Enantioselective bio-hydrolysis of various racemic and meso aromatic epoxides using the recombinant epoxide hydrolase Kau2

Article abstract of DOI:10.1002/adsc.201401164

Abstract Epoxide hydrolase Kau2 overexpressed in Escherichia coli RE3 has been tested with ten different racemic and meso α,β-disubstituted aromatic epoxides. Some of the tested substrates were bi-functional, and most of them are very useful building blocks in synthetic chemistry applications. As a general trend Kau2 proved to be an extremely enantioselective biocatalyst, the diol products and remaining epoxides of the bioconversions being obtained - with two exceptions - in nearly enantiomerically pure form. Furthermore, the reaction times were usually very short (around 1 h, except when stilbene oxides were used), and the use of organic co-solvents was well tolerated, enabling very high substrate concentrations (up to 75 g/L) to be reached. Even extremely sterically demanding epoxides such as cis- and trans-stilbene oxides were transformed on a reasonable time scale. All reactions were successfully conducted on a 1 g preparative scale, generating diol- and epoxide-based chiral synthons with very high enantiomeric excesses and isolated yields close to the theoretical maximum. Thus we have here demonstrated the usefulness and versatility of lyophilized Escherichia coli cells expressing Kau2 epoxide hydrolase as a highly enantioselective biocatalyst for accessing very valuable optically pure aromatic epoxides and diols through kinetic resolution of racemates or desymmetrization of meso epoxides.

Full text of DOI:10.1002/adsc.201401164

FULL PAPERS
DOI: 10.1002/adsc.201401164
Enantioselective Bio-Hydrolysis of Various Racemic and meso
Aromatic Epoxides Using the Recombinant Epoxide Hydrolase
Kau2
Wei Zhao,a Michael Kotik,b Gilles Iacazio,a,* and Alain Archelasa
a
Aix-Marseille Universitꢀ, Centrale Marseille, CNRS, ISM2 UMR7313, Marseille 13397, France
Fax : (+33)-4912-8440;phone: (+33)-4912-82856; e-mail: gilles.iacazio@univ-amu.fr
b
ˇ
Institute of Microbiology, Academy of Sciences of the Czech Republic, Vꢁdenskꢂ 1083, CZ-142 20, Prague, Czech
Republic
Received: December 15, 2014; Revised: March 12, 2015; Published online: && &&, 0000
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201401164.
Abstract: Epoxide hydrolase Kau2 overexpressed in stilbene oxides were transformed on a reasonable
Escherichia coli RE3 has been tested with ten differ- time scale. All reactions were successfully conducted
ent racemic and meso a,b-disubstituted aromatic ep- on a 1 g preparative scale, generating diol- and epox-
oxides. Some of the tested substrates were bi-func- ide-based chiral synthons with very high enantiomer-
tional, and most of them are very useful building ic excesses and isolated yields close to the theoretical
blocks in synthetic chemistry applications. As a gener- maximum. Thus we have here demonstrated the use-
al trend Kau2 proved to be an extremely enantiose- fulness and versatility of lyophilized Escherichia coli
lective biocatalyst, the diol products and remaining cells expressing Kau2 epoxide hydrolase as a highly
epoxides of the bioconversions being obtained – with enantioselective biocatalyst for accessing very valua-
two exceptions – in nearly enantiomerically pure ble optically pure aromatic epoxides and diols
form. Furthermore, the reaction times were usually through kinetic resolution of racemates or desym-
very short (around 1 h, except when stilbene oxides metrization of meso epoxides.
were used), and the use of organic co-solvents was
well tolerated, enabling very high substrate concen-
trations (up to 75 g/L) to be reached. Even extremely Keywords: biotransformations; chiral resolution;
sterically demanding epoxides such as cis- and trans- enantioselectivity; epoxide hydrolase; kau2
Introduction
more, due to the development of genetic engineering
techniques, it is now routine to search for biodiversity
Optically enriched epoxides[1] and diols[2] are key syn- in specific biocatalysts, to overproduce them and to
thons for the generation of chiral compounds of high improve their performance in terms of specific activi-
synthetic value. Within this framework the epoxide ty, enantioselectivity or enantioconvergency by molec-
hydrolase (EH) enzyme family has become a very ular evolution techniques.[5]
popular source of biocatalysts during the last 20
Recently, a new EH termed Kau2, whose gene was
years.[3] Indeed, EHs offer the advantage of being retrieved from environmental DNA of a microbial
largely distributed within the various kingdoms of consortium, was successfully overexpressed in E. coli
life,[4] functioning without added cofactors, and ac- and shown to be of interest for kinetic resolution and
cepting a wide range of epoxide substrates. They are enantioconvergent deracemization of trans- and cis-
used in classical kinetic resolutions of racemates but methylstyrene oxide, respectively.[6] To further explore
can be used also in the desymmetrization of meso-ep- the diversity of asymmetric epoxide and diol synthons
oxides or sometimes in enantioconvergent biotrans- generated through Kau2-catalyzed hydrolysis reac-
formations. The classical limitation of kinetic resolu- tions, various aromatic racemic epoxides were tested,
tion (i.e., a maximum theoretical yield of 50% for some of them being of high potential synthetic inter-
a pure enantiomer) can be overcome in the two latter est.[7] Indeed, some of the tested substrates were bi-
cases, leading to a maximum theoretical yield of functional, expanding thus their synthetic utility.
100% for a pure diol product enantiomer. Further-
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
ÞÞ
These are not the final page numbers!

FULL PAPERS
Wei Zhao et al.
In this paper we describe the kinetic resolution of Results and Discussion
racemic methyl trans- and cis-3-phenylglycidate (1)
and (20), ethyl trans-3-phenylglycidate (3), methyl Kinetic Resolution of Methyl and Ethyl trans-3-
trans-3-(4-methoxyphenyl)glycidate (5), trans- and cis- Phenylglycidate rac-1 and rac-3
3-phenyloxirane-2-carbonitrile (7) and (9), trans-2-
(bromomethyl)- and trans-2-(chloromethyl)-3-phenyl- Initially, rac-methyl trans-3-phenylglycidate 1 was
oxirane (11) and (13), trans-stilbene oxide (16) as well tested as a substrate of the Kau2-EH (Scheme 1) in
as the desymmetrization of cis-stilbene oxide (18) the 1 to 100 g/L concentration range in the presence
using lyophilized cells of E. coli overexpressing the of 5% DMF and 10% isooctane as a water-immiscible
Kau2-EH as a biocatalyst. Compounds 1 and 3 are of co-solvent. A ratio of 1 to 2 (mass of substrate to
synthetic interest since they can be used for the syn- mass of lyophilized Kau2-expressing E. coli cells, m/
thesis of the anticancer taxol side chain[8] or the noo- m) was initially found convenient. For substrate con-
tropic drug (À)-Clausenamide.[9] Methyl 3-(4-methox- centrations lower than or equal to 10 g/L a nearly per-
yphenyl)glycidate (5) in its (2R,3S)-stereochemical fect kinetic resolution was observed, since both
form is a very useful chiral synthon for the access to formed (2R,3R)-2 and remaining (2S,3R)-1 were ob-
the calcium channel blocker Diltiazem, which acts as tained optically pure (ee >99%). Higher substrate
a potent vasodilator,[10] while each enantiomer of 5 as concentrations of 25, 50 and 100 g/L resulted in ees of
well as the corresponding diols can be used for the 98% (c=0.49), 88% (c=0.47) and 69% (c=0.41) for
synthesis of all stereoisomers of isocytoxazone, the the remaining epoxide, respectively, the ee of the
structural isomer of (À)-Cytoxazone, a cytokine mod- formed diol being over 99% in all cases. Even when
ulator of microbial origin.[11] The chloro epoxide 13 the reaction time was extended or new enzyme was
has been used as a synthon to access the naturally oc- added, the ee of the remaining epoxide could not be
curring styryl lactones Leiocarpin C and Gonodiol.[7a] increased for these high substrate concentrations, sug-
Furthermore, the diols 12 and 14 arising from ring- gesting that the reaction was blocked. In order to in-
opening of bromo epoxide 11 and chloro epoxide 13 crease the substrate concentration in these kinetic res-
can easily be transformed under basic conditions olutions, other water-immiscible solvents (10% v/v)
(with retention of configuration) to get access to the than isooctane were tested by analyzing the ee of the
terminal epoxide 15 which is useful for the synthesis remaining epoxide after one hour of reaction time at
of several natural and biologically active com- a substrate concentration of 50 g/L. While ethyl ace-
pounds.[7b,c] In almost all cases the biocatalytic trans- tate was clearly the worst solvent (ee 79%), diisoprop-
formation was of extremely high enantioselectivity, yl ether (ee 95%) proved to be better than isooctane
giving rise to an entire range of enantiomerically pure (ee 87%), the ee of the residual epoxide remaining
epoxides and diols. Furthermore, the tested reactions almost constant for extended reaction times
were successfully performed on a 1 g preparative (Figure 1).
scale, and the reaction times rarely exceeded 60–
Combining these results with the use of a more
80 min. With these water-insoluble substrates an or- active Kau2-EH preparation of 1667 U/g (produced in
ganic water-immiscible solvent was used without seri- a bioreactor) compared to 768 U/g (produced in
ously affecting the enzyme activity. Another major flasks) enabled the substrate concentration to be in-
advantage of such a biphasic biotransformation ap- creased to 50 g/L, ensuring ꢀ99% for both ees of
proach was that higher stabilities of the water-sensi- formed diol product and residual epoxide. Using
tive epoxidic substrates were achieved due to their re- these optimized conditions, the reaction was then con-
duced chemical hydrolysis in the organic phase.
ducted on a preparative scale (1 g) in 80 min, afford-
ing after extraction and purification the remaining ep-
oxide (2S,3R)-1 in 49% isolated yield (ee >99%) and
the formed diol (2R,3R)-2 in 47% isolated yield (ee
>99%). These results compared particularly well with
Scheme 1. Preparative kinetic resolution of methyl and ethyl trans-3-phenylglycidate (rac-1) and (rac-3).
2
asc.wiley-vch.de
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

FULL PAPERS
Enantioselective Bio-Hydrolysis of Various Racemic and meso Aromatic Epoxides
pared favorably[12,14] with other EHs used in the kinet-
ic resolution of rac-3. Essentially the same results as
described above have been reported for a gelozyme
preparation of the Phaseolus radiatus (Mung bean)
EH acting on rac-3: the remaining epoxide (2S,3R)-3
and the formed diol (2R,3R)-4 were obtained in iso-
lated yields of 45 and 40% with ee values of >99 and
94%, respectively.[13] Furthermore, the remaining ep-
oxide (2R,3S)-3 has been obtained either in 95% ee
and 26% yield using whole cells of Pseudomonas sp.
BZS21[14] or in more than 99% ee and 37% yield
using whole cells of Galactomyces geotrichum
ZJUTZQ200.[12] It should be noted that rac-methyl
cis-3-phenylglycidate (20) was not a substrate of the
Kau2-EH, in contrast to its trans counterpart 1.
Figure 1. Comparison of the effects of the various used or-
ganic co-solvents on the final achievable ees of the remain-
ing substrate 1.
Kinetic Resolution of Methyl trans-3-(4-
Methoxyphenyl)glycidate (rac-5)
We then tested the Kau2-EH with rac-5 as a substrate
a recent study[12] that used whole cells of Galactomy- (Scheme 2). Due to the presence of the para-methoxy
ces geotrichum ZJUTZQ200, which enabled the substituent, 5 proved to be prone to chemical hydroly-
access to (2R,3S)-1 in 98.6% ee at 62.5% conversion sis during the time course of the biocatalytic reaction.
from rac-1, resulting in an E value of 19 compared This was detrimental to both final yield in remaining
with an E value >200 in our case.
epoxide as well as the ee of the formed diol. In order
To obtain some additional information about the to minimize the chemical hydrolysis, the effects of the
size of acceptable substrates of the Kau2-EH, com- temperature (178C, 238C or 278C), the percentage
pound 3 was also tested using the optimized reaction (0%, 20% or 40%) of the non-miscible organic co-sol-
conditions defined for the transformation of the vent methyl tert-butyl ether (MTBE), and the pH of
methyl ester 1 (Scheme 1). Essentially the same re- the buffer phase (7.0, 7.5, 8.0, 8.5) were thoroughly
sults were obtained for up to 10 g/L of substrate con- tested. The best results were obtained at 178C with
centration, both ees of remaining epoxide 3 and 40% MTBE at pH 8.5 (see the Supporting Informa-
formed diol 4 reaching >99% at 50% conversion. At tion). Using these optimized reaction conditions, we
25 g/L of substrate concentration, the ee of the epox- tested substrate concentrations ranging from 1 to
ide reached >99% while the diol ee was around 96% 40 g/L using a two-fold concentration of lyophilized
during the entire experiment. At 50 g/L of substrate, biocatalyst (2 to 80 g/L). Generally, the reactions
the ee of 3 did not exceed 78% while the ee of the were quick with >99% ee for the epoxide reached in
diol remained constant at 96% during the whole reac- 1 h, except for low (1 g/L) and high substrate concen-
tion. Thus at very high substrate concentrations the trations (40 g/L). The low mass transfer rate of the ep-
Kau2-EH was found to be less enantioselective to- oxide to the aqueous phase at 1 g/L likely limited the
wards the ethyl ester 3 compared to its methyl ester rate of the enzymatic reaction, whereas at 40 g/L of
counterpart. Nevertheless, after 100 min of a prepara- substrate the presence of organic solvent in conjunc-
tive reaction which was conducted on a 1 g scale at tion with large cell quantities (80 g/L) generated
25 g/L, the remaining (2S,3R)-3 and the formed a jelly-like phase that probably caused some inhibi-
(2R,3R)-4 were isolated in 47 and 44% yields, and tion of the bioconversion reaction. By slightly dimin-
>99 and 94% ee, respectively. Interestingly, the ishing the substrate to catalyst ratio we were able to
Kau2-EH proved to be equally efficient[13] or com- perform a preparative-scale reaction with 1 g (40 g/L)
Scheme 2. Preparative kinetic resolution of rac-methyl trans-3-(4-methoxyphenyl)glycidate (5).
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3
ÞÞ
These are not the final page numbers!

FULL PAPERS
Wei Zhao et al.
Scheme 3. Kinetic resolution of trans- and cis-3-phenyloxirane-2-carbonitrile (rac-7) and (rac-9) and chemical transforma-
tions aimed at determining absolute configurations (see the Supporting Information).
of rac-5 and 70 g/L of lyophilized biomass, affording epoxides (9) were obtained and easily chromato-
in one hour remaining (2S,3R)-5 in 40% isolated yield graphically separated, giving thus the possibility to
and >99% ee, and formed (2R,3R)-6 in 46% isolated test both substrates with the Kau2 EH. An E. coli
yield and 88% ee. Unfortunately and in line with the biomass with a higher EH specific activity (2757 com-
results obtained for glycidates 1 and 3, residual pared to 1667 U/g, see the Experimental Section) ena-
(2S,3R)-5 is not suitable for the synthesis of Diltiazem bled the substrate/enzyme ratio to be reduced to 1/
but could be used in the synthesis of one enantiomer 0.75 (m/m), and 20% MTBE were used to ensure a bi-
of each cis- and trans-isocytoxazone.[11] When com- phasic system. The trans-cyano derivative 7 was tested
pared to previously reported kinetic resolutions of in the concentration range of 1–75 g/L and proved to
rac-5,[12,13] the Kau2 EH proved to be a better catalyst. be an excellent substrate (Scheme 3). Under these re-
Indeed, using whole cells of Galactomyces geotrichum action conditions both residual (2R,3R)-7 and formed
ZJUTZQ200 an E value of only 3 was determined,[12] (2S,3R)-8 were obtained in >99% ee in 80 min for all
while using a gelozyme preparation of the Phaseolus tested concentrations, except for the 1 and 75 g/L con-
radiatus (Mung bean) EH residual (2S,3R)-5 was ob- centrations. The slow mass transfer of the substrate to
tained in 45% yield and more than 99% ee, the the aqueous phase at 1 g/L was probably the reason
formed diol-6 being racemic in this case.
for the lower ee values; on the other hand, some un-
specific inhibition at high substrate and biomass con-
centrations appeared to limit the ee of (2R,3R)-7 to
75%. A preparative reaction run with 1 g of 7 at 50 g/
L afforded after 80 min both residual (2R,3R)-7 and
formed (2S,3R)-8 with ees of >99% in 45 and 46%
Kinetic Resolution of trans- and cis-3-Phenyloxirane-
2-carbonitrile rac-7 and rac-9
Based on the excellent results obtained with the isolated yields, respectively (Scheme 3).
phenyl glycidate derivatives, we decided to test
The Kau2 EH was previously reported to enantio-
whether a cyano functionality could replace the previ- convergently transform cis-b-methylstyrene oxide into
ously used ester functionality. Due to the selected the corresponding (1R,2R)-diol in 97% isolated yield
mode of synthesis, both trans-cyano (7) and cis-cyano- and 98.3% ee.[6] As methyl cis-3-phenylglycidate (20)
4
asc.wiley-vch.de
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

FULL PAPERS
Enantioselective Bio-Hydrolysis of Various Racemic and meso Aromatic Epoxides
was not a substrate of the Kau2 EH, we were pleased (rac-13). It is worthwhile to note that when tested at
to note that closely related cis-3-phenyloxirane-2-car- a concentration of 25 g/L, both rac-11 and rac-13 be-
bonitrile (rac-9) underwent a kinetic resolution using haved similarly in Kau2-mediated hydrolysis reac-
this enzyme. The optimal substrate/enzyme ratio was tions, rac-11 being transformed slightly quicker than
1/1 (m/m), and 20% MTBE were used to ensure the rac-13. Using a substrate/enzyme ratio of 2.5/1 (m/m)
presence of a biphasic system. Once again the kinetic at 1–25 g/L of rac-11, a 2/1 ratio at 50 g/L of rac-11
resolution was almost perfect for substrate concentra- and a 1/1 ratio at 75 g/L of rac-11 in the presence of
tions ranging from 1 to 25 g/L, both residual epoxide 20% MTBE, a nearly perfect kinetic resolution was
and formed diol being recovered essentially optically observed in each case, affording (2S,3R)-11 and
pure after 1 h. At 50 g/L of substrate concentration (2R,3R)-12 with ees exceeding 99% after 1 h of reac-
the reaction stopped after the ee of the remaining ep- tion time.
oxide reached 93%, the obtained diol being nearly
When run on a preparative scale (1 g of rac-11,
optically pure. A preparative reaction was then con- 75 g/L) in the presence of 0.8 g of biocatalyst (60 g/L),
ducted on a 1 g scale at 25 g/L of rac-9, affording the reaction afforded optically pure (2S,3R)-11 and
after 80 min and silica gel purification residual (2R,3R)-12 in 44% and 48% isolated yields, respec-
(2R,3S)-9 and formed (2R,3R)-10 in nearly enantio- tively (Scheme 4). A preparative-scale reaction with
pure form (ee >99%), and 44 and 45% isolated chloro-epoxide-13 [1 g of 13, 50 g/L; 0.5 g of biocata-
yields, respectively. The results described here for the lyst (2011 U/g)], afforded optically pure (2S,3R)-13
Kau2-catalyzed biohydrolysis of both rac-7 and rac-9 (43% isolated yield) and (2R,3R)-14 (44.5% isolated
compared particularly well with those described in yield) (Scheme 4). A (2S,3R) absolute configuration
the literature using whole cells of Mortierella isabelli- was assigned to remaining bromo-epoxide (11) by
na DSM 1414 in an attempt to kinetically resolve analogy with the stereochemistry data obtained for
these epoxides.[15] Indeed, for both rac-7 and of rac-9 chloro-epoxide (13), assuming the same stereochemis-
an E value of only 1 was determined in this case.
try of epoxide opening during the Kau2-mediated hy-
drolysis reaction for both compounds. Such an equiva-
lent behavior was supported by the fact that formed
bromo-diol (12) and chloro-diol (14) possessed the
same absolute configuration (Scheme 4 and the Sup-
porting Information).
Kinetic Resolution of trans-2-(Bromomethyl)- and
trans-2-(Chloromethyl)-3-phenyloxirane (rac-11) and
(rac-13)
To further extend the substrate range of the Kau2
EH, we turned our attention to the very interesting
compounds trans-2-(bromomethyl)-3-phenyloxirane
(rac-11) and trans-2-(chloromethyl)-3-phenyloxirane
Scheme 4. Kinetic resolution of trans-2-(bromomethyl)- and trans-2-(chloromethyl)-3-phenyloxirane (rac-11) and (rac-13)
and chemical transformations aimed at determining absolute configurations of diols (see the Supporting Information).
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
5
ÞÞ
These are not the final page numbers!

FULL PAPERS
Wei Zhao et al.
Kinetic Resolution of trans-Stilbene Oxide (16) and
Desymmetrization of cis-Stilbene Oxide (18)
almost enantiomerically pure (ee >99%) with more
than 90% analytical yield in 24 h. At 50 g/L the yield
of formed 19 dropped considerably to 20% after 24 h
In an attempt to further explore the substrate scope of reaction time. Thus a preparative reaction was run
of the Kau2 EH, sterically more demanding epoxides on a 1 g scale at 25 g/L of 18. After 30 h of reaction
such as trans-stilbene and cis-stilbene oxides were time the epoxide was totally consumed and the
tested. Although these epoxides are not bi-functional, formed (1R,2R)-diol-19 was recovered in 99% ee and
they are of great interest as scouting molecules of the 84.5% isolated yields. This result is comparable to the
active site of Kau2 EH. As anticipated, the sterically data of a previous report dealing with the desymmet-
challenging trans-stilbene oxide (16) was a poorer rization of meso-epoxides using various EHs obtained
substrate than the above-mentioned epoxidic com- from a metagenomic library.[16] Using cell extracts of 4
pounds. Indeed, the remaining epoxide was obtained different EHs (BD 8877, BD 8676, BD 9300, BD
in enantiomerically pure form only after 24–48 h of 9883) (1R,2R)-diol (19) was obtained with ees ranging
reaction time, compared to approximately 1 h for the from 96 to 99.5%, while its antipode (1S,2S)-diol (19)
above-mentioned substrates. With the exception of was obtained in 99% ee using the BD 9196 EH.[16]
1 g/L of substrate concentration, which resulted in an
We can conclude from the above-mentioned results
ee of 93% for the remaining epoxide after 105 h, that the Kau2 EH is a particularly useful enzyme for
higher concentrations (5, 10 and 25 g/L) enabled the kinetic resolution/desymmetrization of 1,2-disub-
enantiomerically pure epoxide to be obtained after stituted epoxides which contain at least one phenyl
24 h (Scheme 5). At the highest substrate concentra- ring substituent. Indeed, exceptionally high stereose-
tion of 50 g/L an enantiomerically pure remaining ep- lectivity was demonstrated with this enzyme, leading
oxide was obtained but only after 48 h. In all cases in all but two cases to ees of at least 99% for both re-
20% MTBE were used as an immiscible organic maining epoxide and formed diol, which represents
phase with the biocatalyst concentration being twice a nearly perfect kinetic resolution (Table 1).
as high as that of the substrate (m/m). The reaction
Furthermore, the Kau2 EH-catalyzed hydrolysis
products of the trans-stilbene oxide transformations was conducted at very high substrate concentrations
comprised meso-diol 17 and residual (1R,2R)-16. (from 25 to 75 g/L) and was generally complete
When run on a preparative scale with 1 g of rac-16 at within 1 h, thus closely approaching industrial needs.
50 g/L for 5 h, the reaction afforded residual (1R,2R)- The enzyme tolerated well the presence of water-im-
16 with an ee of >99% in 45% isolated yield and miscible organic solvents such as di-isopropyl ether
meso-diol 17 in 46% isolated yield.
and MTBE, alleviating or diminishing the chemical
The reaction portfolio was then extended to the de- hydrolysis of the epoxide as well as substrate and
symmetrization of the meso-epoxide cis-stilbene oxide product inhibition. Furthermore, the use of organic
(18) (Scheme 5). The cis-epoxide exhibited quantita- co-solvents enabled the dissolution of very high quan-
tive transformation in the range from 1 to 25 g/L (the tities of substrate and product. The biocatalyst was
biomass concentration being double that of the sub- also tolerant to pH changes in the alkaline region
strate, m/m), the formed (1R,2R)-19 being in all cases (pH 7.0 to pH 8.5), offering the possibility to further
Scheme 5. Kinetic resolution of trans-stilbene oxide (16) and desymmetrization of cis-stilbene oxide (18).
6
asc.wiley-vch.de
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

FULL PAPERS
Enantioselective Bio-Hydrolysis of Various Racemic and meso Aromatic Epoxides
Table 1. Comparison of the preparative scale reaction of the
various tested substrates (1 g/L) using the Kau2 EH.
Substrate eeS [%] (isolated
eeP [%] (isolated
E
yield [%])
yield [%])
1
3
5
7
>99 (49)
>99 (47)
>99 (40)
>99 (45)
>99 (44)
>99 (44)
>99 (43)
99 (45)
>99 (47)
94 (44)
88 (46)
>99 (46)
>99 (45)
>99 (48)
>99 (44.5)
meso (46)
99 (84.5)
>200
>200
>100[a]
>200
>200
>200
>200
>200
199
9
11
13
16
18
20
Figure 2. Comparison of the relative Kau2 EH efficiencies
(vertical bars) for the kinetic resolutions of the various ep-
oxidic substrates at the same concentration of 25 g/L, but
under reaction conditions adapted for each substrate (see
the Experimental Section).
–
no reaction
[a]
The E value is probably lower due to spontaneous chemi-
cal hydrolysis of the substrate.
minimize the spontaneous chemical hydrolysis of leads to an approximately two-fold quicker resolution
fairly unstable epoxides. Some of the tested substrates process (ee of the remaining epoxide >99%) based
are potentially very valuable chiral synthons for the on the same amount of enzyme used. We can con-
synthesis of numerous biologically active target mole- clude: (i) the efficiency appears to be inversely pro-
cules. When compared to a recent report about the ki- portional to the size of the substituent (CH2Brꢁ
netic resolution of glycidates 1, 3 and 5 with whole CH2Cl> CN>COOMe>COOEt>Ph), and (ii)
cells of Galactomyces geotrichum, the Kau2 EH ex- enzyme-catalyzed resolution of trans-configured sub-
pressed in E. coli proved to be an outstanding biocat- strates exhibited higher efficiencies than for their cis-
alyst for getting access to optically pure epoxides.[12] counterparts.
The residual epoxides [(2S,3R)-1 and (2S,3R)-3] are
It clearly appears that a b-substitution at the oxir-
suitable as chiral synthons for accessing the taxol side ane ring (with respect to the phenyl group) plays
chain[17a,b] as well as for being used in the synthesis of a critical role on the outcome of the biohydrolysis.
the nootropic drug (À)-Clausenamide.[9] Unfortunate- Indeed, the presence of a substituent alkyl, phenyl,
ly, the residual epoxide (2S,3R)-5 did not bear the cor- ester, cyano, chloro or bromo) located at the b-carbon
rect configuration to be used as a chiral synthon for atom seems to preclude one enantiomer of the vari-
accessing Diltiazem. We have shown that bi-function- ous tested racemates from reacting with the Kau2
al epoxides can be used as substrates of the Kau2 EH EH, leading to the observed very high enantioselec-
with virtually no noticeable influence of different tivities found in almost all cases. More precisely, with
functionalities such as nitrile, methyl or ethyl ester trans-configured epoxides the reactive enantiomer
and bromine and chlorine on the reactivity. In the always corresponded to the one bearing the benzylic
latter cases, new terminal epoxides were easily acces- carbon atom of S-absolute configuration, the enzy-
sible without any loss in enantiopurity from the enzy- matic attack occurring selectively at this carbon atom
matically formed diol by simply using basic reaction (Scheme 6).
conditions. Even very sterically demanding substrates
In the case of cis-substitution the Kau2 EH proved
such as trans- and cis-stilbene oxides were also trans- to be totally inactive with cis-methyl phenylglycidate
formed by the enzyme; however, lower reaction rates (20). This is astonishing in view of the considerable
were observed. Even these demanding substrates re- enzymatic activities determined with structurally re-
acted to completion on a preparative scale within one lated compounds such as cis-methyl styrene oxide,[6]
day using the Kau2 EH. In order to compare the effi- rac-9, and meso-18. For the time being, no explana-
ciencies of the Kau2-EH in the kinetic resolutions of tion of this fact can be given. We can only speculate
the various substrates described above, we present in that the active site of the Kau2 EH is not compatible
Figure 2 the results of all the tested substrates at the with the methyl ester group of 20, in contrast to the
same concentration of 25 g/L under the various reac- less polar methyl, cyano and phenyl groups. Another
tion conditions described in the Experimental Section. intriguing result was that the replacement of a methyl
This comparison is based on the number of mmoles of group by a cyano group (rac-9) led to the loss of the
remaining epoxide at the end of the resolution pro- high enantioconvergency found previously[6] in favor
cess (epoxide ee >99%) divided by the reaction time of a perfect enantioselectivity. It should be noted in
and the number of enzymatic units used in the reac- that case that, in contrast to its trans-counterpart, the
tion. That means that a two-fold higher efficiency reactive enantiomer corresponded to the one bearing
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
7
ÞÞ
These are not the final page numbers!

FULL PAPERS
Wei Zhao et al.
tively. Luria Bertani (LB) medium contained (per litre) 10 g
of peptone, 5 g of yeast extract and 5 g of NaCl and was ad-
justed to pH 7.5. For solid media 15 g/L of agar were added.
The minimal growth medium was composed of (per litre)
sucrose 10 g, KH2PO4 13.6 g, (NH4)2SO4 4 g, NaOH 3 g,
MgSO4·7H2O 2 g, CaCl2 0.377 g, FeSO4·7H2O 0.01 g and ad-
justed to pH 7.4.
Engineering E. coli RE3 expressing Kau2 EH: The con-
struction of the IPTG-inducible Kau2 EH-encoding plasmid
was described elsewhere.[6] Transformation of E. coli RE3
with this DNA construct resulted in ampicillin-resistant
Kau2-overproducers that were able to grow in minimal
medium with sucrose as the sole carbon source.[18]
Kau2 EH production: Frozen glycerol stock suspensions
of E. coli RE3 cells expressing Kau2 EH were spread on
Petri dishes filled with LB agar containing 100 mg/mL of am-
picillin and grown for 24 h at 288C. One colony was used to
inoculate an Erlenmeyer flask (250 mL) filled with 30 mL of
pre-culture minimal medium or with 30 mL of LB medium,
both containing 100 mg/mL of ampicillin. The pre-culture
was grown overnight at 318C. The bioreactor was filled with
3 L of minimal medium (adjusted to pH 7.4) and was inocu-
lated directly using the entire pre-culture. The cultivation
temperature was 288C, pO2 was maintained above 20% (air
flow 1 VVM), and the stirring speed was set at 300 rpm.
After 8–9 h of cultivation, the temperature was reduced
from 288C to 258C, and subsequently IPTG (1M in distilled
water) was added to a final concentration of 0.45 mM. After
15 h of cultivation at 258C, the cells were harvested by cen-
trifugation and then lyophilized. Two cell samples were ob-
tained with specific activities of 1667 U/g and 2757 U/g (Bio-
Bundle bioreactor from Applikon Biotechnology, 10 L).
Determination of Kau2 EH activity: The EH activity was
determined using (S)-para-chlorostyrene oxide as a substrate.
Fifteen mL of a 400 mM stock solution in acetonitrile (final
concentration 3.0 mM) and lyophilized cells expressing the
Kau2-EH (0.1 g/L) were added to 2 mL of buffer A in a 10-
mL round-bottom flask. The enzymatic reaction was incu-
bated at 278C under stirring (800 rpm). Samples of 200 mL
were withdrawn at regular intervals, diluted with 200 mL of
acetonitrile and centrifuged at 12000 g for 3 min, the super-
natant (200 mL) was then filtered (0.45 mm) and analyzed by
HPLC (Agilent 1100 Series) on a NUCLEOSILꢄ C18
column (250ꢅ4.60 mm, Macherey–Nagel) in isocratic mode
(55% acetonitrile in distilled water, flow rate of 0.7 mL/
min). Residual epoxide and formed diol eluted at 15.5 and
5.5 min, respectively (detection at 220 nm), they were quan-
tified using hydrobenzoin as a standard with a retention
time of 10.1 min.
Scheme 6. Kau2 EH attack at the benzylic carbon atom of
trans-configured 1,2-disubstituted aromatic epoxides.
the benzylic carbon atom of R-absolute configuration,
the enzymatic attack occurring selectively at the b-
carbon atom bearing the cyano group.
Conclusions
In conclusion, we have shown that the Kau2 EH is an
outstanding catalyst, which enables an easy, robust
and efficient access to numerous useful optically pure
diols and epoxides that can be used, for example, as
chiral building blocks in the synthesis of various bio-
logically active chemicals.
Experimental Section
General Remarks
All reagents were used as received from Sigma–Aldrich,
Acros or Fluka. Dimethylformamide (DMF), tetrahydrofur-
an (THF), dichloromethane, methanol, diisopropyl ether,
tert-butyl methyl ether (MTBE) and isooctane were of ana-
lytical grade. Acetonitrile, hexane and isopropyl alcohol
were of HPLC grade. The epoxides rac-methyl trans-3-(4-
methoxyphenyl)glycidate (5), rac-trans-stilbene oxide (16)
and meso-cis-stilbene oxide (18) were from Sigma–Aldrich.
All other reagents were from Sigma–Aldrich, ACROS or
1
Fluka. H and 13C NMR spectra were recorded on a Bruker
Chemical Synthesis
Synthesis of methyl trans-3-phenylglycidate (rac-1):[19] In
Avance 300 Ultrashield NMR spectrometer. Chemical shifts
are given in ppm relative to residual peaks of chloroform-d3
or acetone-d6. The coupling constants J are given in Hertz
(Hz). The abbreviations br, s, d, t and m correspond to
a broad signal, a singlet, a doublet, a triplet and a multiplet,
respectively.
a
100-mL round-bottom flask were weighed 0.896 g
(5.5 mmol) of trans-methyl cinnamate, followed by the addi-
tion of 20 mL of acetonitrile and 5 mL distilled water. Then
1.87 g of dibromoamine-T[20] were added, and the mixture
was stirred at room temperature for 30 min. The formed
bromohydrin was cyclized by adding 5.2 g of K2CO3, leaving
the reaction under stirring at room temperature overnight.
The reaction was stopped by addition of 1.0 g of Na2S2O3.
The mixture was poured into a separatory funnel, extracted
Biocatalyst Production and EH Activity
Buffer and culture media: Buffers A and B were composed
of 50 mM Na-phosphate, adjusted to pH 7.0 and 8.5, respec-
8
asc.wiley-vch.de
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

FULL PAPERS
Enantioselective Bio-Hydrolysis of Various Racemic and meso Aromatic Epoxides
3 times with 50 mL of diethyl ether, the organic phases were
collected, dried over MgSO4, filtered and evaporated under
vacuum. The product was purified by flash chromatography
on silica gel (230–400 mesh) using an 8/2 mixture of pentane
and diethyl ether as eluent, affording rac-methyl trans-3-
phenylglycidate (1) as a colorless liquid; yield: 0.81 g (82%).
1H NMR (CDCl3, 300 MHz): d=7.2–7.3 (brm, 5H), 4.1 (d,
J=1.5 Hz, 1H), 3.8 (s, 3H), 3.5 (d, J=1.6 Hz, 1H);
13C NMR (CDCl3, 75 MHz): d=168.6, 134.9, 129.0, 128.7,
125.8, 58.0, 56.6, 52.6.
2008C under vacuum (0.4 mbar), affording trans-2-(bromo-
methyl)-3-phenyloxirane rac-11 as a slightly yellow liquid;
yield: 0.302 g (49%). H NMR (CDCl3, 300 MHz): d=7.3–
1
7.4 (brm, 5H), 3.7 (d, J=1.8 Hz, 1H), 3.4 (d, J=5.8 Hz,
2H), 3.2 (dt, J=1.8 Hz, J=5.8 Hz, 1H); 13C NMR (CDCl3,
75 MHz): d=136.0, 128.6, 128.6, 125.7, 61.0, 60.3, 31.9.
A similar procedure starting from cinnamyl chloride af-
forded trans-2-(chloromethyl)-3-phenyloxirane rac-13 as
1
a colourless liquid; yield: 0.275 g (52.1%). H NMR (CDCl3,
300 MHz): d=7.2–7.4 (brm, 5H), 3.8 (d, J=1.8 Hz, 1H),
3.72 (dd, J=5.0 Hz, J=11.7 Hz, 1H), 3.66 (dd, J=5.7 Hz,
J=11.7 Hz, 1H), 3.3 (ddd, J=1.9 Hz, J=5.0 Hz, J=5.7 Hz,
1H); 13C NMR (CDCl3, 75 MHz): d=135.9, 128.6, 128.6,
125.7, 61.0, 58.6, 44.4.
Synthesis of ethyl trans-3-phenylglycidate (rac-3):[19] In
a
100-mL round-bottom flask were weighed 0.979 g
(5.5 mmol) of trans-ethyl cinnamate, followed by the addi-
tion of 20 mL of acetonitrile and 5 mL of distilled water.
Then 3.949 g of dibromoamine-T were added, and the mix-
ture was stirred at room temperature for 30 min. The next
preparatory steps were as described above for the synthesis
of 1. Flash chromatography (as described above) afforded
rac-ethyl trans-3-phenylglycidate (3) as a colorless liquid;
Synthesis of methyl cis-3-phenylglycidate (rac-20):[22]
A
mixture of cis-3-phenyloxirane-2-carbonitrile-9 (0.302 g,
2 mmol), dry potassium carbonate (0.289 g, 2.0 mmol) and
dry methanol (10 mL) was stirred at room temperature for
3 h and then acidified with dilute HCl for 2 h at 48C. The re-
action solution was then extracted 3 times by ethyl acetate
(30 mL), the combined organic layers were washed with
brine then water, dried over MgSO4, filtrated and finally
concentrated under vacuum. The crude product was purified
by flash chromatography on silica gel (230–400 mesh) using
a 7/3 mixture of pentane and diethyl ether as an eluent, af-
fordingrac-methyl cis-3-phenylglycidate-20 as a colorless
liquid; yield: 0.287 g (80%). 1H NMR (CDCl3, 300 MHz):
d=7.3–7.4 (brm, 5H), 4.3 (d, J=4.5 Hz, 1H), 3.8 (d, J=
4.5 Hz, 1H), 3.6 (brs, 3H); 13C NMR (CDCl3, 75 MHz): d=
166.3, 133.0, 128.4, 128.3, 126.3, 58.3, 56.0, 53.3.
Chemical hydrolysis of methyl trans-3-phenylglycidate
(rac-1): For access to racemic forms of the various diol
products, acidic opening of the corresponding epoxides was
carried out as exemplified for rac-1. In a 100-mL round bot-
tomed flask were added 100 mg of rac-1, 20 mL of water
and two drops of H2SO4 (98%). The reactions were per-
formed overnight at room temperature under magnetic stir-
ring (800 rpm). The reaction mixture was then extracted
with 20 mL of ethyl acetate, and the formed rac-diols ana-
lyzed by chiral GC or HPLC. The chemical hydrolyses of
rac-3, rac-5, rac-7, rac.9, rac-11, rac-13 and rac-16 were car-
ried out using the same protocol leading to diastereoisomer-
ic mixtures of corresponding diols 4, 6, 8, 10, 12, 14 and 17.
Synthesis of (1R,2S)-1-phenylglycidol (15): In a 25-mL
round-bottom flask were weighed 123.5 mg (7.0 mmol) of
(2R,3R)-12, followed by the addition of 3 mL of THF. Then
56 mg (14.0 mmol) of NaOH were added, and the mixture
was stirred at 08C for 2 h. The mixture was poured into
a separatory funnel, extracted 3 times with 15 mL of ethyl
acetate, the organic phases were collected, dried over
MgSO4, filtered and evaporated under vacuum. The product
was purified by flash chromatography on silica gel (230–400
mesh) using an 7/3 mixture of pentane and diethyl ether as
eluent, affording (1R,2S)-1-phenylglycidol (15) as a colorless
liquid; yield: 57 mg (59%). (2R,3S)-15: 1H NMR (CDCl3,
300 MHz): d=7.3–7.4 (brm, 5H), 5.0 (d, J=2.7 Hz, 1H), 3.3
(dt, J=2.9 Hz, J=3.9 Hz,1H), 3.0 (dd, J=2.8 Hz, J=4.9 Hz,
1H), 2.8 (dd, J=4.1 Hz, J=4.9 Hz, 1H), 2.3 (brs, 1H);
13C NMR (CDCl3, 75 MHz): d=139.5, 128.6, 128.3, 126.4,
70.8, 55.1, 43.6.
1
yield: 0.82 g (78%). H NMR (CDCl3, 300 MHz): d=7.2–7.3
(brm, 5H), 4.28 (q, J=7.2 Hz, 1H), 4.27 (q, J=7.2 Hz, 1H),
4.1 (d, J=1.7 Hz, 1H), 3.5 (d, J=1.7 Hz, 1H), 1.3 (t, J=
7.2 Hz, 3H); 13C NMR (CDCl3, 75 MHz): d=168.2, 135.0,
129.0, 128.6, 125.8, 61.8, 57.9, 56.7, 14.1.
Synthesis of trans- and cis-3-phenyloxirane-2-carboni-
trile (rac-7) and (rac-9):[21] To a solution of benzaldehyde
(380 mg, 3.0 mmol), chloroacetonitrile (450 mg, 6.0 mmol)
and tetrahexylammonium bromide (154 mg, 0.36 mmol) in
THF (15.0 mL) was added KOH (400 mg, 7.2 mmol) at
room temperature. After 22 h under stirring, NaBH4
(200 mg) was added to the mixture to remove the remaining
benzaldehyde. After 5 min, the reaction was quenched with
water, and the mixture was extracted 3 times with 15 mL of
ethyl acetate. The combined organic layers were washed
with brine and water, dried over MgSO4, filtrated and finally
concentrated under vacuum. The crude mixture was purified
by flash chromatography on silica gel (230–400 mesh) using
a 9/1 mixture of pentane and diethyl ether as an eluent, af-
fording trans-7 as a colorless liquid (yield: 0.09 g, 22.5%)
and cis-9 as a white solid (yield: 0.16 g, 40% yield, mp 54–
558C). 7: 1H NMR (CDCl3, 300 MHz): d=7.4–7.5 (brm,
3H), 7.2–7.3 (brm, 2H), 4.3 (d, J=1.8 Hz, 1H), 3.4 (d, J=
1.8 Hz, 1H); 13C NMR (CDCl3, 75 MHz): d=132.8, 129.8,
129.0, 125.8, 116.2, 58.5, 44.7. 9: 1H NMR (CDCl3,
300 MHz): d=7.4–7.5 (m, 5H), 4.3 (d, J=3.7 Hz, 1H), 3.8
(d, J=3.7 Hz, 1H); 13C NMR (CDCl3, 75 MHz): d=131.4,
129.7, 128.7, 126.3, 115.0, 57.7, 45.1.
Synthesis of trans-2-(bromomethyl)- and trans-2-(chloro-
methyl)-3-phenyloxirane (rac-11) and (rac-13): To a solu-
tion of 70% meta-chloroperoxybenzoic acid (mCPBA)
(8.5 mmol) in 20 mL of CH2Cl2 was added 30 mL of a cin-
namyl bromide (571 mg, 2.9 mmol) solution in CH2Cl2. Then
20 mL of 0.5M NaHCO3 were added, and the heterogene-
ous mixture was stirred at 258C for 17 h. The resulting mix-
ture was washed 3 times with 10% sodium bicarbonate, and
the combined aqueous phases were extracted 3 times with
30 mL of diethyl ether. The combined organic layers were fi-
nally washed with brine then water, dried over MgSO4, fil-
trated and then concentrated under vacuum. The crude
product was first purified by flash chromatography on silica
gel (230–400 mesh) using a 9/1 mixture of pentane and di-
ethyl ether as eluent and then by bulb-to-bulb distillation at
The same protocol was used for (2R,3R)-14 leading to the
same (1R,2S)-1-phenylglycidol (15).
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
9
ÞÞ
These are not the final page numbers!

