Highly enantioselective bioreduction of 4-bromoacetophenone

Article abstract of DOI:10.1016/j.tetasy.2011.10.014

Microorganisms were used to reduce 4-bromoacetophenone to (S)-4-bromophenylethanol and (R)-4-bromophenylethanol. After a fractional factorial design to identify the important variables for this reaction, Geotrichum candidum provided a 98.9% conversion with >99% ee of the (R)-isomer, while Rhodotorula rubra led to a 97.6% conversion with a 98.8% ee of the S-isomer.

Full text of DOI:10.1016/j.tetasy.2011.10.014

Tetrahedron: Asymmetry 22 (2011) 1763–1766
Contents lists available at SciVerse ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Highly enantioselective bioreduction of 4-bromoacetophenone
Raquel de Oliveira Lopes a, Joyce Benzaquem Ribeiro a, , Aline de Souza Ramos b, Leandro S.M. Miranda a,
⇑
Ivana Corrêa Ramos Leal a, Selma Gomes Ferreira Leite c, Rodrigo Octavio Mendonça Alves de Souza a
a Biocatalysis and Organic Synthesis Group, Instituto de Quimica, Universidade Federal do Rio de Janeiro, Rio de Janeiro 22941-909, Brazil
b Farmanguinhos, Fundação Oswaldo Cruz, Rua Sizenando Nabuco 100, Manguinhos, Rio de Janeiro, RJ 21041-250, Brazil
c Escola de Química, Universidade Federal do Rio de Janeiro, CT Bloco E, Cidade Universitária, Rio de Janeiro, RJ 21949-909, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
Microorganisms were used to reduce 4-bromoacetophenone to (S)-4-bromophenylethanol and (R)-4-
bromophenylethanol. After a fractional factorial design to identify the important variables for this reac-
tion, Geotrichum candidum provided a 98.9% conversion with >99% ee of the (R)-isomer, while Rhodotorula
rubra led to a 97.6% conversion with a 98.8% ee of the S-isomer.
Received 8 September 2011
Accepted 18 October 2011
Available online 15 November 2011
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
has also been widely studied as a model reaction for ketone biore-
ductions in order to produce a specific chiral alcohol.7 The use of
Biocatalysis has become an increasingly valuable tool for syn-
thetic chemists. Bioreductions are attractive methods, mainly due
to high enantioselectivity, mild and safe reaction conditions and
lower environmental impact compared to conventional reactions
in organic chemistry.1–4 Chiral alcohols with additional functional
groups are important intermediates in the synthesis of enantiome-
rically pure pharmaceuticals and other important chemicals. For
example, optically active phenylethanols and their derivatives are
useful building blocks for the synthesis of complex molecules be-
cause the alcohol functionality can be easily transformed, without
racemization, into other useful functional groups.5 Chiral pheny-
lethanols [(R) or (S)] are interesting compounds with a number
of potential applications, particularly in the drug industry. These
alcohols are used as building blocks for the synthesis of bioactive
compounds, such as pharmaceuticals, agrochemicals, and natural
products.6 The reduction of acetophenone to chiral phenylethanol
microbial whole cells is advantageous because they contain the
necessary co-factors (NADH and NADPH) while the metabolic
pathways for their regeneration is inexpensive.8
Herein we report the use of 14 microorganisms (10 yeasts
strains and 4 filamentous fungi strains) for the asymmetric reduc-
tion of 4-bromoacetophenone (Fig. 1). Some reaction conditions
were studied by a fractional factorial design.
2. Results and discussion
2.1. Bioreduction of 4-bromoacetophenone using resting cells
At first, we tested 14 microorganisms (10 yeasts strains and 4
filamentous fungi strains) for the asymmetric reduction of 4-bro-
moacetophenone using whole cells after growing. The results are
shown in Table 1. All microorganisms were able to reduce 4-bro-
O
OH
OH
whole cell
up to 99% yield
up to 99% e.e.
microogranism
or
Br
Br
Br
2a
(S)-1-(4-bromophenyl)ethanol
2b
(R)-1-(4-bromophenyl)ethanol
1
Figure 1. Reduction of 4-bromoacetophenone.
⇑
Corresponding author.
E-mail address: joyce_benzaquem@yahoo.com.br (J.B. Ribeiro).
0957-4166/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2011.10.014

1764
R. de Oliveira Lopes et al. / Tetrahedron: Asymmetry 22 (2011) 1763–1766
Table 1
Table 2
Bioreduction of 4-bromoacetophenone
Monitoring of the reaction time of 4-bromomacetophenone reduction
Microorganism
Conversiona (%)
eea (%)
Configuration
Microorganism
Time’s reaction (h)
Conversiona (%)
eea (%)
K. marxianus
Hansenula sp.
Pichia sp.
G. candidum
Candida sp.
A. niger
M. ramannianus
T. harzianum
S. cerevisiae 40
S. cerevisiae 60
S. cerevisiae 80
P. chrysosporium
R. rubra
13.2
6
64.5
91.9
59.1
98.9
>99
98.5
6
3.5
0
13.2
96.1
99.3
51
84
(R)
(R)
(S)
(R)
(R)
(R)
(S)
(R)
(S)
(S)
—
A. niger
5
10
15
20
79.9
91.1
94.2
83.4
95.3
98.2
97.4
95.6
89.8
97.4
53.8
>99
20
G. candidum
R. minuta
R. rubra
5
10
15
20
95.7
98.9
99.1
98.2
81.2
>99
>99
98
98
93.2
>99
0
31.6
98.8
98.2
5
10
15
20
74.4
96.8
98.9
98
97.6
98.4
98
(R)
(S)
(S)
98
R. minuta
5
10
15
20
48.9
94.4
99.2
97.8
>99
>99
97.4
>99
a
Measured by GC with the chiral column BETA DEX 225.
a
moacetophenone, but some microorganisms, such as Kluyveromyces
marxianus, Hansenula sp., 3 strains of Saccharomyces cerevisiae, and
Phanerochaete chrysosporium showed low conversions (between 3%
and 13%). All reactions were conducted over 24 h.
Measured by GC with the chiral column BETA DEX 225.
design was employed to simplify the reaction conditions while
maintaining the high conversion and high enantiomeric excess.
Five variables were studied: biomass (Cell), pH, and concentrations
of 4-bromoacetophenone, glucose, and MgCl2. Three central point
replicates were accomplished. The experimental domain and re-
sults are shown in Table 3. Response variables were % conversion
and % ee after 20 h-incubation-periods.
Geotrichum candidum, Aspergillus niger, and Trichoderma
harzianum were able to produce the (R)-enantiomer with excellent
conversions and enantiomeric excess (ee), while Rhodotorula rubra
and Rhodotorula minuta gave the (S)-enantiomer. Numerous
approaches have been documented to improve the stereospecificity
of whole-cell bioconversions. Although recombinant microorgan-
isms have also been used with good results, wild-type strains, such
as the ones used herein, are generally preferred for industrial pro-
cesses mainly because of their robustness.9 Our results are compat-
ible with the wild-typestrains reported in the literature. Kurbanoglu
et al.10 described the reduction of 4-bromoacetophenone using
A. niger; they obtained a 100% conversion and higher than a 99% ee
of the (R)-enantiomer, within 48 h while we obtained a 98.9%
conversion and higher than a 99% ee within 24 h. In another work,
Kurbanoglu et al.5 described the reduction of the same substrate
but using immobilized cells of Rhodotorula glutinis; they obtained
the (S)-isomer with a 100% conversion and higher than a 99% ee.
Immobilization can influence the enantiomeric excess and conver-
sion level, as verified in previous studies with other substrates.8,9,11
We used two species of Rhodotorula genus, R. rubra and, R. minuta,
and obtained the (S)-isomer with a 99.3% conversion and a 98.2%
ee without immobilization. This suggests that some approaches
could enhance the conversion levels and enantiomeric excess.
According to Table 3, all experimental runs using G. candidum
led to an excess of (R)-4-bromophenylethanol. Conversions ob-
tained after 20 h varied from 40.4% to 99.6% while enantiomeric
excesses varied from 20.6% to higher than 99%. When minimum
substrate concentration was used (runs 1, 2, 5, and 6), higher than
98% conversion and high enantioselectivity were obtained (>99%
ee). Under these conditions, the conversions and enantioselectivi-
ties were similar to those obtained with G. candidum (Table 1).
A high conversion is important for a promising industrial pro-
cess, but a high enantioselectivity is essential for obtaining a chiral
building block. Furthermore, the conditions that furnished the
highest stereoselectivities also provided the highest conversions.
Hence only the enantiomeric excess was maximized in the exper-
imental domain studied. Data were evaluated by means of analysis
of variance (ANOVA) and the confidence level was set at 95%
(Table 3). Only the substrate concentration was the significant
parameter (p-value <0.05). Other variables were not significant
parameters in the experimental domain (p >0.05) and were not
considered in the model below (normalized variables):
2.2. Monitoring of reaction time
%ee ¼ 67:9 ꢀ 31:6 ꢁ S
where S = substrate concentration (4-bromoacetophenone).
ð1Þ
A. niger and G. candidum, which produce the (R)-enantiomer and
R. rubra and R. minuta, which produce the (S)-enantiomer, were
selected for monitoring the reaction time. It was possible to verify
high conversion and ee within 10 h with all microorganisms tested.
Thus, it was not necessary to keep the reaction for 24 h. We did not
observe inversion of the configuration. Nakamura et al.12 observed
that the use of G. candidum and acetophenone as the substrate gave
the (S)-isomer in the first 12 h, while a longer reaction time (40 h)
led to the formation of the (R)-enantiomer (Table 2).
The fit of the model is expressed by R2, which was calculated to
be 0.88. This indicates that the model explains 88% of the variabil-
ity in the data. The adjusted R2 statistic was (0.87). Analysis
showed that the curvature was not important to the model (p-
value = 0.89) and that the linear model obtained to describe the
enantiomeric excess was suitable for the observed data.
Since the substrate concentration showed a negative effect on
enantioselectivity while the other variables had no effect on the re-
sponse, the conditions selected for 4-bromoacetophenone reduc-
tion to 4-bromophenylethanol by G. candidum were: substrate,
1 g/L; biomass, 4.0 gdw/L; glucose, 30 g/L; pH, 5.
2.3. Fractional factorial design to simplify reaction conditions
to obtain (R)-4-bromophenylethanol
Fractional factorial designs are frequently used to evaluate the
reaction conditions of bioconversions and thus improve the meth-
ods, conversions, and ee8,9,13 G. candidum was considered to be the
best biocatalyst for obtaining (R)-4-bromophenylethanol, with a
98% ee at high yields (98%) after 20 h. A 25-2 fractional factorial
Experiments (performed in triplicate) were conduced to verify
conversion level and enantiomeric excess under the reaction con-
ditions described as above. After 20 h, the product was obtained
in a 98.9% conversion (SD = 0.5%) and greater than a 99% ee. This
result matched the model prediction.

R. de Oliveira Lopes et al. / Tetrahedron: Asymmetry 22 (2011) 1763–1766
1765
Table 3
Experimental design to simplify the reaction conditions of 4-bromoacetophenone reduction to give (R)-4-bromophenylethanol by G. candidum
Run
Factorsa
pH
Responses (20 h)
Conversionb (%)
Cell (gdw/L)
S (g/L)
Glucose (g/L)
MgCl2 (g/L)
eeb (%)
1
2
3
4
5
6
7
8
4
4
4
1
1
3
3
1
1
3
3
4
6
4
6
4
6
4
6
5
5
5
30
50
50
30
30
50
50
30
40
40
40
0
1
1
0
1
0
0
1
99.4
98.7
40.4
63.2
99.4
99.6
94.5
80
>99
>99
30.4
20.6
>99
>99
45.6
46.2
76.6
49
4
12
12
12
12
8
9
10
11
1.5
1.5
1.5
0.5
0.5
0.5
89.9
92
89.4
8
8
82
Incubation: 30 °C, 150 rpm, 20 h.
a
Cell (biomass concentration–g dry weight/L); S (substrate–4-bromoacetophenone).
Measured by GC with the chiral column BETA DEX 225.
b
For experimental designs, the fit of the model is considered
between both variables were significant parameters (p-value
<0.05). Other variables were not significant parameters in the
experimental domain (p >0.05) and thus were not considered in
the model below (normalized variables):
acceptable when R2 is greater than 0.7 for biological data and
greater than or equal to 0.8 for data of chemical origin.14
2.4. Fractional factorial design to simplify the reaction
conditions to obtain (S)-4-bromophenylethanol
%Conversion ¼ 55:1 þ 18:3 ꢁ Cell ꢀ 16:7 ꢁ S ꢀ 6:3 ꢁ Cell ꢁ S ð2Þ
where Cell = biomass concentration; S = substrate concentration (4-
bromoacetophenone).
R. rubra was chosen to obtain the other isomer, due to the high
conversion of 4-bromoacetophenone to (S)-4-bromophenylethanol
(a 96% yield with a 98.8% ee). The same variables studied in the
reduction of 4-Br-acetophenone catalyzed by G. candidum were
also investigated by a 25-2 fractional factorial design in the reduc-
tion catalyzed by R. rubra. Three central point replicates were
accomplished. The variables, experimental domain, and results
are shown in Table 4. Response variables were % conversion and
% ee after a 20 h-incubation-period.
According to Table 4, all experimental runs using R. rubra led to
up to 99% ee of (S)-4-bromophenylethanol with conversions ob-
tained after 20 h ranging from 26.4% to 97.4%. Only the conversion
level should be maximized in the experimental domain studied.
The highest conversions (P96.9%) were achieved in the experi-
mental runs 5 and 6, when maximum biomass concentration and
minimum substrate concentration were used. Under both condi-
tions, conversion and enantioselectivity were similar to the results
in Table 1 with the same amount of substrate and R. rubra biomass.
Data were evaluated by means of analysis of variance (ANOVA)
and are shown in Table 4. The confidence level was set at 95%. Bio-
mass concentration, substrate concentration, and the interaction
The coefficient of determination (R2) was 0.99, which means
that the model explains 99% of variability in response. The adjusted
R2 statistic was also high (0.98). Analysis showed that the curva-
ture was not important to the model (p-value = 0.17) and that
the linear model obtained to describe the enantiomeric excess
was suitable for the observed data.
Biomass had a positive effect on conversion within the experi-
mental domain. Quite the contrary, substrate and the interaction
between biomass and substrate concentration showed negative ef-
fects. The addition of MgCl2 had no effect on the reduction of 4-
bromoacetophenone catalyzed by R. rubra and G. candidum; this
was also observed in previous studies on the reduction of ethyl
benzoylacetate by R. rubra8 and ethyl 4-chloroacetoacetate by K.
marxianus.9 Variation of glucose concentration had no significative
effect on response and it was used at a minimum level. Based on
these results, the following conditions were selected for the
bioreduction of 4-bromoacetophenone with R. rubra: biomass,
12 gdw/L; substrate, 1 g/L; glucose 30 g/L; pH 5.
Experiments were performed in triplicate in order to verify
conversion level and enantiomeric excess under the reaction
Table 4
Experimental design to simplify the reaction conditions for 4-bromoacetophenone reduction to (S)-4-bromophenylethanol by R. rubra
Run
Factorsa
pH
Responses (20 h)
Conversionb (%)
Cell (gdw/L)
S (g/L)
Glucose (g/L)
MgCl2 (g/L)
eeb (%)
1
2
3
4
5
6
7
8
4
4
4
1
1
3
3
1
1
3
3
4
6
4
6
4
6
4
6
5
5
5
30
50
50
30
30
50
50
30
40
40
40
0
1
1
0
1
0
0
1
47.2
48.6
26.4
27.9
97.4
96.9
49.5
52.8
55.4
51.9
52.2
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
4
12
12
12
12
8
9
10
11
1.5
1.5
1.5
0.5
0.5
0.5
8
8
Incubation: 30 °C, 150 rpm, 20 h.
a
Cell (biomass concentration–g dry weight/L); S (substrate–4-bromoacetophenone).
Measured by GC with the chiral column BETA DEX 225.
b

1766
R. de Oliveira Lopes et al. / Tetrahedron: Asymmetry 22 (2011) 1763–1766
conditions described above. After 20 h, the product was obtained in
a 97.6% conversion (SD = 0.3%) and a 98.8% ee (SD = 0.2%). These re-
sults matched the prediction model.
phase was dried (anhydrous Na2SO4), filtered, and concentrated
under vacuum. Products were analyzed by (chiral) gas chromatog-
raphy (GC), on the column Beta DEX 225 (30 m ꢁ 0.25 mm ꢁ
0.25 lm), at 130 °C (35 min). We confirmed the configuration by
using the (R)-isomer purchased from Aldrich.
3. Conclusions
In order to study the reaction conditions on the reduction of
4-bromoacetophenone by G. candidum and R. rubra, 25,6 fractional
factorial designs13 were used to evaluate five variables in 8 runs with
3 replicates of the central point. Variables and domain were:
Biomass concentration (Cell): 4–12 g dry weight/L; substrate (4-
bromoacetophenone): 1–3 g/L; pH, 4–6; glucose: 30–50 g/L; MgCl2:
0–1 g/L. Response variables were: % conversion and % ee Reactions
were carried out in 500 mL cotton plugged Erlenmeyer flasks
containing 50 mL of medium for 20 h at 30 °C and 150 rpm. After
that period, the medium was treated as described above. Statistical
analyses were performed using Statistica 6.0 (Statsoft Inc., Tulsa, OK,
USA).
A screening study of some wild fungi strains has been carried out
with different enantioselectivities in the reduction of 4-bromoace-
tophenone being obtained. The yeast G. candidum highlighted in
the production of (R)-4-bromophenylethanol and R. rubra were
selected for obtaining the other isomer, (S)-4-bromophenylethanol.
In the reduction catalyzed by G. candidum, the reaction conditions
selected by an experimental design were simpler than those used
in the initial studies and led to excellent conversions (98.9%) and
enantioselectivities (>99% ee) for the (R)-isomer. Biomass, glucose,
MgCl2, and pH were not significant parameters, which are a good
characteristic for industrial processes, while allowing some
variation of the parameters during process without affecting the
enantiomeric excess. In the 4-bromoacetophenone reduction with
R. rubra, the reaction conditions selected by the experimental design
led to a 97.6% conversion and a 98.8% ee for the (S)-4-bromopheny-
lethanol isomer. Further optimization studies could improve these
results.
Acknowledgment
Financial support from CAPES and CNPq–BRAZIL is
acknowledged.
4. Experimental
References
1. Nakamura, K.; Yamanaka, R.; Matsuda, T.; Harada, T. Tetrahedron: Asymmetry
2003, 14, 2659–2681.
2. Kataoka, M.; Kita, K.; Wada, M.; Yasohara, Y.; Hasegawa, J.; Shimizu, S. Appl.
Microbiol. Biotechnol. 2003, 62, 437–445.
3. Yang, Y.; Zhu, D.; Piegat, T. J.; Hua, L. Tetrahedron: Asymmetry 2007, 18, 1799–
1803.
4. Houng, J. Y.; Liao, J. H.; Wu, J. Y.; Shen, S. C.; Hsu, H. F. Process Biochem. 2007, 42,
1–7.
Microorganisms, media, growth conditions, and biotransforma-
tion. Hansenula sp., K. marxianus, Candida sp., Pichia sp.,
G. candidum, 3 strains of S. cerevisiae, R. rubra, R. minuta and
filamentous fungi, A. niger, T. harzianum, Mucor ramannianus, and
P. chrysosporium, belong to the collection of the ‘Departamento de
Engenharia Bioquímica, Escola de Química, UFRJ’ (Cidade Univer-
sitária, CT Bloco E, Rio de Janeiro, Brazil, e-mail selma@eq.ufrj.br)
and are freely available upon request. Cells were allowed to grow
for 48 h, under 150 rpm and 30 °C in a medium containing 1% glu-
cose, 0.5% yeast extract, 0.5% peptone, 0.1% (NH4)2SO4, and 0.1%
MgSO4ꢂ7H2O. After that period, they were harvested by centrifuga-
tion, re-suspended in water, and used for the reaction. After centri-
fugation, the cells (12 g/L, dried weight) were added to the
reduction’s medium containing: glucose (5%), MgCl2 (0.1%) in a final
volume of 50 mL. After 30 min of addition of the microorganisms,
the substrate (50 mg diluted in 1 mL of ethanol 96%) was added
to the medium. The reaction was carried out in 500 mL cotton
plugged Erlenmeyer flasks for 24 h at 30 °C and 150 rpm. After
24 h, the medium was centrifuged again to separate the cells and
the liquid phase was extracted with ethyl acetate. The organic
5. Kurbanoglu, E. B.; Zilbeyaz, K.; Ozdal, M.; Taskin, M.; Kurbanoglu, N. I.
Bioresour. Technol. 2010, 101, 3825–3829.
6. Zilbeyaz, K.; Kurbanoglu, E. B. Bioresour. Technol. 2008, 99, 1549–1552.
7. Abas, F. Z.; Uzir, M. H.; Zahar, M. H. M. J. Appl. Sci. 2010, 10, 3289–3294.
8. Ramos, A. S.; Ribeiro, J. B.; Vazquez, L.; Fiaux, S. B.; Leite, S. G. F.; Ramos, M. C. K.
V.; Aquino Neto, F. R.; Antunes, O. A. C. Tetrahedron Lett. 2009, 50, 7362–7364.
9. Ribeiro, J. B.; Ramos, A. S.; Fiaux, S. B.; Leite, S. G. F.; Ramos, M. C. K. V.; Aquino
Neto, F. R.; Antunes, O. A. C. Tetrahedron: Asymmetry 2009, 20, 2263–2266.
10. Kurbanoglu, E. B.; Zilbeyaz, K.; Kurbanoglu, N. I.; Kilic, H. Tetrahedron:
Asymmetry 2007, 18, 1159–1162.
11. Ribeiro, J. B.; Ramos, M. C. K. V.; Aquino Neto, F. R.; Leite, S. G. F.; Antunes, O. A.
C. J. Mol. Catalysis B: Enzimatic. 2003, 24–25, 121–124.
12. Nakamura, K.; Matsuda, T.; Ohno, A. Tetrahedron: Asymmetry 1996, 7, 3021–
3024.
13. Logothetis, N. Managing for Total Quality—From Deming to Taguchi and SPC;
Prentice Hall International Ltd: London, 1992.
14. Lundstedt, T.; Seifert, E.; Abramo, L.; Thelin, B.; Nyström, A.; Pettersen, J.;
Bergman, R. Chemometrics Intelligent Lab. Systems 1998, 42, 3–40.