Asymmetric acyloin condensation catalyzed by phenylpyruvate decarboxylase

Article abstract of DOI:10.1016/S0957-4166(99)00548-0

Cells obtained from growth of Achromobacter eurydice, Pseudomonas aromatica and Pseudomonas putida on L-phenylalanine containing medium catalyzed the enzymatic acyloin condensation of phenylpyruvic acid 1 and acetaldehyde 2 by phenylpyruvate decarboxylase

Full text of DOI:10.1016/S0957-4166(99)00548-0

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 10 (1999) 4667–4675
Asymmetric acyloin condensation catalyzed by phenylpyruvate
decarboxylase
∗
Zhiwei Guo, Animesh Goswami, K. David Mirfakhrae and Ramesh N. Patel
Enzyme Technology, Process Research, Process Research & Development, Bristol-Myers Squibb Pharmaceutical Research
Institute, New Brunswick, NJ 08903, USA
Received 5 October 1999; accepted 12 November 1999
Abstract
Cells obtained from growth of Achromobacter eurydice, Pseudomonas aromatica and Pseudomonas putida
on L-phenylalanine containing medium catalyzed the enzymatic acyloin condensation of phenylpyruvic acid 1
and acetaldehyde 2 by phenylpyruvate decarboxylase to produce 3-hydroxy-1-phenyl-2-butanone 3. The acyloin
condensation by Achromobacter eurydice and P. aromatica was stereoselective, providing the 3R enantiomer 3a
with enantiomeric excess (ee) of 95% and 84%, respectively. A partially purified enzyme was prepared from the
cell free extract of Achromobacter eurydice. The acyloin product 3a was obtained in 45% yield with an ee of 91%
by using this partially purified preparation of phenylpyruvate decarboxylase. © 2000 Elsevier Science Ltd. All
rights reserved.
1. Introduction
Chiral unsymmetrically substituted acyloins (α-hydroxyketones) are important classes of intermedi-
ates in organic synthesis due to their bifunctional aspect, especially having one chiral center amenable
to further modification. Enzyme mediated acyloin formation could provide an advantageous, environ-
1,2
mentally friendly method to prepare chiral acyloins. Acyloin formations mediated by yeast pyruvate
3
–5
6–9
decarboxylase (EC 4.1.1.1)
and bacterial benzoylformate decarboxylase (EC 4.1.1.7)
have been
reported. Though phenylpyruvate decarboxylase (EC 4.1.1.43)1
0,11
for decarboxylation of phenylpyruvic
acid has been known for a long time, there is no report in the literature on the acyloin condensation
catalyzed by phenylpyruvate decarboxylase. The present work describes the first report of asymmetric
acyloin condensation catalyzed by phenylpyruvate decarboxylase.
∗
Corresponding author. E-mail: patelr@bms.com
0
957-4166/99/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(99)00548-0
tetasy 3183

4
668
Z. Guo et al. / Tetrahedron: Asymmetry 10 (1999) 4667–4675
2. Results and discussion
Several Pseudomonas species and Achromobacter eurydice were grown in medium containing L-
phenylalanine to induce the phenylpyruvate decarboxylase enzyme. The cell mass production was higher
when the cells were grown in medium containing yeast extract. The cells were used to catalyze the
acyloin condensation between phenylpyruvic acid 1 and acetaldehyde 2 to provide the acyloin product
3-hydroxy-1-phenyl-2-butanone 3 (Fig. 1).
Fig. 1. Acyloin condensation with phenylpyruvate decarboxylase
Authentic racemic product 3 was prepared using known12 chemical synthetic schemes. Analytical
methods were developed for identification of product 3. Analytical HPLC methods were also developed
to separate the two enantiomers of the acyloin product 3.
Acyloin condensation was carried out with the cell suspensions of various microorganisms from the
genus Pseudomonas and Achromobacter. The results are shown in Table 1. There was no acyloin product
formed in the absence of cells. In the presence of cells, acyloin products were identified. The yield of
the acyloin product was calculated by comparing the HPLC area with that of a standard curve made by
using the chemically synthesized 3. Product 3 was obtained in 1–7% yield during the initial screening
studies. There are two probable reasons for the low yield. Phenylpyruvic acid 1 decomposes under
the reaction conditions and the amounts of actual phenylpyruvate decarboxylase enzyme in cells used
during the screening experiments were low. Sodium salt of phenylpyruvic acid was found to have better
stability under the reaction conditions. The yield of the acyloin product was improved to 45% using
higher concentration of partially purified enzyme and the sodium salt of phenylpyruvic acid as described
later.
The acyloin product from Achromobacter eurydice SC 16386 was isolated. The structure of product 3
was established by comparison of its NMR and mass spectral data with that of the authentic 3 synthesized
by the chemical route. The ee of this sample was determined to be 95% by HPLC. The retention times
of the major (97.5%) and minor (2.5%) enantiomers were 42.5 min and 27.7 min (HPLC method 4),
respectively. The racemic authentic sample of 3 prepared by chemical synthesis showed two peaks at
27.7 and 42.5 min of 1:1 ratio when analyzed by the same HPLC method. The enantiomeric analysis of
product 3 from Pseudomonas aromatica SC 16387 showed the same major enantiomer as that obtained
with Achromobacter eurydice with an ee of 84%. Thus, the acyloin products were formed with high
enantioselectivity in both cases. The amounts of acyloin products in other cases were too small to make
definitive comments about their enantiomeric purity at this point.
The absolute configuration of the acyloin product was established by isolating the product from the
enzymatic reaction and subsequent chemical reduction to a diol of known stereochemistry. It was reported
that α-hydroxyketones could be reduced by Zn(BH ) to erythro diols as the major and threo diols as
4
2
1
3
the minor products. The chemically synthesized (±)-3 was reduced by Zn(BH ) to the erythro-(±)-4
4
2

Z. Guo et al. / Tetrahedron: Asymmetry 10 (1999) 4667–4675
4669
Table 1
Initial screening of microorganisms for acyloin condensation by phenylpyruvate decarboxylase
as the major (70%) and threo-(±)-5 as the minor (30%) products. All compounds were characterized by
NMR. An HPLC method (method 5) was also developed for the analysis of 4a, 4b, 5a, and 5b.
Achromobacter eurydice SC16386 cells were grown in a large scale fermentor with yeast extract
medium to provide increased cell mass. These cells and modified reaction conditions using phenylpy-
ruvic acid sodium salt were employed for the enzymatic acyloin condensation, and a reaction product
consisting of two compounds was isolated in 18% total yield. These were separated to provide the
acyloin product 3a (10.7% isolated yield, 88% ee by HPLC) and the diol 5a (7.2% isolated yield,
1
diastereomerically and enantiomerically pure by HPLC and H NMR). The diol 5a was probably formed
by reduction of the initially formed acyloin product 3a by the oxidoreductase(s) present in Achromobacter
eurydice SC16386 cells. The enzymatic acyloin product 3a with 88% ee was reduced by Zn(BH ) to the
4
2
erythro-4a and threo-5a diols as the major (70%) and minor (30%) products, respectively. The absolute
configurations of the erythro-4a and threo-5a diols were determined to be 2S,3R and 2R,3R, respectively,
14
by comparing the specific rotation data to those of the literature values. Thus, the absolute configuration
of the acyloin product 3a was determined as 3R.
Cell extracts of Achromobacter eurydice were prepared, and the phenylpyruvate decarboxylase was
partially purified from the cell free extract as described in the Experimental. This partially purified
phenylpyruvate decarboxylase was used for the acyloin condensation to provide the acyloin product
3a and phenylacetic acid as by-product. The latter probably formed by oxidation of phenylacetaldehyde
obtained by decarboxylation of phenylpyruvate. The acyloin product 3a was obtained in 45% yield with
an ee of 91%. Further purification of the enzyme is in progress.
