Polymer enzyme conjugates as chiral ligands for sharpless dihydroxylation of alkenes in organic solvents

Article abstract of DOI:10.1002/cbic.201402339

Conjugates of enzymes and poly(2-methyloxazoline) were used as organosoluble amphiphilic polymer nanocontainers for dissolving osmate, thereby converting the enzymes into organosoluble artificial metalloenzymes. These were shown to catalyze the dihydroxylation of different alkenes with high enantio-selectivity The highest selectivities, found for osmate complexed with laccase polymer-enzyme conjugates (PECs), even exceed those of classical Sharpless catalysts.

Full text of DOI:10.1002/cbic.201402339

DOI: 10.1002/cbic.201402339
Full Papers
Polymer Enzyme Conjugates as Chiral Ligands for
Sharpless Dihydroxylation of Alkenes in Organic Solvents
Stefan Konieczny, Melanie Leurs, and Joerg C. Tiller*[a]
Conjugates of enzymes and poly(2-methyloxazoline) were used
as organosoluble amphiphilic polymer nanocontainers for dis-
solving osmate, thereby converting the enzymes into organo-
soluble artificial metalloenzymes. These were shown to cata-
lyze the dihydroxylation of different alkenes with high enantio-
selectivity. The highest selectivities, found for osmate com-
plexed with laccase polymer–enzyme conjugates (PECs), even
exceed those of classical Sharpless catalysts.
Introduction
The synthesis of enantiomerically pure compounds is essential
for the production of pharmaceuticals and other fine chemi-
cals, and therefore is the focus of research.[1] Considering the
economy of product formation, the use of enantioselective cat-
alysts is the most efficient way to achieve this. Mostly organo-
metallic catalysts (but also enzymes) are applied for the syn-
thesis of chiral compounds.[2] The drawbacks of classical chemi-
cal asymmetric catalysts are their often complex synthesis and
purification, as well as their usually lower selectivity and per-
formance in comparison to enzymatic systems.[3]
cule host: dative anchoring, supramolecular anchoring, and co-
valent anchoring.[7,9]
The supramolecular anchoring strategy is based on the high
affinity of different molecules to the protein host.[10] Some en-
zymes and proteins, for example serum albumin, show a re-
markable ability to bind tightly to some hydrophobic mole-
cules, like fatty acids or porphyrins. Such systems have been
used to catalyze different types of asymmetric reactions,[7] for
example, reduction,[11] sulfoxidation,[12] and epoxidation.[13] An-
other often-applied method is biotin–(strept)avidin technology,
which employs the high affinity of this system.[3]
Enzymes are very attractive as catalysts because of their
unique properties: chemo-, regio-, and stereoselectivity, effi-
ciency, and high reactivity in aqueous media.[4] In addition to
their natural catalytic activities, enzymes have a well-defined
chiral matrix at the active site. This is why the activity of bio-
catalysts, particularly metalloenzymes, can be reassigned by
the formation of a complex between the active site and an or-
ganometallic catalyst. This merging of biocatalysts and synthet-
ic catalysts results in “artificial metalloenzymes”, which have
the molecular recognition ability of proteins and the broad re-
activity scope of small-molecule catalysts.[5,6]
The dative anchoring strategy employs the direct interaction
between the amino acids of the enzyme and the metal ion or
metal complex. Yamamura and Kaiser were the first to gener-
ate such a hybrid catalyst, by replacing the natural zinc of
carboxypeptidase A with copper, and they used the resulting
artificial metalloenzyme in oxidation reactions.[8b]
The goal of covalent anchoring strategies is to selectively
modify amino acids of the proteins with molecules that are
able to act as good ligands for the proposed metal catalyst.
For example, Kaiser and Lawrence generated a potent oxidore-
ductase by covalent modification of the thiol group of cysteine
C25 of Papain with flavins.[14] Another report on artificial metal-
loenzymes for asymmetric catalysis based on the covalent an-
choring strategy was published by Davies and Distefano in
1997; they described the preparation of a 1,10-phenanthroline
covalently linked to adipocyte lipid-binding protein and subse-
quent incorporation of copper.[15]
Artificial metalloenzymes are classified by the type of ap-
plied biomolecule host, the applied catalytically active metal
species, and the anchoring strategy.[7] One kind of artificial
metalloenzyme is derived from natural metalloenzymes, in
which the metal ion in the active center is substituted with
another by “dative anchoring”.[8a,b] Another variant of artificial
metalloenzyme adapts proteins or enzymes that do not natu-
rally contain a metal ion. There are three different strategies
for the incorporation of a metal complex within the biomole-
Artificial metalloenzymes, generated by different anchoring
strategies, have been applied for various kinds of reactions, for
example, sulfoxidation,[16a–c] hydrogenation,[17a–c] epoxida-
tion,[8a,18] and Diels–Alder reactions.[19a,b] There are also a few
examples of dihydroxylation of alkenes by proteins (hosts) in
aqueous media, but only for bovine serum albumin (BSA, 77%)
and streptavidin (95%)—a good enantiomeric excess (ee) was
achieved when, for example, a-methylstyrene was used as sub-
strate.[20]
[a] Dipl.-Ing. S. Konieczny,+ Dipl.-Chem. M. Leurs,+ Prof. J. C. Tiller
Chair of Biomaterials and Polymer Science
Department of Biochemical and Chemical Engineering, TU Dortmund
Emil-Figge-Strasse 66, 44227 Dortmund (Germany)
E-mail: joerg.tiller@tu-dortmund.de
[+] These authors contributed equally to this work.
Because of the natural environment of proteins and en-
zymes, most reactions with artificial metalloenzymes have
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cbic.201402339.
ChemBioChem 2015, 16, 83 – 90
83
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
been carried out in aqueous solutions.[21] There are only a few
examples in which organic cosolvents have been used.[22a–c]
This limits the range of usable substrates. Another way to
apply enzymes in organic solvents and even supercritical CO2
is by immobilizing them on solid carriers, such as acrylate
resins or amphiphilic co-networks, but this introduces diffusion
limitations.[23a–d] This can be avoided by rendering enzymes or-
ganosoluble.
ronment of the enzyme and thereby incorporated by dative
anchoring.
Water-saturated chloroform was chosen as solvent, and the
oxidizing agent was potassium hexacyanoferrate(III). First,
experiments were carried out at room temperature by using
PMOx–a-chymotrypsin conjugate as ligand for the osmate and
To overcome the limitations of artificial metalloenzymes in
aqueous systems, we have investigated organosoluble poly-
mer–enzyme conjugates (PECs) for their potential as artificial
metalloenzymes with, as an example reaction, the osmate-de-
pendent asymmetric dihydroxylation of alkenes.
Results and Discussion
The goal of this work was to prepare and explore organosolu-
ble artificial metalloenzymes. The conceptual basis was the es-
tablished synthesis of PECs by conjugating proteins to poly(2-
methyl-oxazoline) (PMOx). As a result, the enzymes become
soluble in polar aprotic organic solvents such as chloroform[24]
and show activity in both aqueous solutions and organic
media.[25]
PMOx was synthesized by living cationic ring-opening poly-
merization initiated with benzyl tosylate and terminated with
ethylenediamine (Scheme 1). To generate PECs, the resulting
poly(2-oxazoline) with the terminal amino group was coupled
via pyromellitic dianhydride (PDA) to the amino groups of the
respective protein (Scheme 2). The resulting PECs were ana-
lyzed by SDS-PAGE to estimate the extent of their degree of
modification (see Table S1 in the Supporting Information).
We chose osmate-catalyzed Sharpless dihydroxylation of al-
kenes as model reaction (Scheme 3), because this has already
been successfully applied for artificial enzymes in an aqueous
environment.[20,27]
Scheme 2. Strategy for enzyme modification with PMOx and PDA (lysozyme
structure as example from ref. [26]). n=40, R=Me.
One great advantage of organo-soluble PECs is their amphi-
philic core–shell nanocontainer character.[28] Potassium osmate
is not soluble in organic media. When adding this salt to a PEC
solution in an organic solvent, it can only be dissolved by
being complexed to a hydrophilic core, that is, the protein
part of the PEC. Thus, the osmate is forced into the chiral envi-
Scheme 3. Sharpless dihydroxylation. a) K2[OsO2(OH)4], ligand, K3[Fe(CN)6],
K2CO3, CHCl3 sat., 08C.
Scheme 1. Living cationic ring-opening polymerization of 2-methyl-2-oxazoline.
ChemBioChem 2015, 16, 83 – 90
www.chembiochem.org
84
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
styrene as the substrate. The initial activity of this catalytic
system (0.027 mm a-chymotrypsin PEC, RT, Table 1) was found
to be similar to the conversion rate of the dihydroxylation of
a-methylstyrene by osmate–streptavidin complexes in aqueous
environment.[20] In contrast to the latter, no enantioselectivity
is achieved here, that is, the formed dihydroxylate is a race-
mate. In a control experiment, in which pyridine was used as
ligand for osmate instead of the PECs, no product was found
after one day. Thus, we hypothesized that non-selective prod-
Table 1. Enantiomeric access (ee) and TON at different temperatures and
after different reaction times for the dihydroxylation[a] of styrene cata-
lyzed by 0.027 mm[b] of osmate-a-chymotrypsin PECs.
Figure 1. Enantiomeric access (ee) versus reaction temperature for asymmet-
ric dihydroxylation with a-chymotrypsin PEC as ligand. The reaction was car-
ried out with 0.027 mm enzyme, 0.2 mm K2OsO2(OH)4, 300 mm K3Fe(CN)6,
300 mm K2CO3, and 100 mm styrene for seven days.
[c]
[b,d]
[e]
T [8C]
t [d]
TONOs
TONenz
TOFenz [sÀ1
]
ee [%]
RT
9
5
1
7
4
7
6.20
0.10
0.07
0.07
45.93
0.74
0.52
0.52
5.32ꢁ10À4
1.22ꢁ10À6
1.50ꢁ10À6
8.60ꢁ10À7
rac
7.1 S
13.1 S
19.4 S
0
osmate–BSA complex in water was reported to afford an ee of
6% in the dihydroxylation of styrene,[27] whereas the respective
PEC complex in chloroform explored here forms a racemate.
This suggests that the modification of the protein with PMOx
and/or the reaction medium have an influence on enantio-
selectivity. Kokubo et al.[27] postulated that the osmate is coor-
dinated to the protein at the primary amino residues. These
functional groups are blocked by the PMOx chains in the PEC.
This forces the osmate to complex at a different position and
possibly influences the selectivity of the catalysis.
[a] Reaction carried out with 0.2 mm K2OsO2(OH)4, 300 mm K3Fe(CN)6,
300 mm K2CO3 und 100 mm styrene. [b] Calculation based on the enzyme
amount before the modification with PMOx. [c] The TONOs are defined as
mmol product per mmolK2OsO2(OH)4. [d] The TONenz are defined as mmol
product per mmol enzyme. [e] The TOFenz are defined mmol product per
mmol K2OsO2(OH)4 per second.
uct formation is caused by osmate ions randomly distributed
to the PECs, and that the number of these is significantly
higher than for that truly complexed in the protein. Further-
more, the activation energy of the reaction catalyzed by a se-
lective osmate-enzyme complex should be lower than that of
the conversion catalyzed by weakly bound osmate. Therefore,
lowering the reaction temperature should slow the latter and
favor the more selective catalytic centers. When the reaction
temperature is lowered, ee for the S configuration increases
significantly from 0% at room temperature to 19.4% at 08C
(Figure 1). An inevitable drawback of the lower temperature is
the decrease in activity, that is, in the turnover number (TON;
Table 1).
