Resolution of non-protein amino acids via the microbial protease-catalyzed enantioselective hydrolysis of their N-unprotected esters

Article abstract of DOI:10.1016/j.tet.2005.08.033

In the Aspergillus oryzae protease-catalyzed ester hydrolysis, substitution of N-unprotected amino acid esters for the corresponding N-protected amino acid esters resulted in a large enhancement of the hydrolysis rate, while the enantioselectivity was deteriorated strikingly when the substrates employed were the conventional methyl esters. This difficulty was overcome by employing esters bearing a longer alkyl chain such as the isobutyl ester. Utilizing this ester, amino acids carrying an aromatic side chain were resolved with excellent enantioselectivities (E=50 to >200). With amino acids bearing an aliphatic side chain also, good results in terms of the hydrolysis rate and enantioselectivity were obtained by employing such an ester as the isobutyl ester. Moreover, the enantioselectivity proved to be enhanced further by conducting the reaction at low temperature. This procedure was applicable to the case where the enantioselectivity was not high enough even by the use of the isobutyl ester.

Full text of DOI:10.1016/j.tet.2005.08.033

Tetrahedron 61 (2005) 10254–10261
Resolution of non-protein amino acids via the microbial
protease-catalyzed enantioselective hydrolysis of their
N-unprotected esters
*
Toshifumi Miyazawa, Kiwamu Imagawa, Hiroe Minowa, Toyoko Miyamoto
and Takashi Yamada
Department of Chemistry, Faculty of Science and Engineering, Konan University, Higashinada-ku, Kobe 658-8501, Japan
Received 11 July 2005; revised 8 August 2005; accepted 8 August 2005
Available online 26 August 2005
Abstract—In the Aspergillus oryzae protease-catalyzed ester hydrolysis, substitution of N-unprotected amino acid esters for the
corresponding N-protected amino acid esters resulted in a large enhancement of the hydrolysis rate, while the enantioselectivity was
deteriorated strikingly when the substrates employed were the conventional methyl esters. This difficulty was overcome by employing esters
bearing a longer alkyl chain such as the isobutyl ester. Utilizing this ester, amino acids carrying an aromatic side chain were resolved with
excellent enantioselectivities (EZ50 to O200). With amino acids bearing an aliphatic side chain also, good results in terms of the hydrolysis
rate and enantioselectivity were obtained by employing such an ester as the isobutyl ester. Moreover, the enantioselectivity proved to be
enhanced further by conducting the reaction at low temperature. This procedure was applicable to the case where the enantioselectivity was
not high enough even by the use of the isobutyl ester.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
Aspergillus oryzae protease (Amano protease A) and
Bacillus subtilis protease (Amano protease N) for the
resolution of non-protein amino acids via the enantio-
selective hydrolysis of their N-protected esters.5 Recently,
we have also reported the utilization of Aspergillus melleus
protease (Amano protease P) and A. oryzae protease as
catalysts for peptide bond formation via the kinetically
controlled approach6 and of B. subtilis protease via the
thermodynamically controlled approach.7 On these lines,
we have preliminarily reported on the resolution of non-
protein amino acids via the enantioselective hydrolysis of
their N-unprotected esters employing A. oryzae protease.8,9
During the course of the investigation, we have found that
the enantioselectivity of the hydrolysis can be significantly
improved by choosing an appropriate ester grouping and by
lowering the reaction temperature. In this paper, we describe
further relevant details.
Microbial proteases from a variety of sources are
commercially available. They have been employed mainly
for food processing and as detergent ingredients. Although
they possess some properties favorable as synthetic
catalysts—they are inexpensive, stable and easy to handle,
and they require no added cofactors, they have rarely been
exploited for organic syntheses, except subtilisins, the most
important members of which are subtilisin Carlsberg (from
Bacillus licheniformis) and subtilisin BPN0 (from Bacillus
amyloliquefaciens). Not only racemic amino acid esters but
also other racemic carboxylic acid esters have been resolved
with subtilisins.1 Furthermore, the former subtilisin has
been utilized for the transesterification between N-protected
amino acid esters and alcohols2 and for the regioselective
acylation of polyhydroxy compounds such as carbo-
hydrates,3 because of its unique robustness in organic
solvents. The latter subtilisin, including its variants, has
frequently been applied to peptide synthesis.4 We also have
been making a search for the effective use of microbial
proteases in organic syntheses. We exploited previously
Besides some 20 amino acids universally distributed as
protein constituents in living organisms, there are hundreds
of other amino acids of non-protein origin.10 Homochiral
non-protein amino acids, including those synthetic com-
pounds which have not been found in nature, are useful
building blocks for the synthesis of analogs of biologically
active peptides such as toxins, antibiotics, hormones, and
enzyme inhibitors11 and versatile chiral starting materials or
Keywords: Resolution; Non-protein amino acids; Aspergillus oryzae
protease; Enantioselective hydrolysis; Isobutyl ester; Low temperature.
Corresponding author. Tel.: C81 784314341; fax: C81 784352539;
e-mail: miyazawa@base2.ipc.konan-u.ac.jp
*
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.08.033

T. Miyazawa et al. / Tetrahedron 61 (2005) 10254–10261
10255
chiral auxiliaries for other synthetic purposes.12 For the
supply of non-protein amino acids in quantity, the chemical
synthesis of racemic forms followed by their optical
resolution is still the preferable way, though a number of
methods have been developed for the asymmetric syntheses
of amino acids.13 For the resolution of racemic amino acids,
enzymatic methods have often been employed besides
chemical methods based on the formation of diastereomeric
salts or derivatives.14,15 Hydrolytic enzymes have often
been used for the purpose. Acylase I (aminoacylase;
N-acylamino acid amidohydrolase) is the most frequently
used enzyme in the chemoenzymatic preparation of the
L (^S) forms of amino acids from the racemic N-acyl
derivatives. An extensive study has been conducted of the
substrate specificity of porcine kidney acylase and the mold
enzyme from A. oryzae, especially for the resolution of
unnatural and rarely occurring amino acids.16 However, the
most commonly used enzymes such as porcine renal acylase
I are not always applicable to the resolution of non-protein
amino acids. Accordingly, it is worthwhile to develop novel
enzymatic resolution methods for them. Ester hydrolysis has
been utilized for the enzyme-based resolution procedures
for amino acids.17 The mammalian protease a-chymo-
trypsin is one of the most popular enzymes which have been
employed for such a purpose. We also have exploited some
microbial lipases and porcine pancreatic lipase for the
resolution of non-protein amino acids through the enantio-
selective hydrolysis of their N-protected esters.18 The
success of the lipase-mediated methods prompted us to
examine microbial proteases for the same purpose.
Initially, the methyl esters of several phenylalanines bearing
a halogen atom on the benzene ring (1, RZX-PhCH2) were
subjected to hydrolysis with A. oryzae protease19 at the
initial pH of 7 and at 30 8C (Scheme 1). After the desired
degree of conversion (ca. 40%), the liberated amino acid (2)
was isolated and its enantiomeric excess (ee) was
determined by HPLC analysis on a chiral ligand-exchange
column.20 The results are compiled in Table 1. As
anticipated, the hydrolysis of the methyl esters of these
N-unprotected, halogenated phenylalanines proceeded
much more smoothly than that of the corresponding
N-protected methyl esters. Thus, ca. 40% conversion was
reached after dozens of minutes instead of dozens of hours
for the N-protected derivatives. However, the enantio-
selectivities deteriorated strikingly compared with those
observed with the corresponding N-protected derivatives,21
as judged from the values of enantiomeric ratio (E).22
Accordingly, the influence of ester groupings was examined
next using 4-fluorophenylalanine (1, RZ4-F-PhCH2) as a
model amino acid in the expectation of a possible
enhancement of enantioselectivity even at the cost of
some reduction in the hydrolysis rate (entries 1–5, Table 1).
In the resolution of carboxylic acids via the enzyme-
catalyzed ester hydrolysis, the methyl or ethyl ester is
ordinarily a preferable one, because esters carrying a longer
alkyl chain are usually hydrolyzed at reduced rates.
Activated esters such as the 2-chloroethyl or 2,2,2-
trifluoroethyl ester are sometimes employed in order to
enhance the rate of ester hydrolysis. On the other hand, the
influence of the alcohol moiety of esters on the enantio-
selectivity of hydrolysis is not fully understood: relevant
data have scarcely been accumulated in the literature.23 As
can be seen from the table, the use of the isopropyl ester
resulted in a large retardation of hydrolysis with a slight
improvement of enantioselectivity (entry 2). In the case of
the benzyl ester, the hydrolysis proceeded very smoothly
with a distinct enlargement in the E value (entry 5). It was
gratifying to find that the use of the n-butyl (entry 3) or the
isobutyl ester (entry 4) resulted in no retardation of the
hydrolysis and yet a marked enhancement of enantio-
selectivity. Especially with the n-butyl ester, the hydrolysis
proceeded in an almost enantiospecific manner, the E value
being evaluated to be O200.24 Of these two esters, the
isobutyl ester was chosen in order to avoid the handling of
the stinking n-butyl alcohol in the preparation of n-butyl
esters and after their hydrolysis. Thus, the resolution of
other halogenated phenylalanines was examined by employ-
ing their isobutyl esters as substrates. A marked enhance-
ment of enantioselectivity was achieved by replacing the
methyl ester with the isobutyl ester (entries 7, 9, and 11).
The reaction was changed from almost non-enantiospecific
to specific with 4-chlorophenylalanine. Excellent enantio-
selectivities were attained also in all the other cases
employing the isobutyl esters as substrates (entries 12–l4).
The marked enhancement of enantioselectivity by switching
the methyl ester to the isobutyl ester observed here imply
that the role of the ester moiety becomes relatively
important in the substrate recognition by the protease
when an acylamino group is replaced by a free amino group.
2. Results and discussion
As stated above, we investigated previously the resolution
of non-protein amino acids via the enantioselective
hydrolysis of the methyl esters of their N-benzyloxycar-
bonyl (Z) derivatives using two microbial proteases from
B. subtilis (Amano protease N) and A. oryzae (Amano
protease A), the former protease generally yielding better
results than the latter one.5a When B. subtilis protease was
employed, amino acids bearing an aliphatic side chain as
well as those with an aromatic side chain were resolved with
good to excellent enantioselectivities in phosphate buffer
containing 20% (v/v) DMF. In some cases, however, the
hydrolysis rates were rather slow, which makes the method
less practical. Since the N-Z-amino acid methyl esters used
as substrates generally had low solubilities in water, the
addition of such a polar solvent as DMF was inevitable,
which should be responsible for the often observed slow
hydrolysis. Consequently, we intended to examine amino
acid esters bearing a free a-amino group as substrates for the
microbial protease-catalyzed hydrolysis which have better
solubilities than the corresponding N-protected derivatives
in the expectation of an enhancement of the hydrolysis rate.
Furthermore, the influence of ester groupings was also
explored with N-unprotected amino acids bearing an
Scheme 1. A. oryzae protease-catalyzed hydrolysis of amino acid esters (1).
See Tables 1 and 2 for R and R0.

10256
T. Miyazawa et al. / Tetrahedron 61 (2005) 10254–10261
Table 1. A. oryzae protease-catalyzed hydrolysis of aromatic amino acid esters (1)a
R0
Entry
R
Conversion (%)
Time (min)
eeP (%)b
E
1
2
3
4
5
6
7
8
4-F-PhCH2
4-F-PhCH2
4-F-PhCH2
4-F-PhCH2
4-F-PhCH2
3-F-PhCH2
3-F-PhCH2
2-F-PhCH2
2-F-PhCH2
4-Cl-PhCH2
4-Cl-PhCH2
2-Cl-PhCH2
4-Br-PhCH2
PhCH2
Me
48
35
34
47
39
23
33
24
44
43
46
38
52
31
79
230
24
25
5
30
30
45
55
59
30
20
25
24
47
67
99
97
91
84
98
7.5
92
46
98
99
91
O99
4.2
7.1
O200
170
37
15
150
1.2
50
3.7
O200
O200
120
O200
i-Prc
n-Buc
i-Buc
Bzlc
Me
i-Bu
Me
i-Bu
Me
i-Bu
i-Bu
i-Bu
i-Bu
9
10
11
12
13
14
a Reactions were conducted as described in the Section 4 using 0.2 mmol of DL-1 in the form of the hydrochloride (or hydrobromide).
b Enantiomeric excess of the liberated amino acid.
c Hydrobromide.
aliphatic side chain. The results obtained when a series of
esters of norvaline (2-aminopentanoic acid; 1, RZn-Pr)
were subjected to hydrolysis with A. oryzae protease
(Table 2). The pH of the reaction mixture was maintained
at 7 by automatic titration with 0.1 M NaOH during the
hydrolysis. DMSO (20%, v/v) was added to the phosphate
buffer to increase the solubility of the alkyl esters larger than
the n-butyl ester (entries 9–12). In the case of the n-pentyl
ester, it was ascertained that the addition of DMSO had no
harmful effects on the enzymatic hydrolysis: the reaction
was little retarded and the enantioselectivity was not
affected at all (entries 8 and 9). When normal alkyl esters
are compared, the enantioselectivity was increased almost
progressively with the length of the ester alkyl chain from
methyl to n-heptyl (entries 1–6, and 8–11). The longer
n-octyl ester (entry 12) yielded an E value as low as that
obtained with the methyl ester, and in addition, the reaction
was extremely slow. Of all the esters examined, the isobutyl
ester gave an enantioselectivity of the highest rank even in
the shortest reaction time. Therefore, this ester was chosen
for further studies, making it easier to compare the results
with those with the aromatic amino acids mentioned above.
Thus, rather high enantioselectivities were attained in the
hydrolysis of the isobutyl esters of other aliphatic amino
acids (entries 16, 19, 21, and 22). However, they were lower
than those observed with the aromatic amino acid
derivatives. As stated below, we found that conducting the
hydrolysis at low temperature was quite effective in
improving further the enantioselectivity.
In all the cases mentioned above, the preferential hydrolysis
of the L-enantiomers was confirmed by comparison with
Table 2. A. oryzae protease-catalyzed hydrolysis of aliphatic amino acid esters (1)a
R0
Entry
R
Conversion (%)
Time (min)
eeP (%)b
E
1
2
3
4
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Pr
n-Bu
n-Bu
n-Bu
n-Bu
n-Bu
Et
Me
Et
n-Pr
i-Pr
n-Bu
i-Bu
i-Bu
n-Pentyl
n-Pentyl
n-Hexyl
n-Heptyl
n-Octyl
Me
Et
n-Pr
i-Bu
i-Bu
Me
i-Bu
i-Bu
40
40
40
40
40
40
40
40
40
42
40
14
40
39
39
40
41
40
40
40
40
33
33
170
280
250
510
150
77
400
180
190
120
96
23 h
160
200
170
140
19.7 h
390
190
22.3 h
320
280
14.5 h
58
72
75
82
88
92
97
91
91
92
94
70
53
67
86
95
98
38
78
95
94
89
77
5.5
10
12
18
30
45
130
40
40
50
63
5.9
4.6
7.8
24
5
6
7c
8
9d
10d
11d
12d
13
14
15
16
17c
18
19
20c
21
22
23
76
O200
5.5
14
78
66
Et
Et
i-Bu
n-Pentyl
i-Pentyl
i-Bu
i-Bue
i-Bue
27
11
a Reactions were conducted as described in the Section 4 using 0.4 mmol of DL-1 in the form of the hydrochloride (or tosylate) in 0.1 M phosphate buffer
(pH 7.0) without DMSO at 30 8C, unless otherwise noted.
b Enantiomeric excess of the liberated amino acid.
c Reactions were conducted at 5 8C.
d Reactions were conducted in 0.1 M phosphate buffer (pH 7.0) containing 20% (v/v) DMSO.
e Tosylate.

T. Miyazawa et al. / Tetrahedron 61 (2005) 10254–10261
10257
authentic samples prepared form the optically active amino
acids, if available, on HPLC or suggested from the
regularity of elution order of the enantiomers on HPLC.20
This stereochemical preference is the same as that observed
in the hydrolysis of racemic N-Z-amino acid 2-chloroethyl
or 2,2,2-trifluoroethyl esters mediated by lipases18 or
proteases.5,25
acid; 1, RZn-Bu): the reaction was almost enantiospecific
at 5 8C (entry 17).
3. Conclusion
In this work, we have shown that in the hydrolysis of amino
acid esters with a free a-amino group catalyzed by A. oryzae
protease, the enantioselectivity can be enhanced greatly by
employing esters with a longer alkyl chain such as the
isobutyl ester. The enhancement of enantioselectivity by
modifying the alcohol moiety of the substrate ester seems to
be rather general, and this possibility should be considered
when the enantioselectivity is inadequate with the methyl or
ethyl ester conventionally used in the ester hydrolysis
mediated by hydrolytic enzymes. Furthermore, we have
shown that the enantioselectivity of the hydrolysis reaction
catalyzed by the microbial protease can be enhanced
significantly by conducting the reaction at low temperature.
In cases in which Tr exists above the ordinary temperature
(this must be often the case), there must be such a
possibility. The relevant results accumulated recently
support this idea. In the reactions conducted in water with
organic cosolvents or in organic solvents, it is usually easy
to lower the reaction temperature even below 0 8C, which
may lead to a significant enhancement of enantioselectivity.
As the majority of enzymes used for biotransformations are
not thermally stable, reactions at higher temperatures are
largely restricted, while lowering temperature seems to have
less operational limits.
We examined next the possibility of improving the
enantioselectivity by lowering temperature in the A. oryzae
protease-catalyzed hydrolysis of the isobutyl esters of
2-aminobutanoic acid (1, RZEt) and norvaline.26 It is
widely believed that in enzymatic reactions, as in the
conventional chemical reactions, lower temperatures lead to
higher enantioselectivity; this has been supported by a
number of experimental results.27,28 A rational under-
standing of such a temperature effect on enantioselectivity
has been proposed.29 When KRT ln E is plotted against
absolute temperature (T), a straight line should be obtained
(KRT ln EZDDH‡K TDDS‡).30 If the racemic temperature
(Tr), at which no enantiomeric discrimination occurs, is high
above the ordinary temperature, the enantioselectivity will
increase with decreasing temperature. As shown in Figure 1,
when KRT ln E obtained from the E value at each
temperature was plotted against T, straight lines were
obtained for the hydrolysis of the isobutyl esters of both the
amino acids. The temperature range examined here was
5–45 8C. The existence of a linear relationship between
KRT ln E and T indicates that the temperature effect on the
enantioselectivity of the enzymatic hydrolyses is governed
by the above-mentioned equation and the enzyme retains its
active conformation within the temperature range
examined. In both the cases, Tr’s were high (80 8C for
2-aminobutatnoic acid; 93 8C for norvaline); thus, when the
reaction temperature was lowered from 30 to 5 8C, the
hydrolysis time became much longer, but a largely
enhanced enantioselectivity was obtained in each case
(entries 20 and 7, respectively, in Table 2). This is also the
case with the isobutyl ester of norleucine (2-aminohexanoic
4. Experimental
4.1. General
1H NMR spectra were obtained at 300 MHz on a Varian
Unity 300 spectrometer using DMSO-d6 as a solvent with
TMS as an internal standard unless otherwise noted. Mp
were determined on a Yamato MP-21 apparatus and are
uncorrected. Optical rotations were measured using a
JASCO DIP-4 digital polarimeter. All organic solvents
were distilled following standard protocols and dried over
molecular sieves prior to use. Petroleum ether refers to the
fraction with distillation range 30–70 8C. Protease from A.
oryzae (protease A) was supplied by Amano Pharmaceutical
Co. (Japan) and had a specific activity of 10 U/mg solid
(pH 7.0) (one Amano unit is defined as the enzyme quantity
which hydrolyzes casein to produce color equivalent to
400 mg of L-Tyr per 60 min at 37 8C, according to the
supplier).
4.2. Preparation of amino acid esters
1
The physical and H NMR data of the amino acid esters
prepared are compiled in Table 3. Elemental analysis data of
all the new compounds are available as Supplementary
material.
4.2.1. As hydrochlorides (or tosylates). The method for
preparing amino acid methyl esters35 was modified as
illustrated below for the preparation of DL-2-aminobutanoic
acid isobutyl ester hydrochloride. Thionyl chloride (16 ml)
Figure 1. Influence of temperature (T) on the difference in the activation
free energy (DDG‡ZKRT ln E) between the enantiomers for the A. oryzae
protease-catalyzed hydrolysis of isobutyl esters of 2-aminobutanoic acid
(square) and norvaline (circle).

10258
T. Miyazawa et al. / Tetrahedron 61 (2005) 10254–10261
Table 3. Physical and 1H NMR data of DL-amino acid esters
DL-amino acid ester (1) hydro-
chloride
R0
R
Mp (8C)a
1H NMR dH (DMSO-d6)
4-F-PhCH2
Me
169–170 [C]
3.03–3.17 (2H, m), 3.67 (3H, s), 4.26 (1H, t, JZ6.6 Hz), 7.11–7.28 (4H, m), 8.47
(3H, s)
4-F-PhCH2
4-F-PhCH2
4-F-PhCH2
i-Prb
198–199 [C]
116–117 [C]
127–129 [C]
1.22 (3H, d, JZ6.3 Hz), 1.23 (3H, d, JZ6.3 Hz), 3.02–3.17 (2H, m), 3.87 (1H, t,
JZ6.3 Hz), 4.97 (1H, septet, JZ6.3 Hz), 7.13–7.30 (4H, m), 8.59 (3H, s)
0.89 (3H, t, JZ7.2 Hz), 1.20–1.50 (2H, m), 1.51–1.66 (2H, m), 3.03–3.17 (2H, m),
4.08–4.20 (2H, m), 4.26 (1H, t, JZ6.6 Hz), 7.10–7.30 (4H, m), 8.57 (3H, s)
0.76 (3H, d, JZ6.6 Hz), 0.77 (3H, d, JZ6.6 Hz), 1.60–1.81 (1H, m), 3.06–3.27
(2H, d of ABq, JZ14.0, 8.0, 5.7 Hz), 3.75–3.86 (2H, m), 4.33 (1H, dd, JZ8.0,
5.7 Hz), 7.09–7.31 (4H, m), 8.70 (3H, s)
3.03–3.18 (2H, m), 4.27 (1H, t, JZ6.6 Hz), 5.23 (1H, s), 7.11–7.28 (4H, m), 7.33–
7.45 (5H, m), 8.67 (3H, s)
3.09–3.21 (2H, m), 3.67 (3H, s), 4.31 (1H, t, JZ6.6 Hz), 7.05–7.37 (4H, m), 8.58
(3H, s)
n-Bub
i-Bub
4-F-PhCH2
3-F-PhCH2
3-F-PhCH2
Bzlb
Me
199–201 [C]
160–162 [C]
137–138.5 [D]
i-Bu
0.76 (3H, d, JZ6.9 Hz), 0.77 (3H, d, JZ6.9 Hz), 1.66–1.83 (1H, m), 3.05–3.25
(2H, d of ABq, JZ14.0, 7.8, 5.7 Hz), 3.78–3.87 (2H, m), 4.33 (1H, dd, JZ7.8,
5.7 Hz), 7.06–7.40 (4H, m), 8.70 (3H, s)
2-F-PhCH2
2-F-PhCH2
Me
159–160 [D]
139–142 [D]
3.08–3.24 (2H, m), 3.62 (3H, s), 4.16 (1H, dd, JZ6.0, 5.7 Hz), 7.13–7.37 (4H, m),
8.65 (3H, s)
i-Bu
0.71 (3H, d, JZ6.6 Hz), 0.73 (3H, d, JZ6.6 Hz), 1.60–1.76 (1H, m), 3.06–3.26
(2H, d of ABq, JZ14.0, 8.7, 5.7 Hz), 3.70–3.86 (2H, d of ABq, JZ10.5, 6.6 Hz),
4.33 (1H, dd, JZ8.7, 5.7 Hz), 7.14–7.37 (4H, m), 8.74 (3H, s)
3.05–3.18 (2H, m), 3.67 (3H, s), 4.27 (1H, t, JZ6.6 Hz), 7.22–7.40 (4H, m), 8.47
(3H, s)
0.76 (3H, d, JZ6.6 Hz), 0.77 (3H, d, JZ6.6 Hz), 1.65–1.82 (1H, m), 3.02–3.23
(2H, m), 3.77–3.87 (2H, m), 4.24–4.30 (1H, m), 7.25–7.40 (4H, m), 8.70 (3H, s)
0.76 (3H, d, JZ6.6 Hz), 0.77 (3H, d, JZ6.6 Hz), 1.66–1.82 (1H, m), 3.00–3.23
(2H, m), 3.76–3.87 (2H, m), 4.25–4.30 (1H, m), 7.21–7.41 (4H, m), 8.69 (3H, s)
0.78 (3H, d, JZ6.6 Hz), 0.79 (3H, d, JZ6.6 Hz), 1.67–1.85 (1H, m), 3.00–3.20
(2H, d of ABq, JZ14.1, 7.8, 6.0 Hz), 3.77–3.88 (2H, m), 4.25–4.29 (1H, m), 7.18–
7.52 (4H, m), 8.54 (3H, s)
4-Cl-PhCH2
4-Cl-PhCH2
2-Cl-PhCH2
4-Br-PhCH2
Me
183–184 [C], lit.31 177–178
149–151 [D]
i-Bu
i-Bu
i-Bu
128–131 [D]
162–164 [C]
PhCH2
i-Bu
129–131 [B]
0.74 (3H, d, JZ6.6 Hz), 0.75 (3H, d, JZ6.6 Hz), 1.64–1.82 (1H, m), 3.00–3.24
(2H, d of ABq, JZ13.8, 8.1, 5.7 Hz), 3.75–3.86 (2H, d of ABq, JZ10.1, 6.1 Hz),
4.26 (1H, dd, JZ8.1, 5.7 Hz), 7.22–7.35 (5H, m), 8.63 (3H, s)
0.85 (3H, t, JZ7.4 Hz), 1.21–1.45 (2H, m), 1.74 (2H, q-like, JZ ca. 7.4 Hz), 3.72
(3H, s), 3.95 (1H, t, JZ6.3 Hz), 8.62 (3H, s)
0.87 (3H, t, JZ7.2 Hz), 1.22 (3H, t, JZ7.2 Hz), 1.20–1.50 (2H, m), 1.76 (2H, q-
like, JZ ca. 7.5 Hz), 3.92 (1H, t, JZ6.3 Hz), 4.11–4.26 (2H, m), 8.62 (3H, s)
0.87 (3H, t, JZ7.2 Hz), 0.90 (3H, t, JZ7.2 Hz), 1.20–1.51 (2H, m), 1.62 (2H,
sextet, JZ7.0 Hz), 1.70–1.82 (2H, m), 3.96 (1H, t, JZ6.3 Hz), 4.04–4.17 (2H, m),
8.58 (3H, s)
n-Pr
n-Pr
n-Pr
Me
Et
116–117 [C], lit.32 116–117
66–68 [D], lit.33 ca. 65c
68–71 [E]d
n-Pr
n-Pr
n-Pr
i-Pr
96–97 [F]
76–77 [D]
0.87 (3H, t, JZ7.2 Hz), 1.22 (3H, d, JZ6.3 Hz), 1.23 (3H, d, JZ6.3 Hz),
1.20–1.50 (2H, m), 1.69–1.81 (2H, m), 3.87 (1H, t, JZ6.3 Hz), 4.98 (1H, septet,
JZ6.3 Hz), 8.60 (3H, s)
0.87 (3H, t, JZ7.2 Hz), 0.89 (3H, t, JZ7.2 Hz), 1.20–1.50 (4H, m), 1.51–1.65
(2H, m), 1.70–1.81 (2H, m), 3.96 (1H, t, JZ6.3 Hz), 4.08–4.21 (2H, m), 8.55 (3H,
s)
n-Bu
n-Pr
n-Pr
i-Bu
84–86 [A]
78–79 [C]
0.86 (3H, t, JZ7.2 Hz), 0.89 (6H, d, JZ6.6 Hz), 1.22–1.48 (2H, m), 1.76 (2H, q-
like, JZ ca. 7.5 Hz), 1.83–1.96 (1H, m), 3.87–3.98 (3H, m), 8.60 (3H, s)
0.86 (3H, t, JZ7.2 Hz), 0.87 (3H, t, JZ7.2 Hz), 1.19–1.50 (6H, m), 1.55–1.64
(2H, m), 1.69–1.81 (2H, m), 3.96 (1H, t, JZ6.3 Hz), 4.07–4.21 (2H, m), 8.54 (3H,
s)
n-Pentyl
n-Pr
n-Pr
n-Pr
n-Hexyl
n-Heptyl
n-Octyl
78–79 [C]
69–72 [B]
73–74 [G]
0.86 (3H, t, JZ7.2 Hz), 0.87 (3H, t, JZ7.2 Hz), 1.19–1.50 (8H, m), 1.55–1.64
(2H, m), 1.70–1.81 (2H, m), 3.96 (1H, t, JZ6.3 Hz), 4.07–4.21 (2H, m), 8.55 (3H,
s)
0.84 (3H, t, JZ7.2 Hz), 0.86 (3H, t, JZ7.2 Hz), 1.18–1.48 (10H, m), 1.53–1.63
(2H, m), 1.68–1.79 (2H, m), 3.96 (1H, t, JZ6.3 Hz), 4.06–4.20 (2H, m), 8.51 (3H,
s)
0.85 (3H, t, JZ7.2 Hz), 0.87 (6H, d, JZ6.6 Hz), 1.19–1.50 (12H, m), 1.55–1.64
(2H, m), 1.70–1.81 (2H, m), 3.96 (1H, t, JZ6.3 Hz), 4.06–4.21 (2H, m), 8.54 (3H,
s)
0.85 (3H, t, JZ6.9 Hz), 1.17–1.42 (4H, m), 1.71–1.83 (2H, m), 3.73 (3H, s), 3.96
(1H, t, JZ6.3 Hz), 8.54 (3H, s)
0.86 (3H, t, JZ6.9 Hz), 1.22 (3H, t, JZ7.2 Hz), 1.17–1.43 (4H, m), 1.72–1.84
(2H, m), 3.93 (1H, t, JZ6.3 Hz), 4.11–4.27 (2H, m), 8.55 (3H, s)
0.86 (3H, t, JZ6.9 Hz), 0.90 (3H, t, JZ7.2 Hz), 1.17–1.42 (4H, m), 1.63 (2H,
sextet, JZ7.0 Hz), 1.72–1.83 (2H, m), 3.93 (1H, t, JZ6.3 Hz), 4.04–4.18 (2H, m),
8.54 (3H, s)
n-Bu
n-Bu
n-Bu
Me
Et
117–119 [C], lit.32 122–123
81–82 [D]
n-Pr
82–84 [E]
n-Bu
i-Bu
56–57 [B]
0.84 (3H, t, JZ7.2 Hz), 0.89 (6H, d, JZ6.6 Hz), 1.17–1.43 (4H, m), 1.78 (2H, q-
like, JZ ca. 7.3 Hz), 1.82–1.96 (1H, m), 3.87–3.99 (3H, m), 8.55 (3H, s)
0.89 (3H, t, JZ7.5 Hz), 1.82 (2H, d of q, JZ7.5, 6.0 Hz), 3.72 (3H, s), 3.94 (1H, t,
JZ6.0 Hz), 8.65 (3H, s)
Et
Me
146–148 [C], lit.34 139

T. Miyazawa et al. / Tetrahedron 61 (2005) 10254–10261
10259
Table 3 (continued)
DL-amino acid ester (1) hydro-
chloride
R0
R
Mp (8C)a
1H NMR dH (DMSO-d6)
Et
i-Bu
72–73 [A]
0.90 (6H, d, JZ6.3 Hz), 0.92 (3H, t, JZ7.5 Hz), 1.80–1.97 (3H, m), 3.88–4.00
(3H, m), 8.60 (3H, s)
i-Bu
i-Bu
i-Bu
i-Bu
106–108 [B]
oile
0.87–0.90 (12H, m), 1.63 (2H, t, JZ7.0 Hz),1.67–1.79 (1H, m), 1.81–1.97 (1H,
m), 3.88–3.971 (3H, m), 8.55 (3H, s)
n-Pentyl
i-Pentyl
0.85 (3H, t, JZ6.9 Hz), 0.90 (6H, d, JZ6.9 Hz), 1.18–1.46 (6H, m), 1.73–1.83
(2H, m), 1.83–1.97 (1H, m), 3.87–3.99 (3H, m), 8.63 (3H, s)
0.84 (3H, d, JZ6.6 Hz), 0.85 (3H, d, JZ6.6 Hz), 0.91 (6H, d, JZ6.6 Hz),
1.06–1.37 (2H, m), 1.44–1.58 (1H, m), 1.72–1.99 (3H, m), 3.88–4.01 (3H, m),
8.54 (3H, s)
116–118 [B]f
a Recrystallization solvent: A, EtOAc–ether; B, ether–petroleum ether; C, MeOH–ether; D, EtOH–ether; E, 1-propanol–ether; F, 2-propanol–ether; G,
1-octanol.
b Hydrobromide.
c Tosylate, mp 123–124 8C (D).
d Tosylate, mp 85–86 8C (C).
e Tosylate, mp 103–104 8C (D).
f
Tosylate, mp 132–134 8C (D).
was added dropwise below 0 8C under stirring to isobutyl
alcohol (2-methyl-1-propanol; 98 ml) which had been pre-
cooled to K10 8C. After 15 min the DL-amino acid (4.46 g)
was added and the reaction mixture was stirred at 50 8C for
60 h. The mixture was evaporated under reduced pressure,
and the residue was recrystallized from ethyl acetate–
petroleum ether to yield white needles; 7.25 g (86%); mp
66–68 8C. Other DL-amino acid ester hydrochlorides were
prepared in essentially the same manner and purified by
recrystallization from an appropriate solvent (shown in the
table). In some cases, the hydrochlorides obtained were
converted to the tosylates which became crystalline more
easily.
The resulting mixture was shaken (180 strokes per minute)
at 30 8C. The reaction was monitored by HPLC on a
Nucleosil 5C8 column (4.6 mm I.D.!150 mm; Chemco
Scientific Co., Japan). The liquid chromatograph employed
was a Shimadzu LC-10AS instrument equipped with a
Rheodyne 7725i sample injector, an SPD-10A variable
wavelength UV monitor and a Chromatopac C-R6A data
processor under the following conditions: mobile phase, 10–
30% (v/v) solution of CH3CN in water containing
trifluoroacetic acid (0.1%); flow rate, 1.0 ml minK1; column
temperature, 30 8C; detection, UV at 254 nm. After the
desired degree of conversion (ca. 40%), the reaction mixture
was adjusted to pH 9 with 0.5 M NaOH and extracted
several times with ethyl acetate to remove the unreacted
ester. The aqueous layer was adjusted to pH 7 with 1 M HCl
and evaporated to dryness under reduced pressure. The
residue was dissolved in a small amount of distilled water
and analyzed by chiral HPLC (vide infra) to obtain the ee
value of the liberated amino acid. The results are shown in
Table 1.
4.2.2. As hydrobromides. The preparation of DL-4-
fluorophenylalanine isobutyl ester hydrobromide is
described as a typical example. N-Z-DL-4-fluorophenyl-
alanine36 (252 mg) was reacted with isobutyl alcohol (70 ml)
in the presence of 1-ethyl-3-(3-dimethylaminopropyl)-
carbodiimide hydrochloride (170 mg) and 4-dimethyl-
aminopyridine (50 mg) in anhydrous dichloromethane
(3 ml) according to Dhaon et al.37 The crude isobutyl
ester (218 mg) thus obtained was treated with 25% HBr in
acetic acid (3 ml) to yield the debenzyloxycarbonylated
product, which was recrystallized from MeOH–ether; yield,
224 mg (96% from the N-Z-amino acid); mp 127–129 8C.
Other DL-amino acid ester hydrobromides were prepared in
essentially the same manner and purified by recrystalliza-
tion from MeOH–ether.
4.4. A. oryzae protease-catalyzed hydrolysis of aliphatic
amino acid esters
An amino acid ester hydrochloride (or tosylate) (0.4 mmol)
was dissolved in 2.5 ml of 0.1 M phosphate buffer (pH 7.0)
with/without DMSO (0.6 ml). The pH was adjusted to 7.0
with 0.5 M NaOH. The protease solution (0.5 ml) prepared
as above from a protease preparation (20 mg) was added to
the above substrate solution. The resulting mixture was
stirred at a constant temperature in the range of 5–45 8C.
The pH was maintained at 7.0 by automatic titration with
0.1 M NaOH using an AUT-1 automatic pH titrator
equipped with an ABT-1 automatic burette (TOA
Electronics, Japan). The progress of the reaction was
followed by the consumption of the alkali. After the desired
degree of conversion (ca. 40%), the reaction mixture was
treated as above to separate the liberated amino acid, which
was analyzed by chiral HPLC to obtain its ee value. The
results are shown in Table 2.
4.3. A. oryzae protease-catalyzed hydrolysis of aromatic
amino acid esters
An amino acid ester hydrochloride (or hydrobromide)
(0.2 mmol) was dissolved in 2.5 ml of 0.1 M phosphate
buffer (pH 7.0). The pH was adjusted to 7.0 with 0.5 M
NaOH. On the other hand, a protease preparation (20 mg)
was added to 1 ml of the same buffer, mixed up and
centrifuged (12,000 rpm, 20 min), and 0.5 ml of the
supernatant was added to the above substrate solution.

10260
T. Miyazawa et al. / Tetrahedron 61 (2005) 10254–10261
Griengl, H. Synthesis 1992, 365–366. (d) Berger, B.; de Raadt,
4.5. Chiral HPLC analysis
A.; Griengl, H.; Hayden, W.; Hechtberger, P.; Klempier, N.;
Faber, K. Pure Appl. Chem. 1992, 64, 1085–1088.
10. Hunt, S. The Non-protein Amino Acids. In Chemistry and
Biochemistry of the Amino Acids; Barrett, G. C., Ed.; Chapman
and Hall: London, 1985; pp 55–138.
The optical purities of the resulting amino acids were
determined by chiral HPLC on a Sumichiral OA-5000
column (4.6 mm I.D.!150 mm; Sumika Chemical
Analysis Service, Japan).20 HPLC analysis was performed
on the same liquid chromatograph as stated above under the
following conditions: mobile phase, 0.5–5 mM copper(II)
sulfate in water–2-propanol (85/15–99/1, v/v); flow rate,
1 ml minK1; column temperature, 30 8C; detection, UV at
254 nm.
11. Roberts, D. C.; Vellaccio, F. In Unusual Amino Acids in
Peptide Synthesis; Gross, E., Meienhofer, J., Eds.; The
Peptides; Academic: New York, 1983; Vol. 5, pp 341–449.
12. Coppola, G. M.; Schuster, H. F. In Asymmetric Synthesis:
Construction of Chiral Molecules Using Amino Acids; Wiley:
New York, 1987.
13. Duthaler, R. O. Tetrahedron 1994, 50, 1539–1650.
14. Barrett, G. C. Resolution of amino acids. In Chemistry and
Biochemistry of the Amino Acids; Barrett, G. C., Ed.; Chapman
and Hal: London, 1985; pp 338–353.
Acknowledgements
The authors are grateful to Amano Pharmaceutical Co. for
the generous gift of the protease used in this study. This
work was financially supported in part by a Grant-in-Aid for
Scientific Research from the Ministry of Education,
Science, Sports and Culture of Japan. M. Kawagishi is
acknowledged for some preliminary experiments.
15. Williams, R. M. Synthesis of Optically Active a-Amino Acids;
Pergamon: Oxford, 1989.
16. Chenault, H. K.; Dahmer, J.; Whitesides, G. M. J. Am. Chem.
Soc. 1989, 111, 6354–6364.
17. Miyazawa, T. Amino Acids 1999, 16, 191–213.
18. Miyazawa, T.; Iwanaga, H.; Ueji, S.; Yamada, T. Biocatal.
Biotransform. 2000, 17, 445–458.
19. Contrary to the hydrolysis of the methyl esters of N-protected
amino acids,5a A. oryzae protease generally gave better results
than B. subtilis protease towards the esters of N-unprotected
amino acids examined in the present work. Therefore, the
former protease was employed in the following studies.
20. On a Sumichiral OA-5000 column: Miyazawa, T.; Minowa,
H.; Imagawa, K.; Yamada, T. Chromatographia 2004, 60,
45–50.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.tet.2005.08.
033
21. In the case of 3-fluorophenylalanine (1, RZ3-F-PhCH2) alone,
an acceptable enantioselectivity was obtained (EZ15).
22. Chen, C.-S.; Fujimoto, Y.; Girdukas, G.; Sih, C. J. J. Am.
Chem. Soc. 1982, 104, 7294–7299.
References and notes
23. An example of a marked improvement of the enantio-
selectivity can be seen in the lipase-catalyzed hydrolysis of
8-acetoxy-10-bromoundec-9-enoic acid esters by using the
n-butyl ester instead of the methyl ester as the substrate:
Bhalerao, U. T.; Dasaradhi, L.; Neelakantan, P.; Fadnavis,
N. W. J. Chem. Soc., Chem. Commun. 1991, 1197–1198.
24. Faber, K. Biotransformations in Organic Chemistry, 4 ed.;
Springer: Berlin, 2000; p 42.
1. Gais, H. J.; Theil, F. Enzyme Catlaysis in Organic Synthesis.
In Drauz, K., Waldmann, H., Eds.; Wiley-VCH: Weinheim,
2002; pp 407–412.
´
2. Santos, A. M.; Vidal, M.; Pacheco, Y.; Frontera, J.; Baez, C.;
Omellas, O.; Barletta, G.; Griebenow, K. Biotechnol. Bioeng.
2001, 74, 295–308.
3. Altreuter, D. H.; Dordick, J. S.; Clark, D. S. J. Am. Chem. Soc.
2002, 124, 1871–1876.
25. Miyazawa, T.; Iwanaga, H.; Yamada, T.; Kuwata, S.
Biotechnol. Lett. 1994, 16, 373–378.
4. Kullmann, W. Enzymatic Peptide Synthesis; CRC: Boca
Raton, Florida, 1987; pp 47–48.
26. Preliminary communication: Miyazawa, T.; Imagawa, K.;
Yanagihara, R.; Yamada, T. Biotechnol. Tech. 1997, 11,
931–933.
5. (a) Miyazawa, T.; Iwanaga, H.; Yamada, T.; Kuwata, S.
Chirality 1992, 427, 431. (b) Miyazawa, T. In Fluorine-
Containing Amino Acids: Synthesis and Properties; Kukhar’,
v. p., Soloshonok, V. A., Eds.; Wiley: Chichester, 1995.
6. Miyazawa, T.; Hiramatsu, M.; Murashima, T.; Yamada, T.
Biocatal. Biotransform. 2003, 21, 93–100.
27. For a review on the low temperature effect on lipase-catalyzed
reactions, see: Sakai, T. Tetrahedron: Asymmetry 2004, 15,
2749–2756.
28. Pham, V. T.; Phillips, R. S. J. Am. Chem. Soc. 1990, 112,
3629–3632.
7. Miyazawa, T.; Masaki, S.; Tanaka, K.; Yamada, T. Lett. Pept.
Sci. 2003, 10, 83–87.
29. (a) Phillips, R. S. Enzyme Microb. Technol. 1992, 14,
417–419. (b) Phillips, R. S. TIBTECH 1996, 14, 13–l6.
30. The E value is defined as the ratio of the specificity constants,
8. Miyazawa, T.; Minowa, H.; Miyamoto, T.; Imagawa, K.;
Yanagihara, R.; Yamada, T. Tetrahedron: Asymmetry 1997, 8,
367–370.
k
cat/Km, between the enantiomers:22 EZ(kcat/Km)S/(kcat/Km)R.
Thus, KRT ln E corresponds to the difference in the activation
free energy DD(G‡) between the enantiomers: KRT ln EZ
DDG‡. The temperature dependence of DDG‡ can be given by
the equation DDG‡ZDDH‡K TDDS‡.
9. A. oryzae protease has been employed for the enantioselective
hydrolysis of carboxylic acid esters, especially sterically
demanding a,a-disubstituted carboxylic esters: (a) Fu¨lling,
G.; Sih, C. J. J. Am. Chem. Soc. 1987, 109, 2845–2846. (b)
Feichter, C.; Faber, K.; Griengl, H. J. Chem. Soc., Perkin
Trans. 1 1991, 653–654. (c) Yang, S.; Hayden, W.; Faber, K.;
31. Łankiewicz, L.; Kasprzykowski, F.; Grzonka, Z.; Kettmann,
U.; Hermann, P. Bioorg. Chem. 1989, 17, 275–280.

T. Miyazawa et al. / Tetrahedron 61 (2005) 10254–10261
10261
32. Smith, E. L.; Polglase, W. J. J. Biol. Chem. 1949, 180,
1209–1223.
36. Miyazawa, T.; Nakajo, S.; Nishikawa, M.; Hamahara, K.;
Imagawa, K.; Ensatsu, E.; Yanagihara, R.; Yamada, T.
J. Chem. Soc., Perkin Trans. 1 2001, 82–86.
33. Curtius, T. J. prakt. Chem. 1930, 125, 211–302.
34. Curtius, T.; Mu¨ller, E. Ber. 1904, 37, 1261–1279.
35. Brenner, M.; Huber, W. Helv. Chim. Acta 1953, 36,
1109–1115.
37. Dhaon, M. K.; Olsen, R. K.; Ramasamy, K. J. Org. Chem.
1982, 47, 1962–1965.