The acyloin condensation catalyzed by phenylpyruvate decarboxylase enzyme probably proce-
eds the same way as pyruvate decarboxylase and benzoylformate decarboxylase catalyzed acyloin
6
condensations involving enzyme catalyzed decarboxylation of phenylpyruvic acid (PhCH COCO H)
2
2
to the thiamine–enzyme bound complex PhCH CH(OH)–TPP–enzyme, followed by its reaction with the
2
aldehyde (RCHO) to provide the acyloin product PhCH COCH(OH)R. The products of phenylpyruvate
2
16
catalyzed acyloin condensation can be used for the synthesis of biologically active acyloins and various
1
7
substituted carbohydrates and carbohydrate mimetics. The present chemical methods of syntheses of
1
2,18
acyloins
generally give racemic products. The enzyme catalyzed acyloin condensation, such as the
one described here, can provide stereoselective products and are of great value for asymmetric synthesis.

4
670
Z. Guo et al. / Tetrahedron: Asymmetry 10 (1999) 4667–4675
3. Experimental
3.1. Chemicals
Chemicals were purchased from VWR and/or Aldrich.
.2. Microorganisms
3
Microorganisms were obtained from our culture collection. Some microorganisms were obtained from
ATCC. The SC number denotes the number in our culture collection.
3.3. Analytical methods
NMR spectra were recorded in CDCl using a Bruker AC-300 NMR spectrometer operating at 300
3
1
13
1
MHz for H and 75 MHz for C or a Varian 500 MHz NMR spectrometer operating at 500 MHz for H
and 125 MHz for C. LC–MS analysis was performed by BMS analytical department.
1
3
3.4. HPLC
Method 1 was performed on a reversed phase phenylhexyl column (5 µm, 15 cm×0.46 cm, Luna) at
5
2
0°C using 0.1% H PO and CH CN (75:25) as eluent at a flow rate of 1 mL/min and UV detection at
3
4
3
10 nm. Retention times for compounds 1 and 3 were 11.7 min and 10.9 min, respectively.
Method 2 was performed on a reversed phase C-18 column (5 µm, 15 cm×0.21 cm, Advantage 300)
at 50°C with a flow rate of 0.5 mL/min using a gradient elution as follows: 98% solvent A (0.1%
trifluoroacetic acid in water) and 2% solvent B (0.1% trifluoroacetic acid in CH CN) for the first 5 min,
3
then increasing to 85% solvent A and 15% solvent B in 15 min. UV detection was at 210 nm. Retention
times for compounds 1 and 3 were 10.7 min and 14 min, respectively.
Method 3 was performed on a reversed phase C-18 column (5 µm, 15 cm×0.46 cm, Kromasil) at
4
2
0°C using 0.1% H PO and CH CN (75:25) as eluent at a flow rate of 1.0 mL/min and UV detection at
3 4 3
10 nm. Retention times for compounds 1, phenyl acetic acid and 3 were 5.0 min, 6.7 min and 7.5 min,
respectively.
Method 4 for separation of stereoisomers was performed on a Chiralpak AD column (25 cm×0.46
cm, Daicel) at ambient temperature with a mixture of hexane:isopropanol:ethanol:trifluoroacetic acid
(98.5:0.8:0.6:0.1) as eluent at a flow rate of 1.5 mL/min and UV detection at 210 nm. Retention times for
the two isomers of compound 3 were 27.7 min and 42.5 min.
Method 5 for separation of stereoisomers was the same as method 4 except the eluent was a mixture
of hexane:isopropanol:ethanol (95.4:4.0:0.6) and the flow rate was 1.0 mL/min.
HPLC methods 1–3 were used for identification and quantitation of different components while
methods 4 and 5 were used for enantiomeric excess (ee) analyses.
3
.5. Chemical synthesis of 3-hydroxy-1-phenyl-2-butanone 3
To a stirred solution of trimethylsilyl cyanide (20 mmol, 1984 mg, 2667 µL) in 80 mL CH Cl at 0°C,
2
2
acetaldehyde (20 mmol, 880 mg, 1117 µL) was added followed by TEA (2 mmole, 202 mg, 278 µL).
The resulting mixture was stirred at 0°C for 2 h. The solvent was removed under reduced pressure. The
residue was dissolved in 20 mL anhydrous ether and added to a stirred solution of benzylmagnesium

Z. Guo et al. / Tetrahedron: Asymmetry 10 (1999) 4667–4675
4671
chloride (30 mL, 1.0 M in ether) at 24°C. The reaction mixture was refluxed for 2 h, cooled to room
temperature. Water (20 mL) was added with caution followed by 20 mL of 4 M HCl under vigorous
stirring. The progress of the reaction was followed by TLC. The hydrolysis was complete after 1 h. The
product was extracted with ethyl acetate (2×40 mL). The extract was washed successively with water,
5
% sodium bicarbonate, water, 1 M HCl, water, brine, dried over MgSO , and concentrated to dryness.
4
Flash chromatography on silica and elution with CH Cl gave 2.48 g of a colorless oily product which
2
2
still contained some by-product benzyl alcohol as impurity. The pure 3 (390 mg) was obtained by a
second flash chromatography on silica and elution with hexane:MTBE (4:1 to 2:1), followed by a third
flash chromatography on silica and elution with CH Cl :MTBE (9:1). The pure 3 was homogeneous by
2
2
1
TLC and HPLC, H NMR: δ 1.44 (d, J=7.4 Hz, 3H, CH ), 3.42 (1H, OH, D O exchangeable), 3.78 (d,
J=15.5 Hz, 1H, CH -B), 3.84 (d, J=15.5 Hz, 1H, CH -A), 4.38 (m before D O exchange, q after D O
exchange, J=7.4 Hz, 1H, CH), 7.28–7.4 (m, 5H); C NMR δ 19.66 (CH ), 44.45 (CH ), 72.15 (CHOH),
27.2, 128.7 (2C), 129.3 (2C), 133.0, 210.0 (CO). MS m/z 165 (M+H).
3
2
2
2
2
2
13
3
2
1
3.6. Chemical reduction of (±)-3 by Zn(BH4)2
The ether solution of Zn(BH ) (160 mL) was prepared from ZnCl (30 mmol) and NaBH (2.7 g)
4
2
2
4
1
5
as described in the literature. To a stirred solution of (±)-3 (377 mg in 2.3 mL ether) at 0°C, 5 mL
of the above Zn(BH ) solution was added. The resulting mixture was stirred at 0°C and the progress
4
2
of reaction was monitored by TLC. After 2 h, additional 3 mL of Zn(BH ) was added and the reaction
4
2
was continued for one more hour to near completion. Water (1 mL) was added dropwise followed by
3
0 mL of ethyl acetate. The organic phase was washed successively with 1 M HCl (2×5 mL), water (5
mL), brine (5 mL), dried over MgSO , and concentrated to dryness. Flash chromatography on silica and
elution with hexane:ethyl acetate (1:1) gave 265 mg of 1-phenyl-2,3-butanediol. The H NMR spectrum
4
1
of the product showed that it was a mixture of the erythro-4 (70%) and threo-5 (30%) by comparison
1
4
with literature data. TLC and HPLC (method 2, with retention time 12.0 min for 4 and 13.6 min for 5,
respectively) showed the same 7:3 ratio of the two diastereoisomers. This mixture was subjected to flash
chromatography and eluted with CH Cl :MTBE (4:1). The fast moving portion (fraction 29–34, 8 mL
2
2
1
per fraction) gave 5 as a colorless oil, 60 mg, HPLC method 2: 95% 5 and 5% 4, H NMR (CDCl –2%
3
D O): δ 1.28 (d, 3H), 2.78 (dd, 1H), 2.90 (dd, 1H), 3.60 (m, 1H), 3.70 (m, 1H), 7.2–7.4 (m, 5H), which
2
were in agreement with the literature data for threo isomers. This oily sample (±)-5 showed two major
peaks at 19.2 and 22.1 min of 1:1 ratio when analyzed by the HPLC method 5. The slow moving portion
1
(
1
fraction 39–46) gave a white solid, 120 mg, HPLC method 2: >95% 4, H NMR (CDCl –2% D O): δ
3
2
.26 (d, 3H), 2.72 (dd, 1H), 2.85 (dd, 1H), 3.8–3.95 (m, 2H), 7.2–7.4 (m, 5H), which were in agreement
with the literature data for erythro isomers. This solid sample (±)-4 showed two major peaks at 21.3 and
26.0 min of 1:1 ratio when analyzed by the HPLC method 5.
