Asymmetric bioreduction of ethyl 3-halo-2-oxo-4-phenylbutanoate by Saccharomyces cerevisiae immobilized in Ca-alginate beads with double gel layer

Article abstract of DOI:10.1021/op0502497

The asymmetric bioreduction of ethyl 3-halo-2-oxo-4-phenylbutanoate was studied for several microorganisms and especially for Saccharomyces cerevisiae. The highest chemical yield (90%), de (70%) and ee (96-99%) were obtained with S. cerevisiae immobilized in calcium alginate beads with double gel layers, and reaction conditions were optimized by changing matrix of immobilization, concentration of substrate, and feeding with glucose as electron donor. The entrapment of cells with double gel layers was fundamental to achieve high enantio- and diastereoselectivity.

Full text of DOI:10.1021/op0502497

Organic Process Research & Development 2006, 10, 611−617
Asymmetric Bioreduction of Ethyl 3-Halo-2-oxo-4-phenylbutanoate by
Saccharomyces cerevisiae Immobilized in Ca-Alginate Beads with Double Gel
Layer
Humberto M. S. Milagre,† C´ıntia D. F. Milagre,† Paulo J. S. Moran,† Maria Helena A. Santana,‡ and
J. Augusto R. Rodrigues*,†
State UniVersity of Campinas, Institute of Chemistry, CP 6154, CEP 13084-971 Campinas, SP, Brazil, and State
UniVersity of Campinas, School of Chemical Engineering, CP 6066, CEP 13081-970 Campinas, SP, Brazil
Abstract:
disadvantages.9-12 One alternative is to use immobilized cells
for enantioselective reductions of ketones.13 Several methods
for entrapment of living and growing cells have been
developed.14,15 A preparation that provides extremely mild
immobilization conditions is the entrapment within ionotropic
gels, such as calcium alginate.16
Entrapment of cells in calcium alginate is the most widely
used immobilization technique in the biocatalytic production
of chemical compounds.17-21 Alginate is cheap and readily
available, has a high affinity for water, and has the ability
to form gels under mild conditions, which are suitable for
most cells. It is nontoxic and nonpathogenic, what makes it
attractive for applications in the food and the pharmaceutical
industries.20 Several ketone reductions using immobilized
cells have been reported in the literature and, in many cases,
by S. cereVisiae immobilized in calcium alginate.13,22-26 Other
microorganisms27,28 as well as other immobilization matrixes
The asymmetric bioreduction of ethyl 3-halo-2-oxo-4-phenylbu-
tanoate was studied for several microorganisms and especially
for Saccharomyces cereWisiae. The highest chemical yield (90%),
de (70%) and ee (96-99%) were obtained with S. cereWisiae
immobilized in calcium alginate beads with double gel layers,
and reaction conditions were optimized by changing matrix of
immobilization, concentration of substrate, and feeding with
glucose as electron donor. The entrapment of cells with double
gel layers was fundamental to achieve high enantio- and
diastereoselectivity.
Introduction
The production of enantiomeric pure compounds is of
increasing importance in the fine chemical and pharmaceuti-
cal industries in particular.1 Many of these compounds can
be obtained by asymmetric reductions of ketones mediated
by Saccharomyces cereVisiae (baker’s yeast).2 In biocatalytic
conversions,3-7 whole cells are used more often than isolated
enzymes since cofactor regeneration is not required for
sustained catalytic activity. S. cereVisiae is an economically
attractive biocatalyst due to its availability, low cost, ease
of handling and disposal, safety for food and pharmaceutical
applications, as well as its capacity to catalyze a wide range
of stereoselective reductions.3-8 However, this method has
not been considered suitable for large-scale production of
chiral alcohol, due to the low concentration of reagents and
the long process of isolation of products, since the reactions
are generally performed in batch process. Some work with
continuous cell-culture systems tried to overcome these
(9) Chin-Joe, I.; Haberland, J.; Straathof, A. J. J.; Jongejan, J. A.; Liese, A.;
Heijnen, J. J. Enzyme Microb. Technol. 2002, 31, 665-672.
(10) Kometani, T.; Yoshii, H.; Matsuno, R. J. Mol. Catal. B: Enzym. 1996, 1,
45-52.
(11) Goubet, I.; Maugard, T.; Lamare, S.; Legoy, M. D. Enzyme Microb. Technol.
2002, 31, 425-430.
(12) Wendhausen, R., Jr.; Moran, P. J. S.; Joekes, I.; Rodrigues, J. A. R. J. Mol.
Catal. B: Enzym. 1998, 5, 69-73.
(13) Buque, E. M.; Chin-Joe, I.; Straathof, A. J. J.; Jongejan, J. A.; Heijnen, J.
J. Enzyme Microb. Technol. 2002, 31, 656-664.
(14) Park, J. K.; Chang, H. N. Biotechnol. AdV. 2000, 18, 303-319.
(15) Bickerstaff, G. F. Immobilization of Enzymes and Cells; Humana Press:
New Jersey, 1997.
(16) Kierstan, M.; Bucke, C. Biotechnol. Bioeng. 1977, 19, 387-397.
(17) Smidsrød, O.; Skjåk-Braek, G. Trends Biotechnol. 1990, 8, 71-78.
(18) Hannoun, B. J. M.; Stephanopoulos, G. Biotechnol. Bioeng. 1986, 28, 829-
835.
(19) Poncelet, D.; Lencki, R.; Beaulieu, C.; Halle, J. P.; Neufeld, R. J.; Fournier,
A. Appl. Microbiol. Biotechnol. 1992, 38, 39-45.
(20) Martinsen, A.; Storrø, I.; Skjåk-Braek, G. Biotechnol. Bioeng. 1992, 39,
186-194.
* Author for correspondence. E-mail: jaugusto@iqm.unicamp.br.
† Institute of Chemistry.
‡ School of Chemical Engineering.
(21) Hulst, A. C.; Tramper, J.; Van’t Riet, K.; Westerbeek, J. M. M. Biotechnol.
Bioeng. 1985, 27, 870-876.
(1) Straathof, A. J. J.; Panke, S.; Schmid, A. Curr. Opin. Biotechnol. 2002, 13,
548-556.
(2) Nakamura, K.; Yamanaka, R.; Matsuda, T.; Harada, T. Tetrahedron:
Asymmetry 2003, 14, 2659-2681.
(22) Gervais, T. R.; Carta, G.; Gainer, J. L. Biotechnol. Prog. 2003, 19, 389-
395.
(23) Griffin, D. R.; Gainer, J. L.; Carta, G. Biotechnol. Prog. 2001, 17, 304-
310.
(3) Buque, E. M.; Chin-Joe, I.; Straathof, A. J. J.; Jongejan, J. A.; Heijnen, J.
J. Enzyme Microb. Technol. 2002, 31, 656-664.
(4) Brzezin˜ska-Rodak, M.; Z˙ yman´czyk-Duda, E.; Kafarski, P.; Lejczak, B.
Biotechnol. Prog. 2002, 18, 1287-1291.
(24) Fadnavis, N. W.; Vadivel, S. K.; Sharfuddin, M.; Bhalerao, U. T.
Tetrahedron: Asymmetry 1997, 24, 4003-4006.
(25) Naoshima, Y.; Munakata, Y.; Nishiyama, T.; Maeda, J.; Kamezawa, M.;
Haramaki, T.; Tachibana, H. World J. Microbiol. Biotechnol. 1991, 7, 219-
224.
(5) Chen, J.; Wang, K.; Houng, J.; Lee, S. Biotechnol. Prog. 2002, 18, 1414-
1422.
(6) Lourenc¸o, E.; Rodrigues, J. A. R.; Moran, P. J. S. J. Mol. Catal. B: Enzym.
2004, 29, 37-40.
(7) Filho, E. S. P.; Rodrigues, J. A. R.; Moran, P. J. S. Tetrahedron: Asymmetry
2001, 12, 847-852.
(26) Talebnia, F.; Niklasson, C.; Taherzadeh, M. J. Biotechnnol. Bioeng. 2005,
90, 345-353.
(27) Naoshima, Y.; Akakabe, Y.; Watanabe, F. Agric. Biol. Chem. 1989, 53,
545-547.
(8) Dahl, A. C.; Fjeldberg, M.; Madsen, J. Ø. Tetrahedron Asymmetry 1999,
10, 551-559.
(28) Mart´ınez-Lagos, F.; Sinisterra, J. V. J. Mol. Catal. B: Enzym. 2005, 36,
1-7.
10.1021/op0502497 CCC: $33.50 © 2006 American Chemical Society
Published on Web 03/10/2006
Vol. 10, No. 3, 2006 / Organic Process Research & Development
•
611

Scheme 1. Reduction of ethyl 3-chloro-2-oxo-4-phenylbutanoate 1 mediated by baker’s yeast
Table 1. Reduction of ethyl
3-bromo-2-oxo-4-phenylbutanoate 1b with free cells of
several strains
such as polyurethane,29 carrageenan,25 and chrysotile fibers12
have also been described.
Recently, we described the first example of ketoester
reduction mediated by immobilized baker’s yeast in alginate
fibers with double gel layer.30 The immobilization procedure
prevents cell leakage from alginate fibers into the medium
in a device in which the alginate fiber cells were restricted
to the inner layer, whereas the outer layer helped to prevent
leakage and contact with potential inhibitors. This method
of cell entrapment was developed by Tanaka et al. for
continuous fermentation in the production of ethanol 31 and
Chitinase.32,33
Asymmetric reductions of 3-chloro-2-oxoalkanoates with
fermenting baker’s yeast have been reported by Tsuboi et
al.34 The reduction of corresponding esters with baker’s yeast
gave optically active 3-chloro-2-hydroxyalkanoic esters (anti
(2S,3R):syn (2S,3S) ) 52:48-90:10) in 50-85% yields and
>95% ee, except for 43% ee of ethyl syn-(2S,3S)-3-chloro-
2-hydroxy-phenylbutanoate. The reduction of 1a was achieved
with good diastereoselectivity (anti (2S,3R)-3a:syn (2S,3S)-
2a ) 66:34) and high enantioselectivity anti-(2S,3R)-3-
chloro-2-hydroxy-phenylbutanoate in 95% ee (Scheme 1).
