Microbiological transformations. Part 45: A green chemistry preparative scale synthesis of enantiopure building blocks of Eliprodil: Elaboration of a high substrate concentration epoxide hydrolase-catalyzed hydrolytic kinetic resolution process

Article abstract of DOI:10.1016/S0040-4020(00)01032-2

The enantioselective hydrolysis of racemic para-chlorostyrene oxide 2 following a typical 'green chemistry' procedure based on the use of two different Epoxide Hydrolases is described. This allows the preparation of both enantiomers of 2 in very high enantiomeric purity. Furthermore, using a 'one-pot' sequential bi-enzymatic strategy enabling to overcome the 50% yield limitation intrinsic to any resolution process, rac-2 could be transformed into nearly enantiopure (R)-3 with an overall yield as high as 93%. The methodology developed was based on the use of a biphasic reactor at high substrate concentration, which is highly desirable for any potential industrial process. The obtained chirons are valuable building blocks for the synthesis of various biologically active targets, like (R)-Eliprodil.

Full text of DOI:10.1016/S0040-4020(00)01032-2

TETRAHEDRON
Pergamon
Tetrahedron 57 (2001) 695±701
Microbiological transformations. Part 45: A green chemistry
preparative scale synthesis of enantiopure building blocks of
Eliprodil: elaboration of a high substrate concentration epoxide
hydrolase-catalyzed hydrolytic kinetic resolution process
p
K. M. Manoj, A. Archelas, J. Baratti and R. Furstoss
Groupe Biocatalyse et Chimie Fine, ESA 6111 associ e e au CNRS, Universit e de la M e diterran e e, Facult e de Sciences de Luminy, Case 901,
1
63 avenue de Luminy, 13288 Marseille Cedex 9, France
Received 7 September 2000; revised 25 October 2000; accepted 6 November 2000
AbstractÐThe enantioselective hydrolysis of racemic para-chlorostyrene oxide 2 following a typical `green chemistry' procedure based on
the use of two different Epoxide Hydrolases is described. This allows the preparation of both enantiomers of 2 in very high enantiomeric
purity. Furthermore, using a `one-pot' sequential bi-enzymatic strategy enabling to overcome the 50% yield limitation intrinsic to any
resolution process, rac-2 could be transformed into nearly enantiopure (R)-3 with an overall yield as high as 93%. The methodology
developed was based on the use of a biphasic reactor at high substrate concentration, which is highly desirable for any potential industrial
process. The obtained chirons are valuable building blocks for the synthesis of various biologically active targets, like (R)-Eliprodil. q 2001
Elsevier Science Ltd. All rights reserved.
1
. Introduction
associated with its (2)-(R) enantiomer rather than to its
1)-(S) antipode, and that the pharmacodynamic properties
of the two enantiomers were different.
(
4
The search for novel methods enabling the synthesis of
enantiopure pharmaceuticals is a major thrust area in
contemporary ®ne organic synthesis, due to the increasing
Another important long-term issue for contemporary indus-
trial ®ne chemistry is the employment of environmentally
gentle processesÐthe so-called `green chemistry'
approachÐwhere the usage of toxic materials like heavy
metal reactants/catalysts or organic solvents should be
lowered or, ideally, eliminated.
1
demand for safer and more ef®cient drugs. This is aimed to
avoid the possible (sometimes deleterious) side effects of
the less active (or inactive) enantiomer of a given biologi-
cally active molecule. At the start of this work, Eliprodil 1
2
(
Scheme 1) was subject to phase three trials as a highly
promising neuroprotective agent (acting as a glutamate
antagonist), known to possess high ef®ciency against
ischemic stroke. Interestingly, it has been recently demon-
During the last ten years, we (and others) have demonstrated
that Epoxide Hydrolases (EH) (EC 3.3.2.3) are very inter-
esting new tools for the organic chemist since they allow
3
5
strated that the in vivo activity of Eliprodil is essentially
Scheme 1.
Keywords: enantioselective hydrolysis; biotransformation; hydrolytic kinetic resolution; epoxide hydrolase; Aspergillus niger; Solanum tuberosum; Eliprodil;
chiral synthons.