3.7. Growth of microorganisms on L-phenylalanine agar plate
The composition of medium for agar plate was as follows: L-phenylalanine 3.3 g, (NH ) SO 1 g,
4
2
4
agar noble 3 g, 5 mL H33A solution (made by dissolving 100 g each of K HPO and KH PO in
2
4
2
4
water to a volume of 1 L), 5 mL H33B solution (made by dissolving 40 g MgSO ·7H O, 2 g NaCl,
4
2
2
g FeSO ·7H O, 2 g MnSO ·4H O in water to a volume of 1 L), dissolved in water by heating and
4
2
4
2
brought to a volume of 1 L. The pH was adjusted to 7.0. The media was autoclaved at 121°C for 15 min.
Twenty cultures including nine bacteria and 11 yeasts were inoculated on agar plates. Six cultures, one
Pseudomonas sp. and five Pseudomonas putida, grew well on the plate after six days.

4
672
Z. Guo et al. / Tetrahedron: Asymmetry 10 (1999) 4667–4675
3.8. Growth of microorganisms in medium containing L-phenylalanine as the sole carbon source
The first liquid growth stage was in 20 mM L-phenylalanine medium made by dissolving 3.3 g L-
phenylalanine, 2 g KH PO , 1 g (NH ) SO , 0.2 g MgSO in water to a volume of 1 L. The pH was
2
4
4 2
4
4
adjusted to 7.0 and the medium was autoclaved at 121°C for 15 min. Microorganisms grown on the L-
phenylalanine agar plate above were inoculated in 100 mL of L-phenylalanine medium and incubated at
28°C and 200 rpm for four days.
This first stage grown culture (5 mL) was transferred to 100 mL of sterile 60 mM L-phenylalanine
medium made in the same way as above except containing 9.9 g of L-phenylalanine and grown at 28°C
and 200 rpm for four days. The cells and broth were separated by centrifugation at 15,000 g for 25 min
and stored at −70°C before use.
3.9. Growth of microorganisms in medium containing L-phenylalanine and yeast extract
The L-phenylalanine yeast extract medium was made by dissolving 3 g L-phenylalanine, 3 g yeast
extract, 0.9 g L-glutamic acid, 4 g KH PO , 2 g (NH ) SO , 0.2 g MgSO in water to a volume of 1
2
4
4 2
4
4
L. The pH was adjusted to 7.0 and the medium was autoclaved at 121°C for 15 min. Microbial cultures
from thawed frozen vials were inoculated into 100 mL of the sterile medium and grown at 28°C, 200
rpm for two or three days. For the first transfer, the grown culture (5 mL) was transferred to 100 mL of
the sterile medium and grown in the same way. Each successive transfer was done in the same way. After
three transfers, the cells and broth were separated by centrifugation at 15,000 g for 25 min and stored at
−70°C before use.
3.10. Acyloin condensation mediated by microbes grown in medium containing L-phenylalanine as the
sole carbon source
In a 50 mL Erlenmeyer flask, cells (0.4 g wet weight) were suspended in 10 mL 0.1 M sodium
phosphate buffer pH 7.0. MgCl ·6H O (6 mg, final concentration 3 mM), thiamine pyrophosphate
2
2
chloride (TPP, 1 mg, final concentration 0.2 mM) and phenylpyruvic acid (33 mg, final concentration
20 mM) were added. The mixture was incubated for 10 min at 200 rpm at 28°C. Acetaldehyde (56 µL,
final concentration 100 mM) was added and the biotransformation was carried out at 200 rpm and 28°C.
Samples (2 mL each) were taken at 2, 4, 6, and 24 h. To each sample 1 mL 5% citric acid (pH adjusted to
3.5) was added and extracted with 3 mL ethyl acetate. The organic extract was washed with 2 mL brine
and concentrated by evaporation under a stream of nitrogen.
