The effect of a unique halide-stabilizing residue on the catalytic properties of haloalkane dehalogenase DatA from Agrobacterium tumefaciens C58

Article abstract of DOI:10.1111/febs.12238

Haloalkane dehalogenases catalyze the hydrolysis of carbon-halogen bonds in various chlorinated, brominated and iodinated compounds. These enzymes have a conserved pair of halide-stabilizing residues that are important in substrate binding and stabilization of the transition state and the halide ion product via hydrogen bonding. In all previously known haloalkane dehalogenases, these residues are either a pair of tryptophans or a tryptophan-asparagine pair. The newly-isolated haloalkane dehalogenase DatA from Agrobacterium tumefaciens C58 (EC 3.8.1.5) possesses a unique halide-stabilizing tyrosine residue, Y109, in place of the conventional tryptophan. A variant of DatA with the Y109W mutation was created and the effects of this mutation on the structure and catalytic properties of the enzyme were studied using spectroscopy and pre-steady-state kinetic experiments. Quantum mechanical and molecular dynamics calculations were used to obtain a detailed analysis of the hydrogen-bonding patterns within the active sites of the wild-type and the mutant, as well as of the stabilization of the ligands as the reaction proceeds. Fluorescence quenching experiments suggested that replacing the tyrosine with tryptophan improves halide binding by 3.7-fold, presumably as a result of the introduction of an additional hydrogen bond. Kinetic analysis revealed that the mutation affected the substrate specificity of the enzyme and reduced its K_0.5 for selected halogenated substrates by a factor of 2-4, without impacting the rate-determining hydrolytic step. We conclude that DatA is the first natural haloalkane dehalogenase that stabilizes its substrate in the active site using only a single hydrogen bond, which is a new paradigm in catalysis by this enzyme family. The newly-isolated haloalkane dehalogenase DatA from Agrobacterium tumefaciens C58 possesses a unique halide-stabilising tyrosine residue, Y109, in place of the conventional tryptophan. This is the first natural haloalkane dehalogenase that stabilises its substrate in the active site using only a single hydrogen bond, which is a new paradigm in catalysis by this enzyme family.

Full text of DOI:10.1111/febs.12238

The effect of a unique halide-stabilizing residue on the
catalytic properties of haloalkane dehalogenase DatA
from Agrobacterium tumefaciens C58
Khomaini Hasan1,*, Artur Gora1,*, Jan Brezovsky1, Radka Chaloupkova1, Hana Moskalikova1,
Andrea Fortova1, Yuji Nagata2, Jiri Damborsky1 and Zbynek Prokop1
1 Loschmidt Laboratories, Department of Experimental Biology and Research Centre for Toxic Compounds in the Environment,
Faculty of Science, Masaryk University, Brno, Czech Republic
2 Department of Environmental Life Sciences, Graduate School of Life Sciences, Tohoku University, Sendai, Japan
Keywords
Haloalkane dehalogenases catalyze the hydrolysis of carbon–halogen bonds
in various chlorinated, brominated and iodinated compounds. These
catalytic mechanism; enzyme kinetics;
halide-stabilizing residue; haloalkane
enzymes have a conserved pair of halide-stabilizing residues that are impor-
tant in substrate binding and stabilization of the transition state and the
halide ion product via hydrogen bonding. In all previously known haloal-
dehalogenase; substrate specificity
Correspondence
kane dehalogenases, these residues are either a pair of tryptophans or a
tryptophan–asparagine pair. The newly-isolated haloalkane dehalogenase
DatA from Agrobacterium tumefaciens C58 (EC 3.8.1.5) possesses a unique
halide-stabilizing tyrosine residue, Y109, in place of the conventional tryp-
tophan. A variant of DatA with the Y109W mutation was created and the
effects of this mutation on the structure and catalytic properties of the
enzyme were studied using spectroscopy and pre-steady-state kinetic experi-
ments. Quantum mechanical and molecular dynamics calculations were
used to obtain a detailed analysis of the hydrogen-bonding patterns within
the active sites of the wild-type and the mutant, as well as of the stabiliza-
tion of the ligands as the reaction proceeds. Fluorescence quenching experi-
ments suggested that replacing the tyrosine with tryptophan improves
halide binding by 3.7-fold, presumably as a result of the introduction of an
additional hydrogen bond. Kinetic analysis revealed that the mutation
affected the substrate specificity of the enzyme and reduced its K0.5 for
selected halogenated substrates by a factor of 2–4, without impacting the
rate-determining hydrolytic step. We conclude that DatA is the first natural
haloalkane dehalogenase that stabilizes its substrate in the active site using
only a single hydrogen bond, which is a new paradigm in catalysis by this
enzyme family.
Z. Prokop, Loschmidt Laboratories,
Department of Experimental Biology and
Research Centre for Toxic Compounds in
the Environment, Faculty of Science,
Masaryk University, Kamenice 5/A13,
62500 Brno, Czech Republic
Fax: +420 5 4949 6302
Tel: +420 5 4949 6667
E-mail: zbynek@chemi.muni.cz
*These authors contributed equally to this
work.
(Received 14 January 2013, revised 5
March 2013, accepted 6 March 2013)
doi:10.1111/febs.12238
Introduction
Haloalkane dehalogenases (HLDs) are bacterial
enzymes that catalyze the hydrolytic dehalogenation of
various halogenated compounds [1–3]. As a result of
their broad specificity, HLDs have potential applica-
tions in biocatalysis, bioremediation and biosensing
[4–8]. Structurally, the HLDs belong to the a/b hydro-
lase superfamily. Their active site is deeply buried in
the protein core, between the highly conserved a/b
Abbreviations
HLD, haloalkane dehalogenase; MD, molecular dynamics; NAC, near attack configuration; PCA, principle component analysis; QM, quantum
mechanics; SSG, substrate specificity group.
FEBS Journal 280 (2013) 3149–3159 ª 2013 The Authors Journal compilation ª 2013 FEBS
3149

Unique halide-stabilizing residue of DatA
K. Hasan et al.
hydrolase fold main domain and a helical cap domain
[4,9–19]. The key active site residues consist of a cata-
lytic pentad: a nucleophile, a base, a catalytic acid and
two halide-stabilizing residues. The dehalogenation
reaction is initiated by the nucleophilic attack of an
aspartic acid on the sp3-hybridized carbon atom
attached to the halogen. This breaks the carbon–halo-
gen bond, forming a covalently bound alkyl–enzyme
intermediate and a halide ion. The alkyl–enzyme inter-
mediate is then hydrolyzed by a water molecule, which
is activated by the catalytic base. The final reaction
products are a halide ion, a proton and an alcohol
[17–22].
Previous studies have demonstrated that the two
halide-stabilizing residues in the active sites of HLDs
are essential for their catalytic activity [23–26]. These
residues hold the substrate in the configuration
required for the SN2 reaction and then stabilize the
reaction’s transition state. They also stabilize the
halide ion released during the cleavage of the carbon–
halogen bond. These effects are exerted via a pair of
hydrogen bonds, one donated by each of the two resi-
dues (Fig. 1A). The properties of the halide-stabilizing
residues contribute significantly to the enzyme’s stabil-
ization efficiency [23]. The first halide-stabilizing resi-
due is a highly conserved tryptophan located next to
the nucleophile. Its other role is to maintain the nucle-
ophile in the correct orientation for the nucleophilic
attack [24]. The second halide-stabilizing residue is
typically tryptophan or asparagine. Recently, a HLD
named DatA was isolated from Agrobacterium tum-
efaciens C58 (EC 3.8.1.5) (newly Agrobacterium fabrum
C58). Remarkably, this enzyme possesses a unique
tyrosine residue in the position occupied by the (other-
wise fully conserved) first halide-stabilizing tryptophan
residue in other HLDs (Fig. 1B) [27,28].
In the present study, we report the results of an
experimental and computational investigation into the
impact of this unique halide-stabilizing residue on the
catalytic properties of DatA. The mutant DatA01,
which carries the substitution Y109W, was created by
site-directed mutagenesis and then biochemically char-
acterized and compared with the wild-type enzyme.
Spectroscopic analyses, as well as steady-state and tran-
sient kinetic experiments, provided insights into the
effects of the halide-stabilizing residues on halide-bind-
ing, substrate specificity and reaction kinetics. Quantum
mechanics (QM) calculations and molecular dynamics
(MD) simulations were performed to complement the
experimental studies by providing information on how
the active site residues interact with the substrate and
the products formed during the reaction.
Results
Structural and thermostability analysis
The mutant protein Y109W was constructed by site-
directed mutagenesis, expressed in Escherichia coli and
purified to homogeneity using the two-step purification
protocol [28]. The correct folding of the purified mutant
DatA01 was verified by CD spectroscopy. The far-UV
CD spectrum of the native enzyme exhibited two nega-
tive features at 208 and 222 nm, which is typical of
a-helical content (Fig. S1). The spectrum of DatA01
was similar to that of the wild-type enzyme, indicating
that the replacement of the tyrosine with the otherwise-
conserved tryptophan had no effect on the overall
secondary structure of the enzyme. The melting temper-
atures determined for DatA (48.3 °C) and DatA01
(51.8 °C) differed by 3.5 °C. The increased thermosta-
bility of DatA01 could be attributed to the replacement
of tyrosine with the larger tryptophan, which strength-
ened the hydrophobic packing of the structure and
improved the van der Waals contacts between the
amino acid residues in the protein core [29].
A
B
Substrate specificity analysis
Substrate specificity measured with 30 different haloge-
nated substrates revealed different patterns for DatA
and DatA01 (Fig. 2). The specific activity of DatA01
towards most of the tested substrates was lower than
that of DatA. Interestingly, DatA01 showed increased
specific activities towards several substrates: bis(2-chlo-
roethyl)ether, 3-chloro-2-methylpropene, 1,2-dibromo-
ethane, 1,2-dibromopropane, 1,2,3-tribromopropane,
1-bromo-2-chloroethane, 1,2-dibromo-3-chloropropane
or 4-bromobutyronitrile.
Fig. 1. The enzyme–substrate complex in a ‘conventional’ HLDs
(A) and the DatA wild-type (B). The halide-stabilizing residues found
in ‘conventional’ HLDs are either a tryptophan–tryptophan or a
tryptophan–asparagine pair. The halide-stabilizing residues of DatA
are a tyrosine–asparagine pair. Hydrogen bonds are indicated by
dashed lines and hydrophobic contacts are indicated by an arc with
spokes radiating towards the ligand.
3150
FEBS Journal 280 (2013) 3149–3159 ª 2013 The Authors Journal compilation ª 2013 FEBS

K. Hasan et al.
Unique halide-stabilizing residue of DatA
Fig. 2. Comparison of the substrate specificity profiles of DatA wild-type (blue) and DatA01 (red) for 30 different substrates. The insert
shows the clustering of DatA, DatA01 and other characterized HLDs based on their activity towards the tested substrates. The t1/t2 score
plot obtained by principal component analysis explains 31% of the variance in the original dataset.
Changes in the substrate specificity were explored by
principal component analysis (PCA). Although
DatA01 and DatA were in the same substrate specific-
ity group (SSG)-IV, the substitution of tyrosine for
tryptophan significantly modulated the enzyme’s sub-
strate specificity, as indicated by the very different
positions of the two enzymes on the scores plot
(Fig. 2).
determined based on fluorescence measurements upon
excitation at 280 nm. The equilibrium dissociation
constant for the mutant DatA01 (Kd = 34 ꢀ 4 mM)
was 3.7-fold lower than that for the wild-type DatA
(Kd = 127 ꢀ 32 mM). The replacement of the tyrosine
with tryptophan thus stabilized the enzyme–halide
complex.
Steady-state kinetics
Halide-binding analysis
The kinetic constants determined for the wild-type
DatA and the mutant DatA01 under steady-state con-
ditions are summarized in Table 1. Both the wild-type
and the mutant yielded hyperbolic kinetic profiles for
their reactions with 1,3-dibromopropane and
1-bromohexane, suggesting classical Michaelis–Menten
kinetics (data not shown). The sigmoidal profiles
obtained in experiments using 1,3-diiodopropane,
1-chlorohexane and 1-bromobutane indicate that these
substrates are hydrolyzed via a cooperative Hill mech-
anism. Regardless of the kinetic mechanism in opera-
tion or the substrate used, the K0.5 values for DatA01
were lower than those for the wild-type DatA. In par-
ticular, the K0.5 values for brominated substrates with
DatA01 were 2.5–4.5-fold lower than those obtained
with DatA, indicating that the mutant has a higher
affinity for these substrates (Table 1). These data are
The steady-state fluorescence of DatA and DatA01
was measured after mixing with different bromide
concentrations (0–1500 mM). The intensity of fluore-
scence was recorded upon excitation at 280 nm for
both tryptophan and tyrosine and selective excitation
at 295 nm for tryptophan exclusively (Fig. 3). The
fluorescence profile of the DatA wild-type upon exci-
tation at 295 nm indicated only nonspecific fluore-
scence quenching as the bromide concentration
increased, reflecting the effect of the anion as a gen-
eral fluorescence quencher. By contrast, strong addi-
tional quenching was observed for the variant
DatA01, suggesting the existence of a specific interac-
tion between the bromide ion and the introduced
tryptophan residue. The equilibrium constants for
bromide binding (Kd) in DatA and DatA01 were
FEBS Journal 280 (2013) 3149–3159 ª 2013 The Authors Journal compilation ª 2013 FEBS
3151

Unique halide-stabilizing residue of DatA
K. Hasan et al.
tent with the reduced specific activity observed as a
result of the mutation (Fig. 2).
Pre-steady-state kinetics
Transient kinetic analysis was used to obtain more
information on the kinetics of the DatA and DatA01
catalytic cycles and their limiting factors. The pre-
steady-state burst kinetics for 1-bromobutane hydroly-
sis were analyzed by the rapid quench flow technique.
Clear pre-steady-state bursts were observed for
substrate depletion and the corresponding formation
of bromide ions with both DatA and DatA01,
although no burst in alcohol formation was observed
with either enzyme (Fig. 4). This indicates that sub-
strate binding and the cleavage of the carbon–halogen
bond are fast steps, and that hydrolysis of the alkyl–
enzyme intermediate is the rate-determining step in
the conversion of 1-bromobutane by both enzymes.
Rate constants of 2.6 ꢀ 0.2 sꢁ1 and 1.2 ꢀ 0.1 sꢁ1
were obtained for DatA and DatA01, respectively, in
the linear steady-state phase; the corresponding
steady-state turnover numbers were 1.5 ꢀ 0.01 sꢁ1
and 0.7 ꢀ 0.02 sꢁ1 (Table 1). The calculated rates of
halide formation during the burst phase for DatA
Fig. 3. Steady-state fluorescence of DatA (circle) and DatA01
(triangle) at different bromide concentrations (0–1500 mM). The
fluorescence intensity was measured under excitation at 280 nm
for DatA (●) and DatA01 (▲) and at 295 nm for DatA (○) and
DatA01 (Δ). All measurements were conducted in 100 mM glycine
buffer (pH 8.6) at room temperature in triplicate. Solid lines
represent the best fits to the data based on the Stern-Volmer
equation, F/F0 = [1 + (fꢂKQ[X])]/(1 + KBr[X]), in which F/F0 is the
relative fluorescence, f is the relative fluorescence intensity of the
enzyme–halide complex, and KBr is the association equilibrium
constant for the specific binding of bromide. KQ is the quenching
constant, which is the apparent association equilibrium constant for
the nonspecific quenching interaction between bromide and the
fluorophore.
and DatA01 were 117 ꢀ 64 sꢁ1 and 147 ꢀ 20 sꢁ1
,
respectively. The higher burst rate for DatA01 indi-
cates that it is more proficient than the wild-type at
promoting the first step of the reaction (i.e. the
cleavage of the carbon–halogen bond). The overall
effect on Michaelis constant observed after the muta-
tion is most likely associated with both increased rate
of carbon–halogen bond cleavage and better stabiliza-
tion of the enzyme–substrate complex. It has been
suggested previously that the rate of carbon–halogen
bond cleavage and the second-order dissociation
constant for the enzyme and substrate complex are
the main determinants of the Michaelis constant in
HLDs [22].
consistent with the halide-binding results obtained
from the fluorescence spectroscopy experiments, which
indicated that tryptophan was a better halide-stabiliz-
ing residue than tyrosine. Moreover, substrate inhibi-
tion by 1-bromohexane (which suggests strong
binding of multiple substrate molecules to the
enzyme) was observed for DatA01 but not for the
wild-type DatA. The catalytic rates of DatA01 were
2–6-fold lower than those of the wild-type for all of
the tested substrates other than 1,3-dibromopropane
and 1-chlorohexane. This reduction in kcat is consis-
Table 1. Steady-state kinetic parameters for DatA and DatA01 with seven different substrates determined at 37 °C in 100 mM glycine
buffer (pH 8.6). The kinetic data were fitted to the Hill equation, v = kcatꢂSn/(K0.5 + Sn), where n is the Hill cooperativity coefficient and K0.5
n
is the half-maximal activity constant. The kinetic data for DatA were taken from Hasan et al. [28], with the exception of those for the
substrate 1-bromobutane. ND, not determined.
DatA
DatA01
Substrate
K0.5 (mM)
kcat (sꢁ1
)
n
Ksi (mM)
K0.5 (mM)
kcat (sꢁ1
)
n
Ksi (mM)
1,3-Dibromopropane
1,3-Dichloropropane
1,3-Diiodopropane
1-Bromohexane
2.2 ꢀ 0.3
> 4
1.5 ꢀ 0.1
> 0.01
1.0
ND
ND
ND
ND
ND
ND
0.82 ꢀ 0.08
> 4
1.3 ꢀ 0.1
1.0
ND
1.0
> 0.012
1.0
ND
0.20 ꢀ 0.03
0.14 ꢀ 0.09
0.07 ꢀ 0.02
0.53 ꢀ 0.01
1.2 ꢀ 0.1
1.7 ꢀ 0.1
0.04 ꢀ 0.01
1.5 ꢀ 0.1
3.5 ꢀ 1.6
1.0
0.16 ꢀ 0.02
0.03 ꢀ 0.01
0.06 ꢀ 0.01
0.18 ꢀ 0.01
0.72 ꢀ 0.02
0.46 ꢀ 0.07
0.04 ꢀ 0.01
0.71 ꢀ 0.03
2.7 ꢀ 0.8
1.0
ND
0.25 ꢀ 0.01
1-Chlorohexane
1.3 ꢀ 0.2
1.7 ꢀ 0.1
2.2 ꢀ 0.6
2.6 ꢀ 0.1
ND
ND
1-Bromobutane
3152
FEBS Journal 280 (2013) 3149–3159 ª 2013 The Authors Journal compilation ª 2013 FEBS

K. Hasan et al.
Unique halide-stabilizing residue of DatA
A
B
Fig. 4. Pre-steady-state burst of 1-bromobutane conversion by the haloalkane dehalogenases DatA (A) and DatA01 (B). Both reactions were
performed in 100 mM glycine buffer (pH 8.6) at 37 °C. The concentrations of 1-bromobutane (■) and 1-butanol (Δ) were determined by GC
and the bromide (○) concentration was analyzed by ion chromatography. Solid lines represent the best fit of the kinetic data to a linear
equation for the alcohol product and to the burst equation, c = A0[1 ꢁ exp(ꢁk1t)] + k2t, for the substrate and halide, where c is the
concentration, A0 is the burst amplitude, t is the time, and k1 and k2 are rate constants.
confirm the difference in stabilization. In DatA01, the
halogen atom is stabilized by two hydrogen bonds; the
average distance between the N43:2HD2 atom and the
Molecular modelling
The docked poses of enzyme–substrate complexes were
searched for geometries close to the near attack con-
figuration (NAC) [30], in which the distance between
the OD1 atom of the D108 residue and the carbon
ꢀ
bromine atom is 3.0 ꢀ 0.6 A, whereas that between
the W109:HE1 atom and the bromine atom is
ꢀ
2.8 ꢀ 0.4 A. By contrast, only N43 formed a hydro-
ꢀ
atom attached to the halogen is < 3.4 A and the angle
gen bond with the bromine atom in DatA, with an
ꢀ
between the halide, the carbon atom attached to the
halide and the OD1 atom of D108 is between 157º and
180º. All of the chosen geometries fulfilled the NAC
distance requirement; their NAC angles ranged from
105º to 163º. The geometries selected for further QM
calculations are summarized in Table S1.
average length of 2.9 ꢀ 0.5 A. The average distance
between Y109:HH1 and the bromine atom is
ꢀ
4.6 ꢀ 0.4 A, which is considerably longer than would
be expected for a hydrogen bond (Fig. 5).
Discussion
QM calculations were performed to investigate the
effect of mutating the halogen-stabilizing residue on
the rate of the SN2 reaction. A comparison of the cal-
culated reaction profiles showed that the mutation of
the halide-stabilizing residue reduced the activation
energy for both brominated and iodinated compounds
(Fig. S2 and Table S2). The activation energies calcu-
lated for chlorinated substrates were almost equal for
both enzymes. The main interactions that stabilized
the halogen atoms of the substrates were provided by
residues H41, N43, P44, D108, Y109/W109, E127,
F147, F160, G207, P208 and L211 (Fig. S3). The
Y109 residue present in the wild-type enzyme provided
less effective stabilization than the W109 residue in the
mutant. Unexpectedly, inspection of the geometries
during SN2 reaction revealed that the substrate’s halo-
gen atom was stabilized by only one hydrogen bond
(donated by N43) in the wild-type enzyme, DatA. The
second bond that is observed in other dehalogenases
was not detected in the wild-type enzyme but was pres-
ent in the mutant DatA01.
The haloalkane dehalogenase DatA from A. tumefac-
iens C58 is a recently discovered member of the versa-
tile HLD family [28]. A sequence alignment of DatA
with known HLDs revealed that DatA carries a
unique tyrosine residue in the position commonly
occupied by a tryptophan residue situated in close
proximity to the nucleophile and is highly conserved in
other members of the family. This residue has previ-
ously been shown to be important in stabilizing: (a)
the binding of the substrate molecule in the enzyme–
substrate, (b) the transition state of the SN2 reaction,
and (c) the halide ion released after carbon–halogen
bond cleavage [23]. We constructed a variant of this
enzyme, DatA01, in which the tyrosine residue is
replaced with tryptophan with the aim of studying the
mutation’s effect on the enzyme’s structure and
catalytic properties.
The quenching of the intrinsic fluorescence of HLDs
upon halide binding has been reported previously
[19,24–26] and was used to obtain quantitative infor-
mation on how the mutation affected halide binding.
MD simulations of the complexes formed by DatA
and DatA01 with 1-bromohexane were performed to
FEBS Journal 280 (2013) 3149–3159 ª 2013 The Authors Journal compilation ª 2013 FEBS
3153

Unique halide-stabilizing residue of DatA
K. Hasan et al.
interactions of Trp125 and Trp175 with the halogen
atom of the substrate are essential for the cleavage of
carbon–halogen bond. Most of these mutants show
dramatically reduced catalytic performance. Unexpect-
edly, changing the tryptophan to tyrosine resulted in
preservation of turnover number but increased the
Michaelis constant [24]. The W175Y substitution in
DhlA yielded a mutant with a lower binding affinity
as a result of reduced rates of substrate binding and
cleavage of the carbon–halogen bond. Furthermore,
the W175Y substitution increased the rate of bromide
release and the hydrolysis of alkyl–enzyme intermedi-
ate became the rate determining step for the mutant
[26]. Transient kinetic analysis of DatA and DatA01
indicated that the Y109W mutation enhanced the sta-
bility of the enzyme–substrate complex and increased
the rate of the carbon–halogen bond cleavage,
although it did not result in change of rate-limiting
step.
A
B
The computational analysis of the contribution of
active site residues towards ligand stabilization indi-
cated that the N43–W109 pair of the DatA01 mutant
provides better stabilization of the halogen atom than
the N43–Y109 pair of the DatA wild-type. Clearly, the
hydrogens located on the aromatic ring of the tyrosine
residue provide very little stabilization to the halogen
atom. The geometric parameters obtained during QM
calculations of the SN2 reaction and 20-ns MD simula-
tions of the protein–substrate complex revealed that
tyrosine cannot act as a hydrogen bond donor to
stabilize the halogen atom/halide ion as the reaction
proceeds. However, when it is replaced by a trypto-
phan residue, the new hydrogen bond is introduced
(Fig. 5).
Fig. 5. Comparison of the distances between the bromide ion and
the halide-stabilizing residues during MD simulations of (A) DatA
and (B) DatA01 complexes with 1-bromohexane. Lines correspond
to the moving average calculated over a moving ten-point frame.
Red lines represent the distance W109:HE1–Br (DatA) or Y109:
HH1–Br (DatA01); blue lines represent the distance N43:2HD2–Br.
The tryptophan-selective fluorescence spectrum of
DatA01 confirmed that the introduced residue inter-
acts specifically with the halide ion during bromide
binding. Furthermore, the substitution of tyrosine by
tryptophan significantly decreased the enzyme’s equi-
librium dissociation constant: the affinity of the
DatA01 variant for bromide is 3.7-fold higher than
that of the wild-type enzyme. The results from the
halide binding experiments were consistent with those
obtained in steady-state kinetics studies, in which the
Y109W substitution reduced the enzyme’s K0.5 values
for all tested substrates. This suggests that the intro-
duced tryptophan residue increases the mutant
enzyme’s affinity for the alkyl halide substrate, as well
as the halide ion product of the hydrolysis reaction. In
agreement with this finding, it has been previously
reported that tryptophan is a more effective stabilizing
residue than tyrosine [23–26]. The catalytic role of the
tryptophan residues Trp125 and Trp175 forming the
halide-binding site in the active site cavity of HLD
DhlA has been studied by replacing these two residues
by phenylalanine, arginine, tyrosine and glutamine
[24]. The properties of these mutants indicated that the
DatA01 was less active than the wild-type DatA
towards most of the tested substrates. Semi-empirical
QM calculations indicated that the Y109W mutation
in DatA01 introduces the second hydrogen bond,
which helps to stabilize the partial negative charge that
develops on the halide ion as the carbon–bromine
bond breaks (Fig. 5). This decreases the activation
energy and enthalpy of the first step of the reaction.
However, this beneficial effect on the SN2 step is coun-
teracted by the mutation’s negative effect on the rate-
limiting step, namely the hydrolysis of the alkyl–
enzyme intermediate in both DatA and DatA01 when
using 1-bromobutane as the substrate. This observa-
tion explains why DatA can tolerate the ‘loss’ of a
halide-stabilizing residue without compromising its cat-
alytic activity towards the majority of tested haloge-
nated substrates. The hydrolysis of the alkyl-enzyme
intermediate has been previously identified as the
rate-limiting step for LinB when using 1-chlorohexane,
3154
FEBS Journal 280 (2013) 3149–3159 ª 2013 The Authors Journal compilation ª 2013 FEBS

K. Hasan et al.
Unique halide-stabilizing residue of DatA
DpnI-treated PCR product was transformed into E. coli
DH5a and the datA01 gene was subcloned into expression
vector pET21b between the NdeI and HindIII restriction
sites. The expression and purification of DatA and DatA01
were carried out as reported previously [28].
chlorocyclohexane and bromocyclohexane as substrates
[22]. Based on the fact that the substitution of the
unique tyrosine for conserved tryptophan did not
dramatically improve the catalytic activity, we speculate
that datA is not a cryptic gene and that DatA is fully
functional enzyme.
Interestingly, DatA01 exhibited improved activity
towards several 1,2-dihalogenated substrates, compris-
ing an important group of halogenated chemicals that
are commonly used in industry and agriculture and
also occur naturally. We speculate that the SN2 reac-
tion may be the rate-limiting step for 1,2-dihalogenat-
ed substrates as a result of the steric hindrance of the
attacking nucleophile arising from the presence of a
second bulky halogen in the wild-type DatA.
CD spectroscopy
CD spectra were recorded at room temperature (22 °C)
using
a Jasco J-810 spectropolarimeter (Jasco, Tokyo,
Japan). Data were collected from 185 to 260 nm (scanning
speed of 100 nmꢂminꢁ1; response time of 1 s; bandwidth of
2 nm). The spectra were acquired with a cuvette path
length of 0.1 cm and
a
protein concentration of
0.23 mgꢂmLꢁ1. Each spectrum shown is the average of ten
individual scans and was corrected for baseline noise. CD
data were expressed in terms of the mean residue ellipticity
(ΘMRE) using the equation:
Both DatA and DatA01 belong to the SSG-IV
group of HLDs [31] and so the Y109W mutation did
not cause a change in overall substrate specificity.
However, the relative positions of DatA and DatA01
within the group are different and the addition of a
second halide-stabilizing residue in close proximity to
the nucleophile clearly affects the substrate specificity
of DatA. The nature of the halide-stabilizing residues
clearly has a strong effect on substrate specificity in
this enzyme family.
Hobs ꢂ Mw ꢂ 100
HMRE
¼
n ꢂ c ꢂ l
where Θobs is the observed ellipticity in degrees, Mw is
the protein molecular weight, n is number of residues,
l is the cell path length, c is the protein concentration,
and the factor 100 originates from the conversion of
the molecular weight to mgꢂdmolꢁ1
.
In summary, the results of the present study demon-
strate that the novel HLD DatA from A. tumefaciens
C58 catalyses the hydrolytic dehalogenation of haloge-
nated compounds using a single hydrogen bond to sta-
bilize the halogenated substrate, the transition state of
SN2 reaction and the halide product. This has not been
observed with any other native HLD, demonstrating
that DatA is the first representative of a new paradigm
for catalysis by this versatile enzyme family.
The thermal stability of DatA and DatA01 was mea-
sured by heating a solution of the protein from 22 to 80 °C
at 1 °Cꢂminꢁ1. The recorded thermal denaturation curves
were approximately normalized to represent signal changes
between ~ 1 and 0, and fitted to sigmoidal curves using
ORIGIN, version 6.1 (OriginLab, Northampton, MA, USA).
Melting temperatures (Tm) were recorded as the midpoints
of the normalized thermal transition for each protein.
Substrate specificity profiles
Experimental procedures
The enzymes’ substrate specificity was measured in 100 mM
glycine buffer (pH 8.6) at 37 °C using a set of 30 different
halogenated compounds, as reported previously [28]. The
release of halide ions in each reaction was quantified by the
method of Iwasaki et al. [32]. The matrix containing the
specific activity data for DatA, DatA01 and eight well-
characterized wild-type HLDs towards the 30 substrates
was analyzed by PCA as described previously [31].
Mutagenesis and protein purification
Established methods were used for digestion of plasmid
DNA and PCR-amplified DNA fragments using restriction
endonucleases, agarose gel electrophoresis and transforma-
tion into E. coli cells. Plasmid dsDNA was isolated using
the GeneJET Plasmid Miniprep Kit (Fermentas, Burling-
ton, Canada). The datA01 gene that codes for the Y109W
mutant was constructed by inverse PCR using the primers
datA01-f: 5′-GTGCTGCAGGATTGGGGCGCGGCGTTT
G-3′ and datA01-r: 5′-CAAACGCCGCGCCCCAATCCT
GCAGCAC-3′ (substitution underlined) in conjunction
with the Pfu high-fidelity polymerase (Stratagene, La Jolla,
CA, USA) in accordance with the manufacturer’s instruc-
tions. A codon-optimized synthetic datA gene embedded in
the cloning vector pUC18 was used as a template. The
Fluorescence spectroscopy
Steady-state fluorescence was measured in 1-cm quartz
cuvettes using
a Fluoromax-4P (Horiba Jobin Yvon,
Edison, NJ, USA) fluorescence spectrometer. Fluorescence
intensity was recorded at the emission maximum (334 nm)
in response to the excitation of both tryptophan and tyro-
sine at 280 nm and to the selective excitation of tryptophan
FEBS Journal 280 (2013) 3149–3159 ª 2013 The Authors Journal compilation ª 2013 FEBS
3155

Unique halide-stabilizing residue of DatA
K. Hasan et al.
at 295 nm. The fluorescence of wild-type DatA and DatA
01 was recorded at different bromide concentrations rang-
ing from 0 to 1500 mM. All fluorescence spectroscopy
experiments were performed in 100 mM glycine buffer (pH
8.6) at 20 °C.
was taken and divided by the sum of all the log values for
that particular enzyme. The log-transformed, weighted
measure of that enzyme’s activity towards that specific sub-
strate (relative to its activity towards all of the other sub-
strates) was thus obtained. Such transformed data were
used to compare the specificity profile of DatA01 with pre-
viously defined SSGs of HLDs [31], without regard to over-
all specific activity.
Steady-state kinetics analysis
Steady-state kinetics experiments were performed using
seven different substrates in 100 mM glycine buffer (pH 8.6)
at 37 °C. After the addition of the substrate, the reaction
mixture was pre-incubated for 30 min at 37 °C before initi-
ating the reaction. The pre-incubated substrate (1 mL) was
withdrawn and mixed with methanol containing 1,2-dichlo-
roethane as an internal standard for quantification of the
initial substrate concentration. The substrate concentration
was quantified by GC-MS using a Trace 2000 instrument
(Thermo Fisher Scientific, Waltham, MA, USA). The enzy-
matic reaction was initiated by adding an enzyme solution.
The reaction’s progress was monitored every 10 min and the
rate of halide ion formation was quantified by the method
of Iwasaki et al. [32]. The kinetic data were fitted by nonlin-
ear regression using ORIGIN, version 6.1 (Origin Lab).
Molecular docking
The crystal structures of DatA and DatA01 were taken as
starting points for modelling. Atoms with multiple alterna-
tive crystallographic positions were modelled in the alter-
nate conformation A. All crystallographic water molecules
and ions were removed. The polar hydrogens were added
at a simulated pH of 8.5 using the H++ server [33]. Six
substrates (1,3-dibromopropane, 1,3-dichloropropane, 1,3-
diiodopropane, 1-bromohexane, 1-chlorohexane and 1-iod-
ohexane) were docked into the DatA and DatA01
structures using AUTODOCK, version 4.0 [34]. This set of
compounds was selected to permit comparisons of the
effects of substrate size and the identity of the halide ion
on the substrate’s binding geometry. Gasteiger partial
atomic charges, AUTODOCK atom types and solvation
parameters were assigned to the ligands and the enzymes;
nonpolar hydrogens and their charges were merged with
their parent carbon atoms using MGL TOOLS [35]. A grid box
of 64 9 64 9 64 points in the x, y and z dimensions was
Pre-steady-state kinetic analysis
Rapid quench-flow experiments were performed at 37 °C in
a 100 mM glycine buffer (pH 8.6) using a QFM 400 rapid
quench flow instrument (BioLogic, Grenoble, France). The
reaction was started by rapidly mixing 70 lL of enzyme
solution with 70 lL of substrate solution, and quenched
with 100 lL of 0.8 M H2SO4 after time intervals ranging
from 2 ms to 1.6 s. The quenched mixture was directly
injected into 0.5 mL of ice-cold diethyl ether containing
1,2-dichloroethane as an internal standard. After extrac-
tion, the diethyl ether layer containing the noncovalently
bound substrate and the alcohol product was collected,
dried on a short column containing anhydrous Na2SO4,
and analyzed by GC-MS using a Trace 2000 instrument
(Thermo Fisher Scientific, Waltham, MA, USA) and a
DB-5MS capillary column (Agilent Technologies, Santa
Clara, CA, USA). The bromide ion content of the water
phase was measured using an 861 Advanced Compact IC
ion chromatograph equipped with a METROSEP A Supp
5 column (Metrohm, Herisau, Switzerland).
ꢀ
used with a grid spacing of 0.25 A. The grid was positioned
at the coordinates x = 34.99, y = 15.79, z = 14.9 and
x = 35.05, y = 15.89, z = 14.9, respectively, to cover the
entire active site cavities of DatA and DatA01. Two hun-
dred and fifty docking calculations were performed for each
ligand employing the Lamarckian genetic algorithm with
the parameters: initial population size of 300, maximum of
30 000 generations, elitism value of 1, mutation rate of
0.02, and cross-over rate of 0.8. The maximum of energy
evaluations were set to 10 000 000. The local search was
based on the pseudo Solis and Wets algorithm with a maxi-
mum of 300 iterations per local search [36]. Energy of the
unbound system was estimated as the internal energy of the
unbound extended conformation determined from
a
Lamarckian genetic algorithm search. Final orientations
from every docking run were clustered within a clustering
ꢀ
tolerance for an r.m.s.d. of 0.5 A.
PCA
QM calculations
PCA was performed using STATISTICA, version 10.0 (Stat-
Soft, Tulsa, OK, USA). The raw data were log-transformed
and weighted relative to the individual enzyme’s activity
towards other substrates before analysis. Each specific
activity value was incremented by 1 unit to avoid logarith-
mic transformation of zero values; the log of this new value
ꢀ
All residues within 12 A of the bound substrate were
simulated in the quantum mechanical calculations. The
DatA and DatA01 active site models consisted of 55 amino
acids (H41, G42, N43, P44, T45, L49, L68, Y71, L83, Q86,
Q107, D108, Y/W109, G110, A111, A112, F113, E132,
3156
FEBS Journal 280 (2013) 3149–3159 ª 2013 The Authors Journal compilation ª 2013 FEBS

K. Hasan et al.
Unique halide-stabilizing residue of DatA
P133, V134, I138, L143, S144, P145, F147, L156, F163,
V164, F170, L171, L174, F175, P176, F178, F179, A203,
I204, L205, A206, G207, P208, N210, L211, P212, V213,
T219, L224, G246, F247, L248, I273, H274, F275, V276
and Q277) together with the appropriate substrate mole-
cules. The total charge of the system was ꢁ2. The reaction
pathways were mapped using MOPAC 2009 [37] with the
MOZYME algorithm [38] and the DRIVER external subroutine
[39]. The use of semi-empirical modelling made it possible
to perform quantum mechanical calculations despite the
relatively large size of the model, which incorporates the
entire first shell of the active site and many second shell
residues. The SN2 reaction step was modelled by shortening
the distance between the OD1 atom of the nucleophile and
the C1 atom of the substrate to which the halogen atom is
with the rest of the system being harmonically restrained
ꢀꢁ2
with a force constant of 500 kcalꢂmolꢁ1ꢂA , (b) three
rounds of 1000 steps, involving 500 steepest descent steps
followed by 500 steps of conjugate gradient energy minimi-
zation, with decreasing constraints on the protein backbone
ꢀꢁ2
(500, 125 and 25 kcalꢂmolꢁ1ꢂA ), (c) a final minimization
ꢀꢁ2
with no constraints (0 kcalꢂmolꢁ1ꢂA ) applied to the pro-
tein backbone, involving 100 steepest descent steps followed
by 400 steps of conjugate gradient energy minimization, (d)
2 ns of gradual heating of the whole system from 0 to
300 K over 20 ps using a Langevin thermostat with a tem-
perature coupling constant of 1.0 ps in a constant volume
periodic box, and (e) an MD production run with a 2-fs
time step that was applied to the whole system using the
constant pressure periodic boundary conditions. The con-
stant temperature was maintained using the weak-coupling
algorithm for 20 ns of the production simulation time, with
a temperature coupling constant of 1.0 ps. Long-range elec-
trostatic interactions were modelled using the Particle mesh
ꢀ
bound. This distance was reduced by 0.05 A on each driv-
ing step, and the geometry of the system was fully opti-
mized (with the exception of the driven coordinate and the
fixed heavy atoms of the protein backbone). All calcula-
tions were performed using the AM1 Hamiltonian with
Mulliken population analysis for the charge distribution
and the BFGS geometry optimization algorithm. The key-
words used to control the calculation were: GEO-OK,
MMOK, NOINTER, NOXYZ and CHARGE. Data anal-
ysis was performed using TRITON, version 2.0b [40–42]. The
activation energy of the reaction was approximated as the
difference between the heat of formation of the enzyme–
substrate and transition state structures, whereas the
enthalpy of reaction was calculated as the difference
between the heats of formation of the reactants and prod-
ucts. The electrostatic interaction energies (Eel) were calcu-
lated as the sum of all atom–atom interaction energies
between each amino acid and the halogen atoms of the
substrates and the halide ion, respectively, according to
Coulomb’s law, using the geometries and partial charges
obtained from the MOPAC calculations.
ꢀ
Ewald method with a nonbonded cut-off of 10 A and the
SHAKE algorithm [46]. The coordinates were saved at inter-
vals of 1 ps. The resulting trajectories were analyzed using
the Ptraj module of AMBER and were visualized using
PYMOL, version 1.5 [47] and VMD, version 1.8.6 [48].
Acknowledgements
The authors would like to thank Professor M. Tanok-
ura (University of Tokyo, Tokyo, Japan) for providing
the co-ordinates for the DatA and DatA01 structures
before their publication in the Protein Data Bank.
MetaCentrum is acknowledged for providing access to
computing facilities, supported by the Ministry of
Education, Youth and Sports of the Czech Republic
(LM2010005). A.G. was supported by SoMoPro pro-
gramme no. SIGA762 with a financial contribution
from the European Commission within the 7th FP
(FP/2007–2013) under grant agreement No. 229603
and is co-financed by the South Moravian Region.
This work was supported by the Grant Agency of the
Czech Republic (P207/12/0775 and P503/12/0572), the
Grant Agency of the Czech Academy of Sciences
(IAA401630901) and the European Regional Develop-
ment Fund (CZ.1.05/2.1.00/01.0001).
MD simulations
The lowest energy docking poses in the most populated
cluster obtained for the docking of 1-bromohexane with
DatA and DatA01 were selected as the starting configura-
tions for subsequent MD simulations. All crystallographic
water molecules that did not overlap with the docked
ligands were added to the complexes. The system was neu-
tralized by adding sodium cations. Finally, the complexes
were immersed in an octahedral box of TIP3P [43] water
References
ꢀ
molecules with a minimum water layer thickness of 10.0 A.
1 Janssen DB, Oppentocht JE & Poelarends GJ (2001)
Microbial dehalogenation. Curr Opin Biotechnol 12,
254–258.
Approximately 7500 water molecules were added to solvate
each system. Energy minimization and equilibration were
performed using the PMEMD module of the AMBER, ver-
sion 11 [44] with the ff99SB force field [45]. The equilibra-
tion protocol consisted of: (a) 500 steps of steepest descent
energy minimization and 500 steps of conjugate gradient
minimization of the positions of water molecules and ions,
2 Janssen DB, Dinkla IJT, Poelarends GJ & Terpstra P
(2005) Bacterial degradation of xenobiotic compounds:
evolution and distribution of novel enzyme activities.
Environ Microbiol 7, 1868–1882.
FEBS Journal 280 (2013) 3149–3159 ª 2013 The Authors Journal compilation ª 2013 FEBS
3157

Unique halide-stabilizing residue of DatA
K. Hasan et al.
3 Kurihara T & Esaki N (2008) Bacterial hydrolytic
dehalogenases and related enzymes: occurrences,
reaction mechanisms, and applications. Chem Rec 8,
67–74.
16 Ollis DL, Cheah E, Cygler M, Dijkstra B, Frolow F,
Franken SM, Harel M, Remington SJ, Silman I &
Schrag J (1992) The alpha/beta hydrolase fold. Protein
Eng 5, 197–211.
ꢁ
17 Streltsov VA, Prokop Z, Damborsky J, Nagata Y,
4 Prokop Z, Sato Y, Brezovsky J, Mozga T,
Chaloupkova R, Koudelakova T, Jerabek P,
Stepankova V, Natsume R, Van Leeuwen JG et al.
(2010) Enantioselectivity of haloalkane dehalogenases
and its modulation by surface loop engineering. Angew
Chem Int Ed Engl 49, 6111–6115.
Oakley A & Wilce MCJ (2003) Haloalkane
dehalogenase LinB from Sphingomonas paucimobilis
UT26: X-ray crystallographic studies of dehalogenation
of brominated substrates. Biochemistry 42,
10104–10112.
ꢁ
5 Szymanski W, Westerbeek A, Janssen DB & Feringa
ꢁ
18 Verschueren KH, Seljee F, Rozeboom HJ, Kalk KH &
BL (2011) A simple enantioconvergent and
chemoenzymatic synthesis of optically active
a-substituted amides. Angew Chem Int Ed Engl 50,
10712–10715.
Dijkstra BW (1993) Crystallographic analysis of the
catalytic mechanism of haloalkane dehalogenase.
Nature 363, 693–698.
19 Verschueren KH, Kingma J, Rozeboom HJ, Kalk KH,
Janssen DB & Dijkstra BW (1993) Crystallographic and
fluorescence studies of the interaction of haloalkane
dehalogenase with halide ions. Studies with halide
compounds reveal a halide binding site in the active
site. Biochemistry 32, 9031–9037.
6 Swanson PE (1999) Dehalogenases applied to industrial-
scale biocatalysis. Curr Opin Biotechnol 10, 365–369.
7 Bidmanova S, Chaloupkova R, Damborsky J & Prokop
Z (2010) Development of an enzymatic fiber-optic
biosensor for detection of halogenated hydrocarbons.
Anal Bioanal Chem 398, 1891–1898.
20 O’Hagan D & Schmidberger JW (2010) Enzymes that
catalyse SN2 reaction mechanisms. Nat Prod Rep 27,
900–918.
€
8 Reardon KF, Campbell DW & Muller C (2009) Optical
fiber enzymatic biosensor for reagentless measurement
of ethylene dibromide. Eng Life Sci 9, 291–297.
9 Franken SM, Rozeboom HJ, Kalk KH & Dijkstra BW
(1991) Crystal structure of haloalkane dehalogenase: an
enzyme to detoxify halogenated alkanes. EMBO J 10,
1297–1302.
ꢁ
21 Otyepka M, Banas P, Magistrato A, Carloni P &
ꢁ
Damborsky J (2008) Second step of hydrolytic
dehalogenation in haloalkane dehalogenase investigated
by QM/MM methods. Proteins 70, 707–717.
ꢁ
22 Prokop Z, Monincova M, Chaloupkova R, Klvana M,
ꢁ
ꢁ
Nagata Y, Janssen DB & Damborsky J (2003)
10 Gehret JJ, Gu L, Geders TW, Brown WC, Gerwick L,
Gerwick WH, Sherman DH & Smith JL (2012)
Structure and activity of DmmA, a marine haloalkane
dehalogenase. Protein Sci 21, 239–248.
Catalytic mechanism of the halaloalkane dehalogenase
LinB from Sphingomonas paucimobilis UT26. J Biol
Chem 278, 45094–45100.
ꢁ
11 Hesseler M, Bogdanovic X, Hidalgo A, Berenguer J,
ꢁ
23 Bohac M, Nagata Y, Prokop Z, Prokop M, Monincova
ꢁ
ꢁ
M, Tsuda M, Koca J & Damborsky J (2002) Halide-
Palm GJ, Hinrichs W & Bornscheuer UT (2011)
Cloning, functional expression, biochemical
characterization, and structural analysis of a haloalkane
dehalogenase from Plesiocystis pacifica SIR-1. Appl
Microbiol Biotechnol 91, 1049–1060.
stabilizing residues of haloalkane dehalogenases studied
by quantum mechanic calculations and site-directed
mutagenesis. Biochemistry 41, 14272–14280.
24 Kennes C, Pries F, Krooshof GH, Bokma E, Kingma J
& Janssen DB (1995) Replacement of tryptophan
residues in haloalkane dehalogenase reduces halide
binding and catalytic activity. Eur J Biochem 228,
403–407.
12 Holmquist M (2000) Alpha/Beta-hydrolase fold
enzymes: structures, functions and mechanisms. Curr
Protein Pept Sci 1, 209–235.
ꢁ
13 Marek J, Vevodova J, Smatanova IK, Nagata Y,
ꢁ
ꢁ
ꢁ
Svensson LA, Newman J, Takagi M & Damborsky J
(2000) Crystal structure of the haloalkane dehalogenase
from Sphingomonas paucimobilis UT26. Biochemistry
39, 14082–14086.
25 Schanstra JP & Janssen DB (1996) Kinetics of halide
release of haloalkane dehalogenase: evidence for a slow
conformational change. Biochemistry 35, 5624–5632.
26 Krooshof GH, Ridder IS, Tepper AW, Vos GJ,
Rozeboom HJ, Kalk KH, Dijkstra BW & Janssen DB
(1998) Kinetic analysis and X-ray structure of
haloalkane dehalogenase with a modified halide-binding
site. Biochemistry 37, 15013–15023.
14 Newman J, Peat TS, Richard R, Kan L, Swanson PE,
Affholter JA, Holmes IH, Schindler JF, Unkefer CJ &
Terwilliger TC (1999) Haloalkane dehalogenases:
structure of a Rhodococcus enzyme. Biochemistry 38,
16105–16114.
27 Chovancova E, Kosinski J, Bujnicki JM & Damborsky
J (2007) Phylogenetic analysis of haloalkane
15 Mazumdar PA, Hulecki JC, Cherney MM, Garen CR
& James MNG (2008) X-ray crystal structure of
Mycobacterium tuberculosis haloalkane dehalogenase
Rv2579. Biochim Biophys Acta 1784, 351–362.
dehalogenases. Proteins 67, 305–316.
28 Hasan K, Fortova A, Koudelakova T, Chaloupkova R,
Ishitsuka M, Nagata Y, Damborsky J & Prokop Z
3158
FEBS Journal 280 (2013) 3149–3159 ª 2013 The Authors Journal compilation ª 2013 FEBS

K. Hasan et al.
Unique halide-stabilizing residue of DatA
ꢁ
41 Prokop M, Damborsky J & Koca J (2000) TRITON: in
(2011) Biochemical characteristics of the novel
haloalkane dehalogenase DatA, isolated from the plant
pathogen Agrobacterium tumefaciens C58. Appl Environ
Microbiol 77, 1881–1884.
silico construction of protein mutants and prediction of
their activities. Bioinformatics 16, 845–846.
ꢁ
42 Dambosky J, Prokop M & Koca J (2001) TRITON:
29 Vieille C & Zeikus GJ (2001) Hyperthermophilic
enzymes: sources, uses, and molecular mechanisms for
thermostability. Microbiol Mol Biol Rev 65, 1–43.
30 Hur S, Kahn K & Bruice TC (2003) Comparison of
formation of reactive conformers for the SN2
displacements by CH3CO2- in water and by Asp124-
CO2- in a haloalkane dehalogenase. Proc Natl Acad Sci
USA 100, 2215–2219.
graphic software for rational engineering of enzymes.
Trends Biochem Sci 26, 71–73.
43 Jorgensen WL, Chandrasekhar J, Madura JD, Impey
RW & Klein ML (1983) Comparison of simple
potential functions for simulating liquid water. Journal
of Chemical Physics 79, 926–935.
44 Case DA, Darden TA, Cheatham TE, Simmerling CL,
Wang J, Duke RE, Luo R, Crowley M, Walker RC,
Zhang W et al. (2010) Amber 11. University of
California, San Francisco.
31 Koudelakova T, Chovancova E, Brezovsky J,
Monincova M, Fortova A, Jarkovsky J & Damborsky
J (2011) Substrate specificity of haloalkane
45 Hornak V, Abel R, Okur A, Strockbine B, Roitberg A
& Simmerling C (2006) Comparison of multiple Amber
force fields and development of improved protein
backbone parameters. Proteins 65, 712–725.
dehalogenases. Biochem J 435, 345–354.
32 Iwasaki I, Utsumi S & Ozawa T (1952) New
colorimetric determination of chloride using mercuric
thiocyanate and ferric ion. Bull Chem Soc Jpn 25, 226.
33 Gordon JC, Myers JB, Folta T, Shoja V, Heath LS &
Onufriev A (2005) H++: a server for estimating pKas
and adding missing hydrogens to macromolecules.
Nucleic Acids Res 33, W368–W371.
46 Ryckaert J-P, Ciccotti G & Berendsen HJ (1977)
Numerical integration of the cartesian equations of
motion of a system with constraints: molecular
dynamics of n-alkanes. J Comput Phys 23, 327–341.
€
47 Schrodinger L (2010) The PyMOL Molecular Graphics
34 Morris GM, Goodsell DS, Halliday RS, Huey R, Hart
WE, Belew RK & Olson AJ (1998) Automated docking
using a Lamarckian genetic algorithm and an empirical
binding free energy function. J Comp Chem 19,
1639–1662.
System, Version 1.5.
48 Humphrey W, Dalke A & Schulten K (1996) VMD:
visual molecular dynamics. J Mol Graph 14, 27–28.
Supporting information
35 Sanner MF (1999) Python: a programming language
for software integration and development. J Mol Graph
Model 17, 57–61.
Additional supporting information may be found in
the online version of this article at the publisher’s web
site:
Fig. S1. CD spectra of DatA and DatA01.
Fig. S2. Comparison of the reaction energy profiles
obtained from QM calculations.
36 Solis FJ & Wets RJ-B (1981) Minimization by random
search techniques. Math Oper Res 6, 19–30.
37 Stewart JJP (2008) MOPAC2009 Stewart
Computational Chemistry. Colorado Springs, CO,
USA.
Fig. S3. Comparison of the halide ion stabilization by
the active site residues obtained from QM calculations.
Table S1. Geometries of enzyme–substrate complexes
selected from MD snapshots as inputs for QM calcula-
tions.
38 Stewart JJP (1996) Application of localized molecular
orbitals to the solution of semiempirical self-consistent
field equations. Int J Quantum Chem 58, 133–146.
ꢂ
ꢁ
ꢁ
39 Cernohorsky M, Kuty M & Koca J (1997) A
multidimensional driver for quantum chemistry
program MOPAC. Comp & Chem 21, 35–44.
Table S2. Activation energies (Ea) and enthalpies (DH)
for the hydrolysis of various substrates calculated from
the reaction profiles.
ꢁ
ꢁ
40 Prokop M, Adam J, Krız Z, Wimmerova M & Koca J
(2008) TRITON: a graphical tool for ligand-binding
protein engineering. Bioinformatics 24, 1955–1956.
FEBS Journal 280 (2013) 3149–3159 ª 2013 The Authors Journal compilation ª 2013 FEBS
3159