Biocatalytic synthesis of some chiral drug intermediates by oxidoreductases

Article abstract of DOI:10.1007/s11746-997-0237-3

Chiral intermediates were prepared by biocatalytic processes with oxidoreductases for the chemical synthesis of some pharmaceutical drug candidates. These include: (i) the microbial reduction of 1-(4-fluorophenyl)-4-[4-(5-fluoro-2-pyrimidinyl)-1 -piperazi

Full text of DOI:10.1007/s11746-997-0237-3

BIOCATALYSIS SYMPOSIUM
Biocatalytic Synthesis of Some Chiral Drug
Intermediates by Oxidoreductases
Ramesh N. Patel*, Ronald L. Hanson, Amit Banerjee, and Laszlo J. Szarka
Department of Microbial Technology, Bristol-Myers Squibb
Pharmaceutical Research Institute, New Brunswick, New Jersey 08903
ABSTRACT: Chiral intermediates were prepared by biocatalytic companies are aware that, where appropriate, new drugs for
processes with oxidoreductases for the chemical synthesis of
some pharmaceutical drug candidates. These include: (i) the mi-
crobial reduction of 1-(4-fluorophenyl)-4-[4-(5-fluoro-2-pyrim-
idinyl)-1-piperazinyl]-1-butanone (1) to R-(+)-1-(4-fluorophenyl)-
4-[4-(5-fluoro-2-pyrimidinyl)-1-piperazinyl]-1-butanol (2) [R-(+)-
BMY 14802], an antipsychotic agent; (ii) the reduction of
N-4-(1-oxo-2-chloroacetyl ethyl) phenyl methane sulfonamide
the clinic should be homochiral to avoid the possibility of side
effects due to an undesirable stereoisomer. Where the switch
from racemate drug substance to enantiomerically pure com-
pound is feasible, there is the opportunity to double the use of
an industrial process. The physical characteristic of enan-
tiomers vs. racemates may confer processing or formulation
advantages.
Chiral drug intermediates can be prepared by different
routes. One is to obtain them from naturally occurring chiral
synthons, mainly produced by fermentation processes. The
(3) to the corresponding chiral alcohol (4), an intermediate for
(+)-N-4-{1-hydroxy-2-[(-methylethyl)amino]ethyl}phenyl
methanesulfonamide [ -(+) sotalol], a β-blocker with class III an-
tiarrhythmic properties; (iii) biotransformation of Nε-carboben-
zoxy (CBZ)- -lysine (7) to Nε-CBZ-
ate needed for synthesis of (S)-1-[6-amino-2-{[hydroxy(4-phenyl
butyl)phosphinyl]oxy}1-oxohexyl]- -proline (ceronapril), a new
angiotensin converting enzyme inhibitor (6) and (iv) enzymatic
synthesis of -β-hydroxyvaline (9) from α-keto-β-hydroxyisovaler-
ate (16). -β−Hydroxyvaline (9) is a key chiral intermediate
D-
D
L
L-oxylysine (5), an intermedi- chiral pool primarily refers to inexpensive, readily available,
optically active natural products. A second is to carry out the
resolution of racemic compounds. This can be achieved by
preferential crystallization of stereoisomers or diastereoiso-
mers and by kinetic resolution of racemic compounds by
chemical or biocatalytic methods. Finally, chiral synthons can
also be prepared by asymmetric synthesis by either chemical
or biocatalytic processes with microbial cells or enzymes de-
rived therefrom. The advantages of microbial or enzyme-cat-
alyzed reactions over chemical reactions are that they are
stereoselective and can be carried out at ambient temperature
L
L
L
needed for the synthesis of S-(Z)-{[1-(2-amino-4-thiazolyl)-2-
{[2,2-dimethyl-4-oxo-1-(sulfooxy)-3-azetidinyl] amino}-2-oxo-
ethylidene]amino}oxyacetic acid (tigemonam) (10), an orally ac-
tive monobactam.
JAOCS 74, 1345–1360 (1997).
KEY WORDS: Antihypertensive drug (ceronapril), antipsychotic and atmospheric pressure. These minimize problems of iso-
agent [R-(+)-BMY 14802], biocatalysis, β-blocker with class III
merization, racemization, epimerization, and rearrangement,
which generally occur during chemical processes. Biocat-
alytic processes are generally carried out in aqueous solution.
This avoids the use of environmentally harmful chemicals
used in chemical processes and the disposal of solvent waste.
Furthermore, microbial cells or enzymes derived therefrom
can be immobilized and reused in many cycles.
Recently, a number of review articles (1–9) have been pub-
lished on the use of enzymes in organic synthesis. This report
provides some specific examples of the use of oxidoreduc-
tases in stereoselective catalysis and preparation of some chi-
ral drug intermediates required for our antipsychotic, antiar-
rhythmic (β-blocker), antihypertensive, and antibacterial
agents (10–13).
antiarrhythmic properties ( -sotalol), chiral drug intermediates,
D
monobactam (tigemonam), oxidoreductases.
Recently, much attention has been focused on the interaction
of small molecules with biological macromolecules. The
search for selective enzyme inhibitors and receptor agonists
or antagonists is one of the keys for target-oriented research
in the pharmaceutical industry. Increasing understanding of
the mechanism of drug interaction on a molecular level has
led to increasing awareness of the importance of chirality as
the key to efficacy of many drug products. It is now known
that often only one stereoisomer of a drug substance is re-
quired for efficacy, and the other stereoisomer is either inac-
tive or exhibits considerably reduced activity. Pharmaceutical
ANTIPSYCHOTIC DRUG (BMY 14802)
*To whom correspondence should be addressed at Department of Microbial
Technology, Bristol-Myers Squibb Pharmaceutical Research Institute, P.O.
Box 191, New Brunswick, NJ 08903.
Stereoselective microbial reduction of 1-(4-fluorophenyl)-4-
[4-(5-fluoro-2-pyrimidinyl)-1-piperazinyl]-1-butanone (1).
During the past few years, much research effort has been di-
E-mail: Ramesh_N._Patel @ccmail.bms.com.
Copyright © 1997 by AOCS Press
1345
JAOCS, Vol. 74, no. 11 (1997)

1346
R.N. PATEL ET AL.
Microorganisms. Microorganisms (Table 1) were obtained
from the American Type Culture Collection (Rockville, MD).
Microorganisms were stored at −90°C in vials.
Growth of microorganisms. For screening purposes, one
vial of each culture was used to inoculate 100 mL of medium
A [glucose (2%), corn steep solids (3.5%), ammonium sulfate
(0.5%), soybean oil (0.5%) and calcium carbonate (0.35%),
adjusted to pH 6.8 in 1 L of tap water]. Cultures were grown
at 28°C and 280 rpm for 48 h. Cultures were harvested by cen-
trifugation at 20,000 × g for 15 min, washed with 100 mM
potassium phosphate buffer pH 6.9, and used for biotransfor-
mation studies.
SCHEME 1
rected toward understanding the sigma receptor system in
Microbial reduction of compound 1. Cells of microorgan-
brain and endocrine tissue. This effort has been motivated by isms were suspended separately in 10 mL of 100 mM potas-
the hope that the sigma site may be a target for a new class of sium phosphate buffer (pH 6.8) at 20% (wt/vol, wet cells) cell
antipsychotic drugs (14–17). Characterization of the sigma concentration and supplemented with 2 mg/mL of compound 1
system helped to clarify the biochemical properties of the dis- and 70 mg/mL of glucose. The reaction was conducted at 30°C
tinct haloperidol-sensitive sigma binding site, the pharmaco- and 200 rpm for 48 h. Cell suspensions after transformations
logical effects of sigma drugs in several assay systems, and were extracted with 2 vol of methylene chloride/aceto-
the transmitter properties of a putative endogenous ligand for nitrile/isopropanol mixture (60:35:5) After centrifugation, the
the sigma site (18–21).
organic layer was collected and evaporated under a gentle
Compound 2, or R-(+)-1-(4-fluorophenyl)-4-[4-(5-fluoro- stream of nitrogen. The oily residue obtained was dissolved in
2-pyrimidinyl)-1-piperazinyl]-1-butanol [R-(+)-BMY 14802], methanol, filtered through a 0.2-µm LID/X filter (Whatman,
is a sigma ligand, has a high affinity for sigma binding sites, Inc., Fairfield, NJ), and analyzed by gas chromatography (GC).
and can selectively inhibit conditioned avoidance and apomor-
Growth of Mortierella ramanniana in a fermentor. Mor-
phine-induced stereotype in rats predictive of antipsychotic tierella ramanniana ATCC 38191 culture was grown in a
efficacy. Dopamine cell firing in the substantia nigra, caused 380-L fermentor that contained 250 L of medium A. Growth
by the putative sigma receptors, was inhibited by (+)-3H-3-(3- consisted of several inoculum development stages and fer-
hydroxyphenyl)-N-(1-propyl) piperidine (17,18). In this sec- mentation.
tion, we describe the stereoselective microbial reduction of 1
Inoculum development consisted of F1 and F2 stages. In
to yield 2 [R-(+)-BMY 14802] (Scheme 1). The reductase that the F1 stage, frozen vials of M. ramanniana ATCC 38191 cul-
catalyzed the stereoselective reduction of compound 1 to R- tures were inoculated into 100 mL of medium A. Growth was
(+) BMY 14802 (2) has been purified to homogeneity.
carried out in 500-mL flasks at 28°C and 280 rpm for 48 h. In
the F2 stage, 100 mL of F1-stage culture was inoculated into
1.5 L of medium A in a 4-L flask and incubated at 28°C and
180 rpm for 24 h. A germinator with 250 L of medium A was
Materials and Methods
Materials. Standards of compounds 1 and 2 were synthesized inoculated with 1.5 L of F2-stage inoculum and grown for 24 h
by the Chemical Process Research Department, Bristol- at 28°C and 150 rpm with 150 Lpm (liters per min) aeration.
Myers Squibb Pharmaceutical Research Institute (New
Brunswick, NJ) (22,23).
Fermentors with 250 L of medium A were inoculated with
30 L of germinator-grown inoculum. Fermentations were con-
TABLE 1
Microbial Reduction of 1 to 2 [BMY 14802]a
Optical purity of BMY 14802
Substrate 1
BMY 14802
Microorganisms
(g/L)
(g/L)
R-(+)
S-(−)
Mortierella ramanniana
ATCC 38191
0.02
1.95
98.9
1.1
Pullularia pullulans
ATCC 16623
0.04
1.62
1.5
98.5
aMicroorganisms were grown in a 25-L fermentor for 48 h. Cells were harvested by Cepa centrifuge
(New Brunswick Scientific, Edison, NJ) and suspended in 0.1 M phosphate buffer, pH 6.0. Cell sus-
pensions (10% wt/vol, wet cells) were used in the reduction of 1 (2 g/L) at 280 rpm and 28°C for 24
h. The concentrations of 1 and 2 were determined by gas chromatography (GC) and the optical pu-
rity of 2 was determined by chiral high-performance liquid chromatography (HPLC). Abbreviations:
1, 1-(4-fluorophenyl)-4-[4-(5-fluoro-2-pyrimidinyl)-1-piperazinyl]-1-butanone; 2, R-(+)-1-(4-fluo-
rophenyl)-4-[4-(5-fluoro-2-pyrimidinyl)-1-piperazinyl]-1-butanol [R-(+)-BMY 14802]; ATCC, Ameri-
can Type Culture Collection (Rockville, MD).
JAOCS, Vol. 74, no. 11 (1997)

CHIRAL INTERMEDIATES BY OXIDOREDUCTASES
1347
ducted for 40 h at 28°C and 150 rpm with 150 Lpm aeration. tion cycle, when residual glucose was depleted, the biotrans-
To determine the specific activity of cells during fermentation, formation process was initiated by the addition of substrate
cells were periodically harvested by centrifugation from 200 (2 g/L) and glucose (84 g/L) directly to the fermentor. Subse-
mL of culture broth. Cell suspensions (10% wt/vol, wet cells) quently, substrate and product concentrations were monitored.
were prepared in 100 mM potassium phosphate buffer (pH The biotransformation process was considered complete when
5.8) and supplemented with 0.5 mg/mL of compound 1 and 70 the residual substrate concentration was less than 7 mg/L.
mg/mL of glucose. The reaction was conducted at 28°C and Generally, the biotransformation process was completed
280 rpm on a shaker. Periodically, samples were analyzed for in 25 h.
the reduction of compound 1 to compound 2 by GC. The spe-
Isolation of compound 2. Fermentation broth (10 L), con-
cific activity was expressed as micrograms of compound 2 taining 20 g of compound 2, was centrifuged to remove cells.
produced per hour per gram of dry cells. After 45 h of fermen- The clear supernatant was treated with 5 N NaOH to raise the
tation, cells were harvested with the aid of a Sharples cen- pH to 8.0 and allowed to stand at room temperature for 1 h.
trifuge (Alfa-Laval, Springfield, PA), and the wet cell pastes The precipitated solids (70 g) were collected by filtration and
were collected and stored at −60°C until further use. About 12 dissolved in 500 mL acetone. The acetone solution was filtered
kg of wet cell pastes were collected from each fermentation.
and dried over anhydrous sodium sulfate. Acetone was re-
Two-stage biotransformation process. Frozen cells from moved by rotary evaporation, followed by drying at 30°C at
the foregoing batches were used to conduct the reduction of reduced pressure to recover 14.2 g of white solid compound 2
compound 1 in a 5-L BioFlo fermentor (NewBrunswick Sci- in overall 70 mol% yield. The GC area percentage purity of
entific, Edison, NJ). Cell suspensions (20% wt/vol, wet cells) the isolated compound 2 was >99%.
in 3 L of 100 mM potassium phosphate buffer (pH 5.8) were
Preparation and purification of cell extracts. Mortierella
used. Compound 1 (6 g) and glucose (200 g) were supple- ramanniana ATCC 38191 cells were suspended in buffer A
mented to the cell suspensions, and the reduction was con- [0.1 M 2-(N-morpholino)ethanesulfonic acid (MES) buffer,
ducted at 28°C and 600 rpm with 7 Lpm aeration. Periodically, pH 5.5], containing 20% glycerol and 1 mM dithiothreitol
a 2-mL sample was removed and extracted with 2 vol of meth- (DTT), at a 20% (wt/vol, wet cells) cell concentration. The cell
ylene chloride/acetonitrile/isopropanol mixture (60:35:5). suspensions were disintegrated by sonication for 20 min at
After centrifugation, the organic layer was collected and dried 4°C. Disintegrated cells were centrifuged at 20,000 × g for
with a nitrogen stream. The oily residue was dissolved in 40 min at 4°C, and the supernatant solution was referred to as
methanol, filtered through 0.2-µm LID/X filter, and analyzed cell extract.
by GC to determine the percentage conversion of compound 1
Cell extract was loaded onto a DEAE cellulose (DE-52
to compound 2. The optical purity of compound 2 was deter- from Whatman Co., Maidstone, England) column (5.0 × 67
mined by chiral high-pressure liquid chromatography (HPLC). cm). The column was washed with 500 mL of buffer A and
Single-stage fermentation and biotransformation. The fer- eluted with a linear gradient of salt (0–0.5 M sodium chloride)
mentation process consisted of a laboratory sporulation stage in 1 L of buffer A. Fractions that contained reductase activi-
and a fermentation stage in which cells of M. ramanniana ties were pooled and loaded onto a phenyl sepharose column
ATCC 38191 were cultivated. Immediately after fermentation, (25 × 15 cm), which was previously equilibrated with buffer
substrate was added to the fermentor, and biotransformation A. The column was washed with buffer A and then eluted with
was continued.
buffer A containing 1% Triton × 100. Fractions containing re-
In the sporulation stage, frozen vials of M. ramanniana ductase activities were pooled and loaded onto a Pharmacia
ATCC 38191 culture were inoculated into 100 mL of medium (Piscataway, NJ) Mono Q column (0.5 × 5 cm) in a Phast
A contained in a 500-mL flask. The flask was then incubated (Pharmacia) protein liquid chromatography system. The col-
at 25°C and 280 rpm for 48 h. One mL of cell suspension from umn was washed with buffer A and subsequently eluted with
the foregoing flasks was used to inoculate rice flasks to initi- a salt gradient (0–0.1 M sodium chloride in buffer A).
ate sporulation of the culture at 25°C for 144 h under static
Transformation of compound 1 by cell extracts. Cell ex-
conditions. At the end of sporulation time, 100 mL of medium tracts, pooled DE-52 fraction, pooled Mono Q fraction, and
A was added to the flask under aseptic conditions, and the pooled phenyl sepharose fractions were analyzed for their abil-
flask was incubated at 25°C and 280 rpm for 18 h. Spore sus- ity to catalyze the transformation of compound 1 to compound
pension, obtained by the foregoing procedure, was added 2. The reaction mixture contained 0.67 mM NADP, 10 mg/mL
aseptically to 400 mL sterile water and used to inoculate a glucose, 3.4 units of glucose dehydrogenase, 1 mg/mL of com-
20-L fermentor that contained 15 L of medium A. The fermen- pound 1, and an appropriate amount of enzyme fraction in 1
tor was equipped with pH and dissolved-oxygen probes. The mL of buffer A. The reaction was started by the addition of re-
exhaust gas from the fermentor was routed to a mass spec- ductase and continued at room temperature on a shaker at 150
trometer for continuous monitoring of exhaust carbon dioxide rpm. The reaction was quenched by raising the pH of the reac-
and oxygen. Fermentations were conducted at 25°C and 600 tion mixture to 9.5 with 1 N NaOH. The reaction mixture was
rpm with 20 Lpm aeration. During fermentation, samples were extracted with 2 mL toluene, and the separated toluene layer
periodically taken to check sterility and determine the dry cell was analyzed for compounds 1 and 2 by GC. One unit of ac-
weight and glucose concentration. At the end of the fermenta- tivity was defined as the amount of compound 2 formed
JAOCS, Vol. 74, no. 11 (1997)

1348
R.N. PATEL ET AL.
(nmol/min). Protein concentration in various fractions was de-
termined by BioRad (Richmond, CA) assays (24).
TABLE 2
Reduction of 1 to 2 [BMY 14802] by Mortierella ramanniana: Evalua-
tion of Cells During Growth in a 380-L Fermentora
Polyacrylamide gel electrophoresis. Denaturing gels were
run on a Phast system. The purified enzyme was incubated at
90°C for 2 h with 1% sodium dodecyl sulfate (SDS) and 10
mM DTT, then electrophoresed on a 12.5% gel with 0.1% SDS.
Molecular weight (Mr) determination. Mr of the native en-
zyme was determined by gel filtration on a Superose 12 col-
umn (Pharmacia). The column was eluted with 100 mM MES
buffer (pH 6.0) that contained 20% glycerol and 1 mM DTT at
a flow rate of 1 mL/min. The following standards (Sigma, St.
Louis, MO) were used for calibration purpose: sweet potato
amylase (200,000) yeast alcohol dehydrogenase (150,000),
bovine serum albumin (66,000), bovine erythrocyte carbonic
anhydrase (29,000), and horse heart cytochrome (12,384). The
subunit Mr was determined by SDS-polyacrylamide gel elec-
Cell
growth
time (h)
Concentration
of substrate 1
(g/L)
Concentration of 2,
BMY 14802
(g/L)
Optical purity
R-(+) BMY 14802
(%)
7
19
31
43
0.96
0.56
0.04
0.12
0.42
1.1
1.86
1.8
89.5
93.3
99.4
96.8
aDuring fermentation cells were harvested from 500-mL broth samples and
suspended in 100 mM potassium phosphate buffer (pH 5.8). Cell suspen-
sions (20% wt/vol, wet cells) were used to conduct the reduction of 1 (2
mg/mL) at 280 rpm and 20°C for 24 h. Concentrations of 1 and 2 were de-
termined by GC and the optical purity of 2 was determined by chiral HPLC.
For abbreviations, see Table 1.
trophoresis with low-molecular-weight standards from BioRad. 2. Optical purity of >98% was obtained in each reaction
Analytical methods. Analysis of compounds 1 and 2 was (Table 1).
carried out in a Hewlett-Packard (HP) Model 5890 gas chro-
Further research was conducted with M. ramanniana
matograph (Palo Alto, CA) with a flame-ionization detector ATCC 38191 to convert compound 1 to R-(+) compound 2.
(FID). An HP ultra-2 column (25 m × 0.32 mm × 0.1 mm film Cells of M. ramanniana ATCC 38191 were grown in a 280-L
thickness) was used at 230°C injector temperature, 270°C de- fermentor with 250 L of medium. During growth, cells were
tector temperature, and 230–270°C column temperature harvested periodically and used to conduct the reduction of
(4°C/min) over a 10-min run time. The retention time for com- compound 1. As shown in Table 2, cells harvested after 31 h
pound 1 was 5.2 min and for compound 2 was 5.6 min. Sam- growth and used in the bioreduction of compound 1 gave a
ples were prepared by extracting the reaction mixture with higher reaction yield (93%) and optical purity (>99%) of R-
methylene chloride/acetonitrile/isopropanol mixture (60:35:5) (+) compound 2.
and analyzing the separated organic phase directly on GC.
Cells harvested from a 380-L fermentor were evaluated for
The optical purity of compound 2 was determined by chiral the reduction of compound 1 in 5- and 15-L fermentors with
HPLC. A Bakerbond Chiralcel OD column (J.T. Baker, 3- and 10-L cell suspensions (20% wt/vol, wet cells), respec-
Phillipsburg, PA) was used at 20°C. The mobile phase con- tively. Compound 1 was supplied at 2 g/L concentration. Glu-
sisted of hexane/butanol/cyclohexanol (97:2:1). Flow rate was cose was supplemented at 70 g/L concentration. After a 24-h
1 mL/min, and the retention time was 25 min for the R-(+) reaction period, about 99% yield (99.5% optical purity) of R-
enantiomer and 31 min for S-(−) enantiomer.
(+) compound 2 was obtained from both batches. The kinet-
ics of transformation of compound 1 are shown in Table 3.
Isolation of R-(+) compound 2 from the 10-L fermentation
broth was carried out as described in the Materials and Meth-
ods section to obtain 14.2 g of product in overall 70 mol%
yield and 99% GC and HPLC purity. Isolated R-(+) com-
pound 2 gave a melting point of 115°C, specific rotation α
[D]25 of +26.8 with chloroform as solvent, and optical purity
of 99.5% as analyzed by chiral HPLC.
Results
Among microorganisms evaluated for the reduction of com-
pound 1 to 2, M. ramanniana ATCC 38191 predominantly re-
duced compound 1 to R-(+) compound 2, and Pullularia pul-
lulans ATCC 16623 reduced compound 1 to S-(−) compound
TABLE 3
Reduction of 1 to 2 by Cell Suspensions of Mortierella ramanniana ATCC 38191: Prepara-
tive 10-L Batcha
Reaction
time (h)
Concentration of
substrate 1 (g/L)
Concentration of 2 Conversion
Optical purity of
R-(+) BMY 14802 (%)
[BMY 14802] (g/L)
(%)
3
6
12
18
24
1.65
1.3
0.71
0.48
0.04
0.31
0.64
1.25
1.61
1.98
15
32
62
80
99
n.d.
n.d.
n.d.
99.4
99.5
aCells of M. ramanniana ATCC 38191 were suspended in 10 L of 0. 1 M potassium phosphate buffer
(pH 8) at 20% (wt/vol, wet cells) concentration. Glucose (200 g) and 1 (20 g) were supplied, and the
reduction was conducted in a 15-L fermentor at 500 rpm, 28°C, 12 Lpm aeration. Abbreviation:
n.d., not determined. See Table 1 for other abbreviations.
JAOCS, Vol. 74, no. 11 (1997)

CHIRAL INTERMEDIATES BY OXIDOREDUCTASES
1349
with the reaction yield of 99% and the optical purity of 98.9%
for R-(+) compound 2. At the end of the biotransformation
process, cells were removed by filtration, and product was re-
covered from the filtrate. Filtrate was adjusted to pH 8.0 and al-
lowed to stand at room temperature for 1 h. The precipitate was
filtered and dissolved in isopropyl alcohol (IPA). Product was
precipitated as a HCl salt by addition of isopropanolic HCl. Re-
crystallization from IPA yielded 24 g of R-(+) compound 2.
Overall, 80% recovery of product was obtained in 99% HPLC
purity and 98.8% optical purity.
Purification of the enzyme that reduces 1 to R-(+) compound
2 was carried out from cell extracts of M. ramanniana ATCC
38191 (Table 6). Overall, 55-fold purification of enzyme was
achieved, which gave a specific activity [nmol R-(+) compound
2 formed/min/mg protein] of 268. The purified enzyme has a
molecular weight of 29,000 daltons as determined by gel filtra-
tion and consists of a single subunit of 29,000 daltons as judged
from SDS-polyacrylamide gel electrophoresis. The purified en-
zyme was inhibited by p-hydroxy mercuribenzoate, 1,10-
phenanthroline, and 2,2′-bipyridyl at 1 mM concentration of
each inhibitor. This indicates the involvement of a sulfhydryl
group (and perhaps metals) at the enzyme active site.
TABLE 4
Reduction of 1 to [BMY-14802] by Cell Suspensions of Mortierella
ramanianna ATCC 38191: Effect of Glucose Concentrationa
Glucose
added (g/L) [BMY 14802] (g/L)
Concentration of 2 Conversion
Optical purity
R-(+) BMY 14802 (%)
(%)
0
10
20
40
0.48
1.49
1.95
1.93
24
75
98
97
99.4
99.4
99.3
aCells suspensions of M. ramanniana ATCC 38191 (8% wt/vol, wet cells)
were supplied with 1 (2 mg/mL) and glucose as indicated. Reactions were
conducted at pH 5.5, 280 rpm, 28°C, for 20 h. Compounds 1 and 2 were
analyzed by GC, and the optical purity of 2 was determined by chiral HPLC.
See Tables 1 and 2 for abbreviations.
The effect of pH and temperature on the reduction of com-
pound 1 to R-(+) compound 2 by M. ramanniana ATCC 38191
was evaluated in a 100-mL reactor. Cell suspensions (20%
wt/vol, wet cells) were supplemented with 200 mg compound 1
and 7.5 g glucose, and biotransformation was conducted at 28°C
(280 rpm) for 18 h. The optimal pH for the reduction of com-
pound 1 to R-(+) compound 2 is 5.5. The optimal temperature
for the reduction of compound 1 to R-(+) compound 2 is 28°C.
The effect of glucose and substrate concentration on the bio-
transformation of compound 1 to R-(+) compound 2 by cells of
M. ramanniana ATCC 38191 was evaluated. As shown in Table
4, 20 mg/mL of glucose was enough to obtain a 98% reaction
yield. In the absence of added glucose, only 22% reaction yield
was obtained. Substrate at 2 and 4 mg/mL concentration gave
reaction yields of 100% and 90%, respectively. At higher sub-
strate concentration (6 mg/mL), lower reaction yield (63%) and
optical purity (94.5%) were obtained for R-(+) compound 2
(Table 5).
A single-stage fermentation/biotransformation process was
developed for conversion of compound 1 to compound 2 by
cells of M. ramanniana ATCC 38191. Cells were grown in a 20-
L fermentor with 15 L of medium A as described in the Materi-
als and Methods section. After 40 h of growth in a fermentor,
when the residual glucose was depleted (0.1%), the pH of the
medium had dropped to 4.5. Upon total consumption of glucose,
the pH of the medium started to increase, indicating completion
of the fermentation process. When the pH increased to 5.5, com-
pound 1 (2 g/L) and glucose (70 g/L) were added to the fermen-
tor, and the biotransformation process was continued (Fig. 1).
The biotransformation process was completed in a 24-h period,
ANTIARRHYTHMIC AGENT (β-BLOCKER)
Stereoselective reduction of N-4-(1-oxo-2-chloroacetyl ethyl)
phenyl methane sulfonamide (3). Larsen and Lish (25) re-
ported the biological activity of a series of phenethanolamine-
bearing alkyl sulfonamido groups on the benzene ring. Within
this series, some compounds possessed adrenergic and antia-
drenergic actions, including α-adrenergic receptor block or re-
ceptor stimulation, β-adrenergic receptor block or receptor
stimulation.
D
-(+) Sotalol (Scheme 2) is a β-blocker (26).
Unlike other β-blockers, it has class III antiarrhythmic proper-
ties (27). The β-adrenergic blocking drugs, such as propra-
nolol {(1-[(-methylethyl) amino]-3-(1-naphthalenyloxy)-2-
propanol} and sotalol, have been separated chemically into the
dextro and levo rotatory optical isomers, and the activity of the
levo isomer has been demonstrated to be 50 times that of the
corresponding dextro isomer (28). In this section, we describe
the stereoselective microbial reduction of 3 to the correspond-
ing (+)-4, or (+)-N-4-(1-hydroxy-2-chloroacetyl ethyl) phenyl
methane sulfonamide. The compound (+)-4 is a key chiral in-
termediate for the synthesis of D-(+)-sotalol (Scheme 2).
TABLE 5
Reduction of 1 to [BMY-14802] by Cell Suspensions of Mortierella ramanniana:
Effect of Substrate Concentrationa
Concentration of
substrate 1 (g/L)
Reaction time Concentration of 2
Conversion
(%)
Optical purity
R-(+)-2 (%)
(h)
[BMY 14802] (g/L)
2
4
6
24
24
48
2
3.6
3.8
100
90
63
99.2
98.9
94.5
aCell suspensions of M. ramanniana ATCC 38191 (8% wt/vol, wet cells) were supplied with glucose
(75 mg/mL) and substrate as indicated. Reactions were conducted at 28°C, 280 rpm, at pH 5.5. Con-
centrations of 1 and 2 were determined by GC, and the optical purity of 2 was determined by chiral
HPLC. See Tables 1 and 2 for abbreviations.
JAOCS, Vol. 74, no. 11 (1997)

1350
R.N. PATEL ET AL.
FIG. 1. Growth of Mortierella ramanniana in a 15-L fermentor during single-stage fermenta-
tion/biotransformation. (A) Residual glucose and packed cell volume (PCV); (B) substrate 1
and compound 2. Abbreviations: 1, 1-(4-fluorophenyl)-4-[4-(5-fluoro-2-pyrimidinyl)-1-piper-
azinyl]-1-butanone; 2, R-(+)-1-(4-fluorophenyl)-4-[4-(5-fluoro-2-pyrimidinyl)-1-piperazinyl]-
1-butanol [R-(+)-BMY 14802].
tract, and 0.3% peptone. The medium was adjusted to pH 6.8
before sterilization. Cultures were grown at 25°C, 280 rpm for
48 h. Cultures were harvested by centrifugation at 18,000 × g
for 15 min, washed with 10 mM potassium phosphate buffer
pH 6.8, and used for biotransformation studies.
Transformation of compound 3 by microbial cell suspen-
sions. Cells of various microorganisms were suspended in 10
mL of 10 mM potassium phosphate buffer pH 6.8 at 20%
(wt/vol, wet cells) cell concentration and supplemented with
20 mg compound 3 and 750 mg glucose. Transformation of
Materials and Methods
Microorganisms. Microorganisms (Table 7) were obtained
from our culture collection in the Microbiology Department
of the Bristol-Myers Squibb Pharmaceutical Research Insti-
tute and from the American Type Culture Collection. Mi-
croorganisms were stored at −90°C in vials.
Growth of microorganisms. For screening purposes, one vial
of each culture was used to inoculate 100 mL of medium B,
which contained 2% glucose, 1% malt extract, 1% yeast ex-
JAOCS, Vol. 74, no. 11 (1997)

CHIRAL INTERMEDIATES BY OXIDOREDUCTASES
1351
TABLE 6
Purification of BMY 14802 Reductase from Mortierella ramanniana ATCC 38191a
Total units
(nmol/min)
Total protein
(mg)
Specific activity
(units/mg protein)
Recovery
(%)
Purification
(fold)
Step
Volume (mL)
Cell extracts
DE-52 pooled
fraction
800
173
5104
3962
1040
38
4.9
105
100
77.6
1
21.4
Phenyl sepharose
chromatography
Mono-Q column
chromatography
15
6
1700
1494
6.5
6
261
268
33
29
53
55
aReaction mixture for enzyme assay in 1 mL of 2-(N-morpholino)ethanesulfonic acid buffer (pH 5.5) contained 0.67 mM
NADP, 10 mg glucose, 4 units of glucose dehydrogenase, and 1 mg of 1. Reactions were conducted at 25%C, 100 rpm.
Concentrations of 1 and 2 were determined by GC. See Tables 1 and 2 for abbreviations. For additional details on meth-
ods, see the Materials and Methods section.
Analytical methods. The reduction of compound 3 to com-
pound 4 in the reaction mixture samples was monitored by GC
with FID. A Hewlett-Packard fused-silica capillary column
(cross-linked methyl silicone, 15 m long, 0.33-µm film thick-
ness, 0.32 mm i.d.) was used at 250°C detector temperature and
190°C injection temperature. The initial oven temperature was
195°C, and the final temperature was 205°C. Temperature was
increased at 0.5°C/min. Helium was used as the carrier gas at
25 mL/min. The retention time for substrate 3 was 10.2 min,
and for the corresponding reduced product 4 it was 9.6 min.
The optical purity of product 4 was determined by chiral
HPLC. A Bakerbond Chiralcel OB column was used at ambi-
ent temperature. Injection volume was 20 µL, mobile phase
was 65% hexane with 35% n-butanol, flow rate was 0.5
mL/min, and detector wavelength was 230 nm. The retention
time was 16.32 min for the R-(+) enantiomer and 13.63 min for
the S-(−) enantiomer.
Gas chromatography–mass spectrometry (GC–MS) analy-
sis. A GC–MS method was developed to obtain MS data on
compound 4. A Hewlett-Packard ultra 2 column (25 mm, 0.2
mm i.d., 0.33 µm) was used at 260°C injector temperature
and 260°C detector temperature. The initial oven temperature
was 280°C. Temperature was increased at 10°C/min, starting
at 5 min.
SCHEME 2
compound 3 was conducted at 28°C, 200 rpm on a rotary
shaker. Periodically, 2-mL samples were taken and extracted
with 4 mL ethyl acetate. After centrifugation, the ethyl acetate
phase was collected and filtered through a 0.2-µm LID/X filter.
Clarified ethyl acetate filtrate (0.6 mL) was analyzed by GC to
determine the concentrations of compounds 3 and 4. The re-
maining ethyl acetate filtrate (3 mL) was dried under a stream
of nitrogen. The oily residue obtained was dissolved in a
mobile phase (hexane/n-butanol, 65:35) and analyzed by chiral
HPLC to determine the optical purity of compound 4.
Nuclear magnetic resonance (NMR) analysis. The H NMR
TABLE 7
Microbial Reduction of N-4-(1-oxo-2-chloroacetyl ethyl) Phenyl Methane Sulfonamide 3a
Reaction time
(h)
Conversion to
alcohol 4 (%)
Optical purity
of alcohol 4 (%)
Microorganisms
Pichia methanolica ATCC 58403
Pullularia pullulans ATCC 16623
Hansenula polymorpha ATCC 26012
Trichoderma polysporium SC 14962
Rhodococcus sp. ATCC 29675
Rhodococcus sp. ATCC 21243
Nocardia salmonicolor SC 6310
a
40
68
22
22
22
22
22
80
73
95
86
53
64
45
26
75
99
77
91
97.5
99
Cell suspensions (20% wt/vol of wet cells) of each microbial culture in 10 mL of 10 mM phosphate
buffer were supplemented with 20 mg of substrate and 750 mg glucose. Bioreduction of 3 was car-
ried out at 28°C and 200 rpm on a rotary shaker. The concentrations of substrate 3 and product 4
(the corresponding alcohol to 3) were determined by GC, and the optical purity of 4 was determined
by chiral HPLC with a Chiralcel OB column. Abbreviation: 4, (+)-N-4-(1-hydroxy-2-chloroacetyl
ethyl) phenyl methane sulfonamide.
JAOCS, Vol. 74, no. 11 (1997)

1352
R.N. PATEL ET AL.
spectra were recorded at ambient temperature on a Brucker
Reduction of compound 3 by cell extract. One liter of cell
(Karlsruhe, Germany) AP-300 MHZ spectrometer with deuter- extract was supplemented with 2.5 g of compound 3, formate
ochloroform as solvent and 3% tetramethylsilane as an inter- dehydrogenase (500 units), 0.7 mM NAD+, and 25 g sodium
nal standard.
formate. The reaction was carried out in a pH-stat at pH 6.8,
Growth of Hansenula polymorpha ATCC 26012 in a fer- 150 rpm agitation, and 28° C. Periodically, samples were taken
mentor. Hansenula polymorpha ATCC 26012 was grown in a and analyzed for reaction yield and optical purity of compound
380-L fermentor with 250 L of medium B containing 0.25% 4 as described earlier. After complete conversion of 3 to 4, the
antifoam. Growth consisted of inoculum development stages reaction mixture was extracted with 2 vol ethyl acetate, and the
and fermentation.
product was isolated by preparative HPLC as described ear-
Flask inoculum development was carried out in F1 and F2 lier. About 2.1 g of compound 4, with 98% purity and by GC
stages. In the F1 stage, frozen vials of H. polymorpha cultures 99.4% optical purity, was isolated.
were inoculated into 100 mL of medium B in 500-mL flasks
and incubated at 25°C, 280 rpm for 48 h. In the F2 stage, 100
mL of F1-stage cultures was inoculated into 1.5 L of medium
Results
B in 4-L flasks and incubated at 25°C, 180 rpm for 24 h. A seed Microorganisms were screened for the transformation of com-
fermentor with medium B was inoculated with the entire con- pound 3 to compound 4. As shown in Table 7, among cultures
tents of a single F2-stage inoculum. Growth in a seed fermen- evaluated, Rhodococcus spp. ATCC 29675 and ATCC 21243,
tor was conducted at 28°C, 10 psig, 200 Lpm aeration, 120 rpm Nocardia salmonicolor SC 6310, and H. polymorpha ATCC
agitation for 12 h.
26012 gave desired product 4 in >90% optical purity.
Fermentors with 250 L of medium B were inoculated with Hansenula polymorpha ATCC 26012 catalyzed the efficient
13 L of seed fermentor-grown inoculum. Fermentations were conversion of compound 3 to compound 4 in 95% reaction
conducted at 25°C, 150 rpm agitation, 150 Lpm aeration. Upon yield and >99% optical purity.
depletion of carbon source (generally 24 h), cells were har-
Growth of H. polymorpha ATCC 26012 culture was carried
vested with the aid of a Sharples centrifuge. Wet cell pastes out in a 380-L fermentor as described in the Materials and
were collected and stored at −60°C until further use. About 8 Methods section. Cells harvested from the fermentor were
kg of wet cell pastes was collected from each fermentation.
used to conduct transformation in a 3-L preparative batch.
Transformation of 3 by cell suspensions of H. polymorpha. Cells were suspended in 3 L of 10 mM potassium phosphate
Transformation of 3 to 4 was conducted in a fermentor in 3-L buffer pH 7.0 (wt/vol, wet cells) at a cell concentration of 20%.
batches at 25°C, 200 rpm. Compound 3 (4 mg/mL), glucose Compound 3 (12 g) and glucose (225 g) were added to the fer-
(75 mg/mL), and cells (20%, wet cells) were used. During mentor, and the reduction reaction was carried out at 25°C, 200
bioreduction, the substrate and product concentrations were rpm, pH 7. About 95% conversion of compound 3 to com-
determined by in-process GC assays, and the optical purity of pound 4 was obtained in 20-h reaction period (Table 8). The
4 was determined by chiral HPLC.
isolation of compound 4 from the reaction mixture was carried
Isolation of compound 4. Typically, 3 L fermentation broth out by preparative HPLC as described in the Materials and
with 11 g of 4 was extracted twice with equal volumes of ethyl Methods section to obtain 8.2 g of product. The isolated 4 gave
acetate. The organic phase was separated and dried over anhy- a specific rotation [α]D Na of +20 and optical purity of >99%
drous sodium sulfate. The ethyl acetate extract was concen- as analyzed by chiral HPLC. The mass spectra and H NMR
trated under reduced pressure at 40°C. The oily liquid obtained spectra of isolated compound 4 and standard compound 4 were
was dissolved in mobile phase that consisted of 25% methanol virtually identical. Reduction of 3 to 4 was also carried out
in deionized water and purified on a preparative SepTech with cell extracts of H. polymorpha ATCC 26012. Formate de-
HPLC (Wakefield, RI). The column size was 2 × 10 inch and hydrogenase was used to regenerate cofactor NADH, required
was packed with C18 (25 µ) material purchased from SepTech. for the reduction reaction. After 18 h reaction time, 95% con-
The mobile phase was 25% methanol in deionized water. The version of 3 to 4 was obtained. About 2.1 g of 4 was isolated
detection wavelength was 230 nm, and the flow rate was 100 from the reaction mixture. The isolated 4 gave 98% purity by
mL/min. The fraction containing compound 4 was recovered, GC area and 99.4% optical purity.
and methanol was removed under reduced pressure. The water
was removed by freeze-drying. About 8.2 g of compound 4,
with 97% GC purity and >99% optical purity, was obtained.
ANTIHYPERTENSIVE DRUG
Preparation of cell extract. Hansenula polymorpha ATCC Biotransformation of Nε-carbobenzoxy (CBC)-
L
-lysine (7) to
26012 was suspended in buffer B (50 mM potassium phos- Nε-CBZ- -oxylysine (5). As requirements for optical purity of
L
phate buffer, pH 6.8 with 1 mM DTT) at a 20% (wt/vol, wet pharmaceuticals become more stringent, enzymic methods for
cells) cell concentration. Cell suspensions were disintegrated production of chiral synthons are receiving increased attention.
by three passages through a Microfluidizer (Newton, MA) at Compound 5 is an intermediate needed for synthesis of a new
13,000 psi pressure at 4°C. Disintegrated cells were cen- angiotensin-converting enzyme inhibitor with the trivial name
trifuged at 20,000 × g for 40 min at 4°C. The supernatant solu- ceronapril (29), or 6: (S)-1-[6-amino-2-{[hydroxy(4-phenyl-
tion obtained after centrifugation is referred to as cell extract.
butyl)phosphinyl]oxy}1-oxohexyl]-L-proline], which is being
JAOCS, Vol. 74, no. 11 (1997)

CHIRAL INTERMEDIATES BY OXIDOREDUCTASES
1353
The reaction was started by the addition of
L
-amino acid oxi-
TABLE 8
Preparative Scale Reduction of 3 by Hansenula polymorpha
dase, and the increase in absorbance at 460 nm was monitored
(absorbance = 11.3 mM−1cm−1). All continuous spectropho-
tometric assays were performed at 25°C.
Yield of alcohol 4
Optical purity
Reaction time (h)
(g/L)
of alcohol 4 (%)
The reaction solution for L-HIC dehydrogenase, coupled
to L-amino acid oxidase, contained, in 1 mL: 0.1 M potassium
phosphate pH 7.4, 1 mM lysine or lysine derivative, 2000
units bovine liver catalase, and 1.7 units L-HIC dehydroge-
nase. Reactions were started by addition of L-amino acid oxi-
dase, and the absorbance decrease at 340 nm was monitored.
The catalase assay contained, in 1.0 mL: 0.1 M potassium
phosphate pH 7.4, and 0.06% H2O2. The absorbance decrease
after addition of enzyme was monitored at 240 nm (ab-
sorbance = 0.0436 mM−1 cm−1). Protein was determined by
the dye-binding method of Bradford (24) with bovine serum
albumin as standard.
4
8
12
16
20
1.1
1.8
2.4
3.2
3.8
n.d.
n.d.
n.d.
99.4
99.5
aBioreduction of 3 to 4 (the corresponding chiral alcohol to 3) was carried
out in a 5-L fermentor. Cell suspension (3 L) was supplemented with 12 g
of substrate and 100 g of glucose. Reaction was conducted at 28°C, 200
rpm. Substrate 3 and product 4 concentrations were determined by GC,
and the optical purity of 4 was determined by chiral HPLC. See Table 7
for abbreviations.
developed for treatment of hypertension (Scheme 3). The pre-
Analytical methods. HPLC analyses of compound 7 trans-
formations were performed with a Hewlett-Packard Hypersil
C18 20 × 4.6 cm column, with 5 µm particle size. Column
temperature was 40°C, mobile phase was 37% methanol and
63% water with 0.05% H3PO4, flow rate was 1 mL/min, de-
tection wavelength was 215 nm, and injection volume was 5
µL. Retention times were 9.8 min for carbobenzoxy-6-
amino-2-oxohexanoic acid (8), 13.7 min for 5, and 23.3 min
for 7, with the keto acid (8) peak skewed toward higher re-
tention times. Samples were boiled for 2 min, centrifuged,
and filtered before HPLC analysis.
cursor of this intermediate,
prepare an analog of the antibiotic butirosin (30). Compound 5
has been prepared from -lysine by diazotization with sodium
nitrite and sulfuric acid, followed by treatment of the product
-oxylysine with benzyl chloroformate (29) As an alternative
approach, we have explored the enzymic oxidation of -lysine
L-oxylysine, has also been used to
L
L
L
or 7 to the keto acid 8, followed by reduction with a dehydro-
genase of appropriate enantioselectivity.
Materials and Methods
Materials. L-2-Hydroxyisocaproate (L-HIC)dehydrogenase
Optical purity of 5 was determined by derivatization and
separation of diastereomers by GC (32).
was a generous gift from Dr. H. Schütte, Gesellschaft für
Biotechnologische Forschung, Braunschweig, Germany.
Commercial sources were: lysyl oxidase, Yamasa Shoyu Co.,
Microbial growth conditions. Providencia alcalifaciens
SC 9036 was obtained from the Squibb culture collection and
is ATCC strain 13159. It was grown on medium C, described
by Szwajcer et al. (33), which contained 1% peptone, 0.2%
casein hydrolysate, 0.2% yeast extract and 0.6% NaCl at pH
7.2 to 7.4. Growth was at 37°C, 100 rpm in shake flasks. A
200-mL overnight culture was used to inoculate a 15-L tank
that contained the same medium at 37°C, stirred at 200 rpm,
and aerated at 20 Lpm. After 11 h, cells were harvested by
centrifugation, washed with 50 mM potassium phosphate
buffer pH 7.4, and stored frozen at −18° until used for bio-
transformation.
Ltd. (Choshi, Chiba, Japan); L-amino acid oxidase Type 1
from Crotalus adamanteus, Sigma Chemical Company;
polyethyleneglycol (PEG)-2000-NADH, Braunschweiger
Biotechnologie (Braunschweiger, Germany).
Enzyme assays. L-Amino acid oxidase was monitored by
coupling the H2O2 evolved in the reaction to the oxidation of
o-dianisidine catalyzed by horseradish peroxidase (31). The
reaction solution contained in 1 mL: 50 mM potassium phos-
phate pH 7.4, 1 mM
L-lysine or lysine derivative, 0.2 mM
o-dianisidine, and 10 µg (2.17 units) horseradish peroxidase.
SCHEME 3
JAOCS, Vol. 74, no. 11 (1997)

1354
R.N. PATEL ET AL.
Enzyme localization. Cells (1.4 g wet weight) were col- strains of Proteus and Providencia were found to oxidize
lected from a 17-h shake flask culture and sonicated in 15 mL compound 7 to the keto acid 8. Providencia alcalifaciens SC
50 mM potassium phosphate pH 7.4. Debris was removed by 9036 gave the highest conversion to keto acid 8 and was se-
centrifugation for 10 min at 12,000 × g. The extract super- lected for further study.
natant was centrifuged for 1 h at 101,000 × g to give a super-
Compound 7 was converted to 5 in 95% yield by the fol-
-lysine 7 (5.6 g) was added to 1 L
natant and pellet fraction. The pellet was resuspended in 2 mL lowing procedure. CBZ-
L
50 mM potassium phosphate, pH 7.4. Oxidation of 7 was of a solution that contained 0.1 M potassium phosphate pH
measured by incubating 5 mM of 7 with 0.6 mL of extract or 7.4 and 10 g of P. alcalifaciens SC 9036 cells, and the mix-
with 101,000 × g pellet or supernatant fractions in 1.5 mL of ture was incubated at 200 rpm for 27 h at 30°C. Compound 7
0.1 M potassium phosphate pH 7.4 and 3000 units catalase is nearly insoluble at pH 7.4 but went into solution as the re-
for 16 h. The amount of 8 produced by 1 mg of C. adaman- action proceeded. After 24 h, HPLC analysis indicated com-
teus venom -amino acid oxidase under these conditions was plete conversion to the keto acid 8. After cells were removed
L
used as a standard for 100% conversion in the HPLC assay.
by centrifugation, 0.2 M sodium formate, 1 mM NAD, 32
units formate dehydrogenase, and 66 units L-HIC dehydroge-
nase were added, and the solution was incubated for 64 h at
28°C. HPLC analysis indicated conversion to 18.9 mM com-
Results
The enzymatic approach to synthesis of compound 5 is shown pound 5. The isolated product (m.p. 74–75°C, standard m.p.
in Scheme 3. Crotalus adamanteus (rattlesnake) venom 76°C) had 98.4% HPLC homogeneity and 98.5% optical pu-
L
-
amino acid oxidase has been reported to oxidize compound 7 rity. 1H and 13C NMR spectra were identical with those from
to the corresponding keto acid 8 (34), and L-HIC dehydroge- the chemically produced compound: Infrared (IR) (KBr): ν
nase from Lactobacillus confusus converts the keto acid 8 to (O-H) 3335 cm−1; 1H NMR (CD3OD): δ, 1.37–1.85 (m, 6H),
compound 5 (Robison, R.S., M.G. Doremus, and L.J. Szarka, 3.03–3.18 (t, 2H), 4.03–4.13 (q, 1H), 4.75–5.00 (bm, HOD),
unpublished results). Lysyl 2-oxidase from Trichoderma 5.00–5.03 (s, 2H), 7.20–7.40 (m, 5H) ppm; 13C NMR
viride was tested as a microbial alternative to the snake venom (CD3OD): δ, 23.7, 30.9, 35.3, 41.9, 67.6, 71.6, 129.0, 129.2,
enzyme. Substrate specificity studies (Table 9) showed that the 129.7, 138.7, 159.2, 178.1 ppm.
activity of the enzyme is greatly reduced with ε-substituted ly-
sine derivatives. Activity with Nε-formyl-, acetyl-, trifluo-
TABLE 9
roacetyl-, t-butoxycarbonyl (BOC)-, or 7 was less than 2% of
Substrate Specificities of
and
L
-Amino Acid Oxidase
the activity with lysine. There was no activity with L-oxyly-
sine. On the other hand, the snake venom oxidase was active
L
-2-Hydroxyisocaproate (HIC) Dehydrogenase
Specific Activity
(units/mg oxidase)
with the ε-substituted lysine derivatives but had little activity
Enzyme
Substrate
-Lysine
-Oxylysine
-lysine
-lysine
-lysinea
Nε-Trifluoacetyl-
with L-lysine. The methyl and ethyl esters of lysine were sub-
Lysyl oxidase
Trichoderma viride
L
L
1.663
0
0.003
0.033
0
strates for the T. viride oxidase but were not utilized by the
snake venom enzyme.
The product from lysine oxidation by T. viride lysyl oxi-
dase was not a substrate for L-HIC dehydrogenase (Table 9)
or for several other dehydrogenases that were screened at pH
Nε-Acetyl-
Nε-Formyl-
Nε-BOC-
L
L
L
L
-lysine
0.027
0.003
0
Nε-CBZ-
Nε-CBZ-
L
-lysinea
-lysine methyl ester
7.4 or 9. With snake venom
, acetyl-, CBZ-, or trifluoroacetyl-
strates when the oxidation was coupled to L-HIC dehydroge-
nase, but Nε-t-BOC- -lysine was not an effective substrate.
L
-amino acid oxidase, Nε-formyl-
L
L
-lysine were good sub-
L
L
L
-Lysine methyl ester
-Lysine ethyl ester
-Lysine
0.847
1.083
0
L
-Amino acid oxidase
L
Crotalus adamanteus
L-Oxylysine
Nε-Acetyl- -lysine
Nε-Formyl- -lysine
Nε-BOC- -lysine
Nε-Trifluoroacetyl-
0.001
0.405
0.338
0.254
0.406
0.405
The product of lysine oxidation by the T. viride enzyme has
been shown to cyclize to a Schiff base (35) and this may in-
terfere with utilization by L-HIC dehydrogenase.
Alternatively, although L-HIC dehydrogenase shows
rather broad specificity (36) it may require a substituent on
the ε-amino of lysine for activity.
Synthesis of compound 5. HPLC analysis showed that 0.3
units/mL T. viride lysyl oxidase, coupled to 0.8 units/mL L-
HIC dehydrogenase, was able to convert 1 mg/mL compound
7 to 1 mg/mL compound 5 after 48 h in a reaction mixture
that also contained 1 mM NAD, 0.2 M sodium formate, 0.7
units/mL formate dehydrogenase, and 1250 units/mL catalase
in 0.1 M potassium phosphate at pH 8. In an effort to find a
more active 7 oxidase activity, several microbial strains
L
L
L
L
-lysine
Nε-CBZ-
Nε-CBZ-
L
-lysine
L
-lysine methyl ester 0.006
L
L
L
L
-Lysine methyl ester
-Lysine ethyl ester
-Lysine
0.001
0
0
0
0
0.062
0.12
0
T. viride lysyl oxidase
coupled to
-Lysine methyl ester
HIC dehydrogenase L-Lysine ethyl ester
C. adamanteus -lysine
-lysine
-lysine
Nε-Acetyl-
L
L
-amino acid oxidase Nε-Formyl-
L
coupled to
Nε-BOC-
L
HIC dehydrogenase Nε-Trifluoroacetyl-
L
-lysine
0.028
0.12
Nε-CBZ- -lysine
L
aAbbreviations: BOC, t-butoxycarbonyl-; CBZ, carbobenzoxy- .
known to possess L-amino acid oxidase were screened. Four
JAOCS, Vol. 74, no. 11 (1997)

CHIRAL INTERMEDIATES BY OXIDOREDUCTASES
1355
TABLE 10
Localization of Nε-Carbobenzoxy-L-lysine (7) Oxidase and Catalase
Activities in Providencia alcalifaciens
Catalase
Oxidasea
Protein
Fraction
(units/mg)
(% conversion)
(mg/mL)
Cell extract
1328
318
2197
97.6
96.8
0
4.3
5.1
2.4
10,100 × g pellet
101,000 × g
supernatant
aPercentage conversion of 5 mM compound 7 to the keto acid carbobenz-
oxy-6-amino-2-oxohexanoic acid (8), measured by HPLC as described in
the Materials and Methods section. See Table 1 for abbreviations.
L
-Amino acid oxidase has been associated with a particu-
late fraction of the cell in several Proteus species (37,38).
Cells, extracts, or the particulate fraction from Providencia
alcalifaciens were effective in oxidizing 7 to the keto acid 8,
but the 101,000 × g supernatant had no activity (Table 10).
No 7 oxidase activity was found in the 101,000 × g pellet
when H2O2 production was measured. When conversion to
the keto acid was measured by HPLC, 7 oxidase was found
in the particulate fraction and was absent from the supernatant
(Table 10). High catalase activity was found in both the pellet
and supernatant fractions, with higher specific activity in the
supernatant. The high level of catalase accounts for the in-
ability to measure 7 oxidation by H2O2 production.
The pH optimum for oxidation of 7 by P. alcalifaciens cells,
coupled to L-HIC dehydrogenase, was about 8.8 (Fig. 2). The
optimal pH for reduction of the keto acid 8 by L-HIC dehydro-
genase, with NADH, was about 6.6 (Fig. 3). At pH 7.4 and 0.2
mM NADH, the apparent Km for 8 was 10.4 mM, and the
Lineweaver-Burk plot was linear. With 0.2 mM PEG-NADH,
the apparent Km for the keto acid was 2.2 mM and the apparent
Vmax was 17% of the value with NADH. At pH 7.4 and 4 mM
compound 8, apparent Km for NADH was 12 mM, and there
was substrate inhibition by concentrations of NADH above 0.2
mM. Under the same conditions, the apparent Km for PEG-
NADH was 50 mM, and there was substrate inhibition above
0.4 mM. Apparent Vmax under these conditions with PEG-
NADH was 71% of the value with NADH. Schütte et al. (36)
have reported that NADH concentrations greater than 0.24 mM
inhibit L-HIC dehydrogenase.
FIG. 2. Effect of pH on biotransformations. Oxidation of Nε-carbobenz-
oxy (CBZ)-L-lysine (7): 100 mM potassium phosphate buffer, 1 mM 7
and 5 mg (wet weight)/mL Providencia alcalifaciens were incubated at
28°C and 200 rpm for 30 min. A high-pressure liquid chromatography
assay was used to determine activity.
dene]amino}oxyacetic acid (10) (Scheme 4), a new orally ac-
tive monobactam (47–49).
Procedures for the synthesis and resolution of racemic
β-hydroxyvaline (50–53) and the asymmetric synthesis of D-
β-hydroxyvaline have been reported (54), but the direct syn-
thesis of 9 has not been accomplished. In this section, we de-
scribe the direct synthesis of 9 by the reductive amination of
16 with leucine dehydrogenase from Bacillus sphaericus
ATCC 4525. Leucine dehydrogenases from B. sphaericus
(55), B. cereus (56), and B. megaterium (57) have been used
for the synthesis of branched-chain amino acids but not for
hydroxylated amino acids. The B. sphaericus enzyme has
ANTIINFECTIVE DRUG (TIGEMONAM)
Enzymatic synthesis -β-hydroxyvaline (9) from sodium α-
L
keto-β-hydroxyisovalerate (16). During the past several years,
syntheses of α-amino acids have been pursued intensively be-
cause of their importance as building blocks of compounds of
medicinal interest (39–42). New methods have been devel-
oped for the asymmetric synthesis of β-hydroxy-α-amino
acids (43–46) because of their utility as starting materials for
the total synthesis of monobactam antibiotics. Compound 9
is a key chiral intermediate needed for the synthesis of
tigemonam, or S-[Z]-{[1-(2-amino-4-thiazolyl)-2-{[2,2-di-
methyl-4-oxo-1-(sulfooxy)-3-azetidinyl]amino}-2-oxoethyli-
FIG. 3. Reduction of carbobenzoxy (CBZ)-6-amino-2-oxohexanoic
acid (8) by L-hydroxyisocaproate (HIC) dehydrogenase. Reaction
mixture in 1 mL contained 100 mM potassium phosphate buffer, 2
mM Compound 8, 0.3 mM NADH, and 1 µg L-HIC dehydrogenase.
JAOCS, Vol. 74, no. 11 (1997)

1356
R.N. PATEL ET AL.
SCHEME 4
been reported to be inactive for oxidative deamination of ser- 0.2-mL samples were evaluated in two 1-mL reaction systems
ine and threonine (58).
for synthesis of 9: (i) 1 M NH4Cl, 1 M glucose, 2 mM NAD,
and 0.1 M compound 16. (ii) 1 M ammonium formate, 2 mM
NAD, 0.1 M compound 16, and 40 units/mL formate dehy-
drogenase from Candida boidinii. Reactions were carried out
Materials and Methods
Materials. α-Keto-β-bromoisovalerate (11), ethyl α-keto-β- at 30°C for 48 h at an initial pH of 8.2.
bromoisovalerate (12), 2-chloro-3,3-dimethyloxiranecar-
Preparation of leucine dehydrogenase. Bacillus sphaeri-
boxylic acid methyl ester (13) (glycidic ester 13), 2-chloro- cus ATCC 4525 was grown in a 380-L fermentor with 250 L
3,3-dimethyloxiranecarboxylic acid-1-methylethyl ester (14) of medium E, which contained 2% yeast extract, 1% glucose,
(glycidic ester 14), 2-chloro-3,3-dimethyloxiranecarboxylic 0.2% K2HPO4, and 0.25% antifoam, adjusted to pH 7.0.
acid-1,1-dimethylethyl ester (15) (glycidic ester 15), 16, and Growth consisted of inoculum development stages and fer-
9 were synthesized by the Chemical Process Research De- mentation.
partment, Bristol-Myers Squibb Pharmaceutical Research In-
Flask inoculum development was carried out in F1 and F2
stitute as described previously (12). Polyethylene glycol-N6- stages. In the F1 stage, frozen vials of B. sphaericus ATCC
(2-aminoethyl)-NADH (PEG-NADH) and formate dehydro- 4525 cultures were inoculated into 100 mL of medium E, con-
genase were gifts from H. Schütte (Gesellschaft für tained in 500-mL flasks, and incubated at 30°C, 280 rpm for
Biotechnologische Forschung mbH). Dextran NAD (attached 16 h. In the F2 stage, 100 mL of F1-stage cultures were inoc-
through C8 to dextran D4133 with a six-carbon spacer) and ulated into 1.5 L of medium E in a 4-L flask and incubated at
leucine dehydrogenase from Bacillus sp. were purchased 30°C, 180 rpm for 24 h.
from Sigma Chemicals. Glucose dehydrogenase from B.
megaterium was purchased from Amano International En- 6 L of inoculum from F2 stage. Fermentations were con-
zyme Co. (Troy, VA). ducted at 30°C, 150 rpm agitation, 150 Lpm aeration. Upon
Fermentors with 250 L of medium E were inoculated with
Microorganisms. Microorganisms (Table 11) were ob- depletion of carbon source (generally 24 h), cells were har-
tained from our culture collection in the Microbiology vested with the aid of a Sharples centrifuge. Wet cell pastes
Department of the Bristol-Myers Squibb Pharmaceutical Re- were collected and stored at −60°C until further use. Cells
search Institute and from the American Type Culture Collec- were suspended in 10 mM potassium phosphate buffer (pH
tion. Microorganisms were stored at −90°C in vials.
7.0), containing 0.01% mercaptoethanol, and disrupted by
Growth of microorganisms. For screening purposes, one sonication. The sonicate was partially purified by heating for
vial of each culture was used to inoculate 100 mL of medium 20 min at 60°C, followed by centrifugation at 28,000 × g for
D, which contained 0.25% glucose, 0.3% soytone, 1.7% tryp- 10 min. The supernatant was stored at −20°C as source of
tone, 0.5% NaCl, 0.1% yeast extract, and 0.25% K2HPO4 ad- leucine dehydrogenase.
justed to pH 7.0 before sterilization. Cultures were grown at
Enzyme assays. Leucine dehydrogenase activity was mon-
30°C, 280 rpm for 16 h. Cultures were harvested by centrifu- itored by decrease in absorbance at 340 nm. The reaction mix-
gation at 18,000 × g for 15 min, washed with 100 mM potas- ture in 1 mL contained 16 or α-ketoisovalerate, 0.75 M
sium phosphate buffer pH 7.0, and resuspended in 5 mL of NH4Cl–NH4OH buffer, 0.3 mM NADH, and 3–10 µL cell ex-
the same buffer. Cultures were disrupted by sonication, and tract. All components except keto acid were added, and a
JAOCS, Vol. 74, no. 11 (1997)

CHIRAL INTERMEDIATES BY OXIDOREDUCTASES
1357
TABLE 11
Synthesis of
L
-β-Hydroxyvaline (9) by Bacillus Strains
Specific activitya
(units/mg protein)
L
-β-Hydroxyvaline (mM)
Strain
Experiment 1b
Experiment 2c
B. subtilis SC 13794
B. subtilis SC 10253
B. megaterium ATCC 39118
B. megaterium SC 6394
B. sphaericus ATCC 4525
B. cereus SC 14579
0.1289d
0.0343
0.1232d
0.3123d
0.8705
0.1892
0.2079
0.1039
0.5259
0.276
38
11
66.4
61
6.5
68.4
2.5
40.5
2.8
0
73.2
69.8
74.5
66.4
57.7
76.1
68.4
73.7
73.6
34.3
48.3
B. pumilis SC 8513
B. licheniformis SC 12148
B. thuringiensis SC 2928
B. brevis SC 3812
B. coagulans SC 9261
0.0194
0
aAssay contained 10 mM α-keto-β-hydroxyisovalerate (16), 0.3 mM NADH, 0.75 mM
NH4Cl–NH4OH, pH 9.5.
bAfter incubation with 1 M NH4Cl, 1 M glucose, 100 mM compound 16, and 2 mM NAD.
cAfter incubation with 1 M ammonium formate, 100 mM compound 16, 40 units/mL formate dehy-
drogenase, and 2 mM NAD.
dCompound 9 was assayed after 68 h of reaction.
blank value for NADH oxidation was measured before the re- cus ATCC 4525, and 29 units glucose dehydrogenase from
action was started by the addition of keto acid. Further details B. megaterium.
are given in the legends to the Tables and Figures.
HPLC analysis. Samples were diluted with water and
Glucose dehydrogenase activity was monitored by the in- heated in a boiling water bath for 2 min to stop the reaction
crease in absorbance at 340 nm. The reaction mixture in 1 and to precipitate proteins. Compound 9 was assayed with a
mL contained 0.5 M glucose, 0.1 M Tris chloride buffer pH Hewlett-Packard 1090 HPLC equipped with a diode array de-
8.0, 3 mM NAD, and 10 µL cell extract. For determining op- tector. A Bakerbond Chiralpak WH 25 × 0.46 cm column was
timal pH, 0.75 M NH4Cl–NH4OH buffer was substituted for used. Injection volume was 20 mL, mobile phase was 0.3 mM
Tris chloride. The reaction was started by addition of en- CuSO4, flow rate was 1.5 mL/min, temperature was 45°C,
zyme. Protein was determined by the dye-binding method and detection wavelength was 230 nm.
of Bradford (24), with bovine serum albumin as standard.
A solution of 9 (16 mL, 353 mmol obtained by enzymatic
Preparative synthesis of 9. Preparative reactions with reductive amination) was boiled for 2 min, centrifuged to re-
glucose dehydrogenase and leucine dehydrogenase were run move protein, and then chromatographed over Dowex-50 H+
in an initial volume of 16 mL at pH 8.5 and 30°C. pH was (Dow Chemical, Midland, MI) in a column (3 × 13.5 cm). The
maintained by addition of 3 M NH4OH with a Brinkmann material was eluted with H2O (250 mL) and 1 M NH4OH (250
pH-stat. Reactions contained 0.1 to 0.5 M compound 16, 1 mL), and fractions (25 mL each) were monitored by thin-layer
M glucose, 0.2 M NH4Cl, 0.5 mM NAD, 0.01% mercap- chromatography (silica gel; EtOAc/EtOH/AcOH/H2O, 5:2:1:1;
toethanol, 44 units leucine dehydrogenase from B. sphaeri- Rf of 9 = 0.21). Homogeneous fractions were combined, and
SCHEME 5
JAOCS, Vol. 74, no. 11 (1997)

1358
R.N. PATEL ET AL.
HPLC analysis of the reaction mixture indicated a 71% con-
TABLE 12
version of the ester 12 to compound 9. Compound 9 was iso-
lated from the reaction as described in the Materials and
Methods section. HPLC analysis of the recovered material
showed that leucine dehydrogenase produced exclusively the
Effect of Pyridine Nucleotide Derivatives on Synthesis of -β-Hy-
droxyvaline (9)
L
L-β-Hydroxyvaline (mM)
Formate
Glucose
Pyridine nucleotide
dehydrogenasea
dehydrogenaseb
L
-isomer of β-hydroxyvaline. Compound 16 from compound
11 was also converted to compound 9 in 82% reaction yield
by a similar procedure.
1 mM PEG-NADH
1 mM Dextran-NAD
1 mM NAD
82.1
34.1
81.8
10.1
79.9
77.3
Reductive amination of compound 16 by leucine dehydro-
genase, coupled to glucose dehydrogenase, was carried out.
Both enzymes had optimal pH for activity of 8.5. Rates of 9
formation in a coupled system, containing both enzymes,
were similar with 0.5, 1, or 2 mM NAD (data not shown).
When reactions were carried out in a pH-stat as described in
the Materials and Methods section, solutions of 16, prepared
from 0.5 M 12, 0.25 M 13, 0.25 M 14, and 0.25 M 15
(Scheme 5) were converted to 9 in yields of 75, 100, 71, and
aReaction was carried out for 22 h at 30°C with 1 M ammonium formate,
pH 8.5, 100 mM compound 16, 0.01% mercaptoethanol, 5 units/mL for-
mate dehydrogenase, and 6 units/mL leucine dehydrogenase from Bacil-
lus sphaericus.
bReaction was carried out for 22 h at 30°C with 1 M glucose, 1 M ammo-
nium chloride, pH 8.5, 100 mM compound 16, 0.01% mercaptoethanol,
5 units/mL glucose dehydrogenase, and 6 units/mL leucine dehydroge-
nase from B. sphaericus. See Table 11 for abbreviation.
water was evaporated under vacuum. This process was re- 83%, respectively.
peated two more times. The product was dissolved in water (25
When reduced pyridine nucleotide was regenerated with
mL) and lyophilized to give 0.593 g (54% yield) of 9 as a light formate dehydrogenase, PEG-NADH was as effective as
brown solid. The sample was dried over P2O5 at 40°C NAD, and dextran-NAD was less effective for synthesis of 9
under vacuum for 4 h. M.p., 202°C dec; IR (KBr): ν (C=O) by leucine dehydrogenase (Table 12). Conversely, with glu-
1611 cm−1; 1H NMR (D2O): δ, 1.15 (s, 3H), 1.38 (s, 3H), 3.5 cose dehydrogenase to reduce pyridine nucleotide, PEG-
(s, 1H) ppm; 13C NMR (D2O): δ, 24.3, 28.3, 64.4, 70.8, 173.1 NADH was relatively ineffective for synthesis of 9, but dex-
ppm; [α]D +3° (c = 3.5, H2O); +11.9° (1.3, 6 N HCl). Anal. cal- tran-NAD was as effective as NAD.
c’d. for C5H11NO3·0.13 H2O: C, 44.32; H, 8.38; N, 10.34;
The apparent Km for 16 (prepared from 12) was 11.5 mM,
H2O, 1.73; found: C, 44.67; H, 8.28; N, 10.17; H2O, 1.89 (Karl compared to a value of 1.06 mM for α-ketoisovalerate, when
Fischer determination).
using heat-treated extracts of B. sphaericus ATCC 4525 as
source of leucine dehydrogenase. The apparent Vmax with 16
was 41% of the value with α-ketoisovalerate.
Results
Distribution of 16 amination activity in Bacillus strains was
examined. Leucine dehydrogenase activity is mainly found in
Bacillus strains (56). Screening of Bacillus strains was car-
ried out to find a strain with high specific activity for reduc-
tive amination of 16 and to identify any strains able to regen-
erate NADH for the reaction. Of various strains screened
(Table 11), B. sphaericus ATCC 4525 has the highest specific
activity. Bacillus megaterium ATCC 39118 has been selected
for high levels of glucose dehydrogenase (57) and was able
to produce 9 when glucose, NAD, NH4Cl, and 16 were added
to cell extracts. All other strains had less than 6% of the glu-
cose dehydrogenase activity of this strain (data not shown),
but extracts of a few of the strains were also able to produce
9 when provided with glucose, NAD, NH4Cl, and 16. Bacil-
lus cereus ATCC 14579 was the only strain able to produce
significant amounts of 9 when intact cells were provided with
glucose and 16, and gave a maximum conversion of about
50%.
REFERENCES
1. Sih, C.J., and C.S. Chen, Microbial Asymmetric Catalysis:
Enantioselective Reduction of Ketones, Angew. Chem. Int. Engl.
23:570–578 (1984).
2. Jones, J.B., Enzymes in Organic Synthesis, Tetrahedron
42:3351–3403 (1986).
3. Csuz, R., and B.I. Glanzer, Baker’s Yeast Mediated Transfor-
mations in Organic Chemistry, Chem. Rev. 91:49–97 (1991).
4. Crosby, J., Synthesis of Optically Active Compounds: A Large
Scale Perspective, Tetrahedron 47:4789–4846 (1991).
5. Simon, H., J. Bader, H. Gunther, S. Neumann, and J. Thanos, Chi-
ral Compounds Synthesized by Biocatalytic Reductions, Angew.
Chem. Int. Engl. 24:539–553 (1985).
6. Stinson, S.C., Chiral Drugs, Chem. Eng. News, Sept. 28, 46–79
(1992).
7. Santaneillo, E., P. Ferraboschi, P. Grisenti, and A. Manzocchi,
The Biocatalytic Approach to the Preparation of Enantiomerically
Pure Chiral Building Blocks, Chem. Rev. 92:1071–1140 (1992).
8. Margolin, A.L., Enzymes in the Synthesis of Chiral Drugs, En-
zyme Microb. Technol. 15:266–280 (1993).
Enzymatic reductive amination of 16 by leucine dehydro-
genase, coupled to formate dehydrogenase, was carried out
as outlined in Scheme 4. A solution of 0.5 M compound 16
from compound 12, 2 M ammonium formate, 2 mM NAD,
1.5 mg (50 units) leucine dehydrogenase from Bacillus sp.,
and 40 mg (80 units) formate dehydrogenase from C. boidinii
in 16 mL of buffer (pH 8) was incubated at 30°C for 41 h.
9. Wong, C.-H., and G.M. Whitesides, Enzymes in Synthetic Or-
ganic Chemistry, Tetrahedron Organic Chemistry Series, 1994,
Vol. 12, Pergamon, Elsevier Science Ltd., Oxford.
10. Patel, R.N., A. Banerjee, M. Liu, R.L. Hanson, R.Y. Ko, J.M.
Howell, and L.J. Szarka, Microbial Reduction of 1-(4-Fluo-
rophenyl)-4-[4-(5-fluoro-2-pyrimidinyl)-1-piperazinyl]butan-1-
one, Biotechnol. Appl. Biochem. 17:139–153 (1993).
11. Patel, R.N., A. Banerjee, C.J. McNamee, and L. Szarka, Stereo-
JAOCS, Vol. 74, no. 11 (1997)

CHIRAL INTERMEDIATES BY OXIDOREDUCTASES
1359
selective Microbial Reduction of N-(4-(1-Oxo-2-chloroacetyl
Ethyl) Phenyl Methane Sulfonamide, J. Appl. Microbiol Biotech-
nol. 40:241–245 (1993).
[6-Amino-2-[[hydroxy(4-phenylbutyl) phosphinyl]oxy]1-oxo-
hexyl]- -proline, a Novel Orally Active Inhibitor of ACE, J.
Med. Chem. 31:204–212 (1988).
L
12. Hanson, R.L., J. Singh, T.P. Kissick, R.N. Patel, L.J. Szarka, and 30. Haskell, T.H., R. Rodebaugh, N. Plessas, D. Watson, and R.D.
R.H. Mueller, Synthesis of
Hydroxy-Isovalerate Using Leucine Dehydrogenase from Bacil-
lus sp., Bioorg. Chem. 18:116–130 (1990).
L
-β-Hydroxyvaline from α-Keto-β-
Westland, The Preparation and Biological Activity of Novel
Amino Acid Analogs of Butirosin, Carbohydr. Res. 28:263–280
(1973).
13. Hanson, R.L., K.S. Bembenek, R.N. Patel, and L.J. Szarka, Trans- 31. Berezov, T.T., and E.V. Lukasheva, Effect of Substrate and Prod-
-Lysine to CBZ- -Oxylysine Using uct Analogs on the Activity of -Lysine α-Oxidase from Tricho-
formation of N-ε-CBZ-
L
L
L
-
L
Amino Acid Dehydrogenase from Lactobacillus confusus, Appl.
Microbiol. Biotechnol. 37:599–603 (1992).
derma sp., Biochem. Int. 17:529–534 (1988).
32. Jemal, M., and A.I. Cohen, Determination of Enantiomeric Purity
of Z-Oxylysine by Capillary Gas Chromatography, J. Chro-
matogr. 394:388–394 (1987).
14. Ferris, C.D., D.J. Hirsch, B.P. Brooks, and S.H. Snyder, σ-Recep-
tors: From Molecule to Man, J. Neurochem. 57:729–737 (1991).
15. Junien, J.L., and B.E. Leonard, Drugs Acting on σ and Phencycli- 33. Szwajcer, E., P. Brodelius, and K. Mosbach, Production of α-
dine Receptors: A Review of Their Nature, Function, and Possi-
ble Therapeutic Importance, Clin. Neuropharmacol. 12:353–374
(1989).
Keto Acids: 2. Immobilized Whole Cells of Providencia sp. PCM
1298 Containing -Amino Acid Oxidase, Enzyme Microb. Tech-
nol. 4:409–413 (1982).
L
16. Martin, W.R., C.G. Eades, J.A. Thompson, R.E. Huppter, and 34. Meister, A., The α-Keto Analogues of Arginine, Ornithine and
P.E. Gilbert, The Effect of Morphine- and Nalophrine-Like Drugs Lysine, J. Biol. Chem. 206:577–585 (1954).
in the Nondependent and Morphine-Dependent Chronic Spinal 35. Kusakabe, H., K. Kodama, A. Kuninaka, H. Yoshino, H. Misono,
Dog, J. Pharmacol. Exp. Ther. 197:517–530 (1976).
17. Walker, J.M., W.D. Bower, F.O. Walker, R.R. Matsumoto, B.D.
Costa, and K.C. Rice, Sigma Receptors: Biology and Function,
Pharmacol. Rev. 42:355–402 (1990).
18. Steinfels, G.F., S.W. Tam, and L. Cook, Electrophysiological Ef-
fects of Selective σ-Receptor Agonists, Antagonists, and the Se-
lective Phencyclidine Receptor Agonist MK-801 on Midbrain
and K. Soda, A New Antitumor Enzyme, L-Lysine α-Oxidase
from Trichoderma viride. Purification and Enzymological Prop-
erties, Ibid. 255:976–981 (1980).
36. Schütte, H., W. Hummel, and M.R. Kula,
L-2-Hydroxyiso-
caproate Dehydrogenase: A New Enzyme from Lactibacillus con-
fusus for the Stereospecific Reduction of 2-Ketocarboxylic Acids,
Appl. Microbiol. Biotechnol. 19:167–176 (1984).
Dopamine Neurons, Neuropsychopharmacology 2:201–207 37. Pelmont, J., G. Arlaud, and A.M. Rossat, L-Aminoacide oxydases
(1989).
des envelopes des Proteus mirabilis: Properties generales,
Biochimie 54:1359–1374 (1972).
19. Massamiri, T., and S.P. Duckles, Multiple Vascular Effects of
Sigma and PCP Ligands: Inhibition of Amine Uptake and 38. Duerre, J.A., and S. Chakrabarty,
L
-Amino Acid Oxidase of Pro-
teus rettgeri, J. Bacteriol. 121:656–663 (1975).
39. Williams, R.M., Asymmetric Synthesis of
Contractile Responses, J. Pharmacol. Exp. Ther. 253:124–129
(1990).
20. Martinez, J.A., and L. Bueno, Buspirone Inhibits Corticotropin-
Releasing Factor and Stress-Induced Cecal Motor Response in
Rats by Acting Through 5-HT1A Receptors, Eur. J. Pharm.
202:379–383 (1991).
D
-β-Hydroxyvaline, in
Synthesis of Optically Active α-Amino Acids (edited by J.E. Bald-
win and P.D. Magnus), Pergamon, Oxford/New York, 1989, Vol.
7, pp. 185–195.
40. Kamphuis, J., W.H. Boesten, Q.B. Broxterman, H.F.M. Hermes,
J.A.M. van Balken, E.M. Meijer, and H.E. Shoemaker, New De-
velopments in Chemo-Enzymatic Production of Amino Acids,
Adv. Biochem. Eng. Biotechnol. 42:134–186 (1990).
21. Taylor, D.P., M.S. Eison, S.L. Moon, and F.D. Yocca, BMY-
14802: A Potential Antipsychotic with Selective Affinity for σ-
Binding Sites, Adv. Neuropsy. Psychopharm. 1:307–315 (1991).
22. Yevich, J.P., and W.G. Lobeck, Antipsychotic 1-Fluorophenyl- 41. Sykes, R.B., C.M. Cimarusti, D.P. Bonner, K. Bush, D.M. Floyd,
butyl-4-(2-pyrimidinyl)piperazine Derivatives, U.S. Patent
4,605,655 (1986).
23. Yevich, J.P., J.S. New, and D.W. Smith, Synthesis and Biological
Evaluation of 1-(1,2-Benzisothiazol-3-yl)- and (1,2-Benzisoxa-
N.H. Georgopadakou, W.H. Koster, W.C. Liu, W.L. Parker, P.A.
Principle, M.L. Rathnum, W.A. Slusarchyk, W.H. Trejo, and J.S.
Wells, Monocyclic β-Lactam Antibiotics Produced by Bacteria,
Nature (London) 291:489–491 (1981).
zol-3-yl)piperazine Derivatives as Potential Antipsychotic 42. Parker, W.L., J. O’Sullivan, and R.B. Sykes, Naturally Occurring
Agents, J. Med. Chem. 29:359–369 (1986). Monobactams, Adv. Appl. Microbiol. 31:181–205 (1986).
24. Bradford, M.M., A Rapid and Sensitive Method for the Quan- 43. Roemmele, R.C., and H. Rapoport, Chirospecific Synthesis of
titation of Microgram Quantities of Proteins Utilizing the
Principle of Protein Dye Binding, Anal. Biochem. 72:248–254
(1976).
β-Hydroxy-α-Amino Acids, J. Org. Chem. 54:1866–1875
(1989).
44. Ito, Y., M. Sawamura, E. Shirakawa, K. Hayashizaki, and T.
Hayashi, Asymmetric Synthesis of β-Hydroxy-α-alkylamino
Acids by Asymmetric Aldol Reaction of α-Isocyanocarboxy-
lates Catalyzed by Chiral Ferrocenylphosphine-Gold(I) Com-
plexes, Tetrahedron 44:5253–5262 (1988).
25. Larsen, A.A., and P.M. Lish, A New Bio-Isostere: Alkyl-
sulphonamido-Phenethanolamines, Nature 203:1283–1284
(1964).
26. Uloth, R.H., J.R. Kirk, W.A. Gould, and A.A. Larsen, Sulfonali-
dies. I. Monoalkyl- and Arylsulfonamidophenethanolamines. J. 45. Guanti, G., L. Banfi, and E. Narisano, Enantiospecific and Di-
Med. Chem. 9:88–96 (1966)
astereoselective Synthesis of Anti α-Hydrazo- and α-Amino-α-
Hydroxyacids Through “Electrophilic Amination” of β-Hydrox-
yesters, Ibid. 44:5553–5562 (1988).
27. Lish, P.A., J.H. Weikel, and K.W. Dungan, Pharmacological and
Toxicological Properties of Two New β-Adrenergic Receptor An-
tagonists, J. Exp. Ther. 149:161–173 (1965).
28. Somani, P., and T. Bachand, Blockade of Cardiac Effect of Iso-
proterenol by the Stereoisomers of Sotalol, Eur. J. Pharmacol.
7:239–247 (1969).
29. Karanewsky, D.S., M.C. Badia, D.W. Cushman, J.M. DeFor-
rest, T. Dejneka, M.J. Loots, M.G. Perri, E.W. Petrillo, Jr., and
J.R. Powell, (Phosphinyloxy)Acyl Amino Acid Inhibitors of
46. Evans, D.A., E.B. Sjogren, A.E. Weber, and R.E. Conn, Asym-
metric Synthesis of Anti-β-hydroxy-α-amino acids, Tetrahe-
dron Lett. 28:39–42 (1987).
47. Gordon, E.M., M.A. Ondetti, J. Pluscec, C.M. Cimarusti, D.P.
Bonner, and R.B. Sykes, O-Sulfated β-Lactam Hydroxamic Acids
(Monosulfactams). Novel Monocyclic β-Lactam Antiobiotics of
Synthetic Origin, J. Amer. Chem. Soc. 104:6053–6060 (1982).
Angiotensin Converting Enzyme (ACE). 1. Discovery of (S)-1- 48. Parker, W.L., E.M. Cohen, and W.H. Koster, U.S. Patent
JAOCS, Vol. 74, no. 11 (1997)

1360
R.N. PATEL ET AL.
4,741,220 (1988).
ses via Heterocyclic Intermediates xv. Enantioselective Synthe-
-Valine or (S)-O-O-Di-
methyl-α-methyldopa as Chiral Auxiliary Reagents, Synthe-
sis:868–870 (1982).
49. Slusarchyk, W.A., T. Dejneka, J.Z. Gougoutas, W.H. Koster,
D.R. Kronenthal, M.F. Malley, M.G. Perri, F.L. Routh, J.E. Sun-
deen, E.R. Weaver, R. Zahler, J.D. Godfrey, R.H. Mueller, and
sis of (R)-(−)-β-Hydroxyvaline Using
L
D.J. Langen, β-Lactam Synthesis: Chemospecific Sulfonation 55. Wiechmann, R., C. Wandrey, A.F. Buckmann, and M.-R. Kula,
and Cyclization of the β-Hydroxyvaline Nucleus, Tetrahedron
Lett. 27:2789–2792 (1986).
Continuous Enzymatic Transformation in an Enzyme Mem-
brane Reactor with Simultaneous NAD(H)-Regeneration,
Biotechnol. Bioeng. 23:2789–2802 (1981).
50. Godfrey, J.D., R.H. Mueller, and D.J. Van Langen, β-Lactam
Synthesis: Cyclization vs. 1,2-Acyl Migration–Cyclization. The 56. Schütte, H., W. Hummel, H. Tsai, and M. Kula,
L
-Leucine De-
hydrogenase from Bacillus cereus, Appl. Microbiol. Biotechnol.
22:306–317 (1985).
Mechanism of the 1,2-Acyl Migration-Cyclization, Ibid.
27:2793–2796 (1986).
51. Edwards, G.W., and M.L. Minthorn, Jr., Preparation and Prop- 57. Monot, F., Y. Benoit, J. Lemal, A. Honorat, and D. Ballerini,
erties of the Optical Antipodes of β-Methoxyvaline and β-Hy-
droxyvaline, Canad. J. Biochem. 46:1227–1230 (1968).
52. Berse, C., and P. Bessette, A Novel Synthesis of β-Hydroxyva-
line, Canad. J. Chem. 49:2610–2611 (1971).
Leucine Dehydrogenase from Bacillus megaterium, in Proceed-
ings of the European Congress on Biochemistry, edited by O.M.
Neijssel, R.R. van der Meer, and K.C.A.M. Luyben, Elsevier,
Amsterdam, 1987, Vol. 2, pp. 42–45.
53. Beyerman, H.C., L. Maat, D. De Rijke, and J.P. Visser, Absolute 58. Oshima, T., H. Misono, and K. Soda, Properties of Crystalline
Configuration of β-Hydroxyvaline, Recl. Trav. Chim. Pays-Bas
86:1057–1060 (1967).
Leucine Dehydrogenase from Bacillus sphaericus, J. Biol.
Chem. 253:5719–5725 (1978).
54. Schollkopf, U., J. Nozulak, and U. Groth, Asymmetric Synthe-
JAOCS, Vol. 74, no. 11 (1997)