The chiral 3-chloro-2-hydroxyalkanoic esters are very useful
in the synthesis of chiral glycidic esters34-37 and 2,3-epoxi
alcohols,38 which are key intermediates for the synthesis of
important bioactive molecules.34-38
ee (%)
syn anti
yield
(%)
entry
strain
pH
syn:antib
1
2
3
4
5
6
7
8
Rhodotorula minutaa
Rhodotorula minutaa
Rhodotorula glutinisa
Rhodotorula glutinisa
Candida utiliza
4.0
7.0
4.0
7.0
4.0
7.0
4.0
7.0
4.0
7.0
4.0
7.0
4.0
7.0
59
75
39
50
65
62
48
65
55
48
4
10
7
15
87
80
32:68
44:56
34:66
45:55
64:36
65:35
75:25
65:35
68:32
66:34
3:97
0:100
0:100
0:100
69:31
64:36
93 >99
89 94
90 >99
96 >99
85
84
90
90
90
91
89
84
97
74
80
74
Candida utiliza
Pichia canadensisa
Pichia canadensisa
L. mesenteroidesa
L. mesenteroidesa
Serratia rubideac
Serratia rubideac
Serratia marcencesc
Serratia marcencesc
9
10
11
12
13
14
15
16
>99 >99
-
-
>99
>99
>99
95
-
Saccharomyces cereVisiaed 4.0
Saccharomyces cereVisiaed 7.0
83
78
94
a Reaction time ) 72 h. b The diastereomeric ratio was determined by GC-
MS, and enantiomeric excess was determined by GC (see Experimental Section).
c Reaction time ) 96 h. d Reaction time ) 24 h. Reaction conditions are described
in the Experimental Section.
results.34,39-41 The bioreduction was carried out using resting
cells under nonfermenting conditions, i.e. suspended in water
without addition of sugar. All microorganisms reduced 1b
with various ranges of chemical and optical yields. The
results are summarized in Table 1. With the yeasts Rhodo-
torula minuta (entries 1 and 2) and Rhodotorula glutinis
(entries 3 and 4) and the bacteria Serratia rubidea (entries
11 and 12) and Serratia marcences (entries 13 and 14), the
diastereoisomer anti (2S,3R)-3b was predominant in high ee.
Serratia sp. showed excellent diastereo and enantioselectivity,
but low chemical yields. With other yeasts the diastereo-
isomer syn (2S,3S)-2b was predominant, with good ee. A
higher de was observed at pH 4.0 than at pH 7.0 since 1b
was selectively reduced to the ester (2S,3S)-2b, while the
substrates underwent a dynamic kinetic resolution due to
epimerization occurring at C-2 via enol intermediate under
the effect of pH.24 At low pH (pH 4.0) the rate of enolization
is fast, and the syn-alcohol is formed as the major product.
At pH 7.0, when enolization becomes slower, enantioselec-
tivity prevails, but optically pure diastereomers are formed
in a lower ratio.
In this work, we describe the asymmetric bioreduction
of ethyl 3-halo-2-oxo-4-phenylbutanoate with several mi-
croorganisms, and especially with S. cereVisiae, which we
have optimized by changing the matrix of immobilization
and the concentration of substrate, and by feeding with
glucose as electron donor. The entrapment of S. cereVisiae
with double gel layers was fundamental to achieve high
enantio- and diastereoselectivity.
Results and Discussion
Different strains of yeast and bacteria for microbiological
reduction of 1b were selected according to literature
(29) Nakamura, K.; Higaki, M.; Ushio, K.; Oka, S.; Ohno, A. Tetrahedron Lett.
1985, 26, 870-876.
(30) Milagre, H. M. S.; Milagre, C. D. F.; Moran, P. J. S.; Santana, M. H. A.;
Rodrigues, J. A. R. Enzyme Microb. Technol. 2005, 37, 121-125.
(31) Tanaka, H.; Irie; S.; Ochi; H. J. Ferment. Bioeng. 1989, 68, 216-219.
(32) Tanaka, H.; Kaneko, Y.; Aoyagi, H.; Yamamoto, Y.; Funukaga, Y. J.
Ferment. Bioeng. 1996, 81, 220-225.
(33) Tanaka, H.; Aoyagi, H.; Yamamoto, Y.; Funukaga, Y. J. Ferment. Bioeng.
1996, 81, 394-399.
(34) Tsuboi, S.; Furutani, H.; Ansari, H. M.; Sakai, T.; Utaka, M.; Takeda, A.
J. Org. Chem. 1993, 58, 486-492.
High chemical yields and reasonable de were obtained
with S. cereVisiae (entries 15 and 16) which was chosen for
further studies to improve the preliminary results. The
approach was to use free cells of S. cereVisiae without
(35) Rodrigues, J. A. R.; Milagre, H. M. S.; Milagre, C. D. F.; Moran, P. J. S.
Tetrahedron: Asymmetry 2005, 16, 3099-3106.
(39) Nikaido, T.; Matsuyama, A.; Ito, M.; Kobayashi, Y.; Oonishi, H. Biosci.
Biotechnol. Biochem. 1992, 56, 2066-2067.
(40) Oda, S.; Inada, Y.; Kobayashi, A.; Ohta, H. Biosci. Biotechnol. Biochem.
1998, 62, 1762-1767.
(41) De Conti, R. M.; Porto, A. L. M.; Rodrigues, J. A. R.; Moran, P. J. S.;
Manfio, G. P.; Marsaioli, A. J. J. Mol. Catal. B: Enzym. 2001, 11, 233-
236.
(36) Feske, B. D.; Kaluzna, I. A.; Stewart, J. D. J. Org. Chem. 2005, 70, 9654-
9657.
(37) Feske, B. D.; Stewart, J. D. Tetrahedron: Asymmetry 2005, 16, 3124-
327.
(38) Tsuboi, S.; Yamafuji, N.; Utaka, M. Tetrahedron: Asymmetry 1997, 5, 375-
379.
612
•
Vol. 10, No. 3, 2006 / Organic Process Research & Development

Figure 1. (a) Bioreduction of 1b (0.49 mmol) and (b) bioreduction of 1b (0.18 mmol). Both reductions were mediated by
Saccharomyces cereWisiae with free cells suspended in citrate buffer, at pH 4.0 (ee of 2b (9), ee of 3b (b), de 2b/3b (2)).