*
Corresponding author. Tel.: 133-4-91-82-91-58; fax: 133-4-91-82-91-45; e-mail: furstoss@luminy.univ-mrs.fr
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)01032-2

6
96
K. M. Manoj et al. / Tetrahedron 57 (2001) 695±701
Table 1. Chemical hydrolysis (experimental conditions: 4 mM concentra-
tion of rac-2; HPLC analysis of the formed diol, hydrolysis expressed in
nmol/h per mL) of rac-2 as a function of pH and temperature
mental conditions used, the rate of this hydrolysis was deter-
mined as a function of pH and temperature (Table 1) in
order to allow correction of the experimental values
obtained upon enzymatic hydrolysis and optimization of
reaction conditions for preparative scale reactions. Owing
to these results, the experiments were conducted at 08C in
order to avoid signi®cant spontaneous hydrolysis.
pH
08C
68C
288C
6
7
7
8
.7
.5
4
7
5
4
3.5
36
26
20
16
3.5
2.5
2.2
2
.2. Analytical scale studies
ef®cient achievement of the Hydrolytic Kinetic Resolution
HKR) of various epoxides. It must in particular be empha-
When rac-2 (4 mM concentration) was submitted to react
with AnEH at 08C, a conversion ratio of 51% was reached
after 3 h, leading to the remaining (slow reacting) epoxide
showing an ee of 98%. An E value of ,100 was calculated
(
sized that: (a) these hydrolases are cofactor independent; (b)
they are ubiquitous in nature; (c) they can in particular be
found in microorganisms, which makes them very easy to
prepare in large scale (eventually after cloning and over-
expression in an appropriate host); (d) they may be endowed
with substrate- and/or enantio-complementarity, thus allow-
ing to set up, in some cases, ef®cient enantiocomplementary
processes. Studies using such enzymes have been ¯ourish-
(
using c and ee of remaining epoxide), whereas a value of
40 was observed at 278C. Such a temperature dependent E
value enhancement has already been established
,
1
4
previously. The remaining epoxide 2 was shown to be of
S) absolute con®guration by comparison with an authentic
(
1
2
5
sample. Higher reaction time led to enantiopure 2, a fact
which illustrates the attractive feature that such resolution
processes allow to `tune' the ee of the slow reacting enan-
tiomer simply by tailoring the reaction time, i.e. the conver-
sion ratio, a strategy which is not available starting from a
prochiral substrate. After derivatization into its acetonide
derivative, the formed diol 3 could also be analyzed by
chiral gc and was shown to be of 90% ee. Its absolute
con®guration was determined after recyclization into the
corresponding epoxide using a method avoiding loss of
ing over the last years and we have ourselves developed
strategies allowing for the preparation of enantiopure bio-
6
active products. The knowledge about these `new' enzymes
very recently culminated with the publication of the three
®rst X-ray structures of cytosolic EHs from, respectively,
the bacteria Agrobacterium radiobacter, murine liver and
the fungus Aspergillus niger.
7
8
9
In this context, we have explored a way to prepare (R)-1 in
enantiopure form using an EH catalyzed procedure. Since
para-chlorostyrene oxide 2 as well as the corresponding diol
1
5
stereochemical integrity. This indicated that it was of
R) absolute con®guration, which leads to the conclusion
(
3
, are two obvious possible key building blocks for achiev-
that the (R)-2 enantiomer was, at least preferentially,
attacked by the enzyme at the b position, i.e. at the less
substituted carbon atom, thus leading to retention of con®g-
uration. This also con®rms that diol 3 can be a valuable
chiral building block for 1.
ing the synthesis of (R)-1, we embarked on a study aimed to
prepare these chirons in enantiopure form. We describe here
the preparation of each of the two enantiomers of 2 at a
multi-gram scale by using two complementary EHs in
high substrate concentration reactors. Moreover, we also
report a `one-pot' enantioconvergent procedure based on
the sequential use of these two enzymes, which allowed to
overcome the 50% maximum yield intrinsic to a classical
resolution process and afforded (R)-3 at very high yield and
enantiomeric purity.
When rac-2 (4 mM concentration) was submitted to StEH, a
similar behavior was observed. Thus, within 2 h, a conver-
sion ratio of 51% was reached, affording a remaining (slow
reacting) epoxide showing an ee of 97% and the formed diol
of 96% ee. Again an E value of ,100 could be calculated
(
whereas a value of ,40 was observed at 278C). Interest-
ingly, however, both the obtained epoxide and diol were
2
. Results
2.1. Preliminary work
A rapid screening using our `in house' available epoxide
hydrolases led to the conclusion that EHs from Aspergillus
1
0
11
niger (AnEH) and Solanum tuberosum L. (StEH) were
the two best candidates for achieving the resolution of 2.
Interestingly, they were seen to be enantio-complementary,
the AnEH preferentially hydrolyzing the (R) enantiomer
1
2
whereas StEH preferred the (S) antipode. Solubility of 2
in a pH 6.7 phosphate buffer was established by HPLC
analysis to be ,5 mM (0.75 g/L) at 278C. Therefore the
optimum reaction parameters and overall reaction pro®le
studies were ®rst achieved at 4 mM substrate concentration
using each of the two enzymes and will be discussed in a
1
3
separate report.
Figure 1. Analytical enzymatic resolution of rac-2 at 08C using AnEH and
StEH. V c (AnEH); B ee-epox. (AnEH); X ee-diol (AnEH); S c (StEH); A
ee-epox. (StEH); W ee-diol (StEH).
Since rac-2 is also spontaneously hydrolyzed in the experi-

K. M. Manoj et al. / Tetrahedron 57 (2001) 695±701
697
Table 2. Regioselectivity coef®cients for the acid and enzyme catalyzed
hydrolyses
data which was taken into account for correction of the coef®-
cients indicated in Table 2.
a(S) %
b(S) %
a(R) %
b(R) %
Chemical hydrolysis
AnEH
StEH
67
3
97
33
97
3
67
1
8
33
99
92
2.4. Preparative scale experiments
In order to set up an ef®cient preparative scale process for
resolution of rac-2, the following priority of requirements
were decided (a) high ee of the products (b) maximum
possible substrate concentration (c) as high as possible
yield (d) optimum amount of enzyme and (e) reasonable
determined to be of (R) absolute con®guration. This indi-
cated that (a) contrary to the result observed with the A.
niger enzyme, the StEH showed a marked preference for
hydrolyzing the (S)-epoxide (therefore being enantiocom-
plementary to the previous one) and (b) that, again opposite
to the previous result, the water attack preferentially
occurred at the more substituted (benzylic) carbon atom,
leading to inversion of con®guration. Fig. 1 shows the ee
versus time dependent conversion of rac-2 using the two
different enzymes.
1
8
reaction time. Also, the activity and stability pro®le of
each one of the two enzymes had to be taken into account.
1
3
Biphasic high concentration reactor using AnEH. Owing to
the previously described priorities, the reaction was
conducted in a pH 7 phosphate buffer at 08C (the half life
of the biocatalyst was determined to be of about 7 days at
this temperature). The possibility to increase the substrate
concentration (at a constant enzyme/substrate ratio) was
explored by monitoring the ee of the remaining epoxide
2.3. Determination of regioselectivity versus
enantioselectivity
1
8
after 1 and 3 h reaction (Fig. 2).
This `stereochemical ¯exibility', i.e. the possible combination
of various enantio- and regio-selectivities within one single
EH catalyzed hydrolysis is one of the most intriguing and
potentially useful properties of such enzymes, and we have
Interestingly, it appears that (a) the reaction could be
performed using a substrate concentration much above its
maximum solubility (about 5 mM, i.e. 0.75 g/L) (b) appar-
ently no mass transfer limit does occur (c) above 400 mM
substrate concentration, the (high) diol concentration
seemed to slow down the reaction rate. Using the above
de®ned experimental conditions, a preparative scale process
was conducted at 2 M concentration (i.e. 306 g/L) using a
reactor containing 4 g (26 mmol) of rac-2. This was used as
a slurry together with a solution of 540 U in a 9 mL phos-
phate buffer solution. The reaction reached a 50% conver-
sion in about 8 h, after which the reaction mixture was
extracted with pentane to isolate the remaining (S)-2 epox-
ide and with ethyl acetate to extract the resulting (R)-3 diol.
Thus, after puri®cation, 1.9 g (47% yield) of (S)-2 was
obtained which showed a 99% ee, together with 2.15 g
(48% yield) of (R)-3 (92% ee) (overall yield 95%). The
space-time yield of this two-phase reactor was calculated
1
6
previously studied this type of behavior in detail. In order to
accurately establish the speci®c regioselectivity implied for
hydrolysis of rac-2 by AnEH and StEH (i.e. to determine the
1
7
a(R), b(R), a(S) and b(S) coef®cients ), each enantiomer of 2
was submitted independently, in enantiopure form, to hydro-
lysis by these two enzymes. In each case, the corresponding
coef®cients were determined from the ee of the formed diol,
leading to the value shown in Table 2. For AnEH, both enan-
tiomers were essentially attacked at the b carbon atom.
However, the StEH attacked predominantly (R)-2 at the b
position, whereas (S)-2 was attacked at the benzylic a carbon
atom, making this entire process highly enantioconvergent.
The regioselectivity of spontaneous chemical hydrolysis was
also determined starting from each enantiomer. This indicated
a predominant hydrolytic attack at the benzylic carbon atom, a
1
9
Figure 2. Scale-up of substrate concentration for AnEH. Experimental conditions: Reaction volume 0.5 mL; pH 7; t8 08C. The enzyme/substrate ratio was kept
constant through out the scale-up at 0.02 U/mmol of rac-2. A %ee of (S)-2 at tˆ1 h B %ee of (S)-2 at tˆ3 h.

6
98
K. M. Manoj et al. / Tetrahedron 57 (2001) 695±701
Figure 3. Scale-up of substrate concentration for StEH. Experimental conditions: reaction volume 0.5 mL; pH 7; t8 08C. The enzyme/substrate ratio was kept
constant through out the scale-up at 0.0063 U/mmol of rac-2. A %ee of (R)-2 at tˆ1 h B %ee of (R)-2 at tˆ6 h.
to be about 20 g/L/h and the turnover number about
12000 mol of substrate/mole enzyme/h.
an ee higher than 99%, was obtained together with 2.1 g
(47% yield) of (R)-3 (97% ee).
2
0
Biphasic high concentration reactor using StEH. Similar
experimental conditions were chosen for performing the
resolution of rac-2 using StEH. The only difference lies in
the fact that the reaction was conducted at pH 6.7, as it was
demonstrated to be optimum for reactivity and stability of
this enzyme. The possibility to increase the substrate
2.5. Enantioconvergent process using a sequential bi-
enzymatic strategy
As mentioned previously, the major drawback of any reso-
lution process is its intrinsic 50% yield limitation. Fortu-
nately, several ways allowing to overcome this industrially
utmost important problem have been described, including
dynamic resolution, combined chemoenzymatic processes
1
8
concentration was also explored as shown on Fig. 3.
Here again it appears that (a) the reaction can be performed
at substrate concentrations much above its maximum solu-
bility in the medium (b) no mass transfer limitation
occurred. However it clearly appeared that, to the contrary
of what was observed with the AnEH, substrate concentra-
tion above 200 mM were deleterious to the enzyme activity.
In accordance with this inhibition effect, the preparative
scale experiment was conducted at 0.2 M (30.6 g/L)
substrate concentration. Thus, 4 g of rac-2 were hydrolyzed
as a slurry (under mechanical stirring) using a solution of
6
,21
or bi-enzymatic protolcols. In the present case, it clearly
appeared that it could be possible to exploit the enantio- and
regio-complementarity of our two enzymes in order to reach
the ideal `100% yield, 100% ee' goal. For this purpose, we
did set up an optimized procedure using the two enzymes
sequentially rather than simultaneously. This choice was
based on three interesting and complementary facts, i.e.:
(a) as emphasized above, the StEH appeared to be more
sensitive to inhibition by the (R)-diol formed. Therefore,
since the concentration of formed 3 would increase about
twofold if an equivalent amount (in Units) of AnEH would
be present simultaneously, this would lead to a negative
effect on the biohydrolysis rate; (b) since StEH was shown
150 U of StEH in 125 mL phosphate buffer. The reaction
reached a 50% conversion in about 18 h, after which the
reaction mixture was worked up as described above.
After puri®cation, 1.9 g (47% yield) of (R)-2, which showed
Scheme 2.

K. M. Manoj et al. / Tetrahedron 57 (2001) 695±701
699
to be enantioconvergent, formation of the (R) diol would be
nearly exclusive even after consumption of (S)-2; (c) ®nally,
since (owing to the E value of about 100) direct use of AnEH
at the start of the reaction would lead to hydrolysis of some
sample were injected into a Lichrosorb RP 18 column
(Merck) and eluted using a mixture of water/acetonitrile
(3:1) at 1.5 mL/min (UV detector at 220 nm). Under these
conditions, 3 eluted at 5 min (and 2 eluted at 30 min).
(
S)-2 enantiomer to the (S) diol, this would lower the ee of 3
at total conversion. To the contrary, addition of AnEH after
consumption of (S)-2 by the StEH would obviously avoid
this problem.
4
.2. Preparation of the enzymatic extracts
The A. niger and the potato enzymatic extracts were
prepared as previously described.
6
c
Therefore, in a typical experiment, biohydrolysis was
started using 150 U StEH in 125 mL buffer solution. After
about 50% conversion (18 h at 08) 500 U AnEH were added
into the reaction vessel and the reaction was led to comple-
tion (about 2 days in total). The reaction broth was extracted
with ethyl acetate in order to isolate the resulting diol,
affording 4.18 g (R)-3 (93% yield) which showed an ee of
4
.3. Synthesis of rac-p-chlorostyrene oxide 2
This epoxide was obtained by epoxidation of the corre-
sponding ole®n using m-chloroperoxybenzoic acid under
biphasic conditions (CH Cl /phosphate buffer) as
2
2
12
1
previously described (77% yield). H NMR d: 2.75 (dd,
1
96%. This overall strategy is shown in Scheme 2.
H, H2cis, Jgemˆ5.5 Hz, J
ˆ2.6 Hz); 3.15 (dd, 1H,
1±2cis
H2trans, Jgemˆ5.4 Hz, J
ˆ4.1 Hz); 3.84 (dd, 1H, H ,
1±2trans
1
J1±2cisˆ2.6 Hz, J
ˆ4.0 Hz); 7.18±7.26 (m, 2H );
arom
1±2trans
3
. Conclusion
13
.30±7.34 (m, 2H ). C NMR d: 51.3 (C-2); 51.8
arom
7
(
C-1); 126.8; 128.7; 133.9 136.2; 136.2 (C±Ar).
In the course of this work, we have achieved the enantio-
selective hydrolysis of racemic para-chlorostyrene oxide 2
using two different Epoxide Hydrolases. This enabled the
preparation of both enantiomers of 2 in nearly enantiopure
form. Furthermore, using a `one-pot' sequential bi-enzy-
matic strategy, rac-2 could be transformed into (R)-3 with
an overall yield as high as 93%. Thus, this strategy enabled
to overcome the 50% yield limitation intrinsic to any reso-
lution process and to approach the ideal `100% yield, 100%
ee' goal. The methodology developed within this work was
based on the use of a biphasic reactor at high substrate
concentration, which is highly desirable for any potential
industrial process. Moreover, this was performed using a
typical `green chemistry' procedure, a strategy which will
be of utmost importance in the coming century due to
increasing environmental pressure on chemical industry.
The obtained chirons are valuable building blocks for the
synthesis of various biologically active targets, like (R)-
Eliprodil. Work is in progress in our laboratory in order to
further evaluate the scope and limitations of such methodol-
ogies, with the intention of enhancing the utility of micro-
bial Epoxide Hydrolases as industrial catalysts.
4
.4. Analytical reactors
To 3960 mL of pH6.7 buffer containing the appropriate
amount of enzyme (AnEHˆ0.275 mg/mL at 0.26 U/mg
solid or StEHˆ1.837 mg/mL at 0.06 U/mg solid) main-
tained at 08C, 40 mL of DMF solution containing 400 mM
rac-2 and 100 mM meta-bromoacetophenone (MBAP) were
added to give the ®nal concentration at 4 mM rac-2, 1 mM
MBAP, DMF,1% (0.0179 U/mM substrate for AnEH and
0
.2756 U/mM substrate for StEH). Aliquots of 400 mL were
withdrawn at different time intervals; 200 mL methanol
were added to stop the reaction. The epoxide was extracted
with 500 mL isooctane and analyzed by GC for conversion
ratio and ee. A fraction of the diol was also used for conver-
sion ratio determination by HPLC. The remainder was
extracted with ethyl acetate and derivatized for determina-
tion of diol ee.
4.5. Preparative enzymatic resolution of (R/S)-2 using
AnEH
To 9 mL of sodium phosphate buffer (pH 7, 100 mM, at
08C), 2.3 g of AnEH preparation (,540 U, at 0.234 U/mg
solid) were added and NaOH solution was used to adjust the
pH to 7. The resultant solution was ,9.8 mL. To this, 4 g of
rac-2 were added and the slurry stirred at 400 rpm using a
magnetic paddle. A few microliters of the reaction mixture
were withdrawn at appropriate time intervals to check for
the enantiomeric excess of the remaining epoxide and the
formed diol. The reaction was stopped around 8 h and the
reaction mixture extracted with pentane (3£40 mL). The
4
. Experimental
4
.1. General methods
The Aspergillus niger strain used in this work is registered at
the Museum d'Histoire Naturelle under n LCP 521 (Lab.
de Cryptogamie, 12 rue Buffon, 75005 Paris, France). NMR
spectra ( H and C) were recorded in CDCl at 250 and
o.
1
13
3
1
00 MHz, respectively. Chemical shifts are reported in d
from TMS as internal standard. Determination of the enan-
tiomeric excesses of epoxide-2 and diol-3 were performed
using chiral column Chirasil-Dex CB (0.25 mm£25 m,
Chrompack). Other conditions were as follows: 1/100
split, injector and FID at 2508C, column at 1258C for 2
and 1508C for acetonide derivative of 3, Helium at 1 mL/
min. Under these conditions, (R)-2, (S)-2, (R)-3 and (S)-3
eluted at 11, 11.6, 9.8 and 10.7 min, respectively. Reverse
phase HPLC was used to quantify 3. 20 mL of aqueous
collected organic factions were dried over MgSO and
4
concentrated in vacuum. After puri®cation by bulb-to-bulb
distillation 1.9 g (47% yield) of (S)-2 was isolated
2
0
[a] ˆ126 (c 1.29; CHCl ). The remaining aqueous
D
3
phase was extracted with ethyl acetate (3£40 mL), dried
over MgSO and concentrated in vacuum. After ¯ash chro-
4
matography puri®cation (silica gel 60H from Merck and
solvent mixtures consisting of pentane and diethyl ether in
the range of 100% pentane to 100% diethyl ether) 2.15 g

7
00
K. M. Manoj et al. / Tetrahedron 57 (2001) 695±701
2
0
(
CHCl ), mp 82.58C.
48% yield) of (R)-diol 3 were isolated [a] Dˆ258 (c 1.26;
References
3
1
. (a) Schurig, V.; Betschinger, F. Chem. Rev. 1992, 92, 873. (b)
Kolb, H. C.; VanNieuenhze, M. S.; Sharpless, K. B. Chem.
Rev. 1994, 94, 2483±2547.
4.6. Preparative enzymatic resolution of (R/S)-2 using
StEH
2
3
. Sheldon, R. A. Chirotechnology: Industrial Synthesis of Opti-
cally Active Compounds; Marcel Dekker: New York, 1993.
. Scatton, B.; Giroux, C.; Thenot, J. P.; Frost, J.; George, P.;
Carter, C.; Benavides, J. Drugs of the Future 1994, 19, 905±
4.2 g of enzyme powder (t150 U, at 0.037 U/mg solid)
were added to 125 mL buffer (pH 6.7, 100 mM, at 08C)
and pH adjusted. The resultant solution was 127.5 mL. To
this, 4 g of rac-2 were added and this was stirred at 400 rpm
using a magnetic paddle at the bottom and an external
impeller from the top. A few microliters of the reaction
mixture were withdrawn at appropriate time intervals to
check for the enantiomeric excess of the remaining epoxide
and the formed diol. The reaction was stopped around 18 h
and the reaction mixture was extracted with pentane
9
09.
4
. Di Fabio, R.; Pietra, C.; Thomas, R. J.; Ziviani, L. Biorg. Med.
Chem. Lett. 1995, 5, 551±554.
5
. (a) Archer, I. V. J. Tetrahedron 1997, 53, 15617±15662. (b)
Archelas, A.; Furstoss, R. Annu. Rev. Microbiol. 1997, 51,
4
91±525. (c) Archelas, A.; Furstoss, R. Tibtech 1998, 16,
08±116. (d) Archelas, A.; Furstoss, R. In Biocatalysis.
1
From Discovery to Application, Fessner, W.-D., Ed., 1998;
00, pp 159±191. (e) Orru, R. V. A.; Archelas, A.;
Furstoss, R.; Faber, K. Adv. Biochem. Eng. 1999, 63, 145±
67.
(
4£140 mL) for the epoxide and the remaining aqueous
2
phase with ethyl acetate (4£100 mL) for the diol. The
same work-up described above using AnEH allowed iso-
1
2
0
lation of 1.9 g (47% yield) of (R)-2 [a] ˆ224 (c 1.2;
D
6
. (a) Pedragosa-Moreau, S.; Archelas, A.; Furstoss, R. J. Org.
Chem. 1993, 58, 5533±5536. (b) Pedragosa-Moreau, S.;
Morisseau, C.; Baratti, J.; Zylber, J.; Archelas, A.;
Furstoss, R. Tetrahedron 1997, 53, 9707±9714. (c) Cleij, M.;
Archelas, A.; Furstoss, R. J. Org. Chem. 1999, 64, 5029±
2
0
CHCl ) and 2.1 g (47% yield) of (R)-diol 3 [a] ˆ260
3
D
(
c 1.29; CHCl ), mp 84.68C.
3
4.7. Preparative enantioconvergent process using AnEH
and StEH
5
035.
7
8
9
. Nardini, M.; Ridder, I. S.; Rozeboom, H. J.; Kalk, K. H.;
Rink, R.; Janssen, D. B.; Dijkstra, B. W. J. Biol. Chem.
1
.9 g of StEH enzyme powder (,150 U, at 0.079 U/mg
solid) was added to 125 mL buffer (pH 6.7, 100 mM, at
8C) and pH adjusted to 6.7. The resultant was 127.5 mL.
1
999, 274, 14579±14596.
. Argiriadi, M. A.; Morisseau, C.; Hammock, B. D.;
0
Christianson, D. W. Proc. Natl. Acad. Sci. USA 1999, 96,
1
To this, 4 g of rac-2 were added and stirred using a magnetic
paddle at the bottom and an external impeller from the top.
A few microliters of the reaction mix was withdrawn at
appropriate time intervals to check for the enantiomeric
excess of the remaining epoxide and the formed diol.
After 25 h, when almost all of (S)-2 was consumed, 2.18 g
AnEH powder (t500 U, at 0.23 U/mg solid) were added and
the pH was noted to be at 7.1. The reaction was stopped
around 45 h when the concentration of the remaining epox-
ide was lower than 1 mM. The reaction mixture was
extracted with ethyl acetate (4£140 mL) for the (R)-diol
0637±10642.
. Zou, J.-Y.; Hallberg, M.; Bergfors, T.; Oesch, F.; Arand, M.;
Mowbray, S. L.; Jones, T. A. Structure 2000, 8, 111±122.
1
0. (a) We have recently performed and published the puri®cation
and sequence of this enzyme (see Ref. 10b). It also has been
cloned and overexpressed (see Ref. 10c), and its X-ray struc-
ture has been published very recently (see Ref. 9.) (b)
Morisseau, C.; Archelas, A.; Guitton, C.; Faucher, D.;
Furstoss, R.; Baratti, J. C. Eur. J. Biochem. 1999, 263, 386±
3
95. (c) Arand, M.; Hemmer, H.; Durk, H.; Baratti, J.;
Archelas, A.; Furstoss, R.; Oesch, F. Biochem. J. 1999, 344,
73±280.
3
: 4.18 g (93% yield) was isolated after ¯ash chromato-
2
graphy puri®cation (silica gel 60H from Merck and solvent
mixtures consisting of pentane and diethyl ether in the range
1
1. Kindly provided to us by Hammock et al. See: Stapleton, A.;
Beetham, J. K.; Pinot, F.; Garbarino, J. E.; Rockhold, D. R.;
Friedman, M.; Hammock, B. D.; Belknap, W. R. Plant J.
2
0
of 100% pentane to 100% diethyl ether) [a] Dˆ261.4 (c
1.29; CHCl ), mp 83.78C.
3
1
996, 6, 251±258.
1
2. The absolute con®guration of para-chlorostyrene oxide 2 has
been previously established: Pedragosa-Moreau, S.; Moris-
seau, C.; Zylber, J.; Archelas, A.; Baratti, J.; Furstoss, R.
J. Org. Chem. 1996, 61, 7402±7407.
Acknowledgements
1
1
3. To be published elsewhere.
This work has been achieved in the context of the EC
BIOC4-CT95-0005 contract. Financial support of this
work, together with a fellowship to one of us (K. M. M.)
is gratefully acknowledged. We are also extremely grateful
to Professor B. Hammock (University of California, Davis,
USA) for kindly providing us with the baculovirus over-
expression system of the Solanum tuberosum L. EH. We
would also like to very heartily thank Professor M. Arand
and Ms K. H aÈ nel (Institut of Toxicology, University of
Mainz, Germany) for their demonstration and teaching of
the baculovirus expression technique.
4. Sakai, T.; Kawabata, I.; Kishimoto, T.; Ema, T.; Utaka, M.
J. Org. Chem. 1997, 62, 4906±4907.
15. Golding, B. T.; Hall, D. R.; Sarkrikar, S. J. Chem. Soc., Perkin
Trans. 1 1973, 1214±1220.
16. Moussou, P.; Archelas, A.; Baratti, J.; Furstoss, R.
Tetrahedron: Asymmetry 1998, 9, 1539±1547.
17. The percentage of attack on the benzylic carbon atom C1 of
(S)-epoxide (resulting in (R)-diol formation) is called a(S),
while the percentage of attack on the C2 carbon atom is
b(S) (resulting in (S)-diol formation). Similarly, a(R) and

K. M. Manoj et al. / Tetrahedron 57 (2001) 695±701
701
b(R) coef®cients were used in the case of the (R)-epoxide
hydolysis.
results) now allows us to prepare a much more puri®ed enzy-
matic extract which shows speci®c activity of about 22 U/mg.
Therefore, the weight of AnEH necessary to perform these
experiments could be cut down by a factor of 100.
1
8. Since the reaction system was highly heterogeneous, the
conversion ratio could not be determined even with a fair
sense of approximation. Ee was taken as an index to measure
the reaction extent at two different time intervals. The analysis
achieved after 1 h gave an indication of the effect of substrate
concentration and the analysis at 3 h (for AnEH) or at 6 h (for
StEH) indicated the effect of the formed diol on the overall
reaction.
20. We estimate the speci®c activity of pure AnEH to be of about
90 U/mg pure enzyme. Interestingly, comparison of the turn-
over number for a similar substrate (meta-chlorostyrene
oxide) using Jacobsen catalyst gives a value of 1.5 mol of
substrate/mole of catalyst/h: Brandes, B. D.; Jacobsen, E. N.
Tetrahedron: Asymmetry 1997, 8, 3927±3933.
1
9. The enzyme used throughout this work was obtained as a very
crude extract (speci®c activity towards para-chlorostyrene
oxide 2 was about 0.2 U/mg). Our recent work (unpublished
2
1. Orru, R. V. A.; Kroutil, W.; Faber, K. Tetrahedron Lett. 1997,
8, 1753±1754.
3