The reaction product was dissolved in 1 mL CH CN, filtered and analyzed by HPLC (method 1) along
3
with an authentic sample of 3 as a standard (retention time 10.9 min). No acyloin product was formed in
control samples containing no cells.
3.11. Acyloin condensation mediated by microbes grown in medium containing L-phenylalanine and
yeast extract
Cells (0.5–1.0 g) were suspended in 5 mL of 0.1 M sodium phosphate buffer pH 7.0 and placed in a
5
0 mL Erlenmeyer flask. To the mixture, MgCl ·6H O (3 mg, final concentration 3 mM), cocarboxylase
2
2
(TPP, 2.3 mg, final concentration 1 mM) was added followed by a solution of phenylpyruvic acid (8.2
mg in 50 µL ethanol, final concentration 10 mM) under vigorous stirring. One control reaction was set
up without cells. Acetaldehyde was added in three portions: 10 µL after the mixture was incubated for

Z. Guo et al. / Tetrahedron: Asymmetry 10 (1999) 4667–4675
4673
10 min, 45 µL at 4 h and 45 µL at 20 h. The biotransformation was carried out at 200 rpm 28°C for 24
h. The mixture was acidified with 1 mL 1 M HCl and extracted with ethyl acetate (2×4 mL). The ethyl
acetate extract was washed with 5% sodium bicarbonate and water. Removal of ethyl acetate gave the
reaction product. The reaction product was dissolved in CH CN and analyzed by HPLC methods 2 and
3
4.
3.12. Reaction product from Achromobacter eurydice SC 16386 biotransformation
The product obtained from Achromobacter eurydice SC 16386 biotransformation was analyzed by
1
1
HPLC, mass spectrometry and one and two dimensional H NMR. MS m/z 165 (M+H). H NMR: δ
1
.44 (d, J=7.4 Hz, 3H, CH ), 3.42 (bs, 1H, OH), 3.78 (d, J=15.5 Hz, 1H, CH -B), 3.84 (d, J=15.5 Hz,
3
2
1
H, CH -A), 4.38 (m, 1H, CH), 7.2–7.38 (m, 5H).
2
3
.13. Growth of Achromobacter eurydice SC16386 in 100 L medium in fermentor
The medium (per L) contained 3 g L-phenylalanine, 20 g yeast extract, 5 g L-glutamic acid, 4 g
KH PO , 2 g (NH ) SO , and 0.2 g MgSO . The microbial culture of Achromobacter eurydice SC16386
2
4
4 2
4
4
from a thawed frozen vial was inoculated into 100 mL of the sterile medium and grown at 28°C and
00 rpm for two days. This grown culture was transferred (5% inoculum) into fresh sterile medium and
2
grown in the same way for three successive transfers. The scale of each successive transfer was increased
to make 5 L inoculum for final transfer to 100 L sterilized medium in a large fermentor. The fermentor
was also maintained at 28°C for 48 h. The cells were separated from broth by centrifugation at 15,000 g
for 25 min. The wet cells (1076 g) were stored at −70°C before use.
3.14. Isolation of acyloin product 3a and diol 5a from the acyloin condensation mediated by Achromo-
bacter eurydice SC16386 cells grown in fermentor
Achromobacter eurydice SC16386 cells (300 g) were suspended in 200 mL water and 500 mL sodium
phosphate buffer (0.2 M, pH 7.0) in a 2 L glass reactor maintained at 28°C and equipped with a
mechanical stirrer and a condenser (4°C). MgCl ·6H O (600 mg) and TPP (460 mg) were added to
2
2
the reactor. Sodium phenylpyruvate monohydrate was added in two batches (2.04 g each at 0 and 24 h).
Acetaldehyde was added in six batches (2.8 mL each at 0, 4, 8, 24, 28, and 32 h). After 48 h, the reaction
mixture was acidified with 50 mL of 4 M HCl and extracted with ethyl acetate (900 mL). The organic
layer was washed with 5% sodium bicarbonate (2×100 mL), water (100 mL), 1 M HCl (100 mL) and
brine (100 mL), dried over MgSO (15 g) and charcoal (2 g), filtered and concentrated to dryness under
4
reduced pressure. Flash chromatography of the residue on silica with CH Cl :MTBE (4:0 to 4:1) eluent
2
2
gave 510 mg crude product 3a, and a more polar by-product threo-diol 5a (240 mg, 7.2% yield).
