The reaction mechanism of chiral hydroxylation of p -OH and p -NH 2 substituted compounds by ethylbenzene dehydrogenase

Article abstract of DOI:10.1139/cjc-2012-0504

Ethylbenzene dehydrogenase (EbDH; enzyme commission (EC) number: 1.17.99.2) is a unique biocatalyst that hydroxylates alkylaromatic and alkylheterocyclic compounds to (S)-secondary alcohols under anaerobic conditions. The enzyme exhibits a high promiscuity catalyzing oxidation of over 30 substrates, inter alia, para-substituted alkylphenols and alkylanilines. Secondary alcohols with OH and NH₂ substituents in the aromatic ring are highly valuable synthons for many biologically active compounds in the fine chemical industry. EbDH hydroxylates most of the studied compounds highly enantioselectively, except for five substrates that harbour OH and NH₂ groups in the para position, which exhibit a significant decrease in the percent enantiomeric excess (% ee). This phenomenon is inconsistent with the previously suggested enzyme mechanism, but it may be linked to a stabilization of the carbocation intermediate by deprotonation of the OH or NH₂ substituent in the active site that yields a transient quinone (imine) ethide species. This would initiate an alternative reaction pathway involving the addition of a water molecule to a C=C double bond. This hypothesis was cross-validated by density functional theory (DFT) cluster modelling of the alternative reaction pathway with 4-ethylphenol, as well as by experimental assessment of the pH dependency of enantiomeric excesses. The results reported herein suggest that the alternative reaction pathway may significantly contribute to the overall reaction if the carbocation intermediates are stabilized by deprotonation.

Full text of DOI:10.1139/cjc-2012-0504

775
ARTICLE
The reaction mechanism of chiral hydroxylation of p-OH and p-NH2
substituted compounds by ethylbenzene dehydrogenase
Agnieszka Dudzik, Bartłomiej Kozik, Mateusz Tataruch, Anna Wójcik, Daniel Knack, Tomasz Borowski, Johann Heider, Małgorzata Witko,
and Maciej Szaleniec
Abstract: Ethylbenzene dehydrogenase (EbDH; enzyme commission (EC) number: 1.17.99.2) is a unique biocatalyst that hy-
droxylates alkylaromatic and alkylheterocyclic compounds to (S)-secondary alcohols under anaerobic conditions. The enzyme
exhibits a high promiscuity catalyzing oxidation of over 30 substrates, inter alia, para-substituted alkylphenols and al-
kylanilines. Secondary alcohols with OH and NH2 substituents in the aromatic ring are highly valuable synthons for many
biologically active compounds in the fine chemical industry. EbDH hydroxylates most of the studied compounds highly enan-
tioselectively, except for five substrates that harbour OH and NH2 groups in the para position, which exhibit a significant
decrease in the percent enantiomeric excess (% ee). This phenomenon is inconsistent with the previously suggested enzyme
mechanism, but it may be linked to a stabilization of the carbocation intermediate by deprotonation of the OH or NH2
substituent in the active site that yields a transient quinone (imine) ethide species. This would initiate an alternative reaction
pathway involving the addition of a water molecule to a C=C double bond. This hypothesis was cross-validated by density
functional theory (DFT) cluster modelling of the alternative reaction pathway with 4-ethylphenol, as well as by experimental
assessment of the pH dependency of enantiomeric excesses. The results reported herein suggest that the alternative reaction
pathway may significantly contribute to the overall reaction if the carbocation intermediates are stabilized by deprotonation.
Key words: ethylbenzene dehydrogenase, molybdenum enzyme, chiral alcohols, DFT.
Résumé : L’éthylbenzène dehydrogenase (EC 1.17.99.2) est un biocatalyseur unique qui hydroxyle les composés alkylaromatiques
et alkyl hétérocycliques pour donner des énantiomères S d’alcools secondaires dans des conditions anaérobies. L’enzyme manifeste
une importante promiscuité, catalysant l’oxydation de plus de 30 substrats, entre autres des alkylphénols et des alkylanilines
substitués en para. Les alcools secondaires possédant des substituants OH et NH2 sur le noyau aromatique sont des synthons très utiles
pour de nombreux composés biologiquement actifs dans l’industrie des produits de chimie fine. L’éthylbenzène hydroxylase hy-
droxyle de manière très énantiosélective la plupart des composés étudiés, excepté cinq substrats porteurs de groupements OH et NH2
en position para, pour lesquels on observe une diminution significative de l’excès énantiomérique (ee %). Ce phénomène ne
concorde pas avec le mécanisme enzymatique proposé antérieurement, mais il pourrait être lié a` une stabilisation des
carbocations intermédiaires par déprotonation du substituant OH ou NH2 au site actif donnant une espèce éthide de
quinone (imine) transitoire. Cela ouvrirait un autre chemin de réaction comprenant l’ajout d’une molécule d’eau sur une
double liaison C=C. On a procédé a` une validation croisée de cette hypothèse par modélisation, fondée sur la théorie de la
fonctionnelle de la densité (DFT) et un modèle d’agrégat, du chemin de réaction de rechange avec le 4-éthylphénol, ainsi que
par évaluation expérimentale de la dépendance des excès énantiométriques au pH. Les résultats présentés donnent a` penser
que le chemin de réaction de rechange pourrait contribuer considérablement a` la réaction globale si les carbocations
intermédiaires sont stabilisés par déprotonation. [Traduit par la Rédaction]
Mots-clés : éthylbenzène hydrolase, enzyme a` molybdène, alcools chiraux, DFT.
6
Introduction
tor linked by a linear row of five iron–sulfur clusters. The enzyme
active site is deeply buried in the ␣ subunit and is comprised of the
molybdenum cofactor (MGD) and the side chains of aspartic acid
Ethylbenzene dehydrogenase (EbDH) is a molybdoenzyme be-
1
,2
longing to the dimethyl sulfoxide (DMSO) reductase family, and
catalyzes the oxygen-independent, stereoselective hydroxylation
of ethylbenzene to (S)-1-phenylethanol. It is the first known en-
zyme capable of direct anaerobic hydroxylation of a nonactivated
2
23
192
2
23 (Asp ) and histidine 192 (His ), which were proposed to be
6
,7
involved in the catalytic process.
Our previous investigations have shown that EbDH hy-
droxylates aliphatic methylene groups adjacent to aromatic or
heterocyclic rings in many different substrates, forming the re-
1
,3,4
hydrocarbon,
andisinvolvedinthemineralizationofethylben-
zene by Aromatoleum aromaticum, a denitrifying betaproteobacte-
5
8–12
rium. EbDH is a ␣␤␥ heterotrimer with subunits of 96, 43, and 23
spective secondary alcohols.
The implementation of chiral
kDa and contains a molybdenum cofactor and a heme b559 cofac-
chromatography revealed that EbDH exhibits 100% reaction enan-
Received 13 December 2012. Accepted 18 March 2013.
A. Dudzik, M. Tataruch, A. Wójcik, T. Borowski, M. Witko, and M. Szaleniec. Jerzy Haber Institute of Catalysis and Surface Chemistry Polish Academy of Sciences Niezapominajek 8,
0-239 Kraków, Poland.
3
B. Kozik. Department of Organic Chemistry, Jagiellonian University, Ingardena 3, 30-060 Kraków, Poland.
D. Knack. Jerzy Haber Institute of Catalysis and Surface Chemistry Polish Academy of Sciences Niezapominajek 8, 30-239 Kraków, Poland; Laboratory for Microbial Biochemistry, Philipps
University of Marburg, Karl-von-Frisch Strasse 8, D-35043 Marburg, Germany.
J. Heider. Laboratory for Microbial Biochemistry, Philipps University of Marburg, Karl-von-Frisch Strasse 8, D-35043 Marburg, Germany.
Corresponding author: Maciej Szaleniec (e-mail: ncszalen@cyfronet.pl).
This article is part of a Special Issue dedicated to Professor Dennis Salahub in recognition of his contributions to theoretical and computational chemistry.
“
Friendship is the golden ribbon that ties the world together.” Dennis, you are that ribbon.
Can. J. Chem. 91: 775–786 (2013) dx.doi.org/10.1139/cjc-2012-0504
Published at www.nrcresearchpress.com/cjc on 21 March 2013.

7
76
Can. J. Chem. Vol. 91, 2013
Fig. 1. Proposed catalytic mechanism of ethylbenzene hydroxylation by EbDH. The molybdenum cofactor of EbDH is represented only by the
Mo atom with its oxo or hydroxyl ligands. The respective molybdenum cofactor (MoCo) oxidation states are indicated. S, substrate state; TS1,
•
transition state 1; I, intermediate state with bound radical; I , alternative intermediate state with bound carbocation; TS2, transition state 2;
7
P, product state.
tioselectivity (within detection limits) for most compounds,
yielding exclusively (S)-alcoholic products. For some substrates,
however, substantial amounts of (R)-alcohols were also detected,
although always as a minor fraction.10 The theoretical studies of
EbDH led to the formulation of a reaction mechanism, which
appears to be valid for a majority of substrates. In short, the two-
electron oxidation proceeds in two steps: (i) a radical type activa-
tion of the pro(S) C–H bond at C1 (TS1) leading to the formation of
a radical intermediate (I) and reduction of Mo(VI) to Mo(V), and
ents of the substrate analogues. Rotation of the side chain of this
Asp would allow the carboxylic group to form hydrogen bonds
with the substrate and the bound intermediates (Fig. 2).
The plausibility of the standard and the alternative reaction
pathways leading to both (R)- and (S)-alcohols from 4-ethylphenol
was assessed on the basis of density functional theory (DFT) mod-
elling using small active site cluster models.
Experimental section
(ii) a second electron transfer coupled with subsequent OH rebound
Sample preparation
(TS2) leading to the formation of the alcohol product and reduc-
Ethylbenzene dehydrogenase was purified from ethylbenzene-
grown cells of A. aromaticum as described previously.9 The en-
zyme assays were routinely conducted at an optimum pH of 7.5 at
tion of Mo(V) to Mo(IV) (see Fig. 1). The extent of the radical or
carbonation nature of the intermediate seems to depend on the sub-
strate electronic characteristic,13 as recent investigation of the reac-
tion mechanism with the quantum mechanics/molecular mechanics
,24
3
0 °C in 20 mL of 100 mmol/L Tris/HCl containing 200 ␮mol/L
ferricenium(III) tetrafluoroborate, and ethylbenzene dehydrogenase
100–300 ␮L of ϳ1 mg/mL protein solution). The reactions were
(
QM/MM) technique suggests a strong stabilization of the carboca-
(
14
tion form of I introduced by the protein part of the enzyme.
initiated by adding 100 ␮L of a stock solution of the respective
substrate in tert-butanol (a list of substrates, their purities and
producers is available in the Supplementary data). After overnight
incubation, the reaction mixtures where extracted from the water
phase by solid phase extraction using either C18 Polar Plus (Baker)
or polystyrene/divinylbenzene (PS/DVB) copolymer solid-phase ex-
traction (SPE) columns (Strata-X from Phenomenex or the equiva-
lent Chromabond HR-X from Macherey-Nagel), which were eluted
with 0.5 mL of isopropanol.
Pure chiral alcohols are universal synthons for the industrial
production of fine chemicals and can be used as starting material
for the development of new drugs, odorants, food additives, or
even chiral nematic liquid crystals.1
5–19
Therefore, EbDH has great
potential for biotechnological applications and elucidation of its
enantioselectivity with different substrates is of interest. The
production of chiral alcohols by irreversible hydroxylation of hy-
drocarbons by EbDH is novel and complementary to the estab-
lished process using alcohol dehydrogenases, which catalyze the
reduction of ketones in a reversible manner. As a result, com-
pounds with substituents promoting oxidation (i.e., electron do-
nating groups) are enhancing the reaction rate of EbDH, precisely
Organic synthesis of standards
Syntheses of chiral standards and their racemic mixtures of
2-(1-hydroxyethyl)phenol (II), 3-(1-hydroxyethyl)phenol (III), 4-
2
0,21
opposite to their influence on ketone reduction.
The p-OH and
(
(
1-hydroxyethyl)phenol (IV), and 4-(1-hydroxyprop-1-yl)phenol
VI) were previously described. The (S)-2,3-dihydro-1H-indene-
p-NH electron donating effects enhancing reaction rates of EbDH
10
2
are especially noteworthy for the synthesis of potential adrena-
line analogues. However, the enhanced hydroxylation rate by
EbDH observed for 4-ethylphenol (4) and 4-ethylaniline (9) was
associated with a marked decrease of the reaction enantioselec-
tivity.
1,6-diol (XIb) was obtained by enzymatic reduction of 6-hydroxy-2,3-
dihydro-1H-inden-1-one using recombinant (S)-specific phenylethanol
25,26
dehydrogenase from A. aromaticum.
Tests for pH dependency
Therefore, we have investigated, with experimental and theoreti-
The influence of the pH on the enantioselectivity of the reaction
was investigated for 4-ethylphenol (4), 4-ethylaniline (9), and 2,3-
dihydro-1H-inden-5-amine (12) to validate the proposed hypothe-
sis involving a modified mechanism. Reactions were performed in
cal techniques, the influence of the position of OH and NH in the
2
aromatic system on reaction enantioselectivity. The stereochemistry
of the products was determined with chiral liquid chromatography
(LC–mass spectrometry (MS)), while quantum chemical modelling
100 mmol/L Tris/HCl at pH values of 7.0, 7.5, or 8.0. The reaction
was used to understand the experimentally observed phenomena.
Alternative pathways were considered that are analogous to the
one proposed for 4-ethylphenol methylenehydroxylase (4EPMH)
products were separated and quantified by chiral chromatography
using a normal-phase (NP) column with an atmosphere-pressure
chemical ionization (APCI)-LC–MS instrument using both a diode-
array detector (DAD) and selected-ion monitoring (SIM).
2
2
from Pseudomonas putida. 4EPMH hydroxylates 4-ethylphenol in
two steps: (i) two electron oxidation with proton subtraction
yields quinone ethide and (ii) water addition to the double bond
takes place with concomitant protonation of the phenolic group.
Re-evaluation of the structure of the active site of EbDH revealed
a conveniently positioned residue, i.e., Asp4 , which would be
LC/LC–MS
The chiral high-performance liquid chromatography (HPLC)
separations were performed on an Agilent 1100 system in a
normal-phase system on a cellulose tribenzoate polysaccharide
chiral stationary phase (Daicel CHIRALCEL OB-H column, 250 mm ×
85
available to form a hydrogen bond with the p-OH or -NH substitu-
2
Published by NRC Research Press

Dudzik et al.
777
Fig. 2. Model of the structure of the active site of EbDH during C–H activation of 4-ethylphenol (4). The position of the Asp485 side chain was
manually modified by rotation of the C␣ and C␤ dihedral angle by +190°, and the position of the substrate was derived from that of ethylben-
2
3
zene in TS1 from the QM/MM calculations.
4
.6 mm, 5 ␮m). To avoid changes in sample concentrations during
Flexible docking
the analysis, the autosampler was thermostated to 8 °C. Depend-
ing on the polarity of analytes, different types of uniform isocratic
programs were used with mobile phases comprised of mixtures of
n-hexane/isopropanol in ratios ranging from 90:10 up to 65:35 (v/v)
The enzyme–substrate complex with 4-ethylphenol (4) was ob-
tained in a flexible docking performed in Discovery Studio 3.5
(Accelrys) according to the procedure previously described.2
3
A
structure of EbDH ␣ subunit with cofactor geometry obtained in
QM/MM optimization was used as the receptor. The whole model
(see Table 1) at 25 °C and at a flow rate of 0.5 mL/min. Products
2
9
was typed with a CHARMm force field (c36b2) and MM point
were detected by a DAD (Agilent 1100 DAD) with or without cou-
pling to an MS detector (VL MS spectrometer) using an APCI ion
source in positive or negative ion mode (Table S1 in the Supple-
mentary data). The mobile phase used in the normal-phase chro-
matography (i.e., with a high n-hexane concentration) is widely
regarded as incompatible with the APCI due to the high hazard of
n-hexane explosion upon contact with heated nebulizer and high
30
charges were calculated according to the Momany–Rone scheme.
The MM point charges for the substrate, the Mo cofactor, and the
iron–sulfur cluster present in the ␣ subunit of EbDH were fitted to
the electrostatic potential according to the Merz–Singh–Kollman
3
1,32
scheme.
The initial position of the ligand was taken from an overlay with
the position of ethylbenzene in the transition state associated with
2
7
voltage corona discharge. Moreover, a high content of n-hexane
and lack of ions in the mobile phase decreases the chance of
sample ionization with APCI and destabilizes the mass spectrom-
eter (system instability and over polarization due to the uncon-
trolled electric discharges were observed). Therefore, we used a
23
C–H activation (TS1). Then, the geometry of 4-ethylphenol and all
amino acid residues penetrating a 5 Å radius from the substrate
were minimized. Thus, the obtained position of the substrate was
used in DFT modelling.
27,28
modified method of Zavitsanos and co-workers
mal mobile phase with APCI-MS by 1:1 post column addition (i.e.,
.5 mL/min) of 25 mmol/L HCOONH buffer in 75:25 isopropanol/
to couple a nor-
Small cluster DFT modelling
The DFT cluster modelling was performed in Gaussian 09 ac-
cording to the procedure previously established.
3
3
7
,13
0
The choice of
4
water solution. Although such a post column addition dilutes the
sample, it provides stable electrostatic conditions for the mass spec-
trometer, decreases the chance of n-hexane explosion, and intro-
duces buffer ions for efficient APCI. Both scan and SIM modes were
used to detect the product ions of the mass spectra. The characteris-
tic [M – OH] originating from fragmentation of the alcohols by dehy-
droxylation dominated the mass spectra of the products and were
most frequently monitored in the SIM mode.
The absolute configuration of the reaction products was deter-
mined based on commercially available standards (either (S) or (R)
isomers) or self-synthesized standards. In some cases where no
commercial chiral standards were available and no pure products
were accessible by our synthetic reductive procedures (see the
Organic synthesis of standards section), the hypothetic configura-
tions of the products were established based on a quantitative
structure–enantioselective retention relationship (QSERR) model
previously obtained,10 combined with the retention times of frac-
tions from racemic standards. Up-to-date, this approach proved to
be effective in predicting the retention orders for a wide range of
the ethylbenzene derivatives.
DFT follows from its suitability to adequately describe reaction
3
4,35
pathways at a reasonable computational cost.
The Gaussian
3
3
09
package was used with a spin-unrestricted B3LYP func-
3
6
tional. Geometry optimization and vibration calculations were
conducted with the 6-31G(d,p) basis set for light atoms and the
LANL2DZ effective core potential and basis set for Mo. For a proper
description of homolytic C–H cleavage, the ␣–␤ orbital symmetry
was broken with a triplet solution in the intermediate product (I),
followed by conversion to a singlet state. The transition states
were localized by potential energy scans along the reaction coor-
dinates followed by full optimization of transition state (TS) ge-
ometries using the Berny algorithm. Single point energies were
calculated with the lacv3p** basis, both in gas phase and in a
+
model solvent (assuming a polarized continuum model with
37
␧
= 4 ) to correct for a continuous phase present in the enzyme
active site. The obtained electronic energies were corrected by
including zero point energies (ZPEs) obtained from B3LYP/6-31G(d,p)
frequency calculations and employing a scaling factor of 0.98. The
3
8
stable check was performed for each single point energy calcu-
lation, and showed that there are no lower-energy states for the
obtained geometries of stationary points. The energy differences
Published by NRC Research Press

7
78
Can. J. Chem. Vol. 91, 2013
Table 1. LC–MS analyses of EbDH reaction mixtures of OH and NH substituted substrates.
2
No.
Substrate
No.
Product
S (%)
R (%)
% ee
100
QSERR
IPA/n-hexane
1
I
100
0
10:90
1
-Phenylethanol
Ethylbenzene
2
3
II
100
100
0
0
100
100
15:85
15:85
2
-Ethylphenol
-Ethylphenol
2-(1-Hydroxyethyl)phenol
III
3-(1-Hydroxyethyl)phenol
3
4
IV
87–90
13–10
74–80
15:85
4
2
-(1-Hydroxyethyl)phenol
4
-Ethylphenol
-Propylphenol
5
6
V
100
90
S
10:90
15:85
-(1-Hydroxyprop-1-yl)phenol
2
VI
95
5
4
-(1-Hydroxyprop-1-yl)phenol
4-Propylphenol
7
VII
100
20:80
1
1
1
-(2-Aminophenyl)ethanol
-(3-Aminophenyl)ethanol
-(4-Aminophenyl)ethanol
2
-Ethylaniline
-Ethylaniline
8
9
VIII
IX
100
S
S
35:65
30:70
3
80–94
4-Ethylaniline
Published by NRC Research Press

Dudzik et al.
779
Table 1 (concluded).
No.
10
Substrate
No.
Product
S (%)
R (%)
% ee
10
QSERR
IPA/n-hexane
Xa
55
45
15:85
2
,3-Dihydro-1H-inden-5-ol
2
,3-Dihydro-1H-indene-1,5-
diol (66%)
Xb
100
15:85
10:90
2
,3-Dihydro-1H-indene-1,6-
diol (33%)
1
1
XIa
8
S
5
,6,7,8-Tetrahy-
dronaphthalen-2-ol
1,2,3,4-Tetrahydronaphtha-
lene-1,6-diol (53%)
XIb
100
1,2,3,4-Tetrahydronaphtha-
lene-1,7-diol (47%)
1
2
XIIa
3–23
S
S
15:85
15:85
2
,3-Dihydro-1H-inden-5-
amine
5
-Amino-2,3-dihydro-1H-
inden-1-ol
XIIIb
100
6-Amino-2,3-dihydro-1H-
inden-1-ol*
Note: S/R (%) amount of isomers, identified based on chiral standards, percent enantiomeric excess (% ee), quantitative structure–enantioselective retention
relationship (QSERR) identification of the major fraction based on the QSERR model,10 and isopropanol (IPA)/n-hexane composition of mobile phase in LC chiral
chromatography are presented.
were related to the enzyme substrate model (S) in which the po-
sition of 4-ethylphenol was derived from flexible docking. To en-
sure a similar position of the substrate in the cluster model as in
the enzyme active site, the optimization was performed with an
additional constraint between the Mo=O ligand and benzyl car-
bon atom. The transition state describing water displacement
(Fig. 3). The potential formation of a p-quinone ethide intermediate
(i.e., 4-ethylidenecyclohexa-2,5-dien-1-one) from 4-ethylphenol (4)
and a respective p-ethide of cyclohexadienimine (4-ethylidenecy-
clohexa-2,5-dien-1-imine) from 4-ethylaniline (5) as reaction interme-
diates was studied with a cluster model, as was the cofactor-assisted
reaction of the quinone ethide with a water molecule leading to the
formation of both enantiomers of 4-(1-hydroxyethyl)phenol (IV).
4
85
from Asp
by the quinone ethide intermediate was obtained
with a smaller cluster model containing an Asp residue, water
molecule, and substrate intermediate.
Results
Chromatographic studies
The results of chiral LC–MS analysis of EbDH reaction mix-
tures are shown in Table 1. (S)-Alcohol products with 100% ee
Thermochemical calculations
Simple thermochemical descriptors relating ⌬G of carbocation
formation from a particular substrate to ⌬G of carbocation forma-
tion from ethylbenzene (i.e., ⌬⌬G values of the respective carbo-
cations) were calculated in Gaussian 09 at the B3LYP/6-31G(d,p)
(
within the detection limit) were observed for ethylbenzene (1),
2- and 3-ethylphenol (2 and 3, respectively), 2-propylphenol (5),
and 2- and 3-ethylanilines (7 and 8, respectively). A decrease of
enantioselectivity of EbDH-catalyzed hydroxylations was ob-
served in the presence of OH and NH substituents in the para
level of theory for isolated organic compounds according to a
previously established protocol.39
2
Cluster model
position. This was experimentally confirmed with the phenolic
compounds 4-ethylphenol (4, 74%–80% ee), 4-propylphenol
(6, 95% ee), 2,3-dihydro-1H-inden-5-ol (10, 10% ee), and 5,6,7,8-
tetrahydronaphthalen-2-ol (11, 8% ee) and the amines 4-ethylaniline
To address the modified reaction pathway for 4-ethylphenol (4),
7
,13
the usual “high pH” cluster model
was expanded by the side
4
85
chain of Asp
(modelled as acetic acid) and a water molecule
Published by NRC Research Press

7
80
Can. J. Chem. Vol. 91, 2013
Fig. 3. The small cluster model of the active site of EbDH used for
DFT calculations. (A) Schematic representation (gray atoms indicate
frozen coordinates). (B) 3D structure of the active site with bound
Table 2. pH-dependence of hydroxylation enantioselectivity: enantio-
meric excesses for the conversion of 4-ethylphenol (4), 4-ethylaniline
(9), and 2,3-dihydro-1H-inden-5-amine (12).
4
-ethylphenol (4) as a substrate (the dotted line represents an
pH
additional constraint used in optimization of the S state).
Substrate
7.0
7.5
8.0
4
4
2
-Ethylphenol
-Ethylaniline
,3-Dihydro-1H-inden-5-amine
74%
80%
3%
79%
92%
8%
80.2%
96%
23%
Note: Values are the % enantiomeric excess of (S) isomers.
proposed hypothesis of an alternative reaction pathway involved
under these conditions (see below), the % ee values might be sen-
sitive to the pH conditions. Therefore, we determined the % ee
values of the products from some of these compounds at pH val-
ues between 7 and 8. A systematic trend was observed for all three
investigated compounds, which appeared to be hydroxylated with
a higher stereoselectivity with increasing pH (Table 2).
Modelling approaches to explain the loss of
enantioselectivity for p-hydroxy- or p-amino-substituted
substrates
The observed loss of reaction enantioselectivity may be ex-
plained with a modified reaction mechanism for p-OH- or p-NH2-
substituted compounds, which involves the formation of quinone
ethide/imine intermediates and the nonenantioselective addition
of H O. The proposed reaction was modelled by DFT as previously
2
7
,13
reported,
chain of Asp
but using an extended model that included the side
4
85
(see Fig. 3).
The reaction starts with the usual radical cleavage of the pro(S)
C–H bond (TS1) leading to the radical intermediate (I1) as shown in
1
3
our previous studies. The geometry of TS1 is similar to that
previously reported (see Fig. S1 in the Supplementary data), with
the exception of the H-bonding interactions of the p-OH group
4
85
with Asp and a water molecule. The next step is associated with
a transfer of the second electron, which leads to formation of a
carbonation intermediate and reduction of the cofactor to the
Mo(IV) state. At that point (I1), the reaction can either proceed
through the conventional steps (OH rebound via transition state
TS2, see Fig. 7A) to yield 4-((S)-1-hydroxyethyl)phenol and the re-
duced pentacoordinate molybdenum cofactor (MoCo; P1) (Fig. 4)
or, in an alternative route, the carbocation may form a quinone
ethide intermediate (I2). In the conventional pathway, the reac-
tion cycle is closed by the release of the product and coordination
of a water molecule to the active site, followed by oxidation of the
cofactor with concomitant loss of two protons, which restores the
catalytically active Mo(VI)=O species.
Quinone ethide formation should be greatly aided by interac-
4
85
tion with the Asp residue, which may accept the phenolic pro-
ton. To obtain such a species (I2), a second electron has to be
transferred from the radical substrate to the Mo atom and the
phenolic proton must be shifted to the nucleophilic acceptor. Our
(
9, 80%–94% ee) and 2,3-dihydro-1H-inden-5-amine (12, 3%–23% ee).
4
85
For the substituted 2,3-dihydro-1H-indene and tetrahydronaphtha-
lene derivatives (10, 11, and 12), only the major product peaks split
into the two enantiomers during chiral LC, whereas additional
minor product peaks showing the same MS spectra did not split
up. These latter products were assigned as hydroxylation products
of the alternative benzylic carbon of the alicyclic rings, which is in
the meta position relative to the hydroxyl or amino substituents.
The presence of only one enantiomer and coelution of the conver-
sion product from X with the synthesized standard (S)-Xb con-
firmed the (S) chirality of this product and suggested that this is
also the case for XIb and XIIb.
modelling showed that both the carboxylic group of Asp and a
water molecule hydrogen bonded to that group can act as proton
acceptors. The actual transition from radical I1 to quinone ethide
I2 (TS3) was characterized for the structure where the water mol-
ecule acts as an intermediary between the phenol OH group and
4
85
Asp
residue. Along the reaction coordinate (shift of a proton
between oxygen atoms), spin density on the substrate gradually
decreases, i.e., an electron is transferred in TS3. This proton-
coupled electron transfer is also associated with a gradual de-
crease of the S –S –S –S dihedral angle, which indicates a
1
2
3
4
reduction of the molybdenum atom from Mo(V) to Mo(IV). In the
transition state, two protons are being shifted, one from phenol to
From the varying levels of enantioselectivity observed with
-ethylphenol (4), 4-ethylaniline (9), and 2,3-dihydro-1H-inden-5-
4
85
4
water and a second from water to Asp , yielding a protonated
carboxylic group and quionon ethide. The actual energy barrier
associated with the formation of the quinone ethide turns out to
be very small (6 kJ/mol) and disappears when ZPE corrections are
amine (12) in different experiments (Table 1), we assumed that
variations in enzyme preparations or reaction conditions may
influence the enantiomeric purity of the product. Based on the
Published by NRC Research Press

Dudzik et al.
781
Fig. 4. Schematic representation of a conventional reaction pathway for 4-ethylphenol (4). S, enzyme substrate complex; TS1, C–H activation
transition state; I1, radical intermediate; TS2, OH rebound transition state; P1, S product with pentacoordiante MoCo; I2, quinone ethide
intermediate.
introduced (TS3 energy is lower than I1 by 2 kJ/mol). The quinone
ethide state (I2) is significantly more stable than the I1 radical
intermediate (⌬(E + ZPE) = –31.4 kJ/mol).
(Fig. 6C and Figs. 7A, B). Like in the conventional OH rebound
model for ethylbenzene, the steric interactions in the active site
would strongly favour a pro(S) orientation over pro(R). The transi-
4
85
The quinone ethide can further react along four different path-
ways. It can either displace the water molecule from its hydrogen
bond interaction with the Asp4 residue and mobilize it for addi-
tion to the C=C double bond, or a water molecule from the solvent
may enter the active site and become available for the reaction
tion state, one proton is shuttled back from Asp
to a water
molecule and another one from a water molecule to the quinone,
85
−
while the OH group is directly transferred from Mo to the C1
carbon atom. This reaction yields the product in S configuration
and pentacoordinated MoCo identical to that proposed for the
conventional pathway (P1S).
(see the Supplementary data). The energy barrier for the water
displacement, assessed from calculation on the smaller model, is
in the range of 17 kJ/mol (I2–I3, energy of 4.4 kJ/mol). The freed
water molecule can either form energetically favourable interac-
tions with the Mo–OH ligand (I3) or energetically neutral interac-
An analysis of energy profiles (Table 3, Fig. 8) of the resulting
modified mechanisms indicates that C–H activation is the rate
limiting step of the whole process (TS1, 107.6 kJ/mol). Moreover,
the energy of a “standard” rebound reaction of the intermediate
(in a pro(S) orientation) with the OH ligand of the Mo cofactor
(TS2, 35 kJ/mol) is significantly higher than the energy of the
alternative pathway involving water addition of a quinone ethide
intermediate either in the pro(S) or pro(R) conformation (TS4 en-
ergies are 13 and 14 kJ/mol for the pro(S) and pro(R) pathway,
respectively) or the energy for the reaction of the quinone ethide
with the Mo–OH ligand in the pro(S) orientation (TS6, 20.9 kJ/mol).
As there is no energy barrier between I1 and I2, the radical inter-
mediate will very likely evolve into the quinone ethide form in-
stead of directly reacting with the Mo–OH ligand (OH rebound).
From this (quinone ethide-bound) stage, various scenarios are en-
ergetically accessible. The water molecule may either be displaced
2
23
tions with the carboxylic oxygen atoms of Asp
(I4) (see Fig. 5).
The formation of I3 is associated with lowering of the system
energy by –15 kJ/mol, whereas formation of I4 requires 1.8 kJ/mol
more energy compared to I2. These intermediates contain a water
molecule capable of forming strong H bonds with nucleophilic
acceptors that can attack the C1 atom of the quinone ethide. The
orientation of the quinone in I3 allows an easy approach of H O
2
from either side of the benzyl carbon atom, which results in pro(S)
and pro(R) transition states (pro(S) and pro(R) TS4) leading to the
formation of S and R products (P2S and P3R), respectively (Fig. 6A
and Figs. 7C, D). In either case, the water molecule donates one of
its protons to the Mo–OH ligand concomitantly to hydrating the
double bond, while Asp4 reprotonates the quinone group of the
intermediate. It seems that the hydrogen bond between Mo–OH
and the water molecule has a crucial role in activating the latter
for the nucleophilic attack. The transition structures leading to
both enantiomers exhibit C1–O distances of around 2.0 Å, whereas
the ethyl side chain is either directed into the empty hydrophobic
cavity (for the pro(S) conformation) or in the opposite direction.
85
485
from its position at the Asp , further stabilizing the system by
–15 kJ/mol, or yet another water molecule may bind by H-bond
bridges to the Mocofactor. In the former scenario, the absolute bar-
riers associated with TS4s are calculated as 40.5 and 41.4 kJ/mol (for
pro(S) and pro(R), respectively), which are approximately 8 kJ/mol
higher than those for the reaction of the quinone ethide with the
hydroxyl ligand of the molybdenum (TS6). However, these energy
differences are mainly due to system stabilization after water dis-
placement. If another water molecule is involved in the reaction,
both types of transition states attain a similar energy. In such a sce-
nario, the barrier associated with TS6 is only 2 and 6.2 kJ/mol lower
than that with TS4 for pro(S) and pro(R) transition states, respectively
(see the Supplementary data).
223
Due to steric factors, the Asp -assisted attack of a water molecule
is only possible from the pro(R) side, yielding transition state TS5 in
which the (R)-alcohol is formed with the ethyl side chain pointing
toward the hydrophobic pocket (Fig. 6B). Similarly to TS4, the con-
comitant protonation of the quinone results in restoration of the
phenolic aromatic ring. This pathway leads to species P4 with a hexa-
223
coordinate molybdenum cofactor and protonated Asp
.
On the other hand, the reaction with a water molecule inter-
2
23
Finally, a fourth pathway is possible that involves direct transi-
tion from I2 to the product (P1). For this reaction route, we assume
that the quinone intermediate does not displace a water molecule
acting with Asp
is calculated to be associated with a signifi-
cantly higher energy barrier (TS5, 50.7 kJ/mol; absolute barrier
of 63.2 kJ/mol).
Finally, it should be underlined that reaction pathways asso-
ciated with water attack on the quinone ethide (TS4) yield hexa-
4
85
from Asp
but instead approaches the Mo–OH ligand while
forming a transition state (TS6) similar to that observed in TS2
Published by NRC Research Press

7
82
Can. J. Chem. Vol. 91, 2013
Fig. 5. Schematic representation of the steps leading to a proposed alternative pathway: formation of the quinone intermediate and three
different reaction intermediates. TS3, transition state associated with the electron transfer and formation of quinone ethide; I1, radical
4
85
intermediate; I2, quinone ethide intermediate with water bound to Asp ; I3, quinone ethide intermediate with water bound to Mo–OH; I4,
2
23
quinone ethide intermediate with water bound to Asp
.
Fig. 6. Four alternative reaction pathways of the proposed quinone intermediate: (A) Reactions involving a water molecule interacting with
Mo–OH: TS4, it shows either a (a) pro(S) or (b) pro(R) transition state after the benzylic carbon atom is attacked by water; P2, the S product
with a hexacoordinated Mo–H O; P3, the R product with a hexacoordinated Mo–H O. (B) The reaction involving a water molecule interacting
2
2
2
23
with Mo–Asp : TS5, the pro(R) transition state after water attack on the benzylic carbon atom; P4, the R product with a hexacoordinated
2
23
Mo–OH and protonated Asp . (C) The direct reaction of the quinone ethide intermediate with the Mo–OH ligand: TS6, transition state; P1,
the S product and pentacoordinated MoCo.
Published by NRC Research Press

Dudzik et al.
783
Table 3. Energy differences calculated for the 4-ethylphenol (4) reac-
tion via the conventional and modified proposed pathways.
(S)-alcohol, whereas the carbon in the para position is hy-
droxylated to mixtures of (S)- and (R)-alcohols. Besides the posi-
tioning of the substituent to be hydroxylated, the ability to subtract a
Stationary
point
⌬(E + ZPE)
gas phase (kJ/mol)
⌬(E + ZPE)
solvent (kJ/mol)
proton from the OH/NH substituent seems of utmost importance.
2
Indeed, hydroxylation of the nondissociable analog 1-ethyl-4-
S
TS1
I1
0
0
methoxybenzene proceeds with 100% ee to the (S)-alcohol.1
0
107.0
7.5
107.6
18.9
It was our intention to demonstrate by modelling studies
that the alternative reaction pathway(s) involving deprotona-
tion of the phenolic/anilinic group is feasible and may lead
through energetically accessible barriers to an increased
amount of (R)-alcohols. Our hypothetic reaction mechanism
involves stabilization of the reactive radical intermediate (I1) into
a form of the quinone ethide intermediate (I2). Such an interme-
diate may be formed by a direct proton abstraction from the para
TS2
P1
TS3
I2
I2–I3
29.7
34.9
–33.0
17.0
–12.5
4.4
–27.5
13.0
14.2
–55.9
–53.4
–10.8
50.7
18.0
20.9
–35.6
14.7
–2.8
18.0
–12.8
20.9
22.7
–55.7
–54.4
–0.9
60.4
26.9
19.5
a
I3
TS4 pro(S)
TS4 pro(R)
P2S
P3R
I4
TS5 pro(R)
P4R
TS6
substituent of the radical or the short-lived carbocation interme-
485
diate by the Asp
residue or via a water molecule interacting
485
through H bonds with the carboxylic group of Asp . Both pro-
cesses are energetically favourable, but the latter seems more
probable due to the presence of a water molecule in the vicinity of
4
85
Asp
and its likely interaction with the phenolic group during
C–H activation (TS1). The calculations showed no energy barrier
between the radical (I1) and quinone intermediates (I2), suggest-
ing a spontaneous proton-coupled electron transfer from the rad-
ical intermediate to the Mo–Co together with deprotonation of
the phenolic group. The formation of a stabilized quinone ethide
intermediate may then be followed by a nonenantioselective wa-
ter addition reaction proceeding with almost the same energy
barriers for pro(S) and pro(R) orientations (13 vs 14.2 kJ/mol), or by
a slightly enantioselective direct addition of the hydroxyl group of
Mo–OH (most probably preferentially yielding the (S)-alcohol). The
absolute barriers of both pathways (especially assuming the in-
volvement of a second water molecule) are in the same range with
a slight preference for the enantioselective attack of Mo–OH at the
quinone ethide. This result is in agreement with experiments
which still show some excess of the (S)-enantiomer over the (R).
The modelling also showed that the carboxylic group of Asp223,
which coordinates Mo as a protein-derived ligand, can potentially
bind another water molecule (I4), which in turn may attack the
quinone ethide intermediate. However, this process is associated
with a much higher energy barrier (TS5, 50.7 kJ/mol), and the
product (P3) exhibits a higher energy than those involved in the
above discussed pathways. Thus, we conclude that the latter mecha-
nism does not significantly contribute to the observed effect.
Note: The conventional pathway is marked in italic.
Energy estimated from a smaller model.
a
coordinated MoCo with a water ligand, which is ready for
deprotonation and restoration of the catalytically active Mo(VI)=O
species. As a result, P2 and P3 have significantly lower energies
than P1 (by 23 kJ/mol).
In the case of 4-ethylaniline (9), a similar spontaneous forma-
tion of 4-ethylidenecyclohexa-2,5-dien-1-imine (⌬(E + ZPE) =
–
9.5 kJ/mol) may occur after the transfer of the second electron
from the radical intermediate, providing an analogous expla-
nation for the low enantioselectivity (see Fig. S3 in the Supple-
mentary data for the structure of the imine intermediate).
However, we did not calculate the complete pathway for this
substrate.
Discussion
The preliminary modelling studies (data not shown) conducted
for the conventional reaction pathway of EbDH (ethylbenzene
hydroxylation) indicated that the high enantioselectivity of the
reaction is imposed by stereoselective activation of a C–H bond
followed by a rapid reaction of a carbocation intermediate with
the hydroxyl group of Mo–Co. The second step of the hy-
droxylation mechanism (i.e., OH rebound) is believed to be less
enantioselective, because the C1···OH–Mo bond length in TS2 is
longer than the H···O–Mo distance involved in the C–H activation
process (TS1). Due to the longer distance between the reactant and
MoCo, both pro(R) and pro(S) conformations of TS2 may occur in
the active site as the ethyl side change is further remote from the
residues determining enantioselective steric interactions. How-
ever, for most substrates the I1 intermediates seem to be short-
lived species, and their potential pro(S)-pro(R)-rotation does not
seem to contribute significantly to increasing the production of
R)-alcohols through the pro(R) TS2 OH rebound process.
Therefore, the observed severe loss of reaction enantioselectivity
One can also imagine that a water molecule may approach the
quinone ethide from the pro(R) side of the molecule without in-
223
teracting with either Mo–OH or Asp . In such a situation, the
water would not be activated for nucleophilic attack and the re-
sulting product would be a protonated alcohol. However, this
scenario would be energetically unfavorable and therefore was
not considered in this paper.
1
4
The hypothetical alternative quinone–water reaction path-
way presented above is also supported by the observed pH de-
pendence on enantioselectivity for 4, 9, and 12, which may
(
4
85
reflect the (de)protonation grade of Asp
and its ability to
abstract the phenolic proton, or its ability to reprotonate the
in the case of para-substituted phenols and anilines is most probably
associated with changes in the reaction pathway of the second part
of the reaction, i.e., after the enantioselective C–H activation.
The enantioselectivity of EbDH with substrates substituted with
quinone ethide (imine) during H O attack on the C=C bond.
2
Another hypothetical explanation for the observed loss of en-
antioselectivity could imply an extended lifespan of the carboca-
tion intermediate due to stabilizing effects of the p-OH or p-NH2
substituents. This hypothesis would reject the formation of qui-
none/imine ethide intermediates (I2). The extended stabilization
NH or OH groups is clearly dependent on directional interactions
2
of those substituents with a reaction site at the benzylic carbon
atom. In contrast to the para-substituted compounds, meta-
and ortho-substituted analogs are exclusively converted to (S)-
alcohols. This property becomes especially prominent in the case
of substrates with two possible hydroxylation sites either in the m-
or p-position to a hydroxyl or amino substituent (10, 11, 12). The
carbon in the meta position is always hydroxylated to a pure
of a carbocation intermediate might facilitate rotation of the ac-
tivated form in the active site and formation of the (R)-alcohol via
ϩ
the standard OH rebound process. The Hammett ␴ parameters
para
for p-OH and p-NH (–0.92 and –1.3, respectively), as well as calcu-
2
lated ⌬⌬Gcarbocation values for phenols and anilines (–46.6, –48.8,
Published by NRC Research Press

7
84
Can. J. Chem. Vol. 91, 2013
Fig. 7. Geometries of the transition states of alcohol formation from 4-ethylphenol: (A) the conventional pro(S)-oriented OH rebound reaction
TS2); (B) the OH rebound reaction to the quinone ethide intermidiate (TS6); (C) and (D) the modified pro(S)- and pro(R)-oriented TS4 structures
(
of the reaction pathways with the involved water molecules, respectively. Structures of all stationary points are available in the
Supplementary data (Figs. S3 and S4).
Fig. 8. Energy diagram of reaction pathways (⌬(E + ZPE) solvent): Black, conventional pathway (only S enantiomer); green, quinone ethide
pathway with H O bound to Mo–OH (dark green - pro(S) and light green - pro(R) pathways); blue, quinone ethide pathway with H O bound at
2
2
2
23
Asp ; red, quinone ethide pathway with hydroxyl transfer from Mo–OH (only S enantiomer). The energy scale is in kJ/mol.
Published by NRC Research Press

Dudzik et al.
785
–
70, –74, –97, and –118 kJ/mol for 4, 6, 10, 11, 9, and 12, respec-
Project in Nanoscience and Advanced Nanostructures” operated
within the Foundation for Polish Science MPD Programme and cofi-
nanced by the EU European Regional Development Fund.
tively) indicate that p-OH or p-NH groups should indeed princi-
pally stabilize carbocation species. However, this hypothesis is in
2
contrast with the observation that 1-ethyl-4-methoxybenzene,
which should also produce a strongly resonance-stabilized car-
References
ϩ
para
bonation intermediate (␴
= –0.78, ⌬⌬Gcarbocation = –60 kJ/mol),
(1) Kniemeyer, O.; Heider, J. J. Biol. Chem. 2001, 276 (24), 21381–21386. doi:10.
1
0
1074/jbc.M101679200.
2) Hille, R. The Mononuclear Molybdenum Enzymes. Chem. Rev. 1996, 96 (7),
is hydroxylated completely enantioselectivly (100% (S)). More-
over, the significant loss of enantioselectivity due to rotation of
the activated 4-ethylphenol would require much higher energy
barriers between I1 and TS2 than those calculated for ethylben-
zene. Our calculations indicate, however, that these barriers are
approximately the same for both substrates (ϳ40 kJ/mol). There-
fore, the mere stabilization of a carbocation intermediate and its
increased probability to rotate in the active site does not seem to
be a primary reason for enantioselectivity loss, especially in the
(
2757–2816. doi:10.1021/cr950061t.
(3) Johnson, H. A.; Pelletier, D. A.; Spormann, A. M. J. Bacteriol. 2001, 183 (15),
4536–4542. doi:10.1128/JB.183.15.4536-4542.2001.
(
(
4) Heider, J. Curr. Opin. Chem. Biol. 2007, 11 (2), 188–194. doi:10.1016/j.cbpa.2007.
2.027.
0
5) Rabus, R.; Kube, M.; Heider, J.; Beck, A.; Heitmann, K.; Widdel, F.;
Reinhardt, R. Arch. Microbiol. 2005, 183 (1), 27–36. doi:10.1007/s00203-004-
0742-9.
(
6) Kloer, D. P.; Hagel, C.; Heider, J.; Schulz, G. E. Structure 2006, 14 (9),
1377–1388. doi:10.1016/j.str.2006.07.001.
case of OH and NH para-substituted compounds.
2
(7) Szaleniec, M.; Borowski, T.; Schuhle, K.; Witko, M.; Heider, J. J. Am. Chem. Soc.
2010, 132 (17), 6014–6024. doi:10.1021/ja907208k.
(8) Szaleniec, M.; Witko, M.; Tadeusiewicz, R.; Goclon, J. J. Comput. Aided Mol.
Des. 2006, 20 (3), 145–157. doi:10.1007/s10822-006-9042-6.
Summing up, we suggest that the loss of reaction enantioselectiv-
ity for some EbDH substrates (e.g., 4-ethylphenol, 4-ethylaniline,
or 2,3-dihydro-1H-inden-5-ol and 2,3-dihydro-1H-inden-5-amine) is
caused by deprotonation of a carbocation intermediate at a hydroxyl
or amino ligand in the para position yielding a quinone ethide inter-
midiate, followed by water addition to that intermediate.
(
9) Szaleniec, M.; Hagel, C.; Menke, M.; Nowak, P.; Witko, M.;
Heider, J. Biochemistry 2007, 46 (25), 7637–7646. doi:10.1021/bi700633c.
(10) Szaleniec, M.; Dudzik, A.; Pawul, M.; Kozik, B. J. Chromatogr. A 2009, 1216
(34), 6224–6235. doi:10.1016/j.chroma.2009.07.002.
(
11) Knack, D.; Hagel, C.; Szaleniec, M.; Dudzik, A.; Salwinski, A.; Heider, J. Appl.
Environ. Microb. 2012, 78 (18), 6475–6482. doi:10.1128/aem.01551-12.
12) Szaleniec, M. Pharmacol. Rep. 2012, 64, 511–531.
Conclusions
(
The reaction mechanism of EbDH is extended by an alternative
variant that comes into play for para-substituted alkylphenols and
alkylanilines. The observed loss in enantioselectivity for these sub-
strates may be best explained as a nonenantioselective side reaction
of a stabilized quinone ethide intermediate with a water molecule
activated by hydrogen bonding to Mo–OH. The involvement of pro-
ton transfer in the reaction was suggested by the observation of
pH-dependent variations of the enantiomer compositions. This ef-
fect supported our hypothesis and may even be used to further opti-
mize reaction conditions to modify stereoselectivity.
(13) Szaleniec, M.; Salwinski, A.; Borowski, T.; Heider, J.; Witko, M. Int. J. Quantum
Chem. 2012, 112 (8), 1990–1999. doi:10.1002/Qua.23143.
(
14) Szaleniec, M.; Dudzik, A.; Tataruch, M.; Witko, M.; Knack, D.; Heider, J.
Enantioselective hydroxylation by ethylbenzene dehydrogenase – origins of
enantioselectivity; In 6th International Congress on Biocatalysis BIOCAT 2012,
Hamburg; 2012, p. 262
(
(
15) Reider, P. J. Advances in AIDS CHIMIA 1997, 51 (6), 306–308.
16) Gen-Sheng, Y.; Jiang-Yan, X.; Zhi-Min, O.; Shan-Jing, Y. Biotechnol. Lett. 2009,
3
1 (12), 1879–1883. doi:10.1007/s10529-009-0084-4.
(
(
(
17) Suan, C.; Sarmidi, M. R. J. Mol. Catal. B Enzym. 2004, 28 (2-3), 111–119. doi:10.
1016/j.molcatb.2004.02.004.
18) Hanson, R. L.; Goldberg, S.; Goswami, A.; Tully, T. P.; Patel, R. N. Adv. Synth.
Catal. 2005, 347 (7-8), 1073–1080. doi:10.1002/adsc.200505045.
19) Kaneoya, M.; Uchida, M.; Yoshida, N.; Miyazawa, K. 1-Phenyl-1-propanol
derivatives; Edited by EP; Chisso Corporation, CH DE FR GB IT LI SE; 1991, pp.
Supplementary data
1
–17.
Supplementary data (details of the product identification (Table S1),
absolute energies of stationary points for the 4-ethylphenol path-
way (Table S2), energy differences for the pathway with two water
molecules (Table S3), figures of the stationary points for the 4-
ethylphenol reaction pathway (Fig. S1), figures of the stationary
(
(
(
(
20) Pal, S.; Park, D.-H.; Plapp, B. V. Chem-Biol. Interact. 2009, 178 (1–3), 16–23.
doi:10.1016/j.cbi.2008.10.037.
21) Uwai, K.; Konno, N.; Yasuta, Y.; Takeshita, M. Bioorg. Med. Chem. 2008, 16 (3),
1084–1089. doi:10.1016/j.bmc.2007.10.096.
22) Hopper, D. J.; Cottrell, L. Appl. Environ. Microb. 2003, 69 (6), 3650–3652. doi:
10.1128/Aem.69.6.3650-3652.2003.
points for the alternative pathway “quinone – Mo–OH–H O”
2
23) Knack, D. H.; Marshall, J. L.; Harlow, G. P.; Dudzik, A.; Szaleniec, M.;
Liu, S. Y.; Heider, J. Angew. Chem. Int. Ed. Eng. 2013, 52 (9), 2599–2601. doi:10.
1002/anie.201208351.
(
Fig. S2), figures of the stationary points for the alternative path-
2
23
way “quinone – Mo–Asp –H O” (Fig. S3), figure of TS6 for the
2
(
24) Kniemeyer, O.; Heider, J. J. Biol. Chem. 2001, 276 (24), 21381–21386. doi:10.
alternative pathway “quinone – Mo–OH” (Fig. S4), figures of the
1074/jbc.M101679200.
stationary points for the alternative pathway quinone – 2H O–
2
(25) Kniemeyer, O.; Heider, J. Arch. Microbiol. 2001, 176 (1-2), 129–135. doi:10.1007/
s002030100303.
(26) Hoffken, H. W.; Duong, M.; Friedrich, T.; Breuer, M.; Hauer, B.;
Reinhardt, R.; Rabus, R.; Heider, J. Biochemistry 2006, 45 (1), 82–93. doi:10.
4
85
OH–Mo Asp (Fig. S5), figure of the I2 intermediate calculated for
the 4-ethylaniline alternative pathway (Fig. S6), and figures of
small cluster models used for evaluation of the I2–I3 water dis-
placement process (Fig. S7)) are available with the article through
the journal Web site at http://nrcresearchpress.com/doi/suppl/
1
024/bi051596b.
(
27) Alebic-Kolbah, T.; Paul Zavitsanos, A. J. Chromatogr. A 1997, 759 (1-2), 65–77.
doi:10.1016/s0021-9673(96)00756-x.
(28) Zavitsanos, A. P.; Alebic-Kolbah, T. J. Chromatogr. A 1998, 794 (1-2), 45–56.
10.1139/cjc-2012-0504.
doi:10.1016/s0021-9673(97)00892-3.
(
29) Brooks, B. R.; Bruccoleri, R. E.; Olafson, B. D.; States, D. J.; Swaminathan, S.;
Acknowledgements
Karplus, M. J. Comput. Chem. 1983, 4 (2), 187–217. doi:10.1002/jcc.540040211.
The authors acknowledge the financial support of the prior-
ity program 1319 of the German Research Foundation (DFG), the
excellence program The LOEWE Center for Synthetic Microbiology
(30) Momany, F. A.; Rone, R. J. Comput. Chem. 1992, 13 (7), 888–900. doi:10.1002/
jcc.540130714.
(
31) Singh, U. C.; Kollman, P. A. J. Comput. Chem. 1984, 5 (2), 129–145. doi:10.1002/
jcc.540050204.
(LOEWE/SYNMIKRO) from the state of Hessen, Germany, the compu-
(32) Besler, B. H.; Merz, K. M.; Kollman, P. A. J. Comput. Chem. 1990, 11 (4), 431–439.
tational grant MNiSW/SGI3700/PAN/121/2006, the Marian Smo-
luchowski Krakow Research Consortium - a Leading National
Research Centre KNOW supported by the Ministry of Science and
Higher Education, the Polish National Science Center under grant N
N204 269038, as well as the financial support of the project “Biotrans-
formations for pharmaceutical and cosmetics industry” No.
POIG.01.03.01-00-158/09-00, which is in part financed by the Euro-
pean Union (EU) within the European Regional Development Fund.
Anna Wójcik acknowledges project “Kraków Interdisciplinary PhD –
(33) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.;
Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.;
Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.;
Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.;
Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Nor-
mand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi,
J.;Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Moro-
Published by NRC Research Press

7
86
Can. J. Chem. Vol. 91, 2013
kuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
Dapprich, S.; Daniels, A. D.; Farkas, Ö.; Foresman, J. B.; Ortiz, J. V.; Ci-
oslowski, J.; Fox, D. J. Gaussian 09 Revision A.01; Gaussian, Inc., Wallingford
CT, 2009.
(36) Becke, A. D. J. Chem. Phys. 1993, 98 (7), 5648–5652. doi:10.1063/1.464913.
(37) Cossi, M.; Scalmani, G.; Rega, N.; Barone, V. J. Chem. Phys. 2002, 117 (1), 43.
doi:10.1063/1.1480445.
(38) Bauernschmitt, R.; Ahlrichs, R. J. Chem. Phys. 1996, 104 (22), 9047–9052.
doi:10.1063/1.471637.
(
34) Rutkowska-Zbik, D.; Witko, M.; Serwicka, E. M. Catalysis Today 2011, 169 (1),
0–15. doi:10.1016/j.cattod.2010.09.014.
35) Rutkowska-Zbik, D.; Witko, M.; Serwicka, E. M. Catalysis Today 2004, 91-2,
37–141. doi:10.1016/j.cattod.2004.03.022.
1
(
39) Szaleniec, M.; Witko, M.; Heider, J. J. Mol. Catal. A Chem. 2008, 286 (1-2),
(
128–136. doi:10.1016/j.molcata.2008.02.016.
1
Published by NRC Research Press