FULL PAPERS
Wei Zhao et al.
Synthesis of (2S,3R)-2 from cyano-diols (2S,3R)-8 and
(2R,3R)-10: (2S,3R)-8 from the bioconversion of rac-7
(400 mg, 2.45 mmol) was dissolved in acetone dimethyl
acetal (40 mL), and then Dowexꢄ 50WX8 (4 g) was added
to the solution. The reaction was monitored by TLC for
1.5 h, and then the mixture was filtered and the filtrate
evaporated under reduced pressure. The residue was dis-
solved in a mixture of methanol and water (2:1), and KOH
(0.59 g) was added to the solution. Then the reaction solu-
tion was refluxed overnight at 758C. The reaction was
stopped by adding HCl (20%) and adjusted to pH 2. The
mixture was extracted 3 times with 40 mL diethyl ether and
then dried over MgSO4. The diethyl ether was removed by
rotary evaporation, the crude product was dissolved in
methanol, and a few drops of H2SO4 (98%) were added to
the solution. The mixture was stirred at room temperature
for 1.5 h, and was quenched with a saturated aqueous solu-
tion of sodium bicarbonate (20 mL). The mixture was ex-
tracted 3 times with 40 mL diethyl ether, dried over MgSO4
and evaporated under vacuum. The product was purified by
flash chromatography on silica gel (230–400 mesh) using an
50/50 mixture of pentane and diethyl ether as eluent, afford-
ing (2S,3R)-2; yield: 85 mg. The same protocol was used for
(2R,3R)-10, which was obtained from bioconversion of rac-
9, leading to the same (2S,3R)-2.
solid (mp 77–788C) (yield: 0.52 g. 47%; ee >99%). Absolute
configurations were determined by comparison of the opti-
cal rotations found in the literature for (2R,3R)-2 {[a]2D0:
À41.3 (c 0.48, CHCl3)}[23] and (2S,3R)-1 {[a]2D2: +171 (c 1.0,
CHCl3)}[9] with the ones experimentally determined in this
work {[a]22: À44 (c 0.5, CHCl3) and [a]1D5: +171.9 (c 1.0,
CHCl3), Drespectively}. (2R,3R)-2: 13C NMR (CDCl3,
75 MHz): d=172.4, 138.5, 128.3, 128.2, 126.4, 75.0, 74.8,
1
52.4; H NMR (CDCl3, 300 MHz): d=7.2–7.3 (brm, 5H), 4.9
(brdd, 1H), 4.4 (brdd, J=4.4 Hz, J=5.9 Hz, 1H), 3.6 (s,
3H), 2.9 (2brs, 2H).
Bioconversion of ethyl trans-3-phenylglycidate (rac-3):
Analytical scale – The same protocol as described above was
used. The samples were analyzed by chiral GC with a Cyclo-
sil B column (60 m, 0.25 mmꢅ0.25 mm, Agilent, USA), the
flow rate of the carrier gas (H2) was set at 2.67 mL/min. The
column temperature was set at 1508C and kept at this tem-
perature for 20 min and then increased to 1808C at a rate of
108C/min. The retention times were as follows. (2R,3S)-3:
22.7 min, and (2S,3R)-3: 23.0 min. The formed diols were an-
alyzed at 1808C without derivatization, the retention times
were as follows: (2R,3R)-4, 17.7 min; (2S,3S)-4, 18.0 min;
and for the 2 syn-diols-4 18.5 and 18.8 min. A Chiralpak
AD-H column (250ꢅ4.60 mm) in isocratic mode was used
for the determination of the absolute configuration of
(2R,3R)-4 using a mixture of hexane/isopropyl alcohol (85/
15) at a flow rate of 0.7 mL/min. UV detection was set at
230 nm. The retention times were as follows. (2S,3S)-4:
15.9 min, (2R,3R)-4: 16.7 min, and for the 2 syn-diols-4: 18.8
and 20.6 min.
Kau2 EH Enantioselective Hydrolysis of Epoxide
Substrates
Bioconversion of methyl trans-3-phenylglycidate (rac-1):
Analytical scale – A specific volume of a stock solution of
2.8M of rac-1 (final concentration 1–100 g/L) in DMF (final
concentration of DMF adjusted to 5%) and lyophilized bio-
mass with an EH activity of 1667 U/g (final concentration 2–
200 g/L) were added to 0.4 mL of buffer A containing 50 mL
of diisopropyl ether in a 10-mL round bottom flask (final
reaction volume 0.5 mL). The enzymatic reactions were in-
cubated at 278C under magnetic stirring (1200 rpm). Sam-
ples, which were withdrawn at regular intervals, were ex-
tracted with ethyl acetate and analyzed by chiral GC on
a Lipodex-G column (0.25 mm, 25 mꢅ0.25 mm, Macherey–
Nagel) for ee analysis of both residual substrate and formed
diol, 4-bromoacetophenone being used as an internal stan-
dard. The flow rate of the carrier gas (H2) was set at
2.96 mL/min. The following oven temperature program ena-
bled separation of the enantiomers: 1108C for 20 min, fol-
lowed by an increase to 1408C using a rate of 108CminÀ1.
The retention times were as follows. (2R,3S)-1: 17.5 min,
(2S,3R)-1: 17.9 min, (2R,3R)-2: 38.2 min, (2S,3S)-2: 38.6 min,
and for the 2 syn-diols-2: 39.1 and 40.1 min.
Preparative scale – One g of rac-1 dissolved in 1 mL of
DMF was added to 17 mL of buffer A and 2 mL of diiso-
propyl ether in a 100-mL round-bottomed flask. The biocat-
alytic reaction was initiated at 278C by the addition of 1.5 g
of lyophilized biomass with an EH activity of 1667 U/g
under magnetic stirring (1200 rpm). After 60 min, the reac-
tion mixture was extracted 3 times with 40 mL ethyl acetate.
The combined organic phases were dried over MgSO4, fil-
tered and then evaporated under reduced pressure. The
products were purified by flash chromatography (pentane/
diethyl ether, 7/3), affording(2S,3R)-1 as a colorless liquid
(yield: 0.49 g, 49%; ee >99%) and (2R,3R)-2 as a white
Preparative scale – In a 100-mL round-bottomed flask was
added 1.0 g of rac-3 dissolved in 2 mL of DMF, 34 mL of
buffer A and 4 mL of diisopropyl ether. The biocatalytic re-
action was initiated by the addition of 2.0 g of lyophilized
biomass with an EH activity of 1667 U/g. The reaction was
performed at 278C under magnetic stirring (1200 rpm).
After 60 min, the reaction mixture was extracted 3 times
with 40 mL ethyl acetate. The combined organic phases
were dried over MgSO4, filtered and then evaporated under
reduced pressure. The products were purified by flash chro-
matography (pentane/diethyl ether, 7/3), affording (2S,3R)-3
as a colourless liquid (yield: 0.47 g, 47%; ee >99%) and
(2R,3R)-4 as a colorless liquid (yield: 0.49 g, 44%; ee 94%);
[a]2D5: À37.5 (c 1.0, CHCl3). The absolute configuration of
(2S,3R)-3 was determined by comparison of the optical rota-
tion found in the literature {[a]2D0: + 152 (c 1.0, CHCl3)}[24]
with the one experimentally determined in this work {[a]2D5:
+154.7 (c 1.0, CHCl3)}. The absolute configuration of
(2R,3R)-4 was deduced from the published elution order of
(2S,3S)-4 and (2R,3R)-4 on a Chiralpak AD-H column.[13]
(2R,3R)-4: 1H NMR (CDCl3, 300 MHz): d=7.2–7.3 (brm,
5H), 4.9 (brd, J=4.0 Hz, 1H), 4.4 (brs, 1H), 4.1 (q, J=
7.2 Hz, 2H), 3.0 (brs, 2H), 1.1 (t, J=7.2 Hz, 3H); 13C NMR
(CDCl3, 75 MHz): d=171.9, 138.6, 128.3, 128.2, 126.4, 75.0,
74.7, 61.9, 14.0.
Bioconversion of methyl trans-3-(4-methoxyphenyl)glyci-
date (rac-5): Analytical scale – A specific volume of a stock
solution of 100 g/L of rac-5 (final concentration 1–50 g/L) in
MTBE (final concentration of MTBE 40%) and lyophilized
biomass with an EH activity of 1667 U/g (final concentration
2–100 g/L) were added to 0.3 mL of buffer B in a 10-mL
round-bottom flask (final reaction volume 0.5 mL). The en-
10
asc.wiley-vch.de
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

FULL PAPERS
Enantioselective Bio-Hydrolysis of Various Racemic and meso Aromatic Epoxides
zymatic reactions were incubated at 178C under magnetic
stirring (800 rpm). Regularly withdrawn aliquots were satu-
rated with NaCl and extracted with ethyl acetate (200 mL).
Then 100 mL of the organic phase were evaporated under re-
duced pressure, and the residues were dissolved in 100 mL of
isopropyl alcohol. Then 20 mL of this solution were analyzed
by HPLC to determine the ees of both remaining epoxide
and formed diol. A Lux Cellulose-4 chiral column (250ꢅ
4.60 mm, Phenomenex) was used in the isocratic mode with
an 8/2 mixture of hexane/isopropanol at a flow rate of
1.2 mL/min. UV detection was set at 230 nm. The retention
times were as follows: (2S,3R)-5: 8.4 min, (2R,3S)-5:
8.8 min, (2S,3S)-6: 17.7 min, (2R,3R)-6: 19.8 min, and for the
2 syn-diols-6: 21.7 and 27.4 min.
lows: (2R,3R)-7: 16.8 min, (2S,3S)-7: 19.4 min, (2S,3R)-8:
21.6 min, (2R,3S)-8: 25.5 min.
Preparative scale – In a 100-mL round-bottomed flask was
added 1.0 g of rac-7 dissolved in 4 mL of MTBE and 16 mL
of buffer A. The biocatalytic reaction was initiated by addi-
tion of 0.75 g of lyophilized biomass with an EH activity of
2757 U/g, and the reaction mixture was maintained at 278C
under magnetic stirring (800 rpm). After 60 min, the reac-
tion mixture was extracted 3 times with 40 mL ethyl acetate.
The combined organic phases were dried over MgSO4, fil-
tered and then evaporated under reduced pressure. The
products were purified by flash chromatography (pentane/
diethyl ether, 7/3), affording (2R,3R)-7 as a slightly yellow
liquid (yield: 0.45 g, 45%; ee 99%) and (2S,3R)-8 as a white
solid (yield: 0.52 g, 46%; ee 99%). [a]2D2: À44.3 (c 1.0,
EtOH), mp 83–848C. The absolute configuration of
(2R,3R)-7 was determined by comparison of the optical ro-
tation value found in the literature for (2R,3R)-7 {29% ee,
[a]2D5: +24.3 (c 0.86, EtOH)}[15] with the one experimentally
determined in this work {[a]2D0: +150.8 (c 1.1, EtOH)}. The
absolute configuration of the formed diol 8 was determined
by its transformation to the acetonide 8a, which proved to
be of cis-configuration (1H NMR, see the Supporting Infor-
mation), and subsequently to the corresponding syn-methyl
ester diol 2 according to a previously described method
(Scheme 3).[26] It should be noted that during the overnight
reflux an epimerization of cis-acetonide 8a occurred at
carbon-2. Such an epimerization did not occur when starting
Preparative scale – In a 100-mL round-bottomed
flask was added 1.0 g of rac-5 dissolved in 10 mL of
MTBE and 15 mL of buffer B. The biocatalytic reac-
tion was initiated by addition of 1.75 g of lyophilized
biomass with an EH activity of 1667 U/g, and the re-
action mixture was kept at 178C under magnetic stir-
ring (800 rpm). After 60 min, the reaction mixture
was extracted 3 times with 40 mL ethyl acetate. The
combined organic phases were dried over MgSO4, fil-
tered and then evaporated under reduced pressure.
The products were purified by flash chromatography
(pentane/diethyl ether, 7/3), affording (2S,3R)-5 as
a colorless liquid (yield: 0.40 g, 40%; ee >99%) and from the corresponding trans-acetonide (vide infra).[26] The
optical rotation of the formed diol 2 {[a]2D0: À12.7 (c 0.71,
(2R,3R)-6 as a white solid (yield: 0.51 g, 46%; ee
CH2Cl2)} was compared with the one found in the literature
88%), mp 105–1068C. The absolute configurations
for (2S,3R)-syn-methyl-ester-diol 2 {[a]2D0: À16.2 (c 0.69,
were determined by comparison of the optical rota-
CH2Cl2)},[26] establishing the absolute configuration of the
tion values found in the literature for (2R,3S)-5 {[a]2D4:
obtained methyl-ester-diol 2 and consequently the (2S,3R)
À205 (c 1.0, MeOH)}[25] and (2R,3R)-6 {[a]1D8: À43.9 (c
configuration of bio-catalytically formed cyano-diol 8.
1.0, CHCl3)}[23] with the ones experimentally deter-
(2S,3R)-8: 1H NMR (acetone-d6, 300 MHz): d=7.5 (brm,
mined in this work {<[a]2D4: +190.7 (c 1.0, MeOH)
2H), 7.3–7.4 (brm, 3H), 5.7 (d, J=6.8 Hz, 1H), 5.3 (d, J=
and [a]1D8: À39.2 (c 1.0, CHCl3), respectively}. (2R,3R)-
4.3 Hz, 1H), 4.9 (t, J=4.3 Hz, 1H), 4.6 (d, J=5.7 Hz, 1H);
1
13C NMR (acetone-d6, 75 MHz): d=139.9, 128.1, 127.9,
126.7, 118.8, 74.4, 67.1.
6: H NMR (CDCl3, 300 MHz): d=7.2–7.3 (brm, 2H),
6.8–6.9 (brm, 2H), 5.0 (d, J=4.3 Hz, 1H), 4.5 (d, J=
4.3 Hz, 1H), 3.8 (s, 3H), 3.7 (s, 3H), 3.0 (brs, 2H);
13C NMR (CDCl3, 75 MHz): d=172.5, 159.5, 130.1,
127.7, 113.7, 74.8, 74.6, 55.2, 52.5.
Bioconversion of cis-3-phenyloxirane-2-carbonitrile (rac-
9): Analytical scale – A specific volume of a stock solution
of 100 g/L of rac-9 (final concentration 1–50 g/L) in MTBE
(final concentration of MTBE 20%) and lyophilized biomass
with an EH activity of 2757 U/g (final concentration 1–50 g/
L) were added to 0.8 mL of buffer A in a 10-mL round-
bottom flask (final volume of reaction: 1 mL). Enzymatic
reactions were incubated at 278C under magnetic stirring
(800 rpm). Regularly withdrawn aliquots were saturated
with NaCl and extracted with ethyl acetate (200 mL). Then
100 mL of the organic phase were evaporated under reduced
pressure, and the residues were dissolved in 100 mL of iso-
propyl alcohol. Then 20 mL of this solution were analyzed
by HPLC to determine the ees of both remaining epoxide
and formed diol. A Chiralcel OD-H column (250ꢅ4.60 mm)
was used in isocratic mode with a 92.5/7.5 mixture of
hexane/isopropyl alcohol at a flow rate of 1.2 mL/min. UV
detection was set at 215 nm. The retention times were as fol-
lows: (2S,3R)-9: 11.2 min, (2R,3S)-9: 12.0 min, (2R,3R)-10:
20.6 min, (2S,3S)-10: 23.9 min.
Bioconversion of trans-3-phenyloxirane-2-carbonitrile
(rac-7): Analytical scale – A specific volume of a stock solu-
tion of 200 g/L of rac-7 (final concentration 1–100 g/L) in
MTBE (final concentration of MTBE 20%) and lyophilized
biomass with an EH activity of 2757 U/g (final concentration
0.75–75 g/L) were added to 0.8 mL of buffer A in a 10-mL
round-bottom flask (final volume of reaction 1 mL). The en-
zymatic reactions were incubated at 278C under magnetic
stirring (800 rpm). Regularly withdrawn aliquots were satu-
rated with NaCl and extracted with ethyl acetate (200 mL).
Then 100 mL of the organic phase were evaporated under re-
duced pressure, and the residues were dissolved in 100 mL of
isopropyl alcohol. Then 20 mL of this solution were analyzed
by HPLC for ee determination of both remaining epoxide
and formed diol. A Chiralcel OD-H column (250ꢅ4.60 mm)
was used in isocratic mode with a 92.5/7.5 mixture of
hexane/isopropyl alcohol at a flow rate of 1.2 mLminÀ1. UV
detection was set at 215 nm. The retention times were as fol-
Preparative scale – In a 100-mL round-bottomed flask was
added 1.0 g of rac-9 dissolved in 8 mL of MTBE and 32 mL
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
11
ÞÞ
These are not the final page numbers!

FULL PAPERS
Wei Zhao et al.
of buffer A. The biocatalytic reaction was initiated by addi-
tion of 1 g of lyophilized biomass with an EH activity of
2757 U/g. The reaction was performed at 278C under mag-
netic stirring (800 rpm). After 60 min, the reaction mixture
was extracted 3 times with 40 mL ethyl acetate. The com-
bined organic phases were dried over MgSO4, filtered and
then evaporated under reduced pressure. The products were
purified by flash chromatography (pentane/diethyl ether, 7/
3), affording (2R,3S)-9 as a white solid (yield: 0.44 g, 44%;
ee >99%), [a]2D5: +109.8 (c 0.86, EtOH), mp 54–558C and
(2R,3R)-10 as a slightly yellow liquid (yield: 0.51 g, 45.2%;
ee >99%), [a]2D5: À32.1 (c 1.0, EtOH). The absolute configu-
ration of residual epoxide 9 was determined after transfor-
mation of 9 to its cis-methyl-ester counterpart (2S,3S)-20
(Scheme 3).[27] The absolute configuration of (2S,3S)-20 was
determined by comparison of the optical rotation value
found in the literature for (2R,3R)-20 { [a]2D5: +10.8 (c 1.03,
CH2Cl2)}[22] with the one determined experimentally in this
work { [a]2D0: À9.1 (c 0.98, CH2Cl2)}. In order to determine
the absolute configuration of the formed diol 10, the same
protocol was used as for diol 8 (Scheme 3).[26] The optical
rotation of the formed diol 2 { [a]20: À14.7 (c 0.73, CH2Cl2)}
was compared to the one found inDthe literature for (2S,3R)-
syn-methyl-ester-diol 2 { [a]2D0: À16.2 (c 0.69, CH2Cl2)},[26] es-
tablishing the same absolute configuration for the obtained
methyl-ester-diol 2, the (2R,3R) configuration for trans-ace-
tonide 10a and thus the same (2R,3R) configuration for the
biocatalytically formed cyano-diol 10 (Scheme 3). (2R,3R)-
products were purified by flash chromatography (pentane/
diethyl ether, 7/3), affording (2S,3R)-11 as a colourless
liquid (yield: 0.44 g, 44%; ee 99%), [a]2D4: +11.3 (c 1, CHCl3)
and (2R,3R)-12 as a colourless liquid (yield: 0.53 g, 48%; ee
99%), [a]2D0: M
ACTHUNTRGNEU8GN .5 (c 1, CHCl3). Diol 12 was cyclized under
basic conditions to the corresponding 1-phenylglycidol 15,[28]
whose absolute configuration was established as (1R,2S) by
comparison of the optical rotation data found in the litera-
ture for (1R,2S)-15 {[a]25: À100.2 (c 2.31, CHCl3)}[24] with
the one experimentally oDbtained in this work for epoxy alco-
hol 15 arising from bromo-diol {[a]2D5: À104 (c 2.74, CHCl3).
(2R,3R)-12: 1H NMR (acetone-d6, 300 MHz): d=7.3–7.4
(brm, 5H), 4.8 (d, J=5.4 Hz, 1H), 4.0 (ddd, J=3.5 Hz, J=
5.3 Hz, J=7.8 Hz 1H), 3.5 (dd, J=7.7 Hz, J=10.6 Hz, 1H),
3.4 (dd, J=3.4 Hz, J=10.7 Hz, 1H), 2.3–3.1 (brs, 2H);
13C NMR (CDCl3, 75 MHz): d=139.6, 128.7, 128.3, 126.6,
75.0, 74.7, 35.8.
Bioconversion of 2-(chloromethyl)-3-phenyloxirane (rac-
13): Preparative scale – In a 100-mL round-bottomed flask
was added 1.0 g of rac-13 dissolved in 4 mL of MTBE and
16 mL of buffer A. The biocatalytic reaction was initiated
by addition of 0.5 g of lyophilized biomass with an EH activ-
ity of 2757 U/g. The reaction was performed at 278C under
magnetic stirring (800 rpm). After 40 min, the reaction mix-
ture was extracted 3 times with 40 mL ethyl acetate. The
combined organic phases were dried over MgSO4, filtered
and then evaporated under reduced pressure. The products
were purified by flash chromatography (pentane/diethyl
ether, 7/3), affording of (2S,3R)-13 as a colourless liquid
(yield: 0.43 g, 43%; ee >99%) and (2R,3R)-14 as a slightly
yellow liquid (yield: 0.49 g, 44.5%; ee >99%), [a]2D0: À4.6 (c
1, EtOH). The absolute configuration of (2S,3R)-13 was de-
termined by comparison of the optical rotation value found
in the literature for (2S,3R)-13 {[a]2D5: +21.4 (c 0.6,
CHCl3)}[29] with the one experimentally determined in this
work {[a]2D0: +20 (c 0.6, CHCl3)}. Diol 14 was cyclized in
basic conditions to the corresponding 1-phenylglycidol 15,[28]
whose absolute configuration was established as (1R,2S) by
comparison of the optical rotation data found in the litera-
ture for (1R,2S)-15 {[a]2D5: À100.2 (c 2.31, CHCl3)}+[30] with
the one experimentally obtained in this work for epoxy alco-
hol 15 arising from chloro-diol {[a]2D5: À100 (c 2.74, CHCl3)}.
(2R,3R)-14: 1H NMR (CDCl3, 300 MHz): d=7.3–7.4 (brm,
5H), 4.8 (d, J=5.0 Hz, 1H), 3.9 (m, 1H), 3.5 (m, 2H), 2.9–
3.2 (brs, 2H); 13C NMR (CDCl3, 75 MHz): d=139.6, 128.6,
128.3, 126.6, 74.9, 74.6, 46.1.
1
10: H NMR (acetone-d6, 300 MHz): d=7.5–7.6 (brm, 2H),
7.4–7.3 (brm, 3H), 5.7 (brs, 1H), 5.2 (brs, 1H), 4.9 (d, J=
5.8 Hz, 1H), 4.7 (d, J=5.8 Hz, 1H); 13C NMR (acetone-d6,
75 MHz): d=139.5, 128.2, 128.0, 127.3, 118.8, 74.4, 66.4.
Bioconversion of 2-(bromomethyl)-3-phenyloxirane (rac-
11): Analytical scale – A specific volume of a stock solution
of 100 g/L of rac-11 (final concentration 1–100 g/L) in
MTBE (final concentration of MTBE 20%) and lyophilized
biomass with an EH activity of 2757U/g (varying concentra-
tions: see text) were added to 0.8 mL of buffer A in a 10-
mL round-bottom flask (final volume of reaction 1 mL). En-
zymatic reactions were incubated at 278C under magnetic
stirring (800 rpm). Regularly withdrawn aliquots were satu-
rated with NaCl and extracted with ethyl acetate (200 mL).
Then 100 mL of the organic phase were evaporated under re-
duced pressure and the residues were dissolved in 100 mL of
isopropyl alcohol. Then 20 mL of this solution were analyzed
by HPLC to determine the enantiomeric excesses of both
remaining epoxide and formed diol. A Lux Cellulose-4
chiral column (250ꢅ4.60 mm, Phenomenex) was used in iso-
cratic mode with a 92.5/7.5 mixture of hexane/isopropyl al-
cohol at a flow rate of 1 mL/min. UV detection was set at
215 nm. The retention times were as follows: (2S,3R)-11:
8.5 min, (2R,3S)-11: 10.2 min, (2R,3R)-12: 20.6 min, (2S,3S)-
12: 22.6 min, and for the 2 syn-diols: 25.6 and 26.7 min.
Preparative scale – In a 100-mL round-bottomed flask was
added 1.0 g of rac-11 dissolved in 2.6 mL of MTBE and
10.7 mL of buffer A. The biocatalytic reaction was initiated
by addition of 0.8 g of lyophilized biomass with an EH activ-
ity of 2757 U/g. The reaction mixture was kept at 278C
under magnetic stirring (800 rpm). After 60 min, the reac-
tion mixture was extracted 3 times with 40 mL ethyl acetate.
The combined organic phases were dried over MgSO4, fil-
tered and then evaporated under reduced pressure. The
Bioconversion of trans-stilbene oxide (rac-16): Analytical
scale – In all experiments 100 mg of rac-16 were dissolved in
MTBE (from 0.4 to 20 mL) and buffer A was added to
make up a final MTBE concentration of 20% and a final
total concentration of 16 in the range of 1–50 g/L. The reac-
tion was initiated by the addition of 200 mg of lyophilized
biomass with an EH activity of 2757 U/g. The biotransfor-
mations were performed at 278C under stirring (1200 rpm).
Regularly withdrawn aliquots were saturated with NaCl and
extracted with ethyl acetate (200 mL). Then 100 mL of the or-
ganic phase were evaporated under reduced pressure, and
the residues were dissolved in 100 mL of isopropyl alcohol.
Then 20 mL of this solution were analyzed by HPLC to de-
termine the ees of both remaining epoxide and formed diol.
A Lux Cellulose-4 chiral column (250ꢅ4.60 mm, Phenomen-
ex) was used in isocratic mode with a 9/1 mixture of hexane/
12
asc.wiley-vch.de
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

FULL PAPERS
Enantioselective Bio-Hydrolysis of Various Racemic and meso Aromatic Epoxides
isopropyl alcohol at a flow rate of 1 mL/min. UV detection
was set at 215 nm. The retention times were as follows:
(1R,2R)-16: 4.9 min, (1S,2S)-16: 7.9 min, meso-diol-17:
17.3 min.
las Vanthuyne (Aix-Marseille University) for chiral HPLC
analyzes.
Preparative scale – In a 100-mL round-bottomed flask was
added 1.0 g of rac-16 dissolved in 4 mL of MTBE and
16 mL of buffer A. The biocatalytic reaction was initiated
by addition of 2 g of lyophilized biomass with an EH activity
of 2757 U/g. The reaction was performed at 278C under
magnetic stirring (800 rpm). After 5 h, the reaction mixture
was extracted 3 times with 40 mL ethyl acetate. The com-
bined organic phases were dried over MgSO4, filtered and
then evaporated under reduced pressure. The products were
purified by flash chromatography (pentane/diethyl ether, 1/
1), affording (1R,2R)-16 as a slightly yellow solid (yield:
0.45 g, 45%; ee 99%), mp 65–668C and meso-17 as a white
solid (yield: 0.51 g, 46%), mp 122–1248C. The absolute con-
figuration of (1R,2R)-16 was determined by comparison of
the optical rotation value found in the literature for
(1R,2R)-16 {[a]2D0: +357 (c 0.59, benzene)}[31] and the one ex-
perimentally determined in this work {[a]2D0: +348 (c 0.59,
benzene)}. meso-17: 1H NMR (acetone-d6, 300 MHz): d=
7.2–7.3 (brm, 10H), 4.9 (brs, 2H), 4.3 (s, 2H); 13C NMR
(acetone-d6, 75 MHz): d=142.0, 127.4, 127.3, 126.9, 77.7.
Bioconversion of meso-cis-stilbene oxide (18): Analytical
scale – A specific volume of a stock solution of 100 g/L of
meso-18 (final concentration 1–50 g/L) in DMF (final con-
centration of DMF 5%) and lyophilized biomass with an
EH activity of 2757 U/g (final concentration 2–100 g/L)
were added to 1.9 mL of buffer A in a 25-mL round-bottom
flask (final reaction volume: 2 mL). The enzymatic reactions
were performed at 278C under magnetic stirring (1200 rpm).
The next preparatory steps were performed as described for
the analytical scale bioconversion of compound 16. The re-
tention times were as follows: meso-18: 5.6 min, (1S,2S)-19:
23.9 min, and (1R,2R)-19: 26.9 min.
References
[1] P. Kumar, V. Naidu, P. Gupta, Tetrahedron 2007, 63,
2745–2785.
[2] Y. Chen, C. Chen, X. Wu, Chem. Soc. Rev. 2012, 41,
1742–1753.
[3] a) A. Archelas, R. Furstoss, Curr. Opin. Chem. Biol.
2001, 5, 112–119; b) A. Steinreiber, K. Faber, Curr.
Opin. Biotechnol. 2001, 12, 552–558; c) E. J. de Vries,
D. B. Janssen, Curr. Opin. Biotechnol. 2003, 14, 414–
420; d) E. Y. Lee, Biotechnol. Lett. 2008, 30, 1509–1514;
e) W. J. Choi, Appl. Microbiol. Biotechnol. 2009, 84,
239–247; f) M. Kotik, A. Archelas, R. Wohlgemuth,
Curr. Org. Chem. 2012, 16, 451–482.
[4] B. van Loo, J. Kingma, M. Arand, M. G. Wubbolts,
D. B. Janssen, Appl. Env. Microbiol. 2006, 72, 2905–
2917.
[5] a) M. Arand, F. Mꢆller, A. Mecky, W. Hinz, P. Urban,
D. Pompon, R. Kellner, F. Oesch, Biochem. J. 1999,
337, 37–43; b) R. Rink, J. H. Lutje Spelberg, R. J. Piet-
ers, J. Kingma, M. Nardini, R. M. Kellogg, B. W. Dijk-
stra, D. B. Janssen, J. Am. Chem. Soc. 1999, 121, 7417–
7418; c) B. van Loo, J. Kingma, G. Heyman, A. Witte-
naar, J. H. Lutje Spelberg, T. Sonke, D. B. Janssen,
Enzyme Microb. Technol. 2009, 44, 145–153; d) M. T.
Reetz, L.-W. Wang, M. Bocola, Angew. Chem. 2006,
118, 1258–1263; Angew. Chem. Int. Ed. 2006, 45, 1236–
1241; e) M. T. Reetz, D. Kahakeaw, R. Lohmer, Chem-
BioChem 2008, 9, 1797–1804; f) M. T. Reetz, H. Zheng,
ChemBioChem 2011, 12, 1529–1535; g) B. van Loo,
J. H. L. Spelberg, J. Kingma, T. Sonke, M. G. Wubbolts,
D. B. Janssen, Chem. Biol. 2004, 11, 981–990; h) M.
Preparative scale – In a 100-mL round-bottomed flask was
added 1.0 g of rac-18 dissolved in 2 mL of DMF and 38 mL
of buffer A. The biocatalytic reaction was initiated by the
addition of 2 g of lyophilized biomass with an EH activity of
2757 U/g. The reaction was performed at 278C under mag-
netic stirring (800 rpm). After 30 h, the reaction mixture was
extracted 3 times with 40 mL ethyl acetate. The combined
organic phases were dried over MgSO4, filtered and then
evaporated under reduced pressure. The products were puri-
fied by flash chromatography (pentane/diethyl ether, 1/1),
affording (1R,2R)-19 as a slightly yellow solid (yield: 0.92 g,
84.5%; ee 99%), mp 136–1388C. The absolute configuration
was determined by comparison of the optical rotation value
found in the literature for (1R,2R)-19, {[a]2D8: +92.7 (c 1,
EtOH)}[32] with the one experimentally determined in this
ˇ
Kotik, A. Archelas, V. Famerova, P. Oubrechtova, V.
ˇ
Kren, J. Biotechnol. 2011, 156, 1–10; i) H. Zheng, M. T.
Reetz, J. Am. Chem. Soc. 2010, 132, 15744–15751; j) M.
Kotik, W. Zhao, G. Iacazio, A. Archelas, J. Mol. Catal.
B: Enzym. 2013, 91, 44–51.
ˇ
ˇ
[6] M. Kotik, V. Stepꢂnek, M. Grulich, P. Kyslꢁk, A. Arche-
las, J. Mol. Catal. B: Enzym. 2010, 65, 41–48.
[7] a) J. S. Yadav, K. Premalatha, S. J. Harshavardhan, B. V.
Subba Reddy, Tetrahedron Lett. 2008, 49, 6765–6767;
b) R. Palin, D. R. Barn, J. K. Clark, J. E. Cottney, P. M.
Cowley, M. Crockatt, L. Evans, H. Feilden, R. R.
Goodwin, F. Griekspoor, S. J. A. Grove, A. K. Hought-
on, P. S. Jones, R. J. Morphy, A. R. C. Smith, H. Sundar-
am, D. Vrolijk, M. A. Weston, G. Wishart, P. Wren,
Bioorg. Med. Chem. Lett. 2005, 15, 589–593; c) P.
Kumar, A. Dubey, A. Harbindu, Org. Biomol. Chem.
2012, 10, 6987–6994.
1
work {[a]2D8: +93.8 (c 1, EtOH)}. (1R,2R)-19: H NMR (ace-
tone-d6, 300 MHz): d=6.9–7.1 (brm, 10H), 4.52 (brs, 2H),
4.47 (brs, 2H); 13C NMR (acetone-d6, 75 MHz): d=141.7,
127.5, 127.2, 127.1, 78.9.
[8] H. Hamamoto, V. A. Mamedov, M. Kitamoto, N. Haya-
shib, S. Tsuboi, Tetrahedron: Asymmetry 2000, 11,
4485–4497.
[9] G. J. Zheng, Q. Yuan, L. Yang, X. Zhang, J. J. Wang,
W. R. Sun, J. Mol. Catal. B: Enzym. 2006, 43, 133–136.
[10] A. Gentile, C. Giordano, C. Fuganti, L. Ghirotto, S.
Servi, J. Org. Chem. 1992, 57, 6635–6637.
Acknowledgements
We are indebted to Elise Courvoisier-Dezord (Plateforme
AVB, Aix-Marseille University) for performing bioreactor
fermentations. The authors also thank Marion Jean and Nico-
ˇ
ˇ
ˇ
ˇ
´
[11] Z. Hamersak, D. Sepac, D. Ziher, V. Sunjic, Synthesis
2003, 3, 375–382.
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
13
ÞÞ
These are not the final page numbers!

FULL PAPERS
Wei Zhao et al.
[12] C. Wei, J. Ling, H. Shen, Q. Zhu, Molecules 2014, 19,
8067–8079.
[13] A. V. Devi, C. Lahari, L. Swarnalatha, N. W. Fadnavis,
Tetrahedron: Asymmetry 2008, 19, 1139–1144.
[14] C. Li, Q. Liu, X. Song, D. Di, A. Ji, Y. Qu, Biotechnol.
Lett. 2003, 25, 2113–2116.
[22] J. Wang, F. D. Rochon, Y. Yang, L. Hua, M. M. Kayser,
Tetrahedron: Asymmetry 2007, 18, 1115–1123.
[23] B. R. Matthews, W. R. Jackson, H. A. Jacobs, K. G.
Watson, Aust. J. Chem. 1990, 43, 1195–1214.
[24] O. Cabon, D. Buisson, M. Larcheveque, R. Azerad, Tet-
rahedron: Asymmetry 1995, 6, 2211–2218.
[15] M. Zagozda, J. Plenkiewicz, Tetrahedron: Asymmetry
[25] R. Imashiro, M. Seki, J. Org. Chem. 2004, 69, 4216–
4226.
2008, 19, 1455–1460.
[16] L. Zhao, B. Han, Z. Huang, M. Miller, H. Huang, D. S.
Malashock, Z. Zhu, A. Milan, D. E. Robertson, D. P.
Weiner, M. J. Burk, J. Am. Chem. Soc. 2004, 126,
11156–11157.
[26] F. Effenberger, M. Hopf, T. Ziegler, J. Hudelmayer,
Chem. Ber. 1991, 124, 1651–1659.
ˇ
[27] J. Svoboda, Z. Kocfeldovꢂ, J. Palecek, Collect. Czech.
Chem. Commun. 1998, ##63##53, 822–832.
[28] N. K. Jobson, A. R. Crawford, D. Dewar, S. L. Pimlott,
A. Sutherland, Bioorg. Med. Chem. Lett. 2009, 19,
4996–4998.
[29] Z.-X. Wang, Y. Shi, J. Org. Chem. 1997, 62, 8622–8623.
[30] J. M. Palazon, B. Anorbe, V. S. Martin, Tetrahedron
Lett. 1986, 27, 4987–4990.
[17] a) D.-M. Gou, Y.-C. Liu, C.-S. Chen, J. Org. Chem.
1993, 58, 1287–1289; b) A. A. Afon’kin, L. M. Kostri-
kin, A. E. Shumeiko, A. F. Popov, A. A. Matveev, V. N.
Matvienko, A. F. Zadbudkin, Russian Chem. Bull.
2012, 61, 2149–2162.
ˇ
ˇ
[18] M. Grulich, J. Marsalek, P. Kyslꢁk, V. Stepꢂnek, M.
Kotik, Process Biochem. 2011, 46, 526–532.
[19] I. Saikia, B. Kashyap, P. Phukan, Synth. Commun. 2010,
40, 2647–2652.
[20] C. Nair, P. Indrasenan, Talanta 1976, 23, 239–241.
[21] L. Mhamdi, H. Bohli, Y. Moussaoui, R. ben Salem,
Intern. J. Org. Chem. 2011, 1, 119–124.
[31] G. Berti, F. Bottari, P. L. Ferrarini, B. Macchia, J. Org.
Chem. 1965, 30, 4091–4096.
[32] N. Takenaka, G. Xia, H. Yamamoto, J. Am. Chem. Soc.
2004, 126, 13198–13199.
14
asc.wiley-vch.de
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

FULL PAPERS
Enantioselective Bio-Hydrolysis of Various Racemic and
meso Aromatic Epoxides Using the Recombinant Epoxide
Hydrolase Kau2
15
Adv. Synth. Catal. 2015, 357, 1 – 15
Wei Zhao, Michael Kotik, Gilles Iacazio,* Alain Archelas
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢃ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
15
ÞÞ
These are not the final page numbers!