Highly enantioselective kinetic resolution of primary alcohols of the type Ph-X-CH(CH3)-CH2OH by Pseudomonas cepacia lipase: Effect of acyl chain length and solvent

Article abstract of DOI:10.1016/j.tetasy.2003.09.049

Although lipase from Pseudomonas cepacia (PCL) shows high enantioselectivity towards many secondary alcohols, it usually exhibits only low to moderate enantioselectivity towards primary alcohols. To increase this enantioselectivity, we optimised the reaction conditions for the PCL-catalysed hydrolysis of esters of three chiral primary alcohols: 2-methyl-3-phenyl-1- propanol 1, 2-phenoxy-1-propanol 2 and solketal 3. The enantioselectivity towards 1-acetate increased from E=16 to 38 upon changing the solvent from ethyl ether/phosphate buffer to 30% n-propanol in phosphate buffer and increased again to E ≥190 upon changing the substrate from 1-acetate to 1-heptanoate. The same changes increased the enantioselectivity towards alcohol 2 from E=17 to 70, but did not significantly increase the enantioselectivity towards alcohol 3. The best solvent was similar to the solvent used to crystallise the open form of PCL and likely stabilises the open form of PCL. This stabilisation may increase the enantioselectivity by removing kinetic contributions from a non-enantioselective lid-opening step. We determined the kinetic contribution of the lid-opening step by measuring the interfacial activation of PCL. The activation energy for the PCL-catalysed hydrolysis of ethyl acetate was at least 2.6 kcal/mol lower in the presence of a water-organic solvent interface.

Full text of DOI:10.1016/j.tetasy.2003.09.049

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 14 (2003) 3917–3924
Highly enantioselective kinetic resolution of primary alcohols of
the type Ph-X-CH(CH3)-CH2OH by Pseudomonas cepacia
lipase: effect of acyl chain length and solvent
Alessandra Mezzetti, Curtis Keith† and Romas J. Kazlauskas*,‡
Department of Chemistry, McGill University, 801 Sherbrooke Street West, Montre´al, Que´bec, Canada H3A 2K6
Received 1 September 2003; accepted 18 September 2003
Abstract—Although lipase from Pseudomonas cepacia (PCL) shows high enantioselectivity towards many secondary alcohols, it
usually exhibits only low to moderate enantioselectivity towards primary alcohols. To increase this enantioselectivity, we
optimised the reaction conditions for the PCL-catalysed hydrolysis of esters of three chiral primary alcohols: 2-methyl-3-phenyl-1-
propanol 1, 2-phenoxy-1-propanol 2 and solketal 3. The enantioselectivity towards 1-acetate increased from E=16 to 38 upon
changing the solvent from ethyl ether/phosphate buffer to 30% n-propanol in phosphate buffer and increased again to E ]190
upon changing the substrate from 1-acetate to 1-heptanoate. The same changes increased the enantioselectivity towards alcohol
2 from E=17 to 70, but did not significantly increase the enantioselectivity towards alcohol 3. The best solvent was similar to the
solvent used to crystallise the open form of PCL and likely stabilises the open form of PCL. This stabilisation may increase the
enantioselectivity by removing kinetic contributions from a non-enantioselective lid-opening step. We determined the kinetic
contribution of the lid-opening step by measuring the interfacial activation of PCL. The activation energy for the PCL-catalysed
hydrolysis of ethyl acetate was at least 2.6 kcal/mol lower in the presence of a water–organic solvent interface.
© 2003 Elsevier Ltd. All rights reserved.
1. Introduction
example, Miyazawa et al.2 improved the enantioselec-
tivity of lipase AL towards 2-phenoxy-1-propanol 2
from 1.1 to 19 by replacing vinyl trifluoroacetate with
vinyl butanoate as acylating agent. However, for PCL,
vinyl acetate gave higher E (E=42) than the butanoate
(E=35) or the trifluoroacetate (E=1.5).3 Hirose et al.4
improved the resolution of 2-phenyl-1-propanol by
using chiral acyl donors. With lipase QL in hexane
racemic vinyl-3-phenylbutanoate gave E=32, vinyl (R)-
3-phenylbutanoate yielded low enantioselectivity (E=4)
and slow reaction (41 h for 27% conversion), while the
(S) enantiomer provided higher E (E=30) and faster
reaction (1 h for 58% conversion). These results suggest
that the alcohol and acyl chain moieties interact during
enantiorecognition by lipases.
Lipases (triacylglycerol hydrolases, EC 3.1.1.3) usually
successfully resolve chiral secondary alcohols, but reso-
lution of chiral primary alcohols is more difficult due to
low enantioselectivities. Fewer lipases show any enan-
tioselectivity toward primary alcohols and the enan-
tioselectivities are usually low to moderate.1 Such
reactions often require the optimisation of the reaction
conditions to increase the enantioselectivity.
Changing the acyl group often increases the enantiose-
lectivity of lipases towards primary alcohols. Different
acyl groups lead to different structures for the transi-
tion state, which determines the enantioselectivity. For
Changing the solvent also increases the enantioselectiv-
ity, especially for transesterification reactions. Indeed,
transesterification in organic solvents has been more
successful for resolving chiral primary alcohols than the
hydrolysis of the corresponding esters.3–5 Miyazawa et
al.5 increased the enantioselectivity of the lipase AK-
catalysed transesterification of 2 with vinyl butanoate
from 4.5 in benzene to 35 in diisopropyl ether. Hirose
et al.4 improved the enantioselectivity of the lipase
* Corresponding author. Tel.: +1-612-624-5904; fax: +1-612-625-5780;
e-mail: rjk@umn.edu
† Present address: CombinatoRx Inc., 650 Albany Street, Boston, MA
02118, USA.
‡ Present address: Department of Biochemistry, Molecular Biology &
Biophysics and The Biotechnology Institute, University of Minne-
sota, 1479 Gortner Avenue, St. Paul, MN 55108, USA.
0957-4166/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2003.09.049

3918
A. Mezzetti et al. / Tetrahedron: Asymmetry 14 (2003) 3917–3924
SL-catalysed transesterification of 2-phenyl-1-propanol
with racemic vinyl-3-phenylbutanoate from E=40 in
hexane to E=98 in diisopropyl ether. Despite the
numerous attempts to relate the change in enzyme
enantioselectivity to a solvent property such as log P,
trial and error is still the best method to choose the best
solvent.6–9
both the acyl chain length of the substrate and the
reaction conditions. We also verified the rate enhance-
ment of PCL-catalysed hydrolysis in condition of inter-
facial activation using several achiral esters.
2. Results
Herein, we study the PCL-catalysed hydrolysis of three
chiral primary alcohols: 2-methyl-3-phenyl-1-propanol
1, 2-phenoxy-1-propanol 2 and solketal 3. Pure enan-
tiomers of 1 have been used for the preparation of
fungicidal compounds,10 for the synthesis of adenosine
receptor agonists and antagonists11 and for the synthe-
sis of the side chain of zaragozic acid A.12 Pure enan-
tiomers of 2, structurally related to 1, have proven
useful for the synthesis of juvenoids, analogues of the
insect juvenile hormone13 and pure enantiomers of 3
have been used as building blocks in the synthesis of
bioactive compounds such as the b-adrenergic blocker
timolol.14
We investigated how the acyl chain length influences
the enantioselectivity of PCL towards alcohols 1, 2, and
3. We tested the acetate, butanoate and heptanoate of
alcohol 1 and the acetate and heptanoate of alcohols 2
and 3 in potassium phosphate buffer, pH 7.0, contain-
ing 30 vol% of n-propanol (solvent A) (Scheme 1 and
Table 1, entries 1, 6, 7, 9, 10, 12 and 13). A similar
solvent previously promoted the crystallisation of the
open form of PCL.24 Here, however, the concentration
of PCL in solvent A is kept below 10 mg/mL to avoid
its crystallisation while still stabilising the open form.
The concentration of the substrates in all reactions was
above their solubility limit so that some organic phase
was always present to promote interfacial activation of
PCL.
(R)-1 has been prepared in ee >98% by baker’s yeast-
catalysed
reduction
of
(2-dimethoxymethylal-
lyl)benzene.15 High enantioselectivity (E=172) was also
obtained in the Pseudomoas fluorescens lipase-catalysed
transesterification of 1 with vinyl acetate in methylene
chloride.16 Furthermore, while the PCL-catalysed
hydrolysis of 1-acetate proceeded with only moderate
enantioselectivity (E=16),17 the esterification of 1 with
vinyl acetate in chloroform was more successful (E=
11618 and E=12019). Similarly, while the PCL-catalysed
hydrolysis of 2-acetate proceeds with modest enantiose-
lectivity (E=17),20 PCL catalyses the transesterification
of 2 with vinyl acetate in diisopropyl ether with higher
enantioselectivity (E=42).5 The resolution of alcohol 3
has been the subject of extensive research in the past 15
years. The best results have been achieved in the enan-
tioselective oxidation (E >100) of solketal to solketalic
acid by two alcohol dehydrogenases21 and in the lipase
AK-catalysed acylation of 3 with vinyl butanoate at
−40°C (E=55).22 The PCL-catalysed hydrolysis of 3-
benzoate, however, gave low enantioselectivities, rang-
ing from E=3.7 to 9.23
In the case of alcohol 1, PCL showed the highest
enantioselectivity with the heptanoate (E ]190 30,
Table 1, entry 1) and the lowest with the acetate
(E=38 6, Table 1, entry 7). For the butanoate, PCL
exhibited an intermediate enantioselectivity (E=130
30). In the case of alcohol 2 however, the E values were
within experimental error (E=70 20 and E=52 6,
Table 1, entries 9 and 10). We did not detect any
reaction for ( )-3-acetate, while the enantioselectivity of
PCL towards ( )-3-heptanoate (E=5.2 0.1) was signifi-
cantly lower than for substrates 1 and 2 and similar to
what previously reported for the hydrolysis of solketal
benzoate by PCL (E=3.7 to 9).23 For all the alcohols
the reactions were ca. two-fold faster with long acyl
chain esters.
The enantioselectivity for the PCL-catalysed hydrolysis
of 1-heptanoate also varied with the solvent. In bipha-
sic 30 vol% t-butyl methyl ether (t-BME)/potassium
phosphate buffer, pH 7.0 (solvent B) the enantioselec-
tivity dropped to 120 20 and in only potassium phos-
phate buffer, pH 7.0 (solvent C), the enantioselectivity
further dropped to 61 4 (Table 1, entries 4 and 5).
To increase the enantioselectivity of PCL towards esters
of chiral primary alcohols 1, 2 and 3, we optimised
Scheme 1.

A. Mezzetti et al. / Tetrahedron: Asymmetry 14 (2003) 3917–3924
Table 1. PCL-catalysed hydrolysis of esters of 1, 2 and 3 in different solvents
3919
Entry
Substrate
Enzyme source
Solventa
Conv. (%)
Eep (%)
Ees (%)
Eb
1
2
3
4
5
6
7
8
( )-1-Heptanoate
( )-1-Heptanoate
( )-1-Heptanoate
( )-1-Heptanoate
( )-1-Heptanoate
( )-1-Butanoate
( )-1-Acetate
GPCL
A
A
A
B
C
A
A
39.2 0.1
36.4 0.1
35.1 0.1
15.5 0.5
32.0 0.1
17 2
28.0 0.5
44
51.0 0.1
18 3
]98.0 0.2
]98.0 0.2
]98.0 0.2
98.0 0.2
95.0 0.2
98.0 0.2
93 1
79
90 1
95.5 0.4
n.a.
62 2
63.0 0.2
56 5
52.5 0.7
19.0 0.2
45.0 0.2
21 3
35.0 0.3
62
94 1
21 4
n.a.
23 4
–
]190 30 (S)
]180 30 (S)
]170 30 (S)
120 20 (S)
61 4 (S)
130 30 (S)
38 6 (S)
16 (S)
70 20 (S)
52 6 (S)
17 (S)
5.2 0.1 (S)
–
LPL 200S
Lipase PS
GPCL
GPCL
GPCL
GPCL
LPL 200S or PS30
GPCL
GPCL
LPL 200S
GPCL
( )-1-Acetate17
( )-2-Heptanoate
( )-2-Acetate
n.r.c
A
A
B
A
A
9
10
11
12
13
( )-2-Acetate20
( )-3-Heptanoate
( )-3-Acetate
n.a.
27 4
GPCL
n.r.d
–
a A=30 vol% n-propanol in potassium phosphate 0.4 M, pH 7.0; B=30 vol% t-butyl methyl ether in potassium phosphate 0.4 M, pH 7.0;
C=potassium phosphate 0.4 M, pH 7.0
b Determined from eep% and ees% according to Chen et al.;38 enantiomeric purities of the starting materials and products were measured by gas
chromatography directly or after conversion to the 1-acetate or 2-trifluoroacetate; alcohol enantiopreference in parentheses.
c ‘a rapidly stirred suspension of substrate dissolved in a small amount of ether and phosphate buffer (10 mM, pH 7) containing lipase from P.
cepacia…’17
d No enzymatic hydrolysis was observed after 48 h.
All three commercial samples of PCL tested catalysed
the hydrolysis of 1-heptanoate, favouring the (S)-enan-
tiomer, with high enantioselectivity, E ]180 30 (Table
1, average of entries 1–3). PCL from Genzyme Diag-
nostics (GPCL), Amano LPL200S and Amano lipase
PS, all gave E values that were within experimental
error of each other, E ]190 30, E ]180 30 and E
]170 30, respectively.
strate concentration also changes abruptly at the solu-
bility limit. The concentration changes from the soluble
concentration in the aqueous phase to neat substrate
concentration in the insoluble phase. It is not clear how
this concentration change might influence the measured
interfacial activation.
To eliminate the concentration effect, we compared the
rates of hydrolysis of ethyl acetate by PCL at identical
substrate concentrations with and without an interface
present. The ethyl acetate was either dissolved in water
or in a second phase of heptane. The rates with inter-
face were seventy-five times higher than those without
interface (Fig. 1B). The experiment discussed in the
previous paragraphs showed that the rate of hydrolysis
of ethyl acetate increased only 25-folds at the solubility
limit of ethyl acetate. The difference in the two experi-
ments is due to the different organic phase—pure ethyl
acetate or ethyl acetate dissolved in heptane. The hep-
tane interface lowers the activation energy by 2.6 kcal/
mol (RTln75). This value is a lower limit since the
without-interface experiments still contained the
unavoidable air–water interface and the water–reaction
vessel interface. Solubility limits prevented the measure-
ment of the kinetic parameters kcat and Km. Other esters
were not soluble enough in water to measure the with-
out-interface initial rate under pseudo-first order
conditions.
To verify the interfacial activation of PCL, i.e. the rate
enhancement upon changing from a soluble to an insol-
uble substrate form, we measured its reaction rate of
ester hydrolysis in potassium phosphate buffer with
increasing concentration of the substrate for ethyl ace-
tate, isopropenyl acetate, n-pentyl acetate and ethyl
butanoate, as well as for methyl acetate (Fig. 1A).
While methyl acetate, a soluble ester, displayed a steady
increase in reaction rate with increasing concentration,
the other four substrates showed a sharp rate enhance-
ment at a particular substrate concentration. This con-
centration matched the solubility limit of the substrate
as seen by an increase in cloudiness of the solution. The
rate increase ranged from about 10-fold for ethyl
butanoate to about 40-fold for n-pentyl acetate. Isopro-
penyl acetate showed almost a 15-fold increase upon
reaching the solubility limit and ethyl acetate exhibited
a 25-fold increase.
These experiments demonstrate the classical interfacial
activation phenomenon that is characteristic of
lipases,25,26 but only estimate its magnitude. One reason
for an underestimate is that the experiments contained
an interface even at concentrations where the ester was
still soluble. The ester may absorb to the polyethylene
surface of the reaction vessel or to the surface of
bubbles formed by stirring. Since a surface is present
while we measure the soluble rate, this rate is likely
higher than the true soluble rate. A second reason why
the experiments may not be accurate is that the sub-
3. Discussion
The largest change in enantioselectivity, observed dur-
ing the resolution of esters of alcohol 1 (from 38 to
]190, corresponding to a free energy change of ꢀ3.1
kcal/mol in the transition state), occurred upon extend-
ing the ester acetyl group to a heptanoyl group. Weiss-
floch and Kazlauskas17 reported E=16 for the
resolution of 1-acetate in an ether/water mixture. The

3920
A. Mezzetti et al. / Tetrahedron: Asymmetry 14 (2003) 3917–3924
alcohol moiety from binding in this pocket and thus
restricts the flexibility of the alcohol moiety.
The solvent effect on enantioselectivity may involve the
phenomenon of interfacial activation. PCL enantiose-
lectivity towards 1-hepanoate was lowest in aqueous
buffer where the only organic phase present was the
undissolved substrate (E=61, solvent C). Upon adding
t-butyl methyl ether as organic phase the enantioselec-
tivity increased to E=120 (solvent B). Finally, upon
changing to a solvent mixture that favoured crystallisa-
tion of the open form of PCL, its enantioselectivity
increased even further (E ]190, solvent A).
Overbeeke et al.29 suggested a mechanism by which
interfacial activation could contribute to enantioselec-
tivity even though it precedes the transition state for the
chemical reaction. They found a similar increase in
enantioselectivity (from E=8 to 16) for the porcine
pancreatic lipase-catalysed hydrolysis of glycidyl
butanoate when passing from water to a biphasic
water/substrate system, i.e. in conditions of interfacial
activation.30 They showed that when the interfacial
activation is slow (rate similar to that for the chemical
reaction), it will also contribute to the rate of the
overall reaction. Since interfacial activation is pre-
sumably non-enantioselective, this contribution lowers
the overall enantioselectivity. Reaction conditions that
speed up interfacial activation may thus also increase
the enzyme enantioselectivity.29 Increasing the amount
of organic phase (solvent C to B) or using a solvent
mixture that favours the open form of the enzyme
(solvent A) might increase the lipase enantioselectivity
by speeding up its interfacial activation. The higher
reactivity of primary alcohols as compared to sec-
ondary alcohols31 may also make interfacial activation
more important to the enantioselectivity of lipases
towards primary alcohols than towards secondary
alcohols.
Figure 1. Initial rate versus substrate concentration for PCL-
catalysed hydrolysis of several esters. (A) The observed rates
of hydrolysis increase 10 to 40-fold at a particular concentra-
tion (ꢀ50 mM for ethyl butanoate, isopropenyl acetate and
pentyl acetate; ꢀ600 mM for ethyl acetate). These concentra-
tions match the solubility limit of the ester as seen by the
increased cloudiness of the solution. The maximum rate for
isopropenyl acetate was ꢀ300 U/mg, but this is beyond the
y-axis limit of this chart. (B) Initial rate of hydrolysis of ethyl
acetate, either dissolved or present in a separate heptane
phase, as a function of ethyl acetate concentration. At the
same substrate concentration, the initial rate was 75-fold
faster in heptane. This increase is due to interfacial activation.
All experiments used Amano LPL-200S, a purified lipase
preparation.
We also measured the minimum contribution of the
interfacial activation to the reaction rate of PCL-
catalysed hydrolysis of ethyl acetate. At the same sub-
strate concentrations, ethyl acetate in a separate
organic phase reacts at least 75-times faster (]2.6
kcal/mol contribution to the stabilisation of the transi-
tion state) than ethyl acetate dissolved in the aqueous
phase.
fact that PCL exhibited, in all the solvents used, higher
enantioselectivity for 1-heptanoate (E=120 in the very
similar solvent B, E ]190 in solvent A and E=61 in
solvent C) shows not only the important role of the
solvent, but also of the acyl chain length in determining
PCL enantioselectivity.
The value of ]2.6 kcal/mol for the interfacial activa-
tion of PCL is consistent with its proposed molecular
basis. In the closed inactive form,§ the active site is
blocked and a key catalytic residue-Leu17, which sta-
bilises the oxyanion of the tetrahedral intermediate, is
The long heptanoyl chain may reduce the flexibility of
the alcohol moiety in the transition state. X-Ray crystal
structures show that the acyl chain binds to the large
hydrophobic pocket (hydrophobic groove HA) of
PCL.27 This pocket comprises PCL hydrophobic
residues Pro113, Phe119, Leu164, Leu167, Val266 and
Val267. With a short acyl chain like an acetyl, the
substituents from the alcohol moiety can also bind in
this pocket.28 The longer heptanoyl chain prevents the
§ Several groups have solved the X-ray crystal structure of the open
form of PCL,24,32 but not of the closed form. However, we assume
that this structure would be similar to that of Pseudomonas glumae
lipase (PGL), which shares 83% identity with PCL and whose
closed form X-ray crystal structure has been solved.33

A. Mezzetti et al. / Tetrahedron: Asymmetry 14 (2003) 3917–3924
3921
incorrectly oriented. In the open form of PCL,24,32 the
active site is accessible and Leu17 is correctly oriented.
A value of ]2.6 kcal/mol is reasonable for a hydrogen
bond contribution to catalysis, but it is difficult to
predict how much the unblocking of the active site
would contribute to catalysis.
until a constant pH was attained (5 min). Upon addi-
tion of substrate, the recorder was started and the
initial rate measured. The activity of the enzyme was
calculated from these slopes and a [substrate]-rate
profile plotted for each ester assayed. In some cases
with soluble substrates, a magnetic stirrer replaced the
mechanical stirrer.
Another explanation for the increased enantioselectivity
upon changing the solvent is decreased enzyme flexibil-
ity. The increase in enzyme enantioselectivity in organic
solvents has been suggested to be due to decreased
enzyme flexibility.34,35 The lower dielectric constant of
organic solvents as compared to an aqueous environ-
ment increases the strength of the intramolecular elec-
trostatic interactions and thus decreases the flexibility
of the enzyme. The crystallization solvent used in our
experiments may also decrease the flexibility of PCL
since it promotes crystallisation of the lipase open form.
4.2. ( )-2-Methyl-3-phenyl-1-propanol, 1
Lithium aluminium hydride (4.30 g, 110 mmol) was
placed in a flame-dried reaction flask under argon
atmosphere. The flask was cooled to 0°C and anhy-
drous tetrahydrofuran (100 mL) was added. 2-Methyl-
3-phenyl propenal (3.70 g, 25 mmol) dissolved in
anhydrous tetrahydrofuran (25 mL) was added drop-
wise. The reaction was stirred overnight under argon
atmosphere, quenched with a saturated solution of
ammonium chloride, filtered, extracted with ethyl ace-
tate, dried over anhydrous MgSO4 and the solvent was
evaporated under vacuum. The product was purified by
flash chromatography on a silica gel column (eluent: 7:3
hexane:ethyl acetate). Yield 89%. Rf=0.55 (silica gel,
7:3 hexane:ethyl acetate); 1H NMR (CDCl3) l 7.15–
7.30 (m, 5H, Ph), 3.42–3.55 (m, 2H, CH2), 2.75 (dd, 1H,
CH2, J1=6.2 Hz, J2=13.4 Hz), 2.41 (dd, 1H, CH2,
J1=8.2 Hz, J2=13.4 Hz), 1.87–2.00 (m, 1H, CH), 0.91
(d, 3H, CH3, J=6.7 Hz); 13C NMR (CDCl3) l 16.5,
37.9, 39.8, 67.5, 125.9, 128.3, 129.3, 140.9; MS (EI) m/z
(rel. int.) 150 (26, M+), 132 (16, M−H2O), 117 (39,
M+−H2O−CH3), 91 (100, M+−CH(CH3)CH2OH), 77 (8,
Ph).
This is the first time that esters of chiral primary
alcohols 1 and 2 have been resolved with high E via
PCL-catalysed hydrolysis. By the accurate choice of
solvent and acyl chain length, we increased the enan-
tioselectivity of PCL from 1617 to ]190 for the hydrol-
ysis of ( )-1-heptanoate and from 1720 to 70 for the
hydrolysis of ( )-2-heptanoate. Unfortunately this
strategy did not improve the enantioselectivity of PCL
toward alcohol 3. The resolution of esters of chiral
primary alcohols remains more difficult than the resolu-
tion of secondary alcohols, probably due to their
greater flexibility. These results show that the enan-
tioselectivity of PCL is highly substrate dependent and
each primary alcohol must be evaluated individually to
determine the best set of conditions for its resolution.
4.3. ( )-2-Methyl-3-phenyl-1-propyl heptanoate, 1-
heptanoate
4. Experimental
A solution of ( )-2-methyl-3-phenyl-1-propanol (0.50 g,
3.3 mmol) and pyridine (0.42 mL, 5 mmol) in dry ether
(50 mL) was stirred at 4°C. Heptanoyl chloride (0.97
mL, 6.7 mmol) was added dropwise to the solution at
4°C. After 2 h the reaction mixture was washed with a
saturated solution of NaHCO3, water and brine, dried
over anhydrous MgSO4 and the solvent was evaporated
under vacuum. The product was purified by flash chro-
matography on a silica gel column (eluent: 8:2 hex-
ane:ethyl acetate) and isolated as a yellow oil. Yield
All chemicals were purchased from Sigma-Aldrich Co.
(Oakville, ON) unless otherwise specified. Lipases from
Pseudomonas cepacia were purchased from Genzyme
Diagnostics (Cambridge, MA) (2190 U/mg powder)
and Amano Enzyme USA Co., Ltd. (Lombard, IL)
(LPL200S 2260 U/mg powder or Lipase PS ]30 U/mg
powder). ( )-2-Phenoxy-1-propanol was purchased
from TCI America (Portland, OR). Rates of hydrolysis
of achiral substrates were measured by a Radiometer
Copenhagen pH-stat (Copenhagen, Denmark). The chi-
ral column used for the GC analyses was a Chirasil-
DEX CB, 25 m×0.25 mm×0.25 mm (Life Sciences,
Peterborough, ON). The chiral column used for the
HPLC analysis was a Chiracel OD, 25 cm×4.6 mm
(Daicel Chemical Industries, Ltd., Exton, PA). NMR
1
90%. Rf=0.78 (silica gel, 8:2 hexane:ethyl acetate); H
NMR (CDCl3) l 7.07–7.30 (m, 5H, Ph), 3.87–4.00 (m,
2H, CH2), 2.40–2.76 (m, 2H, CH2), 2.31 (t, 2H, CH2,
J=7.4 Hz), 2.04–2.17 (m, 1H, CH), 1.57–1.68 (m, 2H,
CH2), 1.23–1.35 (m, 6H, 3CH2), 0.86–0.93 (m, 6H,
2CH3); 13C NMR (CDCl3) l 14.1, 16.7, 22.6, 25.1, 28.9,
31.6, 34.4, 34.7, 39.9, 68.5, 126.1, 128.3, 129.2, 140.1,
174.0; MS (CI/NH3) m/z (rel. int.) 263 (4, M+H+), 132
(100, M+−C6H13COOH), 91 (44, tropylium), 77 (2, Ph).
1
spectra were collected at 270 Mz for H and 68 MHz
for 13C.
4.1. Interfacial activation
4.4. ( )-2-Methyl-3-phenyl-1-propyl butanoate, 1-
butanoate
Rates of ester hydrolysis were measured by pH-stat in
potassium phosphate buffer (10 mM, pH 7.0), with an
endpoint set at pH 7.15. PCL (Amano LPL200S) in
buffer (10 mL, 0.5 mg lipase/mL) in a polypropylene
cup was stirred vigorously with the mechanical stirrer
A solution of ( )-2-methyl-3-phenyl-1-propanol (0.50 g,
3.3 mmol), butyric anhydride (0.55 g, 3.5 mmol), tri-
ethylamine (0.48 mL, 6.7 mmol) and 4-(dimethyl-
amino)pyridine (10 mg, 0.08 mmol) in methylene

3922
A. Mezzetti et al. / Tetrahedron: Asymmetry 14 (2003) 3917–3924
chloride (25 mL) was stirred at room temperature for 3
h. The reaction mixture was washed with a saturated
solution of NaHCO3, water and brine, dried over anhy-
drous MgSO4 and the solvent was evaporated under
vacuum. The product was purified by flash chromatog-
raphy on a silica gel column (eluent: 8:2 hexane:ethyl
acetate) and isolated as a yellow oil. Yield 89%. Rf=
0.72 (silica gel, 8:2 hexane:ethyl acetate); 1H NMR
(CDCl3) l 7.06–7.30 (m, 5H, Ph), 3.87–4.00 (m, 2H,
CH2), 2.40–2.76 (m, 2H, CH2), 2.29 (t, 2H, CH2, J=7.4
Hz), 2.04–2.17 (m, 1H, CH), 1.59–1.73 (m, 2H, CH2),
0.90–0.98 (m, 6H, 2CH3); 13C NMR (CDCl3) l 13.8,
16.8, 18.6, 34.7, 36.3, 39.9, 68.5, 126.1, 128.4, 129.2,
140.1, 173.8; MS (CI/NH3) m/z (rel. int.) 221 (9, M+H+),
132 (>81, M+−C3H7COOH), 117 (>100, M+-), 91 (>84,
PhCH2), 77 (3, Ph), 71 (27, C4H7O).
