Báo cáo khoa học: Substrate-dependent hysteretic behavior in StEH1-catalyzed hydrolysis of styrene oxide derivatives docx

Substrate-dependent hysteretic behavior
in StEH1-catalyzed hydrolysis of styrene
oxide derivatives
Diana Lindberg, Adolf Gogoll and Mikael Widersten
Department of Biochemistry and Organic Chemistry, Uppsala University, Sweden
Soluble epoxide hydrolases (EC 3.3.2.10) make up a
large group of enzymes catalyzing the hydrolysis of
alkyl and aryl epoxides into the corresponding vicinal
diols [1,2]. They fulfill various functions in host
organisms, including detoxification by hydrolysis of
endogenous and exogenous epoxides, regulation of
cell signaling by hydrolysis of epoxide-containing
bioactive lipids, or participation in the secondary
metabolism of microorganisms. In plants, epoxide
hydrolases have been suggested to contribute to path-
ogen defense systems through hydrolysis of epoxy-
containing hydroxyl fatty acids. The diol products
from the latter reaction show antifungal activity [3]
and are established substrates for several plant iso-
enzymes as precursors in cutin synthesis [4]. The inde-
pendence from cofactors in combination with, in
some cases, high catalytic efficiencies and enantio-
selectivities has created an interest in using epoxide
hydrolases as biocatalysts in the production of fine
chemicals [5,6].
Styrene oxide (SO) and derivatives thereof are
important molecules as chiral and prochiral precursors
in asymmetric synthesis. These compounds are also
relevant for their toxicological impact [7]. Solanum
tuberosum epoxide hydrolase 1 (StEH1) has previously
been investigated using different SO derivatives as

Keywords
epoxide hydrolase; kinetic mechanism; pre-
steady state; regiospecificity; styrene oxide
Correspondence
M. Widersten, Box 576, SE-751 23 Uppsala,
Sweden
Fax: +46 0 18 55 8431
Tel: +46 0 18 471 4992
E-mail:
Website:
(Received 12 September 2008, revised 20
October 2008, accepted 22 October 2008)
doi:10.1111/j.1742-4658.2008.06754.x
The substrate selectivity and enantioselectivity of Solanum tuberosum
epoxide hydrolase 1 (StEH1) have been explored by steady-state and pre-
steady-state measurements on a series of styrene oxide derivatives. A prefer-
ence for the (S)- or (S,S)-enantiomers of styrene oxide, 2-methylstyrene oxide
and trans-stilbene oxide was established, with E-values of 43, 160 and 2.9,
respectively. Monitoring of the pre-steady-state phase of the reaction with
(S,S)-2-methylstyrene oxide revealed two observed rates for alkylenzyme for-
mation. The slower of these rates showed a negative substrate concentration
dependence, as did the rate of alkylenzyme formation in the reaction with
the (R,R)-enantiomer. Such kinetic behavior is indicative of an additional,
off-pathway step in the mechanism, referred to as hysteresis. On the basis of
these data, a kinetic mechanism that explains the kinetic behavior with
all tested substrates transformed by this enzyme is proposed. Regioselectivity
of StEH1 in the catalyzed hydrolysis of 2-methylstyrene oxide was
determined by
13
C-NMR spectroscopy of

18
O-labeled diol products. The
(S,S)-enantiomer is attacked exclusively at the C-1 epoxide carbon, whereas
the (R,R)-enantiomer is attacked at either position at a ratio of 65 : 35 in
favor of the C-1 carbon. On the basis of the results, we conclude that differ-
ences in efficiency in stabilization of the alkylenzyme intermediates by StEH1
are important for enantioselectivity with styrene oxide or trans-stilbene oxide
as substrate. With 2-methylstyrene oxide, slow conformational changes in
the enzyme also influence the catalytic efficiency.
Abbreviations
2-MeSO, 2-methylstyrene oxide; ES, enzyme–substrate; SO, styrene oxide; StEH1, Solanum tuberosum epoxide hydrolase 1; TSO,
trans-stilbene oxide.
FEBS Journal 275 (2008) 6309–6320 ª 2008 The Authors Journal compilation ª 2008 FEBS 6309
substrates, both in buffer and immobilized on different
matrices [8–15].
As with other isoenzymes of the a ⁄ b-hydrolase fold
family, the kinetic mechanism of StEH1 catalysis in
its simplest form consists of the three steps outlined
in Fig. 1. Formation of the Michaelis complex is fol-
lowed by a nucleophilic attack by the active site
Asp105 carboxylate to generate a covalent alkyl-
enzyme intermediate [16,17]. The alkylation half-reac-
tion is also dependent on electrophilic catalysis by
two active-site Tyr residues (Tyr154 and Tyr235)
[13,18,19]. Catalytic turnover is subsequently finalized
by a general base-assisted hydrolysis of the alkylen-
zyme, facilitated by the imidazole of His300, and
product release [10,20–22]. In principle, all of the
described steps may influence the overall stereospeci-
ficity of catalysis.

The active site architecture, including the spatial
arrangement of catalytic groups, is expected to place
restrictions on the productive substrate-binding
modes. For instance, good alignments of the epoxide
for nucleophilic attack on an electrophilic carbon by
the carboxylate and for the Tyr phenols previously
mentioned are crucial for effective catalysis. Alkyl-
enzyme formation depends on efficient activation of
the nucleophile via deprotonation of the carboxyl
group, stabilization of the leaving group oxide
through hydrogen bonding to the Tyr pair, and subse-
quent protonation of the alkylenzyme anion formed
[23]. In the final catalytic step, efficient hydrolysis
requires that the alkylenzyme, following attack by
water, undergoes a structural change in order to stabi-
lize the formed oxyanion through amide backbone
interactions [24–26]. Enzyme structure flexibility is
therefore also expected to influence catalytic efficiency
and enantiopreference.
StEH1 exhibits reasonable stereoselectivity with SO,
with different regiospecificity in the initial nucleophilic
attack by the enzyme on the different enantiomers,
resulting in an enantioconvergent synthesis of (R)-1-
phenylethane-1,2-diol [9]. Other studies on the hydroly-
sis of trans-stilbene oxide (TSO) suggest that the
kinetic mechanisms, under certain conditions, may
differ with substrate enantiomers [15].
Stereodiscrimination with glycidyl-4-nitrobenzoate
displayed by the rat microsomal epoxide hydrolase has
been attributed primarily to the hydrolysis half-reac-

tion [20], whereas it is the alkylation step that has been
shown to mainly influence the enantiospecificity in SO
hydrolysis by the Agrobacterium radiobacter AD1
enzyme [27]. This illustrates that the mechanisms caus-
ing stereospecificity may vary with isoenzyme and⁄ or
substrate.
In addition to the enzyme-afforded catalytic steps, a
further level of reaction complexity is added with SO
derivatives as substrates. The phenyl substituent of
these compounds can assist in stabilizing electron-rich
reaction intermediates, hence affecting the regioselec-
tivity in the ring-opening half-reaction by directing the
attack away from the otherwise more reactive, least
substituted, of the two oxirane carbons [28,29].
In this work, we aimed to systematically analyze the
causes of StEH1 substrate selectivity, using different
SO structural analogs. SO, 2-methyl styrene oxide
(2-MeSO) and TSO, with increasing size and hydro-
phobicity on the C-2 substituent, were used to derive
these structure–function relationships. In addition, the
regioselectivity in StEH1-catalyzed hydrolysis of
2-MeSO was analyzed by
13
C-NMR spectroscopy of
18
O-labeled diol products.
Results and Discussion
Steady-state measurements
The ratio of k
cat

⁄ K
m
values for two substrates is a
direct measure of the enzymatic substrate discrimina-
tion. The lower K
m
values obtained as a result of these
measurements with the (S)- or ( S ,S )-enantiomers
as compared to their (R)or(R,R) counterparts results
in relatively high enantiospecificity values [E =
(k
cat
S
⁄ K
m
S
) ⁄ (k
cat
R
⁄ K
M
R
)] for SO and 2-MeSO; 43-fold
and 160-fold, respectively (Table 1). This degree of
enantiospecificity in favor of (S)-SO is in agreement
with previous studies, in which E-values have been
estimated from specific enzyme activities and product
composition analyses following hydrolysis of racemic
SO [9,12]. When comparing the steady-state parame-
ters derived from the reactions with (S,S)-2-MeSO and

Fig. 1. Kinetic mechanism of StEH1-catalyzed epoxide hydrolysis.
The mechanism includes the formation of a covalent alkylenzyme,
formed with rate k
2
, following the Michaelis complex formation.
The alkylenzyme is subsequently hydrolyzed, with rate k
3
,to
restore the enzyme. K
S
, the equilibrium dissociation constant of the
ES complex, is the ratio of the dissociation and association rate
constants, k
)1
⁄ k
1
.
Enzyme-catalyzed styrene oxide hydrolysis D. Lindberg et al.
6310 FEBS Journal 275 (2008) 6309–6320 ª 2008 The Authors Journal compilation ª 2008 FEBS
the corresponding SO enantiomer, it is clear that
(S,S)-2-MeSO is the favored substrate. The k
cat
(S,S)-2-
MeSO
is the highest recorded for this enzyme with any
substrate, and is approximately 10-fold higher than k
cat
for the (R,R)-enantiomer (Table 1).
The enantiomer discrimination displayed by the
enzyme in SO or 2-MeSO hydrolysis is virtually lost

with TSO. The enzyme exhibits comparably low turn-
over rates with TSO, but low K
m
values compensate to
yield high values of k
cat
⁄ K
m
(Table 1). Hence, energy
gained from favorable enzyme–substrate (ES) interac-
tions is more important than turnover rate in deciding
catalytic efficiency.
A modeling study [26] has suggested that the size and
shape of the active site is well adapted for fitting the
TSO enantiomers, restricting binding to snugly fitted
conformations for both enantiomers. The resulting
free energy realized in the ES complexes can be
calculated from the K
S
parameters (Table 2) to be )25
and )28 kJÆmol
)1
for the (R,R)- and ( S ,S)-enantiomers,
respectively. The corresponding energies for the SO or
(S,S)-2-MeSO substrates are > 6 kJÆmol
)1
less favor-
able in those Michaelis complexes. This ground state
stabilization of the TSO enantiomers will be unfavor-
able for subsequent chemical steps, by increasing the

energy barrier for formation of alkylenzyme, reflected
as comparably lower values of k
2
(Table 2). In the
(S)-enantiomer series, where data are available for all
tested derivatives, the alkylation rate for TSO is the
slowest by a factor of at least three-fold. The snug fit of
the TSO enantiomers in the active site may further act
to prevent the dynamic movement necessary for effi-
cient hydrolysis of the alkylenzyme, which is possibly
reflected in the relatively low values of the k
3
and k
cat
parameters (Tables 1 and 2).
The kinetic mechanism involved in StEH1
hydrolysis of TSO and SO
It is presupposed within the simple kinetic scheme out-
lined in Fig. 1 that a rapid equilibrium forms between
free enzyme, substrate, and the Michaelis complexes.
Another assumption made within the model is that
product release is not rate-limiting for turnover, as
supported by the poor inhibition of the hydrolysis
products in these reactions [9,10]. From the kinetic
model in Fig. 1, it can be seen that the corresponding
steady-state rate law is described by Eqn (1), where
k
cat
is composed of the first factor in the numerator
and K

m
of the first term in the denominator:
v ¼
k
2
k
3
ðk
2
þ k
À2
þ k
3
Þ
½E
0
½S
K
S
ðk
À2
þ k
3
Þ
ðk
2
þ k
À2
þ k
3

Þ
þ½S
ð1Þ
In the pre-steady-state measurements, formation of
the alkylenzyme reaction intermediates was recorded
as concomitant decreases in intrinsic Trp fluorescence
[13]. The fluorescence decrease under pseudo-first-order
conditions of both enantiomers of SO and TSO, as
Table 1. Steady-state kinetic parameters and E-values of the StEH1-catalyzed hydrolysis of styrene oxide and 2-substituted styrene oxide
derivatives.
Substrate k
cat
(s
)1
) K
m
(lM) k
cat
⁄ K
m
(s
)1
ÆlM
–1
) E (S) ⁄ (R)
(S)-SO 10 ± 1.5 220 ± 88 0.047 ± 0.0012 43
(R)-SO > 9 > 400 0.0011 ± 0.000096
(S,S)-2-MeSO 52 ± 2.6 99 ± 1.7 0.53 ± 0.062 160
(R,R)-2-MeSO 5.7 ± 1.5 1400 ± 590 0.0034 ± 0.0018
(S,S)-TSO 0.91 ± 0.063 0.31 ± 0.0091 2.9 ± 0.72 2.9

(R,R)-TSO 15 ± 1.1 16 ± 2.7 1.0 ± 0.11
Table 2. Pre-steady-state kinetic and thermodynamic parameters of StEH1-catalyzed hydrolysis of styrene oxide derivatives.
Substrate K
S
(lM)
k
0
+ k
)0
(s
)1
) k
2
(s
)1
) k
)2
(s
)1
) k
3
(s
)1
)
k
5
+ k
)5
(s
)1

)
k
)2
+ k
3
(s
)1
)
k
2
⁄ K
S
(s
)1
ÆlM
)1
)
(S)-SO > 400 – > 85 – – – 55 ± 2.9 0.11 ± 0.0061
(R)-SO > 600 – > 16 – – – 41 ± 2.2 0.013 ± 0.0031
(S,S)-2-MeSO > 500 80 > 300 – – 20 310 ± 12 0.30 ± 0.036
(R,R)-2-MeSO – 80 – – – 10 – –
(S,S)-TSO 14 ± 7.7 – 29 ± 5.8 14 ± 1.9 1.4 ± 0.36 – – 3.3 ± 1.5
(R,R)-TSO 42 ± 37 – 51 ± 6.9 26 ± 9.4 32 ± 0.37 – – 2.6 ± 2.0
D. Lindberg et al. Enzyme-catalyzed styrene oxide hydrolysis
FEBS Journal 275 (2008) 6309–6320 ª 2008 The Authors Journal compilation ª 2008 FEBS 6311
well as the (R,R)-enantiomer of 2-MeSO, follows first-
order exponential decays, allowing for extraction of
observed rates (k
obs
) for the reactions from the ES

complex to alkylenzyme. Observed rates for the reac-
tions with either enantiomer of TSO show hyperbolic
substrate dependence (Fig. 2A). Fitting of Eqn (2) to
data was used to determine the equilibrium dissocia-
tion constants of the ES complex (K
S
), as well as the
rate constants for formation and decomposition of the
alkylenzyme: k
2
and k
)2
+ k
3
, respectively. The values
of k
–2
and k
3
were extracted from the expression for
k
cat
(Table 2):
k
obs
¼
k
2
½S
K

S
þ½S
þðk
À2
þ k
3
Þð2Þ
The obtained rate constants and K
S
values for the
reaction with (S,S)-TSO agree with previous studies on
the pH dependence of StEH1, which predicted an
increase in k
2
with a concomitant decrease in k
3
when
the pH is changed from 6.8 to 8.0 [13]. The resulting
increase in accumulation of the alkylenzyme contrib-
utes to lower the value of K
m
(S,S)-TSO
from 5 lm at
pH 6.8 to 0.31 lm at pH 8.0. As K
S
(S,S)-TSO
is not
significantly affected by the pH shift (Table 3), it
A B
C

D
Fig. 2. Observed rates, k
obs
, versus substrate concentration for tested SO derivatives. (A) (S,S)-TSO (unfilled squares) and (R,R)-TSO (filled
squares). Solid lines represent the fits to Eqn (2). (B) (S)-SO (unfilled triangles) and (R)-SO (filled triangles). Solid lines represent the fits to
Eqn (3). (C) Substrate dependence on observed higher rate, k
obs1
, in the (S,S)-2-MeSO reaction (unfilled circles). The solid line represents
the fit to Eqn (3). Inset: average of five traces of fluorescence decay in the presence of 1000 l
M (S,S)-2-MeSO and 2 lM StEH1. Lines repre-
sent a single (dashed line) or a double (straight line) exponential fit to the experimental data. Residual plots of the respective fits are shown
in boxes. (D) Substrate dependence of the low rate, k
obs2
, with (S,S)-2-MeSO (half-filled circles) and (R,R)-2-MeSO (filled circles). Solid lines
represent the fits to Eqn (4). See Table 2 for determined values of kinetic parameters.
Enzyme-catalyzed styrene oxide hydrolysis D. Lindberg et al.
6312 FEBS Journal 275 (2008) 6309–6320 ª 2008 The Authors Journal compilation ª 2008 FEBS
follows that the changes in formation and decomposi-
tion rates of the alkylenzyme are mainly responsible
for the lowering of the K
m
value. The lower
K
m
(S,S)-TSO
in turn increases the k
cat
⁄ K
m
(S,S)-TSO

value.
With (R,R)-TSO, however, the increase in pH from 6.8
to 8.0 causes a decrease in k
2
(Table 3), which affects
both k
cat
and K
m
adversely to yield a lower value of
k
cat
⁄ K
m
(R,R)-TSO
at the higher pH. The combined effect
on the hydrolysis of the TSO enantiomers is a shift in
enantiopreference from three-fold in favor of ( R ,R)-
TSO at pH 6.8 [15] to a three-fold preference for the
(S,S)-enantiomer at pH 8.0.
If K
S
>> [S], the hyperbolic concentration depen-
dence given by Eqn (2) simplifies to a linear relation-
ship (Eqn 3). In such cases, as observed in the
reactions with the SO enantiomers (Fig. 2B), the
enzyme is far from saturating substrate concentrations,
and direct determination of the individual parameters
is impossible.
k

obs
¼
k
2
K
S
½Sþðk
À2
þ k
3
Þð3Þ
In spite of the fact that values of k
2
could not be
determined, we propose that in the reactions with SO,
the main contributing factor to the observed enantio-
specificity arises from the different rates of alkyl-
enzyme formation. It can be deduced from Eqn (1)
that the value of K
m
is dependent on the relationship
between k
2
and K
S.
Hence, if K
S
reaches high relative
values (i.e. low ES stabilization), a relatively higher
degree of alkylenzyme accumulation is required

[k
2
>(k
–2
+ k
3
)] in order to account for the deter-
mined value of K
m
(S)-SO
of 220 lm , which is consider-
ably lower than the K
S
(S)-SO
(> 400 lm). With (R)-SO,
the pre-steady-state measurements show that the
increase in observed rates with substrate concentration
is comparably modest (Fig. 2B). This does not rule out
the possibility that the alkylation rate may be similar
to that of (S)-SO, but it would at the same time imply
that K
S
(R)-SO
is substantially higher for this enantio-
mer. The enzyme active site provides a hydrophobic
pocket with few opportunities for specific enthalpy-
contributing interactions with these substrates other
than possibly hydrogen bonding between the oxirane
oxygen and the active site Tyr pair [15,26]. It therefore
appears more likely that ES formation is an entropy-

driven desolvation process whereby increased
hydrophobicity in the substrate promotes ES forma-
tion, lowering the value of K
S
. This agrees with the
log P-values of the tested substrates (TSO, 3.4;
2-MeSO, 1.7; and SO, 1.6), whereby the most hydro-
phobic derivative (TSO) also displays the lowest K
S
(Table 2). Following this assumption, the K
S
values of
the different SO enantiomers are expected to be com-
parable, and the relatively higher value of K
m
(R)-SO
would mainly be an effect of a lower alkylation rate.
As the ordinates of k
obs
at [S] = 0 provide measures
of (k
–2
+ k
3
), the ratio of the slopes of the fitted lines
(k
2
⁄ K
S
) gives estimates of differences in alkylation

rates. The results suggest that StEH1 catalyzes the
formation of the alkylenzyme at an approximately
10-fold lower rate with the (R)-enantiomer (Table 2).
Hence, in the reactions with SO, the main contributing
factor to the observed enantiospecificity arises from
the different rates of alkylenzyme formation.
2-MeSO pre-steady-state measurements suggest
the presence of a hysteretic kinetic mechanism
In order to fit the progression curves of transient Trp
fluorescence quenching during the reaction with (S,S)-
2-MeSO, a double-exponential function had to be
applied (inset in Fig. 2C). This results in two observed
rates with individual amplitudes, suggesting the pres-
ence of two alkylenzyme species. Plotting of the sub-
strate dependence of k
obs1,2
reveals that the faster
observed rate increases linearly with increasing sub-
strate concentration (Fig. 2C), similar to the (S)-SO
case. The slower observed rate displays negative sub-
strate dependence; k
obs2
decreases with increasing sub-
strate concentration (Fig. 2D). The presence of two
observed rates may be explained by two different
models: (a) an additional on-pathway alkylenzyme
species that is formed with a distinct rate and displays
distinct amplitude – this case has been demonstrated
with (R)-SO in the reaction catalyzed by the epoxide
hydrolase from A. radiobacter AD1 [27]; or (b) a

mechanism involving a conformational change in the
free enzyme, referred to as hysteresis [30,31], with alky-
lenzyme species being formed within separated time
frames. These intermediate species may be structurally
Table 3. pH dependence of kinetic parameters of TSO hydrolysis.
Parameter
Substrates
(S,S)-TSO (R,R)-TSO
pH pH
6.8 8.0 6.8 8.0
k
2
(s
)1
)18±2
a
29 ± 5.8 260 ± 56
b
51 ± 6.9
k
3
(s
)1
) 3.2 ± 0.06
a
1.4 ± 0.36 24 ± 3
a
32 ± 0.37
K
S

