Full Papers
tion numbers (ECNs) were lower than those estimated in the
first set of reactions. The optimized space-time yields, the
product yields per unit amount of enzyme, and the straightfor-
ward reaction make this process design attractive for scaling
up, which has already been successfully demonstrated for the
preparative resolution of (+)-cis/trans-limonene oxides cata-
lyzed by Re-LEH.
with calibration curves obtained with authentic substrate/product
standards (2.5–20 mmolL ). One unit of activity (U) is defined as
À1
the enzyme activity that hydrolyzes 1 mmol of substrate per min
under the assay conditions described above. Stereochemical con-
figuration was determined based on commercially available
[26]
authentic standards of 1 and 2 and on Re-LEH reference data.
NMR spectroscopy: NMR spectra were measured in D O at RT on
2
a Bruker Avance III 600 MHz spectrometer equipped with a Broad-
band Observe probe head with z gradient at 600.2 MHz for
1
13
Conclusions
H NMR analysis and 150.9 MHz for C NMR analysis.
The epoxide hydrolase catalyzed resolution of mixtures of
(
+)-cis/trans-limonene oxide and (À)-cis/trans-limonene oxide
Culture conditions, protein expression, and purification
represents a straightforward, economical, and sustainable
route to all limonene oxide enantiomers. The selected toolbox
of recombinant epoxide hydrolases has turned out to be supe-
rior to the best direct separation methods and chemical routes
for the synthesis of all limonene oxide enantiomers. In addi-
tion, the (1S,2S,4R)-limonene-1,2-diol and the (1R,2R,4S)-limo-
nene-1,2-diol are easily accessible. The solvent-free process
design and the achieved product yields make this highly re-
source-efficient one-step reaction not only attractive for large-
scale production but also for further resolutions of cis/trans-ep-
oxide mixtures to obtain valuable enantiopure epoxides and
Cultures of E. coli 10 G harboring plasmids pRhamReLEH, pRham-
[29]
TomskLEH, and pRhamCH55 LEH, respectively, were cultivated in
À1
Luria–Bertani (LB) medium containing 30 mgmL kanamycin, at
T=378C and with shaking at 220 rpm overnight. These cultures
were subsequently used for the inoculation of LB medium
À1
(
500 mL) supplemented with 30 mgmL kanamycin. Cells were
grown at T=378C and with shaking at 220 rpm until they reached
an optical density at l=600 nm of 0.6–0.8; they were then in-
duced by adding rhamnose (0.2% (w/v) final volume) and kept for
2
4 h at T=308C with shaking at 220 rpm. The cells were harvested
by centrifugation (3000 g for 30 min at T=48C), resuspended in
wash buffer (10 mL, 20 mm KPi (potassium phosphate) buffer,
pH 7.0, 500 mm NaCl, 20 mm imidazole), and disrupted by sonica-
tion. Protein purification was performed by Ni-NTA (GE Healthcare)
1
,2-diols.
[29]
chromatography as previously described.
Experimental Section
General
Preparative resolutions under non-optimized conditions
reactions A–D)
(
+)-cis/trans-Limonene oxide (purity 97%, ee 98%) and (À)-cis/
(
trans-limonene oxide (purity 99%, ee 99%) were from Sigma–Al-
drich (catalog numbers 218324 and 218332, respectively).
Preparative-scale resolutions under non-optimized conditions were
performed by adding a 5m solution in CH CN (2 mL) of (+)-limo-
nene oxide (59:41 mixture of 1 and 2) or (À)-limonene oxide
(55:45 mixture of 4 and 5) (1.522 g, 10 mmol) to 25 mm KP buffer
Thin-layer chromatography (TLC) analyses were performed on pre-
coated silica gel 60 F254 plates (Merck) and treated with molybdate
reagent ((NH ) Mo O ·4H O (42 g), Ce(SO ) (2 g), concentrated
3
4
6
7
24
2
4
i
H SO (62 mL), made up to 1 L volume with deionized water).
(pH 8.0, 18 mL) containing the purified LEHs (4 mg of Re-LEH for re-
action A, 30 mg of Tomsk-LEH and CH55-LEH for reactions B and C,
respectively, and 8 mg of Re-LEH for reaction D). Reactions were
performed in plastic tubes (50 mL) and incubated at T=208C with
shaking (180 rpm). GC monitoring and product recovery and purifi-
cation were performed as described above. Resolved epoxide
yields and diol yields are determined according to the total
amount of the initial cis/trans-limonene oxide mixture.
2
4
Product recovery was performed by extraction of the reaction mix-
tures with AcOEt (3 volumes) and purification by flash chromatog-
raphy (silica, petroleum ether/AcOEt, 95:5) on silica gel LC60A (40–
6
3 mesh, Grace).
Unless otherwise stated, all chemicals were of analytical grade and
were purchased from Sigma–Aldrich.
Reaction A: Epoxide 2 was recovered as a transparent oil (0.678 g,
1
13
4
5%) and analyzed by H, C, and HSQC NMR spectroscopy (see
Analytical methods
Figure S3 in the Supporting Information). Diol 3 was recovered as
1
13
GC analysis: GC analyses were performed on a AGILENT 6850 (Net-
work GC System) gas chromatograph equipped with a chiral capil-
lary column (MEGA DEX DAC-BETA, Legnano, Italy; 0.25 mm diame-
ter, 25 m length, and 0.25 mm thickness) and a flame ionization de-
tector. At scheduled time points, reaction samples (100 mL) were
extracted with an equal volume of a 0.025 mgmL benzophenone
solution in AcOEt in the presence of saturated NaCl and injected
into the GC system. The column temperature was initially raised
a transparent oil (0.847 g, 49%) and analyzed by H, C, and
HSQC NMR spectroscopy (see Figure S4 in the Supporting Informa-
tion).
Reaction B: Epoxide 1 was recovered as a transparent oil (0.583 g,
1
13
3
8%) and analyzed by H, C, and HSQC NMR spectroscopy (see
À1
Figure S5 in the Supporting Information). Diol 3 was recovered as
1
13
a transparent oil (0.911 g, 53%) and analyzed by H, C, and
HSQC NMR spectroscopy (see Figure S4 in the Supporting Informa-
tion).
À1
from 80 to 110 8C at a rate of 28Cmin and then was raised from
À1
À1
1
10 to 2008C at a rate of 108Cmin at 2 mLmin flow rate. Re-
tention times were: 1: 9.07 min; 2: 10.66 min; 3: 21.44 min; 4:
.63 min; 5: 10.13 min; 6: 21.28 min; internal standard benzophe-
Reaction C: Epoxide 5 was recovered as transparent oil (0.503 g,
1
13
9
33%) and analyzed by H, C, and HSQC NMR spectroscopy (see
none: 22.45 min. The substrate and product peak areas were nor-
malized to benzophenone, and concentrations were calculated
Figure S6 in the Supporting Information). Diol 6 was recovered as
a transparent oil (1.148 g, 67%) and analyzed by H, C, and
1
13
ChemCatChem 2015, 7, 3171 – 3178
3176
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim