The influence of key residues in the tunnel entrance and the active site on activity and selectivity of toluene-4-monooxygenase

Article abstract of DOI:10.1016/j.molcatb.2010.03.006

Site-directed saturation mutagenesis is a convenient method to fine tune enzyme activity and selectivity at known "hot spots". The objective of this work was to investigate the influence of mutations in the tunnel entrance of toluene 4-monooxygenase (T4MO

Full text of DOI:10.1016/j.molcatb.2010.03.006

Journal of Molecular Catalysis B: Enzymatic 66 (2010) 72–80
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
The influence of key residues in the tunnel entrance and the active site on
activity and selectivity of toluene-4-monooxygenase
Moran Brouk, Netta-Lee Derry, Janna Shainsky, Zohar Ben-Barak Zelas,
Yulia Boyko, Keren Dabush, Ayelet Fishman∗
Department of Biotechnology and Food Engineering, and Institute of Catalysis Science and Technology, Technion-Israel Institute of Technology, Haifa 32000, Israel
a r t i c l e i n f o
a b s t r a c t
Article history:
Site-directed saturation mutagenesis is a convenient method to fine tune enzyme activity and selectiv-
ity at known “hot spots”. The objective of this work was to investigate the influence of mutations in the
tunnel entrance of toluene 4-monooxygenase (T4MO) from Pseudomonas mendocina KR1 on rate, regiose-
lectivity, and enantio-selectivity, in comparison with mutations located in the active site. Residue TmoA
D285, located at the tunnel entrance, distant from the active site was randomized, as well as residue
I100 which is positioned in the vicinity of the diiron center. Both libraries were screened for regioselec-
tive hydroxylation of phenylethanol (PEA), and enantioselective oxidation of methyl p-tolyl sulfide and
styrene. It was discovered that mutations at position 285 enhanced the rate of catalysis but did not affect
the specificity. Substitutions D285I and D285Q were beneficial for the oxidation of the large and bulky
substrates PEA and methyl p-tolyl sulfide (8–11-fold improvement), however variant D285S improved
the activity rate for styrene oxidation (1.7-fold). Thus, the extent of improvement in rate, and the best
amino acid replacement were substrate dependant. In contrast, mutations at position I100 influenced
the activity rate and the selectivity for all three substrates tested. Combining the effects by mutating
both “gates”, to the tunnel entrance along with the active site, resulted in an additive or even synergistic
outcome on enzyme activity and selectivity.
Received 26 November 2009
Received in revised form 19 February 2010
Accepted 14 March 2010
Available online 27 March 2010
Keywords:
Toluene 4-monooxygenase
Saturation mutagenesis
Tunnel
Enantio-selectivity
Biocatalysis
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
evolution, random mutations are generated on the gene of interest
thus creating a diverse library, which is screened for the desirable
Naturally occurring enzymes are remarkable biocatalysts with
numerous potential applications in organic synthesis, industry and
medicine. One of the main advantages of biocatalysts, in addition
to their high regio- and enantio-selectivity, is their ability to accept
substrates other than their natural ones [1,2]. However, enzyme
activity on non-natural compounds is often much lower than on
the natural substrate. Several protein engineering techniques have
been developed in order to overcome the limitations of natural
enzymes and to design enzymes for specific applications. These
techniques can be roughly divided into two main strategies: ratio-
nal design and directed evolution [3].
quality.
In between these two strategies lies site-specific saturation
mutagenesis that can be used to introduce all possible amino
acids at any predetermined position in a gene. This semi-rational
approach can provide more comprehensive information than can
be gained by a single amino acid substitution, as well as overcome
the drawbacks and biases of random mutagenesis (e.g. incom-
pleteness of the amino acid diversity introduced and the need for
high-throughput screening) [6,7]. Site-specific saturation mutage-
nesis enables the fine-tuning of enzyme structure by “finding” the
best amino acid at a specific position.
Rational design uses detailed knowledge about the enzyme
structure and function in order to identify precise residue(s) which
can be mutated for designing or improving a specific quality of
the enzyme [4,5]. In contrast to the comprehensive understand-
ing required for rational design, directed evolution is based on
the Darwinian principles of mutation and selection and does not
require prior knowledge regarding the structure [2,3]. In directed
Often, when changes in the catalytic properties of an enzyme
are sought (such as regio- and enantio-selectivity, substrate speci-
ficity and alternative catalytic activity) mutations close to the active
site appear to be more effective than distant ones. For improving
protein activity and stability (such as thermostability or stability
toward organic solvents), both close and distant mutations have
been shown to be useful [5,8].
Relatively few studies have investigated the effect of mutating
residues located in the tunnel or tunnel entrance, leading to the
active site. Vardar et al. found that mutating a residue located in
the entrance of the tunnel in toluene-ortho-xylene monooxygenase
∗
Corresponding author. Tel.: +972 4 829 5898; fax: +972 4 829 3399.
E-mail address: afishman@tx.technion.ac.il (A. Fishman).
1381-1177/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2010.03.006

M. Brouk et al. / Journal of Molecular Catalysis B: Enzymatic 66 (2010) 72–80
73
of Pseudomonas stutzeri OX1, influences the catalysis rate of the
enzyme, whereas the regiospecificity remained the same [9,10].
Furthermore, it was shown by Schmitt et al. that modifications
made in the tunnel of Candida rugosa lipase can alter the activity
of the enzyme towards fatty acids with different chain lengths (the
activity was decreased when the chain was long enough to reach
the mutated site) [11]. Capila et al. used site-specific mutagenesis
to change a residue within the active site tunnel of chondroitin AC
lyase which affects the binding affinity and causes a change in the
action pattern of the enzyme [12]. Chaloupková et al. reported that
the effect of mutating a residue located at the tunnel entrance of
haloalkane dehalogenase from Sphingomonas paucimobilis UT26, on
the enzyme activity varied with the structure of the substrate, but
in general, the activity increased with the introduction of small and
nonpolar amino acids [13]. Recently, Pavlova et al. investigated key
residues in access tunnels of Rhodococcus rhodochrous haloalkane
dehalogenase and described the influence of these residues on the
substrate and water passage to the active site [14].
In the present study, site-specific saturation mutagenesis was
used to investigate a key residue in the tunnel entrance of
toluene-4-monooxygenase (T4MO) from Pseudomonas mendocina
KR1. T4MO, belonging to the toluene monoxygenase (TMO) family,
is a soluble four-component enzyme which hydroxylates toluene
primarily at the para position forming 96% p-cresol. Other exten-
sively studied TMOs are toluene ortho-monooxygenase (TOM) of
Burkholderia cepacia G4 [15], toluene ortho-xylene monooxygenase
(ToMO) of Pseudomonas stutzeri OX1 [10,16] and toluene para-
monooxygenase (TpMO) of Ralstonia pickettii PKO1 [6,17]. T4MO
consists of a 211-kDa hydroxylase which contains the catalytically
active diiron center and is composed of two subunits in an (␣␤␥)2
quaternary structure [18,19]. The other components are a 36-kDa
NADH oxidoreductase, a 12.5-kDa Rieske-type [2Fe–2S] ferredoxin,
and a 11.6-kDa effector protein [19–21].
described earlier. All experiments were conducted as described
previously [25,26]. Concisely, overnight cells were diluted to an
optical density (OD) at 600 nm of 0.1 and grown to an OD of 1.3. The
exponentially grown cells were centrifuged (8000 × g for 10 min at
25 ◦C) and re-suspended in potassium phosphate buffer (100 mM,
pH 7.0). Cells were subsequently used in biotransformation pro-
tocols as described later on. Expression of TMOs (wild-type [WT]
and variants) by pBS(Kan) vectors within E. coli strains produced
blue or brown cells on agar plates and in broth cultures. The color
is indicative of indigoid compounds formed by oxidation of indole
from tryptophan [27,28].
2.3. Protein analysis and molecular techniques
Protein samples of cells grown with and without 1 mM
isopropyl ␤-D-thiogalactopyranoside (IPTG) were analyzed on
standard 12% Lammeli discontinuous sodium dodecyl sulfate
(SDS)-polyacrylamide gels [22]. All of the highly active enzyme
variants were analyzed by SDS-PAGE to ascertain that the increase
in activity is the result of the desired mutation rather than unex-
pected changes in expression level.
Plasmid DNA was isolated using a Mini Kit (Qiagen, CA, USA), and
DNA fragments were collected using the E-gel® CloneWellTM sys-
tem comprised of 0.8% SYBR SafeTM gels, and E-gel® iBaseTM as the
electrophoresis unit (Invitrogen, Carlsbad, CA). Transformation of
plasmids into E. coli cells was performed via electroporation using a
Micro-Pulser instrument (Bio-Rad, CA, USA) with the program Ec1
(1.8 kV, 1 pulse for a 0.1 cm cuvette).
2.4. Site-specific saturation mutagenesis
A gene library encoding all possible amino acids at position
TmoA D285 in T4MO was constructed as described previ-
ously [26]. To randomize position 285, two degenerate primers,
T4MO 285 Front and T4MO 285 Rear (Table 1), were designed as
well as primers T4MObefEcoRI Front (upstream of the unique EcoRI
site and the start codon of tmoA) and T4MOABRear (downstream of
the stop codon of tmoB and the unique AatII site). The first degen-
erate PCR fragment was amplified using primers T4MO 285 Front
and T4MOABRear, while the second was amplified using primers
T4MObefEcoRI Front and T4MO 285 Rear. The PCR program con-
sisted of an initial denaturation at 94 ◦C for 2 min, followed by 25
cycles of 94 ◦C for 45 s, 55 ◦C for 45 s, and 72 ◦C for 2.2 min, with
a final extension at 72 ◦C for 8 min. The two fragments were com-
bined during the final reassembly PCR in a 1:1 molar ratio using the
It was our objective to investigate the influence of mutations
in the tunnel on rate, regioselectivity, and enantio-selectivity of
T4MO, in comparison with mutations located in the active site. Var-
ious substrates were investigated to study the generality of tunnel
entrance mutagenesis in T4MO.
2. Materials and methods
2.1. Chemicals
2-Phenylethanol (PEA), o-, m-, p-tyrosol, methyl-p-tolyl sulfide,
(R)- and (S)-methyl-p-tolyl sulfoxide and N,O-Bis(trimethylsilyl)-
acetamide were purchased from Sigma–Aldrich Chemical Co.
(Sigma–Aldrich, Rehovot, Israel). Toluene was purchased from Bio-
Lab (Jerusalem, Israel). Hydroxytyrosol was obtained from Cayman
Chemical Co. (MI, USA). Styrene, styrene oxide, (R)-styrene oxide
and p-nitrothiophenolate were purchased from Acros Organics
(Holland Moran, Yahud, Israel). All standards were prepared as
stock solutions in ethanol. All materials used were of the highest
purity available and were used without further purification.
Table 1
Primers used for saturation mutagenesis, site-directed mutagenesis and sequencing
of the D285 region in the tmoA gene in TG1/pBS(Kan)T4MO.
Primer
Nucleotide sequencea
Mutagenesis
T4MObefEcoRI
Front
T4MOABRear
T4MO 285 Front
T4MO 285 Rear
5ꢀ-CCATGATTACGCCAAGCGCG-3ꢀ
5ꢀ-TCCATGCTCTTCACTGTTGAC-3ꢀ
5ꢀ-GGATTACTACACGCCGTTGGAGNNNCGCAGCCAG-3ꢀ
5ꢀ-AACTCCTTGAATGACTGGCTGCGNNNCTCCAACGG-3ꢀ
2.2. Bacterial strains and growth conditions
T4MO 285Q Front 5ꢀ-GGATTACTACACGCCGTTGGAGCAGCGCAGCCAG-3ꢀ
T4MO 285Q Rear
T4MO 285I Front
T4MO 285I Rear
T4MO 100L Front
T4MO 100L Rear
5ꢀ-AACTCCTTGAATGACTGGCTGCGCTGCTCCAACGG-3ꢀ
5ꢀ-GGATTACTACACGCCGTTGGAGATCCGCAGCCAG-3ꢀ
5ꢀ-AACTCCTTGAATGACTGGCTGCGGATCTCCAACGG-3ꢀ
5ꢀ-CACTTTGAAATCCCATTACGGCGCCCTCGCAGTTGG-3ꢀ
5ꢀ-GCTGCATATTCACCAACTGCGAGGGCGCCGTAATGG-3ꢀ
Escherichia coli TG1 (supE hsdꢀ5 thi ꢀ(lac-proAB) F’ [traD36
proAB+ lacIq lacZꢀM15]) with the plasmid constructs was routinely
cultivated at 37 ◦C in Luria-Bertani (LB) medium [22] supple-
mented with kanamycin at 100 ␮g/ml to maintain the plasmids.
To stably and constitutively express the toluene monooxyge-
nase genes from the same promoter, the expression vectors
pBS(Kan)TOM (henceforth TOM) [15] pBS(Kan)TpMO (hence-
forth TpMO) [23], pBS(Kan)ToMO (henceforth ToMO) [24] and
pBS(Kan)T4MO (henceforth T4MO) [23], were constructed as
Sequencing
T4MO seq 1
5ꢀ-CCCGCATGAATACTGTAAGAAGGATCGC-3ꢀ
N stands for A, T, G or C.
Positions subjected to mutagenesis are underlined and bases modified to obtain
the desired point mutation are indicated in bold.
a

74
M. Brouk et al. / Journal of Molecular Catalysis B: Enzymatic 66 (2010) 72–80
outer primers T4MObefEcoRI Front and T4MOABRear. The assem-
bling PCR was programmed similarly to the above PCR program,
with extension at 72 ◦C for 3.15 min instead of 2.2 min. The assem-
bled PCR fragment, containing randomized nucleotides at position
TmoA 285, was ligated into T4MO, after the double digestion of both
vector and insert with EcoRI and AatII, replacing the corresponding
fragment in the original plasmid. The resulting plasmid library was
electroporated into E. coli TG1 cells.
uid phase (the actual initial liquid concentration was 0.25 mM and
1 mM, respectively, based on a Henry’s Law constant of 0.27 [30]).
The vials were shaken at 30 ◦C and the reaction was stopped in
2–5 min intervals by injecting 2 ml ethyl acetate containing 0.5 mM
tert butyl benzene as an internal standard. The vials were mixed
thoroughly to ensure complete extraction of the toluene, and the
organic phase was separated from the aqueous phase by centrifu-
gation. The samples were analyzed by gas chromatography mass
spectra (GC/MS).
The generation of libraries T4MO TmoA I100 and TOM TomA3
V106 was described previously [25,26].
2.7. Analytical methods
2.5. Site-directed mutagenesis
Specific analytical methods were developed for separation of
each of the substrates (and respective products) used in this work.
Site-directed mutagenesis at the T4MO tmoA gene was
performed via overlap extension PCR, in
a similar way to
the site-specific saturation mutagenesis. Shortly, three oligonu-
cleotide primer pairs were designed (Table 1) to generate
the desired mutations. For generating the D285Q, D285I and
I100L mutations, the first mutated PCR fragment was ampli-
fied using primers T4MO 285Q Front, T4MO 285I Front and
T4MO 100L Front, respectively, together with T4MOABRear. The
second mutated PCR fragment was amplified using T4MObefEcoRI
primer together with T4MO 285Q Rear, T4MO 285I Rear and
T4MO 100L Rear, respectively. In contrast with the saturation
mutagenesis protocol, here, T4MO plasmids containing a site muta-
tion (T4MO TmoA I100A, I100G or D285S) were used as templates
to create the double mutants I100A/D285Q or D285I, I100G/D285I
and I100L/D285S, respectively. Each pair of fragments was com-
bined during the final reassembly PCR using the outer primers
T4MObefEcoRI Front and T4MOABRear. The assembled PCR frag-
ment was ligated into WT T4MO, after the double digestion of both
vector and insert with EcoRI and AatII, replacing the corresponding
fragment in the original plasmid.
2.7.1. Hydroxylation of PEA
The progress of enzymatic hydroxylation of PEA was measured
by reverse-phase HPLC, whereas the regiospecificity, in addition
to structure verification, was determined by GC/MS analysis. The
detailed analytical methods were the same as we described previ-
ously [25].
2.7.2. Oxidation of methyl-p-tolyl sulfide
Conversion of methyl-p-tolyl sulfide to (S)- or (R)-methyl-
p-tolyl sulfoxide was determined by gas chromatography (GC)
equipped with a chiral column and flame ionization detector, and
the identity of compounds was confirmed by GC/MS. The detailed
analytical methods were the same as described previously [26].
2.7.3. Conversion of styrene to styrene oxide
The epoxidation of styrene was determined in a similar way to
oxidation of methyl-p-tolyl sulfide using the following changes. The
conversion was determined with a GC 6890N instrument (Agilent
Technologies, Santa Clara, CA), using a 30 m × 0.32 mm × 0.25 ␮m
capillary column packed with ␥-cyclodextrin trifluoroacetyl
(Chiraldex G-TA; ASTEC, Bellefonte, PA) and flame ionization detec-
tor. The temperature was programmed as follows: T1 = 60 ◦C;
dT/dt = 20 ◦C/min, T2 = 120 ◦C; dT/dt = 20 ◦C/min, T3=100 ◦C, 18 min,
split ratio 1:3. Under these conditions, the retention times were:
7.40 min for styrene, 9.67 min for t-butyl benzene, 13.10 min for
(S)-styrene oxide and 13.78 min for (R)-styrene oxide.
2.6. Whole-cell enzymatic biotransformations
Whole-cell activity assays were performed as described previ-
ously [25,26]. When using PEA as a substrate, the biotransformation
was carried out in screw-capped 16-ml glass vials containing 2 ml
cells and 0.25 mM substrate (added from a 100 mM stock solution in
ethanol). Twenty-one-millilitre glass vials containing the cell sus-
pensions (2 ml) were used for styrene and methyl-p-tolyl sulfide
oxidation, due to their volatility. Each vial was sealed with a Teflon-
coated septum and aluminum crimp seal, and 1 mM substrates
were added by injection with a syringe (added from a 400 mM and
200 mM stock solutions in ethanol, respectively). All the vials were
shaken at 600 rpm (Vibramax 100, Heidolph, Nurenberg, Germany)
at 30 ◦C.
The reaction was stopped periodically (a vial was sacrificed)
by filtration of the cells or by extraction with 2 ml ethyl acetate
containing 0.5 mM tert butyl benzene (as an internal standard).
Filtration was used for reactions measured by HPLC, whereas,
extraction was used for samples analyzed by GC. The negative
control used in these experiments was TG1/pBS(Kan) (a plasmid
without the monooxygenase). The initial transformation rates were
determined by sampling at 3–20 min intervals during the first
2–5 h. The specific activity (nmol/min/mg protein) was calculated
as the ratio of the initial transformation rate and the total protein
The identity of styrene and styrene oxide was also con-
firmed by GC/MS, equipped with
a capillary HP-5 column
(30 m × 0.32 mm × 0.25 ␮m, Agilent Technologies) and the temper-
ature was programmed as follows: T1 = 50 ◦C; dT/dt = 20 ◦C/min,
T2 =170 ◦C, 18 min, split ratio 1:10. Under these conditions, the
retention times were: 8.07 min for styrene, 12.49 min for t-butyl
benzene and 13.78 min for styrene oxide.
2.7.4. Hydroxylation of toluene
The hydroxylation of toluene was determined using GC/MS
(Agilent Technologies, Santa Clara, CA), equipped with a capil-
lary HP-5 column (30 m × 0.32 mm × 0.25 ␮m). The temperature
was programmed as follows: T1 = 50 ◦C, 4 min; dT/dt = 30 ◦C/min,
T2 = 280 ◦C; 0.5 min, split ratio 1:10. Under these conditions, the
retention times were: 3.7 min for toluene and 6.8 min for t-butyl
benzene.
content. Total protein content was 0.22 [mg protein/ml/OD600 nm
]
for TOM and ToMO and 0.24 [mg protein/ml/OD600 nm] for T4MO
and TpMO [16,17,29]. Activity data reported in this paper are based
on at least two independent results.
Toluene oxidation was determined in a similar way to styrene
and methyl-p-tolyl sulfide. Two millilitre of the cell suspensions
was sealed in a 21-ml vial. 0.89 mM and 3.56 mM toluene was
added to the vial, calculated as if all the substrate was in the liq-
2.8. Screening methods
2.8.1. One-point-biotransformation as a screening method for
improved hydroxylation, sulfoxidation and epoxidation activities
Screening for mutants with improved activity towards PEA,
methyl-p-tolyl sulfide and styrene was performed in a similar way
to the whole-cell biotransformation with some simplifications: one

M. Brouk et al. / Journal of Molecular Catalysis B: Enzymatic 66 (2010) 72–80
75
colony of each mutant was inoculated to 5 ml LB medium supple-
mented with kanamycin (100 ␮g/ml) and grown for 20 h at 37 ◦C
with shaking at 250 rpm. The cells were centrifuged at 6000 × g for
10 min and re-suspended in 1.5 ml PB (0.1 M, pH 7.0). 0.25 mM PEA,
0.5 mM methyl-p-tolyl sulfide or 1 mM styrene was added to 1 ml of
cell suspension, which was than shaken at 600 rpm, room temper-
ature for 24 h (2 h for styrene). The remaining cell suspension was
used to determine the cell density at 600 nm. The reactions were
stopped by centrifugation or extraction as described previously.
In order to ensure the probability (99%) that all 64 possible out-
comes from the single-site random mutagenesis had been sampled,
at least 292 colonies were screened [31].
2.8.2. Screening for improved epoxidation activity using
p-nitrothiophenolate (pNTP) assay
The detection of styrene conversion to styrene oxide was based
on color loss of pNTP in the assay. This colorimetric assay is a mod-
ification of the method described by Tee et al. [32,33].
The screening of highly active enzymes was preformed in a 96-
plate format, using the epMotion 5070 robotic system. The T4MO
D285 library was screened by, initially, picking colonies from glyc-
erol stocks stored in 96-well plates, using a VP 381 library copier
(V&P Scientific, Inc., San Diego, CA). The colonies were transferred
to 96-deep-well polypropylene plates (ABgene; Thermo Fisher Sci-
entific, Epson, United Kingdom) containing 1.2 ml of LB medium
supplemented with 100 ␮g/ml kanamycin. Cells were grown (18 h,
37 ◦C) and precipitated by centrifugation (3000 × g, 10 min), fol-
lowed by resuspension in 0.4 ml of PB (0.1 M, pH 7.0) containing
1 mM styrene. Tenth ml from each well was removed for measur-
ing the cell density at 600 nm. Following 2 h incubation at room
temperature with styrene, the cells were centrifuged (3000 × g,
10 min) and 0.15 ml supernatant was transferred to each well of a
96-well polystyrene plate. Ten microlitre of 12% (w/v) methylated
␤-cyclodextrin (m␤-CD) in isopropanol and 40 ␮l of freshly pre-
pared 2.5 mM pNTP in 0.25 M NaOH were added to each well. The
plate was shaken for 1 h at room temperature and the absorbance
was measured (443 nm) using a microplate reader (Optimax Molec-
ular Devices, Sunny Vale, CA, USA).
Fig. 1. Positions of key residues in the tunnel entrance (D285) and active site (I100)
of the ␣-subunit of WT T4MO. The tunnel location and key residues, selected for
mutagenesis, are indicated in the figure. The Fe atoms are colored as cyan balls.
Helix E was truncated between residues 199 and 205 to enable better visualization
of the active site. The T4MO hydroxylase was visualized using Swiss-PdbViewer
program (PDB code: 3DHG [19]).
catalytic site, it is not expected to influence the catalytic prop-
erties of the enzyme such as regio- or enantio-selectivity. For
exploring this assumption, site-specific saturation mutagenesis
was used to generate a library of T4MO TmoA D285 variants.
The variants’ activity was screened on three different substrates:
hydroxylation of 2-phenylethanol (PEA), oxidation of methyl-p-
tolyl sulfide and epoxidation of styrene. Since T4MO is a complex
multi-component enzyme which requires the recycling of NADH,
whole-cells expressing the different T4MO variants were used in
all assays.
3.2. The ability of WT T4MO and D285 variants to hydroxylate
PEA
In order to validate the accuracy and reproducibility of the
screening method, the supernatant was also analyzed by GC.
Preliminary results (data not shown) demonstrated that the
total decrease in absorption between 1 mM and 0 mM styrene
oxide (parallel to 0–100% conversion) was 0.9. Therefore, mutants
showing at least 0.3 U (Abs443 nm/OD600 nm) decline in absorbance
compared to wild-type were considered positive in the screen and
chosen for further investigation.
PEA (Fig. 2A) is an abundant flavor and fragrance compound
with a rose-like odor. Recently, we reported the use of PEA as a
starting material for the biosynthesis of hydroxytyrosol, a potent
antioxidant found naturally in olives [25].
Out of the four WT TMOs evaluated for their ability to oxidize
PEA, WT T4MO had the lowest oxidation rate (Table 2). Comparison
between the crystal structures of T4MO [19] and ToMO (exhibiting
68% amino acid identity for the ␣-subunits) [34], which oxidized
PEA 2.2-fold faster, shows a narrower channel entrance in T4MO
and with higher negative electrostatic charge. One of the amino
acid residues responsible for this difference is located at position
285. This position in ToMO is occupied with Ser which is smaller
and less charged than Asp situated in T4MO. Based on pervious
results [25], it is assumed that the bulky and hydrophilic nature
of the PEA side chain collides with the polar and bulky residue of
Asp 285 at the tunnel entrance. Therefore, increasing the width of
the entrance may allow higher conversion rates of PEA by T4MO
variants.
Consequently, the T4MO TmoA D285 library was screened for
increased activity towards PEA. Screening of more than 300 vari-
ants was performed to ensure, with a 99% probability, that each
of the 64 possible codons is sampled [6]. The 14 variants which
oxidized PEA at the highest rate were sequenced to give seven dif-
ferent variants: D285P, D285A, D285Y, D285L, D285C, D285I and
D285Q. Unexpectedly, none of them was Ser, the equivalent residue
in ToMO.
3. Results and discussion
3.1. Site-specific saturation mutagenesis at the tunnel entrance of
T4MO hydroxylase
The alpha-subunit of the T4MO hydroxylase contains the diiron
catalytic center for substrate binding and it was reported to be
responsible for the oxidation activity and selectivity [17,18]. Within
the alpha-subunit, a tunnel connects the diiron center to the sur-
face of the protein [19]. The tunnel comprises of a narrow entrance
with high negative electrostatic charge. At the mouth of the tun-
nel lies residue Asp 285 located ∼26 Å away from the active site
(calculated using the Swiss-PDB viewer).
It was hypothesized that the polar and bulky residue of Asp 285
might control the substrate entrance and product efflux to/from
the active site (Fig. 1). Hence, engineering the narrow and polar
tunnel entrance in T4MO may improve its performance towards
non-natural substrates. Given the entrance distance from the

76
M. Brouk et al. / Journal of Molecular Catalysis B: Enzymatic 66 (2010) 72–80
Table 2
Regiospecific and enantioselective oxidation of PEA, methyl-p-tolyl sulfide and styrene by TG1 cells expressing wild-type TOM, ToMO, T4MO and TpMO.
Substrate
Enzymea
PEAb
Methyl-p-tolyl sulfidec
Styrened
Initial oxidation rate
(nmol/min/mg
protein)
Product distribution
Initial oxidation rate
(nmol/min/mg
protein)
Enantio-selectivity
(%ee, pro S)
Initial oxidation
rate (nmol/min/mg
protein)
Enantio-selectivity
(%ee, pro S)
o-Tyr (%)
m-Tyr (%)
p-Tyr (%)
TOM
0.045
0.051
0.023
0.064
64
–
–
36
99
63
64
–
1
37
36
0.50
NA
0.07
NA
11
0.13
0.10
0.21
0.03
−4
78
23
62
ToMO
T4MO
TpMO
NA
−41
NA
–
NA—not active.
a
Results for TOM and ToMO are based on 0.22 mg protein/ml/OD600 nm and for T4MO and TpMO on 0.24 mg protein/ml/OD600 nm
Reported by Brouk and Fishman with initial substrate concentrations of 0.25 mM and standard deviation of less than 9% [25].
Reported by Feingersch et al. with initial substrate concentrations of 1 mM and standard deviation of less than 10% [26].
.
b
c
d
The initial styrene oxidation rates are based on GC/MS analysis with initial substrate concentrations of 1 mM. Enantio-selectivity (%ee) was determined after 2 h by
analyzing the samples by GC/FID equipped with a chiral column. Results represent an average of at least two independent experiments with the absolute measured error
being less than 10%.
Whole-cell biotransformations of the variants obtained from
the screen (Fig. 3) revealed that replacing the polar Asp residue
at position 285 to the apolar residues Ala, Leu or Ile increased the
enzyme activity by 2.7-, 5.4- and 6.6-fold, respectively. Interest-
ingly, the PEA oxidation rate increased with the hydrophobicity
of the amino acid side chain. Moreover, replacing the negatively
charged Asp 285 with uncharged residues (at pH 7), as Tyr, Pro,
Cys or Gln, resulted with increased activity of 3-, 3.3-, 4- and 10.5-
fold, respectively. Variant D285S, which is the analogous residue
in ToMO (and was not found in the screen), had an activity of
0.017 0.002 nmol/min/mg protein which is only 70% of WT activ-
ity. Despite the increase in oxidation rate for the different D285
variants, there was no change in regioselectivity (Fig. 3B). Sur-
prisingly, examining the activity of the improved D285 variants
on m- and p-tyrosol (Fig. 2A) as substrates (data not shown)
revealed that unlike WT T4MO, they were all capable of oxidizing
the tyrosol isomers to form hydroxytyrosol, but at very low con-
version rates (with the exception of D285P). Likewise, employing
o-tyrosol as a substrate resulted in the formation of two substi-
tuted dihydroxy benzenes, namely, 2,3-dihydroxyphenyl ethanol
and 2,5-dihydroxyphenyl ethanol. These results demonstrate the
ability of T4MO variants to form hydroxytyrosol, and emphasize
the difficulty of polar and bulky substances to enter the tunnel and
reach the active site.
Similarly to our results, position TouA E214 in ToMO, located in
the tunnel entrance, facing residue 285, was found by Vardar et al.
[9] to influence rate but not regioselectivity of o- and p-nitrophenol
hydroxylation. In was shown that the size of the residue at position
214 is most likely to be the factor influencing the oxidation rate,
reaching up to 15-fold improvement for p-nitrophenol oxidation
by ToMO E214G. The regioselectivity of all E214 variants tested
remained the same.
From the results presented in Fig. 3 and the low activity of D285S
it can be concluded that the acidic character of Asp 285 in the tunnel
entrance has a higher impact than its size, on the enzyme activ-
ity toward PEA. As expected, changing the entrance of the tunnel
did not affect the regioselectivity of the enzyme. The ability of the
improved D285 variants to oxidize tyrosol isomers provides further
evidence for the impact of the polar acidic residue of Asp 285 on
the entrance of polar substrates through the tunnel.
3.3. Sulfoxidation of methyl-p-tolyl sulfide by WT and D285
variants
Chiral sulfoxides have been in focus of attention in recent years,
especially in the chemical and pharmaceutical industries [35,36].
In recent work performed in our lab, TMOs were evaluated for
their sulfoxidation activity using two model substrates, thioanisole
and methyl-p-tolyl sulfide [26]. Although all four WT TMOs were
able to oxidize thioanisole, only TOM and T4MO were active on
methyl-p-tolyl sulfide (Table 2). It was shown that the presence
of a side residue on the aromatic ring (as in methyl-p-tolyl sul-
fide) reduces dramatically the activity and the selectivity of TMOs
[26]. For instance, the oxidation rate of methyl-p-tolyl sulfide by
WT T4MO was one order of magnitude lower than the one on
thioanisole, which is 1 Å shorter. Therefore, in the purpose of
improving the enzyme activity towards enlarged substrates, the
T4MO D285 library was screened on methyl-p-tolyl sulfide.
The most active variants were sequenced to give a total of 5
different mutations: D285I, D285V, D285L, D285R and D285Q. The
best performing variant obtained following whole-cell biotrans-
formation was D285I (Fig. 4). It had an 8-fold improvement in
oxidation rate. Substitution of the polar and acidic residue Asp, to
the apolar and aliphatic residues Val and Leu lead to increased enzy-
Fig. 2. The different reactions used in this study. (A) Regioselective hydroxylation
of PEA to form p-tyrosol or m-tyrosol which could be further oxidized to form
hydroxytyrosol, (B) enantioselective sulfoxidation of methyl-p-tolyl sulfide to (S)-
or (R)-methyl p-tolyl sulfoxide, (C) enantioselective epoxidation of styrene to (S)-
or (R)-styrene oxide.

M. Brouk et al. / Journal of Molecular Catalysis B: Enzymatic 66 (2010) 72–80
77
Fig. 3. Initial oxidation rate of PEA (A) by T4MO D285 variants and the product distribution (B). The activity was determined via HPLC analysis over a 2-h time period and
regiospecificity was determined via GC/MS analysis after 24 h, with initial PEA concentration of 0.25 mM.
Fig. 4. Initial oxidation rate of methyl-p-tolyl sulfide (A) by T4MO D285 variants and the enantiomeric excess (proR) (B). The activity and the enantio-selectivity were
determined via GC analysis over 2 and 24 h time period, respectively with initial substrate concentration of 1 mM.
matic activity of 3.7- and 2-fold, respectively. Moreover, replacing
the acidic Asp with a polar but basic residue Arg, or uncharged
(at pH 7) as Gln, resulted in increased activity of 3.2- and 1.7-fold,
respectively. It should be noted that both the orientation and the
size of the changed residue should be taken into consideration.
As hypothesized, the alteration at position D285 lead to increased
activity, but did not change the enantio-selectivity of the variants
(Fig. 4B).
For further improvement of the T4MO activity, the randomized
T4MO TmoA D285 library was screened using the pNTP assay
which is based on the discoloration of pNTP in reaction with
styrene oxide [32,33]. Using this screening method, 352 T4MO
D285 variants were screened and those that showed an absorbance
of 0.3 U below the WT enzyme were screened again in a one-
point-biotransformation procedure using GC analysis. The most
active variant had a smaller and slightly charged residue, Ser, at
position 285. T4MO D285S had an improved epoxidation rate of
0.36 nmol/min/mg protein which is 1.7-fold higher than WT T4MO.
As was expected, the enantio-selectivity of D285S was similar to
that of WT (25% ee, pro S).
3.4. Epoxidation of styrene by WT and the D285 variants
Chiral epoxides are important intermediates in the chemical
synthesis of optically active pharmaceuticals, polymers and agro-
chemicals. Enantiopure styrene oxide, an epoxide derived from
styrene, may be used to synthesize enantiopure pharmaceuti-
cals such as anti-diabetic agents and dopamine agonists [37]. To
date only a single publication dealing with epoxidation reactions
by TMOs has been published. In their article, McClay et al. [38]
reported the oxidation of short-chain alkenes to their correspond-
ing epoxides by several toluene monooxygenases, but styrene
was not included in that work. However, other monooxygenases
such as styrene monooxygenase, cytochrome P450 and p-cymene
monooxygenase have been reported to oxidize styrene selectively
[39–41].
3.5. The effect of mutating the active site vs. the tunnel entrance
As hypothesized and demonstrated by the results, mutating a
residue located in the entrance of the tunnel leading to the active
site, enhanced the oxidation rate of all three substrates but had no
effect on the regio- or enantio-selectivity. We further investigated
this in comparison with mutations in the vicinity of the active site.
Recently, the influence of position I100 in T4MO on PEA oxi-
dation rate and regioselectivity was published [25]. This residue
(Fig. 1) was proposed to act as a gate to the diiron active site pocket
[15,17,18,26]. It was discovered that increasing the size of the active
site pocket and enlarging its entrance, by generating mutations
at position I100, improves the oxidation rate and regioselectiv-
ity. Furthermore, it enabled hydroxytyrosol formation, which WT
T4MO was not capable of producing [25]. Consequently, here we
Initially, four WT TMOs were evaluated for their ability to
oxidize styrene to styrene-oxide (Fig. 2). The results clearly indi-
cated that T4MO has the highest activity rate of 0.21 nmol/min/mg
protein with low enantio-selectivity of 23% ee (pro S) (Table 2).

78
M. Brouk et al. / Journal of Molecular Catalysis B: Enzymatic 66 (2010) 72–80
Table 3
Relative activity and selectivity of T4MO variants mutated in the entrance to the tunnel and/or the active site.
Activity tested
Substrate
T4MO variant
Mutation site
Relative activitya
Product distribution
Enantio-selectivity
(pro S) (%)
o-Tyr (%)
m-Tyr (%)
p-Tyr (%)
Hydroxylation
PEA
WT
D285Q
D285I
I100A
I100A/D285Q
I100A/D285I
–
1
–
–
–
–
–
–
63
57
58
92
91
93
37
43
42
8
9
7
Tunnel
Tunnel
Active site
Both
11.7
7.4
35
85
52
Both
Sulfoxidation
Methyl-p-tolyl sulfide
WT
D285I
I100A
I100G
I100A/D285I
I100G/D285I
–
1
8.3
8.6
11.0
12.6
14.1
−41
−44
79
77
80
Tunnel
Active site
Active site
Both
Both
82
Epoxidation
Styrene
WT
D285S
I100L
–
1
23
25
58
52
Tunnel
Active site
Both
1.7
0.9
1.4
I100L/D285S
a
Relative activity is calculated as the initial oxidation rate normalized to WT T4MO. WT is designated as 1 with an activity of 0.023 0.001, 0.07 0.01 and
0.21 0.02 nmol/min/mg protein for PEA, methyl-p-tolyl sulfide and styrene oxidation, respectivly. Reaction conditions were as described in the legend of Table 2. Results
represent an average of at least two independent experiments with the absolute measured error being less than 10%.
examined the effect of combining efficient mutations at the tun-
nel and the active site. Using site-directed mutagenesis, the D285Q
and D285I substitutions were introduced into the plasmid of I100A
variant, which had the highest activity towards PEA among all the
I100 variants tested. While variant D285Q had improved activity of
11.7-fold with similar regioselectivity as WT, and variant I100A had
improved oxidation rate of 35-fold and preferred regioselectivity
towards the meta position (92%), the double mutant I100A/D285Q
showed improved activity of 85-fold and a regioselectivity similar
to the I100A variant (Table 3). The same tendency was observed for
the I100A/D285I variant which had improved activity of 52-fold
and a regioselectivity of 93% towards the meta position. It can be
concluded that combining mutations in the active site and in the
tunnel entrance resulted in a synergistic effect on PEA oxidation
rate, while the regioselectivity was influenced only by the active
site mutation.
and I100G, oxidized methyl-p-tolyl sulfide 8.6- and 11-fold faster
than WT with enantio-selectivity of 79% and 77% pro-S, respec-
tively (Table 3). D285I, the most active mutant found from the D285
library, had improved activity of 8.3-fold and enantio-selectivity
similar to WT (44% pro-R). Generating double mutants in posi-
tions I100 and D285 resulted in both higher sulfoxidation rate and
high pro-S enantio-selectivity (Table 3). Variant I100A/D285I had
improved activity of 12.6-fold with enantio-selectivity of 80% pro-
S. Likewise, variant I100G/D285I had improved activity of 14.1-fold
with enantio-selectivity of 82% pro-S.
Position I100 was also tested for its influence on epoxidation
activity. A library of T4MO TmoA I100 variants [26] was screened
for improved activity towards styrene. The most active mutant
was found to be T4MO I100L. Further investigation of the mutant
activity showed a similar reaction rate as WT, but with improved
enantio-selectivity of 58% pro-S vs. 25% for WT (Table 3). The sub-
stitution of Asp to Ser at position 285 in the T4MO I100L plasmid,
by site-directed mutagenesis, resulted in increased activity and
enatioselectivity of the double mutant. A 1.4-fold improvement in
the epoxidation activity of variant I100L/D285S was gained by the
D285S mutation while enatioselectivity of 52% pro-S was gained
from the I100L substitution (Table 3).
The importance of position I100 in T4MO was also reported
for sulfoxidation [26]. The activity towards methyl-p-tolyl sulfide
was improved significantly by all the I100 variants tested. While
WT T4MO had a very low transformation rate with pro-R enantio-
selectivity, all the I100 mutants changed their enantio preference
to the S configuration [26]. The most improved variants, I100A
Fig. 5. Initial epoxidation rate of styrene (A) and enantio-selectivity towards the R-enantiomer (B) by WT TOM and TomA3 V106 mutants. Initial substrate concentration
was 1 mM styrene.

M. Brouk et al. / Journal of Molecular Catalysis B: Enzymatic 66 (2010) 72–80
79
Table 4
Relative activity on toluene, PEA, methyl-p-tolyl sulfide and styrene, by TG1 cells expressing T4MO variants.
Relative activity on different substrates
Initial substrate concentration
T4MO variant
0.25 mM
Tolueneb
1 mM
PEAa
Tolueneb
Methyl-p-tolyl sulfidea
Styrenea
WT
1
1
1
1
1
D285Q
D285S
I100A/D285Q
4.6
NE
5.4
11.7
0.7
85
1.6
5.8
NE
1.7
NE
NE
NE
1.7
NE
NE, not examined.
a
Relative activity is calculated as the initial oxidation rate normalized to WT T4MO. WT is designated as 1 with an activity of 0.023 0.001, 0.07 0.01 and
0.21 0.02 nmol/min/mg protein for PEA, methyl-p-tolyl sulfide and styrene oxidation, respectivly. Reaction conditions were as described in the legend of Table 2.
b
Relative activity is calculated as the initial oxidation rate normalized to WT T4MO. WT is designated as 1 with an activity of 1.67 0.01 and 6.6 0.6 nmol/min/mg protein
for initial toluene concentrations of 0.25 mM and 1 mM, respectivly, based on a Henry’s law constant of 0.27 (0.89 mM and 3.56 mM, respectively, added if all the toluene in
the liquid phase). Results represent an average of at least two independent experiments with the absolute measured error being less than 10%.
Residue TomA3 V106 in TOM, the analogous position of T4MO
TmoA I100, was shown to have a major influence on rate and
specificity [6]. Previously, we reported the effect of mutations at
position TOM TomA3 V106 on the enzyme activity towards both
PEA and methyl-p-tolyl sulfide [25,26]. It was found that muta-
tions at position V106 can improve the PEA hydroxylation rate,
compared to WT TOM, up to 39-fold (for V106E variant), and
alter the regiospecificity of the enzyme dramatically [25]. Like-
wise, mutating position V106 affected the sulfoxidation activity
and enantio-selectivity towards methyl-p-tolyl sulfide as a sub-
strate [26]. Screening the TOM TomA3 V106 library for improved
epoxidation activity towards styrene, resulted in five improved
variants: V106C, V106P, V106N, V106S and V106A. Changing the
Val residue at position 106, which is part of the hydrophobic cav-
ity surrounding the diiron active site, affected both activity and
enantio-selectivity of styrene epoxidation (Fig. 5). While WT TOM
oxidized styrene with a rate of 0.13 nmol/min/mg protein and low
enantio-selectivity of only 4% pro-R; a 17-fold improvement in
rate was observed for V106S variant (33% ee, pro-R) and 14-fold
improvement in enantio-selectivity was observed for the V106P
variant (9-fold improvement in epoxidation rate). Additionally,
mutants V106N, V106A and V106C displayed improved activity of
3.7-, 4.1- and 7.6-fold compared to WT TOM with pro-R enatios-
electivity of 19%, 36% and 36%, respectively. These results further
support the major influence that a residue in the entrance to the
active site pocket has on enzyme activity and selectivity.
Altering residues situated in the entrance to the active site
pocket, were previously reported to influence the oxidation rate
of the natural substrate toluene. Fishman et al. reported an initial
toluene oxidation rate of 4.4 nmol/min/mg protein by WT T4MO
(with toluene concentration of 0.09 mM) while a 1.6- and 1.5-fold
improvement was measured for variants I100A and I100S, respec-
tively [18]. Likewise, Rui et al. reported an increased activity on
toluene of different TOM V106 variants with up to 2.1-fold improve-
ment by V106A (with toluene concentration of 0.09 mM) [31]. In
the present work, we further examined the effect of amino acid
substitutions in the tunnel entrance, in addition to the active site
entrance, on toluene oxidation. The initial oxidation rate by WT
T4MO was 1.67 and 6.6 nmol/min/mg protein for 0.25 mM and
1 mM toluene, respectively (Table 4). Expectedly, the toluene oxi-
dation rates were 1–2 orders of magnitude higher than those of
the unnatural substrates, PEA, methyl-p-tolyl sulfide and styrene.
It was found that D285Q, which oxidized PEA (0.25 mM) 11.7-fold
faster than WT, had an improved activity of 4.6-fold on toluene as
well, with an initial oxidation rate of 7.7 nmol/min/mg protein. The
initial oxidation rate of 1 mM toluene by variant D285Q was 1.6-
fold higher than WT similarly to the improvement in oxidation of
1 mM methyl-p-tolyl sulfide. Combining the beneficial mutations,
D285Q and I100A, which led to 85-fold improvement in the oxi-
dation of 0.25 mM PEA, resulted with only 5.4-fold improvement
in toluene oxidation. Thus, the mutations were more valuable for
increasing the rate of the larger substrate, PEA. The most significant
improvement in activity on toluene was obtained by variant D285S,
which exhibited a 1.7-fold improvement in the oxidation of 1 mM
styrene. D285S oxidized toluene 5.8-fold faster than WT, reaching
an initial oxidation rate of 38.4 nmol/min/mg protein. In sum, all
of the variants examined showed an increase in toluene oxidation
although they were screened for different substrates.
In the present work, in order to perform saturation muta-
genesis at position T4MO TmoA D285, the conventional degen-
erate primers, consisting NNN codons, were used (N: ade-
nine/cytosine/guanine/thymine). Recently, Reetz et al. introduced
the use of NDT codon degeneracy (D: adenine/guanine/thymine;
T: thymine) for constructing smaller libraries [42]. This option
involves 12 codons and reduces the number of amino acid replace-
ments to 12 (Phe, Leu, Ile, Val, Tyr, His, Asn, Asp, Cys, Arg, Ser,
Gly). Although the NDT codon degeneracy requires the screening
of only 34 clones in the case of one residue site [42], screening of
T4MO TmoA D285 NDT-library would have hypothetically resulted
with the finding of D285Y, D285L, D285C, D285I, D285V, D285S and
D285R substitutions, but would have missed the D285Q, D285P and
D285A substitutions, that were found to be beneficial. Likewise,
constructing the T4MO TmoA I100 and TOM TomA3 V106 libraries
using NDT primers would have led to the absence of T4MO I100A
and TOM V106A variants that are reported here for their improved
activity. Indeed, NDT codon degeneracy ensures a balanced mix of
amino acids in terms of structural and electrostatic characteristics,
but still, similar amino acids affect differently enzyme activity. For
instance, structurally similar amino acids as Cys and Ser had an
opposite effect on PEA oxidation when situated at position 285.
While the D285C improved the PEA oxidation rate by 4-fold, the
D285S substitution resulted with decreased activity on PEA. Simi-
larly, substituting Asp to Ile resulted with an 8-fold improvement
for methyl-p-tolyl sulfide oxidation rate, while substituting to the
structurally similar Leu improved the enzyme activity by only 2-
fold. Similar amino acids may also have a different affect on the
enantio-selectivity of the enzyme, as in the case of T4MO I100L. A
substitution of Ile with Leu improved the enantio-selectivity from
25% to 58% ee.
4. Conclusions
The results presented in this study, have shed some light on
the role played by a key residue in the tunnel entrance, compared
with the active site pocket entrance. The amino acid residue (D285)
located at the entrance of the tunnel leading to the diiron center was
mutated with the intention of accelerating the substrate/product
flow to/from the active site. The effect of this modification on the
catalytic properties of T4MO was determined by screening for three
different activities: (1) hydroxylation of an aromatic substrate sub-

80
M. Brouk et al. / Journal of Molecular Catalysis B: Enzymatic 66 (2010) 72–80
stituted with a polar and bulky residue (PEA), (2) sulfoxidation on
the side chain of methyl-p-tolyl sulfide and (3) epoxidation on the
double bond side chain of styrene (Fig. 2). Due to the distant location
of D285 from the active site itself, it was hypothesized and demon-
strated that alterationsat this positionaffecttheactivity, but not the
regio- or enantio-selectivity. In contrast, mutating a residue that is
part of the hydrophobic gate in the active site entrance, i.e. position
T4MO I100 and TOM V106, affected both activity and selectivity.
Mutating the entrance of the active site may have two implica-
tions. First, enlarging the hydrophobic gate allows easier access
to/from the active site, which results in higher oxidation rates.
Second, expanding the hydrophobic cavity surrounding the diiron
binding site enables better or different alignment of the substrate
in the active site thus influencing the regio- or enantio-selectivity.
Combining the effects by mutating both “gates”, to the active site
along with the tunnel entrance, results in obtaining an additive or
even synergistic outcome on the enzyme activity.
References
[1] A. Schmid, J.S. Dordick, B. Hauer, A. Kiener, M. Wubbolts, B. Witholt, Nature 409
(2001) 258–268.
[2] T.W. Johannes, H. Zhao, Curr. Opin. Microbiol. 9 (2006) 261–267.
[3] M. Leisola, O. Turunen, Appl. Microbiol. Biotechnol. 75 (2007) 1225–1232.
[4] N.M. Antikainen, S.F. Martin, Bioorg. Med. Chem. 13 (2005) 2701–2716.
[5] K.L. Morley, R.J. Kazlauskas, Trends Biotechnol. 23 (2005) 231–237.
[6] A. Fishman, Y. Tao, G. Vardar, L. Rui, T.K. Wood, in: J.-L. Ramos, R.C. Levesque
(Eds.), Pseudomonas, Springer, US, 2006, pp. 237–286.
[7] J. Sylvestre, H. Chautard, F. Cedrone, M. Delcourt, Org. Process Res. Dev. 10
(2006) 562–571.
[8] S.B. Rubin-Pitel, H. Zhao, Comb.Chem. High Throughput Screen. 9 (2006)
247–257.
[9] G. Vardar, T.K. Wood, J. Bacteriol. 187 (2005) 1511–1514.
[10] G. Vardar, K. Ryu, T.K. Wood, J. Biotechnol. 115 (2005) 145–156.
[11] J. Schmitt, S. Brocca, R.D. Schmid, J. Pleiss, Protein Eng. 15 (2002) 595–601.
[12] I. Capila, Y. Wu, D.W. Rethwisch, A. Matte, M. Cygler, R.J. Linhardt, Biochim.
Biophys. Acta 1597 (2002) 260–270.
[13] R. Chaloupkova, J. Sykorova, Z. Prokop, A. Jesenska, M. Monincova, M.
Pavlova, M. Tsuda, Y. Nagata, J. Damborsky, J. Biol. Chem. 278 (2003) 52622–
52628.
With the intention of rationally designing an enzyme, it is
important to consider, not only the enzyme structure and func-
tion, but also the substrate characteristics. WT T4MO oxidized
the natural substrate, toluene, having a small and hydrophobic
side chain, 1–2 orders of magnitude faster than other structurally
related substrates, as PEA, methyl-p-tolyl sulfide and styrene.
Besides toluene, the fastest oxidation rate was obtained for 1 mM
styrene (0.21 nmol/min/mg protein), which has a relatively small
and rigid allylic side chain, while the oxidation rate of 1 mM PEA,
with the polar and bulky side chain, was the slowest with a rate
of 0.06 nmol/min/mg protein. For the improvement of activity
towards PEA, the removal of the negative electrostatic charge in
the tunnel entrance was more important, whereas for styrene, a
higher impact was achieved by enlarging the width of the tunnel
entrance. In addition to the size of the amino acid side chain, it
is important to consider its rotamer configuration. Generally, the
improvement obtained by mutating the tunnel entrance was lower
for styrene oxidation compared with PEA and methyl-p-tolyl sul-
fide (Table 3). Similarly, altering the active site entrance affected
mostly the enzyme activity toward PEA, with improvement of up
to 35-fold by T4MO I100A, compared with improvement of up to
11-fold for methyl-p-tolyl sulfide oxidation (by T4MO I100G) and
no activation for styrene (by T4MO I100L). Interestingly, while dif-
ferent substitutions, as D285I and D285Q, were beneficial for the
oxidation of both PEA and methyl-p-tolyl sulfide, they affected neg-
atively the enzyme activity towards styrene. Likewise, substitution
D285S, which improved the styrene oxidation rate, had a negative
effect on the enzyme activity towards PEA. The results demon-
strate that altering a key residue can be beneficial for improving
the enzyme’s activity towards diverse substrates, but the extent of
the improvement is different for each one.
[14] M. Pavlova, M. Klvana, Z. Prokop, R. Chaloupkova, P. Banas, M. Otyepka, R.C.
Wade, M. Tsuda, Y. Nagata, J. Damborsky, Nat. Chem. Biol. 5 (2009) 727–
733.
[15] K.A. Canada, S. Iwashita, H. Shim, T.K. Wood, J. Bacteriol. 184 (2002) 344–
349.
[16] G. Vardar, T.K. Wood, Appl. Environ. Microbiol. 70 (2004) 3253–3262.
[17] Y. Tao, A. Fishman, W.E. Bentley, T.K. Wood, J. Bacteriol. 186 (2004) 4705–
4713.
[18] A. Fishman, Y. Tao, W.E. Bentley, T.K. Wood, Biotechnol. Bioeng. 87 (2004) 779–
790.
[19] L.J. Bailey, J.G. McCoy, G.N. Phillips, B.G. Fox, Proc. Natl. Acad. Sci. U.S.A. 105
(2008) 19194–19198.
[20] J.G. Leahy, P.J. Batchelor, S.M. Morcomb, FEMS Microbiol. Rev. 27 (2003)
449–479.
[21] K.H. Mitchell, J.M. Studts, B.G. Fox, Biochemistry 41 (2002) 3176–3188.
[22] J. Sambbrook, D.W. Russell, Molecular Cloning: A Laboratory Manual, third ed.,
Cold Spring Harbor Laboratory Press, New York, 2001.
[23] Y. Tao, A. Fishman, W.E. Bentley, T.K. Wood, Appl. Environ. Microbiol. 70 (2004)
3814–3820.
[24] G. Vardar, T.K. Wood, Appl. Microbiol. Biotechnol. 68 (2005) 510–517.
[25] M. Brouk, A. Fishman, Food Chem. 116 (2009) 114–121.
[26] R. Feingersch, J. Shainsky, T.K. Wood, A. Fishman, Appl. Environ. Microbiol. 74
(2008) 1555–1566.
[27] R.W. Eaton, P.J. Chapman, J. Bacteriol. 177 (1995) 6983–6988.
[28] L. Rui, K.F. Reardon, T.K. Wood, Appl. Microbiol. Biotechnol. 66 (2005) 422–
429.
[29] A. Fishman, Y. Tao, L. Rui, T.K. Wood, J. Biol. Chem. 280 (2005) 506–514.
[30] J. Dolfing, A.J. van den Wijngaard, D.B. Janssen, Biodegradation
261–282.
4 (1993)
[31] L. Rui, Y.M. Kwon, A. Fishman, K.F. Reardon, T.K. Wood, Appl. Environ. Microbiol.
70 (2004) 3246–3252.
[32] K. Tee, O. Dmytrenko, K. Otto, A. Schmid, U. Schwaneberg, J. Mol. Catal. B:
Enzyme 50 (2008) 121–127.
[33] K. Tee, U. Schwaneberg, Comb. Chem. High Throughput Screen. 10 (2007)
197–217.
[34] M.H. Sazinsky, J. Bard, A. Di Donato, S.J. Lippard, J. Biol. Chem. 279 (2004)
30600–30610.
[35] R. Patel, Coord. Chem. Rev. 252 (2008) 659–701.
[36] R. Bentley, Chem. Soc. Rev. 34 (2005) 609–624.
[37] A. Archelas, R. Furstoss, Annu. Rev. Microbiol. 51 (1997) 491–525.
[38] K. McClay, B. Fox, R. Steffan, Appl. Environ. Microbiol. 66 (2000) 1877–1882.
[39] S. Panke, M. Held, M.G. Wubbolts, B. Witholt, A. Schmid, Biotechnol. Bioeng. 80
(2002) 33–41.
[40] E. Farinas, M. Alcalde, F. Arnold, Tetrahedron 60 (2004) 525–528.
[41] T. Nishio, A. Patel, Y. Wang, P.C. Lau, Appl. Microbiol. Biotechnol. 55 (2001)
321–325.
In summary, the amino acid replacement at “hot spots” is sub-
strate dependant. In addition, the location of the key residue in
the protein architecture will determine the nature of the change
in activity. Mutations at the tunnel entrance will most likely affect
the activity rate whereas altering residues near the active site can
affect both rate and selectivity.
[42] M.T. Reetz, D. Kahakeaw, R. Lohmer, Chembiochem 9 (2008) 1797–1804.
Acknowledgment
We thank Prof. Noam Adir for help with analysis of the T4MO
crystal structure.