl2 (2.5 mL) was treated un-
After cooling, evaporation of the solvent afforded a crude product
that was dissolved in Et2O (9.0 mL) and treated, whilst stirring
at 0 °C, with NEt3 (0.67 mL, 4.8 mmol) and EtOCOCl (0.26 g,
der argon with MeC(OMe)3 (0.25 mL, 1.96 mmol) and PPTS 2.4 mmol). After stirring for 3 h at the same temperature, evapor-
(5 mg) and the resulting reaction mixture was stirred at room temp. ation of the filtered solution afforded a crude product that was
for 1.5 h. Evaporation of the solvent afforded a crude product, con- subjected to flash chromatography. Elution with hexane/EtOAc
taining the corresponding intermediate cyclic orthoester,[19] which
was dissolved in anhydrous CH2Cl2 (2.5 mL) and treated with Me3-
SiCl (0.25 mL, 1.97 mmol). After stirring for 18 h at room temp.,
(7:3) afforded pure N-protected aziridine (R)-17 (0.38 g, 42% yield)
as a liquid. 1H NMR (CDCl3): δ ϭ 7.64 (d, J ϭ 8.8 Hz, 2 H),
7.09Ϫ6.88 (m, 4 H), 4.33Ϫ4.15 (m, 2 H), 4.15Ϫ3.94 (m, 2 H), 4.14
evaporation of the solvent afforded a crude reaction product con- (q, J ϭ 6.8 Hz, 2 H), 2.80Ϫ2.62 (m, 1 H), 2.41 (d, J ϭ 6.3 Hz, 1
sisting of chloroacetate (S)-14 (0.71 g, 99% yield), practically pure
as a solid (m.p. 39Ϫ42 °C), which was used directly in the next step
H), 2.27 (t, J ϭ 7.3 Hz, 2 H), 2.11 (d, J ϭ 3.9 Hz, 1 H), 2.06Ϫ1.92
(m, 2 H), 1.92Ϫ1.77 (m, 2 H), 1.77Ϫ1.61 (m, 2 H), 1.61Ϫ1.47 (m,
2 H), 1.45 (s, 9 H), 1.20 (t, J ϭ 6.8 Hz, 3 H) ppm. 13C NMR
without any further purification. IR ν˜ ϭ 1739 (OϪCϭO), 1728
1
(CϭO) cmϪ1. H NMR (CDCl3): δ ϭ 7.64 (d, J ϭ 8.8 Hz, 2 H), (CDCl3): δ ϭ 173.9, 163.7, 157.8, 157.5, 136.1, 129.2, 124.5, 116.6,
7.11Ϫ6.86 (m, 4 H), 5.42Ϫ5.21 (m, 1 H), 4.23Ϫ3.92 (m, 4 H), 3.82 116.3, 106.4, 106.3, 80.2, 67.9, 65.5, 62.7, 49.9, 35.8, 32.3, 31.9,
(dd, J ϭ 11.7, 4.4 Hz, 1 H), 3.69 (dd, J ϭ 11.7, 4.9 Hz, 1 H),
29.2, 28.3, 25.8, 25.1, 14.4 ppm. C27H37NO5 (455.6): calcd. C 71.18,
2.30Ϫ2.12 (m, 4 H), 2.10 (s, 3 H), 1.95Ϫ1.74 (m, 2 H), 1.74Ϫ1.62 H 8.19, N 3.07; found C 71.44, H 7.91, N 3.18.
(m, 2 H), 1.62Ϫ1.47 (m, 2 H), 1.45 (s, 9 H) ppm. 13C NMR
(CDCl3): δ ϭ 173.8, 171.1, 158.4, 157.8, 136.6, 129.9, 129.8, 125.1,
117.2, 116.7, 106.8, 106.8, 80.8, 71.2, 68.4, 64.4, 46.5, 36.2, 32.1,
29.7, 28.9, 26.4, 25.6, 21.7 ppm.
(R)-7-O-(3,4-Aziridinobutyl)-2-O-(tert-butoxycarbonylpentyl)-
naphthalene (18): A stirred solution of N-protected aziridine (R)-17
(0.136 g, 0.29 mmol) in tBuOH (3.0 mL) was treated with two por-
tions of 0.1  tBuOK in tBuOH (1.5 mL ϫ 2), 16 h apart, and
then stirring was prolonged for 32 h at room temp. After dilution
with Et2O (40 mL), evaporation of the washed (water) diethyl ether
solution afforded a crude product, consisting of pure aziridine (R)-
18 (0.114 g, 98% yield), as a liquid that was used directly in the
(S)-2-O-(tert-Butoxycarbonylpentyl)-7-O-(3,4-epoxybutyl)-
naphthalene (15): A stirred solution of chloroacetate (S)-14 (1.36 g,
2.85 mmol) in anhydrous benzene (80 mL) was treated with two
portions of tBuOK (0.48 g ϫ 2, 4.28 mmol ϫ 2), 3 h apart, and
then stirring was prolonged for 20 h. Evaporation of the filtered next step without any further purification. IR (film): ν˜
ϭ
1
solution afforded crude epoxide (S)-15 that was filtered through a
short silica gel column. Elution with hexane/EtOAc (8:2) afforded
epoxide (S)-15 (0.93 g, 81% yield) as a liquid that was sufficiently
pure to be utilized in the next step. An analytical sample of crude
(S)-15 was purifed by flash chromatography. Elution with hexane/
EtOAc (8:2; Rf ϭ 0.29) afforded pure epoxide (S)-15 as a colourless
liquid. [α]2D0 ϭ Ϫ7.63 (c ϭ 1.52, CHCl3). IR (film): ν˜ ϭ 1728 cmϪ1
(CϭO). 1H NMR (CDCl3): δ ϭ 7.65 (d, J ϭ 8.8 Hz, 2 H),
7.10Ϫ6.87 (m, 4 H), 4.22 (t, J ϭ 6.3 Hz, 2 H), 4.06 (t, J ϭ 6.3 Hz,
2 H), 3.27Ϫ3.11 (m, 1 H), 2.85 (t, J ϭ 4.6 Hz, 1 H), 2.61 (dd, J ϭ
5.1, 2.7 Hz, 1 H), 2.27 (t, J ϭ 7.3 Hz, 2 H), 2.21Ϫ2.08 (m, 1 H),
2.08Ϫ1.97 (m, 1 H), 1.97Ϫ1.74 (m, 2 H), 1.74Ϫ1.63 (m, 2 H),
1.63Ϫ1.48 (m, 2 H), 1.45 (s, 9 H) ppm. 13C NMR (CDCl3): δ ϭ
173.7, 158.3, 157.9, 136.6, 129.8, 129.7, 125.0, 117.1, 116.7, 106.8,
106.7, 80.8, 68.4, 65.3, 50.5, 47.8, 36.2, 33.2, 29.7, 28.8, 26.3, 25.6
ppm. C22H28O5 (372.5): calcd. C 70.94, H 7.58; found C 70.76;
H 7.72.
1726 cmϪ1 (CϭO). H NMR (CDCl3): δ ϭ 7.63 (d, J ϭ 8.8 Hz, 2
H), 7.09Ϫ6.81 (m, 4 H), 4.20 (t, J ϭ 6.3 Hz, 2 H), 4.05 (t, J ϭ 6.3
Hz, 2 H), 2.37Ϫ2.12 (m, 3 H), 2.09Ϫ1.87 (m, 1 H), 1.87Ϫ1.34 (m,
7 H), 1.85 (t, J ϭ 6.6 Hz, 2 H), 1.44 (s, 9 H) ppm. 13C NMR
(CDCl3): δ ϭ 173.20, 157.8, 157.6, 136.1, 129.2, 124.5, 116.5, 116.3,
106.3, 80.2, 67.9, 66.3, 49.5, 35.7, 34.2, 29.2, 28.3, 27.7, 25.8, 25.1
ppm.
(R)-7-O-(3-Amino-4-azidobutyl)-2-O-(tert-butoxycarbonylpentyl)-
naphthalene (19):
A solution of aziridine (R)-18 (0.105 g,
0.264 mmol) in tBuOH/H2O (8:1, 2.3 mL) was treated with NaN3
(0.069 g, 1.056 mmol) and NH4Cl (0.056 g, 1.056 mmol) and then
the resulting reaction mixture was stirred at 60 °C for 20 h. After
cooling, dilution with CH2Cl2 and evaporation of the washed (satu-
rated aqueous NaCl) organic phase afforded amino azide (R)-19
(0.097 g, 83% yield, 91% ee), practically pure, as a liquid that was
used directly in the next step without any further purification. IR
1
(film): ν˜ ϭ 2100 (N3), 1726 cmϪ1 (CϭO). H NMR (CDCl3): δ ϭ
(S)-7-O-(4-Azidobutyl-3-hydroxy)-2-O-(tert-butoxycarbonylpentyl)- 7.64 (d, J ϭ 8.8 Hz, 2 H), 7.16Ϫ6.81 (m, 4 H), 4.19 (t, J ϭ 5.8 Hz,
naphthalene (16): A solution of epoxide (S)-15 (0.99 g, 2.48 mmol)
2 H), 4.05 (t, J ϭ 6.3 Hz, 2 H), 3.59Ϫ3.37 (m, 1 H), 3.37Ϫ3.16 (m,
1 H), 2.27 (t, J ϭ 7.3 Hz, 2 H), 2.12Ϫ1.91 (m, 1 H), 1.91Ϫ1.72 (m,
in anhydrous DMF (11 mL) was treated with NaN3 (0.647 g,
9.95 mmol) and then the reaction mixture was stirred at 80 °C for 4 H), 1.72Ϫ1.47 (m, 4 H), 1.45 (s, 9 H) ppm. 13C NMR (CDCl3):
22 h. After cooling, dilution with Et2O (50 mL) and evaporation
of the washed (water) organic solution afforded azido alcohol (S)-
16 (0.85 g, 78% yield), practically pure as a liquid, which was di-
rectly used in the next step without any further purification. IR
δ ϭ 173.3, 157.9, 157.4, 136.1, 129.3, 124.6, 116.7, 116.2, 106.3,
80.3, 67.9, 65.2, 58.7, 49.2, 35.7, 34.6, 29.2, 28.4, 25.9, 25.1 ppm.
(R)-7-O-(4-Azidobutyl-3-N-trifluoroacetylamino)-2-O-(tert-butoxy-
carbonylpentyl)naphthalene (20): A solution of amino azide (R)-19
(0.075 g, 0.169 mmol) in CH2Cl2 (3 mL), containing NEt3 (0.045
mL), was treated at 0 °C with trifluoroacetic anhydride (0.038 g,
0.0254 mL, 0.18 mmol) and then the reaction mixture was stirred
at the same temperature for 2 h. CH2Cl2 and ice-water were added.
Evaporation of the washed (water) organic phase afforded a crude
product that was subjected to flash chromatography. Elution with
hexane/EtOAc (7:3) afforded pure N-protected amino azide (R)-20
1
(film): ν˜ ϭ 2100 (N3), 1726 (CϭO) cmϪ1. H NMR (CDCl3): δ ϭ
7.57 (d, J ϭ 8.8 Hz, 2 H), 7.03Ϫ6.81 (m, 4 H), 4.31Ϫ4.09 (m, 2
H), 4.09Ϫ4.37 (m, 3 H), 3.40 (dd, J ϭ 12.7, 3.9 Hz, 1 H), 3.30 (dd,
J ϭ 12.2, 6.8 Hz, 1 H), 2.20 (t, J ϭ 7.3 Hz, 2 H), 1.95 (q, J ϭ 5.9
Hz, 2 H), 1.87Ϫ1.70 (m, 2 H), 1.70Ϫ1.55 (m, 2 H), 1.55Ϫ1.43 (m,
2 H), 1.38 (s, 9 H) ppm.
(R)-2-O-(tert-Butoxycarbonylpentyl)-7-O-(3,4-N-ethoxycarbonyl-
aziridinobutyl)naphthalene (17): A solution of azido alcohol (S)-16
(0.075 g, 83% yield) as a liquid. [α]2D0 ϭ ϩ11.2 (c ϭ 0.34, CHCl3).
(0.90 g, 1.92 mmol) in anhydrous MeCN (3.0 mL) was treated with IR (film): ν˜ ϭ 2104 (N3), 1724 (CϭO) cmϪ1. H NMR (CDCl3):
1
Ph3P (0.503 g, 1.92 mmol); the reaction mixture was stirred at room δ ϭ 7.66 (d, J ϭ 8.8 Hz, 2 H), 7.09Ϫ6.83 (m, 4 H), 4.47Ϫ4.24 (m,
temp. until N2 developed (0.5Ϫ1 h) and then for 15 h at 80 °C.
2564
1 H), 4.24Ϫ4.09 (m, 2 H), 4.09Ϫ3.92 (m, 2 H), 3.74Ϫ3.50 (m, 2
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2004, 2557Ϫ2566

Fluorescence Assay and Screening of Epoxide Opening by Nucleophiles
FULL PAPER
H), 2.36Ϫ2.07 (m, 4 H), 1.95Ϫ1.72 (m, 2 H), 1.72Ϫ1.47 (m, 4 H),
was stirred at the same temperature for 7 h. A new portion of
1.45 (s, 9 H) ppm. 13C NMR (CDCl3): δ ϭ 173.8, 158.5, 157.2, MCPBA (0.30 g, 1.22 mmol) was added and stirring was prolonged
136.5, 130.1, 129.8, 125.2, 119.3, 117.5, 116.2, 106.9, 106.8, 80.8,
for 12 h at 0 °C. Dilution with CH2Cl2 and evaporation of the
68.4, 65.3, 53.4, 49.3, 36.2, 31.1, 29.6, 28.8, 26.3, 25.5, 14.8 ppm. washed (5% aqueous Na2S2O3, saturated aqueous NaHCO3, and
C26H33F3N4O5 (538.6): calcd. C 57.98, H 6.18, N 10.40; found C
57.71, H 5.93, N 10.27.
saturated aqueous NaCl) organic solution afforded a crude product
that was subjected to flash chromatography. Elution with hexane/
EtOAc/CH2Cl2 (4:4:2) afforded pure epoxide 24 (0.25 g, 46% yield)
Synthesis of Acid (R)-21: Amide (R)-20 (0.054 g, 0.10 mmol) was
treated with CF3COOH/CH2Cl2 (1:1, 0.3 mL) and then the reac-
tion mixture was left at r.t. for 1.5 h. Evaporation of the solvent
afforded acid (R)-21 (0.048 g, 100% yield), practically pure as a
1
as a colorless solid. H NMR (CDCl3): δ ϭ 7.37 (d, J ϭ 9.0 Hz, 2
H), 6.84 (d, J ϭ 9.0 Hz, 2 H), 4.17Ϫ4.02 (m, 2 H), 3.21Ϫ3.09 (m,
1 H), 2.83 (t, J ϭ 4.6 Hz, 1 H), 2.58 (q, J ϭ 2.6 Hz, 1 H), 2.15 (s,
3 H), 2.13Ϫ2.02 (m, 1 H), 2.00Ϫ1.86 (m, 1 H) ppm.
1
solid. m.p. 108Ϫ110 °C. H NMR (CDCl3): δ ϭ 7.66 (d, J ϭ 8.8
Hz, 2 H), 7.10Ϫ6.82 (m, 4 H), 4.45Ϫ4.26 (m, 1 H), 4.26Ϫ4.10 (m,
2 H), 4.06 (t, J ϭ 6.3 Hz, 2 H), 3.72Ϫ3.50 (m, 2 H), 2.42 (t, J ϭ
7.1 Hz, 2 H), 2.21 (q, J ϭ 5.0 Hz, 2 H), 2.00Ϫ1.79 (m, 2 H),
1.79Ϫ1.63 (m, 2 H), 1.63Ϫ1.47 (m, 2 H) ppm. 13C NMR (CDCl3):
δ ϭ 179.3, 158.5, 157.1, 136.5, 130.2, 129.9, 125.3, 117.5, 116.2,
106.9, 106.9, 68.3, 65.4, 53.4, 49.4, 34.4, 31.1, 29.6, 26.4, 25.2 ppm.
(؎)-4-(p-Acetamidophenoxy)-1-chloro-2-butanol (25): A solution of
epoxide 24 (0.10 g, 0.45 mmol) in CH2Cl2 (3.0 mL) was treated
with aqueous 36% HCl (3.0 mL) and then the reaction mixture
was stirred at room temp. for 40 min. Dilution with CH2Cl2 and
evaporation of the washed (saturated aqueous NaHCO3) organic
solution afforded a crude product that was subjected to flash chro-
matography. Elution with hexane/EtOAc (8:2) afforded pure
Conjugation of Acid (R)-21 with KLH and BSA Carrier Proteins: A
1
chlorohydrin 25 (0.080 g, 70% yield) as a colorless solid. H NMR
solution of 1.5
 aqueous N-[3-(dimethylamino)propyl]-NЈ-
(CDCl3): δ ϭ 7.37 (d, J ϭ 9.0 Hz, 2 H), 6.84 (d, J ϭ 9.0 Hz, 2 H),
4.12Ϫ3.96 (m, 3 H), 3.68 (dd, J ϭ 11.2, 4.2 Hz, 1 H), 3.57 (dd, J ϭ
11.0, 6.6 Hz, 1 H), 2.04 (s, 3 H), 2.05Ϫ1.92 (m, 1 H), 1.94Ϫ1.80
(m, 1 H) ppm.
ethylcarbodiimide·HCl (EDC) (40 µL) and 1.6  aqueous N-
hydroxysulfosuccinimide sodic salt (Sulfo-NHS) (39 µL) were ad-
ded in succession to a solution of acid (R)-21 (0.010 g, 0.020 mmol)
in DMF (100 µL). After incubation for 24 h at room temp. (the
reaction was monitored by reverse-phase HPLC), a portion (45 µL)
of the solution containing the activated hapten was added to a
solution of Keyhole-limpet hemocyanin (KHL, 5 mg/mL) in 50 m
phosphate buffer (pH 7.4; 500 µL); a second portion (90 µL) of the
same solution was added to a solution of bovine serum albumine
(BSA, 5 mg/mL) in 50 m phosphate buffer (pH 7.4; 1 mL). The
two solutions, so obtained, were maintained at 4 °C for 24 h (satu-
rated aqueous NaHCO3 was added, if necessary, to obtain pH 7.4)
and then kept at Ϫ18 °C. The conjugates were used in the next step
without any further purification.
Assays with Fluorogenic Epoxide 1: A. Screening of hybridoma cell
culture supernatants. Antibody cell culture supernatant (50 µL) was
added to 20 m aq. borate buffer (50 µL) containing substrate (R)-
1 (200 µM), NADϩ (1 m), HLDH (200 µg/mL), BSA (4 mg/mL),
and 0.2  of either NaCl (291 samples from fusion with hapten 9)
or NaN3 (13 samples from fusions with hapten 8) in the wells of
96-well polypropylene microtiter plates (Costar). A portion of this
solution (50 µL) was immediately transferred to another well con-
taining a 2 m hapten stock solution (0.5 µL). B. Assay for epoxide
hydrolysis on purified anti-9 antibodies: Solutions containing puri-
fied anti-9 antibodies (0.2 mg/mL), epoxide rac-1 (100 m), NaIO4
(1 m), and BSA (2 mg/mL) were monitored at 28 °C in the indi-
vidual wells of 96-well polypropylene microtiter plates by fluor-
escence using a Cytofluor II Fluorescence Plate Reader (Perseptive
Biosystems, filters λex ϭ 360 Ϯ 20 nm, λem ϭ 460 Ϯ 20 nm).
Deprotection of Trifluoroacetamide in Protein Conjugates: The solu-
tions containing the BSA and KLH conjugates of (R)-21, obtained
as described above, were treated independently with 5  aqueous
NaOH (11 and 18 µL, respectively) at pH 14, and then the resulting
reaction mixtures were stirred at room temp. for 16 h (hydrolysis
of the N-trifluoroacetyl group was monitored by TLC, using ninyd-
rine assay). Aqueous 5  HCl was added to both solutions (11 and
18µL, respectively) in obtain pH 7.4 once again. The solution of
the KLH-conjugate was partitioned in four fractions that were sub-
sequently used for immunization. The solution of the BSA-conju-
gate was diluted with PBS and then partitioned into fractions that
were used for the ELISA assay.
HPLC-Assay with 24 and 25: Antibody solution (Ͼ 50 µg/mL) was
obtained by purification of cell culture supernatants as described
previously.[20] Chlorohydrin formation from epoxide 24. A 10 m
stock solution (2 µL) of epoxide 24 in DMF was added to cell
culture supernatant in PBS (160 m NaCl, 10 m phosphate, pH
7.4; 100 µL) in the individual wells of a 96-well polypropylene mic-
rotiter plate (Costar). A portion of this solution (50 µL) was trans-
ferred to another well containing a 2 m stock solution (1 µL) of
the corresponding hapten in PBS. The reactions were incubated
overnight at 25 °C and assayed by RP-HPLC [Vydac C18 218TP54,
0.5 ϫ 22 cm; 1.5 mL minϪ1; eluent: 80% H2O/20% CH3CN;
tR(24) ϭ 4.97 min, tR(25) ϭ 5.65 min], showing, on average, 2.5%
conversion into epoxide. The conversion was the same with or with-
out the hapten. Epoxide formation from chlorohydrin 25: A 10 m
stock solution (2 µL) of chlorohydrin 25 in DMF was added to cell
culture supernatant in PBS (160 m NaCl, 10 m phosphate, pH
7.4; 94.5 µL) in the individual wells of a 96-well polypropylene
microtiter plate (Costar). 1  HCl (4.5 µL) was added to adjust the
pH to ഠ 6. A portion of this solution (50 µL) was transferred to
another well containing a 2 m stock solution of the corresponding
hapten in PBS (1 µL). The reactions were incubated overnight at
4-(Butenyloxy)acetamidobenzene (26): p-Acetamidophenol (3.0 g,
19.9 mmol) was added to a suspension of NaH (0.80 g of a 60%
dispersion in mineral oil, 19.9 mmol) in anhydrous DMF (18 mL)
and then the reaction mixture was stirred at room temp. for 1 h. 4-
Bromo-1-butene (2.2 mL, 21.89 mmol) was added and stirring was
prolonged for 48 h. Dilution with water, extraction with EtOAc,
and evaporation of the washed (aqueous 1  NaOH) organic phase
afforded a crude reaction product consisting of practically pure ole-
fin 26 (0.66 g, 16% yield) that was used in the next step without
any further purification. IR (film): ν˜ ϭ 1670 cmϪ1 1H NMR
.
(CDCl3): δ ϭ 7.37 (d, J ϭ 9.0 Hz, 2 H), 6.82 (d, J ϭ 9.0 Hz, 2 H),
5.96Ϫ5.82 (m, 1 H), 5.20Ϫ5.08 (m, 2 H), 3.98 (t, J ϭ 6.8 Hz, 2 H),
2.56Ϫ2.49 (m, 2 H), 2.12 (s, 3 H) ppm.
(؎)-4-(p-Acetamidophenoxy)-1,2-epoxybutane (24): A solution of
olefin 26 (0.50 g, 2.44 mmol) in CH2Cl2 was treated at 0 °C with 25 °C and assayed, as above, by RP-HPLC. There was, on average,
70% MCPBA (0.67 g, 2.70 mmol) and then the reaction mixture 2.5% conversion into the chlorohydrin and 3.5% conversion into
Eur. J. Org. Chem. 2004, 2557Ϫ2566
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2565

F. Badalassi, G. Klein, P. Crotti, J.-L. Reymond
FULL PAPER
the corresponding 1,2-diol (tR ϭ 2.64 min). The conversion was
[12] [12a]
´
´
E. M. Gonzalez-Garcia, J. Grognux, D. Wahler, J.-L. Re-
[12b]
ymond, Helv. Chim. Acta 2003, 86, 2458.
Grognux, D. Wahler, J.-L. Reymond, Helv. Chim. Acta 2003,
E. Nyfeler, J.
the same with or without the hapten.
86, 2919.
[13] [13a] N. Jourdain, R. Perez-Carlon, J.-L. Reymond, Tetrahedron
Acknowledgments
[13b]
´ `
R. Perez Carlon, N. Jourdain, J.-L.
Lett. 1998, 39, 9415.
[13c]
This work was supported by the University of Bern, the European
COST program (Action D12), the Swiss OFES, the Wander Stif-
tung, the MIUR (Roma) and the University of Pisa. P. C. gratefully
acknowledges Merck Research Laboratories for the generous finan-
cial support derived from the 2002 ADP Chemistry Award.
Reymond, Chem. Eur. J. 2000, 6, 4154.
E. M. Gonzalez-
Garcia, V. Helaine, G. Klein, M. Schuermann, G. Sprenger, W.-
D. Fessner, J.-L. Reymond, Chem. Eur. J. 2003, 9, 893.
F. Badalassi, P. Crotti, J.-L. Reymond, Helv. Chim. Acta 2002,
82, 3090.
[14]
[15] [15a]
D. Wahler, F. Badalassi, P. Crotti, J.-L. Reymond, Angew.
Chem. 2001, 113, 4589; Angew. Chem. Int. Ed. 2001, 40, 4457.
[1]
[15b]
B. Marrs, S. Delagrave, D. Murphy, Curr. Opin. Microbiol.
D. Wahler, F. Badalassi, P. Crotti, J.-L. Reymond, Chem.
[15c]
1999, 2, 241.
Eur. J. 2002, 8, 3211.
J.-L. Reymond, D. Wahler, F.
[2] [2a]
W. P. C. Stemmer, Proc. Natl. Acad. Sci. USA 1994, 91,
Badalassi, H. K. Nguyen, Patent WO0192563, 2001-12-07.
[16] [16a]
10747. [2b] W. P. C. Stemmer, Nature 1994, 370, 389. [2c] L. You,
G. Klein, J.-L. Reymond, Bioorg. Med. Chem. Lett. 1998,
[2d]
[16b]
F. H. Arnold, Protein. Eng. 1996, 9, 77.
M. T. Reetz, K.-E.
8, 1113.
82, 400.
G. Klein, J.-L. Reymond, Helv. Chim. Acta 1999,
[2e]
Jaeger, Chem. Eur. J. 2000, 6, 407.
S. V. Taylor, P. Kast, D.
Hilvert, Angew. Chem. 2001, 113, 3408; Angew. Chem. Int. Ed.
[17]
[18]
C.-H. Wong, G. M. Whitesides, Enzymes in Synthetic Organic
Chemistry, Pergamon, 1994.
2001, 40, 3310.
[3] [3a]
M. T. Reetz, Angew. Chem. 2001, 113, 292; Angew. Chem.
A. Tramontano, K. D. Janda, R. A. Lerner, Science 1986, 234,
[3b]
[18b]
Int. Ed. 2001, 40, 284.
D. Wahler, J.-L. Reymond, Curr.
1566.
S. J. Pollack, J. W. Jacobs, P. G. Schultz, Science
Opin. Chem. Biol. 2001, 5, 152. [3c] F. Beisson, A. Tiss, C. Rivi-
1986, 234, 1570. [18c] P. G. Schultz, J. Yin, R. A. Lerner, Angew.
Chem. 2002, 114, 4607; Angew. Chem. Int. Ed. 2002, 41, 4427.
J. Hasserodt, Synlett 1999, 12, 2007.
Top. Curr. Chem. 1999, 200, 59.
Rep. 1996, 13, 479.
Prod. Rep. 2000, 17, 535.
[3d]
`
ere, R. Verger, Eur. J. Lipid Sci. Technol. 2000, 133.
D.
[18d]
[18e]
Wahler, J.-L. Reymond, Curr. Opin. Biotechnol. 2001, 12, 535.
J.-L. Reymond,
N. R. Thomas, Nat. Prod.
J. D. Stevenson, N. R. Thomas, Nat.
[4] [4a]
[18f]
M. Chini, P. Crotti, C. Gardelli, F. Macchia, Tetrahedron
[4b]
[18g]
1992, 48, 3805.
M. Chini, P. Crotti, L. Favero, F. Macchia,
M. Pineschi, Tetrahedron Lett. 1994, 35, 433 and references
[19]
[20]
[21]
cited therein.
H. C. Kolb, M. S. VanNieuwenhze, K. B. Sharpless, Chem. Rev.
1994, 94, 2483.
N. Bensel, M. T. Reymond, J.-L. Reymond, Chem. Eur. J. 2001,
7, 4604.
[5] [5a]
J. F. Larrow, S. E. Schaus, E. N. Jacobsen, J. Am. Chem.
[5b]
Soc. 1996, 118, 7420.
B. M. Cole, K. D. Shimizu, C. A.
Krueger, J. P. A. Harrity, M. L. Snapper, A. H. Hoveyda, An-
gew. Chem. 1996, 108, 1776; Angew. Chem. Int. Ed. Engl. 1996,
G. Köhler, C. Milstein, Nature 1975, 256, 495.
[5c]
[22] [22a]
35, 1668, 1995.
M. Tokunaga, J. F. Larrow, F. Kakiuchi, E.
F. Zocher, M. M. Enzelberger, U. T. Bornscheuer,
[5d]
N. Jacobsen, Science 1997, 277, 936.
K. D. Shimizu, B. M.
B. Hauer, R. D. Schmid, Anal. Chim. Acta 1999, 391, 345.
Cole, C. A. Krueger, K. W. Kuntz, M. L. Snapper, A. H. Hov-
eyda, Angew. Chem. 1997, 109, 1783; Angew. Chem. Int. Ed.
Engl. 1997, 36, 1704.
[22b] D. Whaler, J.-L. Reymond, Angew. Chem. 2002, 114, 1277;
[22c]
Angew. Chem. Int. Ed. 2002, 41, 1229.
K. Dodere, S. Lutz-
Wahl, B. Hauer, R. D. Schmid, Anal. Biochem. 2003, 321, 131.
C. Morisseau, A. Archelas, C. Guitton, D. Faucher, R. Fur-
stoss, J. C. Baratti, Eur. J. Biochem. 1999, 263, 386.
N. Bensel, PhD thesis, University of Bern, 2001.
[6] [6a]
[23]
[24]
A. Archelas, R. Furstoss, Top. Curr. Chem. 1999, 200, 159.
C. A. G. Weijers, Tetrahedron: Asymmetry 1997, 8, 639 and
[6b]
references cited therein.
[7]
[8]
[25] [25a]
S. C. Sinha, E. Keinan, J.-L. Reymond, J. Am. Chem. Soc.
1993, 115, 4893.
K. D. Janda, C. G. Shevlin, R. A. Lerner, Science 1993,
[25b]
259, 490.
J. Na, K. N. Houk, C. G. Shevlin, K. D. Janda,
[25c]
M. Arand, B. M. Hallberg, J. Zou, T. Bergfors, F. Oesch, M.
J. van der Werf, J. A. de Bont, T. A. Jones, S. L. Mowbray,
EMBO J. 2003, 22, 2583.
R. A. Lerner, J. Am. Chem. Soc. 1993, 115, 8453.
K. D.
Janda, C. G. Shevlin, R. A. Lerner, J. Am. Chem. Soc. 1995,
117, 2659.
[9] [9a]
[26]
A. Kamal, A. B. Rao, M. V. Rao, Tetrahedron Lett. 1992,
The values of ee of (R)- and (S)-5a are supposed to be the same
as those of the corresponding epoxides.
M. Caron, K. B. Sharpless, J. Org. Chem. 1985, 50, 1557.
M. Chini, P. Crotti, L. Favero, F. Macchia, Tetrahedron Lett.
1991, 32, 4775.
[9b]
33, 4077.
35, 81.
M. Mischitz, K. Faber, Tetrahedron Lett. 1994,
[27]
[28]
[10]
[11]
J. H. L. Spelberg, J. E. T. van Hylckama Vlieg, L. Tang, D. B.
Janssen, R. M. Kellogg, Org. Lett. 2001, 3, 41.
F. Badalassi, D. Wahler, G. Klein, P. Crotti, J.-L. Reymond,
Angew. Chem. 2000, 112, 4233; Angew. Chem. Int. Ed. 2000,
39, 4067.
Received January 14, 2004
Early View Article
Published Online May 6, 2004
2566
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 2557Ϫ2566