When the cells were immobilized, the selectivity of these
reactions was better than that observed for free cells, as
shown in Figure 3a. The gradient of concentration of
substrate in the reaction medium is different from the gradient
of concentration inside the calcium alginate beads, where
the substrate is reduced by the S. cereVisiae, due to the slow
diffusive transport.13 Figure 3b shows the results of the
reaction using immobilized cells with addition of glucose.
The yields are almost the same as in Figure 3a, but the
diastereoselectivity was improved since the immobilization
shield plays an important role by creating a barrier to
substrate partition. The glucose acts as an electron donor
since the enzymes (oxidoreductases) involved in the reduc-
Figure 2. Schematic diagram of the concentration profile of
tion use mainly NAD(P)H as cofactor. As S. cereVisiae
contains only a catalytic amount of NAD(P)H, its regenera-
substrates in a spherical immobilized catalyst.13
tions must take place for sustained catalytic activity. With
addition of glucose. Figure 1a shows the results obtained
cell entrapment this effect makes a difference, as can be seen
after 24 h of reaction, in which 60% of substrate was
by comparison between Figure 3a (no glucose, nonfermenting
consumed. The initial concentration of substrate available
condition) and Figure 3b (reaction in the presence of
to the biocatalyst was relatively low because it was not fully
glucose).
soluble in the reaction mixture. Both the initial de and ee
Results of bioreduction of 1b at constant cell concentra-
were higher than at the end (24 h), where the de of 2b was
tion using immobilized beads of different particle sizes are
12.5% while the ee of 3b remained constant. In an attempt
presented in Table 2. The results using free cells and using
to increase the selectivity, the bioreduction was carried out
immobilized cells at dP ) 1.1 mm were comparable;
with a lower concentration of 1b (0.18 mmol instead of 0.49
however, this is not the case anymore at larger particle sizes.
At the highest yield (92%), de and ee both for 2b and 3b
were achieved using immobilized cells at dP ) 3.0 mm. This
latter condition was used in the following experiments.
The yeast Rhodotorula minuta and the bacterium Serratia
marcences were also immobilized. However, for these
mmol), Figure 1b. The selectivity was higher in the initial
stages than at the end, but the ee increased. The de and ee
were both higher in the initial stages when the substrate
concentration was low, but the diastereomerical selectivity
decreased rapidly and then remained constant until the end.
In another attempt to decrease the availability of the
microorganisms, this entrapment process reduces enzymatic
activity. Different matrixes were used for the immobilization
of S. cereVisiae, but entrapment in calcium alginate gave
better results than the other matrixes used (Table 3).
The reaction with S. cereVisiae immobilized in calcium
alginate, at pH 4 and with addition of glucose (entry 1, Table
3), was scaled up (entry 4, Table 3). The de for syn (2S,3S)
dropped from 50 to 40%, and the ee from 96 to 85%, while
the yield remained around 80%. To improve the selectivity
a new procedure was established by decreasing even more
the concentration of substrate available for the biocatalyst,
substrate, the cells were immobilized in alginate beads
creating a barrier at the border of the cellular membrane.13
During the reduction, the concentration levels of these
substrates inside the alginate beads was lower than in the
solution, due to slow diffusive transport13 (see Figure 2). The
lower concentration of substrate in the beads helps the
stereoselectivity, since only the enzyme with lowest Km
(Michaelis constant) is able to react with high Vmax. It is well-
known that baker’s yeast possesses several alcohol dehy-
drogenases with varying substrate selectivities and even
opposing enantioselectivities.13
Vol. 10, No. 3, 2006 / Organic Process Research & Development
•
613

Figure 3. (a) Bioreduction of 1b (0.18 mmol) mediated by S. cereWisiae immobilized in Ca-alginate beads and suspended in citrate
buffer at pH 4.0. (b) Bioreduction of 1b (0.18 mmol) mediated by S. cereWisiae immobilized in Ca-alginate beads and suspended in
citrate buffer at pH 4.0 with addition of glucose (ee of 2b (9), ee of 3b (b), de 2b/3b (2)).
Table 2. Influence of particle size on reduction of 1b using
immobilized S. cereWisiae at constant cell concentrations
ee (%)
dP,mean (mm)
yield (%)
syn : anti
syn
anti
free cells
1.1
2.1
2.3
3.0
87
83
85
88
92
69:31
67:33
73:27
75:25
75:25
83
85
92
92
96
95
95
>99
>99
>99
Figure 4. Schematic diagram of a cross section of Ca-alginate
beads with double gel layers.
Table 3. Bioreduction of 1b mediated by S. cereWisiae
immobilized in different porous matrixes
ee (%)
entry
matrix
yield (%) syn:antic syn
anti
1
2
3
4
Ca-alginatea
κ-carrageenana
Lenticatsa
92
87
79
80
75:25
75:25
75:25
70:30
96
90
90
85
>99
>99
>99
>99
Ca-alginateb
a Bioreduction of 1b (0.18 mmol) mediated by 1.0 g of S. cereVisiae suspended
in citrate buffer (60 mL, pH 4.0) with addition of glucose, reaction time ) 24
h. b Bioreduction of 1b (2,2 mmol) mediated by 5.0 g of S. cereVisiae suspended
in citrate buffer (300 mL, pH 4.0) with addition of glucose, reaction time ) 24
h. c The diastereomeric ratio was determined by GC-MS, and enantiomeric
excess was determined by GC (see Experimental Section).
introducing a second barrier of alginate. The cells were
immobilized with double gel layers, which also prevent cells
from leaking out of the gel beads (Figure 4). It is well-known
that the net pores of Ca-alginate gel beads are so large that
biocatalyst can sometimes leak out from the gel.42 Figure 5
shows the time course of 1b bioreduction mediated by
Saccharomyces cereVisiae immobilized in Ca-alginate beads
with double gel layers. It is worth pointing out that the
selectivity of this reaction was the same during all reaction
times (compare Figure 5 to Figure 3b). The conditions for
bioreduction of 1b were optimized, and enantioselectivity
was improved for both syn and anti enantiomers. Also, the
Figure 5. Bioreduction of 1b (0.18 mmol) mediated by
Saccharomyces cereWisiae immobilized in Ca-alginate beads with
double gel layers suspended in citrate buffer at pH 4.0, with
addition of glucose (1b (f), 2b (0), 3b (O), ee of 2b (9), ee of
3b (b), de 2b/3b (2).
syn:anti ratio increased to 85:15. With double gel layers
another barrier was created, and the gradients of substrate
concentration in the alginate beads affected the overall
reduction performance.
The double gel layer immobilization was used for
bioreduction of ethyl 3-chloro-2-oxo-4-phenylbutanoate 1a
and compared with the results reported by Tsuboi,34 see Table
(42) Won, K.; Kim S.; Kim, K.-J.; Park, H. W.; Moon, S. J. Process Biochem.
2005, 40, 2149-2154.
(43) Okonya, J. F.; Hoffman, R. V.; Johnson, M. C. J. Org. Chem. 2002, 67,
1102.
614
•
Vol. 10, No. 3, 2006 / Organic Process Research & Development

Table 4. Comparison between the bioreduction of 1
mediated by S. cereWisiae immobilized in Ca-alginate and
literature data32
determined at 75.5 MHz (Varian Gemini 300) or 125.7 MHz
(INOVA 500). Chemical shifts are reported in ppm relative
to tetramethylsilane (TMS) in CDCl3. Gas chromatographic
analyses and mass spectra were obtained on a QP 5000-
Shimadzu or an CG 6890/ Hewlett Packard 5973 equipped
with J&W Scientific HP-5 (5% phenylmethylpolysiloxane,
30.0 m × 250 µm × 0.25 µm), Macherey 212117/91
Hydrodex-â 3P (25.0 m × 250 µm × 0.25 µm) capillary
columns. High-resolution mass spectrum was determined on
a VG Auto Spec Micromass. Optical rotations were measured
on a Perkin-Elmer Polarimeter 341. Melting points were
measured on a Microquimica MQ APF-301.
ee (%)
entry
substrate
yield (%)
syn:anti
syn
anti
1a
2b
3b
ref 34
1a
1b
50
85
90
34:66
30:70
85:15
43
85
96
95
>99
>99
a Reaction conditions: 42.5 g of S. cereVisiae suspended in citrate buffer
(300 mL, pH 7.0), 720 mg of 1 and addition of 54.5 g of glucose. b Reaction
conditions: 5.0 g of S. cereVisiae entrapped in Ca-alginate beads with double
gel layers in large scale were suspended in citrate-phosphate buffer (60 mL,
pH 4.0) containing glucose (8 g), 2.2 mmol of 1 (1a or 1b). After each 6 h,
more glucose (1.6 g) was added.
Bromination of Ethyl 2-oxo-4-phenylbutanoate. To a
solution of ethyl 2-oxo-4-phenylbutanoate (5.0 g, 24.2 mmol)
in chloroform (7 mL) was added slowly a solution of bromine
in carbon tetrachloride (1.06 mol‚L-1) until the color of the
solution became red-orange. The excess of bromine was
destroyed with a sodium thiosulphate solution, and the
reaction mixture was extracted with ethyl acetate, dried over
anhydrous MgSO4, and then filtered and evaporated. The
crude material was purified by flash chromatography and
eluted with 20% ethyl acetate in hexane.
4. The diastereomeric ratio was almost the same for both
methods, but for S. cereVisiae entrapped in Ca-alginate beads
with double gel layers, the ee was better than that reported
in the literature (Table 4, entry 3).
The relative configuration of syn-2b was established by
comparison with the known ethyl (2S,3R)-3-chloro-2-hy-
droxy-4-phenylbutanoate,34 and the absolute configuration
was determined by hydrogenolysis with H2 in the presence
of Pd/C 5% of the syn-2b, to give the known ethyl (-)-
(2S)-2-hydroxy-4-phenylbutanoate in 95% yield, [R]20 -32.5
(c 2.0, CHCl3). Bioreduction of 1a by S. cereVisiae entrapped
in alginate beads with double gel layers gave ethyl (2S,3S)-
3-chloro-hydroxy-butanoate 2a (85% ee) and its epimer, the
(2S,3R)-derivative 3a (>99% ee; ratio 2a:3a ) 30:70) in
Ethyl 3-bromo-2-oxophenylbutanoate, 4: pale-yellow
1
oil, 97% yield; H NMR values matched those reported in
the literature.38 1H NMR (300 MHz, CDCl3): δ 1.35 (3H, t,
J ) 7.0 Hz, H-12), 3.24 (1H, dd, J ) 14.5 e 7.5 Hz, H-4),
3.52 (1H, dd, J ) 14.5 e 7.5 Hz, H-4), 4.32 (2H, q, J ) 7.0
Hz, H-11), 5.28 (1H, t, J ) 7.3, H-3) 7.28 (5H, m, H-6 to
H-10).
1
85% yield. H NMR coupling constants for 2-H and 3-H
for syn-2a (δ 4.44, J2,3 ) 1.8 Hz) and for anti-3a (δ 4.24,
J2,3 ) 2.6 Hz) have the same values as for syn-2b (δ 4.50,
J2,3 ) 1.8 Hz) and anti-3b (δ 4.46, J2,3 ) 2.6 Hz), confirming
the proposed absolute configuration for 2b (1H NMR and
13C NMR values matched those reported by Tsuboi).34
In conclusion, we have developed a highly diastereo and
enantioselective method for bioreduction of ethyl 3-halo-2-
oxo-4-phenylbutanoate to produce 3-halo-2-hydroxy-phen-
ylbutanoates. Among the biocatalysts tested, S. cereVisiae
gave the highest product yields, and the entrapment in
alginate beads with double gel layers showed its efficacy in
controlling the substrate concentration which is fundamental
for the enantioselectivity and also for obtaining high syn:
anti ratio of 3-halo-2-hydroxy-phenylbutanoates.
Bioreductions of Ethyl 3-bromo-2-oxo-4-phenylbu-
tanoate. Screening experiments were conducted for yeast
grown for 24 h in an orbital shaker (110 rpm) at 30 °C in
50 mL of a liquid medium containing (per liter), yeast extract
3.0 g, malt extract 3.0 g, bacto-peptone 5.0 g, glucose 20.0
g. The yeast was centrifuged, then resuspended in 1.0 L of
the same medium and grown in an orbital shaker (110 rpm)
at 30 °C for 48 h. The yeast was centrifuged (3000 rpm),
and then washed. Each 4.0 g of biomass were resuspended
in citrate-phosphate buffer (100 mL, pH 4.0), and 143.0
mg (0.5 mmol) of substrate was added. The reaction was
maintained in an orbital shaker (110 rpm) at 30 °C for 24 h.
The bacteria were similarly grown using the following
medium: meat extract 5.0 g, bacto-peptone 5.0 g, glucose
2.0 g, K2HPO4. The reaction was performed in the same way
for all yeasts, except S. cereVisiae. The reactions were
monitored by GC-MS, and at the end of the reaction (24
h), the biomass was centrifuged and washed with ethyl
acetate. The reaction mixture was extracted with ethyl
acetate, dried over anhydrous MgSO4, and then filtered and
evaporated. The crude material was purified by flash
chromatography and eluted with 20% ethyl acetate in hexane.
Ethyl (2S,3S)-3-Bromo-2-hydroxy-phenylbutanoate,
Experimental Section
All reagents and solvents were obtained from commercial
sources. The yeast extract, malt extract, and bacto-peptone
were purchased from Biobra´s (Brazil). The microorganisms
were purchased from the Colec¸a˜o de Culturas Tropical (CCT)
Fundac¸a˜o Andre´ Tosello, Campinas, SP, Brazil, except for
the dry S. cereVisiae, which was purchased from Emulzint
Ltd. (Belgium) and stored in a refrigerator. Thin-layer
chromatography (TLC) analyses were performed with pre-
coated aluminum sheets (silica gel 60 Merck), and flash
column chromatography was carried out on silica (200-400
mesh, Merck). IR spectra were recorded on a FT-IR
BOMEM MB-100 from Hartmann & Braun. 1H NMR
spectra were determined at 300 MHz (Varian Gemini 300)
or 500 MHz (INOVA-500), and 13C NMR spectra were
5: colorless oil, [R]20 -16.8 (c 2.1, CHCl3); IR (film):
D
3503, 3066, 3027, 2983, 2929, 1732, 1596, 1499, 1455, 1367,
1256, 1110, 1017, 740, 701 cm-1; MS m/z (%): 206 (2),
189 (15), 177 (1), 161 (5), 143 (15), 133 (51), 115 (66), 91
1
(100), 77 (15), 55 (20), 51 (15); H NMR (300 MHz,
CDCl3): δ 1.29 (3H, t, J ) 7.1 Hz, H-12), 3.32 (2H, m,
Vol. 10, No. 3, 2006 / Organic Process Research & Development
•
615

H-4), 4.14 (1H, d, J ) 1.8 Hz, H-2), 4.26 (2H, m, H-11),
4.50 (1H, dt, J ) 1.8 e 8.1 Hz, H-3), 7.26 (5H, m, aromatic);
13C NMR (75 MHz, CDCl3): δ 14.8 (CH3), 42.4 (CH2, C-4),
56.9 (CH, C-3), 62.9 (CH2, C-11), 71.5 (CH, C-2), 127.6
(CH, aromatic), 129.1 (2CH, aromatic), 129.7 (2CH, aro-
matic), 138.0 (C, aromatic), 172.2 (CdO). HRMS: calcd
for C12H15BrO3: 286.02046; found: 286.02051.
General Procedure for Reduction Mediated by Sac-
charomyces cereWisiae Entrapped in κ- Carrageenan
Beads. A solution of κ-carrageenan (50 mL, 3%) was added
to a suspension of S. cereVisiae (1.0 g) in distilled water
(6.0 mL) 3% at 40 °C. This mixture was extruded using
syringe nozzles (equipped with external thermostat ribbon
at 40 °C) with inner diameters of 1.0 mm to a solution of
KCl (0.3 mol‚L-1) to give beads 3 mm in diameter. After
20 min the beads were filtered and washed with water. In a
100-mL bioreactor, the beads were suspended in citrate-
phosphate buffer (60 mL, pH 4.0) containing glucose (1.6
g) and stirred at 300 rpm at 30 °C. After activation of the
yeast for 2 h, the pH of the medium was adjusted with a
10% solution of NH4OH, and then the substrate 4 (0.5 mmol
in 5.0 mL of ethanol) was added slowly over 10 h with
stirring; after each 6 h glucose (0.32 g) was added. The
reaction was monitored by GC-MS; at the end of the
reaction (24 h), the beads were filtered and washed with ethyl
acetate, and the reaction mixture was extracted with ethyl
acetate, dried over anhydrous MgSO4, and then filtered and
evaporated. The crude material was purified by flash
chromatography and eluted with 20% ethyl acetate in hexane.
General Procedure for Reduction Mediated by Sac-
charomyces cereWisiae Entrapped in LentiKats. The solu-
tion of LentiKats was prepared according to the procedure
provided by the supplier and kept at 40 °C. A suspension of
S. cereVisiae (1.0 g) in distilled water (6.0 mL) was added
and dispersed homogeneously by mixing with a magnetic
stirrer. For production of the LentiKats, a smooth plate is
needed, preferably one made of polystyrene. Using a standard
syringe with a nozzle (1.0 mm in diameter) droplets were
formed and dripped neatly onto the surface of the plate. The
droplets formed (approximately 3 mm in diameter and 5 mg
in mass) were dried by exposure to air, and the stabilizer
was added. After 2 or 3 min of contact with the stabilizer
solution LentiKats was easily removed from the surface and
put into a bottle containing a 10-fold surplus of LentiKat-
Stabilizer. After stabilization was finished, the supernatant
had to be removed. In a 100-mL bioreactor, the beads were
suspended in citrate-phosphate buffer (60 mL, pH 4.0)
containing glucose (1.6 g) and stirred at 300 rpm at 30 °C.
After activation of the yeast for 2 h, the pH of the medium
was adjusted with a 10% solution of NH4OH, and then
substrate 4 (0.5 mmol in 5.0 mL of ethanol) was added
slowly over 10 h with stirring. After each 6 h, glucose (0.32
g) was added. The reaction was monitored by GC-MS. At
the end of the reaction (24 h), the beads were filtered and
washed with ethyl acetate. The reaction mixture was
extracted with ethyl acetate and dried over anhydrous
MgSO4; the solvent was then evaporated. The crude material
was purified by flash chromatography and eluted with 20%
ethyl acetate in hexane.
Ethyl (2S,3R)-3-Bromo-2-hydroxy-phenylbutanoate, 6:
colorless oil, [R]20 -18.8 (c 2.1, CHCl3) IR (film): 3503,
D
3066, 3027, 2983, 2929, 1732, 1596, 1499, 1455, 1367, 1256,
1110, 1017, 740, 701. MS m/z (%): 206 (2), 189 (15), 177
(1), 161 (5), 143 (15), 133 (51), 115 (66), 91 (100), 77 (15),
1
55 (20), 51 (15) H NMR (300 MHz, CDCl3): δ 1.35 (3H,
t, J ) 7.0 Hz, H-12), 3.30 (2H, m, H-4), 4.24 (2H, q, J )
7.0 Hz, H-11), 4.39 (1H, m, H-2), 4.46 (1H, dt, J ) 2.6, e
7.7 Hz, H-3), 7.26 (5H, m, H-6 to H-10). 13C NMR (75 MHz,
CDCl3): δ 14.8 (CH3), 41.2 (CH2, C-4), 56.8 (CH, C-3),
62.9 (CH2, C-11), 73.8 (CH, C-2), 127.5 (CH, aromatic),
128.9 (2CH, aromatic), 129.7 (2CH, aromatic), 137.9 (C0,
aromatic), 171.3 (C0, aromatic). HRMS: calcd for C12H15-
BrO3 [M+]: 286.02046, found: 286.02051.
Bioreductions of Ethyl 3-bromo-2-oxo-4-phenylbu-
tanoate Mediated by Saccharomyces cereWisiae. In a 100-
mL bioreactor, 1.0 g of S. cereVisiae was suspended in
citrate-phosphate buffer (60 mL, pH 4.0) containing glucose
(1.6 g) and stirred at 300 rpm at 30 °C. After activation of
the yeast for 2 h, the pH of the medium was adjusted with
a 10% solution of NH4OH, and then the substrate 4 (0.5
mmol in 5.0 mL of ethanol) was added slowly over 10 h
with stirring, and glucose (0.32 g) was added every 6 h. The
reaction was monitored by GC-MS, and at the end of the
reaction (24 h), the biomass was centrifuged and then washed
with ethyl acetate. The filtrate was extracted with ethyl
acetate, dried over anhydrous MgSO4, and then filtered and
evaporated. The crude material was purified by flash
chromatography and eluted with 20% ethyl acetate in hexane.
General Procedure for Reduction Mediated by Sac-
charomyces cereWisiae Entrapped in Ca-Alginate Beads.
A solution of sodium alginate (3%) was added to a
suspension of S. cereVisiae (1.0 g) in distilled water (10 mL).
This mixture was extruded using syringe nozzles with inner
diameters of 1.0 mm to a solution of CaCl2 (0.2 mol.L-1) to
give beads with diameters of 3 mm. After 20 min the beads
were filtered and washed with water to remove excess CaCl2.
In a 100-mL bioreactor, the beads were suspended in citrate-
phosphate buffer (60 mL, pH 4.0) containing glucose (1.6
g) and stirred at 300 rpm at 30 °C. After activation of the
yeast for 2 h, the pH of the medium was adjusted with a
10% solution of NH4OH, and then the substrate 4 (0.5 mmol
in 5.0 mL of ethanol) was added slowly over 10 h with
stirring. After each 6 h, glucose (0.32 g) was added. The
reaction was monitored by GC-MS, and at the end of the
reaction (24 h), the beads were filtered and washed with ethyl
acetate, and the reaction mixture was extracted with ethyl
acetate. The combined organic phases were dried over
anhydrous MgSO4, and the solvent was evaporated. The
crude material was purified by flash chromatography and
eluted with 20% ethyl acetate in hexane.
General Procedure for Reduction Mediated by Free
Saccharomyces cereWisiae. In a 100-mL bioreactor, 1.0 g
of S. cereVisiae was suspended in citrate-phosphate buffer
(60 mL, pH 4.0) and stirred at 300 rpm at 30 °C. Then the
substrate 4 (0.5 mmol in 5.0 mL of ethanol) was added, and
the reaction was monitored by GC-MS. At the end of the
616
•
Vol. 10, No. 3, 2006 / Organic Process Research & Development

reaction (24 h), the biomass was centrifuged and washed
with ethyl acetate; the reaction mixture was extracted with
ethyl acetate, dried over anhydrous MgSO4, and then filtered,
and the solvent was evaporated. The crude material was
purified by flash chromatography and eluted with 20% ethyl
acetate in hexane.
and stirred at 300 rpm at 30 °C. After activation of the yeast
for 2 h, the pH of the medium was adjusted with a 10%
solution of NH4OH, and then the substrate 2a-d (2.2 mmol
in 5.0 mL of ethanol) was added with stirring. After each 6
h, glucose (1.6 g) was added. The reaction was monitored
by GC-MS, and at the end of the reaction (24 h), the beads
were filtered and then washed with ethyl acetate. The reaction
mixture was extracted with ethyl acetate, dried over anhy-
drous MgSO4, and then filtered; the solvent was evaporated.
The crude material was purified by flash chromatography
and eluted with 20% of ethyl acetate in hexane.
General Procedure for Reduction Mediated by Free
Saccharomyces cereWisiae with Addition of Glucose. In a
100-mL bioreactor, 1.0 g of S. cereVisiae was suspended in
citrate-phosphate buffer (60 mL, pH 4.0) containing glucose
(1.6 g) and stirred at 300 rpm at 30 °C. After activation of
the yeast for 2 h, the pH of the medium was adjusted with
a 10% solution of NH4OH, and then the substrate 4 (0.18
mmol in 1.0 mL of ethanol) was added with stirring. The
reaction was monitored by GC-MS, and at the end of the
reaction (24 h), the biomass was centrifuged and washed
with ethyl acetate; the reaction mixture was extracted with
ethyl acetate, dried over anhydrous MgSO4 and then filtered,
and the solvent was evaporated. The crude material was
purified by flash chromatography and eluted with 20% of
ethyl acetate in hexane.
General Procedure for Reduction Mediated by Sac-
charomyces cereWisiae Entrapped in Ca-Alginate Beads
with Double Gel Layers in Large Scale. A solution of
sodium alginate (3%) was added to a suspension of S.
cereVisiae (5 g) in distilled water (50 mL). This mixture was
extruded using syringe nozzles with inner diameters of 1.0
mm to a solution of CaCl2 (0.2 mol‚L-1) to produce beads
3 mm in diameter. After 20 min the beads were filtered and
the surfaces dried using filter paper. Afterwards the Ca-
alginate beads were placed into 400 mL of stirred 1.5%
sodium alginate. After 20 min coating time, the beads were
sieved, washed with water, and hardened for 2 h in a solution
of CaCl2 (0.2 mol‚L-1). During the coating procedure, Ca2+
diffused from the Ca-alginate bead into the alginate solution,
and a cell-free Ca-alginate layer was formed surrounding
the biocatalyst. The beads with double gel layers were
washed with water to remove the excess CaCl2. In a 500-
mL bioreactor, the beads were suspended in citrate-
phosphate buffer (300 mL, pH 4.0) containing glucose (8 g)
Ethyl 3-Chloro-2-hydroxy-4-phenylbutanoates, 2 and
3. The reduction of 3-chloro-2-oxo-4-phenylbutanoate was
carried out by the procedure described above for 4. The crude
products were separated by flash chromatography and eluted
1
with hexane/ethyl acetate (5:1). IR and H NMR values
matched those reported in the literature for 2 and 3.34
Ethyl (-)-(2S)-2-Hydroxy-4-phenylbutanoate, 8. So-
dium acetate (57 mg, 0.7 mmol) and 5% palladium on carbon
(16 mg) were added to a solution of 3b (182 mg, 0.7 mmol)
in anhydrous ethanol (3 mL) under argon. After stirring 2 h
at room temperature, the reaction mixture was filtered over
Celite, extracted with ethyl acetate (3 × 50 mL), and washed
with brine; the organic phase was dried in anhydrous MgSO4
and filtered. Then the solvent was distilled under reduced
pressure. The crude product was purified by silica gel
chromatography and eluted with hexane in ethyl acetate
(15%) to give 8 (138 mg, 95%). Yellow oil, [R]20D -32.5 (c
2.0, CHCl3); 1H NMR (300 MHz, CDCl3): δ 1.28 (3H, t, J
) 7.0 Hz, H-12), 1.90-2.18 (2H, m, H-4), 2.77 (2H, m,
H-3), 4.20 (2H, q, J ) 7.0 Hz, H-11), 7.26 (5H, m,
aromatic).12
Acknowledgment
We are grateful to FAPESP, CAPES, UNICAMP-FUN-
CAMP, and CNPq for their financial support.
Received for review December 19, 2005.
OP0502497
Vol. 10, No. 3, 2006 / Organic Process Research & Development
•
617