Enzyme-catalyzed nucleophilic ring opening of epoxides for the preparation of enantiopure tertiary alcohols

Article abstract of DOI:10.1002/adsc.200700146

The halohydrin dehalogenase from Agrobacterium radiobacter AD1 (HheC) catalyzes nucleophilic ring opening of epoxides with cyanide and azide. In the case of 2,2-disubstituted epoxides, this reaction proceeds with excellent enantioselectivity (E values up

Full text of DOI:10.1002/adsc.200700146

FULL PAPERS
DOI: 10.1002/adsc.200700146
Enzyme-Catalyzed Nucleophilic Ring Opening of Epoxides for
the Preparation of Enantiopure Tertiary Alcohols
a
b
a
c
´
Maja Majeric Elenkov, H. Wolfgang Hoeffken, Lixia Tang, Bernhard Hauer,
and Dick B. Janssena,*
a
Biochemical Laboratory, Groningen Biomolecular Sciences and Biotechnology Institute, University of Groningen,
Nijenborgh 4, 9747 AG, Groningen, The Netherlands
Phone: (+31)-50-3634209; fax: (+31) 50-3634165; e-mail: d.b.janssen@rug.nl
Physical Chemistry and Informatics, BASF AG, 67056 Ludwigshafen, Germany
Fine Chemicals and Biocatalysis Research, BASF AG, 67056 Ludwigshafen, Germany
b
c
Received: March 21, 2007; Revised: July 11, 2007
Abstract: The halohydrin dehalogenase from Agro- the enzyme has a broad substrate range and because
bacterium radiobacter AD1 (HheC) catalyzes nucleo- these tertiary alcohols are difficult to prepare in
philic ring opening of epoxides with cyanide and other ways, HheC is an attractive biocatalyst for the
azide. In the case of 2,2-disubstituted epoxides, this production of b-cyano and b-azido tertiary alcohols.
reaction proceeds with excellent enantioselectivity
(E values up to>200), which gives, by kinetic resolu-
tion, access to various enantiopure epoxides and b- Keywords: azides; cyanides; epoxides; halohydrin
substituted tertiary alcohols (ee up to 99%). Since dehalogenase; kinetic resolution; tertiary alcohols.
Introduction
reduced enantioselectivity have been described as
well.[10]
The asymmetric ring opening of epoxides catalyzed
Our previous work on ring opening of epoxides in-
by chiral metal complexes is an important synthetic dicated that the halohydrin dehalogenase from this
strategy for the preparation of valuable chiral build- organism could be a suitable catalyst for the resolu-
ing blocks, and a variety of nucleophiles have been tion of epoxides with azide[6] or cyanide[7] as a nucleo-
employed in these reactions.[1] For the enzymatic ring phile. Since, in the case of 2,2-disubstituted epoxides,
opening of epoxides, halohydrin dehalogenases have the products of such conversions are b-cyano tertiary
recently emerged as a biocatalyst that may comple- alcohols, HheC was considered as a biocatalyst for
ment the use of organometallic catalysts.[2] In the for- the preparation of these alcohols in enantiopure form.
ward direction, halohydrin dehalogenases catalyze the Chiral tertiary alcohols are important building blocks
conversion of vicinal halohydrins to the corresponding for various optically active products, but methods for
epoxides, a reaction that is important for the biodeg- their preparation are limited compared to procedures
radation of some halogenated xenobiotic com- that yield the more easily available primary and sec-
pounds.[3] In the reverse of this reaction, the enzymes ondary alcohols.[11] A frequently used biocatalytic
can accept a vaÀriety of alternative anionic nucleo- method for the resolution of racemic alcohols em-
À
philes such as N3 , NO2 , and CNÀ, to open the epox- ploys the enantioselectivity of lipases.[12] However,
ide ring.[4,5] Depending on the nature of the epoxide, most lipases do not accept tertiary alcohols, and con-
these reactions proceed with good rate and high enan- sequently there are only a few examples of their ap-
tioselectivity, providing a route to enantioenriched plication for resolving enantiomers of these com-
products.[6,7] The structure of the halohydrin dehaloge- pounds.[13] Recently, esterase mutants were reported
nase from Agrobacterium radiobacter strain AD1 to catalyze the hydrolysis of several acetates of terti-
(HheC) has been solved.[8] A recent study indicates ary alcohols with excellent enantioselectivites.[14]
that the performance of this enzyme for the synthesis Some enantiospecific epoxide hydrolases are also able
of the important statin side chain building block (R)- to hydrolyze 2,2-disubstituted epoxides, yielding diols
methyl 4-cyano-3-hydroxybutyrate can be improved containing a tert-alcohol moiety.[15]
by directed evolution.[9] Mutants with enhanced and
Here we report the use of halohydrin dehalogenase
in a new biocatalytic approach to optically pure terti-
Adv. Synth. Catal. 2007, 349, 2279 – 2285
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2279

´
Maja Majeric Elenkov et al.
FULL PAPERS
ary alcohols bearing synthetically useful groups at the of the nature of the R1 substituent, which could be an
b-position.
alkyl, aryl, or ester group.
Results and Discussion
Modelling
Production of Tertiary b-Cyanohydrins with HheC
The remarkable effect of the methyl substituent on
enantioselectivity is most striking for epoxide 1b,
To explore the applicability of HheC for the produc- which has a methyl and an ethyl group on the chiral
tion of optically active b-cyano tertiary alcohols, we centre. This is a rare example of highly efficient chiral
studied the conversion of several 2,2-disubstituted ep- discrimination by an enzyme for a substrate that
oxides (1b–5b) and compared it with the conversion bears substituents with similar electronic properties
of monosubstituted epoxides (1a–5a) (Table 1). Each and with only such as small difference in size. To find
substrate was tested for enantioselective ring opening a structural explanation for the positive effect of the
using HheC as the catalyst and cyanide as the nucleo- additional 2-methyl substituent on halohydrin dehalo-
phile. The resolution of monosubstituted epoxides genase enantioselectivity, substrate 1b was modelled
(1a–5a) proceeded with varying degrees of enantiose- in the active site of the X-ray structure of HheC.
lectivity (E=2–106). In all cases, the (R)-enantiomer Since no structure of the apoenzyme is known, the
was preferentially converted, except for 4b, in which HheC model was derived from a structure of the
case there is a change in priority due to the ester sub- enzyme complexed with (R)-para-nitrostyrene oxide.
stituent. The conversion of all substrates with an addi- In order to mimic the productive complex, the epox-
tional 2-methyl substituent (1b–5b) occurred with a ide oxygen and terminal carbon atom of the sub-
much higher enantioselectivity (E>200). In the mono- strates were positioned in such a way that attack of a
substituted series, enantioselectivity was dependent nucleophile bound in the halide binding site would be
on the R1 substituent of the epoxide, and the highest possible. Using these constraints, and avoiding steric
E value was obtained with 2a. In contrast, when the conflicts, the (R)-enantiomer of molecule 1b fits
additional 2-methyl substituent was present (R2), rather tightly into the active site pocket with the
HheC displayed a high E value (>200) independent methyl group pointing into a small bulge bordered by
the amino acids Trp139, Asn176 and Thr134
(Figure 1). A bulkier group at this position would
lead to steric conflict with those amino acids and thus
the (S)-form of 1b cannot bind in a productive
Table 1. Kinetic resolution of 2,2-disubstituted (1b–5b) and
corresponding monosubstituted epoxides (1a–5a) catalyzed manner. The less sterically demanding monosubstitut-
by halohydrin dehalogenase.[a]
ed (S)-1a can be positioned in a way that is close to
the productive conformation of (R)-1b, which can ex-
plain the observation that it is still a weak substrate
and that the enantioselectivity with 1a is low.
In order to verify this interpretation, all (R) and (S)
enantiomers of the epoxides in Table 1 were docked
Substrate
R1
R2
E[b]
Configuration[c]
into the binding pocket of HheC using the program
GLIDE.[17] The docking solutions were analyzed ac-
cording the binding of the epoxide ring (Table 2). A
solution was considered as productive when the epox-
ide ring was placed in the orientation described
above. Other docking solutions were considered non-
productive. In some cases a conformation near the
productive conformation was observed. No docking
position with the epoxide ring in the epoxide binding
site was found for the (R) form of 4b. This epoxide
carries an ester group and it was found to become
docked in a reverse manner with the ester carbonyl
oxygens interacting with Ser132. Although no protein
flexibility was taken into account, the results in
Table 2 are in agreement with the proposed geometric
explanation for the preference of HheC for (R)-epox-
ides.[18]
1a[d]
1b[d]
2a[d]
2b
3a
3b
Et
Et
c-Hex
c-Hex
CH2Ph
CH2Ph
CO2Me
CO2Me
CH2CO2Me
CH2CO2Me
H
Me
H
Me
H
Me
H
Me
H
Me
29
>200
106
>200
2
>200
n.d.[f]
>200
15
R
R
R
R
R
R
-
4a
4b
S[g]
R
R
5a[e]
5b
>200
[a]
General conditions: 5 mM epoxide in 10 mL Tris-SO4,
pH 7.5, 15 mM NaCN, 0.9 or 3.5 mM HheC.
E values were calculated from eep and ees.
Absolute configuration of the faster reacting enantiomer.
Data from Majeric et al.
Data from Majeric et al.
Not determined due to very low activity.
Change in configuration due to the CIP rule.
[b]
[c]
[d]
[e]
[f]
[7]
´
´
[16]
[g]
2280
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2279 – 2285

Preparation of Enantiopure Tertiary Alcohols
FULL PAPERS
Figure 1. Stereo image of the modelling of (R)-1b in the active site of halohydrin dehalogenase. The side chains of the two
catalytic residues Tyr145 and Ser 132 and the side chains of the amino acids restricting the space for the second substituent
are shown. The substrate is shown in green. The red dots show the two water molecules in the proposed binding site of the
nucleophile.
Table 2. Docking poses obtained with the docking program
and regioselectivity, yielding products with ee ꢀ99%,
GLIDE for the molecules in Table 1.[a]
regardless of the nature of the larger 2-substituent
(R1). The conversions proceeded with retention of the
Substrate
a (R2 =H)
b (R2 =Me)
absolute configuration at the chiral oxirane carbon.
Variations in the electronic or steric nature of the R1
substituent thus had little influence on the enantiose-
lectivity and no effect on the stereopreference of the
reactions. However, the reaction rate strongly de-
pended on the nucleophile and with all epoxides the
highest rate was found with azide. Enzyme activity
was also dependent on the substrate and low activity
in some cases may be a disadvantage for the synthetic
utility of the biotransformation. This problem may be
possibly be overcome by developing mutants with in-
creased activity.[9,10]
(R)
(S)
(R)
(S)
1
2
3
4
5
P
P
~
NP
P
~
P
~
~
~
NP
NP
P
NP
NP
~
no pose
~
NP
NP
[a]
P, productive conformation; NP, non-productive confor-
mation; ~, conformation close to productive.
Azide and Cyanide as Nucleophiles
To demonstrate the synthetic usefulness of the de-
To further explore the potential of halohydrin dehalo- scribed biocatalytic transformations, a preparative
genase for producing enantiopure b-substituted tert- scale conversions were carried out (Table 4). In a typi-
alcohols, we tested a series of 2,2-disubstituted epox- cal experiment, 300 mg of 2b was processed with
ides with azide and cyanide. As illustrated in Table 3, 0.5 equivs. of NaN3 and 9 mg of purified halohydrin
HheC catalyzed the formation of a wide range of tert- dehalogenase in 40 mL Tris-SO4 buffer at room tem-
alcohols both with cyanide and with azide ions.
perature. After 2.5 h product (R)-2d was isolated
All these substrates were hardly sensitive towards from the incubation mixture in 41% yield and very
the non-enzyme-catalyzed ring opening reaction with high ee. Similarly, a reaction with NaCN was per-
cyanide under the conditions used. The same was ob- formed. More enzyme and a longer reaction time
served in our previous work.[7] The insignificance of were required to achieve comparable conversion and
the chemical conversion of the 2,2-disubstituted epox- to extract and purify enantiopure (R)-2c in 40% yield
ides to b-cyano alcohols was reflected in the very high (Table 4). Both with NaCN and NaN3 addition of
product ees (ꢀ 99%). Although azide is a stronger 0.5 equivs. was enough to obtain ca. 45% conversion
nucleophile than cyanide and non-enzyme-catalyzed of epoxide, with no reduction of product ee due to
azidolysis was suspected to be faster than with cya- chemical reaction of epoxide with nucleophile. Also
nide, spontaneous azidolysis was also much slower tertiary alcohol 5c could be isolated in highly enan-
than the enzyme-catalyzed reaction and again had no tioenriched form and good yield after extraction and
effect on optical purity of the product. The only ex- chromatography.
ception was the transformation of 4b where due to
the low enzyme activity the spontaneous ring opening
reaction played a significant role, reflected in the Conclusions
lower ee (90%) of product 4d.
All HheC-catalyzed reactions of 2,2-disubstituted Halohydrin dehalogenase was applied for the kinetic
epoxides proceeded with very high enantioselectivity resolution of racemic 2,2-disubstituted epoxides. By
Adv. Synth. Catal. 2007, 349, 2279 – 2285
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2281

´
Maja Majeric Elenkov et al.
FULL PAPERS
Table 3. Kinetic resolution of 2,2-disubstituted epoxides (1b–5b) catalyzed by HheC on an analytical scale.[a]
Substrate
Nu
Product
t [h]
Conversion [%][c]
34
ees [%][c]
eep [%][c]
1b
CN
1c
5
51
99
1b
2b
N3
1d
0.5
5
50
47
>99
>99
>99
CN
2c
88
2b
N3
2d
1.5
45
82
>99
3b
3b
4b
4b
5b
CN
N3
3c
3d
4c
4d
5c
5d
5
30[b]
47
42
89
26
40
87
89
99
>99
>99
90
2
CN
N3
5
21[b]
29[b]
47
5
CN
N3
3
>99
>99
5b
0.5
47
[a]
General conditions: 5 mM epoxide in 10 mL Tris-SO4, pH 7.5, 15 mM NaCN or 7.5 mM NaN3, 0.9 mM HheC.
3.5 mM of HheC.
Determined by GC or HPLC analysis.
[b]
[c]
nucleophilic ring opening, optically pure tertiary alco- Experimental Section
hols were obtained in good yield at low nucleophile
concentration, in aqueous medium and at room tem- General Remarks
perature. These products contain a synthetically
1H and 13C NMR spectra were recorded on a Varian 400
(100 MHz) spectrometer in CDCl3. Chemical shift values
are denoted in d values (ppm) relative to residual solvent
useful cyano or azido group at the b-position and can
give access to a variety of building blocks in enantio-
pure form. The results obtained open a new biocata-
lytic way for the preparation of enantiomerically pure
tertiary alcohols.
1
peaks (CHCl3, H d=7.26, 13C d=77.16). Enzymatic reac-
tions were monitored by GC or HPLC using commercially
available chiral columns. GC analysis was performed on a
Hewlett–Packard 6890 series gas chromatograph equipped
with an FID detector (set at 2258C), a split injector (set at
2258C), using N2 as a carrier gas and an autosampler. HPLC
analysis was performed on a Merck L-6200 A instrument
with UV detection at 254 nm. Column chromatography was
done using silica gel, Merck type 9385 (230–400 mesh). TLC
was performed on 0.25 mm silica gel 60-F plates, Merck.
2282
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2279 – 2285

Preparation of Enantiopure Tertiary Alcohols
FULL PAPERS
Table 4. Preparative-scale kinetic resolutions of epoxides
with cyanide or azide as the nucleophile.
Preparative-Scale Enzymatic Reactions
Pure epoxides 2b or 5b were dissolved in Tris-SO4 buffer,
followed by addition of NaCN or NaN3, and purified
enzyme. The mixture was stirred at room temperature and
monitored by gas chromatography (GC). The reaction was
stopped at conversion <45% and the mixture was extracted
with ethyl acetate. The organic phase was dried over
Na2SO4 and evaporated. The crude product was chromato-
graphed on a silica gel using pentane/CH2Cl2 (4:6) as eluent
yielding pure alcohols. The NMR data were identical with
synthesized racemic reference compounds.
Sub- t [h] Conv.
strate
Epoxide
tert-Alcohol
Yield eep [%]
[%][b]
[%][a] Yield ees [%]
[%][b]
2b
2b
5b
24
2.5 43
45
42
52
41
37
71 (S) 2c
76 (S) 2d
81 (S) 5c
40
41
35
>99 (S)
>99 (R)
>99 (S)
2
[a]
Determined by GC analysis.
Isolated yield.
[b]
(S)-3-Cyclohexyl-3-hydroxybutyronitrile (2c): The reac-
tion was carried out in 25 mL Tris-SO4 buffer (200 mM,
pH 7.5), for 24 h following the general procedure (42% con-
version) using 200 mg (1.42 mmol) of racemic 2b, 35 mg
(0.71 mmol) of NaCN and 9 mg of halohydrin dehalogenase
(HheC in 1.2 mL buffer). After extraction and purification
by chromatography pure (S)-2b (105 mg, 52%, ee 71%) and
(S)-2c (95 mg, 40%, ee>99%) were isolated. 1H NMR
(CDCl3, 400 MHz): d=0.98–1.33 (m, 5H), 1.30 (s, 3H), 1.50
(m, 1H), 1.60–1.83 (m, 6H), 2.52 (dd, J1 =16.5 Hz, J2 =
1.0 Hz, 1H), 2.55 (d, J=16.5 Hz, 1H); 13C NMR (CDCl3,
100 MHz): d=23.9, 26.1, 26.2, 26.3, 26.8, 27.5, 29.6, 47.1,
73.1, 117.9.
(R)-1-Azido-2-cyclohexylpropan-2-ol (2d): The reaction
was carried out for 2.5 h in 40 mL Tris-SO4 buffer (200 mM,
pH 7.5), following the general procedure (43% conversion)
using 300 mg (2.14 mmol) of racemic 2b, 69 mg (1.06 mmol)
NaN3 and 9 mg of halohydrin dehalogenase (HheC in
280 mL buffer). After extraction and purification by chroma-
tography pure (S)-2b was isolated (155 mg, 51%, ee 76%)
and (R)-2d (163 mg, 41%, ee >99%). 1H NMR (CDCl3,
400 MHz): d=0.92–1.29 (m, 5H), 1.11 (s, 3H), 1.46 (tt, J1 =
12.0 Hz, J2 =3.0 Hz, 1H), 1.65–1.86 (m, 6H), 3.25 (d, J=
12.0 Hz, 1H), 3.36 (d, J=12.1 Hz, 1H); 13C NMR (CDCl3,
100 MHz): d=21.8, 26.7, 26.8, 26.9, 26.9, 28.1, 45.8, 59.9,
75.1.
Bulb-to-bulb distillation was performed using a glass oven
B-585 instrument from Büchi.
The commercial grade reagents and solvents were used
without further purification. Racemic 2-methyl-1,2-epoxybu-
tane (1b) was purchased from Acros Organics. The com-
pounds 1,2-epoxybutane (1a), 2,3-epoxypropylbenzene (3a),
methyl 2-methylglycidate (4b) and (R,R)-N,N’-bis(3,5-di-tert-
butylsalicylidene)-1,2-cyclohexanediaminochromiumACHTRE(UNG III)
chloride were supplied by Aldrich. The epoxides 2-cyclohex-
yl-2-methyloxirane (2b) and 2-benzyl-2-methyloxirane (3b)
were prepared from corresponding ketone using trimethyl-
sulfonium methyl sulfate (Aldrich). Epoxide 4a was pre-
pared by oxidation of the methyl acrylate with NaOCl. Ep-
oxide 5a was prepared by reduction of methyl 4-chloroace-
toacetate with NaBH4, followed by ring closure to methyl
3,4-epoxybutyrate as described.[19] Methyl 2-methyloxirane
acetate (5b) was prepared starting from 3-methyl-3-butenol.
In the first step the alcohol was oxidized by Jones reagent,
followed by esterification and finally epoxidation of methyl
3-methyl-3-butenoate by m-CPBA. Racemic alcohols 1c–5c
and 1d–5d were prepared by ring opening reactions of the
corresponding epoxides with sodium cyanide or sodium
azide in water.
(S)-Methyl 4-cyano-3-hydroxy-3-methylbutanoate (5c):
The reaction was carried out in 30 mL Tris-SO4 buffer
(200 mM, pH 7.5), following the general procedure for 2 h
(45% conversion) using 300 mg (2.14 mmol) of racemic 5b
(200 mg, 1.53 mmol), 38 mg (0.77 mmol) of NaCN and
15 mg of halohydrin dehalogenase (HheC in 280 mL buffer).
After extraction and purification by chromatography pure
(S)-5b (75 mg, 37%, ee 81%) and (S)-5c (84 mg, 35%, ee
>99%) were obtained. 1H NMR (CDCl3, 400 MHz): d=
1.43 (s, 3H), 2.63 (d, J=16.5 Hz, 1H), 2.65 (s, 2H), 2.71 (d,
J=16.5 Hz, 1H), 3.75 (s, 3H); 13C NMR (CDCl3, 100 MHz):
d=27.3, 30.9, 43.6, 52.4, 69.5, 117.3, 172.6.
Halohydrin dehalogenase from Agrobacterium radiobacter
AD1 (HheC, wild-type enzyme, acc. no. AAK92099) was
produced in recombinant form in E. coli cultivated in LB
medium. The enzyme was purified from sonicated cells by
column chromatography as described before.[20]
General Procedure for Enzymatic Reactions on
Analytical Scale
Enzymatic reactions were performed at ambient tempera-
ture (228C). Substrate (0.054 mmol, final concentration
5 mM) was added from a stock solution in DMSO (50 mL,
0.5%) to Tris-SO4 buffer (10 mL, 0.2M, pH 7.5), followed
by addition of a stock solution in water of NaCN (0.75 mL,
162 mmol, final concentration 15 mM) or NaN3 (0.75 mL,
80 mmol, final concentration 7.5 mM). Reactions were initi-
ated by addition of enzyme (0.25 or 1.0 mg) to a final con-
centration of 0.9 or 3.5 mM. The progress of the reaction was
followed by periodically taking samples (1 mL) from reac-
tion mixture. These were extracted with diethyl ether
(1 mL) containing an internal standard (1-chlorohexane or
mesitylene), dried over Na2SO4 and analyzed by GC or
HPLC.
Determination of Enantiomeric Purity
Enantiomeric excesses (ee) were determined by chiral GC
analysis on the Chiraldex G-TA column (30 m”0.25 mm”
0.25 mm) (Column I), or HPLC analysis on Chiralpak AS-H
column (25”0.46 cm) (Column II) (Table 5) or as described
earlier.[7] Compounds are defined in Table 1, except 3-hy-
droxy-4-phenylbutyronitrile (3a’) and methyl 4-cyano-3-hy-
droxybutanoate (5a’).
Adv. Synth. Catal. 2007, 349, 2279 – 2285
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2283

´
Maja Majeric Elenkov et al.
FULL PAPERS
Table 5. Chiral analysis of epoxides, cyano alcohols and b-azido alcohols.
Compound
Column[a]
Conditions
Rt [min]
1d
2b
2c
2d
3a
3a’
3b
3c
3d
4a
4b
4c
4d
5a
5a’
5b
5c
5d
I
I
I
I
I
I
I
I
II
I
I
I
I
I
I
I
I
I
408C 5 min, 108CminÀ1 to 1008C, 7 min at 1008C
1008C 8 min, 158CminÀ1 to 1708C, 6 min at 1708C
1008C 8 min, 158CminÀ1 to 1708C, 6 min at 1708C
1008C 10 min, 108CminÀ1 to 1508C, 5 min at 1708C
1008C 13 min, 158CminÀ1 to 1708C, 10 min at 1708C
1008C 13 min, 158CminÀ1 to 1708C, 10 min at 1708C
1008C 5 min, 108CminÀ1 to 1508C, 16 min at 1508C
1008C 5 min, 108CminÀ1 to 1508C, 16 min at 1508C
2-PrOH/heptane (2:98), 1 mLminÀ1
16.1 (R)/16.3 (S)
7.0 (R)/7.6 (S)
17.4 (S)/17.8 (R)
18.5 (R)/18.7 (S)
12.5 (R)/12.8 (S)
24.9 (R)/25.3 (S)
8.6 (R)/8.7 (S)
24.3 (R)/24.6 (S)
11.2 (S)/12.1 (R)
5.1/6.6
4.7 (S)/5.3 (R)
13.9 (S)/14.6 (R)
11.5 (S)/12.1 (R)
6.9 (R)/8.3 (S)
15.0 (S)/15.4 (R)
5.5 (R)/5.9 (S)
12.2 (S)/12.5 (R)
10.7 (R)/11.0 (S)
1008C 8 min, 158CminÀ1 to 1708C, 3 min at 1708C
1008C 8 min, 158CminÀ1 to 1708C, 3 min at 1708C
1008C 8 min, 158CminÀ1 to 1708C, 3 min at 1708C
1008C 8 min, 158CminÀ1 to 1708C
1008C 8 min, 158CminÀ1 to 1708C, 10 min at 1708C
1008C 8 min, 158CminÀ1 to 1708C, 10 min at 1708C
1008C 6 min, 158CminÀ1 to 1708C, 5 min at 1708C
1008C 6 min, 158CminÀ1 to 1708C, 5 min at 1708C
1008C 6 min, 158CminÀ1 to 1508C, 5 min at 1508C
[a]
Column I (Chiraldex G-TA); Column II (Chiralpak AS-H).
References
Determination of Absolute Configuration
Absolute configurations were assigned by chiral GC analysis
using reference compounds. In the case of epoxides 3a and
3b assignment was based on previously reported data.[21]
The enantiomerically enriched epoxides (R)-2b, (S)-4b, (R)-
5a and (R)-5b were prepared by (R,R)-(salen)CrCl catalyzed
ring opening with TMSN3 according to Label and Jacob-
sen.[22]
[1] E. N. Jacobsen, Acc. Chem. Res. 2000, 33, 421–431.
´
[2] D. B. Janssen, M. Majeric Elenkov, G. Hasnaoui, B.
Hauer, J. H. Lutje Spelberg, Biochem. Soc.,Trans.
2006, 34, 291–295.
[3] J. E. van Hylckama Vlieg, L. Tang, J. H. Lutje Spelberg,
T. Smilda, G. J. Poelarends, T. Bosma, A. E. van Mer-
ode, M. W. Fraaije, D. B. Janssen, J. Bacteriol. 2001, 83,
5058–5066.
[4] J. H. Lutje Spelberg, L. Tang, M. van Gelder, R. M.
Kellogg, D. B. Janssen, Tetrahedron: Asymmetry 2002,
13, 1083–1089.
[5] a) T. Nakamura, T. Nagasawa, F. Yu, I. Watanabe, H.
Yamada, Biochem. Biophys. Res. Commun. 1991, 180,
124–130; b) T. Nakamura, T. Nagasawa, F. Yu, I. Wata-
nabe, H. Yamada, Tetrahedron 1994, 50, 11821–11826.
[6] J. H. Lutje Spelberg, J. E. T. van Hylckama Vlieg, L.
Tang, D. B. Janssen, R. M. Kellogg, Org. Lett. 2001, 3,
41–43.
Calculation of Enantioselectivity
E values were calculated from eep and ees according to for-
mula E=ln
A
ACHTRE[UNG (1+ees)/(1+ees/eep)].
Molecular Modelling
For modelling of substrate binding to HheC, we used the
enzyme complexed with (R)-para-nitrostyrene epoxide
(PDB code: 1ZMT) into which molecule 1b was manually
positioned using the program MAESTRO 7.5 (Schrçdinger,
LLC, New York) after removal of the substrate. The oxygen
of the epoxide ring was positioned within hydrogen bonding
distance of the OH groups of the catalytic residues Tyr145
and Ser132. The terminal carbon atom was placed in the di-
rection of the water molecule that is present in the halide-
binding site.
Docking with the (R) and (S) enantiomers of the mole-
cules listed in Table 1 in the active site of HheC was done
with the program GLIDE 4.0.[17] The molecules were gener-
ated with the Built panel of MAESTRO 7.5 and prepared
with Schrçdingerꢁs program LigPrep according to the user
manual. The protein structure was optimized with the
“pprep” procedure using default values. The docking with
GLIDE was performed in “xp” mode.
´
[7] M. Majeric Elenkov, B. Hauer, D. B. Janssen, Adv.
Synth. Catal. 2006, 348, 579–585.
[8] R. M. de Jong, J. J. W. Tiesinga, H. J. Rozeboom, K. H.
Kalk, L. Tang, D. B. Janssen, B. W. Dijkstra, EMBO J.
2003, 22, 4933–4944.
[9] R. J. Fox, S. C. Davis, E. C. Mundorff, L. M. Newman,
V. Gavrilovic, S. K. Ma, L. M. Chung, C. Ching, S. Tam,
S. Muley, J. Grate, J. Gruber, J. C. Whitman, R. A.
Sheldon, G. W. Huisman, Nat Biotechnol. 2007, 25,
338–344.
[10] L. Tang, D. E. Torres Pazmino, M. W. Fraaije, R. M.
de Jong, B. W. Dijkstra, D. B. Janssen, Biochemistry
2005, 44, 6609–6618.
[11] C. Garcia, V. S. Martin, Curr. Org. Chem. 2006, 10,
1849–1889.
[12] A. Ghanem, H. Y. Aboul-Enein, Chirality 2005, 17, 1–
15.
2284
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2279 – 2285

Preparation of Enantiopure Tertiary Alcohols
FULL PAPERS
[13] a) S. H. Krishna, M. Persson, U. T. Bornscheuer, Tetra-
hedron: Asymmetry 2002, 13, 2693–2696; b) S.-T. Chen,
J.-M. Fang, J. Org. Chem. 1997, 62, 4349–4357; c) D.
OꢁHagan, N. A. Zaidi, R. B. Lamont, Tetrahedron:
Asymmetry 1993, 4, 1703–1708.
[17] R. A. Friesner, J. L. Banks, R. B. Murphy, T. A. Halg-
ren, J. J. Klicic, D. T. Mainz, M. P. Repasky, E. H.
Knoll, M. Shelley, J. K. Perry, D. E. Shaw, P. Francis,
P. S. Shenkin, J. Med. Chem. 2004, 47, 1739–1749.
[18] R. M. de Jong, J. J. W. Tiesinga, A. Villa, L. Tang, D. B.
Janssen, B. W. Dijkstra, J. Am. Chem. Soc. 2005, 127,
13338–13343.
[14] R. Kourist, S. Bartsch, U. T. Bornscheuer, Adv. Synth.
Catal. 2007, 349, 1393–1398.
[15] a) I. Osprian, W. Stampfer, K. Faber, J. Chem. Soc.,
Perkin. Trans. 1, 2000, 3779–3785; b) I. Osprian, W.
Kroutil, M. Mischitz, K. Faber, Tetrahedron: Asymme-
try 1997, 8, 65–71; c) U. Wandel, M. Mischitz, W. Krou-
til, K. Faber, J. Chem. Soc.,Perkin. Trans. 1 , 1995, 735–
736.
[19] J. D. McClure, J. Org. Chem. 1967, 32, 3888–3894.
[20] L. Tang, A. E. van Merode, J. H. Lutje Spelberg, M. W.
Fraaije, D. B. Janssen, Biochemistry 2003, 42, 14057–
14065.
[21] A. F. Dexter, F. J. Lakner, R. A. Campbell, L. P. Hager,
J. Am. Chem. Soc. 1995, 117, 6412–6413.
´
[16] M. Majeric Elenkov, L. Tang, B. Hauer, D. B. Janssen,
[22] H. Label, E. N. Jacobsen, Tetrahedron Lett. 1999, 40,
7303–7306.
Org. Lett. 2006, 8, 4227–4229.
Adv. Synth. Catal. 2007, 349, 2279 – 2285
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2285