(EI) m/z (rel. int.) 194 (8, M+), 101 (89, M+−PhO), 94
(83, PhOH), 77 (19, Ph), 43 (>100, CH3CO).
4.8. ( )-2,2-Dimethyl-[1,3]dioxolan-4-methyl heptanoate,
3-heptanoate
The reaction was performed as for 1-heptanoate. Yield
92% (yellow oil). Rf=0.71 (silica gel, 7:3 hexane:ethyl
1
acetate); H NMR (CDCl3) l 4.23–4.31 (m, 1H, CH),
3.99–4.15 (m, 2H, CH2), 3.67–3.71 (m, 2H, CH2), 2.30
(t, 2H, CH2, J=7.7 Hz), 1.52–1.63 (m, 2H, CH2), 1.39
(s, 3H, CH3), 1.32 (s, 3H, CH3), 1.21–1.30 (m, 6H,
3CH2), 0.84 (t, 3H, CH3, J=6.9 Hz); 13C NMR
(CDCl3) l 14.0, 22.5, 24.9, 25.4, 26.7, 28.8, 31.5, 34.2,
64.5, 66.4, 73.7, 109.8, 173.6; MS (CI/NH3) m/z (rel.
int.) 245 (12, M+1), 229 (61, M+−CH3), 113 (33, M+−
(2,2-dimethyl-[1,3]dioxolan-4-methanol).
4.5. ( )-2-Methyl-3-phenyl-1-propyl acetate, 1-acetate
4.9. ( )-2,2-Dimethyl-[1,3]dioxolan-4-methyl acetate, 3-
acetate
A solution of ( )-2-methyl-3-phenyl-1-propanol (2.50 g,
16.7 mmol), acetic anhydride (1.85 mL, 16.7 mmol),
triethylamine (2.37 mL, 33.4 mmol) and 4-(dimethyl-
amino) pyridine (10 mg, 0.08 mmol) in methylene chlo-
ride (30 mL) was stirred at room temperature for 3 h.
The reaction mixture was washed with a saturated
solution of NaHCO3, water and brine, dried over anhy-
drous MgSO4 and the solvent was evaporated under
vacuum. The product was purified by flash chromatog-
raphy on a silica gel column (eluent: 7:3 hexane:ethyl
acetate) and isolated as a yellow oil. Yield 94%. Rf=
0.76 (silica gel, 7:3 hexane:ethyl acetate); 1H NMR
(CDCl3) l 7.06–7.30 (m, 5H, Ph), 3.85–3.98 (m, 2H,
CH2), 2.40–2.76 (m, 2H, CH2), 2.02–2.11 (m, 1H, CH),
2.01 (s, 3H, CH3), 0.91 (d, 3H, CH3, J=6.7 Hz); 13C
NMR (CDCl3) l 14.3, 16.7, 34.6, 40.1, 68.7, 126.1,
128.4, 129.2, 141.6, 171.1; MS (CI/NH3) m/z (rel. int.)
193 (15, M+H+), 133 (48, M+−CH3COOH), 117 (100,
M+−H2O−CH3), 91 (84, tropylium), 77 (4, Ph).
The reaction was performed as for 1-acetate. Yield 93%
(yellow oil). Rf=0.71 (silica gel, 7:3 hexane:ethyl ace-
tate); 1H NMR (CDCl3) l 4.22–4.31 (m, 1H, CH),
3.99–4.15 (m, 2H, CH2), 3.65–3.70 (m, 2H, CH2), 2.10
(s, 3H, CH3), 1.37 (s, 3H, CH3), 1.29 (s, 3H, CH3); 13C
NMR (CDCl3) l 20.8, 25.4, 26.7, 64.9, 66.3, 73.6,
109.9, 170.8; MS (CI/NH3) m/z (rel. int.) 175 (27,
M+1), 159 (83, M−CH3).
4.10. PCL-catalysed kinetic resolution of esters of alco-
hol 1
Substrates ( )-1-heptanoate, ( )-1-butanoate and ( )-1-
acetate (0.08 mmols) were suspended or dissolved in the
appropriate solvent (3.28 mL) (Table 1). PCL solutions
(0.05 mL) in 0.4 M potassium phosphate buffer at pH
7.0 (GPCL and LPL200S 0.0084 mg/mL, PS 0.2050
mg/mL) were added to the reaction vials. The mixtures
were stirred at room temperature and monitored by
thin layer chromatography (TLC) (hexane:ethyl acetate
8:2) until about 50% conversion (visual estimate). The
reactions were quenched with 2% HCl until pH 2,
extracted with ethyl ether and the solvent evaporated
under vacuum. Starting material ester and product
alcohol were separated by preparative TLC (hex-
ane:ethyl acetate 8:2). The recovered esters 1-hep-
tanoate and 1-butanoate were converted to the
corresponding alcohol 1 by NaOH promoted hydrolysis
in aqueous ethanol (15 w/v% NaOH in 50 vol%
aqueous ethanol) at room temperature. The alcohols
were converted to 1-acetate by pyridine-catalysed ester-
ification using acetic anhydride. The enantiomeric
excesses (ee) of the non-reacted starting materials (1-
heptanoate, 1-butanoate and 1-acetate) and of the
product (1) of the reactions were derived from the ee of
1-acetate measured by gas chromatograpy: 120°C
isothermal for 35 min and raised to 180°C at a rate of
10°C/min, 8.5 psi. Retention times were 31.2 min for
(S)-3-methyl-2-phenyl-1-propyl acetate ((S)-1-acetate)
and 31.9 min for (R)-3-methyl-2-phenyl-1-propyl ace-
tate ((R)-1-acetate); h=1.02. (S)-1 [h]2D0=−11.3 (c
0.124, C6H6), lit.36 [h]D20=−11.1 (c 4.6, C6H6).
4.6. ( )-2-Phenoxy-1-propyl heptanoate, 2-heptanoate
The reaction was performed as for 1-heptanoate. Yield
89% (yellow oil). Rf=0.77 (silica gel, 8:2 hexane:ethyl
1
acetate); H NMR (CDCl3) l 6.90–7.30 (m, 5H, Ph),
4.54–4.65 (m, 1H, CH), 4.12–4.29 (m, 2H, CH2), 2.29
(t, 2H, CH2, J=7.4 Hz), 1.53–1.68 (m, 2H, CH2),
1.22–1.36 (m, 9H, 3CH2+CH3), 0.86 (t, 3H, CH3, J=
5.2 Hz); 13C NMR (CDCl3) l 14.1, 16.9, 22.5, 24.9,
28.8, 31.5, 34.3, 66.8, 71.8, 116.2, 121.2, 129.6, 157.8,
173.8; MS (EI) m/z (rel. int.) 264 (4, M+), 171 (100,
M+−PhO), 113 (27, C6H13CO), 94 (20.6, PhOH), 77 (9,
Ph).
4.7. ( )-2-Phenoxy-1-propyl acetate, 2-acetate
The reaction was performed as for 1-acetate. Yield 91%
(yellow oil). Rf=0.59 (silica gel, 7:3 hexane:ethyl ace-
tate); 1H NMR (CDCl3) l 7.23–7.30 (m, 2H, Ph),
6.88–6.95 (m, 3H, Ph), 4.55–4.66 (m, 1H, CH), 4.25
(dd, 1H, CH2, J1=6.4 Hz, J2=11.6 Hz), 4.15 (dd, 1H,
CH2, J1=4.3 Hz, J2=11.6 Hz), 2.06 (s, 3H, CH3), 1.32
(d, 3H, CH3, J=6.2 Hz); 13C NMR (CDCl3) l 16.7,
20.9, 67.0, 71.8, 116.2, 121.3, 129.6, 157.8, 171.0; MS

A. Mezzetti et al. / Tetrahedron: Asymmetry 14 (2003) 3917–3924
3923
4.11. PCL-catalysed kinetic resolution of esters of alco-
hol 2
2. Miyazawa, T.; Yukawa, T.; Koshiba, T.; Ueji, S; Yanagi-
hara, R.; Yamada, T. Biotechnol. Lett. 2001, 23, 1547–
1550.
Kinetic resolutions were carried out as for the esters of
alcohol 1. Starting ester and product alcohol were
separated by preparative TLC (hexane:ethyl acetate
8:2). The recovered esters 2-heptanoate and 2-acetate
were converted to the corresponding alcohol 2 by
NaOH promoted hydrolysis in aqueous ethanol (15%
NaOH in 50 vol% aqueous ethanol) at room tempera-
ture. The alcohols thus obtained were derivatised to the
corresponding trifluoroacetates by pyridine-catalysed
esterification using trifluoroacetic anhydride. The enan-
tiomeric excesses (ee) of the non-reacted starting mate-
rials 2-heptanoate and 2-acetate and of the product (2)
of the reactions were derived from the ee of the trifl-
uoroacetates measured by gas chromatograpy: 120°C
isothermal for 10 min and raised to 180°C at a rate of
10°C/min, 10 psi. Retention times were 10.3 min for
(S)-2-phenoxy-1-propyl trifluoroacetate and 10.5 min
for (R)-2-phenoxy-1-propyl trifluoroacetate; h=1.03.
The absolute configuration of the resolved alcohol was
determined by oxidising the alcohol product of a
kinetic resolution to the corresponding carboxylic acid
by means of the Jones’ reagent and analysing it by
HPLC using hexane:isopropanol:formic acid (90:10:1)
as mobile phase. Retention times were 10 min for
(S)-2-phenoxypropionic acid and 16.1 min for (R)-2-
phenoxypropionic acid; h=1.61.37
3. Miyazawa, T.; Yukawa, T.; Ueji, S.; Yanagihare, R.;
Yamada, T. Biotechnol. Lett. 1998, 20, 235–238.
4. Hirose, K; Naka, H.; Yano, M.; Ohashi, S.; Naemura,
K.; Tobe, Y. Tetrahedron: Asymmetry 2000, 11, 1199–
1210.
5. Miyazawa, T.; Yukawa, T.; Koshiba, T.; Sakamoto, H.;
Ueji, S.; Yanagihara, R.; Yamada, T. Tetrahedron:
Asymmetry 2001, 12, 1595–1602.
6. Fitzpatrick, P. A.; Klibanov, A. M. J. Am. Chem. Soc.
1991, 113, 3166–3171.
7. Ducret, A.; Trani, M.; Lortie, R. Enzyme Microb. Tech-
nol. 1998, 22, 212–216.
8. Carrea, G.; Ottolina, G.; Riva, S. Trends Biotechnol.
1995, 13, 63–70 (erratum 122).
9. Wescott, C. R.; Klibanov, A. M. Biochim. Biophys. Acta
1994, 1206, 1–9.
10. Martin, C.; Himmele, W.; Siegel, H. US Pat. 4,410,734,
1983.
11. Delinck, D. L.; Margolin, A. L. Tetrahedron Lett. 1990,
31, 6797–6798.
12. Stoermer, D.; Caron, S.; Heathcock, C. H. J. Org. Chem.
1996, 61, 9115–9125.
13. Wimmer, Z.; Saman, D.; Francke, W. Helv. Chim. Acta
1994, 77, 502–508.
14. Weinstock, L. M.; Mulvey, D. M.; Tull, R. J. Org. Chem.
1976, 41, 3121–3124.
15. Ferraboschi, P; Reza-Elahi, S; Verza, E; Santaniello, E.
Tetrahedron: Asymmetry 1999, 10, 2639–2642.
16. Ferraboschi, P; Casati, S.; De Grandi, S.; Grisenti, P.;
Santaniello, E. Biocatalysis 1994, 10, 279–288.
17. Weissfloch, A. N. E.; Kazlauskas, R. J. J. Org. Chem.
1995, 60, 6959–6969.
18. Nordin, O.; Nguyen, B.-V.; Vo¨rde, C.; Hedenstrom, E.;
Ho¨gberg, H.-E. J. Chem. Soc., Perkin Trans. 1 2000,
367–376.
19. Ho¨gberg, H.-E.; Berglund, P.; Edlund, H.; Fggerhag, J.;
Hedenstrijm, E; Lundh, M; Nordin, O.; Servi, S; Vo¨rde,
C. Catal. Today 1994, 22, 591–606.
20. Tuomi, W. V.; Kazlauskas, R. J. J. Org. Chem. 1999, 64,
2638–2647.
21. Geerlof, A.; van Tol, B. A.; Jongejan, J. A.; Duine, J. A.
Biosci. Biotechnol. Biochem. 1994, 58, 1028–1036.
22. Sakai, T.; Kishimoto, T.; Tanaka, Y.; Ema, T.; Utaka,
M. Tetrahedron Lett. 1998, 39, 7881–7884.
23. Bosetti, A.; Bianchi, D.; Cesti, P.; Golini, P. Biocatalysis
1994, 9, 71–77.
24. Schrag, J. D.; Li, Y.; Cygler, M.; Lang, D.; Burgdorf, T.;
Hecht, H.-J.; Schmid, R.; Schomburg, D.; Rydel, T. J.;
Oliver, J. D.; Strickland, L. C.; Dunaway, C. M.; Larson,
S. B.; Day, J.; McPherson, A. Structure 1997, 5, 187–202.
25. Brzozowski, A. M.; Derewenda, U.; Derewenda, Z. S.;
Dodson, G. G.; Lawson, D. M.; Turkenburg, J. P.;
Bjorkling, F.; Huge-Jensen, B.; Patkar, S. A.; Thim, L.
Nature 1991, 351, 491–494.
4.12. PCL-catalysed kinetic resolution of esters of alco-
hol 3
The reactions were performed as for the esters of
alcohol 1 and were submitted to gas chromatographic
analysis to determine the enantiomeric excesses of non-
reacted starting materials and products using 100°C for
10 min, then raised to 180°C at a rate of 5°C/min, 8 psi.
Retention times were 15.2 min for (S)-3, 15.0 min for
(R)-3 (h=1.02), 28.3 min for (S)-3-heptanoate and 28.5
min for (R)-3-heptanoate (h=1.01). The retention time
of the enantiomers was confirmed by comparison with
commercially available samples of pure enantiomers of
(R)-3 and (S)-3.
Acknowledgements
A.M. thanks Les Fonds Que´be´cois de la Recherche sur
la Nature et les Technologies (FCAR) for two post-
graduate fellowships. We thank the National Sciences
and Engineering Research Council of Canada
(NSERC) for financial support and Dr. Eniko¨ Forro
for her work on the PCL-catalysed transesterifications
of alcohols 1 and 2 (unpublished results).
26. Derewenda, U.; Brzozowski, A. M.; Lawson, D. M.;
Derewenda, Z. S. Biochemistry 1992, 31, 1532–1541.
27. Lang, D. A.; Mannesse, M. L. M.; de Haas, G. H.;
Verheij, H. M.; Dijkstra, B. W. Eur. J. Biochem. 1998,
254, 333–340.
References
1. Bornscheuer, U. T.; Kazlauskas, R. J. Hydrolases in
Organic Synthesis; Wiley-VCH: Weinheim, Germany,
1999; pp. 65–106.

3924
A. Mezzetti et al. / Tetrahedron: Asymmetry 14 (2003) 3917–3924
&
28. Luic´, M.; Tomic´, S.; Lesˇcˇic´, I.; Ljubovic´, E.; Sepac, D.;
33. Lang, D.; Hofmann, B; Haalck, L; Hecht, H. J.; Spener,
F.; Schmid, R. D.; Schomburg, D. J. Mol. Biol. 1996, 259,
704–717.
&
Sunjic´, V.; Vitale, L.; Saenger, W.; Kojic´-Prodic´, B. Eur.
J. Biochem. 2001, 268, 3964–3973.
34. Affleck, R.; Haynes, C. A.; Clark, D. S. Proc. Natl. Acad.
Sci. USA 1985, 82, 5167–5170.
35. Klibanov, A. Trends Biotechnol. 1997, 15, 97–101.
36. Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem.
Soc. 1982, 104, 1737–1739.
29. Overbeeke, P. L. A.; Govardhan, C.; Khalaf, N.; Jongejan,
J. A.;Heijnen, J. J. J. Mol. Cat. B:Enzym. 2000, 10, 385–393.
30. Van Tol, J. B. A.; Jongejan, J. A.; Duine, J. A. Biocatal.
Biotransform. 1995, 12, 99–118.
31. Nishizawa, K.; Ohgami, Y.; Matsuo, N.; Kisida, H.;
Hirohara, H. J. Chem. Soc., Perkin Trans. 2 1997, 1293–
1298.
37. Wu, S.-H.; Lay, S.-Y.; Lin, S.-L.; Chu, F.-Y.; Wang, K.-T.
Chirality 1991, 3, 67–70.
38. Chen, C.-S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am.
Chem. Soc. 1982, 104, 7294–7299.
32. Kim, K. K.; Song, H. K.; Shin, D. H.; Hwang, K. Y.; Suh,
S. W. Structure 1997, 5, 173–185.