Aromatic stacking interactions govern catalysis in aryl-alcohol oxidase

Article abstract of DOI:10.1111/febs.13221

Aryl-alcohol oxidase (AAO, EC 1.1.3.7) generates H₂O₂ for lignin degradation at the expense of benzylic and other π system-containing primary alcohols, which are oxidized to the corresponding aldehydes. Ligand diffusion studies on Pleurotus eryngii AAO showed a T-shaped stacking interaction between the Tyr92 side chain and the alcohol substrate at the catalytically competent position for concerted hydride and proton transfers. Bi-substrate kinetics analysis revealed that reactions with 3-chloro- or 3-fluorobenzyl alcohols (halogen substituents) proceed via a ping-pong mechanism. However, mono- and dimethoxylated substituents (in 4-methoxybenzyl and 3,4-dimethoxybenzyl alcohols) altered the mechanism and a ternary complex was formed. Electron-withdrawing substituents resulted in lower quantum mechanics stacking energies between aldehyde and the tyrosine side chain, contributing to product release, in agreement with the ping-pong mechanism observed in 3-chloro- and 3-fluorobenzyl alcohol kinetics analysis. In contrast, the higher stacking energies when electron donor substituents are present result in reaction of O₂ with the flavin through a ternary complex, in agreement with the kinetics of methoxylated alcohols. The contribution of Tyr92 to the AAO reaction mechanism was investigated by calculation of stacking interaction energies and site-directed mutagenesis. Replacement of Tyr92 by phenylalanine does not alter the AAO kinetic constants (on 4-methoxybenzyl alcohol), most probably because the stacking interaction is still possible. However, introduction of a tryptophan residue at this position strongly reduced the affinity for the substrate (i.e. the pre-steady state K_d and steady-state K_m increase by 150-fold and 75-fold, respectively), and therefore the steady-state catalytic efficiency, suggesting that proper stacking is impossible with this bulky residue. The above results confirm the role of Tyr92 in substrate binding, thus governing the kinetic mechanism in AAO.

Full text of DOI:10.1111/febs.13221

Aromatic stacking interactions govern catalysis in
aryl-alcohol oxidase
1
2,†
3
2
1,‡
ꢀ
ꢀ
Patricia Ferreira , Aitor Hernandez-Ortega , Fatima Lucas , Juan Carro , Beatriz Herguedas
,
2
1
Kenneth W. Borrelli3, Victor Guallar3,4, Angel T. Martınez and Milagros Medina
ꢀ
ꢀ
1 Departamento de Bioquımica y Biologıa Molecular y Celular, Facultad de Ciencias, Universidad de Zaragoza, and Instituto de
ꢀ
ꢀ
Biocomputacion y Fısica de Sistemas Complejos, Zaragoza, Spain
ꢀ
ꢀ
2 Centro de Investigaciones Biologicas, Consejo Superior de Investigaciones Cient´ıficas, Madrid, Spain
3 Joint Barcelona Supercomputing Center-Centre for Genomic Regulation-Institute for Research in Biomedicine Research Program in
Computational Biology, Barcelona Supercomputing Center, Barcelona, Spain
ꢀ
4 Institucio Catalana de Recerca i Estudis Avancßats, Barcelona, Spain
Keywords
Aryl-alcohol oxidase (AAO, EC 1.1.3.7) generates H2O2 for lignin degrada-
tion at the expense of benzylic and other p system-containing primary alco-
hols, which are oxidized to the corresponding aldehydes. Ligand diffusion
studies on Pleurotus eryngii AAO showed a T-shaped stacking interaction
between the Tyr92 side chain and the alcohol substrate at the catalytically
competent position for concerted hydride and proton transfers. Bi-substrate
kinetics analysis revealed that reactions with 3-chloro- or 3-fluorobenzyl
alcohols (halogen substituents) proceed via a ping–pong mechanism. How-
ever, mono- and dimethoxylated substituents (in 4-methoxybenzyl and 3,4-
dimethoxybenzyl alcohols) altered the mechanism and a ternary complex
was formed. Electron-withdrawing substituents resulted in lower quantum
mechanics stacking energies between aldehyde and the tyrosine side chain,
contributing to product release, in agreement with the ping–pong mecha-
nism observed in 3-chloro- and 3-fluorobenzyl alcohol kinetics analysis. In
contrast, the higher stacking energies when electron donor substituents are
present result in reaction of O2 with the flavin through a ternary complex,
in agreement with the kinetics of methoxylated alcohols. The contribu-
tion of Tyr92 to the AAO reaction mechanism was investigated by calcula-
tion of stacking interaction energies and site-directed mutagenesis.
Replacement of Tyr92 by phenylalanine does not alter the AAO kinetic
constants (on 4-methoxybenzyl alcohol), most probably because the
stacking interaction is still possible. However, introduction of a tryptophan
residue at this position strongly reduced the affinity for the substrate (i.e.
the pre-steady state Kd and steady-state Km increase by 150-fold and
75-fold, respectively), and therefore the steady-state catalytic efficiency,
suggesting that proper stacking is impossible with this bulky residue. The
above results confirm the role of Tyr92 in substrate binding, thus govern-
ing the kinetic mechanism in AAO.
aromatic stacking; aryl-alcohol oxidase;
catalytic mechanism; GMC oxidoreductases;
steady-state and pre-steady state kinetics
Correspondence
ꢀ
P. Ferreira, Departamento de Bioquımica
ꢀ
y Biologıa Molecular y Celular, Facultad
de Ciencias, Universidad de Zaragoza,
Pedro Cerbuna 12, 50009 Zaragoza, Spain
Fax: +34976762123
Tel.: +34876553774
E-mail: ferreira@unizar.es
Present addresses:
†Manchester Institute of Biotechnology,
University of Manchester, Manchester, UK
‡Medical Research Council Laboratory of
Molecular Biology, Cambridge, UK
(Received 29 October 2014, revised 10
January 2015, accepted 28 January 2015)
doi:10.1111/febs.13221
Abbreviations
AAO, aryl-alcohol oxidase; AAOred, reduced AAO; CTC, charge transfer complex; GMC, glucose/methanol/choline oxidase (oxidoreductase
superfamily); KIE, kinetic isotope effect; MM, molecular mechanics; QM, quantum mechanics.
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS
3091

Aromatic stacking interactions in aryl-alcohol oxidase
P. Ferreira et al.
Introduction
Aryl-alcohol oxidases (AAO, EC 1.1.3.7) are extracel-
lular flavoproteins that typically catalyze the oxidative
dehydrogenation of polyunsaturated alcohols using
molecular oxygen as the final electron acceptor and
producing hydrogen peroxide [1]. This activity was first
reported in Trametes versicolor [2], and was later
described in other white-rot basidiomycetes that are
mainly responsible for lignin degradation in nature,
such as Pleurotus eryngii, Bjerkandera adusta and
Phanerochaete chrysosporium [3–7]. Lignin removal is a
rate-limiting step for carbon recycling in land ecosys-
tems, also playing a central role in paper pulp manu-
facture and the production of chemicals and biofuels
from renewable lignocellulosic biomass [8]. The physio-
logical role of AAO in wood-rotting basidiomycetes is
to provide a continuous supply of extracellular H2O2,
which is required as a substrate for ligninolytic perox-
idases (in white-rot fungi) and as plant polysaccharides
(in brown-rot fungi) [1,9].
AAO from Pleurotus eryngii has been intensively
investigated [1,10–13]. This enzyme contains one mole-
cule of non-covalently bound FAD that acts as a two-
electron acceptor during oxidation of a wide range of
benzylic and other aromatic and aliphatic polyunsatu-
rated primary alcohols (reductive half-reaction). The
oxidative half-reaction, whereby the FAD hydroqui-
none is oxidized by O2, closes the catalytic cycle [12–
14]. AAO also oxidizes some aromatic aldehydes via
their hydrated (gem-diol) forms, suggesting similar
mechanisms for alcohol and aldehyde oxidation [15].
The AAO crystal structure (PDB ID 3FIM) con-
firmed that it shares similar fold topology with other
members of the glucose/methanol/choline oxidase
(GMC) oxidoreductase superfamily [16], and a role in
catalysis of two conserved active-site residues (His502
and His546 in Pleurotus eryngii AAO) was postulated
[17]. Diffusion simulations of the reducing substrate
towards the AAO active site, previously identified in
the crystal structure, suggested an entrance channel
next to the Gln395–Thr406 loop [18]. These studies
additionally suggested that formation of the catalyti-
cally competent complex, requires significant displace-
ments of side chains in the active-site environment
(including Phe397), as well as the p–p stacking interac-
tion with Tyr92 (Fig. 1) [18]. In contrast, the oxidative
substrate migration did not require the above struc-
tural reorganization, even though Phe501 was shown
to be essential for correct O2 positioning [12].
Fig. 1. Detail of the AAO active site with bound alcohol substrate.
Position of 4-methoxybenzyl alcohol after migration simulations to
the AAO active site using PELE software [50] (on the 3FIM crystal
structure). The alcohol is located in the surroundings of two
conserved histidines, three aromatic residues and the FAD
isoalloxazine ring. Carbon atoms are shown in yellow, light pink
and purple for the FAD, the alcohol and the rest of visible residues
respectively. The Gln395–Thr406 loop is shown in magenta.
study on AAO, including analysis of substrate and sol-
vent kinetic isotope effects (KIEs), suggested a con-
certed proton abstraction from the alcohol hydroxyl
and hydride transfer from the alcohol a-carbon to
FAD, without formation of a stable alkoxide interme-
diate [13]. Subsequent research combining site-directed
mutagenesis and quantum mechanics/molecular
mechanics (QM/MM) calculations confirmed the role
of His502 as the AAO catalytic base [18]. Moreover,
the combination of QM/MM profiles and solvent KIE
data showed that the O–H bond cleavage for proton
abstraction from the alcohol hydroxyl precedes the
Ca–H bond cleavage for hydride transfer, although
both processes are highly coupled. These observations
revealed a non-synchronous concerted mechanism for
alcohol oxidation by AAO [18]. This finding contrasts
with the non-concerted mechanism previously reported
for choline and methanol oxidases (EC 1.1.3.17 and
EC 1.1.3.13, respectively), which includes a stable alk-
oxide intermediate, and provides an alternative mecha-
nism for alcohol oxidation in the GMC superfamily
A hydride transfer reaction assisted by a conserved
active-site base is the consensus catalytic mechanism in
the GMC superfamily [19–25]. An initial mechanistic
3092
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS

P. Ferreira et al.
Aromatic stacking interactions in aryl-alcohol oxidase
[26,27]. Subsequent studies indicated that hydride
abstraction from the alcohol substrate by AAO is ste-
reo-selective. Consequently, the interest in the study of
the AAO reaction mechanism has expanded due to the
enzyme’s potential for de-racemization of chiral mix-
tures, as well as to the identification of new oxidative
conversions, such as the recently described 5-hydrox-
ymethyfurfural oxidation to 2,5-furandicarboxylic acid
for production of renewable polyesters [28].
The steady-state kinetic parameters obtained by fit-
ting the experimental data from the bi-substrate kinet-
ics analysis to either Eqn (1) or Eqn (2) (describing
A
In the present study, various aromatic alcohol sub-
strates (Fig. 2) have been used to complete the descrip-
tion of the AAO overall catalytic cycle and to evaluate
the influence of the aromatic stacking interactions on
catalysis. This latter aim was also addressed by using
mutants at position Tyr92 to investigate its contribu-
tion to the AAO reaction mechanism. The results pre-
sented provide new insights into the catalytic
mechanism of AAO, and, by extension, of the GMC
superfamily.
B
Results
Influence of alcohol type on AAO kinetic
mechanism: steady-state results
Double reciprocal plots of the initial rates of AAO
reaction with various substituted benzyl alcohol sub-
strates (Fig. 2) at various O2 concentrations yielded
two kinetic patterns. For 3,4-dimethoxybenzyl and 3-
chloro-4-methoxybenzyl alcohols, the plots were linear
and intersected to the left of the y axis (below zero
with respect to the x axis) (Fig. 3A), as previously
reported for 4-methoxybenzyl alcohol and 2,4-hexadi-
en-1-ol [13]. This is indicative of a sequential reaction
mechanism involving a ternary complex between AAO
and its reducing/oxidizing substrates (Scheme 1, top
loop). In contrast, a parallel line pattern was obtained
for the two other substrates investigated, 3-chloroben-
zyl and 3-fluorobenzyl alcohols (Fig. 3B), indicating a
ping–pong mechanism (Scheme 1, bottom loop).
Fig. 3. Bi-substrate steady-state kinetics for AAO. Double
reciprocal plot for the oxidation of 3-chloro-4-methoxybenzyl alcohol
(A) and 3-chlorobenzyl alcohol (B) by native AAO, measured in
0.1 M phosphate buffer, pH 6, at 12 °C, as a function of the alcohol
substrate concentration at fixed O2 concentrations: 61 lM
(crosses), 152 lM (black triangles), 319 lM (open circles), 668 lM
(black circles) and 1520 lM (open squares). Direct plots are shown
in the corresponding insets.
Fig. 2. Chemical structures of the alcohol
substrates used in this study. (1) 4-
methoxybenzyl alcohol, (2) 3,4-
dimethoxybenzyl alcohol, (3) 3-chloro-4-
methoxybenzyl alcohol, (4) 3-chlorobenzyl
alcohol and (5) 3-fluorobenzyl alcohol.
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS
3093

Aromatic stacking interactions in aryl-alcohol oxidase
P. Ferreira et al.
Scheme 1. Ternary and ping–pong mechanisms in AAO catalysis. The ternary and ping–pong mechanisms are described by the upper and
lower loops, respectively, and the reaction on the right corresponds to aldehyde release in anaerobic experiments.
Table 1. Kinetic parameters for the steady-state reaction and for the pre-steady-state reductive half-reaction of AAO with five different
alcohol substrates.
Steady-state turnover
Reductive half-reaction
kcat
(sꢀ1
Km(Al)
kcat/Km(Al)
(sꢀ1ꢁmM
Km(Ox)
kcat/Km(Ox)
ꢀ1
ꢀ1
Alcohol substrate
)
(lM)
)
(lM)
(sꢀ1ꢁmM
)
kred (sꢀ1
)
Kd (lM)
4-Methoxybenzyl alcohol 1
3-Chloro-4-methoxybenzyl
alcohol
129 ꢂ 5
29 ꢂ 1
25 ꢂ 3
8 ꢂ 1
5160 ꢂ 650
3630 ꢂ 470
348 ꢂ 36
214 ꢂ 30
371 ꢂ 41
136 ꢂ 20
115 ꢂ 3
95 ꢂ 3
31 ꢂ 2
71 ꢂ 10
3,4-Dimethoxybenzyl alcohol
3-Chlorobenzyl alcohol
3-Fluorobenzyl alcohol
57 ꢂ 1 543 ꢂ 39
13 ꢂ 1 62 ꢂ 1
9 ꢂ 1 164 ꢂ 3
105 ꢂ 8
210 ꢂ 4
56 ꢂ 1
119 ꢂ 11
10 ꢂ 1
7 ꢂ 1
479 ꢂ 45
1300 ꢂ 130
1400 ꢂ 160
131 ꢂ 10 838 ꢂ 242
8 ꢂ 1
6 ꢂ 1
58 ꢂ 2
180 ꢂ 7
Measurements were performed in 0.1 M phosphate buffer, pH 6, at 12 °C. Steady-state kinetic constants were determined by varying the
concentrations of both alcohol and O2, and calculated by fitting to Eqn 1 (4-methoxybenzyl, 3-chloro-4-methoxybenzyl and 3,4-dimethoxyben-
zyl alcohols) or Eqn 2 (3-chlorobenzyl and 3-fluorobenzyl alcohols) describing either a ternary or ping-pong mechanism, respectively (see
Materials and Methods). The pre-steady state observed reduction rate constants were fitted to Eqn 3. Means and standard deviations are
provided. 1Data from Ferreira et al. [13].
ternary or ping–pong mechanisms, respectively; see
Experimental procedures), are summarized in Table 1.
Turnover numbers (kcat) calculated under substrate
(alcohol and O2) saturation conditions showed a simi-
lar dependence on the electronic nature of the substitu-
ents on the benzenic ring to those previously reported
under air atmosphere [14]. The highest kcat value
(129 sꢀ1 for 4-methoxybenzyl alcohol) and the lowest
kcat value (13 sꢀ1 for 3-fluorobenzyl alcohol) were
observed in the presence of electron donating and
withdrawing substituents, respectively. However, the
type of alcohol substrate also had a marked influence
kcat/KmðO Þ, is three times higher in the presence of
2
alcohol substrates producing a ping–pong mechanism.
Nevertheless, the catalytic efficiency with regard to the
alcohol substrate, kcat/Km(Al), is considerably decreased
in the ping–pong mechanism due to both a reduction
in kcat and an increase in Km(Al) (Table 1).
AAO redox state during turnover and pre-steady-
state kinetics: stopped-flow results
To investigate the rate-limiting step during oxidation
of the various alcohols, we analyzed the redox state of
the FAD cofactor during the steady-state turnover of
AAO. To achieve this, equal volumes of the enzyme
and of each alcohol at its saturating concentration
were mixed in the stopped-flow equipment under air
atmosphere (Fig. 4). The first spectra obtained after
mixing showed slight displacement of flavin band I
(462 nm), as well as a slight absorbance increase in
the cases of 3-chloro-4-methoxybenzyl, 3-chlorobenzyl
on KmðO Þ, and therefore on the AAO affinity for O2.
2
Thus, the enzyme exhibits a considerably higher affin-
ity for O2 when the reaction takes place with alcohols
that show a ping–pong mechanism in comparison with
those forming a ternary complex: the KmðO Þ values for
2
3-chlorobenzyl and 3-fluorobenzyl alcohols are up to
35 times lower than that for 4-methoxybenzyl alcohol.
As a consequence, the catalytic efficiency for O2,
3094
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS

P. Ferreira et al.
Aromatic stacking interactions in aryl-alcohol oxidase
A
Fig. 4. AAO redox state during turnover with five alcohol
substrates. Aerobic solutions of AAO (approximately 9 lM) were
mixed in the stopped-flow spectrophotometer with equal volumes
of saturating concentrations of the various alcohols in 0.1 M
phosphate buffer, pH 6, at 25 °C and under aerobic conditions. (A)
Time course of the reaction at 462 nm with: 2 mM 3-chlorobenzyl
alcohol (trace 1), 5 mM 3-fluorobenzyl-alcohol (trace 2), 1 mM 3-
chloro-4-methoxybenzyl alcohol (trace 3), 2 mM 4-methoxybenzyl
alcohol (trace 4) and 4 mM 3,4-dimethoxybenzyl alcohol (trace 5).
(B, C) AAO spectral changes upon mixing with 3-chloro-4-
methoxybenzyl alcohol and 3-chlorobenzyl alcohol, respectively.
The oxidized enzyme spectrum before mixing is shown as a dotted
line. Spectra during the steady-state turnover phase (lag phase) are
shown as solid lines at 0.003, 0.012, 0.026 and 0.051s in (A), and
at 0.16, 0.33 and 0.98s in (B). Spectra after the steady-state
turnover phase are shown as dashed lines at 0.24, 0.51, 0.71,
0.97, 1.5, 2.34, 8, 12, 18 and 27 s in (A), and at 3.6, 4.3, 4.59,
4.91, 10 and 60 s in (B). The corresponding insets show the
spectral species (A, B and C) obtained by global analysis of the
spectral evolution once the steady-state turnover phase has been
completed.
B
varies with the alcohol substrate, reflecting the
relative rates of AAO reduction (by each alcohol)
and its oxidation (by O2) (Fig. 4A). For all the
assayed substrates, AAO is predominantly in the oxi-
dized form during this steady-state turnover phase,
and the percentage of reduced fraction depends on
the alcohol type. For 3-chloro-4-methoxybenzyl,
3-chlorobenzyl and 3-fluorobenzyl alcohols, the AAO
oxidized form is highly predominant (95–100%) at
the beginning of turnover, indicating that the oxida-
tive half-reaction is much faster than the reductive
one. Similar results have previously been observed for
4-methoxybenzyl alcohol (approximately 80% is in
the oxidized form) [13]. By contrast, when 3,4-di-
methoxybenzyl alcohol was used, the percentage of
reduced AAO during turnover increased, up to
approximately 40%, indicating that the rates for the
reductive and oxidative half-reactions are almost
balanced in this case, in agreement with the similar
values observed for alcohol and O2 catalytic efficien-
cies (Table 1).
C
The spectral changes after the steady-state turnover
phase (the lag phase, Fig. 4A) are indicative of accu-
mulation of reduced AAO (AAOred) as a consequence
of decreasing the O2 concentration in the reaction
mixture while alcohol is in excess (Fig. 4). When
using 3-chloro-4-methoxybenzyl, 4-methoxybenzyl and
3,4-dimethoxybenzyl alcohols as substrates, such
and 3-fluorobenzyl alcohols, reflecting formation of
the enzyme–substrate complex (Fig. 4B,C) [13,15].
For all substrates, the first spectra produced after
mixing quickly evolved to spectral species that remain
constant for a while and correspond to the steady-
state turnover (lag phase in Fig. 4A). The absorbance
of FAD band I during such periods provides infor-
mation about the flavin redox state during steady-
state turnover. The time duration of these lag periods
spectral evolution best fitted
a two-step process
(A ? B ? C) (Fig. 4B). The first step reflects reduc-
tion of the flavin with concomitant formation of a
charge transfer complex (CTC) between the reduced
enzyme and the aldehyde product of the reaction
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS
3095

Aromatic stacking interactions in aryl-alcohol oxidase
P. Ferreira et al.
one-step process (A ? B), leading directly from oxi-
dized to fully reduced flavin without stabilization of
any CTC intermediate (Fig. 4C). These results with
various substrates under turnover conditions, together
with the steady-state kinetic constants obtained, pro-
vide valuable mechanistic information that may be
correlated with the pre-steady-state studies as
described below.
A
Stopped-flow spectrophotometry has additionally
been used to study the influence of the alcohol type on
the AAO reductive half-reaction by monitoring the
time course of its reduction under anaerobic condi-
tions in the 350–900 nm range. Under the assayed con-
ditions, the five alcohol substrates fully reduced AAO
without production of any FAD semiquinone interme-
diate (Fig. 5), consistent with a hydride transfer mech-
anism as reported for 4-methoxybenzyl alcohol and
2,4-hexadien-1-ol [13].
B
For 3,4-dimethoxybenzyl and 3-chloro-4-methoxyb-
enzyl alcohols, the evolution of spectral changes upon
AAO reduction was consistent with a two-step kinetic
model (A ? B ? C) (Fig. 5A), as previously reported
for 4-methoxybenzyl alcohol [13]. The first process
(A ? B) was fast and hyperbolically dependent on
substrate concentration, accounting for more than
80% of the reaction amplitude. This step is thought to
correspond to substrate oxidation and formation of an
AAOred–product complex. The second process
(B ? C) was concentration-independent and too slow
(3–5 sꢀ1) to be relevant for overall turnover (Table 1).
This step may be related to a slow release of the alde-
hyde product from the reduced AAO active site in the
absence of O2 (k3 in Scheme 1). Formation of such
unstable enzyme–product complexes have also been
reported in other flavin-dependent oxidases that show
a bi-phasic reaction course [29,30]. The hyperbolic
dependence of reduction rates (kobs) for the first pro-
cess was fitted to Eqn (3) or Eqn (4) (see Experimental
procedures), consistent with an essentially irreversible
flavin reduction (krev of approximately 0). The calcu-
lated reduction constants (kred) for the AAO reduction
by 3,4-dimethoxybenzyl and 3-chloro-4-methoxybenzyl
alcohols were up to threefold faster than their turnover
rates, suggesting that the reductive half-reaction is not
the limiting step when these alcohols are oxidized.
For 3-chlorobenzyl and 3-fluorobenzyl alcohols, the
first spectra obtained after mixing clearly exhibit dis-
placement and an absorbance increase of the flavin
band I. These spectral changes were previously attrib-
uted to formation within the dead time of the stopped-
flow assay of an initial Michaelis-Menten complex of
the oxidized enzyme with the alcohol substrate [15].
The subsequent decrease in absorbance best fitted a
Fig. 5. Pre-steady-state kinetics for the reductive half-reaction of
AAO. Spectral time-course of the anaerobic reduction of native
AAO (7.5 lM) by 559 lM 3-chloro-4-methoxybenzyl (A) and 2400 lM
3-chlorobenzyl (B) alcohols. The spectrum of the oxidized enzyme
before mixing is shown as a dashed line. Spectra after mixing are
shown at 4, 19, 39, 80, 330 and 500 ms in (A), and at 2, 39, 78,
129, 219 and 500 ms in (B). The corresponding insets show the
absorbance spectra for the two/three kinetically distinguishable
spectroscopic species (A, B and C) obtained by global analysis of
the spectral evolution. The further insets show the evolution of
these species along the reaction course.
previously characterized by a broad band at approxi-
mately 550–650 nm [13], and supporting the T-shaped
stacking interaction between Tyr92 and the substrate
suggested by ligand migration studies (Fig. 1). The sec-
ond step involves spectral perturbation in the CTC
band and the appearance of a new peak at 490 nm.
These events (CTC formation and spectral perturba-
tions) may be related to the presence of O2 in the
active site, as the AAO anaerobic reduction by these
substrates prevents CTC formation, as described below
(Fig. 5). On the other hand, the spectral changes after
the steady-state turnover phase when using 3-chlorob-
enzyl and 3-fluorobenzyl alcohols occurs through a
3096
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS

P. Ferreira et al.
Aromatic stacking interactions in aryl-alcohol oxidase
Table 2. T-shaped stacking energies between five AAO alcohol
substrates and their corresponding aldehydes with a tyrosine side
chain, and differences between the two calculated energies
(kcalꢁmolꢀ1).
one-step model (A ? B), describing an essentially irre-
versible AAO reduction by 3-chlorobenzyl and 3-fluo-
robenzyl alcohols (Fig. 5B). The observed rates for
this step showed hyperbolic dependence on the sub-
strate concentration, which fitted Eqn (3), indicating
that this step includes rearrangements in the initial
complex to produce the catalytically competent com-
plex as well as the flavin reduction. The kred values for
these two alcohols (8 ꢂ 1 and 6 ꢂ 1 sꢀ1, respectively)
were similar to their reported turnover rates (Table 1),
clearly indicating that, for these substrates, the reduc-
tive half-reaction is the limiting step in catalysis.
Alcohols
Aldehyde
Difference
4-Methoxybenzyl alcohol
3-Chloro-4-methoxybenzyl
alcohol
ꢀ3.27
ꢀ3.16
ꢀ2.83
ꢀ3.00
ꢀ0.44
ꢀ0.16
3,4-Dimethoxybenzyl alcohol
3-Chlorobenzyl alcohol
3-Fluorobenzyl alcohol
ꢀ4.11
ꢀ2.77
ꢀ2.67
ꢀ3.62
ꢀ1.96
ꢀ1.81
ꢀ0.49
ꢀ0.81
ꢀ0.86
Similarly, the AAO oxidative half-reaction was stud-
ied by following the flavin absorbance increase at
462 nm after mixing enzyme reduced in advance by
incubation with a slight excess of the various alcohol
substrates under anaerobiosis with buffer containing
known O2 concentrations. For all reducing substrates,
the rates of flavin reoxidation were linearly dependent
on O2 concentration (data not shown), and similar to
those reported for other oxidases. The values obtained
tants than for the aldehyde products. This difference
facilitates release of the products. Furthermore, impor-
tant differences were observed for 3-chlorobenzyl and
3-fluorobenzyl alcohols. We observed a lower stacking
stabilization for the alcoholic form, and a larger
decrease in the interaction energy with the aldehyde.
The significantly lower stabilization for the aldehyde
product for these two substrates also establishes a
direct correlation between lower p–p stacking stabiliza-
tion energies and the ping–pong mechanism, and the
lack of stabilization of any CTC during turnover.
for second-order rate constants were 7.0 9 105,
ꢀ1
8.0 9 105 and 8.4 9 105
chloro-4-methoxybenzyl and 3,4-dimethoxybenzyl alco-
M
ꢁsꢀ1 for 3-chlorobenzyl, 3-
hol, respectively, similar to that previously reported for
4-methoxybenzyl alcohol (6.7 9 105
Kinetic properties for Tyr92 variants
ꢀ1
M
ꢁsꢀ1) [12, 31].
Therefore, the rates of flavin reoxidation are neither
dependent on the nature of the alcohol substrate nor on
the kinetic mechanism. However, the experimental con-
ditions used do not guarantee the presence of the alde-
hyde product at the active site during flavin reoxidation.
Several site-directed variants of AAO at Tyr92 were
prepared to investigate its role in catalysis. The Y92L,
Y92F and Y92W variants were purified as oxidized
holoproteins, showing an A280/A463 ratio of approxi-
mately 10, similar to the native enzyme. The absorp-
tion spectra of Y92L and Y92F were basically similar
to that of the native enzyme, with absorption maxima
at 387 and 463 nm, while the maxima for the Y92W
variant were slightly displaced at 384 and 457 nm
(data not shown).
Stacking stabilization of the substrate in the
AAO active site
AAO oxidizes mainly aromatic and other p system-con-
taining conjugated primary alcohols. The aromatic nat-
ure of the AAO alcohol substrates suggests the
importance of p–p stacking in substrate stabilization by
AAO. Ligand migration studies showed a T-shaped
interaction between Tyr92 and the alcohol substrate in
the catalytically active configurations [18]. Starting from
this active complex, the Tyr–substrate binding energy
was evaluated using ab initio quantum chemistry.
The catalytic properties of the Y92L, Y92F and
Y92W AAO variants were determined using 4-meth-
oxybenzyl alcohol as substrate (Table 3). While mutat-
ing the tyrosine to phenylalanine only causes a slight
increase in KmðO Þ, its replacement with a leucine pro-
2
duces a decrease in catalytic efficiency for the alcohol
substrate (2.6-fold lower), accompanied by approxi-
mately twofold increases in both Km(Al) and Kd.
Finally, the incorporation of a bulkier residue (Y92W
mutation) causes a strong decrease in catalytic efficien-
cies for both O2 (sixfold lower) and 4-methoxybenzyl
alcohol (860-fold lower). As the turnover rate for the
Y92W variant was reduced tenfold, we conclude that
the main effect of the mutation concerns the availabil-
ity of the alcohol substrate at the AAO active site
(with 75-fold higher Km values). Likewise, the Y92W
mutation prevents CTC formation during enzyme
Table 2 shows the T-shaped interaction energies
between a tyrosine side chain and five AAO substrates:
4-methoxybenzyl, 3-chloro-4-methoxybenzyl, 3,4-di-
methoxybenzyl, 3-chlorobenzyl and 3-fluorobenzyl.
For each substrate, we modeled both the alcohol and
the aldehyde forms. The first clear result is that all the
interaction energies are stabilizing, with similar values
to those obtained by others [32]. In all cases, the sta-
bilization energy is slightly higher for the alcohol reac-
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS
3097

Aromatic stacking interactions in aryl-alcohol oxidase
P. Ferreira et al.
Table 3. Kinetic parameters for steady-state reaction and for the pre-steady-state reductive half-reaction of native AAO and three Tyr92
variants in the oxidation of 4-methoxybenzyl alcohol.
Steady-state turnover
Reductive half-reaction
kred (sꢀ1
Kd (lM)
kcat (sꢀ1
)
Km(Al) (lM)
kcat/Km(Al) (sꢀ1ꢁmM
)
Km(Ox) (lM)
kcat/Km(Ox) (sꢀ1ꢁmM
)
ꢀ1
ꢀ1
)
AAO1
Y92F
Y92L
Y92W
129 ꢂ 5
120 ꢂ 1
100 ꢂ 2
11 ꢂ 0.3
25 ꢂ 3
30 ꢂ 1
5160 ꢂ 650
4450 ꢂ 120
1940 ꢂ 4
6 ꢂ 0.3
348 ꢂ 36
147 ꢂ 4
348 ꢂ 13
181 ꢂ 7
371 ꢂ 41
814 ꢂ 21
286 ꢂ 12
60 ꢂ 3
115 ꢂ 3
119 ꢂ 9
95 ꢂ 2
14 ꢂ 1
31 ꢂ 2
42 ꢂ 12
52 ꢂ 4
51 ꢂ 2
1890 ꢂ 70
4630 ꢂ 520
Measurements were carried out in 0.1 M phosphate buffer, pH 6, at 12 °C. Bisubstrate steady-state constants were determined by varying
the concentrations of both alcohol and O2, and fitting the data to Eqn 1 (native AAO) or Eqn 2 (Y92F, Y92L and Y92W variants) describing
ternary and ping-pong mechanisms, respectively (see Materials and Methods). Means and standard deviations are provided. 1Data from
Ferreira et al. [13].
turnover with 4-methoxybenzyl alcohol, also suggest-
ing that stacking interactions are strongly affected by
this mutation (Fig. 6).
stopped-flow techniques. The spectral evolution
obtained for all variants indicated complete (two-elec-
tron) enzyme reduction, in agreement with a hydride
transfer reaction (Fig. 7). The calculated kred and Kd
values agreed with the steady-state kcat and Km values
The fast reaction of the tyrosine variants with 4-meth-
oxybenzyl alcohol was also investigated by anaerobic
A
B
C
D
Fig. 6. Spectral species obtained after global analysis of the spectral evolution after the steady-state turnover of native AAO (A) and its
Y92L (B), Y92F (C) and Y92W (D) variants. An aerobic solution of AAO (approximately 10 lM, except that for the native protein, which was
approximately 5 lM) was reacted in the stopped-flow equipment with saturating concentrations of 4-methoxybenzyl alcohol (1 mM, except
that for the Y92W variant, which was 10 mM) in 0.1 M phosphate, pH 6, at 25 °C under aerobic conditions. Species A, B and C are shown
as solid, dashed and dotted lines, respectively. The corresponding insets show the evolution of these species along the reaction course.
3098
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS

P. Ferreira et al.
Aromatic stacking interactions in aryl-alcohol oxidase
A
B
C
D
Fig. 7. Pre-steady-state kinetics for the reductive half-reaction of three Tyr92 AAO variants when using 4-methoxybenzyl alcohol as substrate.
(A–C) Spectral evolution of the anaerobic reduction of the Y92F (A), Y92L (B) and Y92W (C) AAO variants (9 lM) by 4-methoxybenzyl alcohol.
Spectra for the oxidized enzyme before mixing are indicated by dashed lines. Spectra after mixing are shown at 0.004, 0.006, 0.009, 0.0115,
0.0141, 0.0192 and 0.032 s in (A), 0.004, 0.006, 0.012, 0.017, 0.029, 0.099 and 0.25 s in (B), and 0.004, 0.006, 0.03, 0.11, 0.18, 0.44 and
0.98 s in (C). The corresponding insets show the spectra for the kinetically distinguishable species (A and B) obtained by global analysis of
the spectral evolution. (D) Dependence of observed rates on the concentrations of 4-methoxybenzyl alcohol by native AAO (black circles) and
the Y92F (open circles), Y92L (black triangles) and Y92W (open triangles) AAO variants, fitted to Eqn (5). Assays were performed at 12 °C.
(Table 3), indicating that the reductive half-reaction is
the rate-limiting step in catalysis for these variants.
To investigate whether Tyr92, by establishing a
stacking interaction with the docked 4-methoxybenzyl
alcohol (Fig. 1), contributes to the hydride transfer
selectivity previously reported for native AAO [18], we
measured the KIEs values for apparent steady-state
kinetic constants for the Y92L variant using three
isotopically labeled preparations of this alcohol:
(R)-[a-2H]-, (S)-[a-2H]- and [a-2H2]-4-methoxybenzyl
significantly higher for the Y92L variant, and therefore
the D(Km(Al)) values were significantly lower than those
of the native AAO. In fact, the small D(Km(Al)) KIE
for native AAO (deuteration mainly affects the
1Hꢀ/2Hꢀ abstraction ratio) was absent for the Y92L
variant (although the KIE on turnover was not modi-
fied), indicating that Tyr92 contributes to 4-methoxyb-
enzyl alcohol binding.
Discussion
alcohol (Table 4). The KIEs on the apparent turnover
D app
values,
(
20%) for the (R) and di-deuterated forms, but were
k ), slightly decreased (approximately
cat
Ping–pong versus ternary mechanism in AAO
and other flavo-oxidases
unaffected for the (S) form. However, the KIEs on the
D app
apparent catalytic efficiency,
(R) form and especially the di-deuterated alcohols were
(
k
/Km(Al)), for the
cat
In comparison with other GMC oxidoreductases, which
catalyze the oxidation of alcohol groups in relatively
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS
3099