(lM)11±6
a
14 ± 7.7 36 ± 22
b
42 ± 37
a
Data from [13].
b
Data from [15].
D. Lindberg et al. Enzyme-catalyzed styrene oxide hydrolysis
FEBS Journal 275 (2008) 6309–6320 ª 2008 The Authors Journal compilation ª 2008 FEBS 6313
identical but kinetically separated due to alternative
pathways for formation.
We have judged the linear model to be unlikely for
two reasons. First, the substrate dependence of k
obs2
displays a decrease with increasing concentration of
(S,S)-2-MeSO, a behavior that requires a rate-limiting
off-pathway conformational transition prior to sub-
strate binding [32]. An on-pathway mechanism would
result in either a substrate-dependent increase in
observed rates, or in no such dependence. Second, as
the mechanism described for the bacterial enzyme fol-
lows a linear reaction pathway, the value of k
obs2
must
be ‡ k
cat
. In the present case, the limiting value of
k

obs2
is approximately 20 s
)1
, which is considerably
lower than the value of k
cat
(52 s
)1
).
Our presently favored model therefore describes a
mechanism similar to previously reported cases with
other enzymes [33,34]. The model invokes two addi-
tional reaction steps: (a) between two conformational
states of the free enzyme, E … E¢; and (b) between the
corresponding Michaelis complexes, ES … E¢S
(Fig. 3A). With (S,S)-2-MeSO, it is assumed that only
ES is capable of productive formation of alkylenzyme
(k
6
> k
2
in Fig. 3D). The higher observed rate (k
obs1
)
reflects the Michaelis mechanism E ‰ ES ‰ E-alky-
l
1
fi E + diol
1
, displaying a substrate dependence

according to Eqn (3), whereas the lower rate (k
obs2
)
reflects the rate of alkylenzyme formation via a pathway
involving a rate-limiting conformational change from E
to E¢:Efi E¢ ‰ E¢S fi ES fi E-alkyl
1
fi E+
diol
1
. With this kinetic model, k
obs2
can be expressed by
Eqn (4) [33,34]. At low substrate concentrations, k
obs2
tends towards k
0
+ k
–0
, and at higher substrate concen-
trations towards k
5
+ k
–5
. Hence, if the transition
E¢S ‰ ES occurs at a lower rate than in the free enzyme
(E¢ ‰ E), i.e. substrate binding stabilizes one con-
former, the observed rate will decrease with increased
substrate concentration.
k

obs
¼
k
0
þðk
5
=K
S
Þ½S
1 þð½S=K
S
Þ
þ
k
À0
þðk
À5
=K
0
S
Þ½S
1 þð½S=K
0
S
Þ
ð4Þ
Kinetic cooperativity in hysteretic enzymes is shown
by characteristic deviations from Michaelis–Menten
kinetics: a burst or lag period in reaction progression
curves and sigmoidal substrate saturation curves [31].

Owing to restrictions in the HPLC-based assay, it is
not possible to resolve a lag period within the present
data, which in the 2-MeSO reactions would be on a
millisecond scale. Also, the data do not allow for a sig-
nificant distinction between cooperative or non-cooper-
ative substrate saturation. The steady-state parameters
given in Table 1 are therefore determined after fitting
the simplest (Michaelis–Menten) model.
In the reactions with (R,R)-2-MeSO, the observed
rates of transient fluorescence quenching follow single
exponential decays, analogous to the reactions with
SO or TSO. The substrate dependence of k
obs
, how-
ever, displays a decrease with increasing substrate
concentration, similar to the slow observed rate in the
(S,S)-2-MeSO reaction (Fig. 2D). Hydrolysis of either
2-MeSO enantiomer therefore appears to be dependent
D
A B C
Fig. 3. (A) Proposed kinetic mechanism of StEH1 hydrolysis of (S,S)-2-MeSO. The mechanism includes a fast, non-rate-limiting route to for-
mation of the alkylenzyme followed by a rate-limiting hydrolysis step. A slow conformational change [E fi E¢ in (D)] is also included,
describing the negative substrate dependence of k
obs
observed in the pre-steady-state measurements. (B) Discarded kinetic mechanism of
(R,R)-2-MeSO hydrolysis, lacking a reaction route to explain the formation of two different product enantiomers. (C) Proposed kinetic mecha-
nism of (R,R)-2-MeSO hydrolysis, including a rate-limiting conformational change of the free enzyme [E fi E¢ in (D)] and the formation of
two distinct alkylenzymes [E-alkyl
1
and E¢-alkyl

2
in (D)]. (D) Proposed kinetic mechanism of StEH1-catalyzed epoxide hydrolysis of 2-MeSO.
The model includes transitions between distinct forms of the free enzyme (E ‰ E¢) and Michaelis complexes (E¢S ‰ ES). In the reaction
with (S,S)-2-MeSO, it is assumed that only the ES form is productive in forming the alkylenzyme, whereas with (R,R)-2-MeSO, both of the
Michaelis complexes may form distinct alkylenzymes, resulting in a mixture of diol products. Simplifications of this model are also applicable
to the catalyzed hydrolyses of SO and TSO enantiomers.
Enzyme-catalyzed styrene oxide hydrolysis D. Lindberg et al.
6314 FEBS Journal 275 (2008) 6309–6320 ª 2008 The Authors Journal compilation ª 2008 FEBS
on similar kinetic mechanisms. The presence of only
one observed rate points towards a simpler model
(Fig. 3B), where k
obs
= k
0
+[(k
)0
K
S
) ⁄ (K
S
+ [S])]
[35]. This model cannot, however, explain the forma-
tion of the two different diol products (see discussion
below), which requires the formation of two distinct
alkylenzymes. Therefore, the model in Fig. 3C has
been applied for fitting to data obtained in the (R,R)-
2-MeSO reaction. The resulting reaction pathways are
E fi E¢ ‰ E¢S ‰ E¢-alkyl
2
and E fi E¢ ‰ E¢S ‰

ES ‰ E-alkyl
1
. The limiting value of k
obs
at high sub-
strate concentrations is approximately two-fold higher
than k
cat
(R,R)-2-MeSO
(k
5
+ k
)5
in Table 2), which sup-
ports the notion that the conformational change occur-
ring prior to substrate binding is rate-limiting in this
reaction.
The physical differences between E and E¢ or ES
and E¢S are at this stage highly speculative, and there
is no evidence for conformational changes in the free
enzyme from structural studies. There is the possibility,
however, that distinct enzyme forms are required for
productive interactions with substrates in different con-
figurations in the respective active sites. It is clear that
the structure of StEH1 provides ample space to allow
2-MeSO to bind in different modes within the active
site, resulting in different ES conformers. One fact sup-
porting this assumption is that both enantiomers of
TSO, a 1,2-bisphenyl substituted epoxide, are readily
accommodated in productive binding modes within the

active site.
Regiospecificity
The basis for the observed regiospecificity in epoxide
ring opening is difficult to rationalize even from ambi-
tious structure–reactivity studies [28,29]. The carbon
subjected to nucleophilic attack will be influenced by
distinct intrinsic reactivities of the oxirane carbons as
dictated by the substituents. Whereas bulky substitu-
ents generally direct the nucleophile to attack the less
hindered carbon, phenyl substituents may provide
favorable electrostatic properties that overcome unfa-
vorable steric features. In an enzyme-catalyzed reac-
tion, the spatial alignment of catalytic groups,
primarily the nucleophilic carboxylate and the electro-
philic Tyr pair, may further affect catalytic rates to a
degree that overcomes intrinsic reactivities.
Monterde et al. have established the regiospecificity
of the StEH1-catalyzed hydrolysis of SO [9]. Their
results show that the carbon that is subject to nucleo-
philic attack by the enzyme Asp is highly dependent
on the enantiomer. The (S)-enantiomer was 98%
attacked at the benzylic C-1, whereas (R)-SO reacted
at the less substituted C-2 (93%). The detailed reasons
for this clear shift are as yet unknown, but the result
implicates different substrate-binding modes in the
active site. Our kinetic studies indicate that alkylen-
zyme formation is more efficient with the (S)-enantio-
mer of SO by a factor of approximately 10, leading to
a higher value of k
cat

⁄ K
m
. The higher reaction rate
may be due to more efficient activation of the epoxide
oxygen through Lewis acid catalysis via more favor-
able positioning of this enantiomer within the active
site by the Tyr phenols and⁄ or more efficient stabili-
zation of the transition state. The fact that C-2 is pri-
marily attacked in the (R)-SO reaction may likewise be
attributed to a binding mode restricting attack on C-1
for steric reasons and ⁄ or a lowering of the catalytic
effect by the enzyme by placing the oxirane oxygen
out of reach of the Tyr phenols.
In this work, we have analyzed the regiospecificity
during hydrolysis of the 2-MeSO enantiomers by
13
C-NMR spectroscopy via incorporation of an
18
O-labeled hydroxyl group at the attacked carbon in
the presence of H
2
18
O. The results are partly in agree-
ment with those observed in the SO reactions: (S,S)-2-
MeSO is exclusively attacked at the benzylic carbon,
with undetectable levels of incorporation of
18
O at C-2
(Table 4, Fig. 4). The regiospecificity in the reaction
with (R,R)-2-MeSO produces a mixture of diol prod-

ucts (Table 4), with 65% (S,R)-diol being formed as a
result of attack at C-1, and the remaining 35% result-
ing from attack at C-2, yielding the (R,S)-diol. Hence,
the same degree of enantioconvergence present in the
reaction with SO is not observed with 2-MeSO under
the assay conditions used. Earlier studies with an epox-
ide hydrolase-containing extract from Aspergillus
terreus demonstrated product ratios after 2-MeSO
hydrolysis similar to those observed in this study [36].
Other reports on the regioselectivity in the ring open-
ing of 2-MeSO catalyzed by different epoxide hydro-
Table 4. Integration of signals in
13
C-NMR spectra of 1-phenylpro-
pane-1,2-diol. ND, not detectable.
Substrate Diol carbon d Integral Ratio Sum
(S,S)-2-MeSO CH
3
17.27 40.67 1.00 1.00
C-2-
18
O ND – – 1.02
C-2-
16
O 71.28 41.35 1.02
C-1-
18
O 77.48 20.59 0.51 1.02
C-1-
16

O 77.51 20.86 0.51
(R,R)-2-MeSO CH
3
17.30 17.66 1.00 1.00
C-2-
18
O 71.25 3.55 0.20 0.99
C-2-
16
O 71.28 13.89 0.79
C-1-
18
O 77.50 5.09 0.29 0.96
C-1-
16
O 77.52 11.82 0.67
D. Lindberg et al. Enzyme-catalyzed styrene oxide hydrolysis
FEBS Journal 275 (2008) 6309–6320 ª 2008 The Authors Journal compilation ª 2008 FEBS 6315
lases have shown a modest preference for attack at
C-1 [37–41].
The difference in kinetic mechanism shown by
StEH1 in the catalyzed hydrolysis of the 2-MeSO
enantiomers indicates that the difference in energy bar-
riers for formation of the distinct alkylenzymes is con-
siderably smaller when the (R,R)-enantiomer is the
substrate. With (S,S)-2-MeSO, the presence of one
single (R,S)-diol product suggests that, with this enan-
tiomer, only the ES (but not the E¢S) complex in
Fig. 3D may be transformed into alkylenzyme and,
subsequently hydrolyzed to product. Hence, the ener-

getic barrier for formation of the (S,R)-alkylenzyme
intermediate resulting from attack at C-2 is consider-
ably higher than that required to form the C-1-linked
alkylenzyme. In the reaction with (R,R)-2-MeSO, the
formation of two diol enantiomers demonstrates that
two different alkylenzymes resulting from nucleophilic
attack at either C-1 or C-2 of the substrate are stabi-
lized, assuming that conformational changes between
ES and E¢S do occur.
To conclude, the enantiospecificity determined from
steady-state kinetic measurements are caused by differ-
ent substrate-dependent mechanisms. Although we
have only been able to study the individual reaction
steps up to the formation of the alkylenzyme interme-
diate, it can be deduced from the values of k
cat
and k
2
that with (S)-SO and TSO, hydrolysis is rate-limiting.
With (R)-SO, k
cat
and k
2
could not be determined, due
to poor enzyme saturation within the solubility range
of the substrate. Therefore, from the lower-limit values
of these parameters (Tables 1 and 2), it cannot be
ruled out that the enzyme alkylation rate does contrib-
ute to k
cat

. We propose that with SO and TSO, differ-
ences in stabilization of the alkylenzyme are mainly
responsible for enantiospecificity. Decreases in K
m
resulting from rapid formation and ⁄ or slow decay of
the alkylenzyme appear to primarily determine selectiv-
ity in these cases, whereas tight substrate binding, as in
the TSO case, works against enantiodiscrimination.
With 2-MeSO, the substrate that is most efficiently
discriminated by StEH1, the more complex kinetic
mechanism requires other interpretations to under-
stand the basis for the enantioselectivity. With this
substrate, slow conformational changes limit the over-
all catalytic efficiencies. The slow, rate-limiting steps
for the (S,S)-enantiomer would escape detection if one
was to study the reaction during the steady state, as
only one diol enantiomer is produced. The amounts of
product formed via either the faster pathway or the
slower pathway are additive, and would be detected as
an apparent Michaelis–Menten reaction with kinetic
parameters described by the dominating faster route. It
follows that, for the overall reaction, hydrolysis of the
(R,S)-alkylenzyme is rate-limiting with (S,S)-2-MeSO.
With (R,R)-2-MeSO, slow conformational changes of
the substrate-free enzyme appear to be directly rate-
limiting, and therefore serve to mask subsequent cata-
lytic steps. Direct analysis of the rates for product
enantiomer formation is required for a complete
understanding of the reaction scheme in this StEH1-
catalyzed transformation.

Experimental procedures
Materials
(R)-SO and (S)-SO (97% and 98% enantiomerically pure,
respectively) and (S,S)-MeSO and (R,R)-2-MeSO were pur-
chased from Aldrich. (S,S)-TSO and (R,R)-TSO were gifts
from P. I. Arvidsson (Department of Biochemistry and
Organic Chemistry, Uppsala University). The purities of
2-MeSO and TSO enantiomers were > 99% as judged by
chiral HPLC analyses.
Protein purification
StEH1 was produced in Escherichia coli XL1-Blue cells and
purified according to a previously described protocol [10].
The protein concentrations of collected fractions were deter-
mined from the absorbance at 280 nm using a molar absor-
bance coefficient of 59 030 m
)1
Æcm
)1
, calculated from the
Fig. 4.
13
C-NMR signals of C–O in 1-phenylpropane-1,2-diol
(125.7 MHz, CDCl
3
solution) obtained by enzymatic hydrolysis of
2-MeSO enantiomers. (A) Hydrolysis products of (S,S)-2-MeSO
with H
2
18
O ⁄ H

2
16
O (1 : 1). (B) Hydrolysis products of (R,R)-2-MeSO
with H
2
18
O ⁄ H
2
16
O (1 : 1). See Table 4 for details of relative
amounts of isotopologs.
Enzyme-catalyzed styrene oxide hydrolysis D. Lindberg et al.
6316 FEBS Journal 275 (2008) 6309–6320 ª 2008 The Authors Journal compilation ª 2008 FEBS
amino acid composition. Once purified, the proteins retained
full activity upon storage at 4 °C over the time period of
analysis. Data for determining steady-state kinetic parame-
ters for catalysis of TSO hydrolysis were adapted from [10].
Steady-state kinetics (data collection)
The steady-state kinetic parameters of both enantiomers of
SO and 2-MeSO were determined by RP-HPLC. Substrates
and the internal standard benzyl alcohol were dissolved in
acetonitrile and added to the reaction vessel containing
0.1 m sodium phosphate (pH 8.0), to a final concentration
of acetonitrile of 1.5% (v⁄ v). The assay pH was chosen to
increase the amount of build-up of alkylenzyme intermedi-
ate, thereby allowing for determinations of the microscopic
rate constants. Alkylenzyme accumulation, at least in the
catalyzed hydrolysis of (S,S)-TSO, is linked to a pH-depen-
dent decrease in hydrolysis rate and a parallel increase in
alkylation rate [13]. Catalyzed reactions were started by the

addition of enzyme solution to the reaction vessels. All
reactions were carried out at 30 °C, and they were termi-
nated at different time points by removing an aliquot of the
reaction mixture and injecting it into methanol to reach a
final concentration of 37% (v ⁄ v) methanol. Substrate
concentrations used were 0.1–0.8, 0.1–0.8, 0.05–0.75 and
0.1–1.2 mm for (S)-SO, (R)-SO, (S,S)-2-MeSO and (R,R)-2-
MeSO, respectively, and the enzyme concentrations used
were 20, 100, 50 and 200 nm for (S)-SO, (R)-SO, (S,S)-
2-MeSO and (R,R)-2-MeSO, respectively. Using a
Water 717 autosampler (SO) or manual injection (2-MeSO)
via a Rheodyne 7125 injector, the methanol ⁄ reaction solu-
tion mixture was injected into the HPLC system. Reaction
constituents were separated on a reverse-phase HighCrom
Kromasil C-18 column (25 cm · 3.2 mm diameter), using a
Waters 515 pump (flow rate 0.6 mLÆmin
)1
), and the peaks
were detected spectrophotometrically with a Waters 484
detector (k = 220 nm). The liquid phase consisted of a
37 : 63 (SO) and 34 : 66 (2-MeSO) mixture of methanol
and sodium phosphate buffer (0.1 m, pH 3.0). For the SO
enantiomers, the retention times for the reaction constitu-
ents and the internal standard were 4.5, 7.1 and 18.4 min
(1-phenylethane-1,2-diol, benzyl alcohol, and SO). For the
2-MeSO enantiomers, retention times of 6.9, 8.9 and
45.1 min (1-phenylpropane-1,2-diol, benzyl alcohol, and
2-MeSO) were observed.
Pre-steady-state kinetics (data collection)
Formation of the alkylenzyme intermediate can be followed

by the decrease in the intrinsic Trp fluorescence of the
enzyme [10]. Transient-state kinetics were determined
in multiple-turnover experiments under pseudo-first-order
conditions. Experiments were performed on an Applied
Photophysics SX.20MV sequential stopped-flow spectro-
photometer at 30 °C, using an excitation wavelength of
290 nm while collecting the fluorescent light through a
320 nm cut-off filter. With all substrates except for (S,S)-2-
MeSO, the apparent rate values, k
obs
, were determined by
fitting a single exponential function with floating endpoint,
f = A exp()k
obs
t)+C, to the progression curve; averages
of five to eight traces were used in all cases. With (S,S)-2-
MeSO, a double exponential function with floating end-
point, f = A
1
exp()k
obs1
t)+A
2
exp()k
obs2
t)+C, was
used. The validity of the higher-order equation was verified
by F-tests. In both cases, f denotes the averaged progres-
sion curve and k
obs

the averaged observed rates. The ampli-
tudes of the fluorescent changes were obtained by
subtracting the initial fluorescence value (A) and the float-
ing endpoint (C). Substrate concentrations were 6.3–800,
100–1100, 25–1000, 85–1000, 0.8–30 and 12–130 lm [(S)-
SO, (R)-SO, (S,S)-2-MeSO, (R,R)-2-MeSO, (S,S)-TSO, and
(R,R)-TSO, respectively], using enzyme concentrations of 1,
10, 2, 10, 0.1 and 1.35 lm [(S)-SO, (R)-SO, (S,S)-2-MeSO,
(R,R)-2-MeSO, (S,S)-TSO, and (R,R)-TSO, respectively].
Substrate stock solutions were prepared in acetonitrile, after
which they, and the enzyme stock solutions, were further
diluted in a sodium phosphate buffer (0.1 m, pH 8.0),
resulting in a 1–2% (v ⁄ v) final concentration of acetonitrile.
Kinetic data analysis
The steady-state parameters k
cat
, K
m
and k
cat
⁄ K
m
were
determined after curve-fitting the Michaelis–Menten equa-
tion to the experimental data with mmfit or rffit in simfit
(fit.man.ac.uk/).
The steady-state rate law (Eqn 1) was derived by the
method of Waley [42], assuming the kinetic mechanism
described in Fig. 1, where K
S

is k
)1
⁄ k
1
, the dissociation
constant of the ES complex, and k
1
is the association rate
of enzyme and substrate.
The rate constants for alkylenzyme formation, k
2
,
together with the dissociation constant for ES complexes,
K
S
, were obtained after fitting Eqn (2) to k
obs
using qnfit
in simfit. The sums of rate constants for alkylenzyme
decomposition, k
)2
+ k
3
, were obtained from the ordinates
of fitted curves. Individual values were subsequently
extracted from the expression for k
cat
in the numerator of
Eqn (1). ES complexes were assumed to be at equilibrium
under the conditions of the measurements. For the slower

observed rate with (S,S)-2-MeSO and k
obs
(R,R)-2-MeSO
, data
were fitted to Eqn (4) [33,34] using qnfit.
18
O-labeling of 2-MeSO hydrolysis products
Initially, control experiments were performed to measure
the catalyzed hydrolysis of (R,R)-2-MeSO in the presence
of 0.1 m ammonium bicarbonate buffer (pH 8.0), replacing
the otherwise used 0.1 m sodium phosphate buffer of
the same pH, to establish that reaction rates were compa-
rable. For the
18
O-labeling experiment, undiluted 2-MeSO
D. Lindberg et al. Enzyme-catalyzed styrene oxide hydrolysis
FEBS Journal 275 (2008) 6309–6320 ª 2008 The Authors Journal compilation ª 2008 FEBS 6317
enantiomer was added, 1 lL at a time, to a 0.1 m ammo-
nium bicarbonate solution composed of a 1 : 1 (v ⁄ v)
H
2
16
O ⁄ H
2
18
O mixture (pH 8.0), 3% (v ⁄ v) acetonitrile, and
1 lm (in the reaction with (S,S)-2-MeSO) or 2.5 lm (in the
reaction with (R,R)-2-MeSO) StEH1. The final volume of
concentrated epoxide added to the reaction vessel was
7 lL, corresponding to a final concentration of 25 mm diol.

The reaction vessel was agitated at 175 r.p.m. and 30 °C,
and the progress of the reaction was followed with RP-
HPLC until completion. Aliquots of the reaction solution
were manually injected using a Rheodyne 7725 injector into
a liquid phase consisting of a 50 : 50 mixture of metha-
nol ⁄ 0.1 m sodium phosphate (pH 3.0). Product mixture
components were separated on a Supelco Ascentis C-18 col-
umn (5 lm,25cm· 4.6 mm diameter) coupled to a Shima-
dzu Prominence LC-20AD pump. Peaks were detected at
220 nm using a Prominence SPD-M20A diode array detec-
tor. After reaction completion, the solvent was evaporated
prior to NMR analysis.
NMR analysis of
18
O-labeled hydrolysis products
NMR spectra were recorded at 500 MHz (
1
H) and
125.7 MHz (
13
C), respectively, for CDCl
3
solutions on a
Varian Inova spectrometer. Chemical shifts were indirectly
referenced to TMS via the solvent signal (
1
H, 7.26,
13
C,
77.0). Signal assignments were made using gradient-

enhanced COSY [43], HSQC [44], and HMBC [45] experi-
ments. For diols labeled with
18
O,
13
C-NMR spectra were
recorded over a spectral range of 30 kHz, using an acquisi-
tion time of 1.3 s, a relaxation delay of 3 s, a pulse flip
angle of 26°, and 9188 transients (total experiment dura-
tion = 11 h), for a final signal ⁄ noise ratio of 50–60. Data
were processed with zero filling to 262 144 points. Relative
peak intensities were determined by integration via decon-
volution analysis (Lorentzian lineshape) of the spectra,
using the manufacturer-supplied software (vnmr 6.1C).
Acknowledgements
The authors thank M. Engman for establishing the
enantiomeric purity of 2-MeSO, A. Gurell for con-
structive criticism during manuscript preparation, and
J. Gurell for the development of a matlab application.
Financial support from the Swedish Research Council,
the Ingegerd Bergh Foundation and the Carl Trygger
Foundation is also gratefully acknowledged. D. Lind-
berg is a Lawski Foundation stipendiate.
References
1 Arand M, Cronin A, Oesch F, Mowbray SL & Jones
TA (2003) The telltale structures of epoxide hydrolases.
Drug Metab Rev 35, 365–383.
2 Moriseau C & Hammock BD (2005) Epoxide hydrolas-
es: mechanisms, inhibitor designs and biological roles.
Annu Rev Pharmacol Toxicol 45, 311–333.

3 Fauth M, Schweizer P, Buchala A, Marksta
¨
dter C,
Riederer M, Kato T & Kauss H (1998) Cutin monomers
and surface wax constituents elicit H
2
O
2
in conditioned
cucumber hypocotyl segments and enhance the activity
of other H
2
O
2
elicitors. Plant Physiol 117, 1373–1380.
4 Kolattukudy PE (1981) Structure, biosynthesis, and
biodegradation of cutin and suberin. Annu Rev Plant
Physiol 32, 539–567.
5 Steinreiber A & Faber K (2001) Microbial epoxide
hydrolases for preparative biotransformations. Curr
Opin Biotechnol 12, 552–558.
6 Archelas A & Furtoss R (2001) Synthetic applications
of epoxide hydrolases. Curr Opin Chem Biol 5, 112–119.
7 Melnick RL (2002) Carcinogenicity and mechanistic
insights on the behavior of epoxides and epoxide-form-
ing chemicals. Ann NY Acad Sci 982, 177–189.
8 Manoj KM, Archelas A, Baratti J & Furstoss R
(2001) Microbiological transformations. Part 45: a
green chemistry preparative scale synthesis of enantio-
pure building blocks of Eliprodil: elaboration of a high

substrate concentration epoxide hydrolase-catalyzed
hydrolytic kinetic resolution process. Tetrahedron 57 ,
695–701.
9 Monterde MI, Lombard M, Archelas A, Cronin A,
Arand M & Furstoss R (2004) Enzymatic transforma-
tions. Part 58: enantioconvergent biohydrolysis
of styrene oxide derivatives catalysed by the Solanum
tuberosum epoxide hydrolase. Tetrahedron Asym 15,
2801–2805.
10 Elfstro
¨
m LT & Widersten M (2005) Catalysis of potato
epoxide hydrolase, StEH1. Biochem J 390, 633–640.
11 Karboune S, Archelas A, Furstoss R & Baratti JC
(2005) Immobilization of the Solanum tuberosum epox-
ide hydrolase and its application in an enantioconver-
gent process. Biocatal Biotransform 6, 397–405.
12 Cao L, Lee J, Chen W & Wood TK (2006) Enantiocon-
vergent production of (R)-1-phenyl-1,2-ethanediol from
styrene oxide by combining the Solanum tuberosum and
an evolved Agrobacterium radiobacter AD1 epoxide
hydrolases. Biotechnol Bioeng 94, 522–529.
13 Elfstro
¨
m LT & Widersten M (2006) Implications for an
ionized alkyl-enzyme intermediate during StEH1-cata-
lyzed trans-stilbene oxide hydrolysis. Biochemistry 45,
205–212.
14 Mateo C, Fernandez-Lafuente R, Archelas A, Guisan
JM & Furstoss R (2007) Preparation of a very stable

immobilized Solanum tuberosum epoxide hydrolase.
Tetrahedron Asym 18, 1233–1238.
15 Thomaeus A, Carlsson J, A
˚
qvist J & Widersten M
(2007) Active site of epoxide hydrolases revisited: a
non-canonical residue in potato StEH1 promotes both
Enzyme-catalyzed styrene oxide hydrolysis D. Lindberg et al.
6318 FEBS Journal 275 (2008) 6309–6320 ª 2008 The Authors Journal compilation ª 2008 FEBS
formation and breakdown of the alkylenzyme inter-
mediate. Biochemistry 46, 2466–2479.
16 Lacourciere GM & Armstrong RN (1993) The catalytic
mechanism of microsomal epoxide hydrolase involves
an ester intermediate. J Am Chem Soc 115, 10466–
10467.
17 Borhan B, Jones AD, Pinot F, Grant DF, Kurth MJ &
Hammock BD (1995) Mechanism of soluble epoxide
hydrolase. Formation of an alpha-hydroxy ester–
enzyme intermediate through Asp-333. J Biol Chem
270, 26923–26930.
18 Yamada T, Morisseau C, Maxwell JE, Argiriadi MA,
Christianson DW & Hammock BD (2000) Biochemical
evidence for the involvement of tyrosine in epoxide acti-
vation during the catalytic cycle of epoxide hydrolase.
J Biol Chem 275, 23082–23088.
19 Rink R, Kingma J, Lutje Spelberg JH & Janssen DB
(2000) Tyrosine residues serve as proton donor in the
catalytic mechanism of epoxide hydrolase from Agro-
bacterium radiobacter. Biochemistry 39, 5600–5613.
20 Tzeng H-F, Laughlin LT, Lin S & Armstrong RN

(1996) The catalytic mechanism of microsomal epoxide
hydrolase involves reversible formation and rate-limit-
ing hydrolysis of the alkyl-enzyme intermediate. JAm
Chem Soc 118, 9436–9437.
21 Tzeng H-F, Laughlin LT & Armstrong RN (1998)
Semifunctional site-specific mutants affecting the hydro-
lytic half-reaction of microsomal epoxide hydrolase.
Biochemistry 37, 2905–2911.
22 Armstrong RN & Cassidy CS (2000) New structural
and chemical insight into the catalytic mechanism of
epoxide hydrolases. Drug Metab Rev 32, 327–338.
23 Thomaeus A, Naworyta A, Mowbray SL & Widersten
M (2008) Removal of distal protein–water hydrogen
bonds in a plant epoxide hydrolase increases catalytic
turnover but decreases thermostability. Protein Sci 17,
1275–1284.
24 Nardini M, Ridder IS, Rozeboom HJ, Kalk KH, Rink
R, Janssen DB & Dijkstra BW (1999) The x-ray struc-
ture of epoxide hydrolase from Agrobacterium radiob-
acter AD1. An enzyme to detoxify harmful epoxides.
J Biol Chem 274, 14579–14586.
25 Argiriadi MA, Morisseau C, Hammock BD & Chris-
tianson DW (1999) Detoxification of environmental
mutagens and carcinogens: structure, mechanism, and
evolution of liver epoxide hydrolase. Proc Natl Acad
Sci USA 96, 10637–10642.
26 Mowbray SL, Elfstro
¨
m LT, Ahlgren KM, Andersson
CE & Widersten M (2006) X-ray structure of potato

epoxide hydrolase sheds light on substrate specificity in
plant enzymes. Protein Sci 15, 1628–1637.
27 Rink R & Janssen DB (1998) Kinetic mechanism of the
enantioselective conversion of styrene oxide by epoxide
hydrolase from Agrobacterium radiobacter AD1.
Biochemistry 37, 18119–18127.
28 Parker RE & Isaacs NS (1959) Mechanisms of epoxide
reactions. Chem Rev 59, 737–799.
29 Whalen DL (2005) Mechanisms of hydrolysis and rear-
rangements of epoxides. Adv Phys Org Chem 40, 247–
298.
30 Frieden C (1979) Slow transitions and hysteretic behav-
ior in enzymes. Annu Rev Biochem 48, 471–489.
31 Neet KE & Ainslie RG Jr (1980) Hysteretic enzymes.
Methods Enzymol 64, 192–226.
32 Fersht AR & Raquena Y (1971) Equilibrium and rate
constants for the interconversion of two conformations
of a-chymotrypsin. The existence of a catalytically inac-
tive conformation at neutral pH. J Mol Biol 60, 270–290.
33 Masson P, Schopfer LM, Froment M-T, Debouzy J-C,
Nachon F, Gillon E, Lockridge O, Hrabovska A &
Goldstein BN (2005) Hysteresis of butyrylcholinesterase
in the approach to steady-state kinetics. Chem Biol
Interact 157-158, 143–152.
34 Kim YB, Kalinowski SS & Marcinkeviciene J (2007) A
pre-steady state analysis of ligand binding to human
glucokinase: evidence for a preexisting equilibrium.
Biochemistry 46, 1423–1431.
35 Fersht A (1999) Structure and Mechanism in Protein
Science. WH Freeman, New York, NY.

36 Moussou P, Archelas A, Baratti J & Furstoss R (1998)
Microbiological transformations. Part 39: determination
of the regioselectivity occurring during oxirane ring
opening by epoxide hydrolases: a theoretical analysis
and a new method for its determination. Tetrahedron
Asym 9, 1539–1547.
37 Bellucci G, Chiappe C, Cordoni A & Marioni F (1994)
Different enantioselectivity and regioselectivity of the
cytosolic and microsomal epoxide hydrolase catalyzed
hydrolysis of simple phenyl substituted epoxides.
Tetrahedron Lett 35, 4219–4222.
38 Pedragosa-Moreau S, Archelas A & Furstoss R (1996)
Microbiological transformations 32: use of epoxide
hydrolase mediated biohydrolysis as a way to enantio-
pure epoxides and vicinal diols: application to substi-
tuted styrene oxide derivatives. Tetrahedron 52,
4593–4606.
39 Williamson KC, Morisseau C, Maxwell JE & Ham-
mock BD (2000) Regio- and enantioselective hydrolysis
of phenyloxiranes catalyzed by soluble epoxide hydro-
lase. Tetrahedron Asym 11, 4451–4462.
40 Lau EY, Newby ZE & Bruice TC (2001) A theoretical
examination of the acid-catalyzed and noncatalyzed
ring-opening of an oxirane by nucleophilic addition of
acetate. Implications to epoxide hydrolases. J Am Chem
Soc 123, 3350–3357.
41 Hopmann K & Himo F (2006) Theoretical study of the
full reaction mechanism of human soluble epoxide
hydrolase. Chem Eur J 12, 6898–6909.
42 Waley SG (1992) An easy method for deriving steady-

state rate equations. Biochem J 286, 357–359.
D. Lindberg et al. Enzyme-catalyzed styrene oxide hydrolysis
FEBS Journal 275 (2008) 6309–6320 ª 2008 The Authors Journal compilation ª 2008 FEBS 6319
43 Horne TJ & Morris GA (1997) P-Type gradient-
enhanced COSY experiments show lower t
1
noise than
N-type. Magn Reson Chem 35, 680–686.
44 Davis AL, Keeler J, Laue ED & Moskau D (1992)
Experiments for recording pure-absorption hetero-
nuclear correlation spectra using pulsed field gradients.
J Magn Reson 98, 207–216.
45 Hurd RE & John BK (1991) Gradient-enhanced pro-
ton-detected heteronuclear multiple-quantum coherence
spectroscopy. J Magn Reson 91, 648–653.
Enzyme-catalyzed styrene oxide hydrolysis D. Lindberg et al.
6320 FEBS Journal 275 (2008) 6309–6320 ª 2008 The Authors Journal compilation ª 2008 FEBS