Investigation of lipase-catalyzed Michael-type carbon-carbon bond formations

Article abstract of DOI:10.1016/j.tet.2009.05.042

Conjugate additions of carbon nucleophiles to appropriate acceptor molecules were investigated with respect to the synthetic potential and stereochemistry of the products. Reactions of short-chain α,β-unsaturated ketones and mono-substituted nitroalkenes

Full text of DOI:10.1016/j.tet.2009.05.042

Tetrahedron 65 (2009) 5663–5668
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Investigation of lipase-catalyzed Michael-type carbon–carbon bond formations
a
a
a
a
b
´
Gernot A. Strohmeier , Tanja Sovic , Georg Steinkellner , Franz S. Hartner , Aleksandra Andryushkova ,
Thomas Purkarthofer a, Anton Glieder a b, Karl Gruber a, Herfried Griengl a
,
,
*
a Research Centre Applied Biocatalysis, Petersgasse 14, 8010 Graz, Austria
b Institute of Molecular Biotechnology, Graz University of Technology, Petersgasse 14, 8010 Graz, Austria
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 6 April 2009
Received in revised form 15 May 2009
Accepted 18 May 2009
Available online 23 May 2009
Conjugate additions of carbon nucleophiles to appropriate acceptor molecules were investigated with
respect to the synthetic potential and stereochemistry of the products. Reactions of short-chain a,b-
unsaturated ketones and mono-substituted nitroalkenes with CH-acidic carboxylic ester derivatives were
catalyzed by various immobilized lipases. Using methyl nitroacetate complete conversion with methyl
vinyl ketone and trans-b-nitrostyrene was obtained. The reactions proceeded without enantioselectivity.
Evidence for enzyme catalysis is provided by the observation that after denaturation of Candida ant-
arctica lipase B or inhibition no reaction took place. Docking studies with the chiral addition product
methyl 2-methyl-2-nitro-5-oxohexanoate did not reveal any specific binding mode for this reaction
product, which would have been the requirement for stereoselective additions. These results support the
experimental findings that the conjugate addition takes place without enantiopreference.
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Bioorganic chemistry
Enzyme catalysis
Lipases
Michael addition
Molecular modelling
1. Introduction
the data from literature it was apparent that only highly activated
donor molecules like acetylacetone and ethyl acetoacetate were able
Conjugate additions, such as the Michael addition, represent
fundamental reactions in synthetic organic chemistry to produce
carbon–carbon and carbon–heteroatom bonds. With this type of
reaction up to three chiral centres can be generated in a single event
with perfect atom-economy. Classical Michael additions are per-
formed under acid/base catalysis.1 Moreover, several asymmetric
procedures using organometallic catalysts or organocatalysts have
emerged in the literature.2
to react. Most of the products reported did not possess a chirality
centre and therefore no conclusions on the stereochemical course of
the reactions could be drawn. To elucidate a potential enzyme-cat-
alyzed Michael addition we decided to investigate this biocatalytic
transformation.
2. Results and discussion
Contrary to the importance for synthetic chemistry in nature
enzyme-catalyzed Michael-type reactions have not been as well
studied yet. Examples are transformations by enoate reductases,3
some ligases4 and a few other enzymes.5 Non-natural enzymatic
reactions6 for conjugate additions were reported with N-ethyl-mal-
2.1. Michael additions of methyl nitroacetate (2a) and
of methyl 2-nitropropionate (2b) to 3-buten-2-one (1)
At the beginning of our investigations it was necessary to ascer-
tain conditions under which enzyme-catalyzed Michael additions
were least affected by non-enzymatic acid/base catalysis. It quickly
turned out that buffered aqueous solutions were not applicable due
to a fairly strong base-induced activation of reactants thus leading to
dominant background reactions. Moving from aqueous conditions to
organic solvents required the careful selection of immobilization
techniques in order to avoid mass transfer limitations associated
with suspended powders or the use of large amounts of enzymes.
We therefore investigated several solid supports, which had already
been used for the immobilization of enzymes by deposition or ad-
sorption. Deposition onto Celite15 was quickly ruled out because of
the strong activity to catalyze our model reaction, the Michael ad-
dition of methyl nitroacetate (2a) to 3-buten-2-one (1). Lewatit VP
OC 160016 and Amberlite XAD-7,17 which are well known supports
eimide7 and hetero-nucleophiles adding to
a-trifluoromethyl acrylic
acid8,9 or cinnamic esters,10 catalyzed by serine hydrolases and
bakers yeast. In the latter cases the authors could demonstrate that
these additions proceeded with moderate to good enantiose-
lectivity.9,10 More recently, several reports on conjugate additions of
hetero-nucleophiles to various acceptors catalyzed by proteases and
lipases were published.11,12 The few successful examples of enzyme-
catalyzed carbon–carbon bond formations via Michael additions
were mainly restricted to acetylacetone, acetoacetates and malonates
in combination with a,b
-unsaturated carbonyl compounds.13,14 From
* Corresponding author. Tel.: þ43 316 8738241; fax: þ43 316 8738740.
E-mail address: herfried.griengl@a-b.at (H. Griengl).
0040-4020/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2009.05.042

5664
G.A. Strohmeier et al. / Tetrahedron 65 (2009) 5663–5668
Table 1
for the immobilization of lipases by adsorption, catalyzed the model
reaction remarkably strong and were therefore also excluded from
the studies. At the end only Accurel MP1000 remained as suitable
support for the investigation of enzyme-catalyzed carbon–carbon
bond formations via Michael addition. Furthermore, using Accurel
MP100015 it was also possible to deposit the lyophilized enzymes on
the carrier. The results obtained with immobilized lipases via de-
position and adsorption after 24 h did not differ significantly in the
Michael-addition reactions between 3-buten-2-one 1 (0.2 M) and methyl nitro-
acetate 2a (0.2 M) in cyclohexane at 20 ꢀC, catalyzed by various immobilized lipases
No.
Manufacturer’s codea
Catalystb
pHc
Conv. [%]d
1
2
3
4
5
6
7
8
ICR-101
ICR-102
ICR-103
ICR-104
ICR-105
ICR-106
ICR-107
ICR-109
ICR-110
ICR-111
ICR-112
ICR-113
ICR-114
ICR-115
ICR-116
ICR-117
ICR-128
Aspergillus sp. lip.
Rhizopus sp. lip.
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
d
8
11
6
9
16
5
7
9
46
7
63
7
27
10
40
14
15
Rhizopus oryzae lip.
Penicillium sp. I lip.
Penicillium sp. II lip.
Candida rugosa lip.
Pseudomonas cepacia lip.
Pseudomonas fluorescens lip.
CALB
case of CALB. Initial screenings of
dehydes in the presence of immobilized CALB and CH-acidic com-
pounds comprising malonates, -ketoesters, acetylacetone and
a,b-unsaturated ketones and al-
b
9
activated acetates revealed only a few successful reactions. More-
over, only the donor molecule methyl nitroacetate (2a) reacted no-
ticeably with a,b-unsaturated carbonyl compounds during the time
span of a few hours. Regarding the acceptor molecules it was not
surprising that acrolein turned out as the most active substrate fol-
10
11
12
13
14
15
16
17
Candida sp. lip.
CALA
Pseudomonas sp. lip.
Porcine pancreas lip.
T. lanuginosas lip.
M. miehei lip.
Alcaligenes sp. lip.
Lipase modified
by directed evolution
for large substrates
CALB WT
lowed by 3-buten-2-one (1) and 1-penten-3-one. b-Substituted an-
alogues like trans-4-phenyl-3-buten-2-one did not react at all. Since
in reactions with methyl nitroacetate (2a) acrolein gave a product
mixture we used 1-buten-3-one (1) as model compound, which gave
a clean Michael addition to a single product. All immobilized lipases
were screened for their efficiency to catalyze the conjugate addition
(see Scheme 1) using 0.2 M solutions of reagents in cyclohexane at
room temperature.
18
19
20
21
22
23
24
25
7.65
7.65
7.65
7.65
7.65
10
26
99e,f
1e
CALB Ser105Ala
CALB (2.5% w/w)
CALB (2.5% w/w) denatured
CALB (2.5% w/w) inhibited
Accurel MP1000
BSA (1.26% w/w) on Accurel
Blank
1e
11e
42
1e
7.00
O
R1
enzyme
cyclohexane
20 °C
COOMe
O
+
COOMe
R2
a
b
c
d
e
f
BioCatalytics Inc.
Immobilized on Accurel MP1000 (1.6% w/w).
pH not adjusted or pH of the sodium phosphate immobilization buffer.
Conversion after 16.5 h at 20 ꢀC.
Conversion after 20 h at 20 ꢀC.
R2
R1
3a,b
1
2a,b
Ph
R1
Methyl nitroacetate (0.4 M) 2a.
enzyme
O2N
O2N
COOMe
R2
+
2a - d
Ph
cyclohexane
20 °C
In contrast 1-penten-3-one gave a conversion of 45% under the
4
5a - d
same conditions indicating strong substrate dependence. Most
noticeably immobilized bovine serum albumin catalyzed the con-
jugate addition fairly strong, which points out that some proteins
without enzyme function can also catalyze conjugate additions.19
Therefore one has to prove that the catalytic activities of the lipases
for Michael-addition reactions did not arise from unspecific pro-
tein-derived activation of the reagents by, e.g., the surface of the
enzyme. It was experimentally verified that inhibition and de-
naturation of CALB caused a complete disruption of the catalytic
activity of the enzyme leading to a similar conversion as found for
the blank reaction. As a consequence one must conclude that the
enzyme-catalyzed Michael addition occurs in the active site of the
enzyme or in close proximity. On the other hand one can assume
that the protein surface of CALB is predominately catalytically
inactive in the process. Having discovered active lipases for the
catalysis of conjugate additions of methyl nitroacetate (2a) to 3-
buten-2-one (1) we became encouraged to investigate the stereo-
chemical course of the transformations. Because the hitherto used
reactions (Scheme 1) furnished products, which tend to racemize in
2, 3, 5
R1
R2
H
CH3
H
NO2
NO2
COMe
a
b
c
d
-(CH2)3CO-
Scheme 1. Enzyme-catalyzed Michael additions of nitroesters (2a) and (2b) to 3-
buten-2-one (1), respectively, nitroesters (2a) and (2b) and -ketoesters (2c) and (2d)
to trans- -nitrostyrene (4).
b
b
The experiments clearly identified Candida antarctica lipases A
and B along with the lipases from Mucor miehei and porcine pan-
creas as efficient catalysts for the Michael-addition reaction. Be-
sides the addition reaction we always observed a loss of methyl
nitroacetate (2a) during the reaction which most likely resulted
from lipase-catalyzed ester hydrolysis caused by traces of water in
the system. This latter fact might be the reason why some enzyme-
containing reactions gave lower conversions than Accurel alone. In
accordance with the literature examples concerning the formation
of carbon–carbon bonds we also noticed a better conversion with
the Ser105Ala mutant13 compared to wild-type CALB. An expla-
nation for this result might be that the partial deletion of the cat-
alytic triad by the replacement of Ser105 against non-coordinating
Ala105 makes the enzyme more prone to non-natural reactions.
