The substrate specificity of the heat-stable stereospecific amidase from Klebsiella oxytoca

Article abstract of DOI:10.1016/j.tet.2003.10.124

The substrate specificity of the heat-stable stereospecific amidase from Klebsiella oxytoca was investigated. In addition to the original substrate, 3,3,3-trifluoro-2-hydroxy-2-methylpropanamide, the amidase accepted 2-hydroxy-2-(trifluoromethyl)-butanamide and 3,3,3-trifluoro-2-amino-2- methylpropanamide as substrates. Compounds with larger side chains and compounds where the hydroxyl group was substituted with a methoxy group, or in which the CF₃ group was substituted by CCl₃, were not accepted. The biotransformation is a new synthetic route to (R)-(+)-3,3,3- trifluoro-2-amino-2-methylpropanoic acid, and its related (S)-(-)-amide, and to (R)-(+)-2-hydroxy-2-(trifluoromethyl)-butanoic acid and its related (S)-(-)-amide.

Full text of DOI:10.1016/j.tet.2003.10.124

Tetrahedron 60 (2004) 747–752
The substrate specificity of the heat-stable stereospecific amidase
from Klebsiella oxytoca
†
*
Nicholas M. Shaw and Andrew B. Naughton
Biotechnology and Chemistry Research and Development, Lonza AG, CH-3930 Visp, Switzerland
Received 29 August 2003; revised 8 October 2003; accepted 24 October 2003
Abstract—The substrate specificity of the heat-stable stereospecific amidase from Klebsiella oxytoca was investigated. In addition to the
original substrate, 3,3,3-trifluoro-2-hydroxy-2-methylpropanamide, the amidase accepted 2-hydroxy-2-(trifluoromethyl)-butanamide and
3,3,3-trifluoro-2-amino-2-methylpropanamide as substrates. Compounds with larger side chains and compounds where the hydroxyl group
was substituted with a methoxy group, or in which the CF3 group was substituted by CCl3, were not accepted. The biotransformation is a new
synthetic route to (R)-(þ)-3,3,3-trifluoro-2-amino-2-methylpropanoic acid, and its related (S)-(2)-amide, and to (R)-(þ)-2-hydroxy-2-
(trifluoromethyl)-butanoic acid and its related (S)-(2)-amide.
q 2003 Elsevier Ltd. All rights reserved.
1. Introduction
extract (0.2 mg/mL). The reaction was followed by chiral
GC analysis of the (R)- and (S)-amide substrates. The
enzyme accepted substrates 2 and 7 (Table 1) and the chiral
GC analyses showed that the reactions were enantiospecific.
The initial rate with 2 as substrate was about a quarter of that
with the original substrate 1, and the time to complete the
reaction was correspondingly longer. However, with sub-
strate 7, the initial rate was about 30 times faster than that
with the original substrate 1, and the time to complete the
reaction was much faster (about 300 times).
(R)- and (S)-3,3,3-trifluoro-2-hydroxy-2-methylpropanoic
acid are intermediates for the synthesis of a number of
potential pharmaceuticals, which include ATP sensitive
potassium channel openers for the treatment of
incontinence,1 and inhibitors of pyruvate dehydrogenase
kinase for the treatment of diabetes.2 A novel heat-stable
stereospecific amidase from Klebsiella oxytoca was
selected, purified, characterised and cloned, and used to
synthesise enantiomerically pure (R)-3,3,3-trifluoro-2-
hydroxy-2-methylpropanoic acid, (S)-3,3,3-trifluoro-2-
hydroxy-2-methylpropanoic acid and (S)-3,3,3-trifluoro-2-
hydroxy-2-methylpropanamide.3,4 We now report on the
substrate specificity of the enzyme.
To isolate the reaction products from substrates 2 and 7,
biotransformations with larger amounts of substrate were
carried out (see Section 5). For substrate 2, the biotrans-
formation was carried out with whole cells of E. coli
containing the recombinant K. oxytoca amidase, and for
substrate 7, the biotransformation was carried out with the
amidase immobilised on the carrier Eupergit C, in the
presence of only substrate and water. This experimental
regime was to facilitate the isolation of the product (see
Section 3). The reactions are shown in Scheme 1, and the
detailed reaction results are shown in Table 2. The amide 8
was obtained with a yield of 34.4% and an ee value of
.98%. The corresponding acid 9 was obtained with a yield
of 39.1% and an ee value of 85.2%. The amide 10 was
obtained with a yield of about 50% and an ee value of
.99%, whereas the corresponding acid 11 was obtained
with a yield of 42% and an ee value of 95.4%.
2. Results
The substrates listed in Table 1 were incubated with the
heat-stable amidase from K. oxytoca. The enzyme used was
partially purified by heat treatment of a cell-free extract
from an Escherichia coli strain that contained the
recombinant K. oxytoca gene.3,4 The biotransformations
were carried out at 40 8C in 100 mM potassium phosphate
buffer, pH 8.0, with substrate (5.0 mg/mL) and enzyme
Keywords: Biotransformation; Amidase; (R)-(þ)-3,3,3-Trifluoro-2-amino-
2-methylpropionic
acid;
(S)-(2)-3,3,3-Trifluoro-2-amino-2-
methylpropanamide; (R)-(þ)-2-Hydroxy-2-(trifluoromethyl)-butanoic
acid; (R)-(þ)-2-Hydroxy-2-(trifluoromethyl)-butanamide.
3. Discussion
*
Corresponding author. Tel.: þ41-279-48-59-37; fax: þ41-279-47-59-37;
e-mail address: nicholas.shaw@lonza.com
Present address: Ciba Specialty Chemicals, Schweizerhalle, Switzerland.
†
There is considerable interest in the synthesis of chiral
0040–4020/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2003.10.124

748
N. M. Shaw, A. B. Naughton / Tetrahedron 60 (2004) 747–752
Table 1. Substrate specificity of the amidase
Substrate
Name
Structure
Initial rate
(mmol/min/mg protein)
(relative)
Time to
complete
conversion
1
2
3,3,3-Trifluoro-2-hydroxy-2-methylpropionamide
1.9 (1)
,5 h
2-Hydroxy-2-(trifluoromethyl)-butanamide
0.52 (0.27)
.80 h
3
4
2-Hydroxy-2-(trifluoromethyl)-pentanamide
0
0
3,3,3-Trifluoro-2-hydroxy-2-phenyl-propionamide
5
6
7
3,3,3-Trichloro-2-hydroxy-2-methylpropionamide
3,3,3-Trifluoro-2-methoxy-2-methylpropionamide
3,3,3-Trifluoro-2-amino-2-methylpropanamide
0
0
54.0 (28.41)
1.0 min
trifluoro-substituted acids as pharmaceutical intermediates1,2
and fluorinated amino acids have been synthesised for use as
mechanism based inhibitors of amino acid decarboxylases
and transaminases.5,6 Trifluoro-substituted acids have been
synthesised chemically1,7,8 and we3,4,9 and another group10
have used enzymatic methods for their synthesis. The
fluorinated amino acid 11 (R)-(þ)-3,3,3-trifluoro-2-amino-
2-methylpropanoic acid (3,3,3-trifluoro-2-methyl-alanine)
has been synthesised chemically,5 and also by enzymatic
resolution.6
compound (at 27% of the rate shown with 1 as substrate),
but there was no measurable activity with the propyl and
phenyl derivatives. It follows that the larger the substrate the
more difficulty it has to fit in the active site. Substituting the
CF3 group with CCl3 also resulted in no activity, again
probably due to the larger size of the CCl3 group and after a
time the pH 8 buffer solution slowly began to hydrolyse the
CCl3 group. The influence of the hydroxyl group on the
substrate was investigated by two further derivatives.
Masking the hydroxyl group with a methoxy group resulted
in no activity. Substituting the hydroxy group with an amino
group resulted in a 28-fold increase in the rate of amide
hydrolysis over substrate 1. The amidase active site
probably requires a direct or indirect proton interaction to
function, either in binding or in a more direct chemical role.
Various derivatives of the original enzyme substrate with
substitutions at each of the three positions other than the
amide group were synthesised, and used to test the substrate
specificity of the amidase. Cell-free extracts were used for
these tests, to ensure that the permeability of the E. coli cell
wall to the various substrates would not influence the
results. When larger groups substituted the methyl group on
the substrate, the enzyme was active with the ethyl
The amides 8 and 10 rotated polarised light to the left (2)
and the acids 9 and 11 rotated to the right (þ). The absolute
configurations of 9 and 11 have been reported as (R)-(þ).1,5,10
Scheme 1. Amidase catalysed resolutions of 2 and 7.

Table 2. Amidase catalysed resolutions of 2 and 7
#
Yielda (%)
Optical rotation
1H NMR (DMSO)
13C NMR (DMSO)
169.3 (s, CONH2),
124.9 (q, JCF¼287 Hz, CF3),
77.2 (q, JCCF¼26.2, quaternary),
24.7 (t, CH2), 6.7 (q, CH3)
Elem. anal. or LC–MS
Mp (8C)
ee (%)
8
34.4
[a]D¼213.2 c¼3.75 (MeOH)
7.60 (br s, 1H, CONH2),
7.45 (br s, 1H, CONH2),
6.46 (s, 1H, OH),
2.00 (m, 1H, CH2),
1.70 (m, 1H, CH2),
Calcd (found) for C5H8F3NO2 C 35.1 (35.4), H 4.7 (4.5),
N 8.2 (8.1)
57–61
.98
0.87 (t, 3H, J¼6.7, CH3)
9
39.1
53c
[a]D¼þ9.75 c¼3.31 (MeOH)b
NaD¼25.4 c¼0.6 (water)
NaD¼þ8.4 c¼0.8 (water)
1.92 (m, 1H, CH2),
1.73 (m, 1H, CH2),
169.7 (s, COOH),
Calcd (found) for C5H7F3O3 C 34.9 (35.2), H 4.1 (4.1)
Mass¼156 Mass found¼m/z 157 ([Mþ1]þ)
Mass¼157 Mass found¼m/z 158 ([Mþ1]þ)
107–111
98–101d
85.2
.99e
95.4e
124.6 (q, JCF¼287 Hz, CF3),
77.4 (q, JCCF¼27 Hz, quaternary),
25.6 (t, CH2), 7.0 (q, CH3)
171.3 (s, CONH2),
126.3 (sq, JCF¼285 Hz, CF3),
60.2 (sq, quaternary C, JCCF¼26 Hz),
20.1 (qq, JCCCF¼2 Hz, CH3)
170.4 (s, COOH),
0.88 (t, 3H, J¼7.0 Hz, CH3)
10
11
7.50 (s, br, 1H, CONH2),
7.44 (s, br, 1H, CONH2),
2.46 (s, br, 2H, NH2),
1.35 (s, 3H, CH3)
5.60 (br, COOH, NH2),
1.35 (s, 3H, CH3)
42.0
126.9 (sq, JCCF¼284 Hz, CF3),
60.1 (sq, quaternary C, JCCF¼25 Hz),
21.0 (q, CH3)
a
The theoretical maximum yield is 50%.
Literature10 (60% ee) [a]D¼þ7.38 (c¼5.33, MeOH).
b
c
d
A yield of 53% is not theoretically possible. The higher figure is probably due to experimental error (trace moisture in the amino acid).
This value was obtained from a separate larger biotransformation containing cell extracts. The sample passed elemental analysis, but the GC shows a trace impurity that may be (S)-1. [a]D (MeOH, c¼1.01) of this
sample was 213.2 (23 8C). Anal. Calcd for C4H7F3N2O: C 30.8, H 4.5, N 17.9. Found C 31.1, H 4.7, N 17.7.
