Transformation of racemic ethyl 3-hydroxybutanoate into the (R)-enantiomer exploiting lipase catalysis and inversion of configuration

Article abstract of DOI:10.1016/j.tetasy.2007.06.030

Racemic ethyl 3-hydroxybutanoate rac-1 was transformed into ethyl (R)-acetoxybutanoate (ee = 92%) with 85-90% chemical yields using enantioselective acylation with isopropenyl acetate in the presence of Candida antarctica lipase B (CAL-B, Novozym 435) under solvent-free conditions, followed by mesylation of the unreacted (S)-alcohol in the reaction mixture and inversion of configuration with cesium acetate in DMF in one pot. When the (R)-acetoxybutanoate was subjected to alcoholysis with ethanol and CAL-B, enantiopure (R)-1 (ee >99%) was produced.

Full text of DOI:10.1016/j.tetasy.2007.06.030

Tetrahedron: Asymmetry 18 (2007) 1682–1687
Transformation of racemic ethyl 3-hydroxybutanoate
into the (R)-enantiomer exploiting lipase catalysis
and inversion of configuration
Mihaela C. Turcu,a Eero Kiljunenb and Liisa T. Kanervaa,*
aDepartment of Pharmacology, Drug Development and Therapeutics/Laboratory of Synthetic Drug Chemistry
and Department of Chemistry, University of Turku, Lemminka¨isenkatu 5 C, FIN-20520 Turku, Finland
bOrion Pharma, PO Box 425, FIN-20101 Turku, Finland
Received 5 June 2007; accepted 21 June 2007
Available online 7 August 2007
Abstract—Racemic ethyl 3-hydroxybutanoate rac-1 was transformed into ethyl (R)-acetoxybutanoate (ee = 92%) with 85–90% chemical
yields using enantioselective acylation with isopropenyl acetate in the presence of Candida antarctica lipase B (CAL-B, Novozym 435)
under solvent-free conditions, followed by mesylation of the unreacted (S)-alcohol in the reaction mixture and inversion of configuration
with cesium acetate in DMF in one pot. When the (R)-acetoxybutanoate was subjected to alcoholysis with ethanol and CAL-B, enan-
tiopure (R)-1 (ee >99%) was produced.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
The reduction of a prochiral b-keto ester in the formation
of an optically active b-hydroxy ester has been performed
using both purified enzymes and microbial cells.6–10
Although significant results have been obtained with
microbial cells, narrow substrate specificity, low stereo-
chemical outcome, and low tolerance for synthetic reaction
conditions still remain as drawbacks. The use of isolated
dehydrogenases, on the other hand, involves high-cost
purification methods and the development of systems for
cofactor regeneration. An interesting method for the prep-
aration of ethyl (R)-3-hydroxybutanoate was previously
reported by hydrolyzing commercial poly[(R)-3-hydroxy-
butanoate] produced from glucose (recovered by enzymatic
hydrolysis of pulp fiber sludge) by fermentation with Alca-
ligenes eutrophus.11,12 Inversion of configuration via the
mesylate ester was described to produce the (S)-
enantiomer.12
The versatility of the enantiomers of 3-hydroxybutanoate
esters as chiral building blocks for the synthesis of bioactive
compounds, for example, for the synthesis of b-lactam
antibiotics,1–3 is well established.4 Especially interesting
are the antimicrobial classes of carbapenems and penems,
many of which are now in clinical phases (Fig. 1). The
interest is based on the high potential of such compounds
against hospital and community pathogens, and more gen-
erally, on their broad antimicrobial spectra. Consequently
much interest has been focused on developing routes to
the synthesis of the b-hydroxybutanoate enantiomers.4,5
OH
OH
O
R1
OH
H
H
H H
N
COOR4
S
R2
R3
N
The combination of broad substrate specificity, usually
high stability and selectivity, effective catalysis without
cofactors, and good commercial availability has made lip-
ases especially attractive for the kinetic resolution of race-
mates. As a further benefit, both the enantiomers are
simultaneously obtained when the reaction is highly enan-
tioselective. On the other hand, a maximum yield of 50%
for one enantiomer is a drawback when the other enantio-
mer is useless. Various types of kinetic resolution reactions
(R
)-3-hydroxybutanoate ester
O
COOH
COOH
4
1
(R
)- (R =Et)
Carbapenem
Penem
Figure 1. Role of (R)-hydroxybutanoate esters in the structure of
carbapenems and penems.
*
Corresponding author. Tel.: +358 2 333 6773; fax: +358 2 333 7955;
e-mail: lkanerva@utu.fi
0957-4166/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2007.06.030

M. C. Turcu et al. / Tetrahedron: Asymmetry 18 (2007) 1682–1687
1683
of racemic ethyl 3-hydroxybutanoate rac-1 (Scheme 1)
using lipases in organic solvents have been reported, leav-
ing room for enantioselectivity enhancements. Thus, ami-
nolysis of rac-1 with various lipases and amines afforded
resolution products (the unreacted 1 and the formed amide
product), the absolute configuration of the more reactive
enantiomer and enzymatic enantioselectivity depending
on the lipase used.13 The alcoholysis of rac-1 with 1-buta-
nol in hexane in the presence of various lipases was shown
to give products with low enantiopurity.14 Lipase-catalyzed
acylations of rac-1 or enantiomerically enriched (S)-1 have
also been studied.14,15 A solvent-free system for acylation
with vinyl acetate was developed in the presence of lipase
B from Candida antarctica (CAL-B), leading to the forma-
tion of the unreacted (S)-1 (ee = 96%) and the formed
enantiomerically enriched (R)-2 (ee = 80%) at 60% conver-
sion.14 The enantiomerically enriched (R)-2 gave (R)-1
(ee = 96%) at relatively low chemical yield through CAL-
B-catalyzed alcoholysis with ethanol in hexane.
holysis of rac-2 with ethanol and to find conditions where
the possible elimination of water from 1 is minimized dur-
ing the chemical steps leading to inversion and during all
enzymatic reaction stages.
2. Results and discussion
2.1. Acylation of ethyl 3-hydroxybutanoate rac-1
In accordance with the previous results,14 CAL-B proved
to be the most enantioselective of all the lipases screened
for the acylation of rac-1. The screenings were performed
with vinyl butanoate as an irreversible acyl donor in tert-
butyl methyl ether (TBME). Vinyl butanoate was used
because lipases often favor butanoate esters over acetate
esters. As shown by conversions, which reached after the
reaction of 5 min, reactivity (conversions 24%, 40%, 40%,
49%, and 54% with 5, 10, 20, 50, and 100 mg mLꢀ1 of
the enzyme) was considerably affected by the CAL-B con-
tent of the model reaction. The order for most favored
reaction media was diisopropyl ether (DIPE) ꢁ TBME >
toluene > hexane > cyclohexane. Optimization was contin-
ued using CAL-B (5 mg mLꢀ1) in TBME. All together,
these studies failed in improving the enantioselective out-
come in the enzymatic acylation of rac-1.
OH
OH
OAc
EtOOC
(i)
COOEt
COOEt
+
(
R
)-2
(S)-1
rac
-
1
(ii)
For the lipase-catalyzed acylation of chiral alcohols
[for instance Scheme 1: rac-1 + RCO2R1 ! (S)-1 + (R)-2 +
R1OH] to lead to enantiopure products, the enzymatic
deacylation of the produced (R)-2 with R1OH back to
the starting alcohol (R)-1 needs to be prevented. This is
usually accomplished by shifting the thermodynamic equi-
librium to the product side, by using a high excess of an
achiral acyl donor (RCO2R1 as a solvent and an acyl
donor), or by using an activated achiral acyl donor, which
either liberates an unstable R1OH [HOCH@CH2 or HOC-
(Me)@CH2] or an alcohol with low nucleophilic character
(HOCH2CF3). With this in mind, acyl donors were
screened for the acylation of rac-1 in the presence of
CAL-B (5 mg mLꢀ1) in TBME (Table 1). In the case of
ethyl esters as acyl donors, enzymatic acylation proceeded
slowly (entries 3 and 6) compared to the reactions with
alkyl activated esters (entries 1, 2, 4, and 5). From the
activated esters, vinyl esters (entries 2 and 4) gave lower
enantiomer ratios, E, than 2,2,2-trifluoroethyl and isopro-
penyl esters (entries 1 and 5). We had expected this type
of behavior of vinyl esters since acetaldehyde (the decom-
position product of vinyl alcohol) is reported to be more
OMs
OAc
EtOOC
COOEt
+
(R
)-2
(S)-3
(iii)
OAc
COOEt
)-2
(iv)
(
R
OH
COOEt
(
R)-1
Scheme 1. Transformation of rac-1 into (R)-1 (i) acylation with isopro-
penyl acetate, 30 mg/mL CAL-B; (ii) MsCl, TEA in DCM; (iii) AcOCs in
DMF; (iv) alcoholysis with 25 mg/mL CAL-B, EtOH in TBME.
Table 1. Effect of acyl donor (0.2 M) on the acylation of rac-1 (0.1 M) in
TBME in the presence of CAL-B (5 mg mLꢀ1) at room temperature
Based on the general benefits of lipase catalysis and on the
reported successful inversion of configuration of (R)-112 we
herein report the transformation of rac-1 into (R)-1
through a three-step protocol: lipase-catalyzed acylation/
inversion of configuration of the unreacted alcohol/
lipase-catalyzed alcoholysis, all in one pot without any
significant purification of the products (Scheme 1). For this
purpose, it was necessary to improve the enantioselectivity
of the lipase-catalyzed acylation of rac-1 from that which
has been reported,14,15 to study the lipase-catalyzed alco-
Entry Acyl donor
Time (h) Conversion (%)
E
1
2
3
4
5
6
PrCO2CH2CF3
PrCO2CH@CH2
PrCO2Eta
0.5
0.5
4
40
49
36
44
48
35
95
64
32 11
65
150
92
1
4
MeCO2CH@CH2
0.5
1
4
3
MeCO2C(CH3)@CH2 0.5
MeCO2Etb
2
a PrCO2Et as an acyl donor and as a solvent.
b MeCO2Et as an acyl donor and as a solvent.

1684
M. C. Turcu et al. / Tetrahedron: Asymmetry 18 (2007) 1682–1687
Table 2. Effect of rac-1 and isopropenyl acetate concentrations on CAL-B-catalyzed acylation at room temperature
Entry
Amount of enzyme (mg mLꢀ1
)
[rac-1]/[MeCO2C(Me) = CH2]a
Time (h)
Conversion (%)
E
1
2
3
4
5
5
5
5
5
5
0.1/0.4
0.1/0.2
0.1/0.1
0.1/0.08
0.1/0.06
0.5
0.5
0.5
0.5
0.5
49
49
49
47
49
150
151
138
181 13
133 13
3
4
3
6
7
8
5
5/5
5/5
5/5
5/5
5/4
5/4
5/3
5
5
5
4.5
5
23
50
52
56
27
109
94
87
98b
117
90
7
5
3
20
30
40
5
30
30
9
10
11
12
3
4
4
5
5/8
50
46/51
102
a [mol Lꢀ1]/ [mol Lꢀ1].
b Due to the fast reaction, the E-value was calculated using only the ee values at one conversion (38%).
harmful to enzymes than for instance acetone (the decom-
position product of isopropenyl alcohol).16 We also found
that E-values for the enzymatic acylation of rac-1 with
achiral butanoate esters started to drop close to 50% con-
versions. In the case of acetate esters, the E-value remained
more constant through the reaction. As the most important
result, enzymatic enantioselectivity was considerably
enhanced when isopropenyl acetate (E = 150, entry 5)
was used as an acyl donor.
