Enantioselective synthesis of non-natural amino acids using phenylalanine dehydrogenases modified by site-directed mutagenesis

Article abstract of DOI:10.1039/b406364c

The substrate scope of three mutants of phenylalanine dehydrogenase as biocatalysts for the transformation of a series of 2-oxo acids, structurally related to phenylpyruvic acid, to the analogous -amino acids, non-natural analogues of phenylalanine, has been investigated. The mutant enzymes are more tolerant than the wild type enzyme of the non-natural substrates, especially those with substituents at the 4-position on the phenyl ring. Excellent enantiocontrol resulted in all cases.

Full text of DOI:10.1039/b406364c

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Enantioselective synthesis of non-natural amino acids using
phenylalanine dehydrogenases modified by site-directed mutagenesis
Patricia Busca,a Francesca Paradisi,b Eamonn Moynihan,a Anita R. Maguire*a and
Paul C. Engel*b
a
Department of Chemistry, Analytical and Biological Chemistry Research Facility,
University College Cork, Cork, Ireland. E-mail: a.maguire@ucc.ie
Department of Biochemistry, Conway Institute of Biomolecular and Biomedical Research,
b
University College Dublin, Belfield, Dublin 4, Ireland. E-mail: paul.engel@ucd.ie
Received 28th April 2004, Accepted 23rd July 2004
First published as an Advance Article on the web 25th August 2004
The substrate scope of three mutants of phenylalanine dehydrogenase as biocatalysts for the transformation of
a series of 2-oxo acids, structurally related to phenylpyruvic acid, to the analogous a-amino acids, non-natural
analogues of phenylalanine, has been investigated. The mutant enzymes are more tolerant than the wild type enzyme
of the non-natural substrates, especially those with substituents at the 4-position on the phenyl ring. Excellent
enantiocontrol resulted in all cases.
Introduction
Enantiopure drugs constitute an increasing proportion (35% of
drugs on the market in 2000) of pharmaceuticals, and thus there
is substantial chemical and economic interest in methods for
asymmetric synthesis.1 Non-natural amino acids, in enantiopure
form, are of considerable interest in the synthesis of alkaloids,
peptides and other compounds with therapeutic applications e.g.
in HIV-protease inhibitors.2 Incorporation of non-natural amino
acids into biologically active peptides and proteins can greatly
improve their activity, stability, bioavailability and binding
specificity.3 In this context access to both enantiopure series, D-
and L-, is important. Non-natural amino acids are increasingly
in demand by the pharmaceutical industry for peptidomimetic
and other single-enantiomer drugs.4 They are also in demand
as precursors to ligands for asymmetric synthesis.2 While some
enantiopure non-natural amino acids are commercially available
they are very expensive.
One of the most important strategies for asymmetric synthesis
involves biocatalysts—the application of biological species such
as microbial cells or enzymes derived therefrom to catalyse
organic reactions.5 Many biocatalysts exhibit high regio-,
chemo- and stereo-selectivity making them superior to chemical
catalysts for asymmetric synthesis.6 Furthermore, biocatalysts
are ideal energy-efficient, environmentally acceptable reagents,
as virtually all reactions proceed under mild conditions and
avoid the use of toxic reagents and disposal of byproducts. Thus
biocatalysts offer a good opportunity to prepare industrially
useful chiral compounds.7
Scheme 1
substrates indicating that the substrate profile of the enzyme was
varied significantly by the mutations. Enhanced discrimination
between phenylalanine and tyrosine in the engineered enzymes
is of particular significance.13b On this basis we envisaged that
engineered PheDH mutants might prove useful as biocatalysts
for the asymmetric synthesis of non-natural amino acids,
especially phenylalanine analogues. In this paper we report our
preliminary results in this area.
The basis of active site design in this project was initially a
recognition of the common chemistry and shared structural
features of the amino acid dehydrogenase family,14 with
glutamate dehydrogenase as the archetype for which a high-
resolution structure had been solved by X-ray crystallography.15
More recently this strategy has been further strengthened by
the direct solution of the structure of a PheDH.16 Site directed
mutagenesis allows alteration of the amino acid residues
surrounding the substrate-binding pocket of the enzyme to
alter the size, shape and polarity of the pocket. The active site of
phenylalanine dehydrogenase (PheDH) from Bacillus sphaericus
has been mutated on the basis of homology modeling13a and
several mutants affecting the contact with the aromatic ring of
the amino acid substrate have been studied.
While naturally occurring enzymes have been widely
employed in asymmetric synthesis,8 there are instances in which
they cannot be readily employed owing to limited substrate
scope. Designer biocatalysts, where the naturally occurring
protein is modified structurally to provide the desired reaction
site selectivity, offer many advantages.9
We focused in particular on mutations of the Asn in position
145 (N145) which showed an increased discrimination between
phenylalanine and tyrosine,13b making the enzyme more selective
towards non-polar substrates.
In this paper we report the activities of those mutants in which
the asparagine has been substituted by an alanine (N145A), a
leucine (N145L) or a valine (N145V).
Cofactor dependent amino acid dehydrogenases catalyse the
interconversion of 2-oxo acids and a-amino acids, illustrated
in Scheme 1 for phenylalanine dehydrogenase (PheDH).10
Indeed phenylalanine dehydrogenase has been employed for the
enantioselective synthesis of L-2-amino-4-phenylbutanoic acid
with excellent enantiopurity by supplying the homologous oxo
acid as illustrated in Scheme 1.6b,11,12
Results and discussion
Substrate synthesis
Seah et al.13 undertook site-directed mutagenesis on PheDH
and reported in 1995 that the resulting enzymes displayed
reduced activity for L-phenylalanine compared to the wild type
enzyme and enhanced activity towards aliphatic amino acid
With the objective of synthesizing a series of phenylalanine
analogues from the appropriate 2-oxo acids through use of the
engineered PheDH mutants, a series of substrates structurally
related to phenyl pyruvate was selected. The compounds (1–13)
2 6 8 4
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 6 8 4 – 2 6 9 1
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4

View Article Online
Table 1 Synthesis of 2-oxo acids—Grignard route
were chosen for this study, including aromatic, aliphatic and
heteroaromatic 2-oxo acids as precursors for the non-natural
amino acids.
Benzyl halide R
Yield of 2-oxo ester
(%)
Yield of 2-oxo acid
(%)
These compounds were chosen to allow exploration of sub-
strate scope in the modified active site of the enzyme. Both steric
and electronic factors have been explored through this series.
Evidently we required ready access to this series of 2-oxo acids. 2-
Oxo acids, especially analogues of the naturally occurring amino
acids are of major importance in intermediary metabolism.
They have been used in the therapy of certain conditions such as
uraemia and nitrogen accumulation disorders.17 They are also of
interestasintermediatesinchemicalsynthesis, inthedevelopment
of enzyme inhibitors and drugs, as model substrates of enzymes
and in other ways. A general review was published in 1983.17
In this work three general synthetic routes to the 2-oxo acids
were employed. The first route involved preparation through
the addition of Grignard reagents to diethyl oxalate followed
by hydrolysis of the resulting 2-oxo ester as illustrated in
Scheme 2.18 This method was employed for the synthesis of the
chloro and fluoro substituted phenylpyruvate derivatives (2)–(6)
and for the saturated analogue (12), see Table 1.
Using a series of benzylic halide derivatives the 2-oxo esters
(20)–(25) were obtained in excellent yields. The Grignard
reagents were prepared in Et2O instead of THF to avoid the
Wurtz coupling side reaction.19 While this approach gave
the oxo esters in high yields, the recovery of the acids after
hydrolysis was poor, which can be attributed to a number of
side reactions. Decarboxylation can occur during hydrolysis
at high temperature or in basic media during the extraction
3-ClC6H4 (14)
4-ClC6H4 (15)
2-FC6H4 (16)
3-FC6H4 (17)
4-FC6H4 (18)
C6H11 (19)
90 (20)a
87 (21) a
95 (22) a
92 (23) a
87 (24) a
81 (25) b
45 (2) c
33 (3) c
42 (4)c
27 (5) c
39 (6) c
33 (12) c
a Yields following chromatography. b Yield following distillation. c Yield
following precipitation from EtOAc/DCM.
process. Rapid decomposition can occur in polar solvents
during concentration of the organic layer at reduced pressure
giving an unidentifiable mixture by NMR. However, the acids
can be isolated in high purity by precipitation from an ethyl
acetate/dichloromethane mixture. H and 13C NMR spectra in
1
CD3OD show that only the enol form is present for (2)–(6).20
This route also led successfully to the cyclohexyl oxo acid (12),
the saturated analogue of phenylalanine. In this case only the
keto form is observed in d6-DMSO.
This method was less successful for the preparation of
compounds containing electron-donating aryl substituents
such as methyl and methoxy, as the benzyl Grignard reagents
bearing these substituents were seen to undergo extensive Wurtz
coupling. For these derivatives (7)–(11) a second method was
employed via the azlactone (Scheme 3 and Table 2), a route
Scheme 2 1. Mg, Et2O, reflux, 10 min; 2. (EtOCO)2, Et2O, 0 °C, 2 h; 3. AcOH, H2SO4 10%, D, 1 h.
Scheme 3 1. N-Acetylglycine, AcONa, Ac2O, D, 1 h; 2. 3 M HCl, D, 3 h.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 6 8 4 – 2 6 9 1
2 6 8 5

View Article Online
Table 2 Synthesis of 2-oxo acids—azlactone route
Ar
Yield of azlactonea
(%)
Yield of 2-oxo acidb
(%)
4-MeC6H4 (26)
4-MeOC6H4 (27)
4-CF3C6H4 (28)
4-pyridyl (29)
2-thienyl (30)
72 (31)
78 (32)
81 (33)
68 (34)
79 (35)
95 (7)
98 (8)
93 (9)
94 (10)
92 (11)
a Yield following precipitation from reaction mixture. b Yield following
crystallisation from reaction mixture.
reported to provide an efficient synthesis of arylpyruvic acids
containing electron-donating groups.21 In this synthesis the
starting aldehyde is condensed with N-acetylglycine in the
presence of sodium acetate in acetic anhydride. The resulting
azlactone is then hydrolysed with 3 M hydrochloric acid to yield
the 2-oxo acid.
While the literature reports the condensation under reflux
overnight, in this work the reaction was found to reach
completion within 1 hour at reflux, monitoring by TLC. The
reaction was then quenched with ice and filtration of the
resulting precipitate afforded the desired azlactones as yellow
solids in good yields and purity.22
The reaction time for the hydrolysis step was also reduced from
a reported 24–48 hours to typically 3 hours again monitoring by
TLC. Filtration of the resultant precipitate gave the 2-oxo acids
in high yields and purity. Again 1H and 13C NMR showed that
only the enol form of each compound (7)–(11)was present in
CD3OD or d6-DMSO.
Fig. 1
substrate per minute under standard conditions) and also nor-
malized relative to activity with phenylpyruvate for each of the
enzymes. Fig. 1 displays the specific activities only.
To test the enantioselectivity of some of these reductive
aminations, a new experiment was set up for each oxo acid
substituted in position 4 of the aryl ring and also for the pyridine
substrate (10): working with an excess of NADH (1 mM) and
with a suitable amount of enzyme a reaction mixture containing
0.5 mM oxo acid was taken to completion and the crude mixture
was then loaded onto a chiral HPLC column. Each enzyme was
tested with all the substrates. The D-amino acid was never
detected, while the L-amino acid was clearly identified in all
cases. Fig. 2 shows an example of the HPLC outcome, in which
all the peaks are clearly separated and the L enantiomer is the
only one detectable.
The 2-chlorophenylpyruvate derivative (1) was synthesized
as illustrated in Scheme 4. Based on earlier work describing
acylation of organocadmium reagents with acid chlorides,23
reaction of an organocadmium reagent with oxalyl chloride
followed by hydrolysis was envisaged as a short synthetic route
to the 2-oxo acid. However, while this route did produce the
2-oxo acid (1) this synthetic method is less satisfactory than the
routes described above so it was not investigated further.
Discussion
The wild type enzyme is able to catalyse the reductive amination
of oxo acids bearing 4-substituted aromatic groups, although
it is necessary to distinguish between halogens and bulky non-
polar groups: while the activity with 4-fluoro (6) and 4-chloro
(3) substitution is certainly satisfactory (114 and 62.3%), it
drops substantially with 4-methyl (7) or 4-methoxy (8) (14.6 and
14.9%) and it is relatively poor with 4-trifluoromethyl (9) (0.9%).