Biotransformation with cell-free extract.
e

750
N. M. Shaw, A. B. Naughton / Tetrahedron 60 (2004) 747–752
It follows, therefore, that the acids 9 and 11 that we isolated
have the (R) configuration, and the amides 8 and 10 the (S)
configuration.
(s, 3H, CH3). 13C NMR (DMSO): 170.7 ppm (s, CONH2);
124.9 ppm (q, CF3, JCF¼286.9 Hz); 74.1 ppm (q, quatern-
ary C, JCCF¼27.5 Hz); 19.8 ppm (q, CH3, JCCCF¼1.6 Hz).
19F NMR (DMSO): 277.9 ppm (s, CF3). Mp 143.6–
145.0 8C (3£ recrys. from AcOEt/Hex). IR: 3452 (s, OH);
3327 (s); 3271 (s); 1695 (sh); 1680 (s, CONH2). Anal. Calcd
for C4H6F3NO2: C 30.6, H 3.9, N 8.9. Found C 30.5, H 3.7,
N 8.6.
Isolation of the product 11 from the reaction mixture was
initially a problem. Although 10 could be easily extracted
into ethyl acetate under basic conditions, 11 could not be
isolated due to the presence of many buffer salts, free
proteins and other cellular components. Extraction of amino
acids such as 11 also requires precise knowledge of their
iso-electric point in the mixture, and this point could not be
determined. These problems were solved by using a reaction
that contained only the substrate 7, immobilised enzyme and
water. The enzyme could easily be removed by filtration
from the reaction after it was complete, 10 was extracted
into ethyl acetate and 11 remained essentially alone in the
water to be collected by evaporation of the water.
5.2.2. 2-Hydroxy-2-(trifluoromethyl)-butanamide 2.
2-Hydroxy-2-(trifluoromethyl)-butanenitrile (Fluorochem)
8 g was added slowly dropwise to conc. H2SO4 (15.3 g)
the mixture was heated to 115 8C for 15 min, cooled to 8 8C
and 21.8 g of water added. Diethyl ether (50 mL) was then
added, and the organic phase was washed with water
(25.0 mL), saturated aqueous NaHCO3 (25.0 mL) and again
with water (25.0 mL). The diethyl ether phase was dried
over Na2SO4, filtered and evaporated. The resulting oil was
treated with n-hexane, and the resulting crystals (4.27 g)
1
were collected by filtration. Appearance: white powder. H
4. Conclusion
NMR (CDCl3): 6.21 ppm (br s, 2H, NH2); 4.09 ppm (s,
OH); 2.01–1.91 ppm (m, 2H,CH2); 0.98 ppm (t, 3H, J¼
7.1 Hz, CH3). 13C NMR (CDCl3): 169.7 ppm (s, CONH2);
124.2 ppm (q, CF3, JCF¼284.9 Hz); 77.6 ppm (q, quatern-
The amidase shows a marked preference for hydrolysis of
only the (R) configured amide group, but it will accept only
a limited range of substrates. However, the substrates that it
did accept resulted in products that were unknown when this
work was carried out (8, 9 and 10) or have considerable
synthetic interest (11). The biotransformation with the
amidase therefore serves as a new synthetic route to these
compounds.11 Since this work was carried out, 9 has also
been synthesised by enzymatic ester resolution,10 and we
were able to assign the configuration of the 9 that we
synthesised based on the results in that paper.
ary C, JCCF¼27.9 Hz); 25.8 ppm (t), 6.5 (q) (CH2CH3). 19
F
NMR (CDCl3): 279.1 ppm (s, CF3). Mp 75.8–78.0 8C
(cryst. from toluene). IR: 3472 (s, OH); 3286 (s); 1695 (s,
CONH2); 1593 (s). MS: m/z (rel. intensity) 172 (Mþ1, 1),
143 (10), 142 (17), 127 (4), 109 (6), 108 (100), 93 (20), 79
(6), 69 (7), 67 (6), 57 (30), 45 (18), 44 (57).
5.2.3. 2-Hydroxy-2-(trifluoromethyl)-pentanamide 3.
1,1,1-Trifluoropent-2-one was synthesised13 and converted
to the corresponding cyanohydrin in a similar way to that
previously described.4 The cyanohydrin was then converted
to the amide (7.80 g, yield 42.7%) by reaction with H2SO4
similarly to the substrate above. Appearance: white powder.
1H NMR (CDCl3): 6.29 ppm (br s, 1H, NH2); 6.24 ppm (br
s, 1H, NH2); 4.15 ppm (s, 1H, OH); 1.95–1.82 ppm (m,
2H); 1.61–1.48 ppm (m, 1H); 1.34–1.21 (m, 1H); 0.97 (t,
3H, J¼7.4, CH3). 13C NMR (DMSO): 170.0 ppm (s,
CONH2); 124.1 ppm (q, CF3, JCF¼285.7 Hz); 77.3 ppm
(q, quaternary C, JCCF¼28.6 Hz); 34.7 ppm (t), 15.7 (t),
13.9 (q). 19F NMR (CDCl3): 279.2 ppm (s, CF3). Mp 77.0–
79.0 8C. IR: 3512 (s, OH); 3486 (s); 3400 (s); 3381 (s); 3253
(s); 1706 (s, CONH2); 1687 (s); 1667 (s). MS: m/z (rel.
intensity) 186 (Mþ1, 2), 148 (6), 143 (Mþ1–C3H7 or
CONH2, 26), 122 (64), 103 (7), 94 (56), 71 (24), 59 (11), 44
(100).
5. Experimental
5.1. Analytical methods—general
The methods for NMR (Varian UNITY-400), chiral GC of
the substrates to monitor the reactions, ammonia determi-
nation to monitor the reactions and chiral GC of the acid
1
products have already been described.3 For H NMR the
frequency was 400, 376 MHz for 19F NMR and 100 MHz
for 13C NMR. For the 1H and 13C NMR the values reported
are in ppm downfield from a reference of TMS, for 19F
NMR the values are reported upfield of the reference
compound (CFCl3). FTIR (Nicolet 20 SXB) was determined
on KBr pellets. Only a sample of the wavenumbers of the
major strong peaks is presented as a reference. Optical
rotations were measured with a Perkin–Elmer PE 241
Polarimeter and LC–MS was carried out with a Finnigan
LCQ Deca. Except where otherwise reported, MS refers to
GC–MS.
5.2.4. 3,3,3-Trifluoro-2-hydroxy-2-phenyl-propionamide
4. The title compound was prepared as described pre-
viously14 and converted to the amide (7.89 g) by reaction
with H2SO4 similarly to the substrate above. Appearance:
pale orange powder. 1H NMR (DMSO): 7.74–7.37 ppm (m,
7H, arom CH and NH2); 3.35 ppm (s, 1H, OH). 13C NMR
(DMSO): 169.5 ppm (s, CONH2); 124.1 ppm (q, CF3,
JCF¼286.5 Hz); 77.5 ppm (q, quaternary C, JCCF¼27.0 Hz);
135.1 ppm (s), 128.7 (d), 128.0 (d), 126.3 (d). 19F NMR
(DMSO): 273.4 ppm (s, CF3). Mp 96.4–97.1 8C. IR: 3506
(s); 3390 (s); 3199 (s); 1700 (s); 1568 (s). Anal. Calcd for
C9H8F3NO2 C 49.3, H 3.7, N 6.4. Found C 49.5, H 3.9, N
6.3.
5.2. Synthesis of racemic substrates for amide
biotransformation
5.2.1. 3,3,3-Trifluoro-2-hydroxy-2-methylpropionamide
1. The title compound was synthesised from ethyl 4,4,4,-
trifluoroacetoacetate by standard chemical methods7,12 in
three steps with an overall yield of 59.4%. Appearance:
white powder. 1H NMR (DMSO): 7.55 ppm (br s, 1H, NH2);
7.53 ppm (br s, 1H, NH2); 6.75 ppm (s, 1H, OH); 1.43 ppm

N. M. Shaw, A. B. Naughton / Tetrahedron 60 (2004) 747–752
751
5.2.5. 3,3,3-Trichloro-2-hydroxy-2-methylpropionamide
5. The title compound was synthesised from 1,1,1-
trichloroacetone via the cyanohydrin (69% yield) and by
reaction with H2SO4 (66% yield) similarly to the substrates
above. Appearance: white powder. 1H NMR (DMSO):
7.48 ppm (br d, 2H, CONH2); 1.64 ppm (s, CH3). 13C NMR
(DMSO): 170.8 ppm (s, CONH2); 105.3 ppm (s, CCl3);
82.5 ppm (s, quaternary C); 22.0 ppm (q, CH3). Mp
172.5.0–175.2 8C (from toluene/hexane, lit.15 174–
176 8C). IR: 3490 (s, OH); 3378 (s); 3196 (s); 3381 (s);
1689 (s, CONH2); 1572 (s). MS m/z (rel. intensity): 208, 206
(Mþ1 Cl-isotopes, 0.6 and 0.4) 179 (2), 165 (6), 163 (16),
161 (17), 128 (62), 126 (100), 91 (13), 88 (20), 62 (15), 43
(73). Anal. Calcd for C4H6Cl3NO2 C 23.3, H 2.9, N 6.8.
Found C 23.0, H 3.1, N 7.1.
the GC method, but it was still possible to determine if there
was conversion of substrate by monitoring the size of the
amide peak or by ammonia determination.
Biotransformations with 2 and 7 as substrates to produce
product for isolation were carried out as follows. For
substrate 2 cells of E. coli XL-1 Blue MRF/pPRS7 were
cultivated in Nutrient Yeast Broth with ampicillin
(100 mg/mL) to an OD650 of 4.7. They were then washed
in 100 mM phosphate buffer, pH 8.0, and re-suspended in
the same buffer to an OD650 of 140. The biotransform-
ation conditions were: racemic substrate, 5.0 mg/mL;
cells, OD650¼10; 100 mM potassium phosphate, pH 8.0;
temperature, 37 8C; volume 1.0 L; stirring at 120 rpm. For
substrate 7 the biotransformation conditions were:
racemic substrate, 5.0 mg/mL; amidase from E. coli
XL-1 Blue MRF/pPRS7, immobilised on Eupergit C,
3.0 g (118 mg protein extract immobilised/g Eupergit C);
solvent, water; temperature, 20 8C; volume 200 mL;
stirring at 120 rpm. For both substrates 2 and 7 the
larger scale biotransformations were monitored by chiral
GC analysis of the substrates, and the reactions were
stopped when all of one of the two amide enantiomers
had been converted to product.
5.2.6. 3,3,3-Trifluoro-2-methoxy-2-methylpropionamide
6. The title compound was synthesised by methylation of
3,3,3-trifluoro-2-hydroxy-2-methylpropionamide
with
dimethylsulfate resulting in a 38% yield of the methyl
ether. Appearance: white powder. 1H NMR (DMSO):
7.65 ppm (br s, 2H, CONH2); 3.35 ppm (s, 3H,¼OCH3);
1.50 ppm (s, 3H, CH3). 13C NMR (DMSO): 168.5 ppm (s,
CONH2); 124.6 ppm (q, CF3, JCF¼286.5 Hz); 80.0 ppm (q,
quaternary C, JCCF¼27.0 Hz); 14.8 ppm (q, CH3). Mp
82.3–86.8 8C (from Tol/Hex).
5.3.1. (S)-(2)-2-Hydroxy-2-(trifluoromethyl)-butana-
mide 8 and (R)-(1)-2-hydroxy-2-(trifluoromethyl)-but-
anoic acid 9. Solution from the biotransformation (902 mL)
was adjusted to pH 10.0 with NaOH, and then extracted 3
times with ethyl acetate (600 mL). The pH of the aqueous
phase was re-adjusted to 10.0 between each extraction.
The three portions of ethyl acetate were combined, dried
over Na2SO4, and the ethyl acetate removed under
reduced pressure at 40 8C. The resulting orange oil was
dissolved in hexane and placed at 218 8C overnight. The
product suspension was filtered, washed with cold hexane
and dried. The product was then recrystallised from hot
toluene and dried to give 1.72 g of 8 as an off-white
solid.
5.2.7. 3,3,3-Trifluoro-2-amino-2-methylpropanamide 7.
The title compound was synthesised via low yielding (3%
overall yield), but reliable, literature method16 from
trifluoroacetone. Appearance: white powder. 1H NMR
(DMSO): 7.50 ppm (br s, 1H, CONH2); 7.43 ppm (br s,
1H CONH2); 2.43 ppm (s, 2H, NH2); 1.35 ppm (s, 3H,
CH3). 13C NMR (DMSO): 171.3 ppm (s, CONH2);
126.3 ppm (q, CF3, JCF¼285.7 Hz); 60.2 ppm (q, quatern-
ary C, JCCF¼26.3 Hz); 20.1 ppm (q, CH3, JCCCF¼2.0 Hz).
19F NMR (DMSO): 276.6 ppm (s, CF3). Mp 81.3–84.8 8C
(sublimation followed by EtOAc recrys, lit.16 84–85 8C).
IR: 3452 (s, OH); 3327 (s); 3271 (s); 1695 (sh); 1680 (s,
CONH2). MS: m/z (rel. intensity) 157 (Mþ1, 1), 156 (Mþ,
1), 113 (4), 112 (100, Mþ–CONH2), 94 (3), 93 (5), 92 (18),
69 (2), 62 (5), 42 (31).
The aqueous phase was adjusted to pH 1.0 with HCl and
extracted 2 times with ethyl acetate. The combined ethyl
acetate fractions were dried over Na2SO4, and the ethyl
acetate removed under reduced pressure at 40 8C. Toluene
(15 mL) was added to the residue and the mixture dried to
give 2.18 g of brown solid. The product was then twice
recrystallised from hot toluene and dried to give 1.97 g of 9
as an off-white solid (mp 104–112 8C). For characteris-
ations of 8 and 9 see Table 2. Additional FT-IR data:
compound 8: 3471 (s, OH), 3295 (br s), 1696 (s, CONH2),
1593 (s), 1277 (s), 1166 (s); compound 9: 3422 (s, OH),
3035 (br s), 1750.4 (s, COOH), 1320 (s), 1268 (s), 1229 (s),
1200 (s), 1185 (s), 1161 (s).
5.3. Biotransformations
Biotransformations to test the substrate specificity of the
enzyme were carried out with a cell-free extract from cells
of E. coli XL-1 Blue MRF/pPRS7. Cells were cultivated in
Nutrient Yeast Broth with ampicillin (100 mg/mL) to an
OD650 of 3.6. They were then washed in 100 mM phosphate
buffer, pH 8.0, and resuspended in the same buffer to an
OD650 of 190. The cells were then broken open by 3 passes
through a French press, after which the cell extract was
heated to 75 8C for 5 min, and cell debris and precipitated
protein removed by centrifugation at 20,000 g. The clear
supernatant had a protein concentration of 9.75 mg/mL and
was used in the biotransformations. To test the various
substrates the conditions were: racemic substrate,
5.0 mg/mL; enzyme extract, 0.2 mg/mL; 100 mM potass-
ium phosphate, pH 8.0; temperature, 40 8C. The reactions
were followed by GC analysis and ammonia determination.3
The same GC method was used for all the substrates. For
substrates 4 and 5, the enantiomers were not separated by
5.3.2. (S)-(2)-3,3,3-Trifluoro-2-amino-2-methylpro-
panamide 10 and (R)-(1)-3,3,3-trifluoro-2-amino-2-
methylpropanoic acid (3,3,3-trifluoro-2-methyl-alanine)
11. Removing the immobilised enzyme by filtration stopped
the biotransformation. The aqueous reaction mixture was
extracted 3 times with ethyl acetate. The combined ethyl
acetate fractions were evaporated to give 0.53 g of 10 as a
white powder. The aqueous phase was also evaporated to
give 0.42 g of 11, also as a white powder.

752
N. M. Shaw, A. B. Naughton / Tetrahedron 60 (2004) 747–752
J.-P.; Zimmermann, B.; Neumu¨ller, R. Org. Proc. Res. Dev.
Acknowledgements
2002, 6(4), 497–504.
4. Brieden, W.; Naughton, A.; Robins, K.; Shaw, N.; Tinschert,
A.; Zimmermann, T. WO 9801568, 1998.
5. Bravo, P.; Capelli, S.; Meille, S. V.; Viani, F.; Zanda, M.;
Kukhar, V. P.; Soloshonok, V. A. Tetrahedron: Asymmetry
1994, 5(10), 2009–2018.
With thank Evelyne Schmid, Veronika Venetz and Urs
Karlen for expert technical assistance, and Dr. Michael
Hauck for analytical support.
6. Keller, J.; Hamilton, B. J. Tetrahedron Lett. 1986, 27,
1249–1250.
References and notes
7. Darrall, R. A.; Smith, F.; Stacey, M.; Tatlow, J. C. J. Chem.
Soc. 1951, 2329–2332.
8. Butlin, R. J. WO 9944618, 1999.
1. Ohnmacht, C. J.; Russell, K.; Empfield, J. R.; Frank, C. A.;
Gibson, K. H.; Mayhugh, D. R.; McLaren, F. M.; Shapiro,
H. S.; Brown, F. J.; Trainor, D. A.; Ceccarelli, C.; Lin, M. M.;
Masek, B. B.; Forst, J. M.; Harris, R. J.; Hulsizer, J. M.; Lewis,
J. J.; Silverman, S. M.; Smith, R. W.; Warwick, P. J.; Kau,
S. T.; Chun, A. L.; Grant, T. L.; Howe, B. B.; Li, J. H.; Trivedi,
S.; Halterman, T. J.; Yochim, C.; Dyroff, M. C.; Kirkland, M.;
Neilson, K. L. J. Med. Chem. 1996, 39, 4592–4601.
9. Brieden, W.; Naughton, A.; Robins, K.; Shaw, N.; Tinschert,
A.; Zimmermann, T. DE 19725802 A1, 1998.
ˇ
10. Konigsberger, K.; Prasad, K.; Repic, O. Tetrahedron:
Asymmetry 1999, 10, 679–687.
11. Naughton, A.; Shaw, N. US 6599721 B2, 2003.
12. Henne, A. L.; Pelley, R. L. J. Am. Chem. Soc. 1952, 74,
1426–1428.
2. Aicher, T. D.; Anderson, R. C.; Bebernitz, G. R.; Coppola,
G. M.; Jewell, C. F.; Knorr, D. C.; Liu, C.; Sperbeck, D. M.;
Brand, L. J.; Strohschein, R. J.; Gao, J.; Vinluan, C. C.; Shetty,
S. S.; Dragland, C.; Kaplan, E. L.; DelGrande, D.; Islam, A.;
Liu, X.; Lozito, R. J.; Maniara, W. M.; Walter, R. E.; Mann,
W. R. J. Med. Chem. 1999, 42, 2741–2746.
13. Gassen, K.-R.; Kirmse, W. Chem. Ber. 1986, 119(7),
2233–2248.
14. Kanagasabapathy, V. M.; Sawyer, J. F.; Tidwell, T. T. J. Org.
Chem. 1985, 50(4), 503–509.
15. Bowman, R. E.; Closier, M. D.; Islip, P. J. J. Chem. Soc. 1964,
3841.
16. Dingwall, J. G. EP 298029 A1, 1989.
3. Shaw, N. M.; Naughton, A.; Robins, K.; Tinschert, A.;
Schmid, E.; Hischier, M.-L.; Venetz, V.; Werlen, J.;
Zimmermann, T.; Brieden, W.; de Riedmatten;, P.; Roduit,