Asymmetric aldol reactions catalyzed by the promiscuous aldo-ketoreductase enzyme

Article abstract of DOI:10.1016/j.tetlet.2014.09.012

The promiscuous aldo-ketoreductase (AKR) enzyme is used as a sustainable biocatalyst for the first time to catalyze asymmetric aldol reactions in aqueous medium. The reactions between aromatic aldehydes and cyclic/acyclic ketones give the corresponding pr

Full text of DOI:10.1016/j.tetlet.2014.09.012

Tetrahedron Letters 55 (2014) 6303–6306
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Asymmetric aldol reactions catalyzed by the promiscuous
aldo–ketoreductase enzyme
Saadi Bayat a,b, Emilia Abd Malek a,b, Normi M. Yahaya , Abu Bakar Salleh , Bimo Ario Tejo ,
Mohd Basyaruddin Abdul Rahman
a
a
a
d
a,b,c,
⇑
Enzyme and Microbial Technology Research Centre, Universiti Putra Malaysia, 43400 Serdang, Selangor Darul Ehsan, Malaysia
Department of Chemistry, Faculty of Science, Universiti Putra Malaysia, 43400 Serdang, Selangor Darul Ehsan, Malaysia
Structural and Synthetic Biology Research Centre, Malaysia Genome Institute, Jalan Bangi, 43000 Kajang, Selangor Darul Ehsan, Malaysia
Center for Infectious Diseases Research, Surya University, Jl. Scientia, Boulevard Blok U/7, Gading Serpong, Tangerang 15810, Banten, Indonesia
b
c
d
a r t i c l e i n f o
a b s t r a c t
Article history:
The promiscuous aldo–ketoreductase (AKR) enzyme is used as a sustainable biocatalyst for the first time
to catalyze asymmetric aldol reactions in aqueous medium. The reactions between aromatic aldehydes
and cyclic/acyclic ketones give the corresponding products in moderate yields and enantioselectivities
in the presence of water. The influence of solvents, the mole ratio of substrates, and enzyme concentra-
tion are investigated. The mechanism of the AKR1A1-catalyzed aldol reaction is also discussed.
Ó 2014 Elsevier Ltd. All rights reserved.
Received 21 April 2014
Revised 23 August 2014
Accepted 3 September 2014
Available online 16 September 2014
Keywords:
Promiscuous enzymes
Aldo–ketoreductase
Asymmetric
Aldol reaction
Biocatalysis has emerged as an elegant and green tool for
modern organic synthesis due to its high efficiency, good
selectivity, and environmental acceptability. Although an enzyme
reported the first aldol reaction catalyzed by CAL-B (lipase from
Candida antarctica).
2
1
1
In this study, we focused on an aldo–ketoreductase (AKR)
enzyme in order to investigate its catalytic activity in asymmetric
aldol reactions. To the best of our knowledge, aldo–ketoreductase
has not been reported as an enzyme for catalyzing carbon–carbon
bond-forming reactions. We therefore used commercially available
AKR1A1 as an enzyme to catalyze the aldol reaction. There are
important tetrad amino acids in the active site of AKR similar to
is capable of catalyzing a specific reaction effectively, some unex-
pected experimental results have indicated that many enzymes
are catalytically promiscuous.2 Promiscuous enzymes have flour-
ished as popular asymmetric biocatalysts for a range of organic
reactions in recent years.3 These types of enzymes can catalyze
different chemical transformations of natural and non-natural sub-
–5
6
–8
22
strates.
Promiscuous enzymes have been used to catalyze the
the aldolase enzyme. The positions of Asp 45, Lys80, Tyr50, and
formation of carbon–carbon and carbon–heteroatom bonds via
classic and widely used organic reactions.9 The aldol reaction is a
fundamental organic reaction for the construction of C–C bonds,
which has been seldom mentioned in published studies of promis-
His113 in the active site pocket are somehow able to impart cata-
lytic activity. Therefore, this similarity encouraged us to choose
AKR as an enzyme for investigation of its catalytic activity in aldol
reactions.
1
0–13
cuous biocatalysts.
Generally, aldolases (class I and II) and
Initially, an aldol experiment was carried out in order to
investigate the catalytic activity of the promiscuous AKR1A1
enzyme, focusing on its ability to catalyze the reaction between
cyclohexanone and 4-nitrobenzaldehyde. The reaction was per-
formed in a mixture of i-PrOH/phosphate–sodium buffer (1:1)
using p-nitrobenzaldehyde (30 mg, 0.198 mmol, 1 equiv), cyclo-
hexanone (23.3 mg, 0.99 mmol, 5 equiv), and AKR1A1 (10 mg).
Generally, a higher pH tends to give the product in a better yield,
but slightly acidic conditions were beneficial for improving the ste-
aldolase antibodies (38C2 and 33F12) are the most common
14
enzymes used in the catalysis of aldol reactions. Promiscuous
enzymes have frequently catalyzed Michael, Mannich and Mark-
ovnikov additions as well as aldol reactions.1
5–19
Asymmetric aldol
reactions between wet acetone and aromatic aldehydes catalyzed
by a lipase enzyme were reported by Wang.20 In 2003, Berglund
2
2
reoselectivity. The reactions catalyzed by AKR1A1 were con-
ducted at different pHs (4.0 to 8.0). The results are summarized
in Table 1. As expected, the pH had a significant influence on the
⇑
Corresponding author. Tel.: +60 03 8946 6798, +60 03 8943 5380.
E-mail address: basya@upm.edu.my (M.B. Abdul Rahman).
http://dx.doi.org/10.1016/j.tetlet.2014.09.012
040-4039/Ó 2014 Elsevier Ltd. All rights reserved.
0

6
304
S. Bayat et al. / Tetrahedron Letters 55 (2014) 6303–6306
yield and stereoselectivity. For pH = 5.0 and 5.5, the corresponding
aldol product was obtained in moderate yield (entries 3 and 4, 51%
and 60%) and moderate enantioselectivity (39% and 40% ee). At
pH = 8.0, the reaction became almost a base-catalyzed process,
therefore, extremely low enantioselectivity (entry 9, 4% ee) and a
higher yield (95%) were obtained. Control experiments showed
that in the absence of AKR1A1, at pH = 5.5, no reaction took place
under these conditions (entry 10). Finally, a pH of 5.5 was chosen
as the optimum pH in terms of the efficiency and selectivity of
the AKR1A1-catalyzed process.
To further optimize the reaction conditions, the effect of the
mole ratio of the substrates on the AKR1A1-catalyzed aldol reac-
tion was investigated. The results indicated that this factor had a
very significant impact on the yield and ee, but changes in the dia-
stereoselectivity were less obvious (Table 2). Consequently, the
mole ratio of substrates, 4-nitrobenzaldehyde/cyclohexanone = 1:5
Table 2
a
Effect of the mole ratio of the substrates on the model aldol reaction
Entry
Mole ratiob
Yieldc (%)
drd (anti:syn)
eed (%) (anti)
1
2
3
4
5
1:1
1:2
1:3
1:5
1:10
15
55
60
68
66
82:18
84:16
85:15
85:15
86:14
20
33
40
43
43
a
Reaction conditions: 4-nitrobenzaldehyde (0.1 mmol), cyclohexanone (0.1, 0.2,
.3, 0.5, 1.0 mmol), AKR1A1 (10 mg), phosphate–sodium buffer/i-PrOH (1:1, 1.0 mL,
0
pH = 5.5), rt, 24 h.
b
Mole ratio of 4-nitrobenzaldehyde/cyclohexanone.
Isolated yield.
ee and dr were determined by HPLC using a chiral OD-H column.
c
d
Table 3
(
entry 4), was chosen as the optimum ratio for further exploration.
The influence of the catalyst loading on the aldol reaction
Effect of enzyme loading on the aldol reaction
Cat. loadinga (mg)
Yieldb (%)
drc (anti:syn)
eec (%), (anti)
Entry
between 4-nitrobenzaldehyde and cyclohexanone was next inves-
tigated. The results showed that an increase in the AKR1A1 loading
1
2
3
4
5
6
7
0
2
5
10
15
20
30
NR
10
45
68
68
69
68
—
—
65:45
75:25
85:15
85:15
84:16
84:16
33
39
43
43
43
43
(
from 2 mg to 20 mg) led to a rise in the yield (from 10% to 69%),
with a slight increase in the selectivity (from 65:45 to 85:15 dr,
and from 39% to 43% ee) (Table 3). Thus, an enzyme concentration
of 10 mg/mL was chosen as the optimum for further investigation.
The reaction medium has been recognized to be one of the most
important factors influencing enzymatic reactions. Thus, the aldol
reaction in various solvents in the presence of a few drops of 1 M
benzoic acid (BA) as an additive for adjustment of the pH was
investigated. The results are shown in Table 4.
a
Reaction conditions: 4-nitrobenzaldehyde (0.1 mmol), cyclohexanone
(
0.5 mmol), AKR1A1 (0, 2, 5, 10, 15, 20, 30 mg), phosphate–sodium buffer/i-PrOH
(
1:1, 1.0 mL, pH = 5.5), rt, 24 h.
b
Isolated yield.
c
ee and dr were determined by HPLC using a chiral OD-H column.
The results clearly indicated that the reaction medium signifi-
cantly influenced the catalytic activity and stereoselectivity of
AKR1A1. Generally, the aldol reaction in high polarity solvents
such as DMF and NMP gave higher yields (entries 1 and 2) than
Table 4
Effect of the solvent on the aldol reaction between 4-nitrobenzaldehyde and
cyclohexanone catalyzed by AKR1A1
3
those in low polarity solvents such as CHCl (entry 4). Interestingly,
a
CHCl and i-PrOH produced very poor enantiomeric excesses and
3
Entry
Solvent
Yieldb
eec (%)
drd (anti:syn)
yields when they were used separately as the solvent. However,
higher enantioselectivities and yields were obtained by adding an
equal volume of H O to these two solvents (entries 6 and 7).
2
1
2
3
4
5
6
7
NMP
DMF
MeCN
50
50
50
20
25
65
52
39
39
34
20
26
43
37
82:18
82:18
86:14
73:27
85:15
91:9
CHCl
i-PrOH
CHCl :H
i-PrOH/H
3
Table 1
3
2
O (1:1)
O (1:1)
Effect of pH on the aldol reaction catalyzed by AKR1A1 in phosphate–sodium buffer/i-
2
82:18
PrOHa
a
Reaction conditions: 4-nitrobenzaldehyde (0.1 mmol), cyclohexanone
(
0.5 mmol), AKR1A1 (10 mg), solvent (1.0 mL, pH = 5.5), rt, 24 h.
Isolated yield.
ee and dr were determined by HPLC using a chiral OD-H column.
b
c,d
Among the surveyed solvents, the best result was obtained with
a mixture of CHCl :H O (1:1, entry 6). In order to pursue asymmet-
ric aldol reactions, CHCl :H O (1:1) was chosen as the reaction
3
2
3
2
medium for further investigation.
Entry
pH
Yieldb (%)
eec (%)
Next, the substrate scope and limitations were investigated
using different aromatic aldehydes and cyclohexanones in the
AKR1A1-catalyzed asymmetric aldol reaction. The results are given
in Table 5. It can be seen that a wide range of aromatic aldehydes
reacted with cyclic and acyclic ketones under the optimized condi-
tions. The electronic and steric effects of the substituents on the
aromatic aldehydes were also investigated.
In general, the aldol products were obtained in higher yields
when the ketone was reacted with aromatic aldehydes bearing
an electron-withdrawing substituent (entries 1–7). In contrast,
only a trace amount of product was obtained when 4-methoxy-
benzaldehyde bearing an electron-donating group was used (entry
8). This can be explained by the fact that electron-withdrawing
groups enhance the electrophilicity of the carbonyl group, which
1
2
3
4
5
6
7
8
9
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
35
40
51
60
68
84
90
92
15
20
39
40
22
16
11
5
8.0
95
Trace
4
0
1
0
No enzyme (pH = 5.5)
a
Reaction conditions: p-nitrobenzaldehyde (0.198 mmol, 1 equiv), cyclohexa-
none (0.59 mmol, 3 equiv), AKR1A1 (10 mg), phosphate–sodium buffer/i-PrOH (1:1,
mL, pH = 4.0–8.0), rt, 24 h.
1
b
Isolated yield after silica gel column chromatography.
ee was determined by HPLC using a chiral OD-H column.
c

S. Bayat et al. / Tetrahedron Letters 55 (2014) 6303–6306
6305
Table 5
Investigation of the reactant scope of the AKR1A1-catalyzed asymmetric aldol
a
reaction in CHCl
3
:H
2
O
Entry
Product
Yieldb (%)
65
eec (%)
43
drc (%)
91:9
OH
O
1
O N
2
NO2 OH
O
2
3
55
54
58
40
96:4
OH
O
82:18
Cl
OH
O
4
5
65
55
60
45
94:6
93:7
F3C
OH
OH
O
O
Br
O
6
7
8
59
40
43
—
75:25
91:9
—
Scheme 1. Proposed mechanism for the stereoselective aldol reaction catalyzed by
AKR1A1.
NC
Cl OH
51
site that the two components of the aldol condensation might
adopt.
OH
O
In conclusion, we have described the first AKR-catalyzed aldol
reactions of a wide range of aromatic aldehydes with cyclohexa-
Trace
28
none and acetone in moderate to good yields. The influence of
MeO
the reaction conditions including the solvent, water content, and
enzyme loading was investigated. Control experiments were also
conducted to demonstrate the specific catalysis of AKR. This
AKR1A1-catalyzed aldol reaction provides a novel case of catalytic
promiscuity which widens the applicability of AKRs in organic syn-
thesis. This case of biocatalytic promiscuity not only broadens the
application of AKR to new chemical transformations, but also
might be expanded into a potentially valuable method for organic
synthesis.
OH
O
9
68
22
—
O N
2
a
Reaction conditions: 4-nitrobenzaldehyde (0.1 mmol), cyclohexanone or ace-
tone (0.5 mmol), AKR1A1 (10 mg), CHCl
few drops), rt, 24 h.
3 2
:H O (1:1, 1.0 mL, pH = 5.5), benzoic acid (a
b
Isolated yield.
c
ee and dr were determined by HPLC using a chiral OD-H column.
Acknowledgment
We gratefully acknowledge the Ministry of Higher Education
MOHE), Malaysia for financial support through the Grant No.
524292.
facilitates the reaction to form a possible enamine intermediate,
while electron-donating groups decrease the electrophilicity.
AKR1A1 catalyzed the aldol addition of cyclohexanone with
aromatic aldehydes to give the desired products in reasonable
yields (up to 65%), enantioselectivities (up to 60% ee), and moder-
ate to excellent diastereoselectivities (up to 96:4, anti:syn). 4-(Tri-
fluoromethyl)benzaldehyde (entry 4) gave the best result. The
reaction of acetone with 4-nitrobenzaldehyde gave a moderate
yield (68%) and poor enantioselectivity (22%) (Table 5, entry 9).
A possible mechanism for the AKR1A1-catalyzed aldol reaction
is illustrated in Scheme 1. We speculate that the ketone compo-
nent may be activated by the formation of an enamine with the
amino group of lysine with removal of water. A proton is removed
(
5
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.tetlet.2014.09.
12.
0
References and notes
1
.
Xie, B.-H.; Li, W.; Liu, Y.; Li, H.-H.; Guan, Z.; He, Y.-H. Tetrahedron 2012, 68,
3160.
2. Hasnaoui, D. G.; Majeri c´ , E. M.; Lutje, S. J. H.; Hauer, B.; Janssen, D. B.
by the basic ring nitrogen of the imidazole of an apposite histi-
dine.23 We also speculate that the carbonyl of the aldehyde compo-
ChemBioChem 2008, 9, 1048.
Carboni-Oerlemans, C.; Domínguez de María, P.; Tuin, B.; Bargeman, G.; van der
Meer, A.; van Gemert, R. J. Biotechnol. 2006, 126, 140.
3.
nent is activated by interaction with the phenolic hydroxyl of
tyrosine. A full discussion of the feasibility of these suggested
interactions, and how they lead to the observed stereoselectivity,
4. Busto, E.; Gotor-Fernández, V.; Gotor, V. Chem. Soc. Rev. 2010, 39, 4504.
5
6
7
.
.
.
Busto, E.; Gotor-Fernández, V.; Gotor, V. Org. Process Res. Dev. 2010, 15, 236.
Chen, Y.-L.; Li, W.; Liu, Y.; Guan, Z.; He, Y.-H. J. Mol. Catal. B: Enzym. 2013, 87, 83.
Taglieber, A.; Höbenreich, H.; Carballeira, J. D.; Mondière, R. J.; Reetz, M. T.
Angew. Chem., Int. Ed. 2007, 46, 8597.
will depend on a detailed analysis of the three-dimensional struc-
ture of the active site pocket2
4–27
and the orientation within that

6
306
S. Bayat et al. / Tetrahedron Letters 55 (2014) 6303–6306
23. Tarle, I.; Borhani, D.; Wilson, D.; Quiocho, F.; Petrash, J. J. Biol. Chem. 1993, 268,
8
9
.
.
Kazlauskas, R. J. Curr. Opin. Chem. Biol. 2005, 9, 195.
Babtie, A.; Tokuriki, N.; Hollfelder, F. Curr. Opin. Chem. Biol. 2010, 14, 200.
25687.
1
0. Branneby, C.; Carlqvist, P.; Hult, K.; Brinck, T.; Berglund, P. J. Mol. Catal. B:
Enzym. 2004, 31, 123.
1. List, B.; Pojarliev, P.; Castello, C. Org. Lett. 2001, 3, 573.
2. Tan, R.; Li, C.; Luo, J.; Kong, Y.; Zheng, W.; Yin, D. J. Catal. 2013, 298, 138.
3. Li, C.; Zhou, Y.-J.; Wang, N.; Feng, X.-W.; Li, K.; Yu, X.-Q. J. Biotechnol. 2010, 150,
24. Xie, Z.-B.; Wang, N.; Jiang, G.-F.; Yu, X.-Q. Tetrahedron Lett. 2013, 54, 945.
25. Khurana, S.; Powers, D. B.; Anderson, S.; Blaber, M. Proc. Natl. Acad. Sci. U.S.A.
1998, 95, 6768.
26. Khurana, S.; Sanli, G.; Powers, D. B.; Anderson, S.; Blaber, M. Proteins Struct.
Funct. Bioinf. 2000, 39, 68.
1
1
1
5
39.
27. Hur, E.; Wilson, D. K. Chem. Biol. Interact. 2001, 130, 527.
28. Typical procedure for the aldol reaction catalyzed by AKR1A1: The initial effort
was performed in i-PrOH/sodium phosphate buffer using p-nitrobenzaldehyde
(30 mg, 0.198 mmol, 1 equiv), cyclohexanone (23.3 mg, 0.99 mmol, 5 equiv),
and AKR1A1 (10 mg) as a test reaction. The resulting mixture was stirred at
ambient temperature for the specified time, and the progress monitored by
thin-layer chromatography (TLC). The reaction was terminated by filtering the
1
4. Wada, M.; Hsu, C.-C.; Franke, D.; Mitchell, M.; Heine, A.; Wilson, I.; Wong, C.-H.
Bioorg. Med. Chem. 2003, 11, 2091.
5. Svedendahl, M.; Hult, K.; Berglund, P. J. Am. Chem. Soc. 2005, 127, 17988.
6. Carlqvist, P.; Svedendahl, M.; Branneby, C.; Hult, K.; Brinck, T.; Berglund, P.
ChemBioChem 2005, 6, 331.
7. Li, K.; He, T.; Li, C.; Feng, X.-W.; Wang, N.; Yu, X.-Q. Green Chem. 2009, 11, 777.
8. Torre, O.; Alfonso, I.; Gotor, V. Chem. Commun. 2004, 1724.
9. Guan, Z.; Fu, J.-P.; He, Y.-H. Tetrahedron Lett. 2012, 53, 4959.
0. Li, C.; Feng, X.-W.; Wang, N.; Zhou, Y.-J.; Yu, X.-Q. Green Chem. 2008, 10, 616.
1. Branneby, C.; Carlqvist, P.; Magnusson, A.; Hult, K.; Brinck, T.; Berglund, P. J.
Am. Chem. Soc. 2003, 125, 874–875.
1
1
1
1
1
2
2
enzyme. The filtrate was diluted with EtOAc (3 mL) and washed with H
(3 mL Â 2). The aqueous phase was back-extracted with EtOAc (3 mL Â 2). The
4
combined organic phase was washed with H O (4 mL), dried over the Na SO ,
2
O
2
2
and concentrated in vacuo. The crude product was purified by flash column
chromatography (EtOAc/n-hexane, 1:3).
2
2. Wu, W.-B.; Xu, J.-M.; Wu, Q.; Lv, D.-S.; Lin, X.-F. Adv. Synth. Catal. 2006, 348,
4
87.