In comparison with these results, the activities of the mutants
are remarkable: while the mutants show decreased activity for
phenyl pyruvate compared to the WT enzyme, the amino acid
substitution in the binding pocket of these enzymes results in
generally increased tolerance of substitution at the 4-position of
the aromatic ring. In the case of the 4-chloro substitution one of
the most striking results is that for N145L (359% as compared
with 62% for the wild-type enzyme). However, in assessing
this result it is important to note that this mutant shows much
the poorest reference activity with phenyl pyruvate (only
22 U mg−1). Thus 359% is only 79 U mg−1 which is in fact the
lowest figure for the 4-chloro oxo acid (3) with the four enzymes
tested. With 4-methyl (7) or 4-methoxy (8) on the other hand,
the improvement in catalytic activity is absolute, in that activities
are not only higher in most cases than with phenyl pyruvate, but
also uniformly much higher than the activity of the wild type
enzyme with the same substrates (3.9-fold for N145L with 4-
methoxy and 4.6-fold for N145A with 4-methyl). In the case of
the 4-trifluoromethyl (9) substitution the N145A mutant shows
a remarkably high activity of 29.4 U mg−1.
Scheme 4
2-Oxocaproic acid (13) is commercially available and was
included in this study as an acyclic derivative.
Biocatalysis
The non-natural 2-oxo acids were initially screened for activ-
ity in the reductive amination of the oxo acid with the wild
type PheDH and three different mutants (N145A, N145L
and N145V). The results are reported in Table 3 and Fig. 1.
The activity is measured under standard conditions at 25 °C24
by following the decrease in absorbance of NADH at 340 nm
(UV spectrophotometer, Cary 50) according to the reaction
in Scheme 5. For convenience each of the results in Table 3 is
expressed both as a specific activity (a unit, U, of enzyme is
defined as the amount of enzyme which converts one lmole of
Despite the fact that these mutants were designed to
discriminate between Phe and Tyr (different group in the 4-
position), the effect of substitution in the 2- and 3-positions has
Scheme 5
2 6 8 6
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 6 8 4 – 2 6 9 1

View Article Online
Table 3 Activity in reductive amination of 2-oxo acids with WT PheDH and engineered mutants
Oxo acids
WTPheDH
N145A
N145L
N145V
phenylpyruvic acid
100 (200.0 U mg−1
)
)
100 (97.0 U mg−1
)
)
100 (22.0 U mg−1
)
)
)
100 (67.0 U mg−1
5.9 (4 U mg−1
34.4 (23.0 U mg−1
)
2-F-phenyl pyruvate (4)
3-F-phenyl pyruvate (5)
4-F-phenyl pyruvate (6)
2-Cl-phenyl pyruvate (1)
3-Cl-phenyl pyruvate (2)
4-Cl-phenyl pyruvate (3)
4-OMe-phenyl pyruvate (8)
4-Me-phenyl pyruvate (7)
4-CF3-phenyl pyruvate (9)
4-pyridyl pyruvate (10)
2-thienyl pruvate (11)
cyclohexyl KA (12)
6.6 (13.2 U mg−1
5.6 (5.4 U mg−1
7.0 (1.5 U mg−1
22.2 (4.9 U mg−1
)
23.0 (46 U mg−1
)
32.1 (31.1 U mg−1
89.3 (86.6 U mg−1
)
)
)
114 (228.0 U mg−1
7.0 (14.0 U mg−1
)
160 (35.2 U mg−1
7.0 (1.5 U mg−1
1.2 (0.3 U mg−1
)
175 (117.3 U mg−1
)
)
)
7.7 (7.5 U mg−1
3.4 (3.3 U mg−1
)
)
)
)
8.0 (5.4 U mg−1
1.0 (0.7 U mg−1
)
0.8 (1.6 U mg−1
)
62.3 (124.5 U mg−1
)
133 (129.0 U mg−1
54.9 (53.3 U mg−1
)
)
)
)
)
359 (79.0 U mg−1
)
133 (88.8 U mg−1
132 (88.4 U mg−1
102 (68.3 U mg−1
)
14.9 (29.8 U mg−1
14.6 (29.2 U mg−1
)
)
522 (114.8 U mg−1
300 (66.0 U mg−1
)
)
)
135 (130.5 U mg−1
30.3 (29.4 U mg−1
18.2 (17.7 U mg−1
Not detectable
)
0.9 (1.8 U mg−1
)
51.5 (11.3 U mg−1
)
4.5 (3.0 U mg−1
1.0 (0.7 U mg−1
Not detectable
)
)
9.0 (18.0 U mg−1
Not detectable
)
10.2 (2.2 U mg−1
Not detectable
)
6.1 (12.2 U mg−1
12.8 (25.6 U mg−1
)
)
17.9 (17.4 U mg−1
)
)
10.3 (2.3 U mg−1
)
22.6 (15.1 U mg−1
85.0 (57.0 U mg−1
)
)
a-ketocaproic acid (13)
112 (108.4 U mg−1
334 (73.6 U mg−1
)
aliphatic substrate indicating a good degree of freedom in the
shape of the binding pocket, especially in N145A and N145V.
In this work we report the activity towards one linear aliphatic
oxo acid (13), but several others have been tested in a previous
study13b and, although the wild type is rather tolerant, all the
mutants provide increased activity, offering more versatile
catalysis.
In conclusion, in this work we explored the possible
application in catalysis of engineered enzymes which in several
cases have markedly improved activities with novel non-natural
2-oxo acid substrates. In contrast the activity of the engineered
enzymes with the natural substrate phenyl pyruvate decreases
significantly. Clearly, site substitution of the asparagine residue
at 145 for the less polar alanine, leucine or valine residues
results in a binding site which can better accommodate
substituted aromatic derivatives of phenyl pyruvate than the
WT enzyme. As a general rule 5% of the wild-type activity
with phenylpyruvate might be regarded as a borderline for
discriminating between a good or a poor catalyst. In cases where
the activity achieved at present falls well below this figure, such
as 3-chlorophenylpyruvate (2), it seems likely that further site-
directed mutagenesis may produce better biocatalysts. However,
it should be borne in mind that even activities of 1–3 U mg−1
represent perfectly usable biocatalysts, especially in view of
the possibility of ‘over-production’ of the biocatalyst in the
bacterial host.
The potential for use of the engineered enzymes as
biocatalysts for the production of non-natural amino acids is
clear from this work. Significantly, we have demonstrated that
there is no decrease in enantioselectivity associated with the
increased activity.
Experimental
General procedures
Fig. 2 HPLC separation of 4-Cl-phenyl pyruvate (3) + DL-4-Cl-Phe
(A) and HPLC chromatogram of the reductive amination of 4-Cl-
phenyl pyruvate (3) with the mutant N145A (B).
Allsolventsweredistilledpriortouseasfollows.Dichloromethane
and ethyl acetate were distilled from phosphorous pentoxide.
Diethyl ether was freshly distilled from sodium in the presence
of benzophenone. Melting points were measured on a Thomas
Hoover Capillary melting point apparatus and are uncorrected.
Elemental analyses were performed by the Microanalysis
Laboratory, University College Cork, using a Perkin Elmer 240
and an Exeter Analytical CE440 elemental analyser. Infrared
(IR) spectra were recorded on a Perkin-Elmer FT-IR Paragon
1000. Liquid samples were examined as thin films interspersed
between sodium chloride plates. Solid samples were dispersed in
potassium bromide and recorded as pressed discs.
also been investigated for the first time: all the enzymes show
the same trend when the substrates are 2-substituted, namely
a decrease of about 15-fold in specific activity, regardless of
the size of the halogen (F or Cl). On the other hand, 3-fluoro
substitution is much better tolerated whereas 3-chloro is a
considerably less acceptable substitution, suggesting a more
precise spatial fit at this position and reflecting the decreased
steric demand of the fluoro substituent.
Finally, a series of substrates which differ more substantially
from the natural substrate, phenyl pyruvate, has been tested:
heteroaromatic 2-oxo acids (pyridine (10) and thiophene (11) in
place of the phenyl ring), and cyclic (cyclohexyl (12) in place of
the phenyl ring) and linear aliphatic derivatives (C-6) (13).
Pyridine oxo acid (10) is well tolerated by N145A and the
specific activity is very close to the activity shown by the wild
type, while the mutant N145V shows very little activity. Quite
surprisingly, each of the enzymes show activity with a cyclic
1H NMR spectra were recorded at 300 MHz on a Bruker
AVANCE 300 spectrometer. 13C NMR spectra were recorded on
a Bruker AVANCE 300 instrument at 75 MHz. Spectra were run
in deuteriochloroform (CDCl3) with tetramethylsilane (TMS) as
internal standard unless otherwise specified. Chemical shifts (dH
and dC) are expressed as parts per million (ppm), positive shifts
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 6 8 4 – 2 6 9 1
2 6 8 7

View Article Online
being downfield from TMS. Splitting patterns in 1H spectra are
designated as s (singlet), bs (broad singlet), bd (broad doublet),
d (doublet), t (triplet), q (quartet), dd (doublet of doublets),
ABq (AB quartet) and (m) multiplet. Coupling constants are
quoted in Hz. For the 13C NMR spectra, assignments are made
from 13C DEPT spectra run in the DEPT-90 and DEPT-135
modes and with the aid of COSY and HETCOR experiments
in some cases. Mass spectra were measured on a Kratos Profile
spectrometer in electron impact (E.I.) mode with an ionisation
voltage of 70 eV.
All organic solutions were dried using magnesium sulfate
unless otherwise specified. Thin layer chromatography was
performed on precoated silica gel (Merck HF254) plates and
compounds were visualised under ultraviolet light. Column
chromatography was performed using Merck silica gel 60
(typical ratio of silica:crude reaction mixture ~ 30:1). Bulb
to bulb distillations were carried out on a Buchi GKR-50
Kugelrohr.
for 2 h at RT, the reaction was quenched with aq. NH4Cl
(1 M, 100 mL). The aqueous layer was extracted twice
with Et2O (2 × 100 mL) and the combined organic layers
were washed with brine (100 mL), dried with MgSO4 and
finally concentrated under reduced pressure, maintaining
the temperature below 25 °C. The excess diethyl oxalate was
removed with bulb to bulb distillation and the residue was
purified by flash chromatography (hexane:EtOAc = 8:2)
to yield the title compound as a colourless oil (7.75 g, 90%).
1H NMR d (CDCl3, ppm) 1.35 (3H, t, J = 7.1 Hz, CH3), 4.33
(2H, q, J = 7.1 Hz, CH2), 6.43 (1H, s, H-3), 6.82 (1H, bs, OH),
7.12–7.98 (4H, m, Ar–H).
The procedure described above for (20) was employed for the
synthesis of each of the esters (21)–(25) using the appropriate
benzyl or alkyl halide in each case. Purification by flash
chromatography using hexane:EtOAc (8:2) gave the pure
esters.
3-(4-Chlorophenyl)-2-oxopropionic acid ethyl ester (21)25
For the Grignard reactions, Et2O was freshly distilled from
LiAlH4 and the molarity of the reagent was determined
by titration of aliquots with 1 M HCl. For the preparation
of organocadmium reagents CdCl2 was stored in a high
temperature oven to keep it dry.
3-(2-Chlorophenyl)-2-oxopropionic acid (1)24
This was prepared as described for (20) using 4-chlorobenzyl
chloride to give the title compound as a colourless oil (87%). 1H
NMR d (CDCl3, ppm) 1.35 (3H, t, J = 7.1 Hz, CH3), 4.33 (2H,
q, J = 7.1 Hz, CH2), 6.43 (1H, s, H-3), 6.82 (1H, bs, OH), 7.29
(2H, J = 8.8 Hz, HA part of ABq, Ar–H), 7.64 (2H, J = 8.8 Hz,
HB part of ABq, Ar–H).
To a suspension of Mg turnings (0.51 g, 20.9 mmol, 1.1 eq) in
Et2O (3 mL) was added 1 mL of a solution of 2-chlorobenzyl
chloride (2.50 mL, 19.0 mmol, 1 eq) in Et2O (17 mL). The
reaction was initiated by heating and the rest of the halide
solution was added dropwise at a rate that maintains reflux.
The mixture was stirred for 10 min, cooled to RT and CdCl2
(1.83 g, 10.0 mmol, 0.5 eq) added portion wise. The suspension
was stirred at reflux for 40 min, cooled to RT and transferred
to a dropping funnel. This was added dropwise to a solution of
oxalyl chloride (4.30 mL, 38.5 mmol, 2 eq) in Et2O (18 mL) pre-
viously cooled to 0 °C. After stirring for 1 h at 0 °C, the reaction
was quenched with ice (approx. 50 mL). The aqueous layer was
extracted twice with Et2O (2 × 50 mL) and the combined organic
layers were washed with brine (50 mL), dried with MgSO4 and
finally concentrated under reduced pressure. Precipitation of
the residue (EtOAc:DCM ≈ 1:19) afforded the title compound
as a pale yellow solid (1.20 g, 32%). 1H NMR d (CD3OD, ppm)
6.90 (1H, s, H-3), 7.17–7.29 (2H, m, Ar–H), 7.38–7.41 (1H,
dd, J = 1.4 Hz, 7.8 Hz, Ar–H), 8.33–8.37 (1H, dd, J = 1.4 Hz,
7.8 Hz, Ar–H); 13C NMR d (CD3OD, ppm) 106.55 (CH-3),
128.52, 130.18, 131.09, 132.96 (4 × CH), 134.80, 134.98 (2 × C),
144.63 (C-2), 168.78 (C-1); mmax /cm−1 (KBr) 3416 and 3200–2500
(OH), 1682 (CO).
3-(2-Fluorophenyl)-2-oxopropionic acid ethyl ester (22)
This was prepared as described for (20) using 2-fluorobenzyl
chloride to give the title compound as a colourless oil (95%). 1H
NMR d (CDCl3, ppm) 1.35 (3H, t, J = 7.1 Hz, CH3), 4.33 (2H, q,
J = 7.1 Hz, CH2), 6.66 (1H, s, H-3), 6.79 (1H, bs, OH), 6.95–7.37
(3H, m, Ar–H), 8.22–8.24 (1H, m, Ar–H).
3-(3-Fluorophenyl)-2-oxopropionic acid ethyl ester (23)26
This was prepared as described for (20) using 3-fluorobenzyl
chloride to give the title compound as a colourless oil (92%). 1H
NMR d (CDCl3, ppm) 1.36 (3H, t, J = 7.1 Hz, CH3), 4.34 (2H, q,
J = 7.1 Hz, CH2), 6.48 (1H, s, H-3), 6.77 (1H, bs, OH), 6.95–6.99
(1H, m, Ar–H), 7.29–7.33 (1H, m, Ar–H), 7.42–7.44 (1H, m,
Ar–H), 7.61–7.80 (1H, m, Ar–H).
3-(3-Chlorophenyl)-2-oxopropionic acid ethyl ester (20)25
3-(4-Fluorophenyl)-2-oxopropionic acid ethyl ester (24)27
To a suspension of Mg turnings (1.02 g, 42.0 mmol, 1.1 eq) in
Et2O (5 mL) was added 2 mL of a solution of 3-chlorobenzyl
chloride (5.00 mL, 38.0 mmol, 1 eq) in Et2O (35 mL). The
reaction was initiated by heating and the rest of the halide
solution was added dropwise at a rate that maintains reflux.
The mixture was stirred for 10 min, cooled to RT and then
transferred into a dropping funnel. This was added dropwise
to a solution of diethyl oxalate (10.00 mL, 76.0 mmol, 2 eq)
in Et2O (75 mL) previously cooled to 0 °C. After stirring
This was prepared as described for (20) using 4-fluorobenzyl
chloride to give the title compound as a colourless oil (87%). 1H
NMR d (CDCl3, ppm) 1.37 (3H, t, J = 7.1 Hz, CH3), 4.35 (2H, q,
J = 7.1 Hz, CH2), 6.48 (1H, s, H-3), 6.64 (1H, bs, OH), 7.01–7.07
(2H, m, Ar–H), 7.71–7.78 (2H, m, Ar–H).
2 6 8 8
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 6 8 4 – 2 6 9 1

View Article Online
3-Cyclohexyl-2-oxopropionic acid ethyl ester (25)28
168.64 (C-1); mmax /cm−1 (KBr) 3478 and 3200–2500 (OH), 1694
(CO).
3-(3-Fluorophenyl)-2-oxopropionic acid (5)29
This was prepared as described for (20) using bromomethyl
cyclohexane to give the title compound as a colourless oil (81%).
1H NMR d (CDCl3, ppm) 0.64–1.34 (5H, m, ring CH2), 1.39 (3H,
t, J = 7.1 Hz, CH3), 1.66–1.72 (5H, m, ring CH2), 1.90–1.98 (1H,
m, CH), 2.70 (2H, d, J = 7.1 Hz, CH2), 4.27 (2H, q, J = 7.1 Hz,
CH2).
Ester (23) was hydrolysed to give acid (5) as a white solid (27%).
1H NMR d (CD3OD, ppm) 6.46 (1H, s, H-3), 6.94–7.04 (1H,
m, Ar–H), 7.27–7.35 (2H, m, Ar–H), 7.43 (1H, d, J = 7.8 Hz,
Ar–H); 13C NMR d (CD3OD, ppm) 110.36 (CH-3), 115.26 (d,
3-(3-Chlorophenyl)-2-oxopropionic acid (2)25
2
2JC,F = 21.8 Hz, C-4), 117.09 (d, JC,F = 23.0 Hz, C-2), 127.02
(C-6), 131.15 (d, 3JC,F = 8.4 Hz, C-5), 139.43 (d, 3JC,F = 8.4 Hz,
C-1), 144.78 (C-2), 167.56 (C-1), C–F not detected; mmax /cm−1
(KBr) 3477 and 3200–2500 (OH), 1698 (CO).
3-(4-Fluorophenyl)-2-oxopropionic acid (6)29
A solution of (20) (7.00 g, 30.1 mmol) in AcOH (20 mL) and
10% aq. H2SO4 (20 mL) was stirred at reflux for 1 hour, under a
Dean–Stark apparatus. The mixture was cooled to RT, diluted
with water (20 mL) and extracted with Et2O (3 × 50 mL). The
combined organic layers were extracted twice with aq. NaHCO3
(2 × 100 mL). The combined aqueous layers were acidified
to pH 1 and extracted three times with Et2O (3 × 100 mL).
The combined organic layers were washed with brine, dried
with MgSO4 and finally concentrated under reduced pressure,
keeping carefully the temperature below 25 °C. Precipitation of
the residue (EtOAc:DCM ≈ 1:19) afforded the title compound
as a white solid (2.67 g, 45%). 1H NMR d (CD3OD, ppm) 6.43
(1H, s, H-3), 7.19–7.32 (2H, m, Ar–H), 7.57–7.59 (1H, dd,
J = 7.5 Hz, 1.2 Hz, Ar–H), 7.89 (1H, t, J = 1.8 Hz, Ar–H);
13C NMR d (CD3OD, ppm) 110.02 (CH-3), 128.46, 129.37,
130.51, 131.07 (4 × CH), 136.28, 139.12 (2 × C), 144.78 (C-2),
162.03 (C-1); mmax /cm−1 (KBr) 3459 and 3200–2500 (OH), 1676
(CO).
Ester (24) was hydrolysed to give acid (6) as a white solid (39%).
1H NMR d (CD3OD, ppm) 6.47 (1H, s, H-3), 6.98–7.08 (2H, m,
Ar–H), 7.76–7.82 (2H, m, Ar–H); 13C NMR d (CD3OD, ppm)
111.07 (CH-3), 116.78 (d, 2JC,F = 21.8 Hz, 2 × C-3), 133.39 (d,
3JC,F = 8.0 Hz, 2 × C-2), 132.90 (C-1), 142.88 (C-2), 162.52 (d,
1JC,F = 245.5 Hz, C-4), 169.02 (C-1); mmax /cm−1 (KBr) 3476 and
3200–2500 (OH), 1699 (CO).
3-Cyclohexyl-2-oxopropionic acid (12)30
2-Oxo acids (3)–(6), (12) were prepared following the
procedure described for (2). Precipitation using EtOAc:DCM
1:19 was used to purify the acids in each case.
Ester (25) was hydrolysed to give acid (12) as a white solid (33%).
1H NMR d (CDCl3, ppm) 0.88–1.31 (5H, m, ring CH2), 1.66–
1.72 (5H, m, ring CH2), 1.90–1.99 (1H, m, CH), 2.79 (2H, d,
J = 7.1 Hz, CH2), 8.69 (1H, bs, OH); 13C NMR d (CD3OD, ppm)
26.30, 26.36, 33.82 (3 × CH2), 33.96 (CH), 45.49 (CH2), 161.40
(C-1), 195.80 (C-2); mmax /cm−1 (KBr) 3428 and 3200–2500 (OH),
1694 (CO).
3-(4-Chlorophenyl)-2-oxopropionic acid (3)25
2-Methyl-4-[4-(methyl)benzylidene]-5(4H)-oxazolone (31)31
Ester (21) was hydrolysed to give acid (3) as a white solid
(33%). 1H NMR d (CD3OD, ppm) 6.44 (1H, s, H-3), 7.31 (2H,
J = 8.8 Hz, HA part of ABq, Ar–H), 7.75 (2H, J = 8.8 Hz, HB
part of ABq, Ar–H); 13C NMR d (CD3OD, ppm) 110.68 (CH-3),
130.15, 132.85 (2 × CH), 133.17, 135.30 (2 × C), 143.98 (C-2),
168.83 (C-1); mmax /cm−1 (KBr) 3466 and 3200–2500 (OH), 1667
(CO).
A mixture of p-tolualdehyde (5.00 mL, 42.4 mmol, 1 eq), N-
acetyl-glycine (6.45 g, 55.1 mmol, 1.3 eq) and sodium acetate
(4.50 g, 55.1 mmol, 1.3 eq) in acetic anhydride (19.00 mL,
0.2 mol, 5 eq) was stirred at reflux for 1 h. The reaction was
quenched with ice (approx. 50 mL) and vigorously stirred for
1 h in an ice bath to allow precipitation. Filtration afforded
3-(2-Fluorophenyl)-2-oxopropionic acid (4)29
1
the title compound as a yellow solid (6.14 g, 72%). H NMR
d (CD3OD, ppm) 1.90 (3H, s, CH3), 2.23 (3H, s, CH3–Ph), 7.23
(2H, HA part of ABq, J = 8.8 Hz, Ar–H), 7.48 (1H, s, H-3),
7.49 (2H, HB part of ABq, J = 8.8 Hz, Ar–H); 13C NMR d
(CD3OD, ppm) 21.84 (CH3), 22.97 (CH3–Ph), 126.51 (C),
130.77, 131.41 (2 × CH), 132.57 (C), 136.25 (CH-3), 141.65
(C), 168.81 (CN), 173.60 (CO); mmax /cm−1 (KBr) 1794, 1773,
1667.
Ester (22) was hydrolysed to give acid (4) as a white solid (42%).
1H NMR d (CD3OD, ppm) 6.68 (1H, s, H-3), 7.02–7.35 (3H, m,
Ar–H), 8.28–8.32 (1H, m, Ar–H); 13C NMR d (CD3OD, ppm)
102.20 (d, 3JC,F = 8.0 Hz, C-3), 116.49 (d, 2JC,F = 22.4 Hz, C-3),
125.87 (d, 3JC,F = 3.5 Hz, C-6), 130.65 (d, 3JC,F = 8.0 Hz, C-4),
1
132.75 (C-5), 146.59 (C-2), 166.17 (d, JC,F = 283.0 Hz, C-2),
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 6 8 4 – 2 6 9 1
2 6 8 9

View Article Online
The procedure described above for (31) was employed for
the synthesis of each of the azlactones (32)–(35) using the
appropriate benzaldehyde in each case. Filtration afforded the
pure azlactones in each case.
A suspension of (31) (2.60 g, 12.9 mmol) in aq. HCl (3 M, 10 mL)
was stirred at reflux for 3 h. The mixture was cooled to RT to
allow crystallisation. Filtration afforded the title compound
1
as an orange solid (2.19 g, 95%). H NMR d (CD3OD, ppm)
2.34 (3H, s, CH3), 6.47 (1H, s, H-3), 7.14 (2H, HA part of ABq,
J = 8.1 Hz, ArH), 7.65 (2H, HB part of ABq, J = 8.1 Hz, ArH);
13C NMR d (CD3OD, ppm) 21.76 (CH3), 112.16 (CH-3), 130.32,
131.17 (2 × CH), 133.96 (C), 138.86 (C), 141.98 (C-2), 168.90
(C-1); mmax /cm−1 (KBr) 3478 and 3200–2500 (OH), 1669 (CO).
2-Oxo acids (8)–(11) were prepared following the procedure
described for (7). Filtration gave the acids in each case.
2-Methyl-4-[4-(methoxy)benzylidene]-5(4H)-oxazolone (32)31
3-(4-Methoxyphenyl)-2-oxopropionic acid (8)25
This was prepared as described for (31) using 4-
methoxybenzaldehyde to give the title compound as a yellow
1
solid (78%). H NMR d (CDCl3, ppm) 2.39 (3H, s, CH3), 3.87
(3H, s, OCH3), 6.96 (2H, HA part of ABq, J = 7.1 Hz, Ar–H),
7.02 (1H, s, H-3), 7.49 (2H, HB part of ABq, J = 7.1 Hz, Ar–H);
13C NMR d (CDCl3, ppm) 16.03 (CH3), 55.84 (OCH3), 114.84
(2 × CH), 126.53 (C), 130.78 (2 × CH), 131.91 (CH-3), 134.65
(C), 162.45 (C), 165.28 (CN), 168.59 (CO); mmax /cm−1 (KBr)
1798, 1775, 1665.
Azlactone (32) was hydrolysed to give acid (8) as an orange solid
(98%). 1H NMR d (CD3OD, ppm) 3.81 (3H, s, OCH3), 6.48 (1H,
s, H-3), 6.96 (2H, HA part of ABq, J = 7.1 Hz, ArH), 7.49 (2H,
HB part of ABq, J = 7.1 Hz, ArH); 13C NMR d (CD3OD, ppm)
56.05 (CH3), 112.15 (CH-3), 115.10 (2 × CH), 129.47 (C), 132.65
(2 × CH), 140.98 (C-2), 160.91 (C), 169.00 (C-1); mmax /cm−1
(KBr) 3458 and 3200–2500 (OH), 1662 (CO).
2-Methyl-4-[4-(trifluoromethyl)benzylidene]-5(4H)-oxazolone
(33)31
3-(4-Trifluoromethylphenyl)-2-oxopropionic acid (9)34
This was prepared as described for (31) using 4-trifluoromethyl-
benzaldehyde to give the title compound as a yellow solid (81%).
1H NMR d (CDCl3, ppm) 2.43 (3H, s, CH3), 7.13 (1H, s, H-3),
7.81 (2H, HA part of ABq, J = 7.1 Hz, Ar–H), 8.06 (2H, HB part
of ABq, J = 7.1 Hz, Ar–H); mmax /cm−1 (KBr) 1797, 1775, 1663.
Azlactone (33) was hydrolysed to give acid (9) as an orange solid
(93%). 1H NMR d (d6-DMSO, ppm) 6.74 (1H, s, H-3), 7.69 (2H,
HA part of ABq, J = 8.1 Hz, ArH), 7.95 (2H, HB part of ABq,
J = 8.1 Hz, ArH), 9.80 (1H, bs, OH); mmax /cm−1 (KBr) 3461 and
3200–2500 (OH), 1667 (CO).
2-Methyl-4-(4-pyridyl)-5(4H)-oxazolone (34)32
3-(4-Pyridyl)-2-oxopropionic acid (10)32
This was prepared as described for (31) using 4-
pyridinecarboxaldehyde to give the title compound as a grey
1
solid (68%). H NMR d (CDCl3, ppm) 2.45 (3H, s, CH3), 7.03
Azlactone (34) was hydrolysed to give acid (10) as a red solid
(94%). H NMR d (d6-DMSO, ppm) 6.57 (1H, s, H-3), 8.24
(2H, HA part of ABq, J = 6.7 Hz, ArH), 8.75 (2H, HB part of
ABq, J = 6.7 Hz, ArH); 13C NMR d (d6-DMSO, ppm) 103.71,
125.33, 141.03 (3 × CH), 152.08 (C), 152.36(C-2) 164.93 (C-1);
mmax /cm−1 (KBr) 3449 and 3200–2500 (OH), 1660 (CO).
(1H, s, H-3), 7.98 (2H, HA part of ABq, J = 6.2 Hz, ArH), 8.80
(2H, HB part of ABq, J = 6.2 Hz, ArH); mmax /cm−1 (KBr) 1793,
1771, 1662.
1
2-Methyl-4-(2-thienyl)-5(4H)-oxazolone (35)33
3-(2-Thienyl)-2-oxopropionic acid (11)33
This was prepared as described for (31) using 2-
thiophenecarboxaldehyde to give the title compound as a yellow
solid (79%). 1H NMR d (CDCl3, ppm) 2.41 (3H, s, CH3), 7.12–
7.15 (1H, m, Ar–H), 7.38 (1H, s, H-3), 7.56 (1H, d, J = 3.6 Hz,
Ar–H), 7.68 (1H, d, J = 5.1 Hz, Ar–H); mmax /cm−1 (KBr) 1793,
1779, 1665.
Azlactone (35) was hydrolysed to give acid (11) as a grey-green
solid (92%). H NMR d (d6-DMSO, ppm) 6.74 (1H, s, H-3),
1
6.99–7.02 (1H, m, ArH), 7.22 (1H, d, J = 3.5 Hz, ArH). 7.50
(1H, d, J = 5.7 Hz, ArH), 9.49 (1H, bs, OH); 13C NMR d (d6-
DMSO, ppm) 105.93, 127.10, 128.18, 128.27 (4 × CH), 137.80
(C), 140.00 (C-2) 166.03 (C-1); mmax /cm−1 (KBr) 3458 and 3200–
2500 (OH), 1664 (CO).
3-(4-Methylphenyl)-2-oxopropionic acid (7)32
Biochemistry
General details. All the reagents were reagent-grade and used
without further purification. HPLC grade solvents were used
2 6 9 0
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 6 8 4 – 2 6 9 1

View Article Online
Table 4 Retention times (min) of the amino acids on a chirobiotic T (MeOH/H2O 70/30, flow rate of 1.0 mL min−1
)
4-F-phe (6)
4-Cl-phe (3)
4-Me-phe (7)
4-MeO-phe (8)
4-CF3-phe (9)
Pyridine (10)
L-enantiomer
D-enantiomer
6.2
7.9
6.9
9.2
6.9
9.0
7.2
10.0
4.8
n.a.
11.3
n.a
6 (a) H. G. Davies, R. H. Green, D. R. Kelly and S. M. Roberts,
Crit. Rev. Biotechnol., 1990, 10, 129–153; (b) E. Santaniello,
P. Ferraboschi, P. Grisenti and A. Manzocchi, Chem. Rev., 1992, 92,
1071–1140.
7 (a) A. L. Margolin, Enzyme Microbial Technol., 1993, 15, 266–280;
(b) K. Mori, Synlett, 1995, 11, 1097–1109; (c) R. N. Patel, Adv. Appl.
Microbiol., 1997, 43, 91–140; (d) R. N. Patel, Adv. Appl. Microbiol.,
2000, 47, 33–78.
for HPLC chromatographic separations. The NADH grade
II was purchased from Roche. The wild type PheDH and the
mutants were over-expressed in E. coli TG1 cells and purified as
described elsewhere.13a The enzymes were stored as precipitates
in 60% ammonium sulfate at 4 °C and desalted before use
through extensive dialysis against Tris buffer pH 8.0, 50 mM.
8 O. P. Ward and A. Singh, Curr. Opin. Biotechnol., 2000, 11(6),
520–526.
Evaluation of enzyme activity—UV assays
9 H. M. Wilks, K. W. Hart, R. Feeney, C. R. Dunn, H. Muirhead,
W. N. Chia, D. A. Barstow, T. Atkinson, A. R. Clarke and J. J.
Holbrook, Science, 1988, 242, 1541–1544.
10 W. Hummel and M. R. Kula, Eur. J. Biochem, 1989, 184, 1.
11 Y. Asano, A. Yamada, Y. Kato, K. Yamaguchi, Y. Hibino, K. Hirai
and K. Kondo, J. Org. Chem., 1990, 55, 5567–5571.
12 C. W. Bradshaw, C.-H. Wong, W. Hummel and M.-R. Kula, Bioorg.
Chem., 1991, 19, 29.
13 (a) S. Y. Seah, K. L. Britton, P. J. Baker, D. W. Rice, Y. Asano
and P. C. Engel, FEBS Lett., 1995, 370, 93–96; (b) S. Y. Seah,
K. L. Britton, D. W. Rice, Y. Asano and P. C. Engel, Biochemistry,
2002, 41, 11390–11397.
Each oxo acid (4 mM) was dissolved to form a component of a
reaction mixture containing NH4Cl (400 mM), KCl (100 mM),
0.1 mM NADH and Tris (50 mM). The pH was adjusted to 8.0
by adding a suitable amount of HCl. 1 mL of reaction mixture
was incubated at 25 °C. The reaction was followed at 340 nm
over 1 minute after adding an appropriate amount of enzyme
to achieve an optimally measurable reaction rate (between 0.01–
0.03 min−1). Each of the reactions was carried out in duplicate
and the average value is reported.
14 K. L. Britton, P. J. Baker, P. C. Engel, D. W. Rice and T. J. Stillman,
J. Mol. Biol., 1993, 234, 938–945.
Determination of enantioselectivity—chiral HPLC conditions
Each oxo acid (~0.5 mM) was dissolved in a solution contain-
ing NH4Cl (400 mM), KCl (100 mM), 1 mM NADH and Tris
(50 mM). The pH was adjusted to 8.0 by adding a suitable
amount of HCl. The solution was filtered through a sterile filter
Acrodisc® 0.45 lm. 1 mL of reaction mixture was incubated at
25 °C and an appropriate amount of enzyme was added to allow
approximate completion of the reaction within about 40 min.
The formation of the corresponding amino acid was followed by
loading 20 lL of the mixture onto a CHIROBIOTIC T, chiral
HPLC column (Table 4). The elution mixture was MeOH/H2O
70/30 at a flow rate of 1.0 mL min−1.
15 (a) P. J. Baker,
K. L. Britton,
P. C. Engel,
G. W. Farrants,
K. S. Lilley, D. W. Rice and T. J. Stillman, Proteins, 1992, 12,
75–86; (b) T. J. Stillman, P. J. Baker, K. L. Britton, D. W. Rice and
H. F. Rodgers, J. Mol. Biol., 1992, 224, 1181–1184.
16 J. L. Vanhooke, J. B. Thoden, N. M. Brunhuber, J. S. Blanchard and
H. M. Holden, Biochemistry, 1999, 38, 2326–2339.
17 A. J. L. Cooper, L. Z. Ginos and A. Meister, Chem. Rev., 1983, 83,
321–358.
18 (a) L. M. Weinstock, R. B. Currie and A. Lovell, Synth. Commun.,
1981, 11(12), 943–946; (b) W. J. Middleton and E. M. Bingham,
J. Org. Chem., 1980, 45, 2883–2887.
19 K. V. Baker, J. M. Brown, N. Hughes, A. J. Skarnulis and A. Sexton,
J. Org. Chem., 1991, 45, 698–703.
20 T. Takai, H. Senda and H.-H. Lee, Spectrosc. Lett., 1998, 31(2),
379–395.
21 H. N. C. Wong, Z. L. Xu, H. M. Chang and C. M. Lee, Synthesis,
1992, 793–797.
Acknowledgements
Patricia Busca and Francesca Paradisi were supported by
Conway Post-doctoral Fellowships under the (Irish) Higher
Education Authority’s Programme for Research in Third-level
Institutions. Eamonn Moynihan and Francesca Paradisi are
employed under Enterprise Ireland’s Advanced Technology
Research Programme. Support from both these funding sources
is gratefully acknowledged.
22 V. Dala, P. Cotelle and J.-P. Catteau, Tetrahedron Lett., 1997, 38(9),
1577–1580.
23 (a) J. Cason, Chem. Rev., 1947, 40, 15–32; (b) D. A. Shirley, Org.
React., 1954, 8, 28–58.
24 H. Hongwen and F. Xianqi, Gaodeng Xuexiao Huaxue Xuebao,
1988, 9, 966–968.
25 H. Yinglin and H. Hongwen, Gaodeng Xuexiao Huaxue Xuebao,
1991, 12(5), 617–620.
26 K. Hirai, E. Ejiri and S. Ooishi, Jpn. Patent no. JP 06157488, 1994.
27 H. F. Schuster and G. M. Coppola, J. Heterocycl. Chem., 1994,
31(6), 1381–1384.
28 S. Brandange, J. Staffan and S. Vallen, Acta Chem. Scand., 1973,
27(10), 3668–3672.
29 M. Tanaka and K. Otsuka, Jpn. Patent no. JP63048244, 1988.
30 D. H. R. Barton, C.-Y. Chern and J. Cs. Jaszberenyi, Tetrahedron,
1995, 51(7), 1867–1886.
31 H. Hoshina, T. Hitoshi, K. Kubo, T. Igarashi and T. Sakurai,
Heterocycles, 2000, 53(10), 2661–2274.
32 K. Yamabe, A. Sawada and K. Miyafuchi, Sasebo Kogyo Koto
Senmon gakko Kenkyu Hokoku, 1988, 25, 119–122.
33 J. Meiwes, M. Schudok and G. Kretzschmar, Tetrahedron
Asymmetry, 1997, 8(4), 527–536.
34 Y. Ueno, K. Morishita, M. Muraoka and N. Ohashi, Jpn. Patent
no. JP10265464 1998.
References
1 (a) R. M. Williams, J. E. Baldwin and P. D. Magnus, Synthesis of
Optically Active Amino Acids, Vol 7 of Organic Chemistry Series;
Pergamon Press, Oxford, 1989; (b) R. O. Duthaler, Tetrahedron,
1994, 50, 1539–1650.
2 A. S. Bommarius, M. Schwarm, K. Stingl, M. Kottenhahn,
K. Huthmacher and K. Drauz, Tetrahedron Asymmetry, 1995, 6,
2851–2888.
3 A. Boto, R. Hernandez, A. Montoya and E. Suarez, Tetrahedron
Lett., 2002, 43, 8269–8272.
4 P. P. Taylor,D. P. Pantaleone,R. F. SenkpeilandI. G. Fotheringham,
Trends Biotechnol., 1998, 16, 412–418.
5 (a) K. Faber, Biotransformations in Organic Chemistry—A text
book, 4th edn., Springer-Verlag, Berlin, 2000; (b) R. N. Patel,
Stereoselective Biocatalysis, Marcel Dekker, New York, 2000.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 2 6 8 4 – 2 6 9 1
2 6 9 1