The crude product 3a was purified on silica eluted with hexane:MTBE (4:1) to afford pure 3a (350 mg,
2
4
1
0.7%), [α]D –90.5 (c 1.54, CHCl ) and –23.6 (c 1.56, MeOH). The ee of this product was determined
3
as 88% by HPLC (method 5). The retention time of the major (94%) and minor (6%) enantiomers were
8.4 and 14.8 min, respectively. The racemic authentic sample of 3 showed two peaks at 14.8 and 18.4
min of 1:1 ratio when analyzed by the same HPLC method.
1
2
4
14
24
1
The threo-diol 5a: [α]D +33.7 (c 1.02, CHCl ) (lit. data [α]
+35.53 (c 1.162, CHCl )), H NMR
3
3
D
(
5
CDCl –2% D O): δ 1.28 (d, 3H), 2.78 (dd, 1H), 2.90 (dd, 1H), 3.62 (m, 1H), 3.70 (m, 1H), 7.2–7.4 (m,
H). The threo-diol 5a showed one single peak only on HPLC (method 5) at 19.2 min.
3
2

4
674
Z. Guo et al. / Tetrahedron: Asymmetry 10 (1999) 4667–4675
3.15. Reduction of acyloin condensation product 3a to diols 4a and 5a by Zn(BH4)2
The acyloin product 3a (164 mg, ee 88%) was reduced with 5 mL of the same Zn(BH ) solution
4
2
as described previously at 0°C for 30 min. The crude product mixture of 4:5 was obtained in about 7:3
ratio. The mixture was separated by flash chromatography on silica, and elution with CH Cl :MTBE
2
2
(
5:1) gave the fast moving diol 5a (42 mg, colorless liquid) and the slow moving diol 4a (50 mg, white
solid).
The fast moving diol 5a: one major peak at 19.2 min by HPLC method 5, [α]
24
+20.1 (c 1.50, CHCl3)
D
1
4
24
(
lit. data for 5a 2R,3R [α]D +35.53 (c 1.162, CHCl )).
3
24
The slow moving diol 4a: one major peak at 26.0 min by HPLC method 5, [α]
–33.6 (c 1.74,
D
1
4
25
CHCl ) (lit. data for 4a 2S,3R [α]D –41.94 (c 1.130, CHCl )).
3
3
3.16. Partial purification of phenylpyruvate decarboxylase enzyme from Achromobacter eurydice
SC16386 cells
Achromobacter eurydice SC16386 cells (100 g) were suspended in 500 mL buffer A (50 mM potassium
phosphate, 2 mM MgSO , 1 mM DTT, 0.1 mM TPP, pH 6.8) and homogenized by three passages through
4
a microfluidizer (Model 110F, Microfluidics Co., Newton, Massachusetts). The resulting suspension was
centrifuged at 15,000 g for 30 min. The supernatant was applied on a DE52 cellulose column (14×5.0
cm) previously equilibrated with buffer A. The column was washed with 450 mL buffer A followed by a
gradient elution from 100% buffer A (300 mL) to 100% buffer B (0.8 M NaCl in buffer A, 300 mL). The
protein concentration and enzyme activity of different fractions were assayed. The major activity was
found in 100 mL of the eluate.
3.17. Acyloin condensation by the partially purified phenylpyruvate decarboxylase enzyme from Achro-
mobacter eurydice SC16386
A portion of the active fraction obtained from the DE-52 cellulose column chromatography was used
for this experiment. Acetaldehyde (11 µL) was added to a mixture of the active fraction (500 µL), water
(
289 µL), potassium phosphate buffer (100 µL, 500 mM, pH 6.8), MgCl (30 µL, 100mM), TPP (20
2
µL, 50 mM), and sodium phenylpyruvate (50 µL, 100 mM) in a screw-capped 4 mL glass vial. After
stirring at 24°C for 20 h, the reaction was terminated by addition of aqueous HCl (50 µL, 1 M) and
MeOH (950 µL). The resulting mixture was filtrated and subjected to HPLC method 3 to determine the
amounts of various components. In order to remove the phenylacetic acid, the aqueuous reaction mixture
was extracted with ethyl acetate (2×3 mL). The ethyl acetate extract was washed with 5% NaHCO (2×2
3
mL) and brine. After removal of solvent, the residue was analyzed by HPLC. HPLC method 3 showed
only one major peak of acyloin 3 at 7.5 min, and the ee value of 91% was determined for the major
enantiomer 3a by HPLC method 5.
Acknowledgements
We are indebted to Mr. Steve Schwarz and Mr. David Brzozowski for their help in growing the
microorganisms, to Dr. Venkata Nanduri for his help in the partial purification of the enzyme and to
the BMS analytical department for providing the NMR and mass spectral analyses.

Z. Guo et al. / Tetrahedron: Asymmetry 10 (1999) 4667–4675
4675
References
1
2
3
4
5
6
7
8
9
. Csuk, R.; Glanzer, B. I. Chem. Rev. 1991, 91, 49–97.
. Servi, S. Synthesis 1990, 1–25.
. Juni, E. J. Biol. Chem. 1961, 236, 2302–2308.
. Crout, D. H. G.; Dalton, H.; Hutchinson, D. W.; Miyagoshi, M. J. Chem. Soc. Perkin Trans. 1991, 1329–1334.
. Rogers, P. L.; Shin, H. S.; Wang, B. Adv. Biochem. Eng. Biotechnol. 1997, 56, 33–59.
. Prosen, E.; Ward, O. P.; Collins, S.; Dewdney, N. J.; Hong, Y.; Wilcocks, R. Biocatalysis 1993, 8, 21–29.
. Hegeman, G. D. J. Bacteriol. 1966, 91, 1140–1160.
. Weiss, P. M.; Garcia, G. A.; Kenyon, G. L.; Cleland, W. W.; Cook, P. F. Biochemistry 1988, 27, 2197–2205.
. Tsou, A. Y.; Ransom, S. C.; Gerlt, J. A.; Buechter, D. D.; Babbitt, P. C.; Kenyon, G. L. Biochemistry 1990, 29, 9856–9862.
1
1
0. Asakawa, T.; Wada, H.; Yamano, T. Biochim. Biophys. Acta 1968, 170, 375–391.
1. Heider, J.; Boll, M.; Breese, K.; Breinig, S.; Ebenau-Jehle, C.; Feil, U.; Gadon, N.; Laempe, D; Leuthner, B.; Mohamed,
M. E.; Schneider, S.; Burchhardt, G.; Fuchs, G. Arch. Microbiol. 1998, 170, 120–131.
1
1
1
1
1
2. Gill, M.; Kiefel, M. J.; Lally, D. A. Tetrahedron Lett. 1986, 27, 1933–1934.
3. Nakata, T.; Tanaka, T.; Oishi, T. Tetrahedron Lett. 1983, 24, 2653–2656.
4. Awano, K.; Yanai, T.; Watanabe, I.; Takagi, Y.; Kitahara, T.; Mori, K. Biosci., Biotechnol., Biochem. 1995, 59, 1251–1254.
5. Gensler, W. J.; Johnson, F.; Sloan, A. D. B. J. Am. Chem. Soc. 1960, 82, 6074–6081.
6. Uchida, R.; Shiomi, K.; Sunazuka, T.; Inokoshi, J.; Nishizawa, A.; Hirose, T.; Tanaka, H.; Iwai, Y.; Omura, S. J. Antibiotics
1
996, 49, 886–889.
1
1
7. Gijsen, H. J. M.; Qiao, L.; Fitz, W.; Wong, C.-H. Chem. Rev. 1996, 96, 443–473.
8. Vankar, Y. D.; Chaudhuri, N. C.; Singh, S. P. Synth. Commun. 1986, 16, 1621–1626.