(ee = 98%) and (R)-2 (ee = 89%) was subjected to the reac-
tion with mesyl chloride (route ii), leading to a mixture of
(R)-2 (ee = 86%) and (S)-3 (ee = 98%). This mixture was
then subjected to reactions with nucleophiles for inversion
to take place.
Hydrolysis under neutral conditions with aqueous CaCO3
was first studied in a mixture of (R)-2 and (S)-3. The idea
was based on the reported inversion of configuration of
(R)-3 under such conditions.12 While (S)-3 in the present
work was inverted, (R)-2 remained the same, leading to a
mixture of (R)-1 and (R)-2. It was shown that only a minor
part of (R)-2 could be hydrolyzed with aqueous CaCO3,
even after refluxing the system for two days. The hydrolysis
of (R)-2 in the mixture needs to be performed under condi-
tions where the possible elimination of water from (R)-1 is
minimized. Thus, enzymatic hydrolysis seemed to be most
reasonable as it may also improve the enantiopurity of (R)-
2. We gave up on this idea, however, since in a mixture of
(R)-1 and (R)-2, enzymatic hydrolysis may lead to the
hydrolysis of the ethyl ester function in addition to that
of the acetate function in (R)-2.
In the next step, the concentration effects on the acylation
of rac-1 with isopropenyl acetate and CAL-B were studied
(Table 2). The enantioselectivity in TBME (0.1 M rac-1,
entries 1–5), on one hand, and the studied solvent-free sys-
tems (5 M rac-1, entries 6–12) on the other hand, were
hardly affected by the amount of isopropenyl acetate.
The enantioselectivity in TBME was somewhat higher than
in the solvent-free cases. This conclusion is justifiable in
spite of the small variations in the E-values within Table
2, entries 1–5 and 6–12, accepting the fact that minor
experimental inaccuracies may cause extensive variations
in relatively high E-values. Thus, it was reasonable to turn
attention to solvent-free systems and at the same time,
increase CAL-B contents in order to obtain faster acylations
(entries 6–9 as well as 10 and 11). For the kinetic resolution
of rac-1 under solvent-free conditions, it was advantageous
that isopropenyl acetate was not used in high excess with
the reactive enantiomer (the unreacted acyl donor needs
to be removed) although the reactivity was not the best
(entries 11 and 12). For a gram-scale kinetic resolution,
rac-1 (5 M) and isopropenyl acetate (3 M) were mixed with
CAL-B (30 mg mLꢀ1), leading to (S)-1 (ee = 95%, isolated
yield 98%) and (R)-2 (ee = 92%, isolated yield 84%).
Next, the transformation of (S)-3 into (R)-2 in their mix-
ture was studied using acetate as a nucleophile (Scheme
1, step iii). Sodium acetate in toluene and 18-crown-6 ether
as a phase transfer catalyst was tested first but the reaction
was very slow even under reflux conditions as detected on
TLC and by GC. Cesium carboxylates have been used
for SN2 displacements of mesylates17 and tosylates.18 Thus,
(S)-3 was dissolved in DMF or in DMSO in order to study
the usability of cesium acetate. The high solubility of
cesium acetate in DMSO induced high basicity favoring
side-reactions, and an unidentified compound [other than
the desired (R)-2] was observed. In DMF, inversion of con-
figuration took place by carefully controlling the refluxing
temperature, allowing the transformation of the mixture of
(R)-2 and (S)-3 into crude (R)-2. This crude material was
directly usable for the enzymatic transformation of (R)-2
into (R)-1. We had created a highly efficient chemoenzy-
matic method where rac-1 was transformed into (R)-2 with
80–90% chemical yields and with 86–94% ee as the repeti-
tion of the process revealed.
2.2. One-pot synthesis of ethyl (R)-3-acetoxybutanoate
The entire protocol in Scheme 1 was performed without
serious purification between the steps. The completion of
each reaction step was monitored by GC in the terms of
conversion and enantiopurity. The protocol was started
by mixing rac-1 with isopropenyl acetate and CAL-B
(route i), and the mixture was shaken at room temperature
as described (see Section 4). The resulting mixture of (S)-1

M. C. Turcu et al. / Tetrahedron: Asymmetry 18 (2007) 1682–1687
1685
Table 4. Effect of (R)-2 (ee = 86%) concentration on the CAL-B
(25 mg mLꢀ1)-catalyzed alcoholysis (1 equiv) in TBME at room
temperature
2.3. Alcoholysis of ethyl 3-acetoxybutanoate
Our aim was to use CAL-B-catalyzed alcoholysis for the
transformation of (R)-2 into (R)-1 under mild conditions
where (R)-1 is stable against the elimination of water. In
order to obtain background information, enzymatic
enantioselectivity and reactivity were first studied by sub-
jecting rac-2 (0.1 M) to the reaction with ethanol (0.2–
0.5 M) and CAL-B (25 mg mLꢀ1) in hexane and in TBME
(Scheme 2). If desired (although not carried out in the
present work), the excellent enantioselectivity observed
(E ꢂ 200 as measured at 40% or lower conversions) in
both the solvents allow the kinetic resolution at 50% con-
version (Table 3). The reaction in TBME was considerably
faster than in hexane. Thus, the reaction with 0.2 M etha-
nol in TBME was at 50% conversion in an hour (entry 4)
while 4 h were necessary for the same reaction in hexane
(entry 1). In both solvents, reactivity decreased when
increasing the ethanol content. Attempts to use high rac-
2 concentrations (accordingly also high ethanol concentra-
tions) caused considerable rate retardation (Table 4).
Moreover, a drop in ee values was evident at conversions
close to 50% because the produced (R)-1 enzymatically
reacts with ethyl acetate back to (R)-2 and ethanol. In
addition, the enzymatic hydrolysis of 2 with water from
the CAL-B preparation can take place. Accordingly the
drop in ee was more pronounced in more hydrophilic
TBME than in hexane.
Entry
(R)-2 (M)
Time (h)
Conversion (%)
ee(R)-1 (%)
1
2
3
4
5
6
7
8
1
1
1
2
2
3
3
3
7/24
7/24
7
7/24
7
7/24
7/24
24/(24+5)e
84/86
64/92a
82b
72/83
83c
52/78
38/84d
76/89
99.8
99.9
99.7
99.6
99.7
99.8
99.8
99.9
a Ethanol (1.2 equiv).
b CAL-B (50 mg mLꢀ1).
c CAL-B (100 mg mLꢀ1).
d Ethanol (2 equiv).
e After 24 h ethanol (0.25 equiv) was added.
(Tables 3 and 4, entry 1 compared to 2 and 6 to 7), the reac-
tion was optimized using only 1 equiv of ethanol. In the-
ory, the highly enantioselective alcoholysis will proceed
to 93% conversion with initial ee(R)-2 = 86%. However,
under equimolar conditions the alcoholysis will stop at
an equilibrium, and in practice, the alcoholysis of (R)-2
(ee = 86%) with ethanol (1 equiv) and CAL-B
(25 mg mLꢀ1) in TBME proceeded to somewhat over
80% conversion with 1–3 M substrate concentrations in
reasonable times. The benefit in the terms of higher conver-
sions with higher ethanol concentrations was minimal
(entry 2 compared to 1 and entries 7 and 8 to 6).
