Activation of pig liver esterase in organic media with organic polymers. Application to the enantioselective acylation of racemic functionalized secondary alcohols

Article abstract of DOI:10.1021/jo0016881

Pig liver esterase (PLE) shows practically no activity in acylation of alcohols with vinylic esters in organic solvents. However, addition of methoxypoly(ethylene glycol) (MPEG), bovine serum albumin (BSA), TentaGelAmino resin (TGA), or aminomethyl polystyrene (AMPS) confers activity to PLE in acylation of alcohols with vinyl propionate in organic solvents of low water content. Polymer-activated PLE showed high enantioselectivities (E > 100) in the acylation of racemic 1-alkoxy-, 1-ethylsulfanyl-, and 1-fluoro-3-aryl-2-propanols as well as racemic 1-phenoxy-2-propanol and racemic 1-methoxy-2-phenoxy-2-propanol. The synthetic utility of polymer-activated PLE has been demonstrated by the gram-scale resolution of 1-methoxy-3-phenyl-2-propanol, 1-ethylsulfanyl-3-phenyl-2-propanol, 1-methoxy-3-p-methoxyphenyl-2-propanol, 1- fluoro-3-phenyl-2-propanol, and 1-methoxy-3-phenoxy-2-propanol. In PLE-catalyzed acylation of alcohols with vinyl propionate, acetaldehyde and propionic acids, both being detrimental to the enzyme, are formed as byproducts. In addition, the water content of the system, which is critical for the activity of pig liver esterase, is lowered because of a competing enzymatic hydrolysis of the acyl donor. The polymers TGA, BSA, and AMPS not only scavenge the aldehyde and the acid through imine formation and neutralization, respectively, but replenish at least in part also the water consumed in the competing hydrolysis of the acyl donor. A recovery of PLE together with the polymer was achieved without major loss of activity through their immobilization on a water-saturated polyaramide membrane, which occurs spontaneously in organic solvents.

Full text of DOI:10.1021/jo0016881

3384
J . Org. Chem. 2001, 66, 3384-3396
Activa tion of P ig Liver Ester a se in Or ga n ic Med ia w ith Or ga n ic
P olym er s. Ap p lica tion to th e En a n tioselective Acyla tion of
Ra cem ic F u n ction a lized Secon d a r y Alcoh ols
Hans-J oachim Gais,* Manfred J ungen, and Vasudev J adhav
Institut fu¨r Organische Chemie der Rheinisch-Westfa¨lischen Technischen Hochschule Aachen,
Professor-Pirlet-Strasse 1, D-52056 Aachen, Germany
Gais@RWTH-Aachen.de
Received December 1, 2000
Pig liver esterase (PLE) shows practically no activity in acylation of alcohols with vinylic esters in
organic solvents. However, addition of methoxypoly(ethylene glycol) (MPEG), bovine serum albumin
(BSA), TentaGelAmino resin (TGA), or aminomethyl polystyrene (AMPS) confers activity to PLE
in acylation of alcohols with vinyl propionate in organic solvents of low water content. Polymer-
activated PLE showed high enantioselectivities (E > 100) in the acylation of racemic 1-alkoxy-,
1-ethylsulfanyl-, and 1-fluoro-3-aryl-2-propanols as well as racemic 1-phenoxy-2-propanol and
racemic 1-methoxy-2-phenoxy-2-propanol. The synthetic utility of polymer-activated PLE has been
demonstrated by the gram-scale resolution of 1-methoxy-3-phenyl-2-propanol, 1-ethylsulfanyl-3-
phenyl-2-propanol, 1-methoxy-3-p-methoxyphenyl-2-propanol, 1-fluoro-3-phenyl-2-propanol, and
1-methoxy-3-phenoxy-2-propanol. In PLE-catalyzed acylation of alcohols with vinyl propionate,
acetaldehyde and propionic acids, both being detrimental to the enzyme, are formed as byproducts.
In addition, the water content of the system, which is critical for the activity of pig liver esterase,
is lowered because of a competing enzymatic hydrolysis of the acyl donor. The polymers TGA, BSA,
and AMPS not only scavenge the aldehyde and the acid through imine formation and neutralization,
respectively, but replenish at least in part also the water consumed in the competing hydrolysis of
the acyl donor. A recovery of PLE together with the polymer was achieved without major loss of
activity through their immobilization on a water-saturated polyaramide membrane, which occurs
spontaneously in organic solvents.
In tr od u ction
enzyme stability in organic solvents.7-14 One important
factor is the water activity and the nature of the organic
solvent. Another one is the formulation of the enzyme.
Enzymes are generally not soluble in organic solvents.
Therefore, catalysis is carried out under heterogeneous
conditions, which restrict not only the mobility of the
enzyme but also of the substrate and products and can,
thus, cause mass transfer limitations. Immobilization on
solid support, addition of hydrated salts, covalent attach-
ment of methoxpoly(ethylene glycol) (MPEG) residues,
lyophilization with organic polymers, cross-linking of
enzyme crystals, sol-gel entrapment, coating with am-
phiphilic molecules, and entrapment in reverse micelles
are the most important formulations that have been
shown to improve the performance of the enzyme.5,7-14
Pig liver esterase (PLE) is one of the most important
enzymes for the enantioselective hydrolysis of prochiral
or racemic esters in aqueous solution. It has, therefore,
seen many applications in asymmetric synthesis.1-6
However, despite its obvious synthetic potential, the
reverse transformation, namely the PLE-catalyzed enan-
tioselective acylation of prochiral or racemic alcohols in
organic solvents, has not gained the importance of the
lipase-catalyzed acylation method.3-5,7 This is due to the
fact that PLE shows only a very low activity in organic
solvents. The esterase differs in this respect considerably
from lipases, which are highly active in organic media.
Many factors have been found to affect the activity and
(1) Ohno, M.; Otsuka, M. Org. React. 1989, 37, 1 and references
therein.
(2) Zhu, L.-M.; Tedford, M. C. Tetrahedron 1990, 46, 6587 and
references therein.
(3) Gais, H.-J .; Elferink, V. H. M. In Enzyme Catalysis in Organic
Synthesis; Drauz, K., Waldmann, H., Eds.; VCH: Weinheim, 1995; Vol.
I, p 165 and references therein.
(4) Biocatalysis for Fine Chemical Synthesis; Roberts, S. M., Casy,
G., Nielsen, M.-B., Wiggins, K., Eds.; Wiley: New York, 1999.
(5) Bornscheuer, U. T.; Kazlauskas, R. J ., Hydrolases in Organic
Synthesis; Wiley-VCH: Weinheim, 1999 and references therein.
(6) For selected recent examples, see: (a) Gais, H.-J .; Griebel, C.;
Buschmann, H. Tetrahedron: Asymmetry 2000, 11, 917. (b) Sousa, H.
A.; Crespo, J . G.; Alfonso, C. A. M. Tetrahedron: Asymmetry 2000,
11, 929. (c) Liang, X.; Lohse, A.; Bols, M. J . Org. Chem. 2000, 65, 7432.
(d) Yu, M. S.; Lantos, I.; Peng, Z.-Q.; Yu, J .; Cacchio, T. Tetrahedron
Lett. 2000, 41, 5647. (e) Beumer, R.; Bubert, C.; Cabrele, C.; Vielhauer,
O.; Pietzsch, M.; Reiser, O. J . Org. Chem. 2000, 65, 8960.
(7) Theil, F. Tetrahedron 2000, 56, 2905 and references therein.
(8) Kise, H. In Synthesis of Biocomposite Materials; Imanishi, Y.,
Ed.; CRC Press: Boca Raton, 1992; p 183.
(9) Adlercreutz, P. In Enzymatic reactions in organic media; Koski-
nen, A. M. P., Klibanov, A. M., Eds.; Blackie Academic & Profes-
sional: London, 1996; p 9.
(10) Yang, Z.; Russell, A. J . In Enzymatic reactions in organic media;
Koskinen, A. M. P., Klibanov, A. M., Eds.; Blackie Academic &
Professional: London, 1996; p 43.
(11) Zaks, A. In Enzymatic reactions in organic media; Koskinen,
A. M. P., Klibanov, A. M., Eds.; Blackie Academic & Professional:
London, 1996; p 70.
(12) Carrea, G.; Riva, S. Angew. Chem. 2000, 112, 2312; Angew.
Chem., Int. Ed. 2000, 39, 2226 and references therein.
(13) Poly(Ethylene Glycol) Chemistry; Harris, J . M., Ed.; Plenum
Press: New York, 1992.
(14) Poly(ethylene glycol) Chemistry and Biological Applications;
Harris, J . M., Zalpinsky, S., Eds.; American Chemical Society: Wash-
ington, 1997.
10.1021/jo0016881 CCC: $20.00 © 2001 American Chemical Society
Published on Web 04/19/2001

Acylation of Racemic Functionalized Secondary Alcohols
J . Org. Chem., Vol. 66, No. 10, 2001 3385
Sch em e 1. P LE/MP EG- a n d P LE/BSA-Ca ta lyzed
Acyla tion of Hom oben zylic Alcoh ols, F lu or in a ted
Alcoh ols, a n d Meth oxy a n d P h en oxy Gr ou p
Bea r in g Alcoh ols
Attempts to confer activity to PLE in organic solvents
by entrapment in water-filled porous supports,15,16 cova-
lent attachment of MPEG residues,17,18 immobilization
on carrageenan19 and Eupergit,20 or entrapment in
polymers21 were met with various degrees of success.22
Recently, we have shown that the simple addition of
MPEG to PLE through colyophilization significantly
enhances the activity and stability of the enzyme in
organic solvents of low to medium polarity and low water
content.23-25 The colyophilisate (CLP) of PLE and MPEG
was successfully applied to the kinetic resolution of
racemic glycerol derivatives, which bear a primary hy-
droxy group, through acylation with vinyl and isoprope-
nyl esters in toluene containing less than 1% water (vide
infra). We noticed, however, in these experiments also a
decrease of the activity of PLE in organic solvents with
increasing reaction time. This time dependency of the
activity of the enzyme had been attributed in part to a
gradual lowering of the critical water content of the
system through a competitive enzymatic hydrolysis of the
vinylic esters under formation of the corresponding
carboxylic acid and carbonyl byproduct.23,24,26 In addition,
earlier studies on the influence of the carbonyl byproducts
acetaldehyde and acetone upon PLE had shown that they
are detrimental to the activity of the enzyme.24 Deactiva-
tion of PLE by acetaldehyde, which may be caused at
least in part by imine formation with the lysine amino
groups,27 was particularly strong. Besides acetaldehyde,
the carboxylic acid being formed in the aforementioned
hydrolysis seems to cause also a deactivation of PLE in
organic solvents.23 Because of the considerable potential
of PLE for enantioselective acylation, it was thus desir-
able to conduct further studies on its polymer activation
in organic solvents in order to circumvent the aforemen-
tioned problem. Besides the question of a further activa-
tion of PLE that of the substrate specificity of the enzyme
in organic solvents was especially important. We were
particularly interested to see whether secondary alcohols
are also substrates for PLE.24 In the present contribution
we report the results of our studies designed to address
(15) Cambou, B.; Klibanov, A. M. J . Am. Chem. Soc. 1984, 106, 2687.
(16) Kirchner, G.; Scollar, M. P.; Klibanov, A. M. J . Am. Chem. Soc.
1985, 107, 7072.
(17) Heiss, L.; Gais, H.-J . Tetrahedron Lett. 1995, 36, 3833.
(18) Ohya, Y.; Sugitou, T.; Ouchi, T. J . Macromol. Sci., Pure Appl.
Chem. 1995, A32, 179.
(19) Kitchell, B. S.; Henkens, R. W.; Brown, P.; Baldwin, S. W.;
Lochmuller, C. H.; O’Daly, J . P. US 5,262,313, 1993; Chem. Abstr. 1994,
120, 157537.
(20) (a) West, J . B.; Wong, C.-H. Tetrahedron Lett. 1987, 28, 1629.
(b) Tacke, R.; Wagner, S. A.; Sperlich, J . Chem. Ber. 1994, 127, 639.
(21) Tor, R.; Dror, Y.; Freeman, A. Enzyme Microb. Technol. 1990,
12, 299.
(22) For further examples, see: (a) Zaks, A.; Klibanov, A. M. J . Am.
Chem. Soc. 1986, 108, 2767. (b) Barton, P.; Laws, A. P.; Page, M. I. J .
Chem. Soc., Perkin Trans. 2, 1994, 2021. (c) Brown, C.; Marston, R.
W.; Quigley, P. F.; Roberts, S. M. J . Chem. Soc., Perkin Trans. 1 2000,
1809.
(23) Ruppert, S.; Gais, H.-J . Tetrahedron: Asymmetry 1997, 8, 3657.
(24) J ungen, M.; Gais, H.-J . Tetrahedron: Asymmetry 1999, 10, 3747.
(25) For studies of the activity of lipase/MPEG in organic solvents,
see: (a) Ottolina, G.; Carrea, G.; Riva, S.; Sartore, L.; Veronese, F. M.
Biotechnol. Lett. 1992, 14, 947. (b) Matsushima, A.; Kodera, Y.; Hiroto,
M.; Nishimura, H.; Inada, Y. J . Mol. Catal. B 1996, 2, 1. (c) Secundo,
F.; Spadaro, S.; Carrea, G. Biotechnol. Bioeng. 1999, 62, 554.
(26) For the observation of lipase catalyzed hydrolysis of vinylic
esters, see: Weber, H. K.; Weber, H.; Kazlauskas. R. J . Tetrahedron:
Asymmetry 1999, 10, 2635.
(27) For imine formation of lipases with acetaldehyde, see: (a)
Berger, B.; Faber, K. J . Chem. Soc., Chem. Commun. 1991, 1198. (b)
Weber, H. K.; Stecher, H.; Faber, K. Biotechnol. Lett. 1995, 17, 873.
(c) Weber, H. K.; Zuegg, J .; Faber, K.; Pleiss, J . J . Mol. Catal. B 1997,
3, 131.
these issues. We show that PLE can be (i) activated in
organic solvents with a number of special organic poly-
mers, (ii) applied successfully as mixture with these
polymers to the enantioselective acylation of racemic
secondary alcohols, and (iii) recovered and reused.
Resu lts a n d Discu ssion
For the determination of the influence of the various
activators on the activity and the enantioselectivity of
PLE in organic solvents, the racemic functionalized
2-propanols rac-1-4, rac-6, and rac-7 were selected as
potential substrates. These alcohols, which all carry a
benzyl group at the sterogenic center (Scheme 1), were
chosen for the following reasons. Previous results ob-
tained in the acylation of rac-4 catalyzed by PLE in the
presence of MPEG (termed in the following PLE/MPEG-
catalyzed acylation) seemed to indicate a particular

3386 J . Org. Chem., Vol. 66, No. 10, 2001
Gais et al.
Ta ble 1. P LE/MP EG-Ca ta lyzed Acyla tion of Alcoh ols
affinity of PLE in organic solvents toward homobenzylic
alcohols.24 Second, the aforementioned alcohols in enan-
tiomerically enriched form are of synthetic value.28-31
Third, their lipase catalyzed kinetic acylation has to the
best of our knowledge not been described. Included into
this study were the racemic alcohols rac-5 and rac-8-
10. These alcohols were selected because of their struc-
tural resemblance to the aforementioned alcohols and
because of the synthetic value of the enantiomerically
enriched compounds.32-42 Of the latter alcohols only rac-
932,36 and rac-1042 had been submitted to a lipase-
catalyzed acylation in organic solvents. Native PLE is a
mixture of isoenzymes having six major components.3,5
Commercially available are the native mixture and three
different enriched isoenzyme mixtures.43 In the over-
whelming number of cases reported in the literature thus
far the native isoenzyme mixture has been applied for
enantioselective synthesis.1-6 We have, thus, used in the
work described herein the native isoenzyme mixture
because of a better comparison with previous results from
our and other laboratories.
P ig Liver Ester a se/Meth oxyp oly(eth ylen e glycol).
Since the colyophilisate PLE/MPEG had shown promise
for the enantioselective acylation of racemic primary
alcohols in organic solvents, it was of interest to see
whether secondary alcohols are also amenable to a PLE/
MPEG-catalyzed acylation. In addition, possible activity
and selectivity improvements of PLE/MPEG were sought.
The commercially available suspension of PLE in am-
monium sulfate was first desalted by ultrafiltration in
order to exclude any salt effect on the activity and
selectivity of the enzyme in organic solvents (vide infra).
Lyophilization of desalted PLE (190 U/mg, 30 mg) and
MPEG5000 (1 g) in aqueous solution gave PLE/MPEG as
a fine, free-flowing, white powder, which contained less
than 1% water according to a Karl Fischer titration. The
enzyme had a specific activity of 170 U/mg as determined
by hydrolysis of ethyl butyrate in aqueous solution.
Storage of the colyophilisate (CLP) at 4 °C for several
weeks did not lead to a decrease of enzyme activity. Thus,
r a c-1-10a
t
convn alcohol ester
entry substrate solvent (d) (%) ee (%) ee (%)
Eb
1
2c
3
4
5
6
7
8
9
rac-1
rac-2
rac-3
rac-3
rac-4
rac-4
rac-5
rac-6
rac-6
rac-7
rac-8
rac-9
toluene
toluene
toluene 0.8
n-octane
toluene
n-octane
toluene
toluene
n-octane
toluene
4
4
47
4
8
44
39
51
3
16
39
1
90
5
8
64
73
99
2
21
47
d
98 >100 (307)
>98 >100
>98 >100
2
2
2
2
2
3
4
93
53
99 >100 (437)
94 >100 (170)
61
4
98 >100
94
d
51
d
10
11
12
13
toluene 1.1
toluene
21
26
47
26
39
90
65
6
2
5
>98 >100
97 >100
rac-10 toluene
a
Conditions: 1 mmol of alcohol, 4 mmol of vinyl propionate,
b
300 U of PLE, 10 mL of solvent containing 0.4% water. Numbers
in parentheses are the calculated values. c The substrate concen-
d
tration was 0.025 mol/L. Not determined.
preparation of PLE/MPEG was accompanied only by an
approximately 10% loss of the activity of the enzyme. We
noticed that MPEG acts as a lyoprotectant since lyo-
philization of PLE in aqueous solution without the
polymer was accompanied by a 30% to 50% activity loss.
An activity preserving effects of MPEG and PEG upon
enzymes had been noticed before in other cases than
PLE.8-14
PLE/MPEG-catalyzed acylation of rac-1-10 was gen-
erally carried out by dissolving the alcohol in an organic
solvent followed by the addition of vinyl propionate as
the acyl donor. Then, the water content of the system
was adjusted to a total of 0.43% (v/v) whereby a tiny
second aqueous phase was formed except in the case of
methyl tert-butyl ether (MTBE) and dimethoxyethane
(DME). Finally, additional solvent and PLE/MPEG,
which dissolved mainly in the aqueous phase, were
added. Because of the solubility of MPEG in some of the
solvents used, part of the polymer may have entered the
organic phase. Acylation was started by rapid stirring of
the two-phase mixture, which caused the aqueous phase
to disperse into small droplets. With increasing reaction
time the second liquid phase disappeared, however, and
solid PLE/MPEG deposited.
High enantioselectivities were recorded in the case of
the 3-phenyl-2-propanols rac-1, rac-2, rac-3, and rac-4,
the fluorinated 3-phenyl-2-propanol rac-6, and the 3-phe-
noxy-2-propanols rac-9 and rac-1024 (cf. Scheme 1) by
using toluene or n-octane as solvent (Table 1). All
reactions are characterized by a selectivity factor E
(enantiomer ratio) being in all cases except two greater
than 100.44 In the case of alcohol rac-1 the point of 47%
conversion was reached after 4 d using toluene as solvent.
The alcohol (R)-1 and the ester (S)-1a were formed with
high selectivity as revealed by an E value of greater than
100 (Table 1, entry 1). Acylation of rac-2, which bears a
isopropoxy group, by using PLE/MPEG proceeded also
with high selectivity. The activity of the enzyme in
toluene toward this substrate was, however, lower than
in the case of rac-1 (Table 1, entry 2). The activity loss
(28) Ferraboschi, P.; Grisenti, P.; Manzocchi, A.; Santaniello, E.
Tetrahdron 1994, 50, 10539.
(29) Bandini, M.; Cozzi, P. G.; De Angelis, M.; Umani-Ronchi, A.
Tetrahedron Lett. 2000, 41, 1601.
(30) Kitazume, T.; Asai, M.; Lin, J . T.; Yamazaki, T. J . Fluor. Chem.
1987, 35, 477.
(31) Bravo, P.; Piovosi, E.; Resnati, G. J . Chem. Soc., Perkin Trans.
1 1989, 1201.
(32) Hoff, B. H.; Anthonsen, H. W.; Anthonsen, T. Tetrahedron:
Asymmetry 1996, 7, 3187.
(33) Waagen, V.; Partali, V.; Hollingsaeter, I.; Huang, M. S. S.;
Anthonsen, T. Acta Chem. Scand. 1994, 48, 506.
(34) Hoff, B. H.; Anthonsen, T. Chirality 1999, 11, 760.
(35) Heuer, C.; Kusters, E.; Plattner, T.; Seidel-Morgenstern, A. J .
Chromatogr. A 1998, 827, 175.
(36) Miyazawa, T.; Yukawa, T.; Ueji, S.; Yanagihara, R.; Yamada,
T. Biotechnol. Lett. 1998, 20, 235.
(37) Yuan, R.; Watanabe, S.; Kuwabata, S.; Yoneyama, H. J . Org.
Chem. 1997, 62, 2494.
(38) Yanase, H.; Arishima, T.; Kita, K.; Nagai, J .; Mizuno, S.;
Takuma, Y.; Endo, K.; Kato, N. Biosci. Biotech. Biochem. 1993, 57,
1334.
(39) Ishikawa, A.; Shibata, T. J . Liq. Chromatogr. 1993, 16, 859.
(40) Fukui, Y.; Ichida, A.; Shibata, T.; Mori, K. J . Chromatogr. A
1990, 515, 85.
(41) Waagen, V.; Hollingsaeter, I.; Partali, V.; Thorstad, O.; An-
thonsen, T. Tetrahedron: Asymmetry 1993, 4, 2265.
(42) Lemke, K.; Lemke, M.; Theil, F. J . Org. Chem. 1997, 62, 6268.
(43) Roche Diagnostics, Germany, offers besides PLE as an isoen-
zyme mixture PLE E-1 and PLE E-2 both as enriched isoenzyme
mixtures. Fluka, Germany, sells PLE as an isoenzyme mixture and
PLE isoenzyme 1 as an enriched isoenzyme mixture, and Sigma offers
PLE as an isoenzyme mixture.
(44) The E values were calculated according to Chen and Sih (Chen,
C. S.; Sih, C. J . Angew. Chem. 1989, 101, 711; Angew. Chem., Int. Ed.
Engl. 1989, 28, 695) by using the equations for an irreversible reaction
although reversibility of acylation under the conditions employed
cannot be rigorously excluded. E values calculated to be greater than
100 are given as E > 100 because of the reduced accuracy of the higher
values.

Acylation of Racemic Functionalized Secondary Alcohols
J . Org. Chem., Vol. 66, No. 10, 2001 3387
F igu r e 1. PLE/MPEG-catalyzed acylation of rac-1-4 with
vinyl propionate in toluene.
could be compensated for by using another polymer for
the activation of PLE (vide infra) thus making a prepara-
tive scale resolution of this alcohol feasible too. The
ethylsulfanyl group bearing alcohol rac-3 could also be
acylated with vinyl propionate by using PLE/MPEG. In
toluene as solvent the enantioselectivity of the acylation
was high (E > 100), but the activity of the enzyme was
very low (Table 1, entry 3). However, the activity could
be increased considerably by using n-octane as solvent
(Table 1, entry 4). Under these conditions the enzyme
exhibited toward rac-3 a good selectivity (E ) 53) as well
as a good activity (44% conversion after 2 d).
To probe the steric requirements of the active site of
PLE in organic solvents and because of synthetic reasons,
alcohol rac-4, which bears a methoxy group in p-position,
was submitted to a PLE/MPEG-catalyzed acylation. The
enzyme showed toward this alcohol the highest selectivity
(E > 100) and a good activity giving ester (S)-4a and
alcohol (R)-4 (Table 1, entries 5 and 6). While in toluene
the selectivity of the enzyme was higher than in n-octane,
in the latter solvent the activity was slightly higher. In
the case of the acylation of the fluorine bearing alcohols
rac-5-7 the activity of the enzyme was the major problem
(entries 7-10). In addition the selectivity of PLE toward
the benzylic alcohol rac-5 was low. The activity problem
could be solved in the case of rac-6 through a solvent
change. In toluene the selectivity of PLE toward this
alcohol was high (E > 100) but its activity was very low
(16% conversion after 2 d). A switch to n-octane as solvent
saw an increase of the activity of the enzyme (39%
conversion after 3 d). Although the selectivity (E ) 51)
decreased with this solvent change, it is still high enough
for synthetic purposes. In the case of the trifluoromethyl
substituted alcohol rac-7 the selectivity of PLE could not
be determined because of a very low activity of the
enzyme. The activities of PLE/MPEG toward rac-1-4 in
organic solvents are illustrated by Figure 1.45 It is
surprising that PLE/MPEG shows toward the structur-
ally closely related alcohols rac-1-3 such different activi-
ties under the conditions employed. We have no convinc-
ing explanation at present for this phenomenon. Product
inhibition as a possible cause can be excluded since PLE
showed toward rac-2 and rac-3 in n-octane (cf. Table 1,
entry 4) in the presence of MPEG or of another activating
polymer (vide infra) much higher activities. The PLE/
MPEG-catalyzed resolution of alcohol rac-4 showed the
F igu r e 2. (a) Gram-scale PLE/MPEG-catalyzed kinetic reso-
lution of rac-4 with vinyl propionate in n-octane. (b) Solvent
dependency of the PLE/MPEG-catalyzed acylation of rac-4
with vinyl propionate. (c) PLE/TGA- and PLE/MPEG//AMPS-
catalyzed acylation of alcohol rac-4 with vinyl propionate in
organic solvents.
characteristic time dependencies of the conversion and
the ee values of a kinetic resolution (Figure 2a), and the
selectivity factor E, which is high, remained practically
constant throughout the acylation.
PLE exhibited toward the alcohols rac-8-10, which
lack a benzylic phenyl group, moderate activities and
varying selectivities. While the selectivity was only low
in the case of alcohol rac-8 (E ) 6), which does not carry
a benzyl group, it was high in the case of the phenoxym-
ethyl group bearing alcohols rac-9 (E >100) and rac-10
(E > 100) (Table 1, entries 11-13). The lipases studied
so far in the acylation of rac-9 and rac-10 showed lower
enantioselectivities than PLE. The Candida antarctica
(45) The dotted lines of Figure 1 and of the following Figures 2 and
3 are only meant to connect the points of measurement and calculation.

3388 J . Org. Chem., Vol. 66, No. 10, 2001
Gais et al.
Ta ble 2. P LE/MP EG-Ca ta lyzed Acyla tion of Alcoh ol
Ta ble 3. P LE/MP EG-Ca ta lyzed Acyla tion of Alcoh ol
r a c-4 in Differ en t Or ga n ic Solven tsa
r a c-4 a t Differ en t Tem p er a tu r esa
t
convnb alcohol ester
T (°C) t (d) convn (%) alcohol ee (%) ester ee (%)
Eb
>100
entry
solvent
(d)
(%)
ee (%) ee (%)
Ec
0
20
40
1
1
1
13
51
45
13
98
99
g98
96
95
1
2
3
4
5
6
7
8
9e
toluene
benzene
n-decane
n-octane
CH2Cl2
CHCl3
DME
MTBE
vinyl prop.
2
2
2
2
2
2
2
2
2
39 (35)
29 (27)
51 (49)
51 (49)
6 (4)
73
50
98
99
9
99 >100 (437)
99 >100 (327)
96 >100 (226)
94 >100 (170)
g98 >100
>100 (226)
>100 (205)
a
Conditions: 1 mmol of alcohol, 4 mmol of vinyl propionate,
b
300 U of PLE, 10 mL of n-octane containing 0.4% water. Num-
bers in parentheses are the calculated values.
5 (3)
d
g98 >100
47 (42)
51 (47)
51 (47%)
86
98
95
88
85
80
43
55
32
MTBE similar activities. This may be due to the fact that
MPEG has in both solvents only a low solubility. The low
activity of PLE in methylene chloride and chloroform may
perhaps be related among other things to the high
solubility of MPEG in these solvents causing initially a
complete separation of PLE and MPEG.
a
Conditions: 1 mmol of alcohol, 4 mmol of vinyl propionate,
300 U of PLE, 10 mL of solvent containing 0.4% water. Numbers
b
c
in parentheses are those for a reaction time of 1 d. Numbers in
parentheses are the calculated values. Not determined. e Vinyl
d
propionate was used also as solvent.
Because of the distinct temperature dependencies of
the activity and selectivity of some lipases in acylation
in organic solvents,5,12 we looked for this parameter in
the PLE/MPEG-catalyzed acylation of rac-4 in n-octane
containing 0.4% water (Table 3). It was found that the
temperature dependence of these parameters is only only
weakly expressed. The selectivity did practically not
change. The acylation rate increased on going from 0 to
20 °C, and it decreased slightly at higher temperatures,
presumably because of a deactivation of the enzyme.
Having recorded favorable selectivities and activities
in the acylation of alcohols rac-1, rac-3, rac-4, rac-6, and
rac-10 by using PLE/MPEG, we applied this CLP to their
gram-scale resolution. Reaction of 10-20 mmol of alco-
hols rac-1, rac-3, rac-4, rac-6, and rac-10 with 4 equiva-
lents of vinyl propionate in toluene or n-octane containing
0.43% water for 5 to 10 d afforded the enantiomerically
enriched esters (S)-1a , (S)-3a , (S)-4a , (R)-6a , and (R)-
10a , respectively, and the enantiomerically enriched
alcohols (R)-1, (R)-3, and (R)-4, respectively, in good
yields (Table 4).
Crucial to the application of PLE/MPEG in enantiose-
lective acylation, besides the choice of the solvent, is the
establishment of the optimal water content of the system.
In most cases the activity and selectivity of PLE are at
maximum if the system contains between 0.2 and 1.4%
water. However, as observed previously in the PLE/
MPEG-catalyzed acylation of primary alcohols,23,24 the
activity and the selectivity of the enzyme are also in the
case of secondary alcohols inversely depended on the
water content. Whereas the activity of PLE is the highest
at the lower limit of the optimal range of water content,
it is the selectivity of the enzyme which is higher at the
upper limit. For gram-scale experiments, a water content
of 0.43% represents a good compromise.
P ig Liver Ester a se/Bovin e Ser u m Albu m in . Al-
though PLE/MPEG showed synthetically useful activities
in the acylation of rac-1-4, rac-6, rac-9, and rac-10, often
the rate of acylation decreased after an initially fast
acylation more strongly with increasing reaction time
than anticipated (cf. Table 2). This reduction of the rate
may have been caused among other things by a decrease
of the water content of the system because of a competing
PLE/MPEG-catalyzed hydrolysis of vinyl propionate and
by a deactivation of the enzyme by acetaldehyde and
propionic acid, both being formed as byproducts, through
functionalization and protonation of the enzyme (Scheme
2). Such a consumption of water was indicated by the
gradual disappearance of the aqueous phase in the
aforementioned PLE/MPEG-catalyzed acylation with
lipase and Pseudomonas sp. lipase catalyzed acylation
of rac-9 features E values of 4-13932 and 35,36 respec-
tively, while the Pseudomonas cepacia lipase catalyzed
acylation of rac-10 is characterized by an E value of 42.42
The enantiomer preference of these lipases toward rac-9
and rac-10 is the same as of PLE. However, the rates of
the lipase-catalyzed acylations of these alcohols are
higher than those of PLE.
Because of the high selectivity and relatively high
activity of PLE toward rac-4, acylation of this alcohol was
studied in more detail by variation of the solvent and the
temperature. As shown by Table 2and Figure 2b, benzene
and toluene are in regard to selectivity the best solvents
for the PLE/MPEG-catalyzed kinetic resolution of rac-4
(Table 2, entries 1 and 2). A disadvantage of these
solvents is the only moderate activity of the enzyme.
Similar solvent dependencies of selectivity and activity
have been made previously in the case of lipases.46 In
n-decane or n-octane the acylation rate of rac-4 was
enhanced as compared to the aromatic solvents (Table
2, entries 3 and 4). Although the selectivity decreased
slightly by using these solvents, it is still high enough
(E > 100) for a preparative kinetic resolution. In meth-
ylene chloride and chloroform the activity of PLE was
low, although the enantioselectivity of the enzyme re-
mained high (entries 5 and 6). A switch to the more polar
solvents 1,2-dimethoxyethane (DME), methyl tert-butyl
ether (MTBE) and vinyl propionate saw decreased but
still high selectivities and good activities (Table 2, entries
7-9). From Figure 2b, no clear-cut picture as to the
influence of the various solvents upon the activity of PLE
emerges. It is obvious, however, that in the case of PLE/
MPEG no linear correlation exists between the activity
of the enzyme and the log P value of the solvent. Such a
correlation had been found for a number of lipases, which
showed low activity in solvents with log P values less
than 2 (DME, MTBE), moderate activity in those with
log P values larger than 2 and less than 4 (toluene,
benzene, CHCl3) and high activity in solvents with log P
values greater than 4 (n-octane, n-decane).47 The rela-
tively fast conversion of rac-4 in vinyl propionate is not
surprising given the presence of the acyl donor in large
excess. Somewhat surprising is the observation that PLE
shows in water miscible DME and in water immiscible
(46) Orrenius, C.; Oehrner, N.; Rotticci, D.; Mattson, A.; Hult, K.;
Norin, T. Tetrahedron: Asymmetry 1995, 6, 1217.
(47) (a) Laane, C. Biocatalysis 1987, 1, 17. (b) Laane, C.; Boeren,
S.; Vos, K.; Verger, C. Biotechnol. Bioeng. 1987, 30, 81.

Acylation of Racemic Functionalized Secondary Alcohols
Ta ble 4. Gr a m -Sca le P LE/MP EG-Ca ta lyzed Resolu tion of Alcoh ols r a c-1, r a c-3, r a c-4, a n d r a c-6a
alcohol ester
yield (%) yield (%)
J . Org. Chem., Vol. 66, No. 10, 2001 3389
entry
substrate
solventb
t (d)
convn (%)
ee (%)
ee (%)
E
1
2
3
4
5
rac-1c
rac-3d
rac-4e
rac-6f
rac-10g
toluene
n-octane
n-octane
n-octane
toluene
5
5
3
7
10
47
49
51
40
31
90
89
99
61
42
41
42
38
59
56
97
92
96
92
97
39
35
39
39
27
>100
71
>100
44
99
a
b
Yields are based on the racemic alcohols. Containing 0.4% water. c 21 mmol of alcohol, 84 mmol of vinyl propionate, 200 mL of
toluene, 5800 U of PLE. 20 mmol of alcohol, 80 mmol of vinyl propionate, 200 mL of n-octane, 7700 U of PLE. e 20 mmol of alcohol, 80
d
mmol of vinyl propionate, 200 mL of n-octane, 7600 U of PLE. f 10 mmol of alcohol, 40 mmol of vinyl propionate, 100 mL of n-octane, 4800
g
U of PLE. 10 mmol of alcohol, 40 mmol of vinyl propionate, 150 mL of n-octane, 6300 U of PLE.
Sch em e 2. F or m a tion of Sid e P r od u cts in
P LE-Ca ta lyzed Acyla tion of Alcoh ols w ith Vin yl
P r op ion a te a n d Th eir Sca ven gin g by Am in o Gr ou p
Bea r in g P olym er s
with vinyl propionate and thus help to retain the activity
of PLE in organic solvents. In addition, BSA would also
at least to some extend neutralize the carboxylic acid
being formed in the competing enzymatic hydrolysis of
the acyl donor.
Lyophilization of desalted PLE (180-200 U/mg, 30 mg)
and BSA (1 g) in aqueous solution gave PLE/BSA as a
fine, free flowing, white powder, which contained 3-4%
water. The enzyme had a specific activity of 150-160
U/mg in aqueous solution, which did not decrease upon
storage of the CLP at 4 °C for several weeks. Thus,
preparation of PLE/BSA is accompanied only by a slight
loss of enzymatic activity.
PLE/BSA-catalyzed acylation was carried out under
the same experimental conditions as that with PLE/
MPEG. Upon addition of PLE/BSA to the two-phase
mixture composed of toluene or n-octane and a small
amount of water the solid colyophilisate dissolved partly
in the aqueous phase. Acylation was started by a rapid
stirring of the mixture, which caused the aqueous phase
to disperse into small droplets containing some solid
material. With increasing reaction time the second phase
gradually solidified, presumably because of the consump-
tion of water, leaving behind the solid CLP. The results
of the PLE/BSA-catalyzed acylation of rac-1-10 in
toluene and n-octane are compiled in Table 5. The data
show that BSA activation of PLE in organic solvents is
similarly effective as that by MPEG. In some cases, PLE/
BSA showed a higher activity than PLE/MPEG, which
was most notable in the case of the acylation of rac-10
(Table 5, entry 20). Furthermore, the rate of acylation
did not slow as much as in the case of PLE/MPEG.
However, a comparison of the data compiled in Tables 1
and 5 clearly reveals also that in most cases the utiliza-
tion of BSA as activator for PLE leads to a significantly
lower enantioselectivity of acylation as compared to the
use of MPEG. A feature of the PLE/BSA-catalyzed
acylation of rac-1-4, rac-6, and rac-8-10, which was also
found in the case of PLE/MPEG, is a constant correlation
between the reaction rate and the selectivity of the
enzyme toward the substrates. For example, in toluene
after 3 d the conversion of rac-1 was 41% giving alcohol
(R)-1 with 62% ee and ester (S)-1a 95% ee, which
corresponds to an E value of 73 (Table 5, entry 1), and
in n-octane after 3 d the conversion of the alcohol was
56% furnishing (R)-1 with 95% ee and (S)-1a with 82%
ee (E ) 37) (Table 5, entry 2). During PLE/BSA-catalyzed
acylation we noticed a color change of the CLP from
colorless to yellow and finally to reddish brown. This
phenomenon was also observed in the attempted PLE/
BSA-catalyzed acylation of the fluorinated alcohols rac-5
and rac-7 with vinyl propionate, where there was no
conversion of the alcohols (Table 5, entries 9, 10, 13, and
increasing reaction time and by the detection of propionic
acid in the reaction mixture. Thus, following the PLE/
MPEG-catalyzed acylation of rac-4 with vinyl propionate
in n-octane (0.43% of water) by GC analysis revealed
after 1.7 h formation of 6% propionic acid, which in-
creased to 43% after 5 d. The importance of maintaining
the critical amount of essential water during the PLE-
catalyzed acylation with vinyl propionate in organic
solvents as well as the need to scavenge the carbonyl
byproducts prompted us to use bovine serum albumin
(BSA) as an inexpensive primary amino group containing
natural polymer48 for the activation of PLE. We hoped
that the ready reaction of BSA with acetaldehyde under
formation of imines and other stable secondary pro-
ducts49-51 might pave the path for a better activation of
PLE through this polymer by suppressing the analogous
reaction of acetaldehyde with the enzyme.52,53 A beneficial
feature of the reaction of BSA with acetaldehyde in the
present context is the formation of water as a byproduct.
This could perhaps serve to maintain to some extend the
critical amount of water of the system during acylation
(48) Delgado, C.; Patel, J . N.; Francis, G. E.; Fisher, D. Biotechnol.
Appl. Biochem. 1990, 12, 119.
(49) Stepuro, I. I.; Zavodnik, I. B.; Ostrovskii, Y. M. Ukr. Biokhim.
Zh. 1982, 54, 123.
(50) Donohue, T. M.; Tuma, D. J .; Sorrell, M. F. Arch. Biochem.
Biophys. 1983, 220, 239.
(51) Harris, W. J . WO 90/1537, 1990; Chem. Abstr. 1999, 113, 4673.
(52) For the use of BSA as an enzyme-stabilizing additive in the
large scale PLE-catalyzed enantioselective hydrolysis of dimethyl cis-
cyclohex-4-ene-1,2-dicarboxylate in water, see: Riefling, B. F.; Bru¨m-
mer, W. K.; Gais, H.-J . NATO ASI Series C 1986, 178, 347.
(53) For a study of the activity of lipase/BSA, see: (a) Yamane, T.;
Ichiryu, T.; Nagata, M.; Ueno, A.; Shimizu, S.; Biotechnol. Bioeng. 1990,
36, 1063. (b) Bosley, J . A. EP 424,130, 1991; Chem. Abstr. 1991, 115,
67527. (c) Matsumoto, Y.; Tsubota, N. J P 06062846, 1994; Chem. Abstr.
1994, 121, 3961. (d) Inoe, S.; Asahi, N.; Hatanaka, T.; Nanba, T. J P
07289257, 1995; Chem. Abstr. 1995, 124, 49502. (e) Pencreac’h, G.;
Baratti, J . C. Appl. Microbiol. Biotechnol. 1997, 47, 630. (f) Rotticci,
D.; Norin, T.; Hult, K. Org. Lett. 2000, 2, 1373.

3390 J . Org. Chem., Vol. 66, No. 10, 2001
Gais et al.
Sch em e 3. Recover y of P LE/BSA th r ou gh
Ta ble 5. P LE/BSA-Ca ta lyzed Acyla tion of Alcoh ols
r a c-1-10a
Im m obiliza tion on P AM a n d Reu se of th e En zym e
alcohol
ester
sub-
t
convn ee yield ee yield
entry strate solventb (d) (%)
(%) (%) (%) (%)
E
1c rac-1 toluene
2c rac-1 n-octane
3c rac-2 toluene
4c rac-2 n-octane
5c rac-3 toluene
6c rac-3 n-octane
7e rac-4 toluene
8e rac-4 n-octane
9c rac-5 toluene
10c rac-5 n-octane
11c rac-6 toluene
12c rac-6 n-octane
13e rac-7 toluene
14e rac-7 n-octane
15c rac-8 toluene
16c rac-8 n-octane
17c rac-9 toluene
18c rac-9 n-octane
19f rac-10 toluene
20f rac-10 n-octane
3
3
5
5
2
2
2
2
3
3
7
7
5
5
1
1
1
2
1
1
49
56
16
40
20
49
20
46
<1
<1
34
54
<1
<1
42
62
22
52
17
58
62
d
95
d
73
37
>100
95 32 82 44
13 g99
60 44 g99 35 >100
7
67
29
68
0
d
d
d
d
d
d
d
d
d
93
88
g99
97
d
d
90
70 45
d
d
8
23
98
91 42
92
88
d
d
d
d
d
d
d
29
31
>100
>100
d
d
26
28
d
0
35
corresponds to an E value of >100 (Scheme 3). Thus,
PLE/BSA immobilized on PAM shows in the acylation of
this alcohol similar characteristics as the nonimmobilized
CLP. To demonstrate that PLE/BSA//PAM can also be
reused, the immobilized CLP was subsequently applied
to the kinetic resolution of the glycerol derivative rac-
11, which was previously successfully resolved by using
PLE/MPEG.24 PLE/BSA//PAM was washed with toluene
following its removal from the reaction mixture contain-
ing (R)-4 and (S)-4a and placed into a solution of rac-11
(1 mM) and vinyl propionate (5 mM) in toluene contain-
ing 0.5% water. After 2 d, a 56% conversion of the racemic
alcohol had occurred furnishing alcohol (S)-11 with 87%
ee and ester (R)-11a with 90% ee (E ) 52). The loss of
activity of the enzyme after its 2-fold use was estimated
to be 20%.
Because of the varying quality of PLE from commercial
sources, an activity and a selectivity test of the enzyme
prior to its use is required. For the utilization of PLE in
hydrolysis in aqueous solution, this can be done by using
ethyl butyrate and dimethyl cis-1,2-cyclohex-4-ene-1,2-
dicarboxylate as substrates.55 We found that for the
application of PLE in acylation in organic solvents, the
rate and the enantioselectivity of the acylation of the
primary alcohol rac-11 with vinyl propionate in toluene
(0.43% water) by using MPEG as activator are good
criteria for the performance of the enzyme.
>99 37
0
0
8
20
23
97 42
19
90
d
d
d
d
d
d
d
d
d
d
d
1
2
>100
89
28
48
d
d
d
d
a
b
Yields are based on the racemic alcohols. Containing 0.4%
water. c 2 mmol of alcohol, 8 mmol of vinyl propionate, 20 mL of
solvent, 600-800 U of PLE. Not determined. e 1 mmol of alcohol,
d
4 mmol of vinyl propionate, 10 mL of solvent, 300-400 U of PLE.
f 2.6 mmol of alcohol, 10.4 mmol of vinyl propionate, 20 mL of
solvent, 600-800 U of PLE.
14). Furthermore, treatment of BSA with acetaldehyde
or vinyl propionate alone resulted also in the color change
of the protein. Thus, we associate this phenomenon in
the case of PLE/BSA to the reaction of BSA with
acetaldehyde, which is formed by the enzymatic hydroly-
sis of vinyl propionate and the enzymatic acylation of the
alcohol.49-51 The observation of a color change upon
treatment of BSA with vinyl propionate containing 0.4%
water indicates that the acyl donor acylates BSA, with
liberation of acetaldehyde, which in turn reacts with the
protein. Analysis of the thus obtained colored BSA by the
method of Kaiser54 showed a strong reduction of the
number of free primary amino groups.
P ig Liver Ester a se/Ten ta GelAm in o a n d P ig Liver
E st er a se/Met h oxyp oly(et h ylen e glycol)//Am in om -
eth yl P olystyr en e. Primary amino group containing
polymers such as aminomethyl polystyrene (AMPS)56 and
TentaGelAmino (TGA)57 have been used previously to
anchor aldehydes and to scavenge acids or excess elec-
trophiles in solid-phase synthesis.58 Therefore, we con-
sidered AMPS and TGA as interesting alternative can-
didates for the activation of PLE in acylation with vinylic
esters in organic solvents. For the anticipated 3-fold role
of the resin, namely to scavenge acetaldehyde, to bind
water and to neutralize propionic acid, TGA was selected
mainly because of its partial structural resemblance to
MPEG. Lyophilization of desalted PLE (180-200 U/mg,
20 mg) in aqueous solution and TGA (2 g) in aqueous
suspension afforded PLE/TGA as a creamy free flowing
powder. The enzyme had a specific activity of 86 U/mg
in aqueous solution, which did not decrease upon storage
Im m obiliza tion a n d Recover y of P ig Liver Es-
ter a se/Bovin e Ser u m Albu m in . We had previously
observed that upon placement of a water saturated
(approximately 5% water) polyaramide membrane (PAM)
in a suspension of PLE/MPEG in toluene a spontaneous
and complete deposition of the CLP on the membrane
occurs.23,24 No major differences were observed in acyla-
tion of alcohols by using either PLE/MPEG//PAM or PLE/
MPEG. Withdrawal of the liquid phase of the reaction
mixture containing the alcohol and ester and washing
PLE/MPEG//PAM with toluene allowed for its reuse in
subsequent batches without major loss of activity. A
similar placement of water saturated PAM in a suspen-
sion of PLE/BSA in toluene led also to a spontaneous
deposition of the CLP on the membrane in form of small
droplets. Practically none of the enzyme remained in the
suspension. Because of the high activity and selectivity
of PLE/BSA toward rac-4 (Table 5, entries 7 and 8), we
selected the acylation of this alcohol as a probe for the
activity of PLE/BSA//PAM. Enzymatic acylation of rac-4
with vinyl propionate in toluene by using PLE/BSA//PAM
led after 5 d to a conversion of the alcohol of 47% giving
(R)-4 with 69% ee and (S)-4a with 97% ee, which
(55) Gais, H.-J .; Lukas, K. L.; Ball, W. A.; Braun, S.; Lindner, H. J .
Liebigs Ann. Chem. 1986, 687.
(56) Mitchell, A. R.; Erickson, B. W.; Merrifield, R. B. J . Am. Chem.
Soc. 1976, 98, 7357.
(57) Bayer, E. Angew. Chem 1991, 103, 117; Angew. Chem., Int. Ed.
Engl. 1991, 30, 113.
(58) Obrecht, D.; Villalgordo, J . M., Solid-Supported Combinatorial
and Parallel Synthesis of Small-Molecule-Weight Compound Libraries;
Pergamon: Oxford, 1998.
(54) Kaiser, E.; Colesett, R. L.; Bossinger, C. D.; Cook, P. I. Anal.
Biochem. 1970, 34, 595.

Acylation of Racemic Functionalized Secondary Alcohols
Ta ble 6. P LE-Ca ta lyzed Acyla tion of Alcoh ols r a c-2 a n d r a c-4 in th e P r esen ce of Am in o Gr ou p Con ta in in g P olym er sa
alcohol ester
yield (%) yield (%)
J . Org. Chem., Vol. 66, No. 10, 2001 3391
entry
substrate
CLP
solventb
t (d)
convn (%)
ee (%)
ee (%)
E
1c
2d
3c
rac-2
rac-2
rac-4
PLE/MPEG//AMPS
PLE/TGA
PLE/MPEG//AMPS
n-octane
n-octane
toluene
9
6
4
49
55
43
80
89
56
47
38
45
g99
94
98
44
38
36
>100
97
>100
a
b
Yields are based on the racemic alcohols. Containing 0.4% water. c 1 mmol of alcohol, 4 mmol of vinyl propionate, 10 mL of solvent,
d
300-400 U of PLE. 2 mmol of alcohol, 4 mmol of vinyl propionate, 20 mL of solvent, 600-800 U of PLE.
Sch em e 4. P LE-Ca ta lyzed Acyla tion of Alcoh ols
in th e P r esen ce of Am in o Gr ou p Bea r in g P olym er s
a n d Re-u se of th e En zym e
water in order to effect the cleavage of the imines. The
loading of the fresh and used AMPS with free amino
groups was determined by the Kaiser test.54 Reaction of
rac-2 with vinyl propionate in n-octane by using PLE/
TGA resulted after 6 d in a 55% conversion of the alcohol
furnishing with high selectivity (E ) 97) (R)-2 with 89%
ee and (S)-2a with g99% ee. The CLP was recovered
simply by filtration and washing with n-octane. Applica-
tion of recovered PLE/TGA to the acylation of rac-2 (2
mM) in n-octane containing 0.4% water saw after 16 d a
36% conversion of the substrate delivering (R)-2 with 50%
ee and (S)-2a with g99% ee.
The use of PLE/MPEG//AMPS in the acylation of rac-4
with vinyl propionate in toluene led after 4 d to a 43%
conversion of the substrate giving with high selectivity
(E > 100) (R)-4 with 56% ee and (S)-4a with 98% ee. PLE/
MPEG//AMPS was recovered simply by decanting the
liquid phase and washing the solid residue with toluene.
The recovered PLE/MPEG//AMPS was used in a subse-
quent acylation of the glycerol derivative rac-11 (1 mM)
with vinyl propionate (5 mM) in toluene containing 0.5%
water. After 3 d a conversion of the substrate of 50% was
achieved giving with good selectivity (E ) 42) alcohol (S)-
11 with 89% ee and ester (R)-11a with 87% ee. The loss
of activity of the enzyme after its 2-fold use was estimated
to be 20%.
of the CLP at 4 °C for several weeks. The water content
of the CLP was 1-2%. Thus, unlike the preparation of
PLE/MPEG and PLE/BSA that of PLE/TGA was ac-
companied by a significant loss of enzymatic activity. We
have no explanation at present for this activity loss.59 It
should be noted, however, that unlike MPEG and BSA
the polymer TGA is not soluble in water. This leads to a
deposition of PLE on the polymer during lyophilization.
In contrast, PLE/MPEG and PLE/BSA are intimate
mixtures. The results of the PLE/TGA- and the PLE/
MPEG//AMPS-catalyzed acylations, which were carried
out under similar experimental conditions as those by
using PLE/MPE, PLE/BSA and PLE/MPEG//PAM, are
listed in Table 6.
A common feature of the PLE-catalyzed acylation in
the presence of the aforementioned polymers is that the
acylation rates are initially high resulting after 3-4 h
in a 30-35% conversion, but slow subsequently more
strongly than anticipated on the basis of the selectivity
factor E (Figure 2c). At present, we have no experimen-
tally verified explanation for this phenomenon. It should
be noted, however, that the selectivity factor remains
practically constant throughout the acylation.
Acylations by using PLE/TGA and PLE/MPEG//AMPS
were, as in the case of PLE/BSA, accompanied by a color
change of the CLP from almost colorless via yellow to
reddish brown. Such a change was also observed upon
treatment of TGA and AMPS with acetaldehyde or vinyl
propionate in the absence of the enzyme. TGA and AMPS
isolated after this treatment contained no free amino
groups according to a Kaiser test.54 Thus, we ascribe the
color change in these cases also to a reaction of the
polymers with acetaldehyde (cf. Scheme 4). Reaction of
TGA and AMPS with acetaldehyde in the case of the use
of vinyl propionate has to be proceeded by an acylation
of the polymers.
The PLE/MPEG-catalyzed acylation of rac-2 in n-
octane in the presence of AMPS resulted after 9 d in a
49% conversion of the alcohol giving with excellent
selectivity (E > 100) (R)-2 with 80% ee and (S)-2a with
g99% ee (Table 6, entry 1) (Scheme 4). PLE/MPEG was
found adhering to the reaction vessel as it was physically
adsorbed by AMPS.60 PLE/MPEG//AMPS was removed
from the reaction mixture containing (R)-2 and (S)-2a
through filtration, washed with n-octane and used in the
acylation of a second batch of the alcohol (0.67 mM) with
vinyl propionate (2.68 mM) in n-octane containing 0.4%
of water. After 18 d, a 17% conversion of the alcohol was
achieved yielding (R)-2 with 9% ee and (S)-2a with g 99%
ee (E > 100). AMPS was recovered first by washing with
water in order to separate PLE and finally by treatment
with oxalic acid in a mixture of tetrahydrofuran and
A comparison of the activity and selectivity of PLE
toward rac-4 in the presence of the polymers tested,
except TGA, is provided by Figure 3. The enantioselec-
tivity of PLE in the presence of the various polymers is
in all cases the same, although there was a slight
decrease of the ee value of the ester (S)-4a with increas-
ing reaction time. Of all the CLPs of PLE and polymers
tested PLE/BSA had the lowest activity toward rac-4.
(59) A reviewer of this paper has suggested that the reason for the
reduced activity might be the strong alteration of the hydronium ion
activity (pH) of the colyophilisate by these polyamines.
(60) For the immobilization of enzymes on polyaminostyrene, see:
Brandenberger, H. Rev-Ferment. Ind. Aliment. 1956, 11, 237.

3392 J . Org. Chem., Vol. 66, No. 10, 2001
Gais et al.
Sch em e 5. P LE/KCl/K2HP O4-Ca ta lyzed Acyla tion
of Alcoh ols
mg) in water gave PLE/KCl/K2HPO4 as a fine white
powder, which contained less than 1% water. The enzyme
had a specific activity of only 90 U/mg in aqueous
solution, thus, revealing a significant activity loss during
colyphilization. The activity of the CLP did, however, not
decrease upon its storage at 4 °C for several weeks.
The PLE/KCl/K2HPO4-catalyzed acylation of rac-2 with
vinyl propionate in n-octane saw only a very low conver-
sion of the alcohol. In addition, the selectivity was also
only low. However, adjustment of the water content of
the reaction mixture to 0.4% resulted not only in an
increase of the reaction rate but also of the selectivity.
Reaction of rac-2 with vinyl propionate in n-octane by
using PLE/KCl/K2HPO4 under these conditions led after
6 d to a 30% conversion of the alcohol affording with high
selectivity (E > 100) (R)-2 with 39% ee and (S)-2a with
g 99% ee (Scheme 5).
Ster eoselectivity of P ig Liver Ester a se in Or ga n ic
Solven ts. The absolute configurations of (S)-1a /(R)-1,24,28
(S)-4a /(R)-4,28 (R)-6a /(S)-6,31 (R)-9a /(S)-9,32 and (R)-10a /
(S)-1033 were assigned by comparison of their optical
rotation with those reported in the literature. Those of
(S)-2a/(R)-2 and (S)-3a/(R)-3 were assigned on the premise
that they have the same sign of optical rotation as (S)-
1a /(R)-1 and (S)-4a /(R)-4. It follows, thus, that in the case
of rac-1-4 the S-alcohol reacts with the acyl enzyme
faster than the (R)-alcohol leading to the (S)-ester,
whereas in the case of the fluoro derivative rac-6 the (R)-
alcohol is the faster reaction enantiomer. The replace-
ment of the p-methoxyphenyl group in rac-4 by a phenoxy
group did not change the enantiomer preference of the
enzyme as shown by the acylation of rac-10. This holds
also true for rac-9 where the methoxy group of rac-10
has been replaced by a H-atom. Thus, the isoenzyme
mixture of PLE seems to exhibit toward secondary
alcohols, which carry a benzyl, p-methoxybenzyl, and
phenoxymethyl group, not only a high selectivity but also
the same enantiomer preferences irrespective of the other
group which is attached to the stereogenic center. The
only exception is rac-6, which carries a fluorine atom.
Several substrate models have been advanced for the
enantiomer and enantiotopic group preference of PLE in
the hydrolysis of esters of prochiral or racemic carboxylic
acids.62-64 These models discuss the quality of the binding
of the chiral or prochiral carboxylic acid part of an alkyl
ester in the active site of PLE during acylation of its
serine hydroxy group. The most advanced model is that
of J ones et al.62,63 This model allows for a rationalization
of the enantiomer and enantiotopic group preference of
PLE in the hydrolysis of most of the esters of prochiral
or racemic carboxylic acids investigated thus far. An
analogous model, which considers the binding of the
chiral or prochiral alcohol part of an ester, whose car-
boxylic acid part is achiral, in the active site of PLE
F igu r e 3. Activity (a) and selectivity (b, c) of PLE/MPEG,
PLE/BSA, PLE/BSA//PAM, and PLE/MPEG//AMPS in the
acylation of rac-4 with vinyl propionate in toluene.
However, immobilization of PLE/BSA on PAM saw a
considerable increase in the activity of the enzyme.
Interpretation of this observation is difficult since several
parameters such as the water content of the system and
the partition of the small PLE/BSA containing phase in
the organic phase have been changed. Of all of the PLE/
polymer combinations tested PLE/MPEG//AMPS exhib-
ited the highest initial activity toward rac-4. However,
a disadvantage of the use of TGA and AMPS is their
competing reaction with the acyl donor, which is also
accompanied, as in the case of BSA, by the formation of
acetaldehyde.
P ig Liver Ester a se/P ota ssiu m Ch lor id e/P ota s-
siu m Hyd r ogen P h osp h a te. High concentrations of
salts have been used to activate enzymes in organic
solvents by maintaining the critical water content as well
as the structure of the enzyme during lyophilization.61
Lyophilization of PLE (180 U/mg, 30 mg), potassium
chloride (2.94 g) and potassium hydrogen phosphate (30
(61) Ru, M. T.; Dordick, J . S.; Reimer, J . A.; Clark, D. S. Biotechnol.
Bioeng. 1999, 63, 34.
(62) Toone, E. J .; Werth, M. J .; J ones, J . B. J . Am. Chem. Soc. 1990,
112, 4946.
(63) Provencher, L.; J ones, J . B. J . Org. Chem. 1994, 59, 2729.

Acylation of Racemic Functionalized Secondary Alcohols
J . Org. Chem., Vol. 66, No. 10, 2001 3393
Sch em e 6. En a n tioselectivity of P LE in Or ga n ic
Solven ts a n d in Wa ter
during deacylation of the acyl enzyme or the acylation
of the enzyme is not available yet. It has been claimed,
however, that the enantiomer preference of PLE in the
hydrolysis of the diacetates of cyclic diols can be correctly
predicted by the J ones model.65 Application of this model
to the PLE-catalyzed kinetic resolution of the alcohols
investigated in this work in a manner as for the afore-
mentioned diacetates led in most cases to a correct
prediction of the faster reacting enantiomer.66 One has
to bear in mind, however, that the J ones model has been
developed on the basis of the selectivities found by using
the native mixture of PLE isoenzymes, which in principle
could have different activities and enantioselectivities.
Hydrolysis studies of esters in water by using single
isoenzymes67-70 and enriched isoenzyme mixtures6a,6e,71
had revealed for some substrates indeed such differences.
It would be, thus, interesting to see whether the PLE
isoenzymes show similar differences also in acylation in
organic solvents. Such studies of the isoenzymes are,
however, at present not feasible because of their non-
availability from commercial sources and because of the
difficulties associated with the separation of the native
isoenzyme mixture on a somewhat larger scale.67-70,72
PLE in organic solvents shows an interesting selectiv-
ity feature (Scheme 6). Previous observations suggested
that PLE exhibits in organic solvents a higher acyclation
enantioselectivity than for hydrolysis in aqueous solu-
tion.24 For example, we recorded for the PLE/MPEG-
catalyzed acylation of the primary alcohol rac-12 with
vinyl propionate in toluene (0.43% water) a selectivity
factor of E ) 24, while for the PLE catalyzed hydrolysis
of the corresponding butyrate rac-12a in water an E
value of only 3-5 was reported.73 Similar results were
obtained in the case of the related primary alcohol rac-
13 and its propionate rac-13a .24,66 Whereas the PLE/
MPEG-catalyzed acylation of rac-13 is characterized by
a selectivity factor of E ) 30, the PLE catalyzed hydroly-
sis of rac-1 3a has an E value of only 2. Having obtained
these results, we were interested to see whether such a
selectivity difference can be found also in the case of one
of the aforementioned 2-propanols. The PLE catalyzed
hydrolysis of butyrate rac-10a in aqueous solution pro-
ceeded nonselective according to a selectivity factor of E
≈ 1.24,66 In contrast, the PLE/MPEG-catalyzed acylation
of alcohol rac-10 was highly selective as expressed by an
E value of > 100. Lipases and proteases were previously
found to exhibit an analogous selectivity behavior as PLE
upon switching from water to an organic solvent.5 Several
explanations for this phenomenon such as differences in
the solvation of the transition state are currently being
discussed, but no one is completely satisfactory.5,12
Finally, Scheme 6 reveals also that in the cases
investigated the enantiomer preference of PLE in acyla-
tion and hydrolysis of related compounds is the same.
Con clu sion
Addition of the polymers MPEG, BSA, TGA, and AMPS
confers activity to the PLE isoenzymes in the acylation
of alcohols with vinyl propionate in toluene and n-octane
containing less than 1% of water. Whereas in the
preparation of PLE/MPEG and PLE/BSA the specific
activity of the enzyme is largely retained, colyophilization
of PLE with TGA is accompanied by a significant loss of
enzymatic activity. In addition, the primary amino group
containing polymers BSA, TGA and AMPS are not inert
against the acyl donor vinyl propionate. In general, the
CLP of choice for a PLE-catalyzed acylation in organic
media is PLE/MPEG either as such or immobilized on
PAM. PLE shows high enantioselectivity toward the
functionalized secondary alcohols rac-1-4, rac-6, rac-9,
and rac-10. The activity and selectivity of the enzyme
were in the case of rac-1, rac-3, rac-4, rac-6 and rac-10
such as to allow for a kinetic resolution on a gram scale.
Recycling of PLE/polymer for a batch-wise application
can be achieved through their spontaneous immobiliza-
tion on a water saturated polyaramide membrane in
organic solvents. The activity enhancement observed for
PLE in the presence of the primary amino group contain-
ing polymers BSA, AMPS, and TGA seems to be associ-
ated at least in part with a scavenging of acetaldehyde
formed as a byproduct in the acylation with vinyl
propionate. However, other factors might contribute too
to the enhanced activity of PLE in the presence of the
polymers. Colyophilization of PLE with the aforemen-
(64) Walser, P.; Renold, P.; N’Goka, V.; Hosseinzadeh, F.; Tamm,
C. Helv. Chim. Acta 1991, 74, 1941.
(65) Moravcova´, J .; Vanclova´, Z.; Cˇ apkova´, J .; Kefurt, K.; Staneˇk,
J . J . Carbohydr. Chem. 1997, 16, 1011.
(66) J ungen, M. Ph.D. Thesis, RWTH Aachen 2001.
(67) Heymann, E.; J unge, W. Eur. J . Biochem. 1979, 95, 509.
(68) J unge, W.; Heymann, E. Eur. J . Biochem. 1979, 95, 519.
(69) O¨ hrner, N.; Mattson, A.; Norin, T.; Hult, K. Biocatalysis 1990,
4, 81.
(70) Lam, L. K. P.; Brown, C. M.; De J eso, B.; Lym, L.; Toone, E. J .;
J ones, J . B. J . Am. Chem. Soc. 1988, 110, 4409.
(71) Daussmann, T.; Gais, H.-J .; Buschmann, H.; Griebel, C. Un-
published results.
(72) Emmer, A.; J ansson, M.; Roeraade, J . J . Chromatogr. A 1994,
672, 231.
(73) J anes, L. E.; Lo¨wendahl, A. C.; Kazlauskas, R. J . Chem. Eur.
J . 1998, 4, 2324.

3394 J . Org. Chem., Vol. 66, No. 10, 2001
Gais et al.
enzyme had a specific activity of 160-180 U/mg, which did
tioned polymers leads to an increase of the surface of the
enzyme, which should increase its activity. In addition
colyophilization may result in a larger intake of water
in the surrounding of PLE thus maintaining the water
activity of the enzyme during acylation. Although no
extensive comparative investigation has thus far been
carried out, it seems, that the activity of PLE/polymer
in acylation of alcohols in organic solvents is lower than
of PLE in hydrolysis of the corresponding esters in
aqueous solution. The enantioselectivity of PLE, however,
may by occasionally higher in organic solvents.
not decrease upon storage of the CLP at 4 °C for several weeks.
P LE/BSA. PLE (30 mg, 180-200 U/mg) was desalted by
ultrafiltration (cutoff 30 kDa) under ice cooling. The remaining
solution/suspension of PLE in water (20 mL) was treated with
BSA (1 g) and the mixture was stirred magnetically for 2 h
under ice cooling. Then, the mixture was frozen in liquid
nitrogen and lyophilized (0.004-0.009 mbar, 48 h) to give PLE/
BSA as a white powder, which contained 3-4% water. The
enzyme had a specific activity of 150-160 U/mg, which did
not decrease upon storage of the CLP at 4 °C for several weeks.
P LE/TGA. PLE (20 mg, 180-200 U/mg) was desalted by
ultrafiltration (cutoff 30 kDa) under ice cooling. The remaining
solution/suspension of PLE in water (20 mL) was treated with
TGA resin (2 g) and the suspension was stirred magnetically
for 2 h under ice cooling. Then, the mixture was frozen in liquid
nitrogen and lyophilized (0.040-0.016 mbar, 48 h) to give PLE/
TGA as a free flowing creamy colored powder, which contained
1-2% water. The enzyme had a specific activity of 86 U/mg,
which did not decrease upon storage of the CLP at 4 °C for
several weeks.
P LE/KCl/K2HP O4. PLE (30 mg, 180-200 U/mg) was de-
salted by ultrafiltration (cutoff 30 kDa) under ice cooling. To
the remaining solution/suspension of PLE in water (20 mL)
was added KCl (2.94 g) and K2HPO4 (30 mg) and the mixture
was stirred magnetically for 2 h under ice cooling. Then, the
mixture was frozen in liquid nitrogen and lyophilized (0.040-
0.016 mbar, 48 h) to give PLE/KCl/K2HPO4 as a free flowing
colorless powder, which contained less than 1% water. The
enzyme had a specific activity of 90 U/mg, which did not
decrease upon storage of the CLP at 4 °C for several weeks.
Exp er im en ta l Section
Gen er a l Meth od s. Chemical shifts are given in ppm
relative to Me4Si: δ ) 0.00 as internal standard. Peaks in the
13C NMR spectra are denoted as ×e3u×e3 for carbons with zero
or two attached protons or “d” for carbons with one or three
attached protons, as determined from the ATP pulse sequence.
Specific rotations are given in grad‚L/dm‚g, c in g/100 mL.
Enzymatic acylations were run at ca. 22 °C and were moni-
tored by GC analysis on a CP-Sil-8 column. PLE (EC 3.1.1.1,
180-200 U/mg, native mixture of isoenzymes) was purchased
from Roche Diagnostics (then Boehringer Mannheim) as a
suspension in 3.2 M NH4SO4 (30 mg PLE in 3 mL). The specific
activity of PLE and of PLE admixed with polymers was
determined by hydrolysis of ethyl butyrate in aqueous buffer
solution (pH 8.0, 25 °C). Enantiomer compositions were
determined by GC analysis on an octakis-(2,3-O-dipentyl-6-
O-methyl)-γ-cyclodextrin column (25 m × 0.25 mm) (Lipodex
γ-6-Me, Macherey Nagel), a permethylated â-cyclodextrin
column (25 m × 0.25 mm) (CP-Chirasil-Dex-CB, Chrompack),
and an octakis-(2,6-di-O-pentyl-3-O-butyl)-γ-cyclodextrin col-
umn (25 m × 0.25 mm) (Lipodex E, Macherey-Nagel) (carrier
gas hydrogen, 100 kPa). Column chromatography was done
on Merck silica gel 60 (230-400 mesh). MPEG5000 was pur-
chased from Sigma, and BSA, fraction V, was obtained from
Roche Diagnostics. TGA (TentaGel S NH2 SS, LC ) 0.3 mmol/
g, 90 µm) and AMPS (LC ) 1.0 mmol/g, 200-400 mesh) were
obtained from Advanced ChemTech. The water content of PLE/
polymer was determined by Karl Fischer titration with Karl
Fischer solutions obtained from Merck and Riedel-de-Hae¨n.
The polyaramide ultrafiltration membrane (UF-PA-5H and
UF-PA-20H) was from Celanese. Alcohols rac-8 and rac-9 were
obtained from Acros Organics. The synthesis of alcohols rac-
1-7 and rac-10 was carried out according to the litera-
ture24,38,41,74 or as described in the Supporting Information.
Standard workup of enzymatic acylation included chromatog-
raphy in order to separate ester and alcohol following Kugel-
rohr distillation to remove MPEG.
Desa ltin g of th e Su sp en sion of P LE in Aqu eou s Am -
m on iu m Su lfa te. A suspension of PLE (30 mg) in aqueous
(NH4)2SO4 (3 mL) was placed in an Amicon ultrafiltration cell
(model 2800, 200 mL, Ø ) 62 mm) equipped with a RC-YM30/
PM30 membrane (Millipore), and ultrafiltration was carried
out under stirring and ice cooling at 1.5 bar positive nitrogen
pressure through repetitive concentration and dilution by
using distilled water (1000 mL). Five times the aqueous
solution/suspension of PLE was concentrated to a volume of
20 mL and taken up again in water (200 mL). Finally a
solution/suspension of PLE (30 mg) in water (20 mL) was
obtained.
P LE/MP EG. PLE (30 mg, 180-200 U/mg) was desalted by
ultrafiltration (cutoff 30 kDa) under ice cooling. The remaining
solution/suspension of PLE in water (20 mL) was treated with
MPEG5000 (1 g), and the mixture was stirred magnetically for
2 h under ice cooling. Then, the mixture was frozen in liquid
nitrogen and lyophilized (0.008 mbar, 60 h) to give PLE/MPEG
as a white powder, which contained less than 1% water. The
P LE/BSA-Ca ta lyzed Acyla tion of r a c-1 in n -Octa n e. To
a stirred solution of vinyl propionate (800 mg, 8 mmol) and
alcohol rac-1 (322 mg, 2 mmol) in n-octane (10 mL) was added
water (0.08 mL) (0.4% v/v). Stirring of the mixture was
continued for 30 min, and PLE/BSA (127 mg, 580 U) and
n-octane (10 mL) were then added. After stirring the mixture
rapidly for 3 d (57% conversion), MgSO4 (1 g) and NaHCO3
(0.5 g) were added and the mixture was filtered through Celite.
The filtrate was concentrated in vacuo and the residue was
purified first by Kugelrohr distillation (0.8 mbar, 100 °C) and
then by chromatography (cyclohexane/EtOAc, 1:1) to give
alcohol (R)-1 (108 mg, 32%) with 95% ee, [R]22D ) +9.2 (c 1.57,
MeOH) and ester (S)-1a (197 mg, 44%) with 82% ee, [R]22
-3.0 (c 2.23, MeOH).
)
D
P LE/BSA-Ca ta lyzed Acyla tion of r a c-2 in n -Octa n e. To
a stirred solution of vinyl propionate (800 mg, 8 mmol) and
alcohol rac-2 (388 mg, 2 mmol) in n-octane (10 mL) was added
water (80 µL) (0.4% v/v). Stirring of the mixture was continued
for 30 min, and PLE/BSA (127 mg, 580 U) and n-octane (10
mL) were then added. After stirring the mixture rapidly for 5
d (40% conversion), MgSO4 (1 g) and NaHCO3 (500 mg) were
added and the mixture was filtered through Celite. The filtrate
was concentrated in vacuo and the residue was purified first
by Kugelrohr distillation (0.8 mbar, 100 °C) and then by
chromatography (cyclohexane/EtOAc, 3:1) to furnish alcohol
(R)-2 (170 mg, 44%) with 60% ee, [R]22D ) +3.2 (c 1.45, MeOH)
and ester (S)-2a (175 mg, 35%) with g99% ee, [R]22D ) -9.4 (c
1.06, MeOH). (R)-2: 1H NMR (400 MHz, CDCl3) δ 7.30 (m,
2H), 7.22 (m, 3 H), 3.98 (m, 1 H), 3.58 (sept, J ) 6.04 Hz, 1
H), 3.44 (dd, J ) 6.0, 3.3 Hz, 1 H), 3.28 (dd, J ) 7.1, 2.2 Hz,
1 H), 2.8 (m, 2 H), 2.48 (s, 1 H), 1.17 (d, J ) 4.9 Hz, 3 H), 1.15
(d, J ) 4.7 Hz, 3 H); 13C NMR (100 MHz, CDCl3) δ 22.0 (d),
22.1 (d), 39.8 (u), 71.2 (u), 71.4 (d), 72.0 (d), 126.2 (d), 128.2
(d), 129.1 (d), 137.9 (u). Anal. Calcd for C12H18O2 (194.13): C,
74.19; H, 9.34. Found: C, 73.90; H, 9.59. (S)-2a : 1H NMR (400
MHz, CDCl3) δ 7.27 (m, 2 H), 7.21 (m, 3 H), 5.16 (m, 1 H),
3.55 (sept, J ) 6.0 Hz, 1 H), 3.46 (dd, J ) 6.0, 4.4 Hz, 1 H),
3.41 (dd, J ) 5.2, 5.2 Hz, 1 H), 2.98 (dd, J ) 7.4, 6.3 Hz, 1 H),
2.89 (dd, J ) 6.9, 6.9 Hz, 1 H), 2.31 (dd, J ) 4.9, 2.5 Hz, 1 H),
2.28 (dd, J ) 5.2, 2.5 Hz, 1 H), 1.15 (d, J ) 2.5 Hz, 3 H), 1.14
(d, J ) 2.5 Hz, 3 H), 1.08 (tr, J ) 7.7 Hz, 3 H); 13C NMR (100
MHz, CDCl3) δ 9.2 (d), 22.0 (d), 22.0 (d), 27.7 (u), 37.0 (u), 67.9
(u), 72.0 (d), 73.4 (d), 126.2 (d), 128.1 (d), 129.3 (d), 137.2 (u),
(74) Biggs, J .; Chapman, N. B.; Wray, V. J . Chem. Soc B 1971, 50,
10539.

Acylation of Racemic Functionalized Secondary Alcohols
J . Org. Chem., Vol. 66, No. 10, 2001 3395
173.7 (u). Anal. Calcd for C15H22O3 (250.16): C, 71.97; H, 8.86.
Found: C, 72.30; H, 9.01.
(d), 26.5 (u), 27.7 (u), 34.6 (u), 39.1 (u), 73.4 (d), 126.4 (d), 128.2
(d), 129.3 (d), 136.9 (u), 173.6 (u). Anal. Calcd for C14H20O2S
(252.38): C, 66.63; H, 7.99. Found: C, 66.91; H, 8.10.
P LE/MP EG//AMP S-Ca ta lyzed Acyla tion of r a c-2. To a
stirred solution of vinyl propionate (400 mg, 4 mmol) and
alcohol 2 (194 mg, 1 mmol) in n-octane (10 mL) was added
water (80 µL) (0.4% v/v). The mixture was stirred for 30 min,
and PLE/MPEG (71 mg; 290 U) and n-octane (10 mL) were
then added. After the mixture was stirred for 30 min, ami-
nomethylated polystyrene (1.12 g, 1.45 mmol) was added and
rapid stirring was continued for 9 d (49% conversion). Subse-
quently, the organic phase was decanted and the residue was
washed thoroughly with toluene, air-dried and stored for a
reuse. The combined organic phases were treated with MgSO4
(1 g) and NaHCO3 (0.5 g) and the mixture was filtered through
Celite. The filtrate was concentrated in vacuo and the residue
was purified first by Kugelrohr distillation (0.8 mbar, 100 °C)
and then by chromatography (cyclohexane/EtOAc, 3:1) to yield
P LE/MP EG-Ca ta lyzed Acyla tion of r a c-4. To a stirred
solution of vinyl propionate (8.0 g, 80 mmol) and alcohol rac-4
(3.97 g, 20 mmol) in n-octane (100 mL) was added water (0.9
mL). The mixture was stirred for 30 min, and PLE/MPEG (1.27
g, 7600 U) and n-octane (100 mL) were then added. After the
mixture was stirred rapidly for 3 d (51% conversion), MgSO4
(10 g) and NaHCO3 (5 g) were added and the mixture was
filtered through Celite. The filtrate was concentrated in vacuo,
and the residue was purified first by Kugelrohr distillation
(1.6 mbar, 180 °C) and then by chromatography (cyclohexane/
EtOAc, 2:1) to give alcohol (R)-4 (1.50 g, 38%) with g99% ee,
[R]23D ) +15.0 (c 1.76, i-PrOH) and ester 4a (2.00 g, 39%) with
96% ee, [R]23 ) +1.9 (c 2.27, i-PrOH). (R)-4: 1H NMR (300
D
MHz, DMSO-d6) δ 7.12 (m, 2 H), 6.83 (m, 2 H), 4.68 (d, J )
5.4 Hz, 1 H), 3.72 (m, 1 H), 3.71 (s, 3 H), 3.24 (s, 3 H), 3.18 (d,
J ) 5.4 Hz, 2 H) 2.60 (dd, J ) 13.7, 5.4 Hz, 1 H), 2.51 (dd, J
) 13.6, 7.6 Hz, 1 H); 13C NMR (75 MHz, DMSO-d6) δ 39.0 (u),
54.8 (d), 58.2 (d), 70.2 (d), 76.1 (u), 113.4 (d), 130.2 (d), 130.9
(u), 157.4 (u). Anal. Calcd for C11H16O3 (196.25): C, 67.32; H,
8.22. Found: C, 67.09; H, 8.01. (S)-4a : 1H NMR (300 MHz,
CDCl3) δ 7.13 (m, 2 H), 6.82 (m, 2 H), 5.14 (m, 1 H), 3.78 (s, 3
H), 3.40 (m, 2 H), 3.35 (s, 3H), 2.85 (dd, J ) 13.9, 6.9 Hz, 1 H),
2.84 (dd, J ) 13.9, 6.9 Hz, 1 H), 2.49 (q, J ) 7.5 Hz, 2 H), 1.09
(tr, J ) 7.5 Hz, 3 H); 13C NMR (75 MHz, CDCl3) δ 9.1 (d), 27.7
(u), 36.1 (u), 55.2 (d), 59.1 (d), 72.6 (u), 73.3 (d), 113.8 (d), 129.2
(u), 130.4 (d), 158.3 (u), 173.9 (u). Anal. Calcd for C14H20O4
(252.31): C, 66.65; H, 7.99. Found: C, 66.91; H, 8.28.
alcohol (R)-2 (91 mg, 47%) with 80% ee, [R]22 ) +5.2 (c 1.44,
D
MeOH) and ester (S)-2a (111 mg, 44%) with g 99% ee, [R]22
) -9.5 (c 1.38, MeOH).
D
P LE/TGA-Ca ta lyzed Acyla tion of r a c-2. To a stirred
solution of vinyl propionate (800 mg, 8 mmol) and alcohol 2
(388 mg, 2 mmol) in n-octane (20 mL) was added water (80
µL) (0.4% v/v). The mixture was stirred for 30 min, and PLE/
TGA (680 mg, 580 U) was then added. After the mixture was
stirred rapidly for 6 d (55% conversion), the organic phase was
decanted and the residue was washed thoroughly with toluene,
air-dried and stored for a reuse. The combined organic phases
were treated with MgSO4 (1 g) and NaHCO3 (500 mg), and
the mixture was filtered through Celite. The filtrate was
concentrated in vacuo and the residue was purified first by
Kugelrohr distillation (0.8 mbar, 100 °C) and then by chro-
matography (cyclohexane/EtOAc, 3:1) to afford alcohol (R)-2
P LE/BSA//P AM-Ca ta lyzed Acyla tion of r a c-4. To a
stirred solution of vinyl propionate (400 mg, 4 mmol) and
alcohol rac-4 (196 mg, 1 mmol) in toluene (20 mL) was added
water (80 µL) (0.4% v/v). Subsequently, PLE/BSA (83 mg, 360
U) and the water-saturated polyaramide membrane (ca. 4 cm2)
were added under gentle stirring to the mixture. After 15 min,
nearly all of the PLE/BSA had been adsorbed on the membrane
and the solution became clear. After the mixture was stirred
gently for 5 d (47% conversion), the organic phase was
decanted and PLE/BSA//PAM was washed with toluene and
stored for a reuse. The combined organic phases were treated
with MgSO4 (1 g) and NaHCO3 (500 mg), the mixture was
filtered through Celite, and the filtrate was concentrated in
vacuo. Purification of the residue first by Kugelrohr distillation
(0.8 bar, 100 °C) and then by chromatography (cyclohexane/
(149 mg, 38%) with 89% ee, [R]22 ) +5.9 (c 1.38, MeOH) and
D
ester (S)-2a (187 mg, 38%) with 94% ee, [R]22 ) -7.4 (c 1.55,
D
MeOH).
P LE/KCl/K2HP O4-Ca ta lyzed Acyla tion of r a c-2. To a
stirred solution of vinyl propionate (800 mg, 8 mmol) and
alcohol rac-2 (388 mg, 2 mmol) in n-octane (10 mL) was added
water (0.08 mL) (0.4% v/v). The mixture was stirred for 30
min, and PLE/KCl/K2HPO4 (0.644 g, 580 U) and n-octane (10
mL) were then added. After the mixture was stirred rapidly
for 6 d (30% conversion), MgSO4 (1 g) and NaHCO3 (0.5 g) were
added and the mixture was filtered through Celite. The filtrate
was concentrated in vacuo, and the residue was purified first
by Kugelrohr distillation (0.8 mbar, 100 °C) and then by
chromatography (cyclohexane/EtOAc, 3:1) to give alcohol (R)-2
EtOAc, 2:1) gave alcohol (R)-4 (89 mg, 45%) with 69% ee, [R]22
D
(208 mg, 53%) with 39% ee, [R]22 ) +2.15 (c 1.975, MeOH)
D
) +1.96 (c 1.37, MeOH) and ester (S)-4a (104 mg, 41%) with
and ester (S)-2a (107 mg, 21%) with g99% ee, [R]22D ) -9.4 (c
97% ee, [R]22 ) -0.81 (c 1.15, MeOH).
D
1.25, MeOH).
P LE/MP EG//AMP S-Ca ta lyzed Acyla tion of r a c-4. To a
stirred solution of vinyl propionate (400 mg, 4 mmol) and
alcohol rac-4 (196 mg, 1 mmol) in toluene (10 mL) was added
water (80 µL) (0.4% v/v). The mixture was stirred for 30 min,
PLE/MPEG (93 mg, 380 U) and toluene (10 mL) were then
added, and stirring was continued for 30 min. Subsequently,
aminomethylated polystyrene (1.5 g, 1.5 mmol) was added to
the mixture. After the mixture was stirred rapidly for 4 d (43%
conversion), the organic phase was decanted and the residue
was washed thoroughly with toluene, air-dried, and stored for
reuse. The combined organic phases were treated with MgSO4
(1 g) and NaHCO3 (500 mg), the mixture was filtered through
Celite, and the filtrate was concentrated in vacuo. Purification
of the residue first by Kugelrohr distillation (0.8 mbar, 100
°C) and then by chromatography (cyclohexane/EtOAc, 2:1)
afforded alcohol (R)-4 (90 mg, 45%) with 56% ee, [R]22D ) +1.63
(c 1.41, MeOH) and ester (S)-4a (90 mg, 36%) with 97% ee,
P LE/MP EG-Ca ta lyzed Acyla tion of r a c-3. To a solution
of vinyl propionate (8.0 g, 80 mmol) and alcohol rac-3 (3.9 g,
20 mmol) in n-octane (100 mL) was added water (0.9 mL). The
mixture was stirred for 30 min, and PLE/MPEG (1.29 g, 7700
U) and n-octane (100 mL) were then added. After the mixture
was stirred rapidly for 5 d (49% conversion), MgSO4 (10 g) and
NaHCO3 (5 g) were added and the mixture was filtered
through Celite. The filtrate was concentrated in vacuo and the
residue was purified first by Kugelrohr distillation (0.6 mbar,
160 °C) and then by chromatography (cyclohexane/EtOAc, 1:1)
to yield alcohol (R)-3 (1.65 g, 42%) with 89% ee, [R]23 +4.7 (c
D
1.52, EtOH), [R]23 ) -26.9 (c 1.45, CHCl3) and ester (S)-3a
D
(1.67 g, 35%) with 92% ee, [R]23 ) -1.5 (c 1.94, THF). (R)-3:
D
1H NMR (400 MHz, DMSO-d6) δ 7.18-7.29 (m, 5 H), 4.84 (d,
J ) 5.2 Hz, 1 H), 3.78 (sext, J ) 5.8 Hz, 1 H), 2.82 (dd, J )
13.5, 5.0 Hz, 1 H), 2.65 (dd, J ) 13.5, 7.4 Hz, 1 H), 2.52 (m, 4
H), 1.15 (tr, J ) 7.4 Hz, 3 H); 13C NMR (100 MHz, DMSO-d6)
δ 14.7 (d), 25.6 (u), 37.7 (u), 42.0 (u), 70.9 (d), 125.5 (d), 127.7
(d), 129.1 (d), 138.9 (u). Anal. Calcd for C11H16OS (196.31): C,
67.30; H, 8.21. Found: C, 67.50; H, 8.32. (S)-3a : 1H NMR (400
MHz, CDCl3) δ 7.18-7.31 (m, 5 H), 5.17 (trtr, J ) 5.9, 5.4 Hz,
1 H), 2.98 (dd, J ) 13.7, 9.1 Hz, 1 H), 2.97 (dd, J ) 13.8, 5.6
Hz, 1 H), 2.66 (d, J ) 6.0 Hz, 2 H), 2.57 (q, J ) 7.4 Hz, 2 H),
2.28 (q, J ) 7.4 Hz, 2 H), 1.23 (tr, J ) 7.4 Hz, 3 H), 1.08 (tr,
J ) 7.5 Hz, 3 H); 13C NMR (100 MHz, CDCl3) δ 9.1 (d), 14.7
[R]22 ) -0.83 (c 1.15, MeOH).
D
P LE/MP EG-Ca ta lyzed Acyla tion of r a c-6. To a stirred
solution of vinyl propionate (4.0 g, 40 mmol) and alcohol rac-6
(1.49 g, 10 mmol) in n-octane (50 mL) was added water (0.4
mL). The mixture was stirred for 30 min, and PLE/MPEG (0.93
g, 4830 U) and n-octane (50 mL) were then added. After the
mixture was stirred rapidly for 7 d (40% conversion), MgSO4
(5 g) and NaHCO3 (3 g) were added and the mixture was

3396 J . Org. Chem., Vol. 66, No. 10, 2001
Gais et al.
filtered through Celite. The filtrate was concentrated in vacuo
and the residue purified first by Kugelrohr distillation (0.4
mbar, 60 °C) and then by chromatography (cyclohexane/EtOAc,
mg, 600 U) and n-octane (10 mL) were then added. After the
mixture was stirred rapidly for 2 d (52% conversion), MgSO4
(1 g) and NaHCO3 (500 mg) were added and the mixture was
filtered through Celite. The filtrate was concentrated in vacuo
and the residue purified first by Kugelrohr distillation (0.8
mbar, 100 °C) and then by chromatography (cyclohexane/
EtOAc, 3:1) to give alcohol (S)-9 (126 mg, 42%) with 97% ee,
[R]22D ) +23.2 (c 1.35, MeOH) and ester (R)-9a (150 mg, 42%)
1:1) to furnish alcohol (S)-6 (880 mg, 59%) with 61% ee, [R]23
D
) +17.1 (c 1.67, CHCl3) and ester (R)-6a (800 mg, 39%) with
92% ee, [R]23 ) -14.1 (c 1.64, CHCl3). (S)-6: 1H NMR (400
D
MHz, CDCl3) δ 7.20-7.34 (m, 5 H), 4.42 (ddd, J ) 47.0, 9.4,
3.5 Hz, 1 H), 4.32 (ddd, J ) 47.8, 9.4, 6.1 Hz, 1 H), 4.07 (m, 1
H), 2.81 (dd, J ) 6.1, 3.5 Hz, 2 H), 2.14 (s, 1 H); 13C NMR (100
MHz, CDCl3) δ 38.7 (d, J ) 7 Hz, u), 71.2 (d, J ) 19 Hz, d),
85.5 (d, J ) 169 Hz, u), 126.6 (d), 128.5 (d), 129.1 (d), 136.9
(u). Anal. Calcd for C9H11OF (154.18): C, 70.11; H, 7.19.
Found: C, 70.02; H, 7.27. (R)-6a :1H NMR (400 MHz, CDCl3)
δ 7.20-7.34 (m, 5 H), 5.22 (dddtr, J ) 23.1, 7.0, 4.4, 3.0 Hz, 1
H), 4.45 (ddd, J ) 46.7, 10.4, 3.0 Hz, 1 H), 4.37 (ddd, J ) 47.0,
9.9, 4.4 Hz, 1 H), 2.98 (dd, J ) 13.7, 7.0 Hz, 1 H), 2.94 (dd, J
) 13.7, 7.0 Hz, 1 H), 2.34 (qd, J ) 7.4, 3.3 Hz, 2 H), 1.11 (tr,
J ) 7.6 Hz, 3 H); 13C NMR (100 MHz, CDCl3) δ 9.0 (d), 27.6
(u), 35.8 (d, J ) 7 Hz, u), 72.7 (d, J ) 21 Hz, d), 82.4 (d, J )
173 Hz, u), 126.7 (d), 128.4 (d), 129.2 (d), 136.1 (u), 173.6 (u).
Anal. Calcd for C12H15O2F (210.25): C, 68.55; H, 7.19. Found:
C, 68.27; H, 7.36.
with 91% ee, [R]22 ) +10.1 (c 1.87, MeOH). (S)-9: 1H NMR
D
(400 MHz, CDCl3) δ 7.26 (m, 2 H), 6.96 (m, 3 H), 4.22 (m, 1
H), 3.93 (dd, J ) 6.3, 3.0 Hz, 1 H), 3.82 (dd, J ) 7.7, 1.7 Hz,
1 H), 2.36 (s, 1 H), 1.28 (d, J ) 6.3 Hz, 3 H); 13C NMR (100
MHz, CDCl3) δ 19.1 (u), 66.6 (d), 73.5 (u), 114.8 (d), 121.4 (d),
129.7 (d), 158.4 (u). Anal. Calcd for C9H12O2 (152.08): C, 71.03;
H, 7.95. Found: C, 71.18; H, 7.75. (R)-9a : 1H NMR (400 MHz,
CDCl3) δ 7.28 (m, 2 H), 6.95 (m, 3 H), 5.26 (m, 1 H), 4.03 (dd,
J ) 5.8, 4.4 Hz, 1 H), 3.97 (dd, J ) 5.8, 4.4 Hz, 1 H), 2.34 (q,
J ) 7.7, 7.4 Hz, 1 H), 1.36 (d, J ) 6.3 Hz, 3 H), 1.14 (t, J ) 7.4
Hz, 3 H); 13C NMR (100 MHz, CDCl3) δ 9.0 (d), 16.8 (d), 27.7
(u), 37.0 (u), 68.5 (d), 69.9 (u), 114.5 (d), 120.9 (d), 129.3 (d),
158.4 (u), 173.8 (u). Anal. Calcd for C12H16O3 (208.11): C, 69.21;
H, 7.74. Found: C, 68.94; H, 7.52.
P LE/BSA-Ca ta lyzed Acyla tion of r a c-6. To a stirred
solution of vinyl propionate (800 mg, 8 mmol) and alcohol rac-6
(308 mg, 2 mmol) in n-octane (10 mL) was added water (0.08
mL) (0.4% v/v). The mixture was stirred for 30 min, and PLE/
BSA (214 mg, 970 U) and n-octane (10 mL) were then added.
After the mixture was stirred rapidly for 6 d (54% conversion),
MgSO4 (1 g) and NaHCO3 (500 mg) were added and the
mixture was filtered through Celite. The filtrate was concen-
trated in vacuo and the residue purified first by Kugelrohr
distillation (0.4 mbar, 60 °C) and then by chromatography
(cyclohexane/EtOAc, 1:1) to yield alcohol (S)-6 (115 mg, 37%)
Ack n ow led gm en t. Financial support of this work
by the Deutsche Forschungsgemeinschaft (SFB 380
“Asymmetric Synthesis with Chemical and Biological
Methods” and TFB 11 “Stereoselective Synthesis of
Biologically Active Compounds”) is gratefully acknowl-
edged. We thank Dr. A. Liese, FZ J u¨lich, for providing
us with the polyaramide membrane.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures for the synthesis of compounds rac-2-7 and rac-2a -
7a , for the acylation of alcohols rac-3 and rac-6-10, and for
the determination of enantiomer composition. This material
is available free of charge via the Internet at http://pubs.acs.org.
with g99% ee, [R]23 ) +4.5 (c 1.12, MeOH) and ester (R)-6a
D
(188 mg, 45%) with 68% ee, [R]23 ) -5.1 (c 1.28, MeOH).
D
P LE/BSA-Ca ta lyzed Acyla tion of r a c-9. To a solution of
vinyl propionate (800 mg, 8 mmol) and alcohol rac-9 (304 mg,
2 mmol) in n-octane (10 mL) was added water (80 µL) (0.4%
v/v). The mixture was stirred for 30 min, and PLE/BSA (130
J O0016881