Moreover, the possibly enhanced activation of the acceptor mole-
cule and easier deprotonation of methyl nitroacetate (2a) by a more
basic His224 might also be the reason for improved Michael-ad-
dition activity.12,13,18 Using a higher loading of the lipase on the
carrier and an excess of methyl nitroacetate (2a) it was possible to
convert 3-buten-2-one (1) almost completely into the corre-
sponding product (entry 20, Table 1).
the
a-positions, thereby loosing all chiral information from po-
tential stereoselective additions, we decided to replace the donor
molecule 2a by methyl 2-nitropropanoate (2b). This starting ma-
terial has a maximum of structural similarity to methyl nitroacetate
(2a) but leads to a racemization-free quaternary centre in the
product. Though the substituents are not very different in size in
this case it seems feasible to obtain enantio-enriched products in
enzyme-catalyzed Michael additions if there was any preferred
stereochemical course of the reaction (see Scheme 1).
As depicted in Table 2 the Michael-addition reactions of 2-
nitropropanoate (2b) to 3-buten-2-one (1) were catalyzed by var-
ious immobilized lipases but the conversions were significantly
lower than with methyl nitroacetate (2a) although the reaction

G.A. Strohmeier et al. / Tetrahedron 65 (2009) 5663–5668
5665
Table 2
formation although a certain type of interaction in the active site of
enzymes must have occurred. This assumption is again supported by
the observation that inhibited CALB (CALB treated with methyl p-
nitrophenyl n-hexylphosphonate) was not catalytically active at all
(conversion <1%). The reaction takes place via the attack of methyl 2-
Michael-addition reactions between 3-buten-2-one (1f) (0.6 M) and methyl 2-ni-
tropropanoate (2b) (0.2 M) in cyclohexane at 20 ꢀC, catalyzed by various immobi-
lized lipases, pH as in Table 1
No.
Manufacturer’s codea
Catalystb
Conv. [%]c
e.e. [%]
1
2
3
4
5
6
7
8
ICR-101
ICR-102
ICR-103
ICR-104
ICR-105
ICR-106
ICR-107
ICR-109
ICR-110
ICR-111
ICR-112
ICR-113
ICR-114
ICR-115
ICR-116
ICR-117
ICR-128
Aspergillus sp. lip.
Rhizopus sp. lip.
R. oryzae lip.
Penicillium sp. I lip.
Penicillium sp. II lip.
C. rugosa lip.
P. cepacia lip.
P. fluorescens lip.
CALB
Candida sp. lip.
CALA
Pseudomonas sp. lip.
Porcine pancreas lip.
T. lanuginosas lip.
M. miehei lip.
Alcaligenes sp. lip.
Lipase modified
by directed evolution
for large substrates
Accurel MP1000
4
6
2
9
2
2
1
1
9
1
16
2
10
5
10
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
nitropropanoate (2b) at the
b-position of 3-buten-2-one (1), forming
a single chiral centre at the quasi
g-position. The chiral centre itself
stems exclusively from the donor molecule, which is initially
deprotonated, thereby loosing the original chiral information, and
then covalently bound to the acceptor molecule. Apparently the lack
of enantioselectivity must be attributed to the unselective addition
of methyl 2-nitropropanoate (2b) to the a,b-unsaturated ketone (1),
9
where within the active site no attack from a preferred direction
occurs. The donor molecule’s substituents might not be sufficiently
diverse to ensure a discrimination of reaction pathways during the
conjugate addition. All further attempts to gain additional in-
10
11
12
13
14
15
16
17
formation on the selectivity at the b-position failed because the al-
ternative substrates benzalacetone and cinnamic aldehyde did not
react under these conditions. It is worth mentioning that the liter-
ature on CALB and CALB S105A catalyzed reactions of thiol hetero-
nucleophiles with similar acceptor substrates also reports the lack of
any enantioselectivity.12
10
18
2
0
a
b
c
BioCatalytics Inc.
Immobilized on Accurel MP1000 (1.6% w/w).
Conversion after 40 h at 20 ꢀC using 0.2 M methyl 2-nitropropanoate (2b) and
2.2. Michael additions to trans-b-nitrostyrene (4)
0.6 M 3-buten-2-one (1) in cyclohexane.
Since in conjugate addition reactions
a,b
-unsaturated ketones
time and the amount of unsaturated ketone were increased (0.2 M
2b and 0.6 M 1 as compared to 0.2 M 2a and 0.2 M 1). The most
active enzymes in the reaction with methyl 2-nitropropanoate (2b)
match exactly those four most efficient ones mentioned in Table 1.
However, in product formation no enantioselectivity was found.
Taking these results one must conclude that the Michael addi-
tions investigated are completely unselective in terms of enantiomer
turned out to be rather unreactive substrates we were seeking some
more reactive Michael acceptors. During our initial screenings of
activated olefins with various CH-acidic compounds in the presence
of Novozyme 435 we had discovered high reactivities with aliphatic
and aromatic nitroalkenes. Due to the higher reactivity of these
substrates it became feasible to study the stereochemistry of con-
jugate addition to give products with more than one stereocentre.
Table 3
Michael-addition reactions between trans-
b-nitrostyrene (4) (0.1 M) and methyl nitroacetate (2a) (0.1 M), methyl 2-nitropropanoate (2b) (0.1 M), methyl 2-oxo-
cyclopentanecarboxylate (2d) (0.2 M) or methyl 3-oxobutanoate (2c) (0.2 M) in cyclohexane at 20 ꢀC, catalyzed by various immobilized lipases
No.
Manufacturer’s codea
Catalystb
5a
5b
5d
5c
Conv. [%]c
d.r.d
Conv. [%]c
1
d.r.d
2.8
Conv. [%]c
3
d.r.d
6.7
Conv. [%]c
d.r.d
1
2
3
4
5
6
7
8
ICR-101
ICR-102
ICR-103
ICR-104
ICR-105
ICR-106
ICR-107
ICR-109
ICR-110
ICR-111
ICR-112
ICR-113
ICR-114
ICR-115
ICR-116
ICR-117
ICR-128
Aspergillus sp. lip.
Rhizopus sp. lip.
R. oryzae lip.
Penicillium sp. I lip.
Penicillium sp. II lip.
C. rugosa lip.
P. cepacia lip.
P. fluorescens lip.
CALB
26
7
4
4
10
3
1.3
1.1
1.3
1.2
1.2
1.4
1.4
1.2
1.0
1.1
1.1
1.1
1.1
1.1
1.2
1.1
1.1
6
16
40
7
67
29
28
70
89
37
37
9
17
1.2
8
4.9
10
11
12
13
14
15
16
17
Candida sp. lip.
CALA
34
1
3
3
8
1.2
2.2
2.0
1.3
1.2
2.5
2.6
35
3
6
17
3
5
4.0
5.8
4.4
4.9
4.9
4.7
3.9
65
0.9
Pseudomonas sp. lip.
Porcine pancreas lip.
T.s lanuginosas lip.
M. miehei lip.
Alcaligenes sp. lip.
Lipase modified
by directed evolution
for large substrates
CALB WT
CALB Ser105Ala
CALB (2.5% w/w)
CALB (2.5% w/w) denatured
CALB (2.5% w/w) inhibited
Accurel MP1000
BSA (1.26% w/w) on Accurel
Blank
2
2
5
18
19
20
21
22
23
24
25
59
52
91
1
1.1
1.1
1.1
8
19
1.8
1.8
5
15
4.3
5.0
3
7
1.0
1.1
2
8
13
0
1.1
1.1
0
13
0
d
1
17
1
d
1
2
0
d
1.2
d
5.8
d
1.0
d
Reference substance
1.1
1.2
4.1
1.4
a
b
c
BioCatalytics Inc.
Immobilized on Accurel MP1000 (1.6% w/w).
Conversion after 20 h determined by chiral HPLC.
Diastereomeric ratio determined by chiral HPLC.
d

5666
G.A. Strohmeier et al. / Tetrahedron 65 (2009) 5663–5668
This special interest originates from a report of Kitazume et al. on
substituted 3,3,3-trifluoropropanoic acids, whose chiral -positions
that the amounts of enzyme or concentrations of reagents used in
these investigations were much higher.13,14 The published data
are therefore not fully comparable with our results. On the other
hand using higher amounts of enzymes, higher concentrations of
reagents or modified reaction conditions one can probably find
even more, hitherto undiscovered examples of enzyme-catalyzed
conjugate additions. However, stereoselective enzyme-catalyzed
Michael additions with carbon-nucleophiles remain unknown so
far.
a
were created via Michael additions of hetero-nucleophiles to 2-
trifluoromethyl acrylate in the presence of hydrolases.9
The Michael-addition experiments were run in cyclohexane at
20 ꢀC using 0.1 M trans-
b
-nitrostyrene (4f) and 0.1 M nitroester 2a
or 2b, respectively, 0.2 M -ketoester 2c and 2d, in the presence of
20 mg/mL of immobilized enzyme. This was half the concentration
used in the reactions with -unsaturated ketones. Successful
conjugate additions of this type (see Scheme 1) lead to products
with two chiral centres in the - and -position with respect to the
acceptor molecule, thus providing more information on the steric
course of the reaction. The transformation of trans- -nitrostyrene
b
a,b
b
g
3. Molecular modelling
b
A modelling of the interaction of the substrates with the amino
acids of the active site was performed for an interpretation of the
experimental result that the Michael additions investigated were
catalyzed by the enzymes applied but no enantioselectivity was
encountered.
(4) and methyl nitroacetate (2a) turned out to be the most reactive
combination with conversions up to 90% after 20 h (Table 3).
Conversions greater than 60% were found with immobilized li-
pases from M. miehei, Thermomyces lanuginosas and C. antarctica
(type Aand B). The lipases from Aspergillus sp., Pseudomonas sp.,
porcine pancreas, Alcaligenes sp. and the lipase ICR-128 gave con-
versions around 30–40%. C. antarctica lipase B and the S105A
mutant, both expressed in Pichia pastoris, displayed roughly the
same activity (Table 3, entries 19 and 20). All these conversions are
significantly higher than those obtained with pure Accurel or
immobilized albumin. Denatured and inhibited CALB were almost
completely inactive. Apart from the conversions it is even more
interesting to look at the stereochemical outcome of the reactions,
which did not proceed with any preferred stereoselectivity. All
products had a diastereomeric ratio of around 1.1 and no clear
indications of any enantioselectivity. The product composition was
virtually the same as obtained by non-stereoselective base cata-
lyzed Michael addition and was obviously hardly affected by side
reaction caused by the lipases’ ability to catalyze partial hydrolysis
with traces of water in the system. In the cases of other donor
molecules we observe similar results as for methyl nitroacetate
(2a) (Table 3).