Aromatic stacking interactions in aryl-alcohol oxidase
P. Ferreira et al.
Table 4. KIEs on apparent steady-state parameters for the
oxidation of three a-deuterated p-methoxybenzyl alcohols by native
AAO and its Y92L variant under O2 saturation conditions. Steady-
state constants were estimated in O2-saturated (1.279 mM) 0.1 M
phosphate, pH 6, at 25 °C. KIEs are the ratio between the activity
on protiated and deuterated substrate calculated by fitting
constants to Eqn (6). Values are means ꢂ standard deviations.
1Data for native AAO are from Ferreira et al. [13] and Hernandez-
Ortega et al. [34].
state kinetic mechanism (Scheme 1, bottom loop). In
contrast, a sequential mechanism operates for alcohols
with electron donor substituents (methoxylated benzyl
alcohols), in which O2 reacts with the AAOred–aldehyde
complex, forming a ternary complex prior to aldehyde
product release (Scheme 1, top loop).
Vanillyl-alcohol oxidase (EC 1.1.3.38), another versa-
tile flavoenzyme that is able to oxidize aromatic alcohols,
although belonging to a different superfamily to AAO,
also exhibits substrate-dependent overall catalysis [35].
This behavior differs from other fungal GMC oxidore-
ductases, such as glucose oxidase (EC 1.1.3.4), pyranose
2-oxidase (EC 1.1.3.10), cholesterol oxidase and cellobi-
ose dehydrogenase (EC 1.1.99.18), for which the ping–
pong mechamism has been reported as the general mech-
anism [23,36–38]. Interestingly, detailed investigations
on pyranose 2-oxidase recently indicated that its steady-
state mechanism alters as a function of pH [39].
D app
(
k /
cat
D app
(
D app
(
k
)
cat
K
)
m(Al)
Km(Al))
Native AAO1
(R)-[a-2H]-4-methoxybenzyl 5.0 ꢂ 0.0 1.3 ꢂ 0.1
3.8 ꢂ 0.1
1.3 ꢂ 0.1
4.1 ꢂ 0.1
alcohol
(S)-[a-2H]-4-methoxybenzyl 1.3 ꢂ 0.0 1.0 ꢂ 0.1
alcohol
[a-2H2]-4-methoxybenzyl
alcohol
7.6 ꢂ 0.1 1.9 ꢂ 0.1
Y92L variant
(R)-[a-2H]-4-methoxybenzyl 4.0 ꢂ 0.0 1.0 ꢂ 0.1
4.3. ꢂ 0.1
1.2 ꢂ 0.1
5.8 ꢂ 0.1
alcohol
(S)-[a-2H]-4-methoxybenzyl 1.3 ꢂ 0.0 1.1 ꢂ 0.1
Stacking interactions govern AAO catalytic
mechanism
alcohol
[a-2H2]-4-methoxybenzyl
alcohol
6.2 ꢂ 0.1 1.1 ꢂ 0.1
Classical force fields semi-quantitatively reproduce T-
shaped stacking dispersion forces [40]. Moreover, high-
level QM studies have indicated the importance of
T-shaped interactions in molecular stacking, for exam-
ple in benzene dimers [41]. In AAO, previous ligand
migration simulations (using classical force fields) [18]
suggested a T-shaped substrate–Tyr92 interaction. Our
QM calculations (Table 2) confirm such stabilizing
interactions for the various substrate (alcohol)/product
(aldehyde) benzenic rings. In addition, stacking ener-
gies estimated for the various substrates/products
correlate fairly well with the electron-donating/
withdrawing properties (to the ring p cloud) of their
various substituents. The nucleophilicity of methoxy-
benzene is stronger than that of benzene due to reso-
nance, increasing the electronic density in the ring.
These effects explain why dimethoxylated alcohols/
aldehydes have the largest stacking interaction ener-
gies. On the other hand, for 3-chloro- and 3-fluoro-
substituted alcohols/aldehydes, the halogen acts as an
electron withdrawer (inductive effect), resulting in the
lowest stabilization energies. Moreover, the slightly
different interaction energies for these two halogenated
compounds follow the expected trend for the increased
electronegativity of the fluoride compared to chloride,
and thus stronger inductive electron-withdrawing
effects. Thus, the lower interaction energies for the
two halogenated aldehydes compared with the meth-
oxylated aldehydes, together with the higher alcohol–
aldehyde differential interaction energies, contribute to
release of the aldehyde product, favoring the
specific reactions, AAO shows a broad electron donor
substrate specificity, oxidizing aromatic and other p sys-
tem-containing substrates with conjugated primary
alcohols (including benzylic, naphthylic and aliphatic
polyunsaturated alcohols) [5,14,33]. These n systems
increase the electron availability at the benzylic position,
enabling hydride abstraction by flavin N5. Moreover,
the architecture of the AAO active site prevents the oxi-
dation of secondary alcohols that cannot be accommo-
dated at adequate distances from the catalytic histidine
and the above-mentioned flavin N5 atom due to the
presence of Phe501 [34].
Here we have performed a detailed study on the AAO
kinetic mechanism using various alcohol substrates. In
all cases, the AAO catalytic reaction may be divided
into a reductive half-reaction, in which two electrons are
transferred via a hydride ion to the oxidized FAD, and
an oxidative half-reaction, in which two electrons are
transferred from the reduced flavin to O2, yielding
hydrogen peroxide. However, based on the results of bi-
substrate kinetics analysis with various benzylic alco-
hols, the overall AAO catalytic cycle appears to be
highly influenced by the chemical nature of the substitu-
ents in the substrate benzenic ring. For electron-with-
drawing substituents (in 3-chloro- and 3-fluorobenzyl
alcohols), both half-reactions are independent, whereby
the aldehyde product dissociation occurs before the O2
reaction (the second substrate), in a ping–pong steady-
3100
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS

P. Ferreira et al.
Aromatic stacking interactions in aryl-alcohol oxidase
ping–pong mechanism. The stacking interactions were
intermediate for the 3-chloro-4-methoxybenzyl alcohol/
aldehyde due to the combination of electron donor
and withdrawing substituents.
substrate showing similar catalytic efficiencies for
alcohol and O2. Finally, for 3-chloro-4-methoxybenzyl
alcohol oxidation, the estimated k5 value (42 sꢀ1) sug-
gests that product release may be the rate-limiting step
(Scheme 1, top loop). In fact, previous studies on 3-
chloro-4-methoxybenzyl alcohol oxidation by AAO
indicate that a proportion of the aldehyde formed was
oxidized to the corresponding acid, rather than being
released from the active site [15].
Taking together the kinetic data, the stacking energy
calculations and the available structural and computa-
tional information reported previously, the overall
AAO catalytic cycle is consistent with the proposed
kinetic model of Scheme 1. Substrate diffusion studies
showed that alcohol and O2 molecules access the AAO
active site through the same hydrophobic channel
[12,18]. However, the alcohol diffusion pathway
requires chain displacements and interaction with resi-
dues Phe397, Tyr92 and Phe501 to reach the AAO
active site, while O2 diffusion does not require any
rearrangement. This continuous O2 supply supports
the finding that, for the sequential mechanism, where
the aldehyde product is still bound at the active site
when O2 reaches it, some arrangement of the aldehyde
product for flavin reoxidation is required (for more
detail, see Movie S1 in [12,18]). However, the O2 affin-
ity increases when the reaction occurs with alcohols,
leading to a ping–pong mechanism. This may be
caused by prior dissociation of product aldehyde, leav-
ing more space at the active site for O2 binding.
Regarding the oxidative half-reaction, reoxidation
of AAO, previously reduced by different alcohols,
yieldeꢀd1bi-molecular rate constants (7.0 9 105 – 8.4 9
105
M
ꢁsꢀ1) in the range typical of other flavo-oxidas-
es, and approximately three orders of magnitude larger
that the values reported for non-enzymatic reoxidation
of free flavins (250 M ꢁsꢀ1) [31,42]. This is consistent
ꢀ1
with enhanced O2 reactivity of AAO when the
conserved active site His502 is protonated during the
oxidative half-reaction, reducing the singlet/triplet
energy gap [10]. The importance of a positively
charged group for O2 activation has been reported for
flavoprotein oxidases [31,43].
Role of Tyr92 in AAO catalysis
Finally, the kinetic data obtained here for the Tyr92
variants strongly suggest that this residue is involved
in alcohol substrate stabilization at the AAO active
site. The Y92F variant exhibits similar rates for 3-
methoxybenzyl alcohol oxidation to the native enzyme,
while the leucine and tryptophan substitutions mainly
alter the alcohol substrate affinity. In the case of the
Y92W variant, the substrate binding significantly
weakened, with 76- and 150-fold higher Km and Kd
values, respectively. However, changes to the affinity
constants for the Y92L variant were modest, suggest-
ing that some stacking is still possible. In fact, the
Y92L and Y92F mutants are capable of forming CTC,
indicating that both proteins stabilize stacking interac-
tions. In a previous study, Tyr92 substitution with ala-
nine (fully removing the stacking interaction) did not
yield active enzyme, supporting involvement of the
above-mentioned p–p interaction in AAO catalysis
[17]. Interestingly, Tyr92 is not conserved in the GMC
oxidoreductase superfamily. However, tyrosine, phen-
ylalanine and leucine residues are frequently found at
this position in the putative AAO sequences from vari-
ous basidiomycetes genomes (Fig. 8). This suggests
that stacking may be a common catalytic strategy for
alcohol substrate oxidation in AAO enzymes.
Rate-limiting step(s) in AAO catalysis
Rapid kinetic experiments provided information on
the reductive and oxidative half-reactions for each
alcohol substrate. For the 3-chlorobenzyl and 3-fluo-
robenzyl alcohols, the reductive half-reaction is the
rate-determining step in catalysis, and is responsible of
the low efficiency of substrate oxidation.
However, for 3,4-dimethoxybenzyl and 3-chloro-4-
methoxybenzyl alcohols, the flavin reduction rate is far
from being the limiting state. In the first case, the
apparent second-order constant estimated from pre-
steady kinetic data (kred/Kd = 156 ꢂ 46 sꢀ1ꢁmMꢀ1) is
in agreement with the steady-state alcohol catalytic
efficiency obtained (105 ꢂ 8 sꢀ1ꢁmMꢀ1). In the ternary
complex mechanism shown in Scheme 1 (top loop),
kcat is a combination of catalytic steps in which the
3,4-dimethoxybenzyl alcohol is oxidized to 3,4-dimeth-
oxybenzaldehyde (k2 approximately equal to kred) and
the product is released (k5) [kcat = k2 k5/(k2 + k5)]. The
estimated value for k5 (101 sꢀ1) is indicative of a par-
tially rate-limiting step for catalysis, suggesting that
another kinetic step (after hydride transfer and before
product release) also contributes to limiting the reac-
tion rate. This step may be atributted to the flavin
reoxidation, as suggested by the fact that 40% of
AAO is reduced during enzyme turnover with this
In conclusion, the study of the overall catalytic cycle
of AAO presented here shows that the stacking-stabi-
lizing interaction of the aromatic substrate/product by
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS
3101