Table 3. Alcoholysis of rac-2 (0.1 M) with ethanol in the presence of
CAL-B (25 mg mLꢀ1) in hexane and TBME
Entry
Ethanol
(M)
Time
(h)
Conversion
(%)
ee(S)-2
(%)
ee(R)-1
(%)
Hexane
3. Conclusions
1
2
3
0.2
0.3
0.5
4
24
24
50
50
50
97
97
98
99
96
96
In conclusion, a CAL-B-catalyzed kinetic resolution meth-
od under solvent-free conditions was developed which to-
gether with the inversion of (S)-1 (the unreacted hydroxy
ester enantiomer) through mesylation followed by the
nucleophilic substitution with cesium acetate allowed the
transformation of ethyl 3-hydroxybutanoate (rac-1) into
the acetate product (R)-2 in one pot with chemical yields
of 85–90% and close to 90% ee. The CAL-B-catalyzed alco-
holysis of the obtained (R)-2 with ethanol in TBME gave
enantiopure (R)-1 (ee >99%).
TBME
4
6
7
0.2
0.3
0.5
1
1
5
50
49
50
95
94
97
98
98
96
Finally, the CAL-B-catalyzed alcoholysis of the above
crude product (R)-2 (ee = 86%) furnished (R)-1 in an enan-
tiopure form (ee >99%) with ethanol in TBME (Scheme 1,
route iv, Table 4). When the substrate was of high enantio-
purity already in the beginning of the alcoholysis, the risk
for drops in ee values in concentrated TBME solutions
was not critical compared to the benefit obtained from fas-
ter reactions. As reactivity was shown to drop significantly
even with a small increase in the ethanol concentration
The studies of the CAL-B-catalyzed alcoholysis of rac-2
with ethanol in TBME revealed excellent enantioselectivity
(E ꢂ 200). It is clear (although not studied here) that a
three-step protocol, including chemical acylation of rac-1/
lipase-catalyzed alcoholysis with ethanol/inversion of con-
figuration will allow the transformation of rac-1 into (S)-2.
OH
OAc
OAc
COOEt
EtOOC
COOEt
CAL-B
TBME
+
EtOH
+
EtOAc
+
(R)-1
rac-2
Scheme 2. Alcoholysis of rac-2 with ethanol.
(S)-2

1686
M. C. Turcu et al. / Tetrahedron: Asymmetry 18 (2007) 1682–1687
4. Experimental
<98%12}. 1H NMR (CDCl3, 500 MHz), [(R)-2]: d = 1.24
(t, J = 7.1 Hz, 3H, CH2CH3); 1.29 (d, J = 6.33 Hz, 3H,
CH3); 2.02 (s, 3H, COCH3); 2.4 (dd, J = 5.7 Hz, J = 15.4
Hz, 1H, CH2); 2.6 (dd, J = 7.1 Hz, J = 15.4 Hz, 1H,
CH2); 4.12 (q, J = 6.5 Hz, 2H, –OCH2CH3); 5.25–5.29
(m, 1H, CH). 13C NMR (CDCl3, 126 MHz): d = 14.15
(–OCH2CH3); 19.87 (C1), 21.14 (–COCH3), 40.86 (C3),
60.58 (–OCH2CH3); 67.32 (C2); 170.21 (–OCOCH2CH3);
170.22 (–OCOCH3).
4.1. Materials and methods
Racemic ethyl 3-hydroxybutanoate rac-1, methanesulfonyl
chloride (mesyl chloride, MsCl), cesium acetate, vinyl ace-
tate, and isopropenyl acetate were products of Aldrich,
Fluka or Acros Organics. Ethyl 3-acetoxybutanoate was
prepared from commercial alcohol using acetic anhydride,
triethyl amine, and DMAP. The solvents were of the high-
est analytical grade and were dried over molecular sieves
˚
4.3. Transformation of rac-1 into (R)-1
(4 A, 16 mm pellets) before use. CAL-B (C. antarctica
lipase B, NovozymÒ 435) was a generous gift from Novo-
zymes. Preparative chromatographic separations were per-
formed by column chromatography on Merck Kieselgel 60
(0.063–0.200 lm). Analytical thin layer chromatography
(TLC) was carried out on Merck Kieselgel 60F254 sheets
and compounds were visualized using potassium perman-
ganate dip. All enzymatic reactions were carried out at
room temperature (23 °C).
rac-1 (20.0 g, 151 mmol, 19.7 mL) was mixed with isopro-
penyl acetate (9.08 g, 90.8 mmol, 9.9 mL) and CAL-B
was added (30 mg mLꢀ1). The mixture was stirred at room
temperature. The reaction was stopped by filtering off the
enzyme at 52% conversion (ee(R)-2 = 89% and ee(S)-1
=
98%). The enzyme was washed with TBME and the solvent
was evaporated.
1
The H NMR and 13C NMR spectra were recorded on a
The above mixture of (S)-1 and (R)-2 was introduced in
dichloromethane (100 mL), after which methanesulfonyl
chloride (9.2 g, 80.2 mmol, 6.2 mL) was added and the mix-
ture stirred in an ice bath until the temperature of 0 °C was
reached. Triethylamine (8.1 g, 80.2 mmol, 11.3 mL) was
added slowly within 40 min, while keeping the temperature
below 10 °C. The reaction mixture was filtered, and the fil-
trate was evaporated until a white slurry was obtained. The
slurry was introduced in diethyl ether (50 mL) and the pre-
cipitate formed was removed by filtration. According to
Bruker 500 spectrometer with tetramethylsilane (TMS) as
an internal standard. Spectroscopic data for 16 and (S)-
312 were in accordance with those given in the litera-
ture. Optical rotations were measured with a Perkin–
Elmer 341 Polarimeter, and [a]D values are given in units
of 10ꢀ1 deg cm2 gꢀ1. The determination of the E-value
is based on the equation E = ln[(1 ꢀ c)(1 ꢀ eeS)]/
ln[(1 ꢀ c)(1 + eeS)] with c = eeS/(eeS + eeP) using linear
regression (E as the slope of the line ln[(1 ꢀ c)(1 ꢀ eeS)]
versus ln[(1 ꢀ c)(1 + eeS)]).19 Decane was used as an exter-
nal standard to estimate the conversion during the alcohol-
ysis of (R)-2.
GC analysis, the obtained mixture of (R)-2 (ee(R)-2
=
86%) and (S)-3 (ee(S)-3 = 98%) did not contain any (S)-1.
The mixture of (R)-2 and (S)-3 was introduced in DMF
(140 mL) and cesium acetate (15.2 g, 79.4 mmol) was
added. The reaction mixture was gently refluxed for 2 h.
Diethyl ether was added and the product was isolated by
aqueous extraction.
In a typical small-scale experiment, CAL-B was added to
the solution of rac-1 (0.1–5 M) or rac-2 (0.1–3 M) and an
acyl donor or ethanol in an organic solvent or without a
solvent. The reaction mixture was shaken at room temper-
ature. The progress of the reaction and the ee values of the
products were followed by taking samples (0.1 mL) at
intervals and analyzing them by GC on a Varian CP-chira-
sil-Dex CB column (25 m). For a good baseline separation,
the unreacted hydroxyl group in the sample was derivatized
with propionic or acetic anhydride in the presence of
pyridine containing 1% 4-N,N-dimethylaminopyridine
(DMAP).
The obtained crude (R)-2 (ee = 86%) was subjected to
enzymatic alcoholysis with ethanol (6.32 g, 137.3 mmol,
8 mL) in TBME in the presence of CAL-B (25 mg/mL).
The reaction was stopped after 25 h at 82% conversion.
Conversion was determined using decane as an external
standard. After column chromatography with petrol
ether/ethyl acetate (9:1) as an eluent, (R)-1 {ee = 99%, yield
25
60%, ½aꢃD ¼ ꢀ43:1 (c 1, CHCl3)} was obtained.
4.2. Gram-scale kinetic resolution of rac-1
rac-1 (5.0 g, 37.8 mmol, 4.91 mL) was mixed with isopro-
penyl acetate (2.27 g, 22.7 mmol, 2.46 mL) and CAL-B
was added (30 mg/mL). The mixture was stirred at room
temperature for 7 h. The reaction was stopped by filtering
off the enzyme at 51% conversion, and the enzyme was
washed with TBME. After evaporation, the unreacted
(S)-1 and the formed (R)-2 were separated on silica gel
Acknowledgments
The authors acknowledge Tekes (Finnish Funding Agency
for Technology and Innovation) and PCAS Finland Oy for
financial support.
by elution with petrol ether/ethyl acetate (9:1), the elution
25
sequence being (R)-2 {16.3 mmol, ee = 92%, ½aꢃD ¼ þ2:9 (c
References
25
1, CHCl3)} before (S)-1 {18.1 mmol, ee = 95%, ½aꢃD ¼ þ42
22
(c 1 CHCl3); the literature data ½aꢃD ¼ þ36:9 (c 2.23,
1. Bush, K.; Macielag, M.; Weidner-Wells, M. Curr. Opin.
Microbiol. 2004, 7, 466–476.
CHCl3), ee = 89.5%14 and [a]D = +40 (c 1, CHCl3), ee

M. C. Turcu et al. / Tetrahedron: Asymmetry 18 (2007) 1682–1687
1687
2. Kobayashi, R.; Konomi, M.; Hasegawa, K.; Morozumi, M.;
Sunakawa, K.; Ubukata, K. Antimicrob. Agents Chemother.
2005, 49, 889–894.
3. Dahloff, A.; Janjic, N.; Echols, R. Biochem. Pharmacol. 2006,
71, 1085–1095.
4. Berks, A. H. Tetrahedron 1996, 52, 331–375.
5. Ham, W.-H.; Oh, C.-Y.; Lee, Y.-S.; Jeong, J.-H. J. Org.
Chem. 2000, 65, 8372–8374.
6. Inoue, K.; Makino, Y.; Itoh, N. Tetrahedron: Asymmetry
2005, 16, 2539–2549.
11. Zhang, S.; Norrlo¨w, O.; Wawrzynczyk, J.; Dey, E. S. Appl.
Environ. Microbiol. 2004, 70, 6776–6782.
12. Carnell, A. J.; Head, R.; Bassett, D.; Schneider, M. Tetra-
hedron: Asymmetry 2004, 15, 821–825.
13. Garcia, M. J.; Rebolledo, F.; Gotor, V. Tetrahedron: Asym-
metry 1992, 12, 1519–1522.
14. Fishman, A.; Eroshov, M.; Dee-Noor, S. S.; Mil, J.; Cogan,
U.; Effenberger, R. . Biotechnol. Bioeng. 2001, 74, 256–
263.
15. Sugai, T.; Ohta, H. Agric. Biol. Chem. 1989, 53, 2009–
2010.
7. Ribeiro, J. B.; Ramos, M. C. K.; Aquino Neto, F. R.; Leite,
S. G. F.; Antunes, O. A. C. J. Mol. Cat. B: Enzym. 2003, 24–
25, 121–124.
16. Berger, B.; Faber, K. J. Chem. Soc., Chem. Commun. 1991,
1198–1200.
8. Rodriguez, S.; Schroeder, K. T.; Kayser, M. M.; Stewart, J.
D. J. Org. Chem. 2000, 65, 2586–2587.
17. Shimizu, T.; Hiranuma, S.; Nakata, T. Tetrahedron Lett.
1996, 37, 6145–6148.
9. Kometani, T.; Yoshii, H.; Kitatsuji, E.; Nishimura, H.;
Matsuno, H. J. Ferm. Bioeng. 1993, 76, 33–37.
10. Nakamura, K.; Yamanaka, R.; Matsuda, T.; Harada, T.
Tetrahedron: Asymmetry 2003, 14, 2659–2681.
18. Dijkstra, G.; Kruizinga, W. H.; Kellogg, R. M. J. Org. Chem.
1987, 52, 4230–4234.
19. Chen, C.-S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am.
Chem. Soc. 1982, 104, 7294–7299.