Schiff bases of amino acid esters as new substrates for the enantioselective enzymatic hydrolysis and accompanied asymmetric transformations in aqueous organic solvents

Article abstract of DOI:10.1021/jo941734v

The enzyme (lipases and chymotrypsin)-catalyzed hydrolysis of Schiff bases derived from racemic amino acid esters and aromatic aldehydes has been investigated. The reactions were successfully carried out in different aqueous organic solvents at ambient temperature, but the aqueous acetonitrile (5.4% water content by volume) was the solvent of choice. The L-amino acid (ee 98%) precipitated out from the solution as the reaction progressed, and the liberated aldehyde and unhydrolyzed D-ester (ee 40-98%) remained in the solution. The range of substrates included amino acids having different types of side chains. The addition of an organic base (DABCO) into the solution resulted in the racemization of the remaining D-ester and the additional hydrolysis of the substrate, thus leading to the effective asymmetric transformation of the initial ester. Upto 87.5% of the initial racemate was converted into the L-enantiomer.

Full text of DOI:10.1021/jo941734v

J . Org. Chem. 1996, 61, 1223-1227
1223
Sch iff Ba ses of Am in o Acid Ester s a s New Su bstr a tes for th e
En a n tioselective En zym a tic Hyd r olysis a n d Accom p a n ied
Asym m etr ic Tr a n sfor m a tion s in Aqu eou s Or ga n ic Solven ts1,2
Virinder S. Parmar,*,† Amarjit Singh,† Kirpal S. Bisht,† Naresh Kumar,† Y. N. Belokon,‡
K. A. Kochetkov,‡ N. S. Ikonnikov,‡ S. A. Orlova,‡ V. I. Tararov,‡ and T. F. Saveleva‡
Department of Chemistry, University of Delhi, Delhi-110 007, India,
and A.N. Nesmeyanov Institute of Organoelement Compounds, 28 Vavilov, 117813 Moscow, Russia
Received October 17, 1994 (Revised Manuscript Received December 1, 1995X)
The enzyme (lipases and chymotrypsin)-catalyzed hydrolysis of Schiff bases derived from racemic
amino acid esters and aromatic aldehydes has been investigated. The reactions were successfully
carried out in different aqueous organic solvents at ambient temperature, but the aqueous
acetonitrile (5.4% water content by volume) was the solvent of choice. The L-amino acid (ee 98%)
precipitated out from the solution as the reaction progressed, and the liberated aldehyde and
unhydrolyzed D-ester (ee 40-98%) remained in the solution. The range of substrates included
amino acids having different types of side chains. The addition of an organic base (DABCO) into
the solution resulted in the racemization of the remaining D-ester and the additional hydrolysis of
the substrate, thus leading to the effective asymmetric transformation of the initial ester. Upto
87.5% of the initial racemate was converted into the L-enantiomer.
In tr od u ction
as catalysts and Schiff bases derived from aromatic
aldehydes and racemic amino acids esters as substrates.
The envisaged advantages of these substrates are the
lability of the N-protective group, their increased solubil-
ity in organic and aqueous-organic solvents, and high
R-C-H bond acidity of the amino acid moiety. The latter
properties might be employed to bring about the racem-
ization of the unhydrolyzed enantiomer and thus induce
asymmetric transformation9 of the initial racemic Schiff
base during the course of hydrolysis. This paper reports
a successful application of this approach to the resolution
of Phe, substituted Phe, Ala, nor-Val, and R-Me-Phe. It
has been shown that the asymmetric transformation of
Phe could be carried out, and L-Phe was obtained in
87.5% yield from the racemic Schiff base derived from
p-chlorobenzaldehyde and D,L-Phe ester when the chy-
motrypsin-catalyzed hydrolysis was carried out in the
presence of DABCO in aqueous organic solvents.
Enzymes have been widely used for the resolution of
racemic organic compounds.3,4 The use of acylases
derived from different sources for the resolution of
R-amino acids is well documented and broadly employed
both in chemical laboratories and industry.5 Recently it
has been shown from our laboratories and by other
groups of workers that some proteases and lipases are
good catalysts for the enantioselective hydrolysis of amino
acid derivatives.6 Another traditional proteolytic enzyme,
chymotrypsin, was used to resolve hydrophobic N-pro-
tected amino acids esters and free amino acids esters,7
but its application was limited in scope due partly to high
instability of the free amino acids esters; after a week,
more than 50% of the free amino acid ester was converted
to diketopiperazine and other polycondensation prod-
ucts.8
We put ourselves a task of elaborating a new resolution
procedure based on the use of chymotrypsin and lipases
Resu lts a n d Discu ssion
† University of Delhi.