Aromatic stacking interactions in aryl-alcohol oxidase
P. Ferreira et al.
Fig. 8. Sequence logo for 70 putative AAO sequences from various basidiomycetes corresponding to positions 83–97 of mature P. eryngii
AAO generated using WebLogo [51]. The compared sequences were taken from the following genomes available at the Joint Genome
Institute (www.jgi.doe.gov); the numbers of AAO sequences are indicated in parentheses: Bjerkandera adusta (11), Dichomitus squalens (9),
Fomitiporia mediterranea (1), Fomitopsis pinicola (1), Ganoderma sp. (7), Gelatoporia subvermispora (4), Gloeophyllum trabeum (2),
Laccaria bicolor (1), Phanerochaete chrysosporium (3), Phlebia brevispora (3), Punctularia strigosozonata (6), Rhodonia placenta (2),
Stereum hirsutum (15) and Trametes versicolor (3). The numbers indicate the position in the mature P. eryngii AAO sequence, the overall
height of each letter reflects the sequence conservation at that position, and the heights of symbols in the same column indicate the
relative frequency of each amino acid.
GATGCG-30, and (c) Y92W, 50-GGGTCTAGCTCTGTTC
active-site Tyr92 governs AAO catalysis, switching
ACTGGATGGTCATGATGCG-30 [44]. Mutations were
between the ping–pong and the ternary mechanism
confirmed by sequencing using a GS-FLX sequencer (Roche,
Basel, Switzerland), and the mutated variants were obtained
as described for recombinant AAO. The naturally oxidized
AAO concentration was determined using the molar absor-
bances of native AAO and of its Y92L, Y92F and Y92W
depending on the stacking stabilization energies. The
importance of Tyr92 for alcohol substrate binding was
also demonstrated by site-directed mutagenesis, kinet-
ics and KIE studies, suggesting a common role in
other AAO proteins where this residue is conserved or
replaced by others that are able to establish similar
stacking interactions.
variants (e463 =ꢀ111 050 M ꢁcmꢀ1, e463 = 11ꢀ1240 M ꢁcmꢀ1
,
ꢀ1
ꢀ1
e463 = 10 044 M ꢁcmꢀ1 and e457 = 10 693 M ꢁcmꢀ1, respec-
tively) calculated by heat denaturation and estimation of the
ꢀ1
free FAD released (e450 11 300 M ꢁcmꢀ1) [44].
Experimental procedures
Steady-state kinetic measurements
Chemicals
Steady-state kinetics data were spectrophotometrically
obtained by following the oxidation of the alcohol sub-
strate to the corresponding aldehyde as previously
described [14]. Two-substrate steady-state kinetics measure-
ments were performed by simultaneously varying the con-
centrations of O2 and alcohol substrate in 0.1 M phosphate
buffer, pH 6, at 12 °C. Due to large differences in AAO
affinities for these alcohols, the concentrations were in the
ranges of 8.7–279, 31.3–4000, 7.8–2000 and 39–4000 lM for
3-chloro-4-methoxybenzyl, 3,4-dimethoxybenzyl, 3-chlorob-
enzyl and 3-fluorobenzyl alcohols, respectively. The reac-
tions were performed in a screw-cap cuvette in which the
buffer solution was first equilibrated at the desired concen-
tration of O2 (61, 152, 319, 668 and 1520 lM at 12 °C) by
bubbling with the appropriate O2/N2 gas mixture for 10–
15 min. Then, reactions were started by addition of the
alcohol substrate (approximately 5–10 lL) and AAO
(5 lL, 0.03 lM final concentration) into a reaction mixture
with a final volume of 1.5 mL. Initial rates were calculated
during the linear phase of alcohol oxidation to the corre-
sponding aldehyde, and analyzed by fitting to Eqn (1) or
Eqn (2), describing ternary complex and ping–pong mecha-
nisms, respectively:
4-methoxybenzyl, 3,4-dimethoxybenzyl (veratryl), 3-chlo-
robenzyl and 3-fluorobenzyl alcohols were purchased from
Sigma-Aldrich (St Louis, MO, USA). 3-chloro-4-methoxyb-
enzyl alcohol, (R)-[a-2H]-4-methoxybenzyl alcohol, (S)-[a-2H]-
4-methoxybenzyl alcohol and [a-2H2]-4-methoxybenzyl
alcohol were synthesized at the Instituto de Ciencia de Ma-
ꢀ
teriales de Aragon (Zaragoza, Spain).
Protein production and purification
Native AAO from P. eryngii was obtained by expression in
Escherichia coli of the mature AAO cDNA (GenBank
accession number AF064069), followed by in vitro activa-
tion in the presence of the FAD cofactor, and purification
by ion-exchange chromatography as described previously
[44]. Mutated variants were prepared using a QuikChange
site-directed mutagenesis kit (Stratagene, Santa Clara, CA,
USA). For the PCR reactions, cDNA cloned into the
pFLAG1 vector was used as the template, and the follow-
ing oligonucleotides (direct sequences) bearing mutations
(shown in bold italics) as primers: (a) Y92L, 50-GGGT
CTAGCTCTGTTCACCTCATGGTCATGATGCG-30, (b)
Y92F, 50-GGGTCTAGCTCTGTTCACTTCATGGTCAT
3102
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS

P. Ferreira et al.
Aromatic stacking interactions in aryl-alcohol oxidase
v
kcatAB
KmðOxÞA þ KmðAlÞB þ AB þ KiðAlÞKmðOxÞ
are the limiting rates for hydride transfer from the sub-
strate to the flavin and for the reverse reaction, respec-
tively, at saturating substrate concentrations, and Kd is
the dissociation constant for the enzyme–substrate
complex.
¼
ð1Þ
ð2Þ
e
v
e
kcatAB
¼
KmðOxÞA þ KmðAlÞB þ AB
Rate constants for the oxidative half-reaction were mea-
sured at 12 °C in 0.1 M phosphate buffer by monitoring the
absorbance increase at 462 nm upon mixing the previously
anaerobically reduced enzyme with buffer equilibrated at
various O2 concentrations. The enzyme was reduced either
by mixing the oxidized protein under anaerobic conditions
with a 1.2-fold excess of the corresponding alcohol sub-
strate or by photoreduction in the presence of 2 mM
5-deazariboflavin and 3 mM EDTA [14]. The apparent
second-order rate constants for the oxidative half-reaction
(kox) were determined as a function of the O2 concentra-
tions, and calculated using Eqn (5), where the kobs is the
experimentally observed rate constant associated with flavin
oxidation at any given O2 concentration:
where v represents the observed initial rate, e is the
enzyme concentration, kcat is the maximal turnover, A
is the alcohol substrate concentration, B is the O2 con-
centration, Km(Al) and Km(Ox) are the Michaelis-Men-
ten constants for alcohol and O2, respectively, and Ki
is the alcohol dissociation constant.
(Al)
Stopped-flow measurements: enzyme turnover
and pre-steady-state kinetics
Stopped-flow experiments were performed on an Applied
Photophysics SX17.MV spectrophotometer using SX18.MV
or XSCAN software for experiments with single wavelength
or photodiode-array detection, respectively (Applied Photo-
physics, Leatherhead, Surrey, UK). For enzyme-monitored
turnover experiments, air-saturated enzyme and substrate
solutions were mixed, and the evolution of the flavin redox
state was monitored in the range 350–900 nm.
app
kobs
¼
kox½O2ꢃ
ð5Þ
KIEs for p-methoxybenzyl alcohol oxidation
Reductive half-reaction studies were performed under
anaerobic conditions. Tonometers containing enzyme and
substrate solutions were made anaerobic by successive evac-
uation and flushing with argon. These solutions also con-
tained glucose (10 mM) and glucose oxidase (10 UꢁmLꢀ1) to
ensure anaerobiosis. Drive syringes in the stopped-flow
apparatus were made anaerobic by sequential passing
through of a sodium dithionite solution and O2-free buffer
[45]. The substrate concentrations used were in the range
0.195–12.5, 0.035–0.559, 0.037–2.4 and 0.039–9.9 mM for
3,4-dimethoxybenzyl, 3-chloro-4-methoxybenzyl, 3-chlorob-
enzyl and 3-fluorobenzyl alcohols, respectively. Measure-
ments were performed in 0.1 M phosphate buffer, pH 6, at
12 °C. All given concentrations are those after mixing an
equal volume of substrate and enzyme (i.e. final concentra-
tions). Spectral evolution was studied by global analysis and
numerical integration methods using PRO-K software
(Applied Photophysics). Observed rate constants (kobs) from
traces recorded at 462 nm were calculated from exponential
fits. Rate constants were obtained by non-linear fitting of
kobs at various substrate concentrations to either Eqn (3) or
Eqn (4):
The substrate KIEs on apparent steady-state kinetic con-
stants due to p-methoxybenzyl alcohol a-deuteration (R, S
and di-deuterated forms) were measured in O2-saturated
(1.279 mM) 0.1 M phosphate buffer, pH 6.0 at 25 °C. The
substrate KIEs were calculated by fitting the initial rates to
Eqn (6), which describes a mechanism with separate iso-
tope effects on kcat and kcat/Km:
v
e
kcat
S
¼
ð6Þ
Kmð1 þ FiE cat Þ þ Sð1 þ FiEk
k
K
Þ
cat
m
where S is the substrate concentration, Fi is the atom
fraction of deuterium label in the substrate (0.98 in the
present case), and Ek
and Ek are the isotope
cat
cat=Km
effect minus 1 on the two kinetic constants [46].
QM calculations
Second-order Moller–Plesset (MP2) [47] QM calculations
were performed using the Gaussian03 program (Gaussian
Inc., Wallingford, CT, 2004). Triple-zeta split-valence basis
sets with polarization orbitals (6-311G*) were used
throughout. The structures of the T-shaped complexes
between a model tyrosine side chain and the 4-methoxyben-
zyl alcohol were optimized using MP2 in gas phase, and the
most stable structure was selected. To isolate the stacking
interactions and avoid distortions from the gas-phase mini-
mization for the remaining complexes, the H atom in the
meta position of the methoxybenzyl structure was replaced
by a chloride or a methoxy group. The 3-chlorobenzyl alco-
kred
A
kobs
¼
ð3Þ
Kd þ A
kred
A
kobs
¼
þ krev
ð4Þ
Kd þ A
where kobs is the observed rate for reduction of the
enzyme at a given alcohol concentration, kred and krev
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS
3103

Aromatic stacking interactions in aryl-alcohol oxidase
P. Ferreira et al.
hol was obtained by removing the para methoxyl from the
3-chloro-4-methoxybenzyl, while the 3-fluorobenzyl alcohol
structure was obtained by a chloride to fluoride change in
3-chlorobenzyl alcohol. At this point, the benzene rings are
maintained frozen and all substituents are optimized. This
procedure was performed for both the alcohol and alde-
hyde forms. Binding energies were approximated from the
interaction energies: AB – (A + B) (where AB stands for
the tyrosine-substrate complex) for all complexes, and
basis-set superposition error (BSSE) corrections [48] were
included at the same level of theory. Several studies on
substituted benzene dimers have shown that this level of
theory is capable of a qualitative description of differences
in stabilization energies [49].
4 Sannia G, Limongi P, Cocca E, Buonocore F, Nitti G
& Giardina P (1991) Purification and characterization
of a veratryl alcohol oxidase enzyme from the lignin
degrading basidiomycete Pleurotus ostreatus. Biochim
Biophys Acta 1073, 114–119.
5 Guillen F, Martinez AT & Martinez MJ (1992)
Substrate specificity and properties of the aryl-alcohol
oxidase from the ligninolytic fungus Pleurotus eryngii.
Eur J Biochem 209, 603–611.
6 Asada Y, Watanabe A, Ohtsu Y & Kuwahara M (1995)
Purification and characterization of an aryl-alcohol
oxidase from the lignin-degrading basidiomycete
Phanerochaete chrysosporium. Biosci Biotechnol Biochem
59, 1339–1341.
7 Muheim A, Waldner R, Leisola MSA & Fiechter A
(1990) An extracellular aryl-alcohol oxidase from the
white-rot fungus Bjerkandera adusta. Enzyme Microb
Technol 12, 204–209.
8 Ruiz-Duenas FJ & Martinez AT (2009) Microbial
degradation of lignin: how a bulky recalcitrant polymer
is efficiently recycled in nature and how we can take
advantage of this. Microb Biotechnol 2, 164–177.
9 Martinez AT, Speranza M, Ruiz-Duenas FJ, Ferreira P,
Camarero S, Guillen F, Martinez MJ, Gutierrez A &
del Rio JC (2005) Biodegradation of lignocellulosics:
microbial, chemical, and enzymatic aspects of the fungal
attack of lignin. Int Microbiol 8, 195–204.
Acknowledgements
This work was supported by grants from the Spanish
Ministry of Economy and Competitiveness (MINECO)
[grant number BIO2013-42978-P (to M.M.), grant
number BIO2011-26694 (to A.T.M.), a ‘Juan de la
Cierva’ sub-program grant (to F.L.) and grant number
CTQ2010-18123 (to V.G.)] and by European projects
INDOX (grant number KBBE-2013-7-613549, to
A.T.M.) and PELE (grant number ERC-2009-Adg
25027, to V.G.).
10 Hernandez-Ortega A, Lucas F, Ferreira P, Medina
M, Guallar V & Martinez AT (2012) Role of active
site histidines in the two half-reactions of the aryl-
alcohol oxidase catalytic cycle. Biochemistry 51,
6595–6608.
11 Hernandez-Ortega A, Ferreira P, Merino P, Medina M,
Guallar V & Martinez AT. Stereoselective hydride
transfer by aryl-alcohol oxidase, a member of the GMC
superfamily. ChemBioChem 13, 427–435.
12 Hernandez-Ortega A, Lucas F, Ferreira P, Medina M,
Guallar V & Martinez AT (2011) Modulating O2
reactivity in a fungal flavoenzyme: involvement of aryl-
alcohol oxidase Phe-501 contiguous to catalytic
histidine. J Biol Chem 286, 41105–41114.
Author contributions
All of the authors performed the research and dis-
cussed the results obtained. P.F. conceived the study,
performed some kinetics studies, produced the figures,
and wrote the paper, A.H.-O. designed and con-
structed the mutants, and performed some kinetic
studies. F.L., K.W.B. and V.G. performed the compu-
tational part of the work, and wrote the corresponding
sections of the paper. J.C. produced and characterized
the protein variants and produced the figures. B.H.
performed some kinetic studies. A.T.M. and M.M.
integrated the results and wrote the paper.
13 Ferreira P, Hernandez-Ortega A, Herguedas B,
Martinez AT & Medina M (2009) Aryl-alcohol oxidase
involved in lignin degradation: a mechanistic study
based on steady and pre-steady state kinetics and
primary and solvent isotope effects with two alcohol
substrates. J Biol Chem 284, 24840–24847.
14 Ferreira P, Medina M, Guillen F, Martinez MJ, Van
Berkel WJ & Martinez AT (2005) Spectral and catalytic
properties of aryl-alcohol oxidase, a fungal flavoenzyme
acting on polyunsaturated alcohols. Biochem J 389,
731–738.
References
1 Hernandez-Ortega A, Ferreira P & Martinez AT (2012)
Fungal aryl-alcohol oxidase: a peroxide-producing
flavoenzyme involved in lignin degradation. Appl
Microbiol Biotechnol 93, 1395–1410.
2 Farmer VC, Henderson ME & Russell JD (1960)
Aromatic-alcohol-oxidase activity in the growth
medium of Polystictus versicolor. Biochem J 74,
257–262.
15 Ferreira P, Hernandez-Ortega A, Herguedas B,
Rencoret J, Gutierrez A, Martinez MJ, Jimenez-
Barbero J, Medina M & Martinez AT (2009) Kinetic
3 Bourbonnais R & Paice MG (1988) Veratryl alcohol
oxidases from the lignin-degrading basidiomycete
Pleurotus sajor-caju. Biochem J 255, 445–450.
3104
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS

P. Ferreira et al.
Aromatic stacking interactions in aryl-alcohol oxidase
and chemical characterization of aldehyde oxidation by
unspecific peroxygenase. FEBS J. doi: 10.1111/
febs.13177.
fungal aryl-alcohol oxidase. Biochem J 425, 585–593.
16 Fernandez IS, Ruiz-Duenas FJ, Santillana E, Ferreira
P, Martinez MJ, Martinez AT & Romero A (2009)
Novel structural features in the GMC family of
oxidoreductases revealed by the crystal structure of
fungal aryl-alcohol oxidase. Acta Crystallogr D Biol
Crystallogr 65, 1196–1205.
17 Ferreira P, Ruiz-Duenas FJ, Martinez MJ, van Berkel
WJ & Martinez AT (2006) Site-directed mutagenesis of
selected residues at the active site of aryl-alcohol
oxidase, an H2O2-producing ligninolytic enzyme. FEBS
J 273, 4878–4888.
29 Pollegioni L, Wels G, Pilone MS & Ghisla S (1999)
Kinetic mechanisms of cholesterol oxidase from
Streptomyces hygroscopicus and Brevibacterium
sterolicum. Eur J Biochem 264, 140–151.
30 Prongjit M, Sucharitakul J, Wongnate T, Haltrich D &
Chaiyen P (2009) Kinetic mechanism of pyranose 2-
oxidase from Trametes multicolor. Biochemistry 48,
4170–4180.
31 Mattevi A (2006) To be or not to be an oxidase:
challenging the oxygen reactivity of flavoenzymes.
Trends Biochem Sci 31, 276–283.
18 Hernandez-Ortega A, Borrelli K, Ferreira P, Medina
M, Martinez AT & Guallar V (2011) Substrate
diffusion and oxidation in GMC oxidoreductases: an
experimental and computational study on fungal aryl-
alcohol oxidase. Biochem J 436, 341–350.
19 Hecht HJ, Kalisz HM, Hendle J, Schmid RD &
Schomburg D (1993) Crystal-structure of glucose-
oxidase from Aspergillus niger refined at 2.3 angstrom
resolution. J Mol Biol 229, 153–172.
20 Yue QK, Kass IJ, Sampson NS & Vrielink A (1999)
Crystal structure determination of cholesterol oxidase
from Streptomyces and structural characterization of
key active site mutants. Biochemistry 38, 4277–4286.
21 Lario PI, Sampson N & Vrielink A (2003) Sub-atomic
resolution crystal structure of cholesterol oxidase: what
atomic resolution crystallography reveals about enzyme
mechanism and the role of the FAD cofactor in redox
activity. J Mol Biol 326, 1635–1650.
22 Kass IJ & Sampson NS (1998) Evaluation of the role
of His447 in the reaction catalyzed by cholesterol
oxidase. Biochemistry 37, 17990–18000.
23 Hallberg BM, Henriksson G, Pettersson G, Vasella A
& Divne C (2003) Mechanism of the reductive half-
reaction in cellobiose dehydrogenase. J Biol Chem 278,
7160–7166.
24 Hallberg BM, Henriksson G, Pettersson G & Divne C
(2002) Crystal structure of the flavoprotein domain of
the extracellular flavocytochrome cellobiose
32 Mutasem OS & Sherrill DC (2003) Unexpected
substituent effects in face-to-face p-stacking
interactions. J Phys Chem 107, 8377–8379.
33 Romero E, Ferreira P, Martinez AT & Martinez MJ
(2009) New oxidase from Bjerkandera arthroconidial
anamorph that oxidizes both phenolic and
nonphenolic benzyl alcohols. Biochim Biophys Acta
1794, 689–697.
34 Hernandez-Ortega A, Ferreira P, Merino P, Medina M,
Guallar V & Martinez AT (2012) Stereoselective
hydride transfer by aryl-alcohol oxidase, a member of
the GMC superfamily. ChemBioChem 13, 427–435.
35 Fraaije MW & van Berkel WJ (1997) Catalytic
mechanism of the oxidative demethylation of 4-
(methoxymethyl)phenol by vanillyl-alcohol oxidase.
Evidence for formation of a p-quinone methide
intermediate. J Biol Chem 272, 18111–18116.
36 Gibson QH, Swoboda BE & Massey V (1964) Kinetics
and mechanism of action of glucose oxidase. J Biol
Chem 239, 3927–3934.
37 Artolozaga MJ, Kubatova E, Volc J & Kalisz HM
(1997) Pyranose 2-oxidase from Phanerochaete
chrysosporium further biochemical characterisation.
Appl Microbiol Biotechnol 47, 508–514.
38 Kujawa M, Ebner H, Leitner C, Hallberg BM, Prongjit
M, Sucharitakul J, Ludwig R, Rudsander U,
Peterbauer C, Chaiyen P et al. (2006) Structural basis
for substrate binding and regioselective oxidation of
monosaccharides at C3 by pyranose 2-oxidase. J Biol
Chem 281, 35104–35115.
39 Rungsrisuriyachai K & Gadda G (2009) A pH switch
affects the steady-state kinetic mechanism of pyranose
2-oxidase from Trametes ochracea. Arch Biochem
Biophys 483, 10–15.
40 Sherrill CD, Takatani T & Hohenstein EG (2009) An
assessment of theoretical methods for nonbonded
interactions: comparison to complete basis set limit
coupled-cluster potential energy curves for the benzene
dimer, the methane dimer, benzene-methane, and
benzene-H2S. J Phys Chem A 113, 10146–10159.
41 Dinadayalane TC & Leszczynski J (2009) Geometries
and stabilities of various configurations of benzene
dehydrogenase. J Mol Biol 315, 421–434.
25 Kass IJ & Sampson NS (1998) The importance of
GLU361 position in the reaction catalyzed by
cholesterol oxidase. Bioorg Med Chem Lett 8,
2663–2668.
26 Menon V, Hsieh CT & Fitzpatrick PF (1995)
Substituted alcohols as mechanistic probes of alcohol
oxidase. Bioorg Chem 23, 42–53.
27 Fan F & Gadda G (2005) On the catalytic mechanism
of choline oxidase. J Am Chem Soc 127, 2067–2074.
ꢀ
28 Carro J, Ferreira P, Rodrıguez L, Prieto A, Serrano A,
ꢀ
ꢀ
Balcells B, Arda A, Jimenez-Barbero J, Gutierrez A,
Ullrich R et al. (2014) 5-Hydroxymethylfurfural
conversion by fungal aryl-alcohol oxidase and
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS
3105

Aromatic stacking interactions in aryl-alcohol oxidase
P. Ferreira et al.
dimer: details of novel V-shaped structure revealed.
Struct Chem 20, 11–20.
47 Wilson AK, Woon DE, Peterson KA & Dunning TH
(1999) Gaussian basis sets for use in correlated
molecular calculations. IX. The atoms gallium through
krypton. J Phys Chem 110, 7667–7676.
48 Boys SF & Bernardi F (1970) The calculation of small
molecular interactions by the differences of separate
total energies. Some procedures with reduced errors.
Mol Phys 19, 553–566.
49 Lee EC, Kim D, Jurecka P, Tarakeshwar P, Hobza P
& Kim KS (2007) Understanding of assembly
phenomena by aromatic-aromatic interactions: benzene
dimer and the substituted systems. J Phys Chem A 111,
3446–3457.
50 Borrelli KW, Cossins B & Guallar V (2010) Exploring
hierarchical refinement techniques for induced fit
docking with protein and ligand flexibility. J Comput
Chem 31, 1224–1235.
42 Massey V (1994) Activation of molecular oxygen by
flavins and flavoproteins. J Biol Chem 269, 22459–22462.
43 Gadda G (2012) Oxygen activation in flavoprotein
oxidases: the importance of being positive. Biochemistry
51, 2662–2669.
44 Ruiz-Duenas FJ, Ferreira P, Martinez MJ & Martinez
AT (2006) In vitro activation, purification, and
characterization of Escherichia coli expressed aryl-
alcohol oxidase, a unique H2O2-producing enzyme.
Protein Expr Purif 45, 191–199.
45 Medina M, Martinez-Julvez M, Hurley JK, Tollin G &
Gomez-Moreno C (1998) Involvement of glutamic acid
301 in the catalytic mechanism of ferredoxin-NADP+
reductase from Anabaena PCC 7119. Biochemistry 37,
2715–2728.
46 Sobrado P, Daubner SC & Fitzpatrick PF (2001)
Probing the relative timing of hydrogen abstraction
steps in the flavocytochrome b2 reaction with primary
and solvent deuterium isotope effects and mutant
enzymes. Biochemistry 40, 994–1001.
51 Crooks GE, Hon G, Chandonia JM & Brenner SE
(2004) WebLogo: A sequence logo generator. Genome
Res 14, 1188–1190.
3106
FEBS Journal 282 (2015) 3091–3106 ª 2015 FEBS