Control experiments were performed with the unmodified
commercial enzymes a-chymotrypsin and laccase, in water and
in chloroform. Also, PMOx was tested in both solvents, and ad-
ditionally a mixture of PMOx and laccase where PMOx was not
attached to the enzyme (Table 3). In all cases with the excep-
tion of the PMOx–laccase mixture, the product was obtained
but no enantioselectivity was found. Evidently, the enantio-
selectivity (above) is attributable to the catalytic activity of the
osmate–PEC complexes.
The fact that temperature influences the activity of the reac-
tion to a great extent suggests that a large number of non-
specifically bound osmate ions are accessible in the reaction
In order to broaden the concept, PECs derived from the con-
jugation of PMOx with various
enzymes were tested for asym-
Table 2. TON and ee for the dihydroxylation[a] of styrene catalyzed by various osmate-PECs at 08C.
metric dihydroxylation at 08C
(Table 2). In all cases, the dihy-
droxylation of styrene results in
product formation, but the
enantioselectivities and reaction
rates differ widely. PECs based
on lysozyme, BSA, and a lipase
from Candida antarctica catalyze
styrene conversion, but only rac-
emic products were isolated. All
other PMOx–enzyme conjugates
(with the exception of one
based on laccase) afford a favora-
ble formation of the S configura-
tion (Table 2). Interestingly, an
[b]
[c]
[b,d]
[e]
Enzyme–PMOx conjugate
cenz [mm]
t [d]
TONOs
TONenz
TOFenz 10À6 [sÀ1
]
ee [%]
lysozyme
BSA
0.048
0.011
0.027
0.024
0.018
0.021
0.012
0.026
0.016
0.018
7
7
7
6
8
8
7
7
7
7
0.29
0.08
0.07
0.08
0.37
0.19
0.07
0.11
0.09
0.17
1.21
1.45
0.52
0.67
4.11
1.81
1.17
0.85
1.13
1.89
2.00
2.40
0.86
1.29
5.95
2.62
1.93
1.41
1.87
3.13
rac
rac
19.4 S
6.2 S
9.9 S
rac
11.6 S
15.2 R
11.8 S
12.1 S
a-chymotrypsin
proteinase K
lipase (R. miehei)
lipase (Candida antartica)
lipase (Candida rugosa)
laccase
peroxidase
ADH
[a] Reaction carried out with 0.2 mm K2OsO2(OH)4, 300 mm K3Fe(CN)6, 300 mm K2CO3, and 100 mm styrene at
08C. [b] Calculation based on the enzyme amount before the modification with PMOx. [c] TONOs is defined as
mmol product per mmol K2OsO2(OH)4. [d] TONenz is defined as mmol product per mmol enzyme. [e] The TOFenz
are defined mmol product per mmol K2OsO2(OH)4 per second.
ChemBioChem 2015, 16, 83 – 90
www.chembiochem.org
85
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
In some cases enantioselectivi-
ty changes over the course of
a reaction.[29a,b] In order to check
this, the dihydroxylation of sty-
rene at 08C with an osmate/PEC
ratio of 3.8:1 was quenched
after three days instead of
seven. The isolated product
shows an ee of 99.4% for the R
product. This ee for 1-phenyl-
ethane-1,2-diol is even higher
than that found for the commer-
cially available Sharpless catalyst
Table 3. TON and ee for the dihydroxylation[a] of styrene for the control experiments at 08C, seven days.
[c]
[b,d]
[e]
Ligand
cligand [mm]
Solvent
TONOs
TONenz
TOFenz [sÀ1
]
ee [%]
a-chymotrypsin[b]
laccase[b]
0.055
0.026
2.29
0.026
0.026
2.29
CHCl3 sat.
CHCl3 sat.
CHCl3 sat.
CHCl3 sat.
water
0.08
0.07
0.09
–
43.33
0.45
0.29
0.54
–
–
333.31
–
4.79ꢁ10À7
rac
rac
rac
–
rac
rac
8.93ꢁ10À7
PMOx
–
–
PMOx+laccase[b]
laccase[b]
5.51ꢁ10À4
–
PMOx
water
[a] Reaction carried out with 0.2 mm K2OsO2(OH)4, 300 mm K3Fe(CN)6, 300 mm K2CO3, and 100 mm styrene.
[b] Calculation based on the enzyme amount before the modification with PMOx. [c] The TONOs are defined as
mmol product per mmol K2OsO2(OH)4. [d] The TONenz are defined as mmol product per mmol used enzyme.
rac: racemate. [e] The TOFenz are defined as mmol product per mmol K2OsO2(OH)4 per second. –: No product
formation could be observed by HPLC analysis.
under
standard
conditions
(97%),[30a,b] thus showing the
mixture. In order to enlarge the content of specifically com-
plexed osmate, the osmate/PEC ratio was lowered. PECs of a-
chymotrypsin and PMOx were chosen for initial experiments. It
was found that decreasing the osmate/PEC ratio from 7.4:1 to
1.8:1 results in an ee change from 17.4 to 36%, and the
osmate activity increases by a factor of 7 (TONOs =0.49,
TONenz =0.88). Given the assumption that osmate ions are
more active and selective when specifically complexed, the
experiment indicates that a lower osmate/PEC ratio might
indeed result in a higher content of specifically bound osmate
ions.
great potential of the laccase–PEC/osmate artificial metalloen-
zymes in organic solvents. The discovery that the ee unfortu-
nately decreases with increasing reaction time indicates that
the reaction takes place in a progressively less enantiocontrol-
ling environment. This might be due to enzyme denaturation
over time by the co-oxidizing agent potassium hexacyanoferra-
te(III) which is present in every used reaction mixture to regen-
erate the osmate catalyst during the reactions.
Next, the PECs with laccase were investigated more thor-
oughly because this is the only enzyme that preferentially di-
rects the production of the R enantiomer. In addition, the fact
that laccase is a metalloenzyme offers the possibility of metal–
metal exchange, and thus the osmate would be inserted into
a well-defined chiral environment that might strongly direct
the enantioselectivity of the reaction. Such substitutions were
successfully achieved for different enzymes with other metal
ions, for example, a zinc–copper exchange in carboxypepti-
dase A[8b] or the substitution of zinc with manganese in car-
bonic anhydrase.[8a]
Figure 2. Dependency of the ee on osmate/PEC ratio for asymmetric di-
hydroxylation with laccase–PEC. Reactions were carried out with 0.2 mm
K2OsO2(OH)4, 300 mm K3Fe(CN)6, 300 mm K2CO3 and 100 mm styrene at 08C
for seven days.
First, the osmate/laccase PEC ratio was varied by using con-
stant osmate concentration. The ee for the R product rises
dramatically with decreasing osmate/PEC ratio (maximum ee
92.6% at osmate/PEC ratio 1.9:1;
(Figure 2). Interestingly, the ac-
Table 4. TON and ee for the dihydroxylation[a] of styrene catalyzed by osmate-laccase PECs at different osmate/
tivity of the osmate also increas-
es dramatically (>60-fold) from
the highest to the lowest
osmate/PEC ratio (Table 4). In
order to further improve the ee
value of the products, the reac-
tion temperature was lowered to
À158C with an osmate/PEC ratio
of 3.8:1. As expected, the ee in-
creases significantly (from 84.6
to 94.2%; Table 4), but the
osmate activity drops more than
20-fold (TON, 0.6 per seven
days).
enzyme ratios and at two temperatures.
[b]
[c]
[b,d]
cenz [mmolLÀ1
]
Osmate/PEC ratio
T [8C]
TONOs
TONenz
TOFenz [sÀ1
]
ee [%]
0.026
7.7
5.1
3.8
3.8
1.9
1.9
3.8
0
0
0
0
0
0
À15
0.11
0.12
0.62
16.02
7.07
41.76
1.78
0.85
0.62
2.34
60.45
13.47
79.56
6.72
1.41ꢁ10À6
1.03ꢁ10À6
9.03ꢁ10À6
1.00ꢁ10À4
2.23ꢁ10À5
1.32ꢁ10À4
3.70ꢁ10À6
15.2 R
43.3 R
99.4 R
84.6 R
92.6 R
98.4 R
94.2 R
0.039
0.053[e]
0.053
0.105
0.105[f]
0.053[g]
[a] Reaction carried out with 0.2 mm K2OsO2(OH)4, 300 mm K3Fe(CN)6, 300 mm K2CO3 and 100 mm styrene over
seven days. [b] Calculation based on the enzyme amount before the modification with PMOx. [c] The TONOs are
defined as mmol product per mmolK2OsO2(OH)4. [d] The TONenz are defined as mmol product per mmol
enzyme. [e] Reaction time of three days. [f] Enzyme conjugate was dialyzed against 10 mm EDTA for 24 h, fol-
lowed by further dialysis against water. [g] Reaction time of 21 days. The TOFenz are defined as mmol product
per mmol K2OsO2(OH)4 per second.
ChemBioChem 2015, 16, 83 – 90
www.chembiochem.org
86
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Table 5. TON and ee for the dihydroxylation[a] of alkenes with laccase
PECs (osmate/enzyme 1.9:1) at 08C.
[c]
[b,d]
[e]
Alkene
TONOs
7.07
TONenz
13.47
TOFenz [sÀ1
]
ee [%]
92.6 R
2.23ꢁ10À5
8.66ꢁ10À6
3.39ꢁ10À5
1.98ꢁ10À6
2.78
10.86
0.64
5.24
20.49
1.20
70.8 1R,2S[f]
75.4 1R,2S[f]
87.4 R
–
–
–
–
–
–
–
–
Figure 3. A) Ribbon diagram of laccase according to Piontek et al.;[33] Copper
atoms are shown as spheres. B) Suggested scheme of copper type I to
osmate exchange by EDTA.
1.13
2.13
3.52ꢁ10À6
rac
[a] Reaction carried out with 0.2 mm K2OsO2(OH)4, 300 mm K3Fe(CN)6,
300 mm K2CO3, and 100 mm styrene. [b] Calculation based on the enzyme
amount before the modification with PMOx. [c] TONOs is defined as mmol
product per mmolK2OsO2(OH)4. [d] TONenz is defined as mmol product
per mmol enzyme. [e] The TOFenz are defined as mmol product per mmol
K2OsO2(OH)4 per second. [f] ee values calculated by comparison of the
amounts of the cis enantiomers (1R,2S and 1S,2R).
In order to explore the possibility that osmate might be lo-
cated in the active site, copper was deliberately removed by
dialysis against aqueous EDTA (Figure 3B). The success of this
method was judged by the loss of the brownish color of the
protein that arises from the copper ions.[33] The resulting
apoenzyme was used as ligand in the osmate-catalyzed dihy-
droxylation of styrene under the optimal conditions for enan-
tioselectivity and activity (as described above). EDTA pre-treat-
ment of laccase enables the protein to act as an even more
enantioselective ligand (ee increase from 92.6 to 98.4%), and
the activity is also dramatically increased (sixfold; Table 4). Both
observations support the hypothesis that osmate is located at
the active site of laccase in the copper type I position, thus
forming a new artificial, organo-soluble metalloenzyme that is
active and highly enantioselective in the dihydroxylation of di-
verse alkenes in organic solvents. This experiment also shows
that osmate is not able to fully exchange copper type I with-
out the aid of EDTA.
Because of the good results with styrene as a reactant,
a range of other alkenes as substrates for Sharpless dihydroxy-
lation were explored. All reactions were carried out in the pres-
ence of of PMOx–laccase PECs with an osmate to enzyme ratio
of 1.9:1 (Table 5). 1H-Indene (70.8% ee), 1,2-dihydronaphtalene
(75.4% ee), and allylphenylether (87.4% ee) are converted with
higher enantiomeric excess values than previously reported.
Kçhler et al. achieved 45, 41, and 82% ee, when using strepta-
vidin mutants in an aqueous environment.[20] For the standard
Sharpless system maximum ee values of 53% for 2,3-dihydro-
1H-indene-1,2-diol (1R,2S) and 56% ee for 1,2,3,4-tetrahydro-
naphthalene-1,2-diol (1R,2S) were found.[31] Two of the applied
alkenes, a-methylstyrene and p-methylstyrene, are not convert-
ed under the tested reaction conditions. This contrasts with
previous work, where the authors applied streptavidin as
ligand for this reaction and found no activity difference for the
osmate–enzyme complex towards these substrates (and, for
instance, 1H-indene).[20] It indicates that, here, osmate is rather
specifically located within the PMOx—laccase PECs.
Conclusion
Conjugates of enzymes and poly(2-methyloxazoline) are solu-
ble in organic media such as chloroform. This offers the possi-
bility to use them as amphiphilic polymer nanocontainers for
dissolving inorganic salts in this solvent. As the ions can only
be dissolved by being complexed to the hydrophilic protein
core of the nanocontainers, they are forced to interact specifi-
cally with the protein as a chiral ligand. We could show that
osmate dissolved in this way in chloroform catalyzes the di-
hydroxylation of alkenes and affords highly enantioselective
product formation. Optimizing the reaction conditions with
laccase PECs strongly indicates that osmate is in the active site,
thereby affording product enantioselectivity that exceeds that
of classical Sharpless catalysts. Thus, we could transfer the con-
cept of artificial metalloenzymes from water to organic sol-
Laccase is a metalloprotein with four copper ions in the
active site; one is known to be easily removed by complexing
agents such as EDTA (Figure 3).[32] Other complexing agents
could be the ethylenediamine (EDA) groups of poly(2-methyl-
oxazoline) that were used to form PECs with laccase. The
cavity resulting from removal of the copper ion could be a loca-
tion to fit the added osmate.
ChemBioChem 2015, 16, 83 – 90
www.chembiochem.org
87
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
500 GC (PerkinElmer; column: CP-Sil 8 CB (Varian)) with nitrogen as
carrier gas (1 mLminÀ1) was used. The injection temperature was
2308C (split ratio 5:1); the FID detector temperature was set to
2808C. The following temperature program was applied: 1008C
(1 min), heating to 2508C (20 KminÀ1), 2508C (2 min).
vents with even higher selectivity. The osmate PECs show high
selectivity, but their activity is rather low. A Sharpless catalytic
system based on osmate with a (DHQD)2 PHAL ligand showed
a reaction rate in the pseudo first-order dihydroxylation of sty-
rene in tBuOH of 3.3ꢁ10À2 sÀ1 [31]
This is about 200 times faster
.
than the highest achieved with the present system. However,
taking into account that the enzyme ligand concentration is 20
times lower and the reaction temperature is 258C lower, the
activities of the two systems are comparable. Optimization is
thinkable by variation of the polymer–enzyme linkers, purifica-
tion of the enzyme conjugates, and further adjustment to the
osmate/enzyme ratio. In future work we will extend the scope
of reactions by varying the conjugating polymer, the enzyme,
and the metal, in order to broaden this new concept.
The turnover numbers of the dihydroxylation of all other alkenes
were determined quantitatively by HPLC (LaChrom Elite, Hitachi)
with a Purospher RP-18 column (Merck Millipore, LiChroCART 125–
4,6 cartridge, 5 mm) and a diode array detector (210 nm). Water/
acetonitrile (1 mLminÀ1) was used as the eluent, with the following
gradient: 10% acetonitrile (5 min), increase to to 90% acetonitrile
(over 10 min), 90% acetonitrile (5 min), reduce to 10% acetonitrile
(over 5 min).
The enantiomeric excess was determined by a LaChrom HPLC
system equipped with a CHIRALPAK IC column (Daicel Corp, Osaka,
Japan). n-Heptane/isopropanol (9:1) was used as the eluent at
0.5 mLminÀ1 (1-phenylethane-1,2-diol) or 1 mLminÀ1 (all other al-
kenes). A diode array detector was used at 216 nm (1-phenyl-
ethane-1,2-diol) or 210 nm (all other diols).
Experimental Section
Materials: Chloroform was dried with activated alumina and sub-
sequent distillation. The obtained chloroform contained less than
1 ppm water (determined by Karl–Fischer titration) and was stored
under argon over a 4 ꢂ molecular sieve. Monomeric 2-methyl-2-ox-
azoline (MOx) was distilled twice with CaH2 under reduced pres-
sure and an argon atmosphere, and stored under argon at À208C
over a 4 ꢂ molecular sieve. N,N-dimethylformamide (DMF) was
dried, freed from amine, and stored under argon over a 4 ꢂ molec-
ular sieve. All chemicals were purchased from Sigma–Aldrich, Ap-
plichem (Darmstadt, Germany), Merck Millipore, ABCR (Karlsruhe,
Germany), Acros, Armar Chemicals (Leipzig Germany), and Carl
Roth, and were of analytical grade or purer and used without fur-
ther modification (Table 6).
Synthesis of benzyl tosylate: A modified procedure of Kazemi
et al was used.[34] Benzyl alcohol (1.04 mL, 10 mmol) and tosyl chlo-
ride (TsCl; 2.86 g, 15 mmol) were added to dry K2CO3 (5 g,
36 mmol) in a mortar and ground vigorously for 5 min. Powdered
KOH (2.81 g, 50 mmol) was added and again vigorously ground for
2 min. The product was dissolved in diethyl ether, and the solids
were filtered off. The crude product was further purified by two-
fold recrystallization from n-heptane.
Synthesis of poly(2-methyl-oxazoline): All reactions were carried
out under argon. Benzyl tosylate (513.2 mg, 1.96 mmol) and 2-
methyl-2-oxazoline (5 mL, 58.75 mmol) were dissolved in dry
chloroform (20 mL) at room temperature, then heated (1008C, 4 h)
in a closed vessel in a Discover microwave synthesis reactor (CEM
Microwave Technology, Buckingham, UK). After cooling to room
temperature, ethylenediamine (EDA; 5 mL, 4.5 g, 75 mmol) was
added, and the reaction mixture was stirred for 48 h at 408C. The
raw polymer was purified by precipitation (ꢁ3) in diethyl ether/
chloroform, and further dialyzed against distilled methanol for two
days by using ZelluTrans cellulose membranes (Carl Roth; MWCO
1000 Da). After dialysis the polymers were dried under reduced
pressure to obtain slightly yellow solids that were characterized by
1H NMR spectroscopy (DPNMR 40) and size-exclusion chromatogra-
Measurements: 1H NMR spectra were recorded in [D6]DMSO in
a Bruker DRX-400 spectrometer (5 mm sample head, 400.13 MHz).
Size-exclusion chromatography (SEC) was performed on a Viscotek
GPCMax system (Malvern Instruments, Malvern, UK) equipped with
a refractive index (RI) detector in saline DMF (20 mm LiBr) at 608C
with a flow rate of 0.7 mLminÀ1. Two TSKgel GMHHR-M 7.8ꢁ
300 mm columns and one precolumn were used. The calibration
was performed with polystyrene standards (Viscotek).
The turnover number of the dihydroxylation of styrene was deter-
mined quantitatively by gas chromatography (GC) measurements
with n-tetradecane (0.001 mm) as the internal standard. A Clarus
Table 6. Purchased enzymes.
Enzyme (organism)
EC
Activity[a] [UmgÀ1] (pH, T, substrate)
Supplier
lipase (C. rugosa)
lipase (C. antartica)
lipase (R. miehei)
lysozyme (hen eggwhite)
alcohol dehydrogenase (Saccharomyces cerevisiae)
BSA
3.1.1.3
3.1.1.3
3.1.1.3
3.2.1.17
1.1.1.1
ꢀ700 (pH 7.2, 378C, triglyceride)
ꢀ9 (pH 8.0, 408C, butyric acid)
ꢀ20
Sigma–Aldrich
Sigma–Aldrich
Novozymes
AppliChem
Sigma–Aldrich
Roth
ꢀ20000[b] (pH 6.2)
ꢀ300 (pH 8.8, 258C, ethanol)
laccase (Trametes versicolor)
proteinase K (Tritirachium album)
peroxidase (horseradish)
a-chymotrypsin (bovine pancreas)
1.10.3.2
3.4.21.64
1.11.1.7
3.4.21.1
ꢀ10 (pH 4.5, 258C, catechol)
Sigma–Aldrich
Sigma–Aldrich
Sigma–Aldrich
Sigma–Aldrich
ꢀ30[c] (pH 7.5, 378C, hemoglobin)
150–250[d] (pH 6.0, 208C, pyrogallol)
ꢀ40 (pH 7.8, 258C, N-benzoyl-l-tyrosine ethyl ester)
[a] 1 U corresponds to the amount of enzyme that converts 1 mmol substrate per minute under the given reaction conditions, if not otherwise indicated.
[b] One unit will produce a DA450 of 0.001 per min at pH 6.24 and 258C when using a suspension of Micrococcus lysodeikticus as substrate in a 2.6 mL reac-
tion mixture (1 cm light path). [c] One unit will hydrolyze urea-denatured hemoglobin to produce color equivalent to 1.0 mmol of tyrosine per min at
pH 7.5 and 378C (color by Folin–Ciocalteu reagent). [d] One pyrogallol unit will form 1.0 mg purpurogallin from pyrogallol in 20 s at pH 6.0 and 208C.
ChemBioChem 2015, 16, 83 – 90
www.chembiochem.org
88
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
phy (SEC, Mn =4200 Da, Mw =4600 Da, DPSEC 48, PDI 1.1). 1H NMR
data for PMOx are as follows.
K3Fe(CN)6 (980 mg, 3 mmol), and finally alkene (1 mmol) were
added, and the reaction mixture was vigorously stirred for 7 days
at 08C. Product was isolated as for the dihydroxylation of styrene.
After removal of the solvent, the obtained product was character-
ized by HPLC.
PMOx: 1H NMR ([D6]DMSO): d=7.40–7.20 (brs, 5H; CH2-C6H5),
4.60–4.50 (brs, 2H, CH2-C6H5), 3.65–3.10 (brs, n·4H, N(CH2)2), 2.70–
2.55 (brs, 6H, CH2-NH-CH2-CH2-NH2), 2.10–1.85 ppm (brs, n·3H,
NCO-CH3).
Acknowledgements
Enzyme purification: Rhizomucor miehei lipase was purified by di-
alysis (15 kDa MWCO) against acetate buffer (100 mm, pH 5.7) for
four days and then acetate buffer (1 mm, pH 5.7) for two days and
subsequent lyophilization. All other enzymes were used as com-
mercial powder without further purification.
We thank Thorsten Moll for the SEC measurements. This research
has received funding from the Ministry of Innovation, Science
and Research of North Rhine–Westphalia within the CLIB-Gradu-
ate Cluster Industrial Biotechnology, contract no: 314–108 001
08.
Conjugation of enzymes with poly(2-methyl-oxazoline): The
polymer amount (tenfold molar excess) was calculated according
to the protein/enzyme primary amino groups (Table S1). The
amount (0.8 equiv) of pyromellitic dianhydride (PDA) was calculat-
ed by reference to the primary amino group at the end of the
polymer chain. PDA was dissolved in dry and amine-free DMF
(5 mL), and PMOx was added. The mixture was allowed to react
(12 h, RT). The enzyme (7 mg) was dissolved in carbonate buffer
(1 mL, 0.5 mm, pH 9.65) followed by the addition of the polymer/
PDA mixture (5 mL). The mixture was allowed to react for 24 h at
RT. Then, the mixture was dialyzed for four days against water in
ZelluTrans cellulose membranes (MWCO=10000 Da), followed by
lyophilization.
Keywords: artificial metalloenzymes · asymmetric catalysis ·
organic solvents · polymer enzyme conjugates · Sharpless
dihydroxylation
[1] S. C. Stinson, Chem. Eng. News 1999, 77, 57–65.
[2] R. Dittmeyer, W. Keim, G. Kreysa, A. Oberholz, Chemische Technik, Vol. 1:
Prozesse und Produkte, 5th ed., Wiley-VCH, Weinheim, 2004.
[3] T. R. Ward, Chem. Eur. J. 2005, 11, 3798–3804.
[4] A. J. J. Straathof, S. Panke, A. Schmid, Curr. Opin. Biotechnol. 2002, 13,
548–556.
[5] F. Yu, V. M. Cangelosi, M. L. Zastrow, M. Tegoni, J. S. Plegaria, A. G. Tebo,
C. S. Mocny, L. Ruckthong, H. Qayyum, V. L. Pecoraro, Chem. Rev. 2014,
114, 3495–3578.
[6] P. J. Deuss, R. den Heeten, W. Laan, P. C. J. Kamer, Chem. Eur. J. 2011, 17,
4680–4698.
[7] J. Steinreiber, T. R. Ward, Coord. Chem. Rev. 2008, 252, 751–766.
[8] a) A. Fernꢃndez-Gacio, A. Codina, J. Fastrez, O. Riant, P. Soumillion,
ChemBioChem 2006, 7, 1013–1016; b) K. Yamamura, E. T. Kaiser, J.
Chem. Soc. Chem. Commun. 1976, 830–831.
[9] C. Letondor, T. R. Ward, ChemBioChem 2006, 7, 1845–1852.
[10] R. Krꢄmer, Angew. Chem. Int. Ed. 2006, 45, 858–860, Angew. Chem.
2006, 118, 872–874.
Sharpless dihydroxylation of styrene with pyridine as a chiral
ligand: The dihydroxylation of styrene was performed under modi-
fied Sharpless conditions.[30a] K2CO3 (410 mg, 3 mmol), ground
K3Fe(CN)6 (980 mg, 3 mmol), pyridine (0.82 mL, 0.01 mmol), and
K2OsO2(OH)4 (0.74 mg, 0.002 mmol) were added to water-saturated
chloroform (10 mL; water content ~1300 ppm as determined by
Karl–Fischer titration). Finally, styrene (115 mL, 1 mmol) was added,
and the reaction mixture was stirred vigorously at room tempera-
ture for one day. The solvent was evaporated, the residue was
dissolved in water (25 mL), and the solution was extracted with
distilled ethyl acetate (25 mL, ꢁ4). The organic phases were com-
bined, and the solvent was evaporated. The obtained solid was
characterized by HPLC and GC.
[11] T. Sugimoto, T. Kokubo, Y. Matsumura, J. Miyazaki, S. Tanimoto, M.
Okano, Bioorg. Chem. 1981, 10, 104–113.
[12] S. Colonna, N. Gaggero, M. Leone, Tetrahedron 1991, 47, 8385–8398.
[13] S. Colonna, A. Manfredi, Tetrahedron Lett. 1986, 27, 387–390.
[14] E. T. Kaiser, D. S. Lawrence, Science 1984, 226, 505–511.
[15] R. R. Davies, M. D. Distefano, J. Am. Chem. Soc. 1997, 119, 11643–11652.
[16] a) P. K. Das, J. M. M. Caaveiro, S. Luque, A. M. Klibanov, Biotechnol.
Bioeng. 2002, 78, 104–109; b) C. Esmieu, M. V. Cherrier, P. Amara, E. Gir-
genti, C. Marchi-Delapierre, F. Oddon, M. Iannello, A. Jorge-Robin, C.
Cavazza, M. Mꢅnage, Angew. Chem. Int. Ed. 2013, 52, 3922–3925;
Angew. 2013, 125, 4014–4017; c) K. Ogura, M. Fujita, H. Iida, Tetrahedron
Lett. 1980, 21, 2233–2239.
[17] a) J. Collot, J. Gradinaru, N. Humbert, M. Skander, A. Zocchi, T. R. Ward,
J. Am. Chem. Soc. 2003, 125, 9030–9031; b) M. T. Reetz, J. J.-P. Peyera-
lans, A. Maichele, Y. Fu, M. Maywald, Chem. Commun. 2006, 4318–4320;
c) M. E. Wilson, G. M. Whitesides, J. Am. Chem. Soc. 1978, 100, 306–307.
[18] K. Okrasa, R. J. Kazlauskas, Chem. Eur. J. 2006, 12, 1587–1596.
[19] a) J. Podtetenieff, A. Taglieber, E. Bill, E. J. Reijerse, M. T. T. Reetz, Angew.
Chem. Int. Ed. 2010, 49, 5151–5155; b) M. T. Reetz, N. Jiao, Angew.
Chem. Int. Ed. 2006, 45, 2416–2419; Angew. Chem. 2006, 118, 2476–
2479.
[20] V. Kçhler, J. Mao, T. Heinisch, A. Pordea, A. Sardo, Y. M. Wilson, L. Knçrr,
M. Creus, J.-C. Prost, T. Schirmer, T. R. Ward, Angew. Chem. Int. Ed. 2011,
50, 10863–10866; Angew. Chem. 2011, 123, 11055–11058.
[21] C. M. Thomas, T. R. Ward, Chem. Soc. Rev. 2005, 34, 337–346.
[22] a) M. Allard, C. Dupont, V. M. Robles, N. Doucet, A. Lledꢆs, J.-D. Marꢅ-
chal, A. Urvoas, J.-P. Mahy, R. Ricoux, ChemBioChem 2012, 13, 240–251;
b) S. Colonna, A. Manfredi, M. Spadoni, Tetrahedron Lett. 1987, 28,
1577–1580; c) W. Laan, B. K. MuÇoz, R. den Heeten, P. C. J. Kamer,
ChemBioChem 2010, 11, 1236–1239.
Sharpless dihydroxylation of styrene with polymer–enzyme con-
jugates: PECs were dissolved in different amounts in water-saturat-
ed CHCl3 (10 mL; water content ~1300 ppm, as determined by
Karl–Fischer titration). K2OsO2(OH)4 (0.74 mg, 0.002 mmol), K2CO3
(410 mg, 3 mmol), ground K3Fe(CN)6 (980 mg, 3 mmol), and finally
styrene (115 mL, 1 mmol) were added, and the reaction mixture
was vigorously stirred at different temperatures for between one
day at RT and three weeks at À158C. The reaction was stopped by
the addition of sodium sulfite (700 mg, 5.5 mmol), and the temper-
ature was slowly increased to room temperature (1 h). The solvent
was evaporated, and the obtained residue was dissolved in water
(25 mL) followed by extraction (as above). After removal of the sol-
vent, the obtained solid was characterized by HPLC and GC.
Additional experiments were performed with pure enzyme
powder, pure amine-terminated PMOx as well as a mixture of both
as ligands instead of PEC, with water and water-saturated chloro-
form as the solvent.
Sharpless dihydroxylation of different alkenes with polymer–lac-
case conjugates: Polymer–laccase conjugate (laccase (0.53 mmol);
calculation based on the enzyme amount before the modification
with PMOx) was dissolved in water-saturated CHCl3 (10 mL; water
content ~1300 ppm, as determined by Karl–Fischer titration)).
K2OsO2(OH)4 (0.37 mg, 0.001 mmol), K2CO3 (410 mg, 3 mmol),
ChemBioChem 2015, 16, 83 – 90
www.chembiochem.org
89
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[23] a) P. Adlercreutz, Chem. Soc. Rev. 2013, 42, 6406–6436; b) N. Bruns, W.
Bannwarth, J. C. Tiller, Biotechnol. Bioeng. 2008, 101, 19–26; c) G. Carrea,
S. Riva, Angew. Chem. Int. Ed. 2000, 39, 2226–2254; Angew. Chem. 2000,
112, 2312–2341; d) I. Schoenfeld, S. Dech, B. Ryabenky, B. Daniel, B. Glo-
wacki, R. Ladisch, J. C. Tiller, Biotechnol. Bioeng. 2013, 110, 2333–2342.
[24] S. Konieczny, C. P. Fik, N. J. H. Averesch, J. C. Tiller, J. Biotechnol. 2012,
159, 195–203.
[25] S. Konieczny, C. Krumm, D. Doert, K. Neufeld, J. C. Tiller, J. Biotechnol.
2014, 181, 55–63.
[26] N. Sukumar, B. K. Biswal, M. Vijayan, Acta Crystallogr. Sect. D Biol. Crystal-
logr. 1999, 55, 934–937.
[30] a) H. C. Kolb, M. S. Van Nieuwenhze, K. B. Sharpless, Chem. Rev. 1994, 94,
2483–2547; b) K. B. Sharpless, W. Amberg, Y. L. Bennani, G. A. Crispino,
J. Hartung, K.-S. Jeong, H.-L. Kwong, K. Morikawa, Z.-M. Wang, D. Xu, X.-
L. Zhang, J. Org. Chem. 1992, 57, 2768–2771.
[31] H. Becker, S. B. King, M. Taniguchi, K. P. M. Vanhessche, K. B. Sharpless, J.
Org. Chem. 1995, 60, 3940–3941.
[32] A. I. Yaropolov, O. V. Skorobogat’ko, S. S. Vartanov, S. D. Varfolomeyev,
Appl. Biochem. Biotechnol. 1994, 49, 257–280.
[33] K. Piontek, M. Antorini, T. Choinowski, J. Biol. Chem. 2002, 277, 37663–
37669.
[34] F. Kazemi, A. R. Massah, M. Javaherian, Tetrahedron 2007, 63, 5083–
5087.
[27] T. Kokubo, T. Sugimoto, T. Uchida, S. Tanimoto, M. Okano, J. Chem. Soc.
Chem. Commun. 1983, 769–770.
[28] C. Hon Ho, M. Thiel, S. Celik, E. K. Odermatt, I. Berndt, R. Thomann, J. C.
Tiller, Polymer 2012, 53, 4623–4630.
[29] a) E. Keinan, E. K. Hafeli, K. K. Seth, R. Lamed, J. Am. Chem. Soc. 1986,
108, 162–169; b) D. Vasic-Racki, U. Kragl, D. Conrad, C. Q. Wandrey,
Chem. Biochem. Eng. 1998, 12, 87–95.
Received: June 27, 2014
Published online on November 24, 2014
ChemBioChem 2015, 16, 83 – 90
www.chembiochem.org
90
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim