Enzymatic synthesis of L-6-hydroxynorleucine

Article abstract of DOI:10.1016/S0968-0896(99)00158-3

L-6-Hydroxynorleucine, a key chiral intermediate used for synthesis of a vasopeptidase inhibitor, was prepared in 89% yield and >99% optical purity by reductive amination of 2-keto-6-hydroxyhexanoic acid using glutamate dehydrogenase from beef liver. In an alternate process, racemic 6- hydroxynorleucine produced by hydrolysis of 5-(4-hydroxybutyl)hydantoin was treated with D-amino acid oxidase to prepare a mixture containing 2-keto-6- hydroxyhexanoic acid and L-6-hydroxynorleucine followed by the reductive amination procedure to convert the mixture entirely to L-6-hydroxynorleucine, with yields of 91 to 97% and optical purities of >99%.

Full text of DOI:10.1016/S0968-0896(99)00158-3

Bioorganic & Medicinal Chemistry 7 (1999) 2247±2252
Enzymatic Synthesis of L-6-hydroxynorleucine
Ronald L. Hanson, a,* Mark D. Schwinden, b Amit Banerjee, a David B. Brzozowski, a
Bang-Chi Chen, c Bharat P. Patel, b Clyde G. McNamee, a Gus A. Kodersha, c
David R. Kronenthal, b Ramesh N. Patel a and Laszlo J. Szarka a
aDepartment of Microbial Technology, Bristol-Myers Squibb, One Squibb Drive, New Brunswick, NJ 08903, USA
bChemical Process Research, Bristol-Myers Squibb, P.O. Box 4000, Princeton, NJ 08540, USA
cChemical Process Development, Bristol-Myers Squibb, One Squibb Drive, New Brunswick, NJ 08903, USA
Received 24 June 1998
AbstractÐl-6-Hydroxynorleucine, a key chiral intermediate used for synthesis of a vasopeptidase inhibitor, was prepared in 89%
yield and >99% optical purity by reductive amination of 2-keto-6-hydroxyhexanoic acid using glutamate dehydrogenase from beef
liver. In an alternate process, racemic 6-hydroxynorleucine produced by hydrolysis of 5-(4-hydroxybutyl)hydantoin was treated
with d-amino acid oxidase to prepare a mixture containing 2-keto-6-hydroxyhexanoic acid and l-6-hydroxynorleucine followed by
the reductive amination procedure to convert the mixture entirely to l-6-hydroxynorleucine, with yields of 91 to 97% and optical
purities of >99%. # 1999 Elsevier Science Ltd. All rights reserved.
Introduction
methyl ester has been reported, but very low reactant
and product concentrations were required for success of
one of the steps.11
l-6-Hydroxynorleucine is a chiral intermediate useful
for the synthesis of a vasopeptidase inhibitor now in
clinical trials1 and for the synthesis of C-7 substituted
azepinones as potential intermediates for other anti-
hypertensive metalloprotease inhibitors.2 It has also
been used for the synthesis of siderophores,3±5 indo-
spicine,6 and peptide hormone analogues.7,8 Previous
synthetically useful methods for obtaining this inter-
mediate have involved synthesis of the racemic com-
pound followed by enzymatic resolution. d-Amino acid
oxidase has been used to convert the d-amino acid to
the ketoacid leaving the l-enantiomer which was iso-
lated by ion exchange chromatography.9 In a second
approach, racemic N-acetylhydroxynorleucine has been
treated with l-amino acid acylase to give the l-enantio-
mer.1,7 Both of these resolution methods give a maxi-
mum 50% yield and require separation of the desired
product from the unreacted enantiomer at the end of
the reaction. Enzymatic hydroxylation of l-norleucine
by an activated form of phenylalanine hydroxylase has
also been reported to give 6-hydroxynorleucine on an
Reductive amination of ketoacids using amino acid
dehydrogenases has been shown to be a useful method
for synthesis of natural and unnatural amino acids.12,13
We have previously reported the synthesis of l-b-
hydroxyvaline, an intermediate needed for the synthesis
of the monobactam tigemonam, from the corresponding
ketoacid using leucine dehydrogenase from Bacillus
species.14 We report here the synthesis and conversion
of 2-keto-6-hydroxyhexanoic acid to l-6-hydroxy-
norleucine by reductive amination using beef liver glu-
tamate dehydrogenase and glucose dehydrogenase from
Bacillus sp. for regeneration of NADH. To avoid the
lengthy chemical synthesis of the ketoacid, a second
route was developed to prepare the ketoacid by treat-
ment of racemic 6-hydroxynorleucine (readily available
from hydrolysis of 5-(4-hydroxybutyl)hydantoin15) with
d-amino acid oxidase from porcine kidney or Trigo-
nopsis variabilis and catalase followed by the reductive
amination procedure to convert the mixture to l-6-
hydroxynorleucine.
analytical scale, but not on a preparative scale.10
A
synthesis of l-6-hydroxynorleucine starting from lysine
Results and Discussion
Key words: l-6-Hydroxynorleucine; glutamate dehydrogenase; phenyl-
alanine dehydrogenase; d-amino acid oxidase; 2-keto-6-hydroxy-
hexanoic acid.
A variety of ketoacids can be converted to l-amino acids
by treatment with a suitable amino acid dehydrogenase.
* Corresponding author.
0968-0896/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(99)00158-3

2248
R. L. Hanson et al. / Bioorg. Med. Chem. 7 (1999) 2247±2252
2-Keto-6-hydroxyhexanoic acid was prepared by the
route described under Experimental. Initial screening
using HPLC analysis, with formate dehydrogenase for
regeneration of NADH, showed that phenylalanine
dehydrogenase from Sporosarcina sp. and beef liver
glutamate dehydrogenase converted 0.1 M 2-keto-6-
hydroxyhexanoic acid, sodium salt mixture completely
to l-6-hydroxynorleucine. Additional screening of
amino acid dehydrogenases with spectrophotometric
enzyme assays using the 2-keto-6-hydroxyhexanoic acid,
sodium salt substrate mixture showed that glutamate
dehydrogenases from Candida utilis and Proteus sp.
were active using NADPH but not NADH. Leucine
dehydrogenase partially puri®ed from Bacillus sphaer-
icus ATCC 452514 and alanine dehydrogenase from
Bacillus subtilis were not active. Of 132 cultures
screened, extract from Thermoactinomyces intermedius
ATCC 33205 was the most active (Table 1). This
strain has been shown to be a source of thermostable
phenylalanine dehydrogenase16 as well as leucine
dehydrogenase.17
l-6-hydroxynorleucine had risen to >99%, the reduc-
tive amination procedure was used to convert the mix-
ture containing 2-keto-6-hydroxyhexanoic acid and l-6-
hydroxynorleucine entirely to l-6-hydroxynorleucine
with yields of 91±97% and optical purities of >99%.
Sigma porcine kidney D-amino acid oxidase and beef
liver catalase or Trigonopsis variabilis whole cells
(source of oxidase and catalase)18 were used successfully
for this transformation.
Conclusion
l-6-Hydroxynorleucine was readily prepared from the
corresponding ketoacid by reductive amination using
beef liver glutamate dehydrogenase. Phenylalanine
dehydrogenase from Sporosarcina sp. or Thermo-
actinomyces intermedius extract were also e€ective.
Either formate dehydrogenase or glucose dehy-
drogenase was useful for NADH regeneration. Pre-
paration of the ketoacid required several steps, but a
shorter route was found in which racemic 6-hydroxy-
norleucine was prepared by hydrolysis of 5-(4-hydroxy-
butyl)hydantoin. Conversion of the d-enantiomer to the
ketoacid using d-aminoacid oxidase followed by in situ
reductive amination gave nearly complete conversion to
l-6-hydroxynorleucine.
Beef liver glutamate dehydrogenase was used for pre-
parative reactions at 10% total substrate concentration.
As depicted in Scheme 1, 2-keto-6-hydroxyhexanoic
acid, sodium salt, in equilibrium with 2-hydroxy-
tetrahydropyran-2-carboxylic acid, sodium salt, is con-
verted to l-6-hydroxynorleucine and the reaction
requires ammonia and reduced nicotinamide adenine
dinucleotide (NADH). Nicotinamide adenine dinucleo-
tide (NAD) produced during the reaction was recycled
to NADH by the oxidation of glucose to gluconic acid
using glucose dehydrogenase from Bacillus megaterium.
The optimum pH for glutamate dehydrogenase with
this substrate was determined to be about 8.76. We
previously reported that glucose dehydrogenase had a
broad pH optimum centered at about 8.5.14 The kinetics
of the reaction are shown in Figure 1. Reaction was
complete in about 3 h with reaction yields of 89±92%,
and optical purity was >99%.
Experimental
HPLC. Analysis of optical purity and amount of 6-
hydroxynorleucine was performed with a Chiralpak
WH (Daicel Chemical Industries, Ltd.) 25Â0.46 cm col-
umn; mobile phase was 0.3 mM CuSO4; ¯ow rate was
1 ml/min; temperature was 40ꢀC; and detection was at
230 nm.
Screening of strains. Strains inoculated from frozen vials
were shaken at 220 rpm at 28ꢀC for 48 h in 50 mL med-
ium containing (g/L): l-phenylalanine, 10; peptone, 10;
yeast extract, 5; K2HPO4, 2; NaCl, 1; and MgSO4
.7H2O, 0.2. Cells were harvested by centrifugation,
washed with 50 mM potassium phosphate bu€er pH 7,
and resuspended in 5 mL of the same bu€er containing
1 mM dithiothreitol. Cells were sonicated for 2 min,
then centrifuged for 20 min at 101,000 g. Extracts were
assayed spectrophotometrically for reductive amination.
The assay for amino acid dehydrogenase contained in a
volume of 1 mL: 0.75 M NH4OH adjusted to pH 8.75
with HCl, 0.4 mM NADH, 25 mg/mL 2-keto-6-hydro-
xyhexanoic acid salt mixture,19 and extract. Absorbance
change per min at 340 nm was used to calculate the
activity of the enzyme.
Chemical synthesis and isolation of 2-keto-6-hydroxy-
hexanoic acid required several steps. In a second more
convenient process (shown in Scheme 2), the ketoacid
was prepared by treatment of racemic 6-hydroxy-
norleucine (produced by hydrolysis of 5-(4-hydroxy-
butyl) hydantoin) with d-amino acid oxidase and
catalase. After the optical purity of the remaining
Table 1. 2-Keto-6-hydroxyhexanoic acid dehydrogenase in microbial
strains
Strain
No.
Units/mLa
Bacillus licheniformis
Bacillus licheniformis
Rhodococcus sp
Sporosarcina ureae
Sporosarcina ureae
Thermoactinomyces intermedius
SC12772
SC12148
SC13810
ATCC 6473
ATCC 13888
ATCC 33205
0.0099
0.0368
0.0177
0.0103
0.0613
0.2161
Materials. 5-(4-Hydroxybutyl)hydantoin was obtained
from Hampshire Chemical Company. Enzymes were
purchased from the following sources: d-amino acid
oxidase, catalase, glutamate, phenylalanine, and alanine
dehydrogenases, Sigma; formate dehydrogenase, Boeh-
ringer Mannheim; glucose dehydrogenase, Amano.
a
Cells from a 50 mL culture were assayed as described in Experi-
menal, except that Thermoactinomyces intermedius was grown at 53ꢀC
and assayed at 50ꢀC.

R. L. Hanson et al. / Bioorg. Med. Chem. 7 (1999) 2247±2252
2249
Scheme 1. Conversion of 2-keto-6-hydroxyhexanoic acid to l-6-hydroxynorleucine.
stirred at room temperature for 30 min. The homo-
geneous reaction mixture was concentrated in vacuo
until a white solid precipitated. The mixture was slur-
ried in hexane (1000 mL) and then ®ltered. The solid
was washed twice with hexane (2Â350 mL). The com-
bined ®ltrates were concentrated in vacuo to a thick
slurry. The mixture was diluted with hexane (800 mL)
and ®ltered. The solid was washed with hexane
(250 mL). The combined ®ltrates were washed with
water (3Â400 mL), saturated NaHCO3 (400 mL) and
saturated NaCl (400 mL). The hexane layer was dried
over MgSO4, ®ltered and concentrated in vacuo to 182 g
of a light yellow oil. The oil was distilled (75±78ꢀC/
2 mm Hg) through a 5 cm Vigreux column to give 1
(165 g, 93% yield) as a clear, colorless oil: TLC Rf=0.77
(hexane:EtOAc, 3:1); 1H NMR (CDCl3, 270 MHz) d
3.60 (2H, t, J=5.9 Hz, CH2), 3.52 (2H, t, J=6.4 Hz,
CH2), 1.80 (2H, m, CH2), 1.62 (2H, m, CH2), 0.84 (9H,
s, C(CH3)3), 0.01 (6H, s, Si(CH3)2); 13C NMR (CDCl3,
270 MHz) d 62.2, 45.0, 30.0, 29.3, 25.9, 18.3, 5.4;
CIMS m/z 223 (M+H+, C10H23ClSiO requires 223);
Elem. Anal. calcd for C10H23ClSiO: C, 53.90; H, 10.40.
Found: C, 54.53; H, 10.52; K-F moisture 0.20.
4-tert-Butyldimethylsilyloxy-1-chlorobutane (1). A 2 L
¯ask equipped with a mechanical stirrer, addition fun-
nel, internal temperature probe and argon inlet was
charged with tert-butyldimethylsilyl chloride (124.8 g,
82.78 mmol) and dichloromethane (350 mL). The solu-
tion was cooled to 0ꢀC, and 4-chloro-1-butanol (86.4 g,
79.60 mmol) was added via the addition funnel. The
addition funnel was rinsed twice with dichloromethane
(2Â50 mL), and the rinses were added to the reaction.
N,N-Diisopropylethylamine (153 mL, 87.56 mmol) was
added slowly over 1.75h so that the internal temperature
did not exceed 6ꢀC. After the addition was complete,
the reaction was stirred at 0ꢀC for 1 h. The reaction was
2-Keto-6-hydroxyhexanoic acid, sodium salt (4,5). A 2 L
¯ask equipped with a large magnetic stir bar was
charged with crushed magnesium turnings (15.23 g,
626.5 mmol). The ¯ask was ¯ame-dried under high
vacuum and allowed to cool under argon. In a separate
¯ame-dried 2 L ¯ask, 1 (103.4 g, 464.1 mmol) was dis-
solved in THF (600 mL). A portion of the solution of
1 (ꢁ150 mL) was added via cannula into the reaction
¯ask containing the magnesium. With stirring, 1,2-
dibromoethane (4 mL, 46.4 mmol) was added dropwise
(Caution!! Exothermic up to 45±50ꢀC). With external
Figure 1. Kinetics of reductive amination using beef liver glutamate
dehydrogenase. The reaction conditions are described under Experi-
mental.

2250
R. L. Hanson et al. / Bioorg. Med. Chem. 7 (1999) 2247±2252
Scheme 2. Conversion of racemic 6-hydroxynorleucine to l-6-hydroxynorleucine.
heating to maintain a re¯ux, the remaining solution of 1
was added via cannula over 1 h. The ¯ask containing the
solution of 1 was rinsed with THF (100 mL), and the
rinse was added to the reaction. After the addition was
complete, the reaction was re¯uxed for 2 h. The reaction
was then allowed to cool to room temperature.
allowed to cool back to 0ꢀC. A solution of crude 2
(70.7 g, 201.7 mmol of 2 +63.4 mmol of diethyl oxalate;
see above) in methanol (300 mL) was added dropwise
over 1 h (maximum internal temperature was 2ꢀC). The
¯ask containing the solution of 2 was rinsed with
methanol (50 mL), and the rinse was added to the reac-
tion. After the addition was complete, the reaction was
stirred for 20 min. The pH of the reaction mixture was
adjusted to 7.15 with 2 M KHSO4 (195 mL). The
methanol was removed in vacuo, and the mixture was
transferred into a separatory funnel with a water rinse
(100 mL) of the reaction ¯ask. Water (740 mL) was
added to the mixture, and the resulting solution was
washed with MTBE (2Â425 mL). The aqueous layer
was transferred into a clean 3 L ¯ask and cooled to
<5ꢀC. The pH of the aqueous layer was adjusted to 4
with 2 M KHSO4. EtOAc (565 mL) was added with
stirring, and the pH of the aqueous layer was further
adjusted to 2.45 with 2 M KHSO4 (a total of 176 mL of
2 M KHSO4 was used). The EtOAc layer was separated,
and the aqueous layer was further extracted with EtOAc
(2Â565 mL). The combined EtOAc extracts were dried
over MgSO4, ®ltered and concentrated in vacuo to give
A ¯ame-dried 2 L ¯ask equipped with a large magnetic
stir bar was charged with diethyl oxalate (76 mL,
556.9 mmol) and THF (400 mL). The solution was
cooled to 75ꢀC. The Grignard reagent solution was
added via cannula to the diethyl oxalate solution over
1 h (maximum internal temperature was 65ꢀC). After
the addition was complete, the reaction was kept at
60ꢀC for 2 h. The cold reaction mixture was poured
into a mixture of ice (530 g), 2N HCl (360 mL) and
saturated NaCl (360 mL). The mixture was shaken and
immediately extracted with EtOAc (1Â1000 mL and
3Â500 mL). The combined extracts were washed with
half-saturated NaHCO3 (780 mL) and saturated NaCl
(780 mL). The organic layer was dried over Na2SO4, ®l-
tered and concentrated in vacuo to a light yellow oil.
The oil was concentrated twice from hexane, and the
residual solvents were removed under high vacuum (1 h)
to give a yellow oil (152 g) which by 1H NMR consisted
of 82 wt% 2 (94% corrected crude yield), 3 wt% 1, and
13 wt% diethyl oxalate: TLC Rf=0.43 (hexane:EtOAc,
1
a yellow oil (46 g) which by H NMR consisted of 87
wt% 3 (76% corrected crude yield), 2 wt% 6, and 11
wt% EtOAc: 1H NMR (DMSO-d6, 270 MHz) d 3.55
(2H, t, J=6.2 Hz, CH2), 2.77 (2H, t, J=7.3 Hz, CH2),
1.46 (4H, m, CH2CH2), 0.84 (9H, s, C(CH3)3), 0.0 (6H,
s, Si(CH3)2); 13C NMR (DMSO-d6, 270 MHz) d 197.8,
164.2, 63.6, 61.1, 39.4, 32.8, 27.1, 19.3, 4.04.
5:1); 1H NMR (CDCl3, 270 MHz)
d 4.30 (m,
CH2+diethyl oxalate), 3.58 (2H, t, J=6.2 Hz, CH2),
2.83 (2H, t, J=7.3 Hz, CH2), 1.65 (2H, m, CH2), 1.52
(2H, m, CH2), 1.32 (m, CH3+diethyl oxalate), 0.85
(9H, s, C(CH3)3), 0.0 (6H, s, Si(CH3)2); 13C NMR
(CDCl3, 270 MHz) d 194.8, 161.4, 63.3, 62.8, 39.2, 32.9,
26.1, 19.7, 18.5, 14.2, 5.2.
A 2 L ¯ask equipped with a large magnetic stir bar and
an internal temperature probe was charged with metha-
nol (700 mL) and cooled to 0ꢀC. 2.0 N NaOH (345 mL,
690 mmol) was added slowly, and the solution was
A 1 L ¯ask equipped with a magnetic stir bar, internal
temperature probe, and addition funnel was charged
with crude 3 (45.2 g, 150 mmol of 3 +1.7 mmol of 6; see
above) and CH3CN (450 mL). The solution was cooled

R. L. Hanson et al. / Bioorg. Med. Chem. 7 (1999) 2247±2252
2251
to 2ꢀC. The addition funnel was charged with 0.10 M
KHSO4 (150 mL, 15 mmol). The KHSO4 solution was
added to the reaction over 3±5 min. The maximum
temperature of the reaction during addition was 2.7ꢀC.
The reaction was stirred 10 min more, and then was
warmed to room temperature over 1 h. After stirring for
1 h at room temperature the pH was adjusted to 7.85
with 1 N NaOH (180 mL). The CH3CN was removed in
vacuo, and the resulting heterogeneous mixture was
transferred into a separatory funnel with an MTBE
rinse (150 mL) and two water rinses (2Â50 mL). The
mixture was shaken, and the aqueous layer (now pH
7.25) was separated and extracted again with MTBE
(1Â150 mL). The aqueous layer was transferred to a
clean 1 L ¯ask and placed on a rotary evaporator for
45 min to remove volatile organics. The aqueous solu-
tion was frozen at 78ꢀC and lyophilized for 22 h to
give a white solid (28.5 g) which consisted of 81 wt%
2-keto-6-hydroxyhexanoic acid, sodium salt (91% cor-
rected crude yield; 33 wt% keto form 4 and 48wt%
hemiketal form 5 (1H NMR)), 2.5 wt% 7, 6 wt% 8
L-6-Hydroxynorleucine from ketoacid. NH4OH (2.02 mL
of 14.8 M, 30 mmoles) in 10 mL of water was adjusted
to pH 8.75 with 12.1 M HCl. Glucose (5.404 g,
30mmol), substrate19 (3 g, 13.12 mmol), NAD (20.56mg,
30 mmol), and dithiothreitol (4.62 mg, 30 mmol) were
added, and the volume was brought to 27 mL with
water. Glucose dehydrogenase (2 mg containing
72.9 units/mg) and beef liver glutamate dehydrogenase
(3 mL containing 19 mg protein/mL, 40 units/mg) were
added to start the reaction. The reaction was carried out
at 30ꢀC and the pH was maintained at 8.75 by addition
of 3 M NH4OH using a Brinkmann pH stat. HPLC
analysis indicated that the reaction was complete after
3 h (Fig. 1). After 25 h the reaction was stopped, and the
yield, determined by HPLC analysis, was 11.62 mmol
(88.6%). After the reaction was complete, the mixture
was heated at 100ꢀC for 4±5 min, then centrifuged at
5000 g for 15 min to remove precipitated proteins. The
volume was then reduced by removing water and am-
monium hydroxide at 25±28 psi and 58±68ꢀC. The pH
of the concentrated solution was adjusted to 2 with 6 N
HCl, followed by chromatography over a Dowex-50W
X8-200 column (4Â42 cm). l-6-Hydroxynorleucine was
eluted from the column with 1M NH4OH. The product
was isolated from the combined rich fractions by
removing water and NH4OH. The isolated yield was
80% and the optical purity was >99%. 1H NMR
(D2O) d 1.64 (m, 2 H), 1.82 (m, 2H), 2.05 (m, 2H), 3.82
(t, 2H), 3.91 (t, 1H); 13C NMR (D2O) d 19.0, 27.5,
29.2, 44.6, 56.5, 172.5: ms, m/z: 148 (m+H), 295
(2m+H)
1
(both by H NMR), and 8 wt% NaKSO4 (according to
theory): mp (uncorrected) 139±141ꢀC (decomp.); 1H
NMR (D2O, 270 MHz) d 3.86 (0.6H, m, OCH hemi-
ketal form), 3.68 (0.6H, m, OCH hemiketal form), 3.59
(0.8H, t, OCH2 keto form), 2.75 (0.8H, t, CH2 keto
form), 1.6 (5.2H, complex m); 13C NMR (D2O,
270 MHz) d 209, 179, 172, 98, 64, 63, 41, 34, 33, 26, 21,
20; IR (KBr pellet vmax 3400, 2932, 1734 (weak), 1628,
1
1423 cm
; FAB±MS m/z 167 (M-H, C6H9O4Na
requires 167); K-F moisture 2.30; Elem. Anal. calcd for
C6H9O4Na.0.03 C6H7O3Na.0.04 C12H16O7Na2.0.27
H2O.0.11 NaKSO4: C, 38.55; H, 5.05; Na, 13.49.
Found: C, 38.05; H, 4.90; Na, 13.70.
L-6-Hydroxynorleucine from racemic 6-hydroxynor-
leucine via porcine kidney oxidase. Racemic 6-hydroxy-
norleucine (0.5 g) in 20 mL 50 mM potassium phosphate
bu€er (pH 7) was treated with porcine kidney d-amino
acid oxidase (49 u, 350 mg) and beef liver catalase (182
u, 8 mg) to give a mixture of ketoacid and 0.236 g l-6-
hydroxynorleucine with 97% o.p. The enzymes were
added in increments and the reaction took 10 days to
complete. The reductive amination was then carried out
as described above. 0.484 g of 6-hydroxynorleucine was
produced (97% yield) with >99% o.p. by HPLC assay.
L-6-Hydroxynorleucine from racemic 6-hydroxynor-
leucine via Trigonopsis oxidase. Trigonopsis variabilis
ATCC10679 was grown in a 250 L fermentor at 28ꢀC on
the medium containing (g/L): KH2PO4, 5; MgSO4.
7H2O, 1; CaCl2, 0.5; H3BO3, 0.1; (NH4)2MoO4, 0.04;
MnSO4.4H20, 0.04; ZnSO4.7H2O; 0.04; CuSO4.5H2O;
0.045; FeSO4.7H2O, 0.025; dl-methionine, 3; biotin,
0.02; thiamine, 0.1; cysteine, 0.72; cerelose hydrate, 22.
Cell paste (129 g) was harvested 50 h after inoculation
and stored frozen. Racemic 6-hydroxynorleucine
(0.514 g) in 70 mL 50 mM potassium phosphate bu€er,
pH 7, was shaken with 14 g washed Trigonopsis var-
iabilis ATCC 10679 cells for 21 h at 28ꢀC and 200 rpm
to give l-6-hydroxynorleucine with o.p. >99%. Cells
were removed by centrifugation, and the supernatant
was treated with glutamate dehydrogenase, glucose
dehydrogenase, NH3, glucose and NAD as described
above. l-6-Hydroxynorleucine (0.469 g, 91% yield) was
obtained with >99% o.p. After the completion of the
6-Hydroxynorleucine from 5-(4-hydroxybutyl)hydantoin.
In a 400 mL Parr pressure reactor was placed (34.4 g,
0.2 mol) of 5-(4-hydroxybutyl)hydantoin, 80 mL of DI
water, (14.8 g, 0.2 mol) of calcium hydroxide and (8 g,
0.2 mol) of sodium hydroxide. The reaction mixture was
sealed, heated to 140ꢀC and stirred for 5 h. After cooling
to room temperature, the salt was ®ltered and the cake
was washed with 2Â35 mL of deionized water. To the
®ltrate was added 20 g of ammonium carbonate. The
mixture was heated to re¯ux for 10 min, cooled to room
temperature and ®ltered. The cake was washed with
20 mL of water. The pH of the ®ltrate was adjusted
from 9.8 to 5.9 using acetic acid. Water was then
removed on rotary evaporator to give about 70 g wet
solid. 200 mL of absolute ethanol was added. The slurry
was stirred for 30 min and ®ltered. The white cake was
washed with 2Â50 mL absolute ethanol and dried to
a€ord 18.1 g (56%) of racemic 6-hydroxynorleucine.

2252
R. L. Hanson et al. / Bioorg. Med. Chem. 7 (1999) 2247±2252
reaction, l-6-hydroxynorleucine was isolated as descri-
bed above. The isolated yield was 0.394 g (76.6%) and
the optical purity was 99%. The H NMR spectrum
agreed with the structure.
8. Dreyfuss, P. J. Med. Chem. 1974, 17, 252.
9. Kern, B. A.; Reitz, R. H. Agric. Biol. Chem. 1978, 42, 1275.
10. Kaufman, S.; Mason, K. J. Biol Chem. 1982, 257, 14667.
11. Belyaev, A. A. Tetrahedron Lett. 1995, 36, 439.
12. Bommarius, A. S. In Enzyme Catalysis in Organic Synth-
esis; Drauz, K.; Waldmann, H., Eds.; VCH: Weinheim, 1995;
Vol II, pp 633±641 and cited references.
1
References and Notes
13. Galkin, A.; Kulakova, L.; Yoshimura, T.; Soda, K.; Esaki,
N. Appl. Environ. Microbiol. 1997, 63, 4651.
1. Robl, J. A.; Sun, C.; Stevenson, J.; Ryono, D. E.; Simpkins,
L. M.; Cimarusti, M. P.; Dejneka, T.; Slusarchyk, W. A.;
Chao, S.; Stratton, L.; Misra, R. N.; Bednarz, M. S.; Asaad,
M. M.; Cheung, H. S.; Aboa-O€ei, B. E.; Smith, P. L.; Mathers,
P. D.; Fox, M.; Schae€er, T. R.; Seymour, A. A.; Trippodo,
N. C. J. Med. Chem. 1997, 40, 1570.
14. Hanson, R. L.; Singh, J.; Kissick, T. P.; Patel, R. N.;
Szarka, L. J.; Mueller, R. H. Bioorganic Chemistry 1990, 18,
116.
15. Gaudry, R. Can. J. Res. (B) 1948, 26, 387.
16. Oshima, T.; Takada, H.; Yoshimura, T.; Esaki, N.; Soda,
K. J. Bacteriol. 1991, 173, 3943.
2. Robl, J. A.; Cimarusti, M. P. Tetrahedron Lett. 1994, 35,
1393.
3. Maurer, P. J.; Miller, M. J. J. Org. Chem. 1981, 46, 2835.
4. Maurer, P. J.; Miller, M. J. J. Am. Chem. Soc. 1982, 104,
3096.
17. Oshima, T.; Nishida, N.; Bakthavatsalam, S.; Kataoka,
K.; Takada, H.; Yoshimura, T.; Esaki, N.; Soda, K. Eur. J.
Biochem. 1994, 222, 305.
18. Wei, C.; Huang, J.; Tsai, Y. Biotechnol. Bioeng. 1988, 34,
570.
5. Maurer, P. J.; Miller, M. J. J. Am. Chem. Soc. 1983, 105,
240.
6. Culvenor, C. C. J.; Foster, M. C.; Hegarty, M. P. Aust. J.
Chem. 1971, 24, 371.
7. Bodanszky, M.; Martinez, J.; Priestly, G. P.; Gardner, J. D.;
Mutt, V. J. Med. Chem. 1978, 21, 1030.
19. This material is a mixture containing 6.6 wt% 3,4-dihydro-
2H-pyran-6-carboxylic acid, sodium salt, 48.4 wt% 2-hydroxy-
tetrahydropyran-2-carboxylic acid, sodium salt, 25.1 wt% 2-
keto-6-hydroxyhexanoic acid, sodium salt, 14 wt% NaCl, and
6 wt% NaHCO3. The actual substrates for the enzyme reaction
are underlined and comprise 73.5% of this mixture.