Hydrolysis of hydrophobic esters in a bicontinuous microemulsion catalysed by lipase B from Candida antarctica

Article abstract of DOI:10.1002/chem.201405335

Selective enzyme-catalysed biotransformations offer great potential in organic chemistry. However, special requirements are needed to achieve optimum enzyme activity and stability. A bicontinuous microemulsion is proposed as reaction medium because of its

Full text of DOI:10.1002/chem.201405335

DOI: 10.1002/chem.201405335
Full Paper
&
Biocatalysis
Hydrolysis of Hydrophobic Esters in a Bicontinuous
Microemulsion Catalysed by Lipase B from Candida antarctica
[b]
[a]
[c]
[a]
Anne K. Steudle, Mireia Subinya, Bettina M. Nestl, and Cosima Stubenrauch*
Abstract: Selective enzyme-catalysed biotransformations
offer great potential in organic chemistry. However, special
requirements are needed to achieve optimum enzyme activi-
ty and stability. A bicontinuous microemulsion is proposed
as reaction medium because of its large connected interface
between oil and water domains at which a lipase can adsorb
and convert substrates in the oil phase of the microemul-
sion. Herein, a microemulsion consisting of buffer–n-octane–
nonionic surfactant CE was used to investigate the key fac-
highest CalB activity was found around 448C in the absence
of NaCl and substrates with larger alkyl chains were better
hydrolysed than their short-chained homologues. The CalB
activity was determined using two different co-surfactants,
namely the phospholipid 1,2-dioleoyl-sn-glycero-3-phospho-
choline (DOPC) and the sugar surfactant decyl b-d-glucopyr-
anoside (b-C G ). The results show the CalB activity as linear
10
1
function of both enzyme and substrate concentration with
an enhanced activity when the sugar surfactant is used as
co-surfactant.
i
j
tors that determine hydrolyses of p-nitrophenyl esters cata-
lysed by the lipase B from Candida antarctica (CalB). The
Introduction
steps and reduced waste output, biotechnological processes
[6]
are often preferred to purely chemical approaches. Lipases
are known to be versatile biocatalysts and numerous reviews
can be found describing the properties and applications of li-
Lipases (EC 3.1.1.3) hydrolyse in vivo triglycerides at an oil/
[
1]
water interface and belong to the class of hydrolases. Most li-
pases possess a lid that covers the active site in an aqueous
environment and opens in contact with a hydrophobic
[1,7]
pases in industry and research.
However, some biotechno-
logical processes still lack competitive space-time yields, sus-
tainable reaction conditions or high enantiopurities.
[
2]
phase, so-called interfacial activation. Since bicontinuous mi-
[
3]
croemulsions have a very large oil/water interface, we investi-
gated them as the reaction media for hydrolytic reactions cata-
lysed by lipases. The general composition of the microemul-
sion was buffer–n-octane–surfactant. Lipase B from Candida
antarctica (CalB) was chosen because of its high stereoselectiv-
In this work we focus on lipase-catalysed hydrolysis reac-
tions. Applications of this type of reaction are the kinetic reso-
lution of racemic mixtures to generate small, optically pure
[8]
molecules like chiral alcohols and amines and the direct
asymmetric synthesis of drugs, for example profens, which are
among the most important non-steroidal anti-inflammatory
[
4]
ity, its well-known crystal structure, reaction mechanism and
conformational stability in both hydrophilic and hydrophobic
[9]
drugs. A challenge is often the limited solubility of non-polar
substrates, even in mixed aqueous–organic systems. We intend
to overcome these solubility limits by using bicontinuous mi-
croemulsions as reaction medium in which the aqueous phase
hosts the enzyme, whereas the substrate is dissolved in the oil
phase. Contact between enzyme and substrate is facilitated
through the interfacial layer.
[
5]
environments. In contrast to other lipases, CalB does not un-
dergo conformational changes by being adsorbed at an inter-
face. This can be attributed to the fact that CalB does not pos-
sess a lid at the entrance of the active site.
Due to the fact that biocatalysts provide high substrate spe-
cificity, good yields, almost no side-products, less purification
Lipase activity has been studied intensely in water/oil (w/o)-
[10,11]
microemulsions
but only few studies have yet been report-
[
a] Dr. M. Subinya, Prof. Dr. C. Stubenrauch
ed on enzyme-catalysed reactions in bicontinuous microemul-
Institute of Physical Chemistry, University of Stuttgart
Pfaffenwaldring 55, 70569 Stuttgart (Germany)
E-mail: cosima.stubenrauch@ipc.uni-stuttgart.de
[12]
sions. We chose to use the latter one because, in contrast to
droplet microemulsions, bicontinuous microemulsions provide
a connected interface between water and oil. In this paper, we
want to identify key parameters for lipase-catalysed hydrolysis
reactions in bicontinuous microemulsions. For this purpose,
microemulsions prepared with alkyl polyglycol ethers CE as
[b] A. K. Steudle
Chemistry Department, Durham University
South Road, DH1 3LE Durham (UK)
[
c] Dr. B. M. Nestl
Institute of Technical Biochemistry, University of Stuttgart
Allmandring 31, 70569 Stuttgart (Germany)
i
j
amphiphilic compound were chosen to take advantage of
a wide range of accessible homologues without changing the
overall nature of the surfactant.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201405335.
Chem. Eur. J. 2015, 21, 2691 – 2700
2691
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
For this endeavour we address four key points: 1) The deter-
mination of the phase behaviour and the partitioning of the
substrate in the microemulsion, 2) the effect of temperature
and NaCl content on the CalB activity, 3) the investigation of
the reaction kinetics by using different enzyme and substrate
concentrations and 4) the effect of co-surfactants on the CalB
activity. The hydrophilic sugar surfactant n-decyl-b-d-glucopyr-
anoside (b-C G ) and the hydrophobic phospholipid 1,2-dio-
the CalB molecules are located at the interfacial layer, whereas
the rest is located in the water domain of the microemul-
[16]
sion. In the following, the influence of one substrate and
two products on the phase behaviour of microemulsions con-
taining buffer/4 wt% NaCl/p-nitrophenol–n-octane/p-nitro-
phenyl palmitate/palmitic acid–C E is presented. Note that
10 5
throughout this work, 100 mm TRIS pH 7 (buffer) is used as an
aqueous phase to prepare the microemulsion. By way of an ex-
ample, the influence of p-nitrophenyl palmitate, p-nitrophenol
and palmitic acid is shown in Figure 1. All additives were dis-
solved in the aqueous or oil phase of the microemulsion at
a concentration of 20 mm. The oil-soluble additives p-nitro-
phenyl palmitate and palmitic acid were found to shift the X-
point of the microemulsion to lower phase inversion tempera-
10
1
leoyl-sn-glycero-3-phosphocholine (DOPC) were used as co-sur-
factants.
