Stereoselective benzylic hydroxylation of alkylbenzenes and epoxidation of styrene derivatives catalyzed by the peroxygenase of Agrocybe aegerita

Article abstract of DOI:10.1039/c1gc16173c

Here we report on the stereoselective benzylic hydroxylation and C1-C2 epoxidation of alkylbenzenes and styrene derivatives, respectively, by a heme-thiolate peroxygenase (EC 1.11.2.1) from the fungus Agrocybe aegerita. Benzylic hydroxylation led exclusively to the (R)-1-phenylalkanols. For (R)-1-phenylethanol, (R)-1-phenylpropanol and (R)-1-tetralol, the ee reached >99%. For longer chain lengths, the enantiomeric excesses (ee) and total turnover numbers (TTN) decreased while the number of by-products, e.g. 1-phenylketones, increased. Epoxidation of straight chain and cyclic styrene derivatives gave a heterogeneous picture and resulted in moderate to excellent ee values and TTN: e.g., in the case of (1R,2S)-cis-β-methylstyrene oxide formation, an ee >99% and a TTN of 110000 was achieved. Hydroxylation and epoxidation were true peroxygenations, which was demonstrated by the incorporation of ¹⁸O from H₂¹⁸O₂ into the products. The use of fed-batch devices and varying feeding strategies for the substrate and co-substrate turned out to be a suitable approach to optimize peroxygenase catalysis.

Full text of DOI:10.1039/c1gc16173c

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2012, 14, 440
www.rsc.org/greenchem
PAPER
Stereoselective benzylic hydroxylation of alkylbenzenes and epoxidation of
styrene derivatives catalyzed by the peroxygenase of Agrocybe aegerita†
Martin Kluge,*a Rene´ Ullrich,a Katrin Scheibnerb and Martin Hofrichtera
Received 20th September 2011, Accepted 11th November 2011
DOI: 10.1039/c1gc16173c
Here we report on the stereoselective benzylic hydroxylation and C1–C2 epoxidation of
alkylbenzenes and styrene derivatives, respectively, by a heme-thiolate peroxygenase (EC 1.11.2.1)
from the fungus Agrocybe aegerita. Benzylic hydroxylation led exclusively to the (R)-1-phenyl-
alkanols. For (R)-1-phenylethanol, (R)-1-phenylpropanol and (R)-1-tetralol, the ee reached
>99%. For longer chain lengths, the enantiomeric excesses (ee) and total turnover numbers (TTN)
decreased while the number of by-products, e.g. 1-phenylketones, increased. Epoxidation of
straight chain and cyclic styrene derivatives gave a heterogeneous picture and resulted in moderate
to excellent ee values and TTN: e.g., in the case of (1R,2S)-cis-b-methylstyrene oxide formation,
an ee >99% and a TTN of 110 000 was achieved. Hydroxylation and epoxidation were true
peroxygenations, which was demonstrated by the incorporation of 18O from H218O2 into the
products. The use of fed-batch devices and varying feeding strategies for the substrate and
co-substrate turned out to be a suitable approach to optimize peroxygenase catalysis.
tive epoxidation of terminal olefins such as unsubstituted styrene
still remains deficient by conventional catalysis. Approaches
like the Shi epoxidation applying a fructose derived chiral
Introduction
Optically pure a-hydroxy alkylbenzenes and epoxides as such
are of high importance as building blocks used in the production
ketone revealed enantiomeric excesses of up to 81% but require
of fine chemicals, odorous substances and bioactive molecules
high percentages of chiral ketone catalysts for control of
like pharmaceuticals and antibiotics.1,2 Chiral epoxides are
configurational orientation.14,15 The use of biomimetic catalysts
of particular interest, since they can undergo stereospecific
like iron porphyrins is efficient but still laborious and requires
ring opening reactions3 wherein two adjacent stereocenters are
unfavorably high ratios of catalyst and reactant.16
formed in one reaction.4–6
Classical preparation of optically active secondary alcohols
Over the last decades, biotechnological research in this field
has been focusing on biotransformations using whole microbial
including a-hydroxyalkyl benzenes applies catalysis in reductive
asymmetric transfer hydrogenation of prochiral ketones using
transition metal catalysts like Ru-, Rh- and Pt-based complexes
possessing chiral ligands.7,8 The disadvantages of these processes
are the use of expensive and environmentally problematic heavy
metals and harsh reaction conditions that affect the preparation
of labile compounds, even though progress in asymmetric
hydrogenation in aqueous solution has been made.9 Asymmetric
epoxidation has made much progress via the Sharpless10 and the
Jacobson11 route to obtain optically active epoxides from allylic
alcohols and isolated alkenes, respectively.12,13 The stereoselec-
cells or isolated enzymes.17 Highly stereoselective benzylic
hydroxylation can be achieved using fungal mycelium18 or bac-
terial suspensions.19 Recent enzymatic approaches to synthesize
chiral alcohols comprise oxidative kinetic resolution of racemic
secondary alcohols by dehydrogenases,20 the reductive stereos-
elective conversion of prochiral a-ketones by dehydrogenases,21
the stereoselective esterification by esterases22 and the enantiose-
lective hydrolysis of the appropriate esters by lipases.23 Optically
active alcohols and epoxides can be prepared from non-activated
hydrocarbons by enzymatic oxy-functionalizations (hydroxyla-
tion/epoxidation) using cytochrome P450s,24–26 chloroperoxi-
dase (CfuCPO) from Caldariomyces (Leptoxyphium) fumago27
and ethylbenzene dehydrogenase (EBDH) from Aromatoleum
aromaticum.28,29
aDepartment of Bio- and Environmental Sciences, International Graduate
School of Zittau, Markt 23, 02763, Zittau, Germany.
The extracellular Agrocybe aegerita aromatic peroxygenase
(AaeAPO) belonging to a novel sub-subclass of peroxidases
transferring peroxide-borne oxygen to substrates (EC 1.11.2),
also referred to as Agrocybe aegerita peroxidase/peroxygenase
(AaP)30–32 or unspecific peroxygenase (EC 1.11.2.1; www.chem.
E-mail: kluge@ihi-zittau.de; Fax: 49 3583 612734; Tel: 49 3583 612719
bUnit of Enzyme Technology, Faculty of Chemsitry, Biotechnology and
Process Engineering, Lausitz University of Applied Sciences,
Großenhainer Str. 57, 01968, Senftenberg, Germany
† Electronic supplementary information (ESI) available: Experimental
and analytical details, preparation of reference compounds. See DOI:
10.1039/c1gc16173c
qmul.ac.uk/iubmb/enzyme/EC1/11/2/1.html)
catalyzes
440 | Green Chem., 2012, 14, 440–446
This journal is The Royal Society of Chemistry 2012
©

View Article Online
Table 1 Results of the enzymatic conversion of benzylic substrates into chiral products by AaeAPO. Reactions were carried out using 0.5 U mL-1
(155 nM) AaeAPO, 1 mM of the couple substrate and hydrogen peroxide in 10 mM potassium phosphate buffer (pH 7.0) containing 20% MeCN
Reaction scheme
Entry
R =
Conversion (%)
TTN (103)
Config.
ee (%)
n-Alkylbenzenes
1
2
3
4
CH3
95
64
52
8.4
10.6
7.1
5.8
0.9
R-(+)
R-(+)
R-(+)
n.d.a
>99
>99
~40
n.d.
C2H5
n-C3H7
n-C4H9
Cycloalkylbenzenes
5
6
CH2
(CH2)2
77
85
8.6
9.4
R-(-)
R-(-)
87
>99
Styrenes
R1 R2 R3
7
H
Me H
H
H
H
H
H
71
96
19
95
7.9
10.7
2.1
n.d.
n.d.
S,S-(-)
1R,2S-(+)
7
29
>98
>99
8
9b
10
Me H
Me
H
10.6
Cycloalkenyl benzenes
11
12
13c
CH2
(CH2)2
—
96
95
10.7
10.6
1S,2R
n.d.
—
2.3
32
—
~100
11.1
a n.d. not determined; b main products 3-phenyl-propen-1-ol and 3-phenyl-propen-1-one; c substrate: 1,4-dihydronaphthalene.
a variety of reactions actually being accredited to P450s.33–36
AaeAPO shares some structural, spectroscopic and catalytic
properties with CfuCPO.32,37 The new outcome here is the
efficient stereoselective benzylic oxygenation of alkylbenzenes
and the epoxidation of the C1–C2 double bond of styrene
derivatives by this enzyme. In order to prove the applicability
of AaeAPO in a process-oriented approach, we optimized the
yield of (R)-1-phenylethanol per enzyme in a fed-batch reactor.
10 was converted with virtually perfect stereoselectivity for
the epoxide formed (and with high TTN up to 110 000).
trans-b-Methylstyrene 9 was only oxidized to some extent
and preferably at the terminal carbon; a minor fraction was
converted into (S,S)-trans-b-methylstyrene oxide (>98% ee).
In contrast, cis-b-methylstyrene was completely converted and
only into one enantiomer, (1R,2S)-cis-b-methylstyrene oxide
(>99% ee). Cycloalkenyl benzenes were epoxidized at the double
bond similar to styrenes. For 1,2-dihydronaphthalene oxide
32% ee and for (1S,2R)-indene oxide, almost no enantiomeric
excess (2.3%) was observed. In a side reaction accounting
for about 5% of conversion, 1,2-dihydronaphthalene 12 was
oxidized to 1- and 2-hydroxy-1,2-dihydronaphthalene, which
spontaneously dehydrated to naphthalene (data not shown).
Naphthalene in turn served as substrate for AaeAPO and the
resulting naphthalene 1,2-oxide rearranged to give 1- and 2-
naphthol as described earlier.35 1,4-Dihydronaphthalene 13 was
subject to epoxidation as well, which yielded the achiral 2,3-
oxide (confirmed by GC-MS; compare also Table S1 in the
supplementary material†).
Results
The stereoselectivity of enzymatic hydroxylation of saturated
and unsaturated alkylbenzenes and cycloalkylbenzenes by
AaeAPO is illustrated in Table 1. The hydrocarbons with
saturated side chains, i.e. ethyl- 1, and propylbenzene 2, indane 5
and tetralin 6, were converted to the corresponding conjugated
secondary benzyl alcohols with high ees up to 99% and traces
of a-ketones. The following tendency was observed for n-
alkylbenzenes 3 and 4: with increasing chain length, turnovers
and ee values decrease and the number of side products formed –
probably other alcohols – increases. The side products detected
are listed in Table S1 (supplementary material†). The bicyclic
hydrocarbons with saturated side chains, tetralin and indane,
were hydroxylated in a similar manner giving the same (R)-
configuration.
The C1–C2 double bond was found to be prone to epoxidation
and in the case of styrene 7, with only poor enantiomeric
excess (7%). The chiral composition of methyl substituted
styrene oxides was unexpected. Thus, a-methylstyrene 8 was
converted mainly to a-methylstyrene oxide that further par-
tially rearranged to a-methyl phenylacetaldehyde. While for a-
methylstyrene oxide, the ee was yet 29%, cis-b-methylstyrene
Indene 11 was epoxidized at the double bond. The observed
1- and 2-indanones could be formed by rearrangement under
the experimental conditions or in the gas chromatograph. In the
HPLC elution profile, only one predominant product, indene
oxide, was found. This peak was collected and could be resolved
into the enantiomers using a chiral reversed phase HPLC
approach.
Incorporation of oxygen
Oxygen originating from 18O-labeled hydrogen peroxide (H218O2)
was completely incorporated into (R)-1-phenylethanol and
This journal is
The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 440–446 | 441
©

View Article Online
remained at the benzylic carbon also during the second oxi-
dation step forming acetophenone. Fig. 1 shows the 70 eV EI
mass spectra of (R)-1-phenylethanol (A) and acetophenone (B)
formed from ethylbenzene by AaeAPO that reacted with either
H218O2 or (overlaid) with natural abundance hydrogen peroxide
(H216O2). In the mass spectra of 18O-(R)-1-phenylethanol and
18O-acetophenone (Fig. 1), characteristic shifts by +2 m/z for
the molecular ion and certain oxygen containing fragment peaks
can be seen, and vice versa, the respective 16O-related peaks are
lacking. Surprisingly, the oxidation of 16O-(R)-1-phenylalcohol
and 16O-(S)-1-phenylalcohol showed no difference in turnover
but only 25% incorporation of 18O into the resulting acetophe-
none when reacted with H218O2. This ratio was affected neither
by the pH in the range from 3 to 9 nor by the presence of radical
scavengers (10 mM 4-oxo-TEMPO or 25 mM ascorbic acid,
data not shown) at pH 7. Oxygenation of cis-b-methylstyrene
proceeded virtually completely. The 70 eV EI mass spectrum of
cis-b-methylstyrene oxide strongly resembles that of the trans-
isomer from the NIST spectra library. It shows a molecular ion
peak at 134 m/z and a fragment ion of 133 m/z caused by H-loss.
When cis-b-methylstyrene oxide was formed in the presence of
H218O2, these peaks were completely replaced by peaks at 136
m/z and 135 m/z.
conversion of ethylbenzene and propylbenzene. The representa-
tion of the so called “integrated Menten differential equation”
was nearly linear vs. time (R2 = 0.9952) proving conformity.
The KM- (and kcat)-values obtained from nonlinear regression
were 694 mM (409 s-1) for ethylbenzene and 480 mM (194 s-1)
for propylbenzene. The values obtained by linear regression
are in good agreement with these data and fit well the Hanes
approximation [706 mM (415 s-1) for ethylbenzene and 497 mM,
(197 s-1) for propylbenzene] (Fig. 2). These kinetic constants
for benzylic hydroxylation are in the same range as those for
the peroxygenation of naphthalene and aryl alcohols or for the
peroxidation of 2,6-dimethoxyphenol (Table 2).
Optimization of (R)-1-phenylethanol production
With regard to a possible implementation of the AaeAPO-
catalyzed hydroxylation of ethylbenzene in an enzyme-based
process, the reaction was optimized by a fed-batch reaction
design towards the following parameters: high product con-
centration in relation to the amount of applied enzyme and
prevention of acetophenone formation by short reaction times.
As the result of this optimization, we achieved a maximum
TTN of 43 000 related to (R)-1-phenylethanol and a space–time
yield (STY) of approximately 60 g L-1 d-1 with an excess of
ethylbenzene (nH2O2/nEB = 0.6) in the feed solution delivered at a
flow rate of 3 mL h-1 (corresponding to 10 min reaction time).
Over-oxidation to acetophenone was minimized as can be seen
in Fig. 3A and 3C. The reduced amount of H2O2 in relation
to ethylbenzene prevented the unwanted catalase-like activity
that otherwise would have led to an irreversible inactivation of
Determination of kinetic parameters for ethylbenzene and
propylbenzene
First the conformity with the Michaelis–Menten model was
proved by following the time course of the AaeAPO-catalyzed
Fig. 2 Initial conversion rates of ethylbenzene oxidation by 8.4 nM
AaeAPO in the presence of 2 mM hydrogen peroxide vs. ethylbenzene
concentration in 10 mM potassium phosphate buffered 20% (v/v)
MeCN. The line was theoretically derived from nonlinear regression.
Margins are given as confidence intervals for P = 0.95. Inset: Hanes plot
with linear regression.
Fig. 1 70 eV EI spectra of 1-phenylethanol (A) and acetophenone
(B) resulting from incorporation of either 16O (blue, back) or 18O
(red, in front) into ethylbenzene confirming a proposed fragmentation
mechanism.39
442 | Green Chem., 2012, 14, 440–446
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
Table 2 Catalytic constants for benzylic hydroxylation in comparison to values of other activities of AaeAPO published earlier
Substrate/product
KM [mM]
kcat [s-1]
k
cat/KM [M-1 s-1]
Reference
Ethylbenzene/(R)-1-phenylethanol
Propylbenzene/(R)-1-phenylpropanol
Benzylalcohol/benzaldehyde
2,6-Dimethoxyphenol/coerulignone
Naphthalene/1-naphthol
Veratryl alcohol/veratraldehyde
Pyridine/pyridine N-oxide
694
480
1001
298
320
2,367
69
410
194
269
108
166
85
5.90 ¥ 105
4.05 ¥ 105
2.69 ¥ 105
3.61 ¥ 105
5.17 ¥ 105
3.58 ¥ 104
3.04 ¥ 103
herein
herein
30
30
38
30
33
0.21
active site in the right manner. Binding of aromatic rings
may be realized by hydrophobic and p–p interactions. Recent
structure determination of AaeAPO revealed a conical funnel
leading from the surface of the protein to the active site,
which is lined with several phenylalanine and other apolar
residues to guide lipophilic substrates to the heme.40,41 The
conformation of the substrate side-chain containing the benzylic
carbon then controls the configuration of the hydroxylated
product. For n-alkylbenzenes, the alkyl group may first rotate
into a favorable position prior to hydroxylation. This implies
that chiral determination must be directed to an approaching
substrate molecule by an asymmetric environment at the distal
vicinity of the heme center that allows for pro-(R) hydroxylation
only. If we consider hydroxylation to proceed via a concerted
mechanism, oxygen will be inserted into the exposed pro-(R) C–
H bond and the alcohol will be released. In the case of hydrogen
abstraction, that is also discussed for P450s,26 the forming sp2-
hybridized radical becomes planar. During this step, orientation
and accessibility of the benzylic carbon to the heme oxygen
does not inevitably get lost. Thus, this works well with ethyl-
and propylbenzene, barely with butylbenzene and hardly with
pentylbenzene. If this decrease of the ee had been caused by
inversion, the rearrangement would have efficiently competed
with the rebound mechanism (which was seemingly not the
case). Also, an effect of the chain length on the rate of inversion
appears rather unlikely. Alternatively, one can imagine that the
bulkiness of the side-chain within the active-site cage may affect
the rebound rate. This may be established by increasing the
distance between the heme oxygen and the benzylradical, which
in turn can reduce the rebound rate and thus allows a more
pronounced inversion.
Fig. 3 Influences of certain parameters on final concentrations of ethyl-
benzene (EB, blue), (R)-phenylethanol (PE, green) and acetophenone
(AC, red). Variation of the feed flow rate A, enzyme concentration B,
ratio of peroxide and EB in feed C (3 mL h-1), D (1 mL h-1). Other
parameters were kept constant: 0.25 U AaeAPO, feed flow rate: 1 mL h-1,
n
H2O2/nEB equal one. Blue, green and red curves refer to the concentration
ordinate, brown graphs to the right ordinate.
AaeAPO. A similar high TTN was reached when the batch was
fed with an equal amount of reactants (nH2O2/nEB = 1) and 0.25–
0.5 U AaeAPO (Fig. 3B) at a flow rate of only 1 mL h-1 but with
one third of the STY.
Discussion
Asymmetric hydroxylation of ethyl- propyl- and butylben-
zene was already reported for CfuCPO42 leading to 1-(R)-
phenylethanol (97% ee), 1-(S)-phenylpropanol (88% ee) and 1-
(S)-phenylbutanol (90% ee). CfuCPO was also shown to possess
prochiral selectivity for the stereoisomers of C1-monodeuterated
benzyl alcohol,43 the hydroxyl group of which binds in the
same orientation as the methyl group of ethylbenzene when
being converted into 1-(R)-phenylethanol. This substrate ori-
entation towards the oxygen to be inserted may also be the
same as for thioanisole that is oxidized to (R)-methylphenyl
sulfoxide by AaeAPO44 and CfuCPO.45 Unlike CfuCPO, the
(R)-configuration was preserved in the course of alkylbenzene
hydroxylation by AaeAPO, i.e. a substantial formation of (S)-
isomers was not observed. Furthermore the ee values for 1-
(R)-phenylethanol and 1-(R)-phenylpropanol were higher in
AaeAPO-catalyzed reactions than in CfuCPO-catalyzed ones,
and the TTNs of AaeAPO were at least two orders of magnitude
The peroxygenase of the fungus Agrocybe aegerita (AaeAPO)
acts as an efficient monooxygenase that can transfer one
peroxide-borne oxygen atom to diverse organic substrates. In
the present study, the AaeAPO-catalyzed oxygenation yielded
secondary Ca-alkanols from alkyl benzenes and different epox-
ides from alkenyl benzenes. Some of these reactions are highly
stereo- and/or regioselective and hold an interesting potential
for organic synthesis. In this context, fed-batch devices using
AaeAPO and varying feeding strategies for substrates and the co-
substrate may be a suitable approach to transfer peroxygenase
catalysis into practice.
Stereoselectivity of hydroxylation
To obtain such capacious stereoselectivity, substrate binding
is of high importance, since it orientates the substrate at the
This journal is
The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 440–446 | 443
©

View Article Online
higher than those reported for CfuCPO,42 even under not
optimized conditions without continuously delivered peroxide.
Intracellular CYP P450BSb (syn. fatty acid peroxygenase, EC
1.11.2.4) from Bacillus subtilis has been reported to efficiently use
H2O2 for the generation of reactive heme-species via a “peroxide
shunt”-like pathway.46 It catalyzes the Ca/Cb-hydroxylation of
long-chain fatty acids such as myristic acid. In the presence of
decoy molecules (short-chain carboxylic acids from C4 to C8),
also other substrates including ethylbenzene were found to be
hydroxylated; ee values for 1-(R)-phenylethanol ranged between
35% and 68% in the presence of butanoic and heptanoic acid,
respectively.47 A comparable ee of 73% was reported for the
formation of 1-(R)-phenylethanol from ethylbenzene by P450cam
using NADH and O2 as co-substrates.48
are fixed. For CfuCPO, different simulations were performed to
explain the stereoselectivity of cis-b-methylstyrene epoxidation,
but they revealed contradictory results.53 There are also no indi-
cations for cis–trans-epimerization during the epoxidation of cis-
b-methylstyrene by AaeAPO or CfuCPO, which was described
for dissolved or electrode-bound hemoproteins like P450cam and
myoglobin54,55 or Mn(salen)-supported catalysts.11 This either
implies a concerted mechanism or radical intermediates that are
sterically hindered from inversion into the energetically favored
trans-configuration.
The retention of stereochemistry is important for the imple-
mentation of new substrates for the stereoselective production of
drugs, in particular when the conversion of the trans-isomers is
discriminated. A laborious preparation of optically active cis-b-
methylstyrene oxides from L-(-)- or D-(+)-pseudoephedrine was
already described by Witkop and Foltz in 1957.56 Stereospecific
epoxidation of cis-b-methylstyrene via the Jacobsen-Katsuki
route required chiral Mn-salen catalysts and yielded up to
85% epoxide with 92% ee but performed only 25 turnovers in
total.13 Other authors demonstrated the epoxidation of cis-b-
methylstyrene to the corresponding (1S,2R)-isomer with 80%
ee by Coprinus cinereus peroxidase (CiP, EC 1.11.1.7) but with
about ten cycles only; in addition to the epoxide, similar amounts
of benzaldehyde from C1–C2 bond cleavage were formed in the
course of this reaction.57 CfuCPO was reported to catalyze the
epoxidation of cis-b-methylstyrene into the same enantiomer
in good yields with 96% ee and a TTN of approximately 1600
when H2O2 was added slowly.25 AaeAPO catalyzed the efficient
epoxidation of cis-b-methylstyrene with TTNs up to 110 000 and
>99% ee for the (1S,2R)-isomer.
In a similar way as ethylbenzene, tetralin and indane were
hydroxylated by AaeAPO resulting in high ee values for the
(R)-isomers (99% and 87%, respectively). Comparable ee values
(94% and 88%) in the same configuration were obtained with
49
P450cam
.
To our best knowledge, there is no report on the
conversion of tetralin or indane by CfuCPO. Besides C1-C2
epoxidation 1,2-dihydronaphthalene was proven to be hydrox-
ylated at the C3 and C4 to form 1- and 2-hydroxy dihydron-
aphthalene that in turn dehydrated into naphthalene, neither of
which were the case for 1,4-dihydronaphthalene. Formation of
naphthalene from 1,2-dihydronaphthalene was also described
for the metabolism of the Pseudomonas putida mutant UV4.50–52
It was shown to proceed via the dioxygenase-catalyzed formation
of the chiral arene hydrates of naphthalene51 and subsequent
release of water. The high asymmetry of this catalysis suggests
future attempts to determine absolute configuration of these
labile compounds formed by AaeAPO.
The products of the conversion of unsubstituted styrene
showed almost no enantiomeric excess implying the enzyme’s
inability to bind the molecule in a similar way as the structurally
related ethylbenzene, maybe due to the side-chain’s inability of
free rotation.
Stereoselectivity of epoxidation
Epoxidation of styrene and its derivatives by AaeAPO proceeded
in a different manner from the hydroxylation of alkylbenzenes.
Although conversions varied between 19% and almost 100%,
the ee values for the resulting epoxides were poor with the
exception of cis-b-methylstyrene that turned out to be a good
substrate both in terms of conversion (95%) and ee (>99%). The
configuration of the (1R,2S)-epoxide obtained was unexpected,
since the orientation of the oxygen relative to the carbon
network is vice versa to that of phenylalkanols formed during
alkylbenzene hydroxylation by AaeAPO. trans-b-Methylstyrene
bearing a long and torsionally rigid side-chain was a bad
substrate in terms of epoxidation and only traces (<1%) of
the S,S-epoxide were found (even though with high ee). It is
conceivable that this molecule was restrained from entering the
active site in a way that would facilitate the contact of the double
bond to the heme oxygen; as the result, trans-b-methylstyrene
was preferably oxidized at the terminal methyl group (C3-
position) and only to a moderate extent (19% conversion).
Comparison of the results with those of the bicyclic derivatives
of cis-b-methylstyrene reveals no general principle of substrate
binding, i.e. stereoselective peroxygenation may not simply be
achieved by structural homology. Possibly, the side chain of
cis-b-methylstyrene is more flexible and able to bend- to some
extent- outside the plane of the aryl ring, while the respective
carbons in 1,2-dihydronaphthalene and in particular in indene
Mechanism of peroxygenation
The results of the 18O-incorporation studies display a true
peroxygenation mechanism for alkyl hydroxylation and alkenyl
epoxidation, i.e. the transferred oxygen exclusively originated
from the peroxide (H218O2). The same was shown before for
the AaeAPO-catalyzed aromatic oxygenation,35 sulfoxidation,58
N-oxidation33 and O-dealkylation.34 All these reactions could be
useful to prepare specifically labeled molecules (e.g. radioactively
or isotopically labeled drugs) for diagnostic purposes.
On the other hand, the results of 18O-incorporation into 16O-
1-phenylethanol to give the corresponding ketone remain incon-
sistent with one single reaction mechanism. If the oxidation of 1-
phenylethanol solely proceeded via a geminal diol intermediate
(gem-diol) that subsequently releases water, one would expect
that 18O and 16O are evenly present in the resulting acetophenone.
However, we observed a ratio of 1 : 3 for 18O/16O, which – only
to a small extent – may be attributed to a kinetic isotope effect
(not more than 2% isotopic excess of 16O). To overcome this
discrepancy, we suggest an alternative mechanism that explains
the observed excess of 16O-acetophenone (Fig. 4). It may involve
caged ketyl radicals as recently described for P450s formed
through hydrogen abstraction and one-electron oxidation
444 | Green Chem., 2012, 14, 440–446
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
due to the very low apparent KM of EBDH (0.45 mM), the
apparent kcat was only 0.26 s-1.62
Potential for application
In the case of ethylbenzene hydroxylation, a fed-batch design
that varied three process parameters revealed a considerable
increase of TTN and STY, which could be even further improved
(up to a TTN of 43 000 and an STY of approximately 60 g L-1
d-1). In order to make this process more eco-friendly and to
ease downstream processing, the amount of toxic MeCN can
be reduced to 5% (vol/vol) in the final reaction mixture using a
higher concentrated fed solution.
In a special experiment following the epoxidation of cis-b-
methylstyrene (2 mM substrate/peroxide, started by addition
of 18 nM AaeAPO, 3 min reaction time at RT), an STY of
approximately 130 g L-1 d-1 and a TTN of 110 000 were achieved.
cis-b-Methylstyrene oxide is known to be unstable in aqueous
media.56 Therefore, we suggest a suitable ring opening reaction,
in the course of which the two stereocenters will be retained
during one-pot synthesis. A variety of bonds including C–C can
be formed using certain reagents (e.g. nucleophiles).5,63 Finally,
an extraction step may be necessary prior to further processing
the enantio- and diastereomerically pure products.
Fig. 4 Proposed reaction scheme for the peroxygenation of ethylben-
zene and 1-phenylethanol by AaeAPO in the presence of either H216O2 or
H218O2, proposed formation of compound I and dual mode mechanism
for the oxidation of 1-phenylethanol. Oxidation may proceed either
without oxygenation via a caged ketyl radical and a second oxidation
step or via oxygenation and a short-lived gem-diol intermediate formed
by rebound and subsequent decay.
without oxygen transfer.59 In the next step, either a one-electron
oxidation could directly lead to 16O-acetophenone or a rebound
mechanism may result in the formation of a gem-diol. However,
the isotopic ratio was not affected by pH or the presence of
radical scavengers, which admittedly would have strengthened
a dual mechanism hypothesis. Under the conditions used, 1-
phenylethanol and acetophenone did seemingly not exchange
oxygen either with water or dissolved oxygen. This was further
proven by the incubation of acetophenone with an equimolar
amount of H218O, which did not result in any oxygen exchange
via hydration within three days. Only when adding concentrated
hydrochloric acid, was an exchange of about 10% observed.
Conclusions
Asymmetric benzylic hydroxylations and C1–C2 epoxidations
are challenging oxy-functionalizations in organic chemistry.
AaeAPO has been shown to effectively catalyze these reac-
tions and some thereof are commended for application in
(chemo)enzymatic synthesis for the production of e.g. O-labeled
fine chemicals. The kinetic data and the tolerance against high
initial H2O2 concentrations allows for high TTN, STY and ee
values in a simple reaction design under mild conditions. Thus
AaeAPO is considered superior to any other biocatalyst for these
purposes known so far.
Kinetics of oxygen transfer
Acknowledgements
The maximum turnover rates (apparent kcat) of AaeAPO for
ethylbenzene and propylbenzene were calculated to be 409 s-1
and 194 s-1, respectively, and a value in the same range is
assumed for styrene conversion. Reported rates for P450BSb from
Bacillus subtilis are by three to two orders of magnitude lower
for ethylbenzene (max. 0.5 s-1) and styrene (max. 6.6 s-1).47 For
CfuCPO compound I kinetic data are available as second-order
rate constants.60 Based on them, kcat-values can be calculated
with respect to substrate concentration. Thus, given a substrate
concentration of 5 mM, kcat comes out to approximately 5 s-1 for
ethylbenzene and yet approximately 300 s-1 for styrene oxidation.
In applications, the formation of compound I may be the rate
limiting step. For a second-order rate constant of 2.4 ¥ 106 M-1
s-1,61 this means approximately 50 s-1 for the rate of compound I
formation with 2.5 mM CfuCPO and 10 mM hydrogen peroxide
(when delivered by a syringe pump). The periplasmic EBDH (EC
1.17.99.2) from Aromatoleum aromaticum (strain EbN1) was
reported to catalyze the anaerobic water-assisted hydroxylation
of ethylbenzene to (S)-1-phenylethanol. The enzyme displayed
nearly the same catalytic efficiency (kcat/KM) as AaeAPO but,
The authors would like to thank the European Social Fund (ESF
Appl. # 080935557) and the EU project Peroxicats (KBBE-2010-
4-265397) for financial support. We also thank U. Schneider and
M. Brand for technical assistance and S. Fra¨nzle for useful and
inspirational discussions.
Notes and references
1 D. Sinou, Adv. Synth. Catal., 2002, 344, 221–237.
2 M. Breuer, K. Ditrich, T. Habicher, B. Hauer, M. Keßeler, R. Stu¨rmer
and T. Zelinski, Angew. Chem., Int. Ed., 2004, 43, 788–824.
3 J. Gorzynski Smith, Synthesis, 1984, 8, 629–656.
4 H. Shibuya, K. Kawashima, M. Ikeda and I. Kitagawa, Tetrahedron
Lett., 1989, 30, 7205–7208.
5 G. Guanti, V. Merlo and E. Narisano, Tetrahedron, 1994, 50, 12245–
12258.
6 S. Denmark, P. Barsanti, G. Beutner and T. Wilson, Adv. Synth.
Catal., 2007, 349, 567–582.
7 C. Saluzzo and M. Lemaire, Adv. Synth. Catal., 2002, 344, 915–928.
8 J. Blacker and J. Martin, in Asymmetric catalysis on industrial scale:
challenges, approaches and solutions, ed. H. U. Blaser and E. Schmidt,
Wiley-VCH, Weinheim, Germany, 2004, pp. 201–202.
This journal is
The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 440–446 | 445
©

View Article Online
9 J. Mao, B. Wan, F. Wu and S. Lu, Tetrahedron Lett., 2005, 46, 7341–
7344.
38 M. G. Kluge, R. Ullrich, K. Scheibner and M. Hofrichter, Appl.
Microbiol. Biotechnol., 2007, 75, 1473–1478.
39 A. Uzura, T. Katsuragi and Y. Tani, J. Biosci. Bioeng., 2001, 91,
217–221.
10 T. Katsuki and K. B. Sharpless, J. Am. Chem. Soc., 1980, 102, 5974–
5976.
11 H. Zhang, Y. M. Wang, L. Zhang, G. Gerritsen, H. C. L. Abbenhuis,
R. A. van Santen and C. Li, J. Catal., 2008, 256, 226–236.
12 W. Zhang, J. L. Loebach, S. R. Wilson and E. N. Jacobsen, J. Am.
Chem. Soc., 1990, 112, 2801–2803.
13 E. N. Jacobsen, W. Zhang, A. R. Muci, J. R. Ecker and L. Deng, J.
Am. Chem. Soc., 1991, 113, 7063–7064.
40 K. Piontek, R. Ullrich, C. Liers, K. Diederichs, D. A. Plattner and
M. Hofrichter, Acta Crystallographica Section F, 2010, 66, 693–
698.
41 K. Piontek, personal communication.
42 A. Zaks and D. R. Dodds, J. Am. Chem. Soc., 1995, 117, 10419–
10424.
14 Z.-X. Wang, Y. Tu, M. Frohn, J.-R. Zhang and Y. Shi, J. Am. Chem.
Soc., 1997, 119, 11224–11235.
15 H. Tian, X. She, J. Xu and Y. Shi, Org. Lett., 2001, 3, 1929–1931.
16 J. P. Collman, Z. Wang, A. Straumanis, M. l. Quelquejeu and E.
Rose, J. Am. Chem. Soc., 1998, 121, 460–461.
17 D. Vasic-Racki, in Industrial biotransformations, ed. A. Liese, K.
Seelbach and C. Wandrey, Wiley-VCH Verlag, Weinheim, 2nd edn,
2006, pp. 1–36.
43 E. Baciocchi, L. Manduchi and O. Lanzalunga, Chem. Commun.,
1999, 1715–1716.
44 A. Horn, PhD, University of Rostock, 2009, http://rosdok.uni-
rostock.de/resolve?urn=urn:nbn:de:gbv:28-diss2009-0181-1.
45 A. Conesa, F. van de Velde, F. van Rantwijk, R. A. Sheldon, C. A.
M. J. J. van den Hondel and P. J. Punt, J. Biol. Chem., 2001, 276,
17635–17640.
46 I. Matsunaga, M. Yamada, E. Kusunose, Y. Nishiuchi, I. Yano and
K. Ichihara, FEBS Lett., 1996, 386, 252–254.
18 S. Yadav, R. S. S. Yadav, S. Yadava and K. D. S. Yadav, Catal.
Commun., 2011, 12, 781–784.
47 O. Shoji, T. Fujishiro, H. Nakajima, M. Kim, S. Nagano, Y. Shiro
and Y. Watanabe, Angew. Chem., Int. Ed., 2007, 46, 3656–3659.
48 D. Filipovic, M. D. Paulsen, P. J. Loida, S. G. Sligar and R. L.
Ornstein, Biochem. Biophys. Res. Commun., 1992, 189, 488–495.
49 M. P. Mayhew, A. E. Roitberg, Y. Tewari, M. J. Holden, D. J.
Vanderah and V. L. Vilker, New J. Chem., 2002, 26, 35–42.
50 D. R. Boyd, R. A. S. McMordie, N. D. Sharma, H. Dalton, P.
Williams and R. O. Jenkins, Journal of the Chemical Society, Chemical
Communications, 1989.
19 Y. Chen, F. Lie and Z. Li, Adv. Synth. Catal., 2009, 351, 2107–2112.
20 J. B. Arterburn, Tetrahedron, 2001, 57, 9765–9788.
21 K. Nakamura, R. Yamanaka, T. Matsuda and T. Harada, Tetrahe-
dron: Asymmetry, 2003, 14, 2659–2681.
22 N. Krebsfa¨nger, K. Schierholz and U. T. Bornscheuer, J. Biotechnol.,
1998, 60, 105–111.
23 R. Patel, A. Banerjee, V. Nanduri, A. Goswami and F. Comezoglu,
J. Am. Oil Chem. Soc., 2000, 77, 1015–1019.
24 J. H. Dawson and M. Sono, Chem. Rev., 1987, 87, 1255–1276.
25 E. J. Allain, L. P. Hager, L. Deng and E. N. Jacobsen, J. Am. Chem.
Soc., 1993, 115, 4415–4416.
51 R. Agarwal, D. R. Boyd, R. A. S. McMordie, G. A. O’Kane, P.
Porter, N. D. Sharma, H. Dalton and D. J. Gray, J. Chem. Soc.,
Chem. Commun., 1990, 1711–1713.
52 D. R. Boyd, N. D. Sharma, N. A. Kerley, R. A. S. McMordie, G. N.
Sheldrake, P. Williams and H. Dalton, J. Chem. Soc., Perkin Trans.
1, 1996, 67–74.
26 B. Meunier, S. P. de Visser and S. Shaik, Chem. Rev., 2004, 104,
3947–3980.
´
27 S. Aguila, R. Vazquez-Duhalt, R. Tinoco, M. Rivera, G. Pecchi and
J. B. Alderete, Green Chem., 2008, 10, 647–653.
28 H. A. Johnson, D. A. Pelletier and A. M. Spormann, J. Bacteriol.,
2001, 183, 4536–4542.
29 O. Kniemeyer and J. Heider, J. Biol. Chem., 2001, 276, 21381–21386.
30 R. Ullrich, J. Nu¨ske, K. Scheibner, J. Spantzel and M. Hofrichter,
Appl. Environ. Microbiol., 2004, 70, 4575–4581.
31 M. J. Pecyna, R. Ullrich, B. Bittner, A. Clemens, K. Scheibner, R.
Schubert and M. Hofrichter, Appl. Microbiol. Biotechnol., 2009, 84,
885–897.
53 M. Sundaramoorthy, J. Terner and T. L. Poulos, Chem. Biol., 1998,
5, 461–473.
54 P. R. Ortiz de Montellano and L. A. Grab, Biochemistry, 1987, 26,
5310–5314.
55 X. Zu, Z. Lu, Z. Zhang, J. B. Schenkman and J. F. Rusling, Langmuir,
1999, 15, 7372–7377.
56 B. Witkop and C. M. Foltz, J. Am. Chem. Soc., 1957, 79, 197–201.
57 A. Tuynman, J. L. Spelberg, I. M. Kooter, H. E. Schoemaker and R.
Wever, J. Biol. Chem., 2000, 275, 3025–3030.
32 M. Hofrichter, R. Ullrich, M. J. Pecyna, C. Liers and T. Lundell,
Appl. Microbiol. Biotechnol., 2010, 87, 871–897.
33 R. Ullrich, C. Dolge, M. Kluge and M. Hofrichter, FEBS Lett., 2008,
582, 4100–4106.
34 M. Kinne, M. Poraj-Kobielska, S. A. Ralph, R. Ullrich, M.
Hofrichter and K. E. Hammel, J. Biol. Chem., 2009, 284, 29343–
29349.
35 M. Kluge, R. Ullrich, C. Dolge, K. Scheibner and M. Hofrichter,
Appl. Microbiol. Biotechnol., 2009, 81, 1071–1076.
36 E. Aranda, R. Ullrich and M. Hofrichter, Biodegradation, 2010, 21,
267–281.
58 E. Aranda, M. Kinne, M. Kluge, R. Ullrich and M. Hofrichter, Appl.
Microbiol. Biotechnol., 2009, 82, 1057–1066.
59 H. L. R. Cooper and J. T. Groves, Arch. Biochem. Biophys., 2011,
507, 111–118.
60 R. Zhang, N. Nagraj, D. S. P. Lansakara-P, L. P. Hager and M.
Newcomb, Org. Lett., 2006, 8, 2731–2734.
61 T. Araiso, R. Rutter, M. M. Palcic, L. P. Hager and H. B. Dunford,
Biochem. Cell Biol., 1981, 59, 233–236.
62 M. Szaleniec, C. Hagel, M. Menke, P. Nowak, M. Witko and J.
Heider, Biochemistry, 2007, 46, 7637–7646.
63 A. Degl’Innocenti, A. Capperucci, A. Cerreti, S. Pollicino, S.
Scapecchi, I. Malesci and G. Castagnoli, Synlett, 2005, 20, 3063–
3066.
37 M. Hofrichter and R. Ullrich, Appl. Microbiol. Biotechnol., 2006, 71,
276–288.
446 | Green Chem., 2012, 14, 440–446
This journal is
The Royal Society of Chemistry 2012
©