Concerning the stereochemistry one can see that the di-
astereomeric ratio of products from experiments with decent
conversions match those of the chemically synthesized reference
substances. Minor deviations of the diastereomeric ratios were only
found in cases of reactions with low yields. Disappointingly, we
could not detect any enantioselectivity in any of the experiments
performed. The S105A mutant of CALB was generally more active
than the corresponding wild-type enzyme from the same expres-
sion host (Table 1, entries 19 and 20). Moreover it was also apparent
that Accurel-supported bovine serum albumin was often among the
best catalysts tested but could not introduce any chiral selection
into the reaction products, either. Since trans-
b-nitrostyrene (4)
was, at least in combination with methyl nitroacetate (2a), a highly
reactive acceptor molecule we also tested some other aromatic and
an aliphatic analogues. More crowded trans-
styrene and trans-4-fluoro- -nitrostyrene did not react with methyl
nitroacetate (2a) in the presence of CALA and higher-loaded CALB.
trans-4-Dimethylamino- -nitrostyrene, although only little solu-
b-methyl-b-nitro-
b
b
ble, gave a conversion of less than 10% in toluene under similar
conditions. CALB catalysis for the reaction of the aliphatic nitro-
alkene trans-1-nitrohept-1-ene and methyl nitroacetate (2a) at
a concentration of 0.2 M in cyclohexane gave a conversion of 35%
after 20 h. Again in none of these further successful examples we
could observe a stereoselective course of the reaction.
All attempts to convert 2-trifluoromethyl acrylate and the corre-
sponding methyl ester with methyl nitroacetate (2a) in the presence
of immobilized CALB and cyclohexane or toluene as solvents failed.
Surprisingly, although acetylacetone is a reactive C-nucleophile, it
did not react with trans-b-nitrostyrene (4) under our conditions.
Considering other successful examples for enzyme-catalyzed
Michael additions in the literature one must take into account
Figure 1. Docking results of methyl 2-methyl-2-nitro-5-oxohexanoate (3b) in CALB
S105A. (a) Active site pocket with the (R)-enantiomer. The most frequent binding mode
is coloured in green and the docked product located near the putative active site
residues is coloured in cyan. (b) Active site pocket with the (S)-enantiomer. The most
frequent binding mode is again coloured in green and the docked structure located
near the active site residues is represented in cyan. The presentation of the docking
results was created by using the PyMOL visualization program.

G.A. Strohmeier et al. / Tetrahedron 65 (2009) 5663–5668
5667
The docking of both enantiomers of methyl 2-methyl-2-nitro-5-
oxohexanoate (3b) into the active site of the S105A mutant of the
lipase from C. antarctica was carried out using the program AutoDock
v. 3.05.20 With the settings applied, the docked structures showed
only insignificant energetic differences and many different binding
modes (18 binding modes for the (R)-form and 14 binding modes for
the (S)-form of 3b). In each of the figures are depicted two examples
for the (R)-enantiomer (Fig. 1a) and the (S)-enantiomer (Fig. 1b).
The most frequent binding mode is clustered at the opposite side
of the putative active site residues Ala105, His224, Gln106 and
Thr40 and is shown in greenwithin the active site pocket in Figure 1.
The binding mode closest to the putative active site residue (His224)
with the lowest energy is coloured in cyan in Figure 1a for the (R)-
and in Figure 1b for the (S)-enantiomer. The oxyanion hole is formed
by the main chain polar hydrogens of the amino acids Gln106 and
Thr40 as well as by the Thr40 side chain polar hydrogen.
A common feature of the docking results is that the carboxyl
group binds in the oxyanion hole, which may indicate the activa-
tion of the acceptor via polarization. His224 may play a role as acid/
base catalyst. Both enantiomers in fact have similar docked ener-
gies and similar preferred binding modes. The active site is very
voluminous and our calculations do not indicate differential bind-
ing of (R)- and (S)-enantiomers of the docked product methyl 2-
methyl-2-nitro-5-oxohexanoate (3b). Therefore we conclude that
a stereoselective course of the enzyme-catalyzed Michael addition
is very unlikely from a theoretical point of view. All these facts
support the experimental findings that all reactions presented
herein proceeded without detectable stereoselective guidance by
the enzymes used.
Due to the fact that only esters were accepted, it is not out of the
question that an interim binding to the serine of the catalytic triad
via transesterification is involved, although the CALB S105A mutant
was still found catalytically active. As conclusion one must point
out that the biocatalysis of Michael additions, originating from
promiscuous activities of enzymes, is basically feasible. Enzymes
that are capable of performing the reaction stereoselectively still
remain to be discovered from natural sources or by creation with
molecular biology techniques.
5. Experimental section
5.1. Methods and materials
Chemicals were purchased from Aldrich or Fluka. The hydrolytic
enzymes kit was obtained from BioCatalytics Inc. (Pasadena, Cal-
ifornia, USA) and CALB from Aspergillus oryzae was from Fluka.
Accurel MP1000 was kindly donated by Membrana GmbH (Obern-
burg, Germany). Details on the synthesis of reference materials and
spectroscopic data are provided in Supplementary data. Thin-layer
chromatography was performed on Merck silica gel 60 F254 plates.
Flash chromatography was performed by using Merck silica gel 60H.
1H (500 MHz) and 13C (125 MHz) NMR spectra (referenced by using
the solvent peak) were recorded on a Varian Inova-500 instrument.
Chiral HPLC analyses were performed on an Agilent 1100 and the GC
analysis on a Hewlett Packard 6890 instrument.
5.2. Immobilization by deposition
The enzymes ICR-101 to ICR-128 (each of 4 mg and 20 mL for ICR-
4. Conclusion
128) were dissolved in distilled water (1.5 mL) and after adding the
polypropylene support Accurel MP1000 (250 mg) allowed to stand
for 3 h until the mixture was lyophilized. The immobilized enzymes
(1.6% w/w loading) were stored at 4 ꢀC until use.
Several novel lipase-catalyzed Michael-type carbon–carbon
bond formations for the addition of CH-acidic esters to small size
a,b-unsaturated ketones and nitroalkenes are reported. Methyl
nitroacetate (2a) turned out as the most active donor molecule,
5.3. Immobilization by adsorption
which led to almost complete conversion in reactions with methyl
vinyl ketone (1) and trans-
b
-nitrostyrene (4) within 20 h. More
Solutions of wtCALB and CALB S105Awere prepared by dissolving
the appropriate amounts of lyophilized enzyme (contains phos-
phate; protein content determined by absorbance assay) in 3 ml of
distilled water to obtain a loading of 2.5% (w/w) and subsequent
adjustment of the pH to 7.65. Commercially available CALB from A.
oryzae (12.4 mg) was dissolved in potassium phosphate buffer (3 ml,
pH 7.65, 50 mM). After adding Accurel MP1000 (500 mg) the mix-
tures were slightly stirred during 24 h at 20 ꢀC. Finally the solid
supports were filtered off, washed (buffer and distilled water) and
dried (20 mbar, CaCl2 followed by P2O5). The photometric lipase
activity assays based on the cleavage of p-nitrophenyl-butyrate
revealed less than 2% of the initial activity remaining in solution. Due
to a lack of an appropriate assay the immobilization behaviour of
CALB S105A was assumed as the same as for wtCALB.
sterically demanding donor and acceptor substrates were less re-
active or even completely unreactive. Only a limited number of
enzymes tested, such as CALA, CALB and the lipases from M. miehei
and T. lanuginosas, catalyze all or most of the investigated conjugate
additions appropriately. The reactions required strong electron
withdrawing substituents at the alkene substrate and highly nu-
cleophilic CH-acidic compounds to proceed. The activity of the ester
derivatives corresponded to the inverse order of pKa values,
descending from methyl nitroacetate (2a) to methyl acetoacetate
(2c). Enzyme induced stereoselectivity was not found, which in-
dicates that the catalytic mechanism is not as highly specific as the
natural ester hydrolysis reaction. Taking into account that dena-
tured or inhibited CALB is catalytically inactive one must conclude
that the enzyme-catalyzed conjugate addition occurs in or in close
proximity of the enzyme’s active site. One must assume that the
unique arrangement of amino acids in the protein cavity attributes
to the activation of the substrates in a certain way to facilitate the C–
5.4. Denaturation
Immobilized C. antarctica lipase B (50 mg) was treated with urea
(8 M) in potassium phosphate buffer (1 mL, pH 7.18, 50 mM) over
a period of 24 h at 4 ꢀC. After filtering the solid support was washed
(potassium phosphate buffer pH 7.65 and distilled water) and dried
(20 mbar, CaCl2 followed by P2O5). The p-nitrophenyl-butyrate as-
say confirmed less than 1% of remaining activity compared with
untreated immobilized enzyme.
C-bond formation. The absence of a reaction of trans-b-nitrostyrene
(4) with acetylacetone, which is a fairly reactive C-nucleophile, may
also point out that any unspecific catalysis by, e.g., the protein sur-
face may not be sufficient. These facts let us believe that a certain
type of substrate recognition must be involved in the process.
Docking simulations indicate that the enantiomers of methyl 2-
methyl-2-nitro-5-oxohexanoate (3b), the conjugate addition prod-
uct of methyl 2-nitropropanoate (2b) and 3-buten-2-one (1), have
not clearly distinguished binding modes in the voluminous active
site of the CALB S105A mutant. These data allow the conclusion that
a stereoselective enzyme-catalyzed Michael addition is unfavoured.
5.5. Covalent inhibition
Immobilized C. antarctica lipase B was treated with a solution of
methyl p-nitrophenyl n-hexylphosphonate (80 mM)21 in cyclohexane/

5668
G.A. Strohmeier et al. / Tetrahedron 65 (2009) 5663–5668
toluene 10:1 (1 mL for 120 mg immobilized lipase) during a period of
18 h at 20 ꢀC. After filtering the solid support was washed (cyclo-
hexane) and dried under vacuum. The remaining activity was less
than 1% as determined by a p-nitrophenyl-butyrate assay.
References and notes
1. Mather, B. D.; Viswanathan, K.; Miller, K. M.; Long, T. E. Prog. Polym. Sci. 2006, 31,
487–531.
2. (a) Marcelli, T.; van Maarseveen, J. H.; Hiemstra, H. Angew. Chem. 2006, 118,
7658–7666; Angew. Chem., Int. Ed. 2006, 45, 7496–7504; (b) List, B. Chem.
Commun. 2006, 819–824; (c) Christoffers, J.; Koripelly, G.; Rosiak, A.; Roessle, M.
Synthesis 2007, 1279–1300; (d) Pellissier, H. Tetrahedron 2007, 63, 9267–9331;
5.6. Michael-addition reactions
´
(e) Sulzer-Mosse, S.; Alexakis, A. Chem. Commun. 2007, 3123–3135; (f) Erkkilae,
Typical experiments with
a,b-unsaturated ketones were con-
A.; Majander, I.; Pihko, P. M. Chem. Rev. 2007, 107, 5416–5470; (g) Harutyunyan,
S. R.; den Hartog, T.; Geurts, K.; Minnaard, A. J.; Feringa, B. L. Chem. Rev. 2008,
108, 2824–2852; (h) Alexakis, A.; Backvall, J. E.; Krause, N.; Pamies, O.; Dieguez,
M. Chem. Rev. 2008, 108, 2796–2823.
3. (a) Simon, H. Chem. Bio. Chem. 1991, 2, 317–328; (b) Stuermer, R.; Hauer, B.; Hall,
M.; Faber, K. Curr. Opin. Chem. Biol. 2007, 11, 203–213.
4. Viola, R. E. Adv. Enzymol. Relat. Areas Mol. Biol. 2000, 74, 295–341.
5. (a) Carreras, C. W.; Santi, D. V. Annu. Rev. Biochem. 1995, 64, 721–762; (b) Kato,
Y.; Asano, Y. Arch. Microbiol. 1997, 168, 457–463; (c) Tai, C.-H.; Cook, P. F. Acc.
Chem. Res. 2001, 34, 49–59; (d) Sleeman, M. C.; Schofield, C. J. J. Biol. Chem.
2004, 279, 6730–6736; (e) Goldberg, I.; Rokem, J. S.; Pines, O. J. Chem. Technol.
Biotechnol. 2006, 81, 1601–1611; (f) Kusebauch, B.; Busch, B.; Scherlach, K.; Roth,
M.; Hertweck, C. Angew. Chem. 2009, 121, 1–5.
6. For reviews on enzyme promiscuity see the following: (a) Bornscheuer, U. T.;
Kazlauskas, R. J. Angew. Chem. 2004, 116, 6156–6165; Angew. Chem., Int. Ed.
2004, 43, 6032–6040; (b) Khersonsky, O.; Roodveldt, C.; Tawfik, D. S. Curr. Opin.
Chem. Biol. 2006, 10, 498–508; (c) Hult, K.; Berglund, P. Trends Biotechnol. 2007,
25, 231–238; (d) DePristo, M. A. HFSP J. 2007, 1, 94–98.
ducted by adding immobilized enzymes (40 mg/mL) to a solution of
reagents (0.2 M) in cyclohexane. In the case of nitro-olefins the
reactions were conducted at concentrations of 0.1 M. All reactions
were run under weak stirring at 20 ꢀC. Samples were taken directly
from the solution or after adding a co-solvent (acetone) to the re-
action to ensure complete dissolution of all reagents and products.
The measurements were generally conducted after evaporation of
cyclohexane and re-dissolving in n-heptane/2-propanol (HPLC) or
dichloromethane (GC). Analysis was done by HPLC (Chiralpak AD-H
and Chiralcel OD-H using n-heptane/2-propanol as solvent) or GC
(Chiraldex G-TA) using chemically prepared products as reference
material. All products were confirmed by NMR spectroscopy. The
isolated yields of products usually matched those determined by
chromatography.
7. (a) Flavin, M.; Slaughter, C. Biochemistry 1964, 3, 885–893; (b) Flavin, M. Bio-
chem. Biophys. Res. Commun. 1965, 20, 652–654; (c) Flavin, M.; Slaughter, C.
Biochemistry 1966, 5, 1340–1350; (d) Flavin, M.; Slaughter, C. J. Biol. Chem. 1969,
244, 1434–1444.
5.7. Computational methods
8. (a) Kitazume, T.; Ishikawa, N. Chem. Lett. 1984, 1815–1818; (b) Kitazume, T.;
Murata, K.; Kokusho, Y.; Iwasaki, S. Chem. Express 1987, 2, 357–360; (c) Kita-
zume, T.; Murata, K.; Kokusho, Y.; Iwasaki, S. J. Fluorine Chem. 1988, 39, 75–86.
9. Kitazume, T.; Ikeya, T.; Murata, K. J. Chem. Soc., Chem. Commun. 1986, 1331–1333.
10. Rama Rao, K.; Nageswar, Y. V. D.; Sampath Kumar, H. M. Tetrahedron Lett. 1991,
32, 6611–6612.
11. (a) Torre, O.; Alfonso, I.; Gotor, V. Chem. Commun. 2004, 1724–1725; (b) Cai, Y.;
Yao, S.-P.; Wu, Q.; Lin, X.-F. Biotechnol. Lett. 2004, 26, 525–528; (c) Cai, Y.; Wu,
Q.; Xiao, Y.-M.; Lv, D.-S.; Lin, X.-F. J. Biotechnol. 2006, 121, 330–337; (d) Qian, C.;
Xu, J.-M.; Wu, Q.; Lv, D.-S.; Lin, X.-F. Tetrahedron Lett. 2007, 48, 6100–6104; (e)
Priego, J.; Ortı´z-Nava, C.; Carrillo-Morales, M.; Lo´pez-Munguı´a, A.; Escalante, J.;
Castillo, E. Tetrahedron 2009, 65, 536–539; (f) Wang, J.-L.; Xu, J.-H.; Wu, Q.; Lv,
D.-S.; Liu, X.-F. Tetrahedron 2009, 65, 2531–2536; (g) de Souza, R. O. M. A.;
Matos, L. M. C.; Gonçalvesa, K. M.; Costa, I. C. R.; Babics, I.; Leite, S. G. F.; Oes-
treicher, E. G.; Antunes, O. A. C. Tetrahedron Lett. 2009, 50, 2017–2018.
12. Carlqvist, P.; Svedendahl, M.; Branneby, C.; Hult, K.; Brinck, T.; Berglund, P.
ChemBioChem 2005, 6, 331–336.
13. Svedendahl, M.; Hult, K.; Berglund, P. J. Am. Chem. Soc. 2005, 127, 17988–17989.
14. (a) Xu, J.-M.; Zhang, F.; Liu, B.-K.; Wu, Q.; Lin, X.-F. Chem. Commun. 2007, 2078–
2080; (b) Xu, J.-M.; Zhang, F.; Wu, Q.; Zhang, Q.-Y.; Lin, X.-F. J. Mol. Catal. B:
Enzym. 2007, 49, 50–54.
15. Persson, M.; Wehtje, E.; Adlercreutz, P. Biotechnol. Lett. 2000, 22, 1571–1575.
16. Lewatit VP OC 1600 is the solid support for the manufacture of Novozyme 435.
For an application see: P. Eigtved (Novo Industri A/S, Denmark), WO 8902916,
1989 [Chem. Abstr. 1989, 111, 92861].
Docking calculations were carried out using the Autodock 3.05
package.20 For the docking the lipase B from C. antarctica was
chosen (PDB Code 1TCA).22 The protein structure was checked and
prepared for the use in the docking experiments. We utilized the
NQ-Flipper server23 to recognize unfavourable rotamers of Asn and
Gln residues in the protein structure. The results were checked and
the protonation states of the histidines were assigned by visual
inspection. The mutation S105A was introduced by replacing the
Ser105 with an alanine using PyMOL.24 Additionally we used the
program protonate provided by the amber package25 to add polar
hydrogens to the amino acids. All water molecules were discarded
from the structure before docking. SYBYL v.7.326 was used to con-
struct the substrate (methyl 2-methyl 2-nitro-6-oxohexanoate) for
docking in both enantiomeric forms. Gasteiger–Hu¨ckel partial
charges were assigned to the atoms and the structures were en-
ergy-minimized using the Tripos force field.
Autodock 3.05 was used to dock the substrates to the receptor
protein applying a genetic algorithm augmented by a local search
(Solis & Wets). The Genetic Algorithm (GA) parameters were set as
follows: The number of individuals in the population was set to 50
and the maximum number of energy evaluations was set to
500,000 leading to a typical number of generations of 250. The
results of 50 independent runs were clustered.
17. Nakaoki, T.; Mei, Y.; Miller, L. M.; Kumar, A.; Kalra, B.; Miller, M. E.; Kirk, O.;
Christensen, M.; Gross, R. A. Ind. Biotechnol. 2005, 1, 126–134.
18. (a) Branneby, C.; Carlqvist, P.; Magnusson, A.; Hult, K.; Brinck, T.; Berglund, P.
J. Am. Chem. Soc. 2003, 125, 874–875; (b) Branneby, C.; Carlqvist, P.; Hult, K.;
Brinck, T.; Berglund, P. J. Mol. Catal. B: Enzym. 2004, 31, 123–128.
19. A strong catalytic activity of bovine serum albumin was found for a pro-
miscuous Morita–Baylis–Hillman reaction: Reetz, M. T.; Mondie`re, R.; Carbal-
leira, J. D. Tetrahedron Lett. 2007, 48, 1679–1681.
20. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart, W. E.; Belew, R. K.;
Olson, A. J. J. Comput. Chem. 1998, 19, 1639–1662.
Acknowledgements
21. Rotticci, D.; Norin, T.; Hult, K.; Martinelle, M. Biochim. Biophys. Acta, Mol. Cell
Biol. Lipids 2000, 1483, 132–140.
¨
The Osterreichische Forschungsfo¨rderungsgesellschaft (FFG),
the Province of Styria, the Styrian Business Promotion Agency (SFG)
and the city of Grazdwithin the framework of the Kplus pro-
gramdare acknowledged for financial support.
22. Uppenberg, J.; Hansen, M. T.; Patkar, S.; Jones, T. A. Structure (Cambridge, MA,
United States) 1994, 2, 293–308.
23. Weichenberger, C. X.; Sippl, M. J. Nucleic Acids Res. 2007, 35, W403–406.
24. DeLano, W. L. The PyMOL Molecular Graphics System; DeLano Scientific: San
Carlos, CA, USA, 2002.
25. Pearlman, D. A.; Case, D. A.; Caldwell, J. W.; Ross, W. S.; Cheatham, T. E., III;
DeBolt, S.; Ferguson, D.; Seibel, G.; Kollman, P. Comput. Phys. Commun. 1995, 91,
1–42.
26. Tripos International, SYBYL 7.3, 1699 South Hanley Rd., St. Louis, Missouri,
63144, USA.
Supplementary data
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.tet.2009.05.042.