The well-described hydrolysis of p-nitrophenyl esters to p-ni-
trophenol and fatty acid was chosen as the model reaction
(
Scheme 1) because this reaction is typically used to determine
[
13,14]
lipase activities in both aqueous reaction media
and w/o-
[
15]
˜
microemulsions. This reaction allows the photometric deter-
mination of the reaction rate by monitoring the p-nitrophenol
concentration as a function of time at the isosbestic wave-
length.
tures T and lower surfactant mass fractions g. This effect of
polar additives on the phase behaviour is well known and
[17]
hence expected.
The water-soluble additive p-nitrophenol
showed no effect on the efficiency of the microemulsions (i.e.,
˜
on g˜ ), however the phase inversion temperature T decreased.
To avoid changes in the composition of the microemulsion
during the reaction, the surfactant mass fraction and tempera-
ture at which the reaction was carried out were chosen such
that the one-phase region is not left during the reaction. In
other words, by choosing a surfactant mass fraction that was
high enough to overlap the one-phase regions of substrate-
containing and CalB-containing microemulsions, changes of
the phase behaviour during the reaction could be prevented.
Additionally, we determined the reaction rates from the first
Scheme 1. Lipase-catalysed hydrolysis of the p-nitrophenyl esters with differ-
ent chain lengths to p-nitrophenol and the respective fatty acid.
20 min of the reaction to minimise the influence of the prod-
ucts on the phase behaviour. Whether the one-phase region
was indeed not left was checked visually after each measure-
ment.
Results
Phase behaviour of microemulsions
Partitioning of substrates
The phase behaviour of microemulsions prepared with equal
volumes of oil and water (f=0.5) can be measured as a func-
tion of the temperature T and the surfactant mass fraction g.
To gain a deeper understanding of the reaction kinetics of the
CalB-catalysed hydrolysis, the location of the substrate in the
microemulsion was investigated. The aim is to facilitate the hy-
drolysis of a hydrophobic carboxylate ester, which is located in
the oil domain of the microemulsion and is converted by the
CalB at the interfacial layer. To prove that p-nitrophenyl palmi-
tate is only located in the oil domain of the microemulsion,
samples were prepared in the three-phase region. The advant-
age of a three-phase system is the fact that the compositions
of oil and water excess phases can be approximated to be the
same as the respective domains in the middle-phase microe-
mulsion possessing a bicontinuous microstructure.
˜
At the phase inversion temperature T, a three-phase region is
formed at low surfactant mass fractions g< g˜ , and a single
phase microemulsion at higher surfactant mass fractions g> g˜ .
˜
Defined by g˜ and T, the X-point can be used to describe and
compare the phase behaviour of different microemulsions. It is
to be noted that a substrate or an enzyme added to a microe-
mulsion might change the phase behaviour. The same holds
true for products generated during the reaction. Accordingly, it
is extremely important to study the influence of substrate,
product and enzyme on the phase behaviour prior to the reac-
tion. This allows counteracting and thus preventing any
changes of the phase behaviour, that is, of the microstructure,
during the reaction. The influence of CalB on the phase behav-
Therefore, a three-phase system was prepared and each of
the phases analysed by UV/Vis spectroscopy. The three-phase
system consisted of buffer/4 wt% NaCl–n-octane/5 mm p-nitro-
phenyl palmitate–C E (phase diagram in Figure 1) with a sur-
iour of the microemulsion system H O/NaCl (CalB)–n-octane–
2
10 5
C E was published previously and showed that an increase of
factant mass fraction of g=0.08. Figure 2 shows the spectra of
p-nitrophenyl palmitate in the oil excess phase, the microemul-
sion middle phase and the water excess phase. The oil excess
phase shows the highest absorbance, according to which the
p-nitrophenyl palmitate is mainly dissolved in the oil domain.
The water excess phase did not show any absorbance indicat-
10 5
CalB concentration in the microemulsion system shifts the
˜
phase inversion temperature T to lower values and that g˜
[16]
passes a minimum with increasing CalB concentration. The
phase behaviour of the CalB-containing microemulsions com-
bined with the partitioning studies revealed that 80–90% of
Chem. Eur. J. 2015, 21, 2691 – 2700
www.chemeurj.org
2692
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 2. UV/Vis spectra of the individual phases of three-phase system. The
microemulsions were prepared at g=0.08 and f=0.5. The sample consisted
of buffer pH 7/4 wt% NaCl–n-octane/5 mm p-nitrophenyl palmitate–C10E .
5
peratures at constant e as well as similar phase inversion tem-
peratures at different e are accessible.
We examined first the influence of the NaCl content e on
the CalB activity. For NaCl, contents of e=0, 0.04, 0.11 and 0.18
in the aqueous phase of the microemulsion the hydrophilicity
of the surfactant was increased stepwise to obtain phase inver-
sion temperatures around 368C (Table 1). Phase diagrams of
microemulsions consisting of buffer/NaCl–n-octane/5 mm p-ni-
trophenyl palmitate–C E were measured and the CalB activity
Figure 1. Phase diagrams of microemulsions consisting of buffer pH 7/
wt% NaCl–n-octane–C10 as a function of T and g. Influence of the oil-
soluble additives p-nitrophenyl palmitate and palmitic acid (top) as well as
of the water-soluble additive p-nitrophenol (bottom). The oil-to-water ratio
was kept constant at f=0.5.
4
E
5
10 j
in the respective microemulsion was determined.
ing no p-nitrophenyl palmitate in the water domain of the mi-
croemulsion. The microemulsion phase showed about half the
absorbance of the oil excess phase due to the volume fraction
of oil being f=0.5. Additionally, the wavelength of maximum
absorbance remained at 267 nm for all spectra, which is the
same as observed in a stock solution of p-nitrophenyl palmi-
tate prepared in n-octane. This indicates that the polarity of
the environment did not change. These results demonstrate
that p-nitrophenyl palmitate is only located in the oil domain
of the microemulsion, and neither dissolved in the water
phase nor adsorbed at the interfacial layer. The same result
was obtained for substrates with different lengths of alkyl
chains, namely p-nitrophenyl laurate and caprylate, as can be
seen in Figure S1 in the Supporting Information.
Table 1. X-points of the systems consisting of buffer pH 7/NaCl–n-
octane/5 mm p-nitrophenyl palmitate–C10E.
j
[
a]
j
e
g˜
g
Reaction
T˜
[8C]
k
2
ꢀ
3
ꢀ1 ꢀ1
[10 Ls g ]
4
5
6
7
.5
0
0.121
0.128
0.156
0.159
0.141
0.148
0.176
0.179
34
36
37
35
0.0817ꢁ0.0064
0.0610ꢁ0.0014
0.0074ꢁ0.0004
0.0033ꢁ0.0002
0.04
0.11
0.18
[a] For each system, the salt content e and the number of ethylene glycol
groups j is indicated. The CalB activity in each system is given by the ob-
served second order rate constant k .
2
The g used for the reactions was gReaction = g˜ +0.02. The CalB
activities are given as observed second order rate constants k2,
which were calculated from the initial reaction rate divided by
the enzyme and substrate concentration. The results are sum-
marized in Table 1, in which the X-point of the phase diagrams
are given for each NaCl content along with the observed
Effect of the NaCl content and temperature on the CalB ac-
tivity
Temperature and ionic strength are known to influence enzy-
matic activities. Kinetic measurements over a range of temper-
atures or at different salt contents can clarify the optimal reac-
tion conditions of the enzyme. To independently study the
effect of temperature and NaCl content e on the CalB activity
in bicontinuous microemulsions, the hydrophilicity of the sur-
factant was changed such that different phase inversion tem-
second order rate constants k . As one can see, the observed
2
second order rate constant k decreases significantly with in-
2
creasing NaCl concentrations and thus the addition of NaCl
was avoided wherever possible. With respect to the influence
of the temperature (measurements were carried out between
Chem. Eur. J. 2015, 21, 2691 – 2700
www.chemeurj.org
2693
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
34 and 378C, see Table 1) it was shown that the CalB activity
slightly increases with temperature until it starts to unfold and
[
13,18]
loose activity at about 45–558C.
However, in the study at
hand, the highest CalB activity was measured at the lowest
temperature. Thus temperature cannot be the dominating pa-
rameter in our case. One also sees that the g˜ -values and thus
the gReaction-values shift towards slightly higher values with in-
creasing NaCl concentration. To test whether changes of the
surfactant mass fraction g influence the CalB activity, we per-
formed reactions in a range from gReaction = g˜ +0.01 to gReaction
=
g˜ +0.05. Since we found no differences in the activity, we can
readily say that the k values reflect the dependence on the
2
NaCl concentration.
To study the effect of temperature on the CalB activity, all
microemulsions were prepared without NaCl (e=0). The stud-
ied systems consisted of buffer–n-octane/5 mm p-nitrophenyl
palmitate–C E with j=4, 4.5, 5 and 5.5 (j=4.5 means a 1:1
10 j
mixture of C E and C E ). The respective phase diagrams are
10
4
10 5
presented in Figure 3 (top), in which the surfactant mass frac-
tions used for the reaction gReaction are also indicated. As can be
seen in Figure 3 (top) the phase diagrams are shifted such that
measurements at T=23, 34, 44 and 518C could be carried out.
As mentioned above, slight changes in g˜ do not influence the
activity. Figure 3 (bottom) illustrates the observed second
order rate constants k for each temperature. The CalB activity
2
shows a bell-shaped dependence on the temperature, which is
well known from enzyme-catalysed reactions. This shape is ex-
plained by the increase of the reaction rate with temperature
as described by Arrhenius, which is counteracted by the ther-
mal deactivation of the enzyme at higher temperatures. The
highest CalB activity was obtained in the microemulsion pre-
pared with buffer–n-octane/5 mm p-nitrophenyl palmitate–
Figure 3. Top: phase diagrams of microemulsions consisting of buffer pH 7–
n-octane (5 mm p-nitrophenyl palmitate)–C E with j=4; 4.5; 5; 5.5. The oil-
to-water ratio was kept constant at f=0.5. The phase inversion temperature
T˜ and the surfactant mass fraction gReaction = g˜ +0.02 were used for the kinetic
1
0 j
measurements. Bottom: influence of the temperature on the CalB activity
measured in the respective microemulsions. The enzyme activity was mea-
ꢀ
1
sured at each temperature with 0.5, 1, 2 and 3 mgmL of CalB.
C E at T=448C and g =0.175. Consequently, we have
Reaction
10
5
chosen this system to be our model system for the following
studies wherever possible and feasible.
nitrophenol formed for each substrate throughout the reac-
tion.
The reactions were monitored until completion to detect
any inhibitory effects that could be caused by the substrates
or products. The yield obtained after 10 h indicates a clear de-
pendency on the chain length of the substrate, with 75% for
p-nitrophenyl caprylate, 80% for p-nitrophenyl laurate and
90% for p-nitrophenyl palmitate. However, the observed
Influence of the chain length of the substrate and the con-
centration on the CalB activity
To elucidate the impact of the length of the alkyl chain and
concentration of the substrate on the CalB activity, the hydrol-
ysis of substrates with different chain lengths were investigat-
ed. Moreover, the concentration was increased for one of
them. Three substrates were tested, namely p-nitrophenyl pal-
mitate, p-nitrophenyl laurate and p-nitrophenyl caprylate. To
study CalB activity with increasing substrate concentrations, p-
nitrophenyl laurate was used. In conventional aqueous reac-
tion media, these substrates cannot be solubilised without the
addition of detergents, and especially the p-nitrophenyl palmi-
tate shows severe solubility problems. The reactions were car-
ried out in microemulsions at 448C consisting of buffer–n-
second order rate constants k were found to be similar,
2
ꢀ
3
ꢀ1 ꢀ1
namely 0.14ꢁ10 Ls
g
for p-nitrophenyl caprylate, 0.17ꢁ
ꢀ
3
ꢀ1 ꢀ1
ꢀ3
ꢀ1 ꢀ1
10 Ls
g
for p-nitrophenyl laurate and 0.13ꢁ10 Ls g
for p-nitrophenyl palmitate. To study the CalB activity with in-
creasing substrate concentrations, p-nitrophenyl laurate had to
be used instead of p-nitrophenyl palmitate due to solubilisa-
tion problems of the latter in n-octane. The microemulsions
used were of the same composition as before with the only
difference that the amount of substrate in the oil phase was
varied. The resulting phase diagrams can be seen in Figure 5
(top). As can be seen, the phase inversion temperature de-
creases with increasing substrate concentration, which is why
the reaction temperatures were slightly different, or, more pre-
cisely, between 44 and 418C. Figure 5 (bottom) shows the ob-
tained initial reaction rates for the hydrolysis of p-nitrophenyl
octane/substrate–C E (g =0.175), which we have identi-
Reaction
10
5
fied to be the most promising system regarding the activity of
CalB. Phase diagrams containing 5 mm of the different sub-
strates did not exhibit any differences (see also Figure S2 in
the Supporting Information). Figure 4 shows the amount of p-
Chem. Eur. J. 2015, 21, 2691 – 2700
www.chemeurj.org
2694
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 4. The CalB-catalysed hydrolysis of p-nitrophenyl palmitate, -laurate
and -caprylate in a bicontinuous microemulsion liberates p-nitrophenol. The
amount of formed p-nitrophenol is calculated from the absorbance at the
isosbestic wavelength and is plotted as a function of time. The microemul-
5
sions consisted of buffer pH 7–n-octane/5 mm substrate–C10E . The enzyme
ꢀ
1
concentration was 3 mgmL in the aqueous phase and the microemulsion
was prepared at f=0.5, g=0.175 and T=448C.
laurate in a concentration range of 5–60 mm. The CalB activity
was found to be a linear function of the substrate concentra-
tion. To find out whether the small temperature differences in-
fluence the initial reaction rate, three substrate concentrations
were measured at two temperatures: 20 mm at 44 and 438C,
Figure 5. Top: phase diagrams of microemulsions consisting of buffer pH 7–
n-octane/p-nitrophenyl laurate–C10E with increasing p-nitrophenyl laurate
5
concentrations from 5–60 mm. Bottom: initial reaction rates of the CalB-cata-
lysed hydrolysis of p-nitrophenyl laurate in the microemulsion as a function
of the substrate concentration. The microemulsions were prepared at
f=0.5 and g=0.175 and at temperatures between T=44 and 418C accord-
ing to the location of the X-point at the respective p-nitrophenyl laurate
concentration. Please note that the initial reaction rate is normalised by the
4
0 mm at 43 and 428C and 60 mm at 42 and 418C. The ob-
tained initial reaction rates are also plotted in Figure 5
bottom) and it is obvious that temperature differences of
18C have only a very small effect on the CalB activity.
(
ꢁ
ꢀ1
CalB concentration, which is given in mgmL
.
Influence of the concentration of CalB on its activity
Previous kinetic studies of lipase-catalysed reactions performed
cient amount of CalB is adsorbed, the interfacial tension gab
drops sharply, until an equilibrium interfacial tension is
reached. At high CalB concentrations, the induction time and
sometimes even the time period in which gab drops down to
the equilibrium value is out of the detectable time range. At
[19]
in w/o-microemulsions reported the presence of lag times,
which are explained by the use of the injection method (i.e.,
injecting an enzyme solution into an already formed microe-
mulsion). Delays or lag times are observed when the adsorp-
[20]
ꢀ1
tion of the enzyme is slower than the reaction kinetics. As
we also used the injection method, we studied first possible
delays in the reaction kinetics. For this purpose, the adsorption
of CalB at a buffer/n-octane interface was followed by interfa-
cial tension measurements and then compared with reaction
kinetics measured at different CalB concentrations, focussing
on very small CalB concentrations. Figure 6 (top) shows the in-
terfacial tension measurements between n-octane and buffer
the highest CalB concentration of 3 mgmL only the equilibri-
um interfacial tension value could be detected.
After having observed that even small concentrations of
CalB adsorb relatively fast at the buffer/n-octane interface, the
CalB activity in the microemulsion was measured. By way of an
ꢀ
1
example, for 0.25 mgmL of CalB the equilibrium value is
reached after 100 s. As interfacial tension measurements and
adsorption times are sensitive to temperature, we could not
use our model system since the phase inversion temperature
solutions containing 0.025, 0.075, 0.25, 0.4, 0.5,
1
and
ꢀ
1
˜
3
mgmL CalB, which were carried out at room temperature.
of the model system is T=448C, whereas the interfacial ten-
In Figure 6 (bottom) the corresponding CalB activities in the
microemulsions are presented. The curves seen in Figure 6
sion study was carried out at room temperature. Thus the mi-
croemulsion consisting of buffer–n-octane/5 mm p-nitrophenyl
(
top) are typical for interfacial tension measurements with pro-
palmitate–C E (g
=0.141) was used for the kinetic meas-
10
4
Reaction
[
21]
teins. At low CalB concentrations, the induction time can be
observed, that is the time it takes for CalB to adsorb without
urements since it has a phase inversion temperature close to
˜
room temperature (T=238C) so that interfacial tension and ini-
significantly reducing the interfacial tension g . Once a suffi-
tial reaction rates can be compared directly. Figure 6 (bottom)
ab
Chem. Eur. J. 2015, 21, 2691 – 2700
www.chemeurj.org
2695
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
liquid crystalline phases at a minimum, only a co-surfactant
mass fraction of d (DOPC)=0.05 could be used. Sugar surfac-
tants, on the other hand, have a strongly hydrated head group
at nearly all temperatures and the maximum possible co-sur-
factant mass fraction (without losing the temperature-sensitivi-
ty of the microemulsions), was d (b-C G )=0.50. Due to very
10
1
high phase inversion temperatures in the presence of the
sugar surfactant, it was necessary to add 18 wt% NaCl to the
aqueous phase of these microemulsions to decrease the phase
inversion temperature so that the system can be compared
with the other systems in the present study. As the CalB activi-
ty has been proven to be quite temperature-dependent, the
influence of the co-surfactants was tested over the same tem-
perature range as before the C E microemulsions. The microe-
10 j
mulsions were prepared as explained previously by using dif-
ferent C E surfactants to obtain phase inversion temperatures
10 j
in the range of 20–508C.
The resulting systems were buffer–n-octane/5 mm p-nitro-
phenyl palmitate–C E/DOPC and buffer/18 wt% NaCl–n-
10 j
octane/5 mm p-nitrophenyl palmitate–C E/b-C G . Table 2
10
j
10
1
shows the X-points of each microemulsion, indicating the
number of polyethylene glycol groups j of the C E surfactants
10 j
for each system. The respective phase diagrams can be found
in Figure S3 in the Supporting Information. Each of these sys-
tems was then used to measure the CalB activity towards the
hydrolysis of p-nitrophenyl palmitate at the respective phase
Figure 6. Top: interfacial tension gab between buffer (pH 7) and n-octane as
a function of time. The samples contained either no or up to 3 mgmL
CalB in the aqueous phase. Bottom: initial reaction rates of the CalB-cata-
˜
inversion temperature T and the surfactant mass fraction
ꢀ
1
gReaction determined from the phase diagrams.
lysed hydrolysis of p-nitrophenyl palmitate in a microemulsion consisting of
buffer pH 7–n-octane/5 mm p-nitrophenyl palmitate–C10E at 238C as a func-
4
tion of the initial CalB concentration cCalB. The microemulsion was prepared
at g=0.141 and f=0.5.
Table 2. X-points of the systems consisting of buffer pH 7–n-octane/
5
j
mm p-nitrophenyl palmitate–C10E/DOPC and buffer pH 7/18 wt% NaCl–
[a]
n-octane/5 mm p-nitrophenyl palmitate–C10
E
j
/b-C10
G
1
.
shows the initial reaction rates plotted against the CalB con-
centration. Since no lag time was observed during the meas-
urements, we conclude that the reaction immediately started
at all studied CalB concentrations (resolution limit was a lag
time of less than one minute). As can be seen in Figure 6, the
CalB activity was found to be a linear function of the CalB con-
centration. The adsorption of the CalB at the interfacial layer of
the microemulsion and the rearrangement of the microemul-
sion system after the perturbation of adding the CalB solution
was therefore fast enough to not cause any lag time in the de-
tectable time range.
C E/DOPC
10 j 10 1
C E/b-C G
1
0
j
j
g˜
T
˜
[8C]
g˜
T˜ [8C]
4
4.5
5
5
6
0.089
0.092
0.112
0.127
–
23
32
43
50
–
0.088
0.117
0.127
0.142
0.160
24
31
39
44
50
.5
[
a]For each system the number of ethylene glycol groups j is indicated.
Figure 7 shows the resulting CalB activities as observed
second order rate constants k plotted against temperature.
2
The system containing only C E surfactant without co-surfac-
10 j
tants is plotted again for comparison. In the observed temper-
ature range, CalB was active in all investigated systems. The
highest CalB activity was found at temperatures of 40–458C re-
gardless of the composition of the interfacial layer. The abso-
lute values for the observed second order rate constants k2
vary with the composition of the interfacial layer. The co-sur-
factant DOPC was found to negatively influence the CalB activ-
ity, that is, lower CalB activities were observed compared to
values obtained in the DOPC-free C E systems. On the other
Co-surfactants
The addition of co-surfactants to the system can help to un-
derstand how the composition of the interfacial layer affects
the CalB activity. Two different co-surfactants were chosen,
namely the hydrophobic phospholipid 1,2-dioleoyl-sn-glycero-
3-phosphocholine (DOPC) and the hydrophilic sugar surfactant
n-decyl-b-d-glucopyranoside (b-C G ). Due to different proper-
1
0
1
ties, they could not be used at the same co-surfactant mass
fraction d. Phospholipids on the one hand are strongly amphi-
philic and form very efficient microemulsions, but also tend to
stabilize liquid crystalline phases. To keep the formation of
10 j
hand, the co-surfactant b-C G enhanced the CalB activity, that
10
1
is, higher CalB activities than those observed for the C E sys-
1
0 j
tems were found.
2696
Chem. Eur. J. 2015, 21, 2691 – 2700
www.chemeurj.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
previously reported for lecithin reverse micelles and was ex-
[25]
plained with a different water activity.
Studying the influence of temperature and ionic strength on
the activity of CalB in bicontinuous microemulsions was only
possible by changing the hydrophilicity of the surfactant. This,
unfortunately, also affects the domain sizes of the microemul-
sion because different surfactants have different efficiencies.
However, by looking at the X-points listed in Table 1, we see
that the efficiency does not change significantly, which, in
turn, means that the domain size stays more or less the same.
Nevertheless, a systematic study on the influence of the
domain size of the bicontinuous microemulsion on the enzy-
matic activity would be the next logical step.
The dependence of the CalB activity on the alkyl chain
length of the substrate revealed two interesting points. Firstly,
all the substrates are dissolved in the oil phase of the microe-
mulsion. Therefore, and in contrast to what happens in con-
ventional reaction media, no solubility problems are observed.
This is why the differences between the initial rates are small.
The slightly higher CalB activity towards the p-nitrophenyl lau-
rate can be explained by the fact that the active site of CalB
Figure 7. Influence of the temperature on the observed second order rate
constants k of the CalB-catalysed hydrolysis of p-nitrophenyl palmitate. The
microemulsions consisted of buffer pH 7–n-octane/5 mm p-nitrophenyl pal-
mitate–C10 /DOPC (d=0.05) and buffer pH 7/18 wt% NaCl–n-octane/5 mm
p-nitrophenyl palmitate–C10 /b-C10 (d=0.50). The different temperatures
were realised by changing j in the C10 surfactants. The microemulsions
were prepared at f=0.5 and gReaction = g˜ +0.02. The temperature depend-
ence of the k values obtained for the microemulsion containing only C10
2
E
j
E
j
G
1
E
j
[26]
2
E
j
fits exactly for 13 C atoms and that therefore the binding of
p-nitrophenyl laurate is the most favourable one. The second
interesting point concerns the obtained yield after 10 h. The
results show that the longer the alkyl chain of the substrate,
the better is the obtained yield, which can be explained with
the inhibition of CalB by short-chain fatty acids. Inhibition of
CalB and other lipases by alcohols and fatty acids are well
known from esterification reactions, also for octanoic acid and
surfactants is plotted for comparison.
Discussion
CalB activity in bicontinuous microemulsions is affected by
both NaCl content and temperature. The addition of NaCl de-
creases the enzymatic activity because it changes the ionic in-
teractions. These changes could decrease the adsorbed
amount of CalB molecules at the interfacial layer or affect the
CalB conformation. The reduced amount of adsorbed lipase
molecules at interfaces through increase of the NaCl content
has been reported for different systems, such as a membrane
[27]
lauric acid, which are formed as products in this study.
It is also very interesting to note that a linear relationship
between the substrate concentration and the CalB activity was
found. In the present study no saturation of the CalB activity
and therefore no Michaelis–Menten kinetics were observed.
[
22]
[23]
reactor system and an ionic liquids system. To investigate
if the NaCl content is influencing the activity of CalB in a two-
phase system, we carried out a reaction in a two-phase system
containing a stock solution of substrate in n-octane and
This is probably due to a quite high K as found in w/o-micro-
M
[28]
emulsions,
so that the studied substrate concentrations
were much lower than KM ([S]!K ). In the present study, we
M
did not further increase the substrate concentration due to the
large differences in the resulting X-points, which would have
made a comparison of the data very difficult.
100 mm TRIS buffer pH 7 with either 0, 4 or 18 wt% NaCl. The
results clearly showed a decrease of the obtained yields when
analysed after 24 h, as shown in Figure S4 in the Supporting
Information. Due to the fact that in the two-phase system the
only possible contact between enzyme and substrate is at the
macroscopic interface, desorption of CalB from this macroscop-
ic interface could explain the observed trend. With regards to
the influence of the NaCl content on the conformation of CalB,
circular dichroism measurements in aqueous solutions showed
no significant changes of the secondary conformation of CalB,
even in the presence of 18 wt% NaCl (Figure S5 in the Sup-
porting Information). Nevertheless, it is possible that the con-
formation of the CalB molecules that are adsorbed at the inter-
facial layer is more sensitive to unfavourable ionic interactions.
The correlation between temperature and CalB activity re-
vealed a maximum CalB activity at 448C, which is higher than
in conventional reaction media. For instance, in aqueous solu-
tion the maximum activity of a mixture of CalA and CalB is at
Studying the CalB adsorption at an oil/water interface we
found that at most CalB concentrations the interfacial tension
needs about 100 s to reach the equilibrium value. This relative-
ly fast adsorption time (compared to other proteins) correlates
well with the kinetics measured at different CalB concentra-
tions, in which no lag time was observed. The conclusion from
these findings is that CalB adsorption at the oil/water interface
is fast and that the reaction rate directly depends on the CalB
concentration. The observed reaction rate therefore depends
on both enzyme and substrate concentration, and no retarding
effects are observed.
The composition of the interfacial layer influences the reac-
tion rate of CalB-catalysed hydrolysis of p-nitrophenyl palmi-
tate. Compared with the CalB activity observed in microemul-
sions formulated with C E alone, the addition of the co-surfac-
10 j
tant DOPC led to lower CalB activities, while the addition of
[
24]
358C. An enhancement of the thermostability of lipases was
the sugar surfactant b-C G enhanced the CalB activity. These
10
1
Chem. Eur. J. 2015, 21, 2691 – 2700
www.chemeurj.org
2697
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
results can be explained by the hydrophilicity of the surfactant
head group and its hydration level. The results showed that
the CalB activity increases with the hydrophilicity of the surfac-
tant. For instance, in the microemulsions prepared with sugar
surfactant as co-surfactants, which is more hydrophilic than
C E the activity of CalB was higher compared to the b-C G -
Experimental Section
Enzyme and materials
All surfactants were purchased from Sigma (UK), namely tetraethy-
lene glycol monodecyl ether (C10
) with a purity of ꢂ97%, penta-
ethylene glycol monodecyl ether (C E ) with a purity of ꢂ97%,
E
4
10
5
10
5
10
1
hexaethylene glycol monodecyl ether (C E ) with a purity of
1
0 6
free system. These results are in agreement with studies per-
formed in w/o-microemulsions, which identified hydrophilic
surfactants with large head groups as promising surfactants for
ꢂ99%, octaethylene glycol monodecyl ether (C E ) with a purity
1
0 8
of ꢂ98% and decyl b-d-glucopyranoside (b-C G ) with a purity of
1
0
1
ꢂ98%. The substrates p-nitrophenyl palmitate, p-nitrophenyl lau-
[
11,15]
lipase-catalysed reactions.
The strong hydration of the
rate and p-nitrophenyl caprylate and the products p-nitrophenol
sugar surfactant head group can also play an important role in
providing water molecules especially at the interface and in
stabilising the CalB even at high (18 wt%) NaCl contents. As
a consequence, the strongly hydrated sugar surfactant head
group might also prevent desorption of CalB from the inter-
face, which was assumed to be the reason for the low CalB ac-
tivity in microemulsions prepared with C E at high e.
(
in spectrophotometric grade) and palmitic acid (ꢂ99%) were also
purchased from Sigma, UK. The lipase B from Candida antarctica
was purified and provided by the Institute of Technical Biochem-
istry, Stuttgart, Germany. The purification was performed by ion-ex-
change chromatography and the lyophilised purified protein was
used to prepare the stock solutions. For the preparation of the
buffer solution, tris(hydroxymethyl)aminomethane with a purity of
ꢂ99.9% was used. For the preparation of the microemulsions bi-
distilled water, sodium chloride (ꢂ99.5%) from Sigma, UK and n-
octane (ꢂ99%) purchased from Sigma, UK, was used.
[31]
10 j
To better understand the systems, we determined the activa-
tion energies of reactions for the co-surfactant-free microemul-
sion as well as for the two microemulsions prepared with sur-
factant/co-surfactant mixtures. The apparent activation energy
E was calculated according to the Arrhenius Equation and it
a
Preparation of the microemulsions
ꢀ
1
was found to be 32 kJmol for the C E /b-C G system,
10
5
10
1
ꢀ
1
The parameters describing the composition of the microemulsion
are described in the following Equations. The mass fraction of the
surfactant in the total mixture is given by:
5
6
0 kJmol
for the co-surfactant-free C E system and
10 5
ꢀ
1
0 kJmol for the C E /DOPC system. The lowest apparent
10 5
activation energy found for the microemulsion prepared with
C E /b-C G is comparable to the activation energies that
10
5
10
1
were reported for lipase-catalysed reactions in droplet microe-
mSurfactant
g ¼ m
Surfactant þ mOil þ mBuffer
ð1Þ
[
25,29,30]
mulsions.
The length of the alkyl chain of the substrate
can also influence the activation energy. Indeed, the activation
energy for hydrolysis of the esters seems to increase with in-
The mass fraction of the co-surfactant in the surfactant mixture is
given by:
ꢀ
1
creasing chain length of the ester, for example, 20 kJmol for
ꢀ
1
p-nitrophenyl butyrate and 44 kJmol for p-nitrophenyl capry-
[
30]
late. Additionally, one should note that the different gReaction
m
Coꢀsurfactant
d ¼ m
ð2Þ
of the systems cannot cause the differences in activities since
Coꢀsurfactant þ mSurfactant
it
holds
gReaction(C E /DOPC)<g
(C E /b-C G )<
Reaction 10 5 10 1
10
5
gReaction(C E ) whereas the activity follows k (C E /DOPC)<
The oil-to-water ratio is either given by the mass fraction of the oil
in the mixture of oil and aqueous phase
10
5
2
10 5
k (C E )<k (C E /b-C G ).
2
10
5
2
10
5
10
1
Conclusion
mOil
a ¼ m
ð3Þ
Oil
þ mBuffer
We have found a temperature dependence of the CalB activity
in bicontinuous microemulsions that does not depend on the
interfacial composition with a maximum activity around 448C.
The influence of NaCl was found to be more complex, with
high NaCl contents found to be unfavourable for the CalB ac-
tivity when the reactions were carried out in microemulsions
prepared with C E surfactants, but not in the presence of
or by the corresponding volume fraction
VOil
ꢀ
¼ V
ð4Þ
Oil þ VBuffer
10 j
sugar surfactant. Interfacial tension measurements between
CalB solutions and n-octane revealed fast adsorption at nearly
all concentrations, which supports our previous results that
showed CalB preferably located at the interface. The linear de-
pendence of the CalB activity on both enzyme and substrate
concentrations showed that CalB-catalysed biotransformations
are feasible in bicontinuous microemulsions and benefit from
fast mass transport and enhanced enzyme–substrate contact
at the interface.
The mass fraction of NaCl in the total mass of the aqueous phase
is:
mNaCl
e ¼ m
ð5Þ
NaCl
þ mBuffer
The concentration of lipase in the water phase of the microemul-
sion is:
Chem. Eur. J. 2015, 21, 2691 – 2700
www.chemeurj.org
2698
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
mCalB
VBuffer
at this wavelength to a small extent. The absorbance at a specific
wavelength is described as
cCalB
¼
ð6Þ
For the preparation of bicontinuous microemulsions, equal vol-
umes (f=0.5) of a buffer solution with appropriate amounts of
NaCl and n-octane were used. Where possible, stock solutions of
substrate in n-octane were prepared, as well as stock solutions of
lyophilised purified CalB in buffer. For the phase diagrams, 0.3 mL
of oil phase and 0.3 mL of water phase was used, giving a total
volume of 0.6 mL of oil and water. For the reactions, 0.075 mL of
oil and 0.075 mL of water was used, giving a total volume of
X
A ¼ d ꢄ
ei ꢄ c
i
ð9Þ
i
in which A is the absorbance at a specific wavelength, d is the
path length of the cuvette, e is the molar absorptivity and c is the
concentration. The index i represents all components absorbing at
this wavelength, in our case the substrate or the product p-nitro-
phenol. We can now write for the concentration of p-nitrophenol
at a time x during the reaction
0
.15 mL of oil and water. The corresponding surfactant mass was
calculated and the three components were weighed in with an ac-
ꢀ
3
curacy of 0.5ꢁ10 g.
Optical detection of phase behaviour
Alðisosb:Þ
ꢀ
eSubstrate ꢄ c
d
Substrate;t¼0
c
¼
ð10Þ
pNP;t¼x
ðe ꢀ e
Þ
The microemulsions were prepared in a glass vial, which was
closed with a plastic stopper and sealed with laboratory film. The
vial was placed in a thermostated water bath equipped with a mag-
netic stirrer and a digital thermometer (accuracy 0.018C) next to
the vial. The phase boundaries enclosing the one phase region at
surfactant mass fractions g> g˜ were optically detected in an accu-
racy of ꢁ0.58C and extrapolated to the X-point. Anisotropic
phases were detected by using crossed polarisers.
pNP
Substrate
In the present work, all concentrations are initial concentrations in
either the oil or the aqueous phase used to prepare the microe-
mulsion.
To determine the isosbestic wavelength of the product, scans of p-
nitrophenol in microemulsions containing buffer solutions at differ-
ent pH values were taken. The molar absorptivity e measured ac-
cording to the Beer–Lambert law was correlated with the initial
concentration of p-nitrophenol in the aqueous phase and is there-
fore smaller than in aqueous solution due to the presence of oil
and surfactant in the mixture. Hence, the molar absorptivity e was
measured for each system.
Localisation of compounds in a three-phase microemulsion
To investigate the localisation of compounds in a microemulsion,
a three-phase microemulsion was prepared at a surfactant mass
fraction g ! g˜ (in the present study g=0.08). The microemulsion
was equilibrated at the phase inversion temperature taken from
the respective phase diagram. The samples were prepared in GC
vials, sealed with laboratory film. After 24 h the individual phases
were carefully removed through the septum, using a syringe with
a needle. The spectra of the individual phases were measured in
a Cary 100 spectrophotometer with a thermostated cell holder.
Each of the phases were equilibrated and measured at the phase
inversion temperature, to ensure that the microemulsion (i.e., the
middle phase) was homogeneous and transparent.
The measurements were carried out using a Cary 100 spectropho-
tometer with a thermostated cell holder. For the measurement of
the enzyme activity, the microemulsion was thermostated in the
cell holder for 5 min before the reaction was started by adding the
CalB solution. For this purpose, the microemulsion was initially pre-
pared with 12.5 mL less aqueous phase than oil phase, which was
then added as a concentrated CalB solution to obtain the desired
CalB concentration in the aqueous phase. Scans were taken from
600–200 nm, at least every 2 min. After the measurements, the
sample was checked again to be in the one-phase region. An ex-
ample of a kinetics measurement is shown in Figure S6 in the Sup-
porting Information.
Measurement of reaction kinetics
According to the Michaelis–Menten kinetics, for K @[S], the initial
reaction rate v can be described as
M
Acknowledgements
We thank the EPRSC for funding. We also thank Dr. Sandra En-
gelskirchen, Dr. Bernd Nebel and Prof. Dr. Bernhard Hauer for
fruitful discussions. The support of Dr. AnnMarie O’Donoghue
to interpret the kinetic data is greatly acknowledged.
kcat½Eꢃ½Sꢃ
v ¼
ð7Þ
K
M
and the second-order rate constant k =k /K can be introduced:
2
cat
M
Keywords: biocatalysis · enzymes · esters · hydrolases ·
hydrolysis
v ¼ k ½Eꢃ½Sꢃ
ð8Þ
2
[
1] P. Reis, K. Holmberg, H. Watzke, M. E. Leser, R. Miller, Adv. Colloid Inter-
face Sci. 2009, 147–148, 237–250.
Please note that the enzyme concentration [E]=c
is given in
g . The initial re-
CalB
ꢀ1 ꢀ1
ꢀ
1
[2] R. Verger, Trends Biotechnol. 1997, 15, 32–38.
mgmL and thus the resulting unit for k is Ls
2
[
[
3] R. Strey, Colloid Polym. Sci. 1994, 272, 1005–1019.
4] E. M. Anderson, K. M. Larsson, O. Kirk, Biocatal. Biotransform. 1998, 16,
action rate v is given by the slope of the linear increase of concen-
tration versus time. For this purpose, the absorbance of the prod-
uct p-nitrophenol was followed at the isosbestic wavelength for
181–204.
[
5] a) J. Uppenberg, M. T. Hansen, S. Patkar, T. A. Jones, Structure 1994, 2,
293–308; b) J. Uppenberg, N. Oehrner, M. Norin, K. Hult, G. J. Kleywegt,
S. Patkar, V. Waagen, T. Anthonsen, T. A. Jones, Biochemistry 1995, 34,
2
0 min. To calculate the concentration of p-nitrophenol from the
absorbance one needs to consider that the substrate also absorbs
Chem. Eur. J. 2015, 21, 2691 – 2700
www.chemeurj.org
2699
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
6838–16851; c) C. Li, T. Tan, H. Zhang, W. Feng, J. Biol. Chem. 2010,
[17] S. Burauer, T. Sachert, T. Sottmann, R. Strey, Phys. Chem. Chem. Phys.
1999, 1, 4299–4306.
285, 28434–28441; d) M. Martinelle, M. Holmquist, K. Hult, Biochim. Bio-
phys. Acta Lipids Lipid Metab. 1995, 1258, 272–276; e) V. Ferrario, C.
Ebert, L. Knapic, D. Fattor, A. Basso, P. Spizzo, L. Gardossi, Adv. Synth.
Catal. 2011, 353, 2466–2480.
6] A. Schmid, J. S. Dordick, B. Hauer, A. Kiener, M. Wubbolts, B. Witholt,
Nature 2001, 409, 258–268.
[18] A. L. Ong, A. H. Kamaruddin, S. Bhatia, W. S. Long, S. T. Lim, R. Kumari,
Enzyme Microb. Technol. 2006, 39, 924–929.
[19] M. B. Stark, P. Skagerlind, K. Holmberg, J. Carlfors, Colloid Polym. Sci.
1990, 268, 384–388.
[20] S. Avramiotis, P. Lianos, A. Xenakis, Biocatal. Biotransform. 1996, 14,
299–316.
[
[
7] a) V. Gotor-Fernꢂndez, R. Brieva, V. Gotor, J. Mol. Catal. B 2006, 40, 111–
1
6
20; b) T. Tan, J. Lu, K. Nie, L. Deng, F. Wang, Biotechnol. Adv. 2010, 28,
28–634; c) M. M. Soumanou, M. Pꢃrignon, P. Villeneuve, Eur. J. Lipid
[21] R. Miller, V. B. Fainerman, A. V. Makievski, J. Krꢅgel, D. O. Grigoriev, V. N.
Kazakov, O. V. Sinyachenko, Adv. Colloid Interface Sci. 2000, 86, 39–82.
[22] W. Pronk, G. Boswinkel, K. van’t Riet, Enzyme Microb. Technol. 1992, 14,
214–220.
Sci. Technol. 2013, 115, 270–285; d) C. M. Clouthier, J. N. Pelletier, Chem.
Soc. Rev. 2012, 41, 1585–1605; e) P. Adlercreutz, Chem. Soc. Rev. 2013,
4
1
2, 6406–6436; f) K. Jaeger, B. Dijkstra, M. Reetz, Annu. Rev. Microbiol.
999, 53, 315–351.
[23] Y. Jiang, C. Guo, H. Xia, I. Mahmood, C. Liu, H. Liu, J. Mol. Catal. B 2009,
58, 103–109.
[
[
8] M. Breuer, K. Ditrich, T. Habicher, B. Hauer, M. Keßeler, R. Stꢄrmer, T. Ze-
linski, Angew. Chem. Int. Ed. 2004, 43, 788–824; Angew. Chem. 2004,
[24] M. Adamczak, Pol. J. Food Nutr. 2003, 12, 3–8.
[25] J.-P. Chen, K.-C. Chang, J. Ferment. Bioeng. 1993, 76, 98–104.
[26] J. Pleiss, M. Fischer, R. D. Schmid, Chem. Phys. Lipids 1998, 93, 67–80.
[27] a) C. G. Lopresto, V. Calabrꢆ, J. M. Woodley, P. Tufvesson, J. Mol. Catal. B
2014, 110, 64–71; b) G. D. Yadav, P. S. Lathi, J. Mol. Catal. B 2004, 27,
113–119; c) N. A. Serri, A. Kamaruddin, W. Long, Bioprocess Biosyst. Eng.
2006, 29, 253–260.
1
16, 806–843.
9] R. Kourist, P. Dominguez de Maria, K. Miyamoto, Green Chem. 2011, 13,
607–2618.
10] a) H. Stamatis, A. Xenakis, F. N. Kolisis, Biotechnol. Adv. 1999, 17, 293–
18; b) K. Martinek, A. V. Levashov, N. Klyachko, Y. L. Khmelnitski, I. V. Be-
2
[
3
rezin, Eur. J. Biochem. 1986, 155, 453–468.
11] D. Das, S. Roy, R. N. Mitra, A. Dasgupta, P. K. Das, Chem. Eur. J. 2005, 11,
[28] a) R. Schomaecker, B. H. Robinson, P. D. I. Fletcher, J. Chem. Soc. Faraday
Trans. 1 1988, 84, 4203–4212; b) T. P. Valis, A. Xenakis, F. N. Kolisis, Bioca-
tal. Biotransform. 1992, 6, 267–279.
[
[
4881–4889.
12] a) M. Laupheimer, S. Engelskirchen, K. Tauber, W. Kroutil, C. Stuben-
rauch, Tenside Surfactants Deterg. 2011, 48, 28–33; b) M. Sathishkumar,
R. Jayabalan, S. Mun, S. Yun, Bioresour. Technol. 2010, 101, 7834–7840;
c) S. Wellert, B. Tiersch, J. Koetz, A. Richardt, A. Lapp, O. Holderer, J.
Gꢅb, M.-M. Blum, C. Schulreich, R. Stehle, Eur. Biophys. J. 2011, 40, 761–
[29] D. Prazeres, F. Garcia, J. Cabral, J. Chem. Technol. Biotechnol. 1992, 53,
159–164.
[30] G. E. Crooks, G. D. Rees, B. H. Robinson, M. Svensson, G. R. Stephenson,
Biotechnol. Bioeng. 1995, 48, 190–196.
[31] P. Trodler, J. Nieveler, M. Rusnak, R. D. Schmid, J. Pleiss, J. Chromatogr. A
2008, 1179, 161–167.
774.
[13] N. Zhang, W. C. Suen, W. Windsor, L. Xiao, V. Madison, A. Zaks, Protein
Eng. 2003, 16, 599–605.
[14] U. K. Winkler, M. Stuckmann, J. Bacteriol. 1979, 138, 663–670.
[15] D. Das, P. K. Das, Langmuir 2003, 19, 9114–9119.
[16] M. Subinya, A. K. Steudle, B. Nestl, B. Nebel, B. Hauer, C. Stubenrauch, S.
Engelskirchen, Langmuir 2014, 30, 2993–3000.
Received: September 18, 2014
Published online on December 15, 2014
Chem. Eur. J. 2015, 21, 2691 – 2700
www.chemeurj.org
2700
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim