Desymmetrization of cbz-serinol catalyzed by crude pig pancreatic lipase reveals action of lipases with opposite enantioselectivity

Article abstract of DOI:10.1016/j.molcatb.2012.09.007

Lipase preparations from pig pancreas are widely used in industrial production of fine chemicals and pharmaceuticals. In our attempts to synthesize enantiopure building blocks for production of iminocyclitols, we used a crude preparation of pig pancreatic

Full text of DOI:10.1016/j.molcatb.2012.09.007

Journal of Molecular Catalysis B: Enzymatic 85–86 (2013) 134–139
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Desymmetrization of cbz-serinol catalyzed by crude pig pancreatic lipase reveals
action of lipases with opposite enantioselectivity
Elisabeth Egholm Jacobsena,∗, Aleksander Liea, Marte Marie Hansen Frigstada,
Mohammed Farrag el-Behairyb, Torbjørn Ljonesa, Roland Wohlgemuthc, Thorleif Anthonsena
a Norwegian University of Science and Technology, Department of Chemistry, NO-7491 Trondheim, Norway
b Medicinal and Pharmaceutical Chemistry Department, Pharmaceutical and Drug Industries Research Division, National Research Centre, Dokki, Giza 12622, Egypt
c Sigma-Aldrich, Industriestrasse 25, CH-9470 Buchs, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 28 March 2012
Received in revised form 23 August 2012
Accepted 12 September 2012
Available online 19 September 2012
Lipase preparations from pig pancreas are widely used in industrial production of fine chemicals
and pharmaceuticals. In our attempts to synthesize enantiopure building blocks for production of
iminocyclitols, we used a crude preparation of pig pancreatic lipase in acetylation of carboxybenzyl-
2-amino-1,3-propanediol (cbz-serinol) with vinyl acetate in tetrahydrofuran. The enantiomeric excess
of the product monoester changed from (2S)-monoacetate to (2R)-monoacetate after prolonged reaction.
By using vinyl acetate both as solvent and acyl donor only the (2R)-monoacetate was obtained. Purified
enzyme preparation gave only the (2S)-monoacetate. It is likely that the crude enzyme preparation con-
tains lipases with opposite stereoselectivity, acting at different rates in the desymmetrization reaction.
The same enzyme preparation was used in hydrolysis of the diacetylated propanediol in phosphate buffer
and acetonitrile as co-solvent. After 24 h the (2S)-monoacetate was obtained in high enantiomeric excess.
Our results show that monitoring the progress of biocatalyzed reactions are of utmost importance, in
particular when crude enzyme preparations are used as catalysts.
Keywords:
Biotransformations
Carboxybenzyl-2-amino-1,3-propanediol
Crude PPL
Pure PPL
(R)- and (S)-selective lipases
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
in transesterification of a series of secondary alcohols, while the
E-value increased by increasing conversion in hydrolysis of the cor-
Hydrolytic enzymes have for many years been used with great
success to obtain enantiopure compounds, both in hydrolysis of
esters and transesterification reactions in organic media [1,2].
Kinetic resolution of racemic mixtures is a very common method,
but desymmetrization of a meso or prochiral substrate is also widely
used. During kinetic resolution, the enantiomeric excess of the
product fraction (eep) decreases as the reaction progresses, but
the enantiomeric excess of the remaining substrate (ees) increases
and will always reach 100% at some conversion higher than 50%.
The important parameter for a kinetic resolution process is the E-
value, which for practical purposes is the ratio of the rate constants
for reaction with the two enantiomeric substrates. For desym-
metrization the important parameter is the enantiomeric excess
(ee) of the product. The E-value in kinetic resolutions and the
ee in desymmetrizations are parameters considered to be con-
stant throughout the reaction [1]. However, we have previously
shown that in kinetic resolutions catalyzed by lipase B from Candida
antarctica (CALB) the E-value decreased by decreasing conversion
responding butanoates. This shows that monitoring the reactions
is important in all biocatalyzed reactions. We found that the chang-
ing concentrations of the (R)-alcohols in the medium were causing
this effect by a possible allosteric inhibition (activation) of the lipase
[3,4]. We have also shown that acetaldehyde from the acyl donor
was not causing any effect on CALB [5].
Lipase preparations from pig (porcine) pancreas (PPL) are used
in industrial production of fine chemicals and pharmaceuticals
(DSM, Lonza, Sigma–Aldrich, etc.), f. inst. glycidol and cyclo-
propanediol [6,7]. Enzyme extracts from pig pancreas (Creon®,
Solvay Pharmaceuticals, Switzerland) have also been used for many
years as aid for digestion of patients suffering from pancreas dys-
functions and malnutrition, among them 70,000 Cystic Fibrosis
patients worldwide (http://www.cff.org/AboutCF).
Lipase preparations of animal origin tend to consist of a collec-
tion of enzymes which catalyze fat hydrolysis. Although the crude
preparation PPL Type II from Sigma–Aldrich has been purified,
protease and amylase activity are still observed. Monitoring the
progress of crude PPL catalyzed reactions is therefore important in
order to verify the action of competing enzymes. Fractionation of
crude PPL Type II from Sigma–Aldrich has been performed [7–9].
Three lipases were identified as 25-kDa, 33-kDa and the most
∗
Corresponding author. Tel.: +47 98843559; fax: +47 73550877.
E-mail address: elisabeth.e.jacobsen@ntnu.no (E.E. Jacobsen).
1381-1177/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcatb.2012.09.007

E.E. Jacobsen et al. / Journal of Molecular Catalysis B: Enzymatic 85–86 (2013) 134–139
135
esterase) was purchased from Böhringer Mannheim, Lot 10315523-
22, activity not given.
OH
H
k1
XHN
k3
OAc
OH
OH
OAc
OAc
(2S)-2
2.2. Chemicals
H
H
XHN
+
XHN
k2
k4
Analytical grade chemicals were purchased from Sigma–
Aldrich. Solvents were from VWR International. Purification and
separation of products were performed using VersaFlashTM from
Supelco, Sigma–Aldrich, column VersaPakTM (silica, spherical
23 × 110 mm, 23 g), Cat. No. 97758-U, max. pressure 60 psi, eluent
pentane:EtOAc, 4:1. Except for entry C (see Table 1) the THF used in
enzymatic reactions was dried over molecular sieve. The dry THF
used for reaction C was distilled under N2 from Na/benzophenone.
OAc
OH
H
1
3
XHN
X = Cbz
(2R)-2
Scheme 1. Acetylation of prochiral carboxybenzyl-2-amino-1,3-propanediol (1)
with vinyl acetate catalyzed by pig pancreatic lipase (PPL).
2.3. Chromatographic analyses
abundant PPL (52-kDa) [9]. These fractions have been compared in
the resolution in aqueous media of different chiral compounds that
are precursors of drugs [10]. In immobilized forms, the PPL enzyme
fractions have been tested as catalysts for kinetic resolution of sec-
ondary substrates, f. inst. ethyl 2-hydroxy-4-phenylbutanoate [9].
The (2S)-enantiomer is the faster reacting enantiomer in hydrolysis
reactions catalyzed by all three enzymes. However, the E-values are
low (7–10), indicating that this is not a good substrate for the lipases
in crude PPL. Both (R)- and (S)-monoesters have been made from
PPL catalyzed hydrolysis of different prochiral diesters, however,
the configuration will depend of the substrate structure [11,12].
Based on the actual kinetic mechanism Van Tol et al. have made a
kinetic model system for kinetic resolutions catalyzed by crude PPL
[13]. In inhibition experiments of PPL Sigma Type VI-S (purified
enzyme) a proteinaceous inhibitor from wheat flour caused non-
competitive inhibition. This indicates that this enzyme may have a
mechanism of action different from that of the reported soybean-
derived lipase inhibitors which caused competitive inhibition [14].
Enantiopure derivatives of 1,3-propanediols are used as precur-
sors for bioactive compounds such as iminocyclitols (glycosidase-
inhibitors), phospholipids, platelet activating factor and rennin
inhibitors [15]. Desymmetrization of prochiral 1,3-propanediols
and diacetates in the presence of lipases has become a practi-
cal approach for preparation of chiral compounds due to high
specificity and reproducibility [16]. Monoacetates are in principle
formed at unequal rates, and also react further at unequal rates
in a double step process (see Scheme 1). The first step is desym-
metrization, the second, kinetic resolution. It is anticipated that if
k1 > k2 then k4 > k3, and moreover, that the ratio k1/k2 is constant
throughout the reaction [1].
Chiral HPLC analyses in entry C were performed with two
instruments; a Varian HPLC coupled to a Varian 2550 UV detector
with a Varian Prostar 4000 autosampler and an Agilent 1100 Series:
G1379A Degasser, G1311A QuatPump with G1313A coupled to
a Bruker UV detector and an ALS autoinjector. Column Daicel
Chiralcel OD-R (i.d. 4.6 mm, 25 cm, particle size 10 ␮m). Eluent
A: H2O (40%) + CH3CN (59.9%) + TFA (0.1%). Eluent B: H2O + TFA
(0.1%). Gradient elution: 65% A + 35% B to 95% A + 5% B over
25.72 min (flow 1.0 mL min−1). tR 1: 5.2 (5.4) min, tR (2S)-2: 7.8
(8.1) min, tR (2R)-2: 10.9 (11.9) min, and tR 3 (diacetate) 20.8 min.
Chromatograms were recorded and analyzed using Varian Star
Chromatography Workstation 6.0 software (chromatograms in
Fig. 1). Reaction mixtures from entries A and C–G were analyzed
by GLC using a Varian 3380 Gas Chromatograph with a Varian
CP-8410 autoinjector and Varian Star Chromatography Worksta-
tion 6.0 software, CP-Chirasil DEX column (i.d. 0.32 mm, 25◦m,
film density 0.25 ␮m), carrier gas H2 (5.0), injection temp. 250 C,
FID temp. 270 ◦C, column pressure 7.5 psi, split flow 60 mL min−1
.
Temp. ◦ progr. 80–92 ◦C/30 ◦C min−1 92–98 ◦C/3 ◦C min−1(2),
,
98–104 C/3 ◦C min−1
,
104–140 ◦C/30 ◦C min−1 140–200 ◦C/
,
3 ◦C min−1. tR diol 1: 11.8 min, monoacetates: tR (2S)-2 17.3 min,
tR (2R)-2: 18.8 min, monobutanoates: tR (2S)-4:17.4 min, tR (2R)-4:
18.7 min and tR diacetate 3: 33.1 min. RS > 1.5 for monoester enan-
tiomers in all chromatographic analyses, (2R)-2 was identified in
accordance with Choi and Borch [17]. Thin layer chromatography
was performed using Merck Silica Gel 60 F254 TLC plates using UV
light at 312 nm.
2.4. Immobilization of crude PPL
2. Experimental
Powdered polypropylene (Accurel MP1000, 1.5 g) was added to
EtOH (96%, 3 mL). PPL (L 3126, Type II (crude) Steapsin, 1.5 g) was
suspended in phosphate buffer (20 mM, 30 mL). The two solutions
were mixed and stirred for 22.5 h at RT. The immobilized enzyme
was filtered off, washed with dist. H2O and lyophilized for 16 h.
2.1. Enzyme catalysts
Preparations of pig pancreatic lipase (PPL, EC 3.1.1.3 and CAS
9001-62-1) were purchased from Sigma–Aldrich: crude PPL L3126
Type II (Steapsin), Lot 096K0747 and Lot 096K0721. Analyses (by
Sigma–Aldrich, Buchs, Switzerland) of Lot 096K0747 show 38%
protein, 358 U/mg protein (olive oil, 30 min incubation)/50 U/mg
protein (triacetin), amylase activity: 147 U/mg, general protease
2.5. Enzymatic acetylation acetylation of 1
See Table 1, entries A–G for amounts. Substrate
1
activity: 2.5 × 10−4 U/mg, ␣-chymotrypsin activity: 7.2 × 10−2
.
(0.18–1.8 mmol) and vinyl acetate or vinyl butanoate
(20 ␮L–0.72 mL) were mixed in solvent, (2–10 mL). The reac-
tions were initiated by addition of enzyme (0.0012–0.252 g). The
reactions were stirred at 200 rpm under N2 at RT in a closed
flask. Samples (50 ␮L) were withdrawn at various intervals before
halting the reactions by filtering off the enzyme catalyst with
addition of EtOAc (entries A and B) or CH2Cl2 (entries C–G). All
reactions were performed in duplicate. In entry B the reaction
mixtures were analyzed by chiral HPLC, in entries A and C–G by
Pure PPL L0382, Type VI-S, lyophilized powder, ≥20,000 U/mg, 50%
protein. Immobilized Lipase B from Candida antarctica (Novozym
435, 10 U/mg) was a gift from Novozymes AS, (Bagsværd, Denmark).
Unit definition: One unit will hydrolyze 1.0 ␮equiv. of fatty acid
from olive oil in 1 h at pH 7.7 at 37 ◦C. Trypsin (Type II crude) from
pig pancreas T-8128, was purchased from Sigma–Aldrich, Lot 96F-
06891. Trypsin activity: 1740 U/mg (BAEE units) and chymotrypsin
activity: 1370 U/mg (ATEE units). Esterase from hog liver (pig liver

Table 1
Enzymatic esterification of 1, entries A–G. Hydrolysis of diacetate (3) entry H. Entries B and H were analyzed by chiral HPLC, entries A and C–G by chiral GLC. Duplicates not shown. Duplicates with PPL Type II Lot 096K0721 gave
similar results as in entry D. Yield in % are shown for isolated products.
Entry
Substr./amount
Enzyme
Enz. amount
Solvent
VA: vinyl acetate
VB: vinyl butanoate
Rx. time (h)
3
Monoester in
excess at rx. end
Yield %
AFigs. 2 and 3
1/0.0811 g,
0.360 mmol
Untreated PPL L3126,
Sigma Type II Steapsin
Lot 096K0747
180 mg/mmol
substr., 0.0654 g
THF, 2 mL
VA: 35 ␮L,
0.376 mmol (1.0×
exc.)
(2R)-2, 70% ee
(2R)-2, 98% ee
(2R)-2, 98% ee
358 U/mg
B
C
Fig. 1
1/0.4086 g,
1.814 mmol
Immob. PPL L3126,
Sigma Type II Steapsin
Lot 096K0747
140 mg/mmol
substr., 0.252 g
THF, 10 mL
VA: 0.72 mL,
7.73 mmol (4.3×
exc.)
9
358 U/mg
1/0.0563 g,
0.250 mmol
Untreated PPL L3126,
Sigma Type II Steapsin
Lot 096K0747
300 mg/mmol
substr., 0.0755 g
Vinyl acetate,
5 mL,
53.7 mmol
1.5
62
[␣]20 (EtOAc, c
D
0.797): −7.4. [␣]20
(CHCl3, c 1.0):
+1.84
D
358 U/mg
D
E
F
1/0.0407 g,
0.181 mmol
Untreated PPL L3126,
Sigma Type II Steapsin
Lot 096K0747
360 mg/mmol
substr., 0.0651 g
THF, 1 mL
THF, 2 mL
THF, 2 mL
VA: 20 ␮L,
0.215 mmol (1.2×
exc.)
17
6
(2R)-2, 98% ee
358 U/mg
1/0.0816 g,
0.362 mmol
Untreated Sigma Type
VI-S, lyophilized
powder
33 mg (pure
enz)/mmol substr.,
0.0012 g
VA: 40 ␮L,
0.432 mmol (1.2×
exc.)
(2S)-2, 100% ee
(2S)-4, 100% ee
20,000 U/mg
1/0.0816 g,
0.362 mmol
Untreated crude
trypsin/chymotrypsin
from pig pancreas
Sigma Type II Lot
96F-06891
180 mg/mmol
substr., 0.065 g
VB: 40 ␮L,
0.432 mmol
19
Trypsin activity:
1740 U/mg (BAEE
units) and
chymotrypsin activity:
1370 U/mg (ATEE
units)
G
H
1/0.0816 g,
0.362 mmol
Untreated pig liver
esterase Böhringer
Mannheim Lot
10315523-22
Untreated PPL L3126,
Sigma Type II Steapsin
Lot 096K0747
180 mg/mmol
substr., 0.065 g
THF, 2 mL
VB: 40 ␮L,
0.432 mmol
19
24
(2S)-4, 100% ee
(2S)-2, 91% ee
3/0.830 g,
3.69 mmol
Substrate
1020 mg/mmol
substr., 2.8 g
Phosphate-
buffer,
CH3CN
60
[␣]20 (CHCl3, c
D
1.0): −1.82
358 U/mg

E.E. Jacobsen et al. / Journal of Molecular Catalysis B: Enzymatic 85–86 (2013) 134–139
137
Fig. 1. Left: HPLC chromatogram of acetylation of 1 with crude PPL and vinyl acetate in THF after 30 min, see Table 1, entry B, tR 1: 5.2 min, tR (2S)-2: 7.8 min, tR (2R)-2:
10.9 min. Right: HPLC chromatogram of the same reaction after 3 days tR 1: 5.4 min, tR (2S)-2: 8.1 min, tR (2R)-2: 11.9 min and tR 3: 20.8 min. After 9 days the ee of (2R)-2 was
98%.
chiral GLC. The HPLC samples were diluted with H2O:CH3CN, 1:1
before injection.
2.7.2. N-Benzyloxycarbonyl-2-aminopropane-1,3-diyl diacetate
(3)
The diacetate 3 was synthesized by reacting 1 (0.9013 g) in
vinyl acetate (5 mL) with lipase B from Candida antarctica (CALB,
Novozym 435) (0.0401 g, 10 mg/mmol substrate) for 23 h. The
diester was purified by flash chromatography using EtOAc:hexane,
2:1 to 100% purity (GLC) in 90% yield. Mp 47.1–49.1 ◦C. 1H NMR
(300 MHz, CDCl3): ı 2.07 (s, 6H), 4.07–4.28 (m, 5H), 5.10 (s, 2H), 5.2
(br s, 1H), 7.31–7.43 (m, 5H). TLC: Rf 0.07 (EtOAc:hexane, 2:1).
2.6. Enzymatic hydrolysis of 3
See Table 1, entry H. The anion exchange material Macro-Prep
High Q support was filtered with dist. H2O (20 mL), phosphate
buffer (50 mL, 2 mM, pH 7) then washed with dist. H2O until
the filtrate reached pH 6.7. Diacetylated cbz-serinol (3) (0.830 g,
3.69 mmol) was dissolved in CH3CN (110 mL) and dist H2O
(100 mL). Then crude PPL (2.8 g, 1020 mg enzyme/mmol substrate)
and the anion exchange material (100 mg) were added. The reac-
tion mixture was stirred on water bath (25 ◦C) for 24 h. The enzyme
and anion exchange material was filtered off and the solvent was
removed under reduced pressure. EtOAc (100 mL) was added and
the reaction mixture was extracted with H2O/NaCl-solution (10/90,
2 × 50 mL), then dried over Na2SO4 before the solvent was removed
under reduced pressure. The product ((2S)-2) was adsorbed on sil-
ica before flash chromatography.
2.7.3. (2R)-N-Benzyloxycarbonyl-2-amino-3-hydroxypropyl
acetate ((2R)-2)
1H NMR (300 MHz, CDCl3): ı 2.08 (s, 3H), 2.39 (br s, 1H), 3.69 (m,
2H), 3.98 (m, 1H), 4.23 (d, 2H), 5.13 (s, 2H), 5.24 (br s, 1H), 7.32–7.43
(m, 5H). TLC: Rf 0.4 (EtOAc:hexane, 2:1), ee 98% (GLC), 60% yield
[␣]20D = −7.4 (EtOAc, c 0.797), [␣]20D = +1.84 (CHCl3, c 1.0).
2.7.4. (2S)-N-Benzyloxycarbonyl-2-amino-3-hydroxypropyl
acetate ((2S)-2)
1H NMR (300 MHz, CDCl3): ı 2.08 (s, 3H), 2.39 (br s, 1H), 3.69 (m,
2H), 3.98 (m, 1H), 4.23 (d, 2H), 5.13 (s, 2H), 5.24 (br s, 1H), 7.32–7.43
(m, 5H). TLC: Rf 0.4 (EtOAc:hexane, 2:1), ee 91% (GLC), 62% yield,
[␣]20D = −1.82 (CHCl3, c 1.0).
2.7. Characterization of compounds
Optical rotations were determined using a Perkin-Elmer 243 B
instrument (Wales, UK), concentrations in g/100 mL. 1H and 13C
NMR spectra (Bruker Avance DPX instruments 300/75 MHz and
400/100 MHz) were obtained from solutions of CDCl3 or DMSO-d6,
chemical shifts are in parts per million and rel. to TMS. Coupling
constants are in Hertz (Hz). The spectra were processed using the
software XWIN-NMR 3.5 from Bruker Biospin. Mass spectra were
recorded using a MAT 95 XL (ThermoQuest Finnigan, San Jose, CA,
USA) with EI probe (DIP). Melting points were obtained from a cap-
illary tube melting point apparatus from Sanyo Gallenkamp (USA).
Characterizations of compounds were in accordance with earlier
reported data [17–19].
3. Results and discussion
Acetylation
of
prochiral
carboxybenzyl-2-amino-1,3-
propanediol (cbz-serinol, 1) with vinyl acetate as acyl donor
catalyzed by different preparations of Pig pancreatic lipase in
organic solvent was monitored by chiral HPLC/GLC as the reaction
progressed (Scheme 1 and Fig. 1). Hydrolysis of diacetate 3 in
phosphate buffer with acetonitrile as co-solvent and crude PPL
gave the (2S)-monoacetate in 91% ee after 24 h (entry H in Table 1,
reaction not shown in Scheme 1).
Composition of the reaction mixture during the first 3 h of acety-
lation of 1 is shown in Fig. 2 (Table 1, entry A). Both monoacetate
enantiomers were formed, but virtually no diacetate. However, a
small amount of diacetate (3) was observed after prolonged reac-
tion time (3 days) (entry B).
2.7.1. N-Benzyloxycarbonyl-2-aminopropane-1,3-diol (1)
Mp 110.3–112.2 ◦C. 1H NMR (300 MHz, CDCl3): ı 2.33 (br s, 2H,
OH) 3.74–3.95 (m, 5H), 5.14 (s, 2H), 5.50 (br s, 1H), 7.32–7.55 (m,
5H). 13C (100 MHz, DMSO-d6): ı 55.4 (1C, CH), 60.9 (1C, CH2) 65.5
The enantiomeric excess plot in Fig. 3 (entry A) shows that the
(2S)-monoacetate (2S)-2 is in high excess in the early stage of the
reaction. At 85 min of conversion, the enantiomeric excess drops to
zero. During further reaction the ee increases again, but now the
(1C, CH2), 128.1 (3C), 128.7 (2C), 138.5 (1C), 156.3 (1C,
C ). TLC:
Rf 0.8 (EtOAc: Hexane, 2:1). MS (EI) (m/z) M•+ 224.8, 193.9, 107.9,
90.9 (100%), 78.9, 64.8, 80.8, 42.9.

138
E.E. Jacobsen et al. / Journal of Molecular Catalysis B: Enzymatic 85–86 (2013) 134–139
100
80
60
40
20
0
that vinyl acetate acts as an inhibitor of the (S)-selective enzyme.
Both (2S)-2 and (2R)-2 were produced when vinyl acetate was used
in equimolar amounts (entry A), or 1.2 times excess (entries D and
E) and 4.3 times excess (entry B). Another batch of crude PPL (Lot
096K0721) was also used in desymmetrization reactions of 1 in
THF both with vinyl acetate and vinyl butanoate, showing similar
results as with crude PPL Lot 096K0747.
Characterization of compounds is in accordance with earlier
reported data [17–19]. The absolute configuration of (2R)-2 was
determined by comparing optical rotation with previously reported
values [17].
(2R)-2
Diol 1
(2S)-2
It is still interesting to reveal which of the enzymes in the crude
preparation are responsible for the (R)-enantiomer. The PPL Type II
enzyme preparation (Lot 096K0747, 358 U/mg) shows a slight chy-
motrypsin activity (7.2 × 10−2 U/mg) and general protease activity
(2.5 × 10−4 U/mg). However, these enzymes are not responsible for
the (R)-monoesters: Only the (S)-monobutanoate ((2S)-4, 100% ee,
not shown in scheme) was produced in the reaction of 1 in THF
with trypsin/chymotrypsin from pig pancreas as the catalyst and
vinyl butanoate as the acyl donor (entry F). Pig liver esterase also
produced solely the (S)-monobutanoate ((2S)-4, 100% ee, entry G)
under the same conditions. As mentioned, Segura et al. reported
that in kinetic resolution of ethyl 2-hydroxy-4-phenylbutanoate
all three fractionated lipases (25-kDa, 33-kDa and PPL (52-kDa))
favored production of the (S)-monoester, however with low E-
values [9]. Since the enantioselectivity of PPL towards this substrate
is low, both enantiomers are actually produced. The fact that the
enantiomeric excess calculations are based on the ee only at one
single point of conversion, which is common, may also play a role.
As mentioned, we have earlier observed that a change in enantiose-
lectivity in kinetic resolutions was due to influence on the enzyme
catalyst by (R)-alcohols in the medium, a possible non-competitive
(or allosteric inhibition) [3,4]. Chiral compounds in the reaction
medium is one of the oldest ways of obtaining enantioselectivity.
Forty years ago Grignard reactions gave 70% ee when the reaction
was carried out in the presence of 1,2:5,6-di-O-isopropylidene-␣-
d-glucofuranose [22].
Opposite from resolution, in desymmetrizations of prochiral
compounds there are no chiral molecules present in the reaction
mixture at the start of the reaction, but as the reaction proceeds,
the concentration of pure enantiomers increases. Inhibition of
enzymes by the formed products may also occur. In our experi-
ments the slow-down of the (S)-selective enzyme may be a result
of non-competitive inhibition by vinyl acetate and/or the produced
(R)-monoester.
Although enantioselectivity and enantiomeric excess per def-
inition are constant, concentrations of substrate and product
enantiomers change throughout a reaction possibly affecting both
catalysts and reaction medium. Our results show that it is always
very important to monitor the reaction progress.
85 min
Diacetate
Time, min
Conversion, %
0 5
40
30
45
60
54
90
60 62
120
70
150
75
180
85
Fig. 2. Composition of reaction mixture in % of each compound, during acetylation
(180 min) of 1 (circles) catalyzed by crude PPL in THF with vinyl acetate (equimolar
amounts) (entry A, Table 1). (2S)-2 (squares), (2R)-2 (diamonds), diacetate (trian-
gles).
(2R)-monoacetate, (2R)-2, is predominant (70% ee). Since no diac-
etate was observed during the first 3 h of reaction, the changing ee
is not due to interconversion of (2S)-monoacetate via diacetate 3 to
the (2R)-monoacetate. Acyl migration was shown to be negligible,
which is anticipated due to the apolar solvent. Fadel and Arzel have
reported that monoesters of serinol undergo racemization in acidic
environments but not in apolar solvents [20]. Similar results have
been obtained in our group [21].
When the acetylation of 1 with crude PPL was carried out in
pure vinyl acetate (entry C, as previously reported by Choi and
Borch [17]), only the (2R)-monoacetate, (2R)-2, was produced (in
98% ee) after 1.5 h. When the desymmetrization was catalyzed by
purified PPL (Sigma Type VI-S, L0382, 20,000 U/mg) only the (2S)-
monoacetate, (2S)-2, was formed (100% ee, entry E). This indicates
that the (R)-selective enzyme is present in the crude PPL prepa-
ration. That the changing ee in reactions of the crude enzyme
preparation is due to action of two different enzymes is supported
by the fact that the (2R)-monoacetate was formed in excess ((2R)-2,
98% ee) when vinyl acetate was used as solvent. In this case it seems
100
80
OH
H
XHN
60
OAc
(2S)-2
4. Conclusion
40
Action of lipases with opposite stereoselectivity seems to be the
reason for the changing enantioselectivity of crude PPL in acety-
lation of 1. Since crude PPL is a mixture of enzymes for digestive
purposes, it is reasonable that it is able to hydrolyze both enan-
tiomers of fats. However, the most abundant lipase in pig pancreas
is (S)-selective towards 1. It seems that this enzyme is the only
active enzyme in the hydrolysis reaction of diacetyl cbz-serinol
(3), since the only product was (2S)-2. Experiments are under way
in order to establish which one of the 25-kDa and the 33-kDa
lipase is giving rise to the (R)-monoesters of 1. The (S)-selective
enzyme is the faster reacting enzyme. Vinyl acetate acts as an
inhibitor of the (S)-selective enzyme when used in large excess as
OAc
H
XHN
OH
(2R)-2
20
85 min
0
180
5
120
Time, min
30
60
150
90
Fig. 3. Enantiomeric excess (ee) of monoacetates (2S)-2 (circles) and (2R)-2
(squares), entry A. At 85 min of reaction ee is zero. (2S)-2 in excess before 85 min,
(2R)-2 in excess at 180 min (70% ee).

E.E. Jacobsen et al. / Journal of Molecular Catalysis B: Enzymatic 85–86 (2013) 134–139
139
solvent, giving rise only to the (R)-monoacetate. However, since
the (S)-selective enzyme “slows down” as the (2R)-monoester is
produced when vinyl acetate is used in only 1.0–4.3 times excess,
the (2R)-monoester may also play a role in the inhibition of the
(S)-selective enzyme. Trypsin and chymotrypsin (from pancreas)
and esterase from pig liver catalyzed the conversion of the (2S)-
monobutanoate ((2S)-4) in 100% ee. Our experiments demonstrate
that the crude PPL preparation gives rise to both monoester enan-
tiomers in high enantiopurity depending on reaction conditions.
When crude enzyme preparations are used, monitoring of the reac-
tion progress is of utmost importance to determine the reaction of
several enzymes in order to obtain the wanted enantiomer.
[2] U.T. Bornscheuer, R.J. Kazlauskas, Hydrolases in Organic Synthesis. Regio- and
Stereoselective Biotransformations, 2nd ed., WILEY-VCH, 2005.
[3] E.E. Jacobsen, E.W. van Hellemond, A.R. Moen, L.C.V. Prado, T. Anthonsen, Tetra-
hedron Lett. 44 (2003) 8453.
[4] E.E. Jacobsen, L.S. Andresen, T. Anthonsen, Tetrahedron: Asymmetry 16 (2005)
847.
[5] E.E. Jacobsen, B.H. Hoff, T. Anthonsen, Chirality 12 (2000) 654.
[6] W.E. Ladner, G.M. Whitesides, J. Am. Chem. Soc. 106 (1984) 7250.
[7] M. Pregnolato, M. Terreni, I.E. de Fuentes, A. Alcantara Leon, P. Sabuquillo,
R. Fernandez-Lafuente, J.M. Guisán, J. Mol. Catal. B: Enzym. 11 (2001)
757.
[8] T. Vorderwülbecke, K. Kieslich, H. Erdmann, Enzyme Microb. Technol. 14 (1992)
631.
[9] R.L. Segura, J.M. Palomo, C. Mateo, A. Cortes, M. Terreni, R. Fernández-Lafuente,
J.M. Guisan, Biotechnol. Progr. 20 (2004) 825.
[10] T.A. Savidge, Biotechnology of Industrial Antibiotics, Marcel Decker, New York,
1984.
[11] V. Kerscher, W. Kreiser, Tetrahedron Lett. 28 (1987) 531.
[12] D. Breitgoff, K. Laumen, M.P. Schneider, J. Chem. Soc. Chem. Commun. (20)
(1986) 1523.
[13] J.B.A. Van Tol, J.A. Jongejan, J.A. Duine, Biocatal. Biotransform. 12 (1995) 99.
[14] H. Tani, H. Ohishi, K. Watanabe, J. Agric. Food Chem. 43 (1995) 2796.
[15] J.D. Vaghjiani, T.S. Lee, G.J. Lye, M.K. Turner, Biocatal. Biotransform. 18 (2000)
151.
[16] K. Takabe, Y. Iida, H. Hiyoshi, M. Ono, Y. Hirose, Y. Fukui, H. Yoda, N. Mase,
Tetrahedron Lett. 11 (2000) 4825.
[17] J.Y. Choi, R.F. Borch, Org. Lett. 9 (2007) 215.
[18] T. Wharton, L. Wilson, J. Bioorg. Med. Chem. 10 (2002) 3545.
[19] X. Cong, F. Hu, K.-G. Liu, Q.-J. Liao, Z.-J. Yao, J. Org. Chem. 70 (2005) 4514.
[20] A. Fadel, P. Arzel, Tetrahedron: Asymmetry 8 (1997) 283–291.
[21] B.J. Sjursnes, L. Kvittingen, T. Anthonsen, J. Am. Oil Chem. Soc. 72 (1995)
533.
Acknowledgments
We thank The Research Council of Norway (RCN) for a research
fellowship (contract Grant Number 202903/11) to M.F.B. Doctors
Pere Clapes and Jesús Joglar (Consejo Superior de Investiga-
ciones Científicas Dpto. Química Pèptids i Proteïns, Instituto de
Investigaciones Químicas y Ambientales, Barcelona) are gratefully
acknowledged for assistance. Lipase B from Candida antarctica was
a gift from Novozymes AS, Bagsværd, Denmark.
References
[22] T.D. Inch, G.J. Lewis, G.L. Sainsbury, D.J. Sellers, Tetrahedron Lett. 41 (1969)
3657.
[1] K. Faber, Biotransformations in Organic Chemistry, 6th ed., Springer-Verlag,
Berlin, 2011, p. 42.