‡ A. N. Nesmeyanov Institute of Organoelement Compounds.
X Abstract published in Advance ACS Abstracts, J anuary 15, 1996.
(1) Dedicated with regards and good wishes to Dr. A. V. Rama Rao
on the occasion of his 60th birthday.
The Schiff bases were prepared according to the
literature procedures from the corresponding amino acid
ester and the aldehyde.10 The Schiff bases of 4-F-D,L-
Phe, and 2-F-D,L-Phe were prepared by alkylation of the
corresponding Schiff base with 4-fluoro- and 2-fluoroben-
zyl bromide, respectively. The Schiff bases of D,L-Ala, D,L-
nor-Val, D,L-Phe, 4-F-D,L-Phe, 2-F-D,L-Phe, and D,L-R-Me-
Phe ethyl esters with benzaldehyde were also prepared
by direct alkylation of the Schiff bases derived from the
corresponding aldehyde and the glycine ester11 (Scheme
1). The Schiff bases of D,L-Ph-Gly-OEt, D,L-nor-Val-OEt,
D,L-Trp-OEt, D,L-4-F-Phe-OEt, and D,L-2-F-Phe-OEt have
been prepared for the first time.
(2) Part of the results have been presented at the 10th IUPAC
International Conference on Organic Synthesis, Bangalore, India,
December 11-16, 1994.
(3) (a) Wong, C. H. Science 1989, 244, 1145. (b) Duthaler, R. O.
Tetrahedron 1994, 50, 1539. (c) Klibanov, A. M. Acc. Chem. Res. 1990,
23, 114.
(4) (a) Lickefett, H.; Krohn, K.; Konig, W. A.; Gehrcke, B.; Syldatk,
C. Tetrahedron:Asymmetry 1993, 4, 1129. (b) Yamazaki, T.; Ohnogi,
T.; Kitazume, T. Tetrahedron:Asymmetry 1990, 1, 215.
(5) (a) Enzymes as Catalysts in Organic Synthesis. Series C.
Mathematical and Physical Sciences; Schneider, M. P., Eds.; NATO
ASI Series; Boston, 1986; Vol. 178, p 263. (b) Chenault, H. K.; Dahmer,
J .; Whitesides, G. M. J . Am. Chem. Soc. 1989, 111, 6354.
(6) (a) Tyagi, O. D.; Boll, P. M.; Parmar, V. S.; Taneja, P.; Singh, S.
K. Indian J . Chem. 1992, 31B, 851. (b) Miyazawa, T.; Iwanaga, H.;
Ueji, Sh.; Yamada, T.; Kuwata, S. Chem. Lett. 1989, 2219. (c)
Miyazawa, T.; Takitani, T.; Ueji, Sh.; Yamada, T.; Kuwata, S. J . Chem.
Soc., Chem. Commun. 1988, 1214.
The reaction protocol was a very simple one. Chymot-
rypsin (10-7-10-6 mol) or a lipase was added as an
insoluble powder to the aqueous organic solvent (1:19,
(7) (a) Ricca, J . M.; Crout, D. H. G. J . Chem. Soc., Perkin Trans. 1
1993, 1225. (b) Huby, N. J . S.; Kinsman, R. G.; Lathbury, D.; Vernon,
P. G.; Gallagher, T. J . Chem. Soc., Perkin Trans. 1 1991, 145. (c)
Chenevert, R.; Letourneau, M.; Thiboutot, S. Can. J . Chem. 1990, 68,
960.
(8) Greenstein, J . P.; Winitz, M. Chemistry of the Amino Acids; J ohn
Wiley and Sons Inc.: New York, 1961; Vol. 2, pp 782 and 797.
(9) Fulling, G.; Sih, Ch. J . J . Am. Chem. Soc. 1987, 109, 2845.
(10) Stork, G.; Leong, A. Y. W.; Touzin, A. M. J . Org. Chem. 1976,
41, 3491.
(11) Ghosez, L.; Antoine, J . P.; Deffense, E.; Navarro, M.; Libert,
V.; O’Donnell, M. J .; Bruder, W. A.; Willey, K.; Wojciechowski, K.
Tetrahedron Lett. 1982, 23, 4255.
0022-3263/96/1961-1223$12.00/0 © 1996 American Chemical Society

1224 J . Org. Chem., Vol. 61, No. 4, 1996
Parmar et al.
Sch em e 1
Sch em e 2
The enantiomeric purities of the product amino acids
were insensitive to the method of preparation of the
Schiff bases. However, it was desirable to distill the
aldimine substrate in vacuo before applying the resolu-
tion procedure. The use of the alkylation reaction
mixture without prior distillation decreased the yields
of both L- and D-amino acids.
Ta ble 1. Hyd r olysis of Sch iff Ba ses of r-Am in o Acid s
Eth yl Ester s (see Sch em e 1) Ca ta lyzed by Ch ym otr yp sin
in Aqu eou s Or ga n ic Solven ts a t Am bien t Tem p er a tu r e
precipitatea filtratea
L-amino
acid
D-amino
acidb
aqueous
time solvent yield, ee, yield, ee,
run
X
R1
R2
(h)
(1:19)
%
%
%
%
Chymotrypsin retained its activity for several days in
the aqueous organic solvents as was shown by addition
of new portions of the fresh substrate after the initial
one was completely hydrolyzed. The formation of the
additional amino acid precipitate was detected even if
the fresh substrate was added after 5 days since the
beginning of the experiment.
1
2
3
4
5
6
7
8
9
Cl
Cl
Cl
H
H
H
H
H
H
H
H
PhCH2
PhCH2
PhCH2
PhCH2
PhCH2
p-FC6H4CH2
o-FC6H4CH2
Me
24 t-BuOH 14
65 t-BuOH 36.5 98
260 t-BuOH 46.5 97
98
79
11
69
81
-
95
95
49
36
40
29
H
H
H
H
H
H
H
H
59
52
-
30 MeNO2
24 MeCN
48 MeCN
24 MeCN
108 MeCN
208 MeCN
200 MeCN
240 MeCN
240 MeCN
200 MeCN
20
38
37
21
30
33
25
29
36
0
98
99
99.5 40
98.5 59
85
92
99.5 50
89
5
-
37
46
40
C3H7
The rate of the reaction depends on the type of the
solvent used and was approximately proportional to the
rate of dissociation of the initial Schiff base into the
components (free ester and aldehyde), as measured by
1H NMR spectroscopy under the experimental conditions
without adding the enzyme. The results may be ratio-
nalized assuming that the real substrates were the free
amino acids esters that were liberated from the Schiff
bases by reaction with water (see Scheme 2) and the
dissociation of the Schiff bases was the rate-limiting
process. An equimolar mixture of D,L-phenylalanine
ethyl ester and p-chlorobenzaldehyde (instead of the
preformed Schiff base of the two), under similar condi-
tions in the presence of added DABCO, yielded only
40.5% of L-phenylalanine. The reaction stopped after this
conversion, and no additional L-phenylalanine could be
obtained. The low yield might be due to the instability
of the free amino acid ester and its tendency to rearrange
into diketopiperazine kind of compounds.8 Further, the
dissociation of Schiff bases in the reaction mixture, which
is the rate-limiting process, also controls the concentra-
tion of the free amino acid esters so that their rearrange-
ment/cyclization is minimized.
Different solvents were successfully used for the enan-
tioselective hydrolysis as illustrated by the results in
Table 2. Aqueous acetonitrile was probably the solvent
of choice providing faster reaction rates, better chemical
yields, and enantiomeric purity of both the amino acid
and the unhydrolyzed ester.
The Schiff bases of aromatic, as well as of aliphatic
amino acid esters were utilized as substrates for the
chymotrypsin-catalyzed hydrolysis, although the enzy-
matic hydrolysis of aliphatic amino acid esters proceeds
1 order of magnitude slower than that of the aromatic
10
11
12 Cl
13
CH3 PhCH2
H
H
H
H
Ph
Ph
40.5 5.5
37
100
2
0
H
PhCH(OH)
a
Chemical yields and enantiomeric analyses were performed
by GLC; however, the isolated yields were only 2-3% less from
b
those determined by GLC. The D-amino acid was obtained by
hydrolysis of the remaining Schiff’s base by 6 N HCl.
8-10 mL) immediately followed by the substrate (10-2
-
10-3 mol) with stirring at ambient temperature. The
addition of any salts or buffer components into the
reaction mixture was found to be unnecessary. The
amount of water in the organic solvent was kept at a 20-
30 fold molar excess over the substrate and it was enough
for the hydrolysis to proceed. At the same time, the
content of water in the organic solvent was not sufficient
to dissolve the precipitated amino acid or chymotrypsin.
As the reaction progressed, the amount of white precipi-
tate in the reaction mixture increased. The chiral GLC
1
and H NMR spectral analysis of the precipitate and the
filterate indicated that it was the free L-amino acid (ee
98%) which was insoluble, and derivatives of D-amino
acid ester stayed in the solution. No side reaction
products were detected by NMR either in solution or in
the precipitate. The enantiomeric purity of the unhy-
drolyzed amino acid ester depended on the extent of the
reaction, as illustrated by the data on the chymotrypsin-
catalyzed hydrolysis of the Schiff base of phenylalanine
ester (Table 1) in aqueous t-BuOH. Table 1 shows the
chemical yields of the resolved amino acids as calculated,
assuming that the theoretically possible yields of the
enantiomers of amino acids were 50% for the L-enanti-
omer and 50% for the D-enantiomer.

Enantioselective Enzymatic Hydrolysis
J . Org. Chem., Vol. 61, No. 4, 1996 1225
Ta ble 2. Hyd r olysis of th e Sch iff’s Ba se Der ived
fr om p-Ch lor oben za ld eh yd e a n d D,L-P h e-OEt w ith
Differ en t En zym es
Ta ble 3. Hyd r olysis of Sch iff’s Ba se of D,L-P h e-OEt w ith
p-Ch lor oben za ld eh yd e (see Sch em e 3) in Aqu eou s
Or ga n ic Solven ts a t Am bien t Tem p er a tu r e in th e
P r esen ce of DABCO
aqueous
solvent
(1:19)
L-Phea
D-Phea
time
(h)
precipitateb filtrateb
enzyme
yield, % ee, % yield, % ee, %
L-amino
acid
D-amino
acid
PPL
PPL
PPL
PPL
PPL
PPL
PPL
MeCN
Me2CO
t-BuOH
MeCN
MeCN
t-BuOH
THF
70
24
70
29.5b
26.5b
41b
14
18
14
92
99
91
99.5
99
97
98.5
96
69b
23b
52b
63
23
60
75.5
45.5
61b
14
49
16
14
49
28
11
34
6
aqueous
time solvent yield, ee, yield, ee,
c
run
enzyme
DABCOa (h) (1:19)
%
%
%
%
66
1
2
3
4
5
6
7
8
9
chymotrypsin
chymotrypsin
chymotrypsin
chymotrypsin
chymotrypsin
PPL
5
10
5
24 MeCN
24 MeCN
96 Me2CO 62
96 t-BuOH 58
96 t-BuOH 63
240 Me2CO 27
240 MeCN
230 t-BuOH 45.5 92
168 MeCN
96 MeCN
30
59.5 87.5 14.5
90 32
99.5 30
94 25
91.5 65
92.5 41.5 14
120
144
120
96
3
0
4
1
19
2
19
5
43.5b
18b
10
10d
10d
10d
10
10
Amano P t-BuOH
CCL
Me2CO
144
73
PPL
17
96
40
-
13
a
The enantiomeric purity and chemical yields were established
Amano P
-
chymotrypsine
78
86
3.5g
by using chiral GLC; however, the isolated yields were only 2-3%
b
10 chymotrypsinf
87.5 90
6.8 0.6g
less from those determined by GLC. Determined by weighing the
sample before purifying it on Sephadex G-25.
a
Mole percent in relation to the substrate. b The chemical yields
and enantiomeric analyses were performed by GLC; however, the
isolated yields were only 2-3% less from those determined by
GLC. c The ee values are lower, probably because the racemic
Schiff base (obtained after the attainment of optimum equilibrium
ones. The only Schiff base that failed to hydrolyze was
the one derived from phenylserine ester; the possible
explanation of the phenomenon might be the formation
of the oxazolidine ring via the interaction of the CdN
bond with the OH group of the substrate side chain.
However, the compound (Schiff base) was stable under
the experimental conditions, and no dissociation produc-
ing the amino acid ester (i.e. phenyl serine ethyl ester)
took place.
Partial racemization of the remaining ester accompa-
nied the resolution of the Schiff base derived from
phenylglycine ester and benzaldehyde catalyzed by chy-
motrypsin on prolonged incubation (run 11, Table 1). A
greater C-H bond acidity of phenylglycine relative to
other R-amino acids might be the underlying reason for
this observation.9 A better electron-withdrawing alde-
hyde component viz. p-chlorobenzaldehyde under similar
conditions brought about, in addition, the racemization
of the L-amino acid itself (run 12, Table 1). The PPL,
CCL, and Amano P-lipases were also used for the
resolution purposes (Table 2), but the effectiveness of the
process was not as good as in the case of chymotrypsin-
catalyzed reactions, probably because the ethyl esters are
not that good substrates for the lipases.6
We believe that the choice of appropriate reaction
conditions and use of additional bases might induce
selective racemization of the esters leaving the precipi-
tated amino acids intact, since the acidities of the amino
acid Schiff bases were, at least, 3 orders of magnitude
smaller than the acidity of the corresponding ester
derivatives and because the precipitation of the amino
acid removes it from the reaction mixture and its
recombination with the aromatic aldehyde gives insig-
nificant amount of the corresponding Schiff base. For
amino acids which are weak C-H acids and for short
intervals of the reaction, the effect is almost negligible.
The case of phenylglycine hydrolysis is an exception
because the amino acid is a relatively strong C-H acid
and significant racemization of the precipitated amino
acid was observed (run 11, Table 1).
Keeping in view the acidities of Schiff bases and the
fact that the enzymatic hydrolysis of the initial substrate
takes place enantioselectively, we reasoned that the
remaining unhydrolyzed substrate might undergo base-
catalyzed racemization in the presence of the enantio-
merically enriched amino acid without any loss of enan-
tiomeric purity of the latter. This concept has been
proved as we have successfully carried out the asym-
metric transformation of the Schiff base derived from
d
state) is chemically hydrolyzed by 6 N HCl. 1,8-Diaminooctane
(10 mol %) in relation to the substrate was used in this case.
e Reaction was run for 96 h without adding DABCO, and then
enzyme was filtered off and filtrate concentrated in vacuo. The
concentrated filtrate was subjected to hydrolysis with a fresh lot
of R-chymotrypsin and DABCO (10 mol %), and the contents were
stirred for 72 h. The yield is reported for the combined amino
acid obtained in two lots. f Reaction was run for 48 h without
adding DABCO, and then the precipitate was filtered off and the
filtrate concentrated in vacuo; DABCO solution in MeCN was
added to the residue and the mixture was kept for 48 h. Then a
fresh portion of aqueous chymotrypsin solution was added, and
g
the mixture was agitated for another 48 h. ee of L-enantiomer.
racemic phenylalanine ethyl ester and p-chlorobenzal-
dehyde in the case of chymotrypsin-catalyzed hydrolysis.
As can be seen from the results summarized in Table 3,
more than 50% (up to 63%) of the initial racemic ester of
phenylalanine was converted into L-phenylalanine when
DABCO (5-10 mol % relative to the substrate) was added
into the solution in an aqueous organic solvent. In fact
the reactions simply stopped after these conversions
(about 60%), and no change in the amount of the
remaining ester was observed. One reason for this kind
of behavior might be that the water from the hydration
shell of the enzyme was consumed to hydrolyze the ester,
the liberated ethanol went into the solution to shift
equilibrium toward the initial ester, and the reaction
stopped at some stage. In a water-methanol mixture,
no reaction was observed which might serve as an
argument to support the equilibrium hypothesis. It was
also anticipated that over the period of time, DABCO
might cause loss in the catalytic activity of the enzyme.
In fact it was found that sequential hydrolysis carried
out by addition of DABCO and a fresh lot of R-chymot-
rypsin to the D-Schiff base, obtained by evaporation of
the filtrate of the reaction mixture instead of its addition
in the beginning of the reaction, increased the total yield
of the precipitated amino acid. The chemical yield and
enantiomeric purity of the material could be significantly
increased if the racemization of the remaining D-PheOEt
was carried out in a separate stage. First, the resolution
procedure was carried out as described above without
adding any base. In a typical experiment, after 96 h the
precipitated L-Phe was filtered, the filtrate was concen-
trated to remove ethanol, and the unreacted aldimine
was again hydrolyzed under the same conditions with
added DABCO and a fresh lot of R-chymotrypsin. The

1226 J . Org. Chem., Vol. 61, No. 4, 1996
Parmar et al.
Sch em e 3
total isolated yield of L-Phe was as high as 78% in 86%
ee (Table 3, run 9).
enzymatic hydrolysis in a mixture of D2O/CD3CN. The
1H NMR spectral analysis of the precipitated L-amino
acid showed that approximately 40% of its R-proton was
substituted by deuterium. As could have been expected,
even larger enrichment was detected (upto 60%) in the
phenylalanine ester remaining in the solution, and this
further supported our argument.
The effect of other bases, i.e. Et3N, (i-Pr)2NH, 1,8-
diaminooctane, and 1,9-diaminononane on the selective
racemization and asymmetric transformation of the
Schiff base, derived from racemic phenylalanine ethyl
ester and p-chlorobenzaldehyde or benzaldehyde in chy-
motrypsin-catalyzed hydrolysis has been studied. It was
found that ee of the precipitated amino acid in all the
cases was less than 92%.
In an optimized experiment, the precipitate was fil-
tered after 48 h to give L-Phe in 46.5% chemical yield
and 98% ee; the solution was evaporated, a new portion
of organic solvent (without water) and DABCO added
to the residue, and the mixture kept at ambient temper-
ature for 48 h to ensure complete racemization of the
remaining Schiff base of D-PheOEt (as monitored by
polarimetry). Then a fresh portion of chymotrypsin,
dissolved in the required amount of water, was added to
the solution, and the mixture was agitated at ambient
temperature for 48 h after which time another 41% of
L-Phe (83% ee) was filtered off. The filterate contained
6.8% of almost racemic Phe (0.6% ee in L-enantiomer).
The data is summarized in Table 3 (run 10).
The increase in the concentration of DABCO enhanced
the rate of the reaction. Aqueous MeCN, Me2CO, and
t-BuOH were used as solvents in the reaction success-
fully. In all the cases studied, the ester remaining in
solution was either racemic or had an admixture of
L-amino acid, which was extracted into the solution by
DABCO. Similar trends were observed in the experi-
ments conducted with PPL, which gives increase in the
yields of the amino acids (Table 3). A 10% of 1,8-
diaminooctane in the reaction mixture was also found to
cause racemization, and Amano P-lipase in t-BuOH was
used to affect the resolution.
In conclusion, the resolution method elaborated in this
work relies on use of inexpensive chemicals and reagents
and experimentally it is very simple. The asymmetric
transformation, which can be carried out successfully,
opens a new perspective in the synthesis of enantiomeri-
cally enriched amino acids.
Exp er im en ta l Section
Gas chromatographic analyses were performed with a chiral
glass capillary column (l ) 41 m, i.d. ) 0.21 mm) of diamide
polysiloxane phase type “Chirasil-Val” (synthesized in the
laboratory) and an FID detector. The carrier gas was helium,
temp 147 °C, flow rate 1 mL/min, and inlet pressure was 1.6
bar. Amino acids were analyzed as N-trifluoroacetyl deriva-
tives12 of their isopropyl esters. Chymotrypsin from bovine
pancreas, lyophilized powder, was obtained from Fluka. Por-
cine pancreatic lipase (PPL, type-II) and Candida cylindracea
lipase (CCL, type-VII) were acquired from Sigma. Amano
P-lipase was obtained from Amano, J apan.
We believe that the transformation might be consid-
ered as being a combination of three processes: (a) The
chemical dissociation of the aldimine bond, giving some
amount of free amino acid ester and aldehyde. (b) The
enantioselective enzyme-catalyzed hydrolysis of the free
L-amino acid ester accompanied by the precipitation of
L-amino acid. (c) The base-induced racemization (via the
intermediate carbanion) of the remaining racemic aldi-
mine of D-amino acid ester, followed by the enzymatic
hydrolysis via a and b (see Scheme 3).
P r ep a r a tion of Sch iff Ba ses. The Schiff bases of amino
acid esters were synthesized according to the literature
procedure in excellent yields starting from the correspond-
ing amino acid ester and the corresponding benzaldehyde.10
To check if the additional amount of L-Phe came from
the partial racemization of the D-ester, we performed the
(12) Nokoporksin, S.; Birrell, P.; Gil-Av, E. J . Chromatogr. Sci. 1970,
8, 177.

Enantioselective Enzymatic Hydrolysis
J . Org. Chem., Vol. 61, No. 4, 1996 1227
Sch iff ba se of D,L-P h e-OEt w ith ben za ld eh yd e: viscous
colorless liquid, bp 182-84 °C (∼1 Torr), yield 47% (after
derived from p-chlorobenzaldehyde and Phe-OEt (0.3 g, 10-3
mol) in MeCN (8.75 mL) was added PPL (0.3 g) or chymot-
rypsin (0.010 g, 4 × 10-7 mol) in 0.5 mL of H2O with stirring,
and the contents were stirred for several hours (see Table 1)
at ambient temperature. The precipitate containing the
L-amino acid and the enzyme was filtered and washed with
1% aqueous ammonia (twice), the solution was evaporated in
vacuo, standard amino acid (L-Ile) solution was added to the
residue, and the mixture was purified first on Sephadex G-25
(medium) with ammonia (1% solution; instead a 5% solution
of chloroacetic acid was also used to separate the enzymes as
precipitates from the solution) and then on a column of ion-
exchange resin DOWEX-50W (H+). The amino acids were
eluted from the resin with 5% aqueous ammonia, the solution
was evaporated, and the residue was analyzed without further
purification as described above. For the preparation of D-
amino acids, the MeCN filterate was evaporated in vacuo and
hydrolyzed by stirring with 6 N HCl at 20 °C for 1 h, the
mixture was extracted with toluene to remove the initial
benzaldehyde. The water layer was refluxed for 5 h and
evaporated. The residue containing D-amino acid was ana-
lyzed after purification on DOWEX-50W (H+). Isolated yields
of both L- and D-amino acids do not differ significantly (2-
3% less) from those determined by GLC in the case of the
R-chymotrypsin-catalyzed reaction (Table 1), and the purity
of the amino acids (precipitate or in the solution) was checked
1
distillation; H NMR (CDCl3, 200 MHz) δ 7.90 (s, 1H), 7.10-
7.70 (m, 10H), 4.10 (q, 2H), 3.85-4.05 (m, 1H), 3.05-3.40 (m,
2H), 1.20 (t, 3H); 13C NMR (CDCl3, 50 MHz) δ 172.1, 162.5,
129.7, 129.2, 128.1, 127.7, 126.6, 126.2, 66.0, 60.5, 45.3, 14.0.
Sch iff ba se of D,L-P h -Gly-OEt w ith ben za ld eh yd e:
colorless solid crystallized from hexane, mp 61-62 °C, bp 140-
42 °C (2 Torr), yield 51% (after distillation); IR (KBr) 1730,
1
1630 cm-1; H NMR (CDCl3, 200 MHz) δ 8.20 (s, 1H), 7.15-
7.75 (m, 10H), 5.10 (s, 1H), 4.10 (q, 2H), 1.05 (t, 3H). Anal.
Calcd for C17H17NO2: C, 76.40; H, 6.37; N, 5.24. Found: C,
77.21; H, 6.42; N, 5.00.
Sch iff ba se of D,L-n or -Va l-OEt w ith ben za ld eh yd e:
colorless viscous oil, yield 81%; IR (KBr) 1740, 1645 cm-1; 1H
NMR (CDCl3, 200 MHz) δ 8.15 (s, 1H), 7.25-7.70 (m, 5H), 4.05
(q, 2H), 3.85 (t, 1H), 1.85 (m, 2H), 1.15 (m, 2H), 1.05 (t, 3H),
0.90 (t, 3H). Anal. Calcd for C14H19NO2: C, 72.10; H, 8.15;
N, 6.01. Found: C, 71.85; H, 8.41; N, 5.88.
Sch iff ba se of D,L-Tr p -OEt w ith ben za ld eh yd e: colorless
solid crystallized from benzene, mp 129-30 °C, yield 57%; IR
1
(KBr) 1735, 1635 cm-1; H NMR (CDCl3, 200 MHz) δ 8.25 (s,
1H, NH), 7.75 (s, 1H, CHdN), 6.70-7.65 (m, 5H, Ind), 4.05
(q, 2H, CH2O), 3.95-4.20 (m, 1H, CH), 3.05-3.50 (m, 2H, CH2-
Ind), 1.10 (t, 3H, CH3). Anal. Calcd for C20H20N2O2: C, 75.00;
H, 6.25; N, 8.75. Found: C, 74.28; H, 6.25; N, 8.77.
1
Gen er a l P r oced u r e of Alk yla tion of Sch iff Ba se of Gly-
OEt w ith Su bstitu ted Ben zyl Br om id e.11 A heterogenous
mixture of Schiff base of Gly-OEt with benzaldehyde10 (1.91
g, 10 mmol), tetrabutylammonium bromide (0.16 g, 0.5 mmol)
in CH2Cl2 (5 mL), and 30% NaOH (2 mL) was stirred at 0 °C,
and a solution of benzyl bromide (11 mmol) in CH2Cl2 (5 mL)
was added dropwise during 1 h. The stirring was continued
for 1 h more and then for 3 h at room temperature before the
reaction mixture was filtered and solvent removed. The
residue was extracted with CCl4 (3 × 20 mL). The CCl4
extracts were combined and washed with cold water (2 × 10
mL) and saturated aqueous NaCl (2 × 10 mL) and dried over
Na2SO4. Solvent was evaporated and the product purified by
distillation in vacuo.
from H NMR spectra and TLC. Large scale experiments could
easily be carried out using the same proportions of the reagents
and solvents.
Asym m etr ic Tr a n sfor m a tion of D,L-P h e-OEt in th e
P r esen ce of a n Ad d ed Ba se d u r in g En zym a tic Hyd r oly-
sis of Its Sch iff Ba se w ith p-Ch lor oben za ld eh yd e. The
reaction was carried out as described above, the only difference
being that DABCO was added into the reaction mixture. The
amino acid recovered from the organic filtrate as described
above was separated from DABCO by ion-exchange chroma-
tography on DOWEX-50 and analyzed (see Table 3).
Ba se-Ca ta lyzed Deu ter iu m Exch a n ge of th e r-P r oton
of P h en yla la n in e Ester d u r in g En zym a tic Hyd r olysis of
th e Sch iff Ba se w ith p-Ch lor oben za ld eh yd e. The reaction
was performed as described above, the only exception being
that it was carried out in the mixture of CD3CN/D2O in the
presence of DABCO (5%). As the reaction in the mixture was
slower than usual, it took 87 h for 42% of L-Phe (81% ee) to
precipitate. The workup of the reaction mixture was made in
the usual manner. No internal standard was added to the
amino acid. The products were analyzed by GLC on chiral
column and by 1H NMR. Approximately 40.4% of R-protons
of L-Phe was substituted by deuterium, and 60.6% of deute-
rium was found in the D-Phe-OEt (10% ee) remaining in the
solution.
Sch iff ba se of D,L-4-F -P h e-OEt w ith ben za ld eh yd e:
colorless viscous oil, bp 183-85 °C (∼1 Torr), yield 48% (after
1
distillation); IR (KBr) 1730, 1630 cm-1; H NMR (CDCl3, 200
MHz) δ 7.85 (s, 1H, CHdN), 7.05-7.45 (m, 5H, Ph), 6.80 (m,
4H, C6H4), 3.85-4.0 (m, 1H, CH), 3.95 (q, 2H, CH2O), 2.95-
3.20 (m, 2H, CH2Ph), 1.05 (t, 3H, CH3). Anal. Calcd for
C18H18NO2F: C, 72.24; H, 6.02; N, 4.68; F, 6.35. Found: C,
72.42; H, 6.21; N, 4.30; F, 6.14.
Sch iff ba se D,L-2-F -P h e-OEt w ith ben za ld eh yd e: color-
less viscous oil, bp 185-90 °C (∼1 Torr), yield 31% (after
1
distillation); IR (KBr) 1730, 1630 cm-1; H NMR (CDCl3, 200
MHz) δ 7.85 (s, 1H, CHdN), 6.85-7.65 (m, 9H, aromatic
protons), 4.05 (q, 2H, CH2O), 3.70-3.85 (m, 1H, CH), 2.90-
3.25 (m, 2H, CH2Ph), 1.05 (t, 3H, CH3). Anal. Calcd for
C18H18NO2F: C, 72.24; H, 6.02; N, 4.68; F, 6.35. Found: C,
72.51; H, 6.30; N, 4.25; F, 6.21.
Schiff bases of D,L-Ala; D,L-nor-Val and D,L-R-Me-Phe esters
with benzaldehyde were also obtained by direct alkylation of
the intermediate Schiff base derived from glycine ester and
benzaldehyde11 with corresponding aryl/alkyl bromide under
phase transfer catalytic conditions.
Ack n ow led gm en t. The work was supported by the
Department of Science and Technology (DST, India) and
the Russian Academy of Sciences (Russia) under the
Indo-Russian ILTP Programme, and the Council of
Scientific and Industrial Research (CSIR, India). We
also wish to thank the referees of the paper for the
suggestions to carry out a few more experiments to
rationalize our findings.
Typ ica l P r oced u r e of th e En zym a tic Hyd r olysis w ith
Lip a se or Ch ym otr yp sin . To the solution of the Schiff base
J O941734V