Structural insights into the substrate specificity of bacterial copper amine oxidase obtained by using irreversible inhibitors

Article abstract of DOI:10.1093/jb/mvr125

Copper amine oxidases (CAOs) catalyse the oxidation of various aliphatic amines to the corresponding aldehydes, ammonia and hydrogen peroxide. Although CAOs from various organisms share a highly conserved active-site structure including a protein-derived cofactor, topa quinone (TPQ), their substrate specificities differ considerably. To obtain structural insights into the substrate specificity of a CAO from Arthrobacter globiformis (AGAO), we have determined the X-ray crystal structures of AGAO complexed with irreversible inhibitors that form covalent adducts with TPQ. Three hydrazine derivatives, benzylhydrazine (BHZ), 4-hydroxybenzylhydrazine (4-OH-BHZ) and phenylhydrazine (PHZ) formed predominantly a hydrazone adduct, which is structurally analogous to the substrate Schiff base of TPQ formed during the catalytic reaction. With BHZ and 4-OH-BHZ, but not with PHZ, the inhibitor aromatic ring is bound to a hydrophobic cavity near the active site in a well-defined conformation. Furthermore, the hydrogen atom on the hydrazone nitrogen is located closer to the catalytic base in the BHZ and 4-OH-BHZ adducts than in the PHZ adduct. These results correlate well with the reactivity of 2-phenylethylamine and tyramine as preferred substrates for AGAO and also explain why benzylamine is a poor substrate with markedly decreased rate constants for the steps of proton abstraction and the following hydrolysis. The Authors 2011. Published by Oxford University Press on behalf of the Japanese Biochemical Society. The Authors 2011. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved.

Full text of DOI:10.1093/jb/mvr125

J. Biochem. 2012;151(2):167–178 doi:10.1093/jb/mvr125
Structural insights into the substrate specificity of bacterial
copper amine oxidase obtained by using irreversible inhibitors
Received August 22, 2011; accepted September 29, 2011; published online October 7, 2011
Takeshi Murakawa1, Hideyuki Hayashi1,
4-hydroxybenzylhydrazine; PEA, 2-phenylethylamine;
PHZ, phenylhydrazine; TPQ, topa quinone; TPQpsb,
product Schiff base of TPQ; TPQssb, substrate Schiff
Masayasu Taki2, Yukio Yamamoto2,
Yoshiaki Kawano3, Katsuyuki Tanizawa4 and
Toshihide Okajima1,4,
*
base of TPQ; TPQox, oxidized form of TPQ; TPQred
reduced (aminoresorcinol) form of TPQ; TPQsq,
semiquinone radical form of TPQ.
,
1Department of Biochemistry, Osaka Medical College, Osaka
2
569-8686; Graduate School of Human and Environmental Studies,
Kyoto University, Kyoto 606-8501; 3Institute of Physical and
Chemical Research, RIKEN Harima Institute/SPring-8, Hyogo
679-5148; and 4Institute of Scientific and Industrial Research, Osaka
University, Osaka 567-0047, Japan
*Toshihide Okajima, Institute of Scientific and Industrial Research,
Osaka University, Ibaraki, Osaka 567-0047, Japan.
Tel: þ81-6-6879-4292, Fax: þ81-6-6879-8460,
Copper amine oxidases (CAOs) catalyse the oxidative
deamination of various primary aliphatic amines to the
corresponding aldehydes with concomitant generation
of ammonia and hydrogen peroxide (1, 2). All CAOs
that have been structurally well characterized form
homodimers with each subunit between 70 and
95-kDa in size. Each subunit contains a Cu2þ ion
and a quinone cofactor designated topa quinone
(TPQ) that is produced post-translationally from a
specific tyrosine residue encoded in the polypeptide
chain (3-5). The overall catalytic reaction of CAOs
commonly proceeds through the ping—pong bi—bi
mechanism and consists of reductive and oxidative
half-reactions involving the 2 eꢀ-reduction of TPQ by
the substrate and re-oxidation of TPQ by molecular
oxygen, respectively (Scheme 1) (6-8). In the reductive
half-reaction, the protonated substrate amine first
forms a Michaelis complex with the enzyme containing
the oxidized cofactor (TPQox) and a proton derived
from the substrate is presumably captured by the con-
served Asp residue (catalytic base), which is initially
deprotonated [We have recently obtained evidence
that the C4 hydroxyl group of TPQox (but not the
conserved Asp) accepts a proton from the substrate
amino group and that TPQssb is formed without
participation of the conserved Asp (Murakawa et al.,
unpublished data)]. Nucleophilic attack of the depro-
tonated amino group of the substrate onto the C5 car-
bonyl group of TPQox yields a carbinolamine-like
intermediate of TPQ (step A ! B). The substrate
Schiff base intermediate (TPQssb) is then formed with
the release of a water molecule (step B ! C). The
deprotonated Asp residue abstracts the Ca proton of
the substrate in a stereospecific manner (9), thereby for-
email: tokajima@sanken.osaka-u.ac.jp
Copper amine oxidases (CAOs) catalyse the oxidation
of various aliphatic amines to the corresponding alde-
hydes, ammonia and hydrogen peroxide. Although
CAOs from various organisms share a highly conserved
active-site structure including a protein-derived cofac-
tor, topa quinone (TPQ), their substrate specificities
differ considerably. To obtain structural insights into
the substrate specificity of a CAO from Arthrobacter
globiformis (AGAO), we have determined the X-ray
crystal structures of AGAO complexed with irrevers-
ible inhibitors that form covalent adducts with TPQ.
Three hydrazine derivatives, benzylhydrazine (BHZ),
4-hydroxybenzylhydrazine (4-OH-BHZ) and phenylhy-
drazine (PHZ) formed predominantly a hydrazone
adduct, which is structurally analogous to the substrate
Schiff base of TPQ formed during the catalytic reac-
tion. With BHZ and 4-OH-BHZ, but not with PHZ,
the inhibitor aromatic ring is bound to a hydrophobic
cavity near the active site in a well-defined conform-
ation. Furthermore, the hydrogen atom on the hydra-
zone nitrogen is located closer to the catalytic base in
the BHZ and 4-OH-BHZ adducts than in the PHZ
adduct. These results correlate well with the reactivity
of 2-phenylethylamine and tyramine as preferred sub-
strates for AGAO and also explain why benzylamine
is a poor substrate with markedly decreased rate
constants for the steps of proton abstraction and the
following hydrolysis.
Keywords: copper amine oxidase/irreversible inhibi-
tor/substrate specificity/topa quinone/X-ray crystal
structure.
ming the product Schiff base intermediate (TPQpsb
)
with the reduction of the cofactor ring (step C ! D).
TPQpsb is hydrolysed to form the first product alde-
hyde and the reduced cofactor in an aminoresorcinol
form (TPQred) (steps D ! D0 ! E). Thus, the con-
served Asp residue plays an indispensable role during
the reductive half-reaction. In the oxidative half-
reaction, TPQred is oxidized by molecular oxygen to
Abbreviations: AGAO, BSAO, ECAO and HPAO,
are copper amine oxidases from Arthrobacter
globiformis, bovine serum, Escherichia coli and
Hansenula polymorpha, respectively; BHZ,
benzylhydrazine; CAO, copper amine oxidase; 2-HP,
2-hydrazinopyridine; 4-OH-BHZ,
ß The Authors 2011. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved
167

T. Murakawa et al.
Scheme 1 Proposed reaction mechanism of CAO. The carbinolamine-like intermediate shown in parenthesis has not been structurally identified.
regenerate TPQox via the semiquinone and iminoqui-
none forms of TPQ (TPQsq and TPQimq, respectively)
with the concomitant release of hydrogen peroxide and
ammonia (steps E ! F ! G ! A) (10). Alternatively,
under the steady-state conditions with excess substrate,
TPQssb is formed by direct transimination between
TPQimq and the substrate (step G ! C) (11). The
bound copper ion plays an essential role in the oxida-
tive half-reaction not only by mediating electron trans-
amines like methylamine (20). These structural features
of the substrate-binding pocket may be reflected in the
K
m values for each substrate. However, detailed inves-
tigation in view of the rate constants that could affect
overall substrate specificities given by the second order
rate constant, kcat/Km, has not been carried out, except
for a recent kinetic and structural analysis of the sub-
strate specificity of the two HPAO isozymes (16).
Hydrazine derivatives such as phenylhydrazine
(PHZ) and benzylhydrazine (BHZ) are well-known ir-
reversible inhibitors for CAOs as they form covalent
complexes with the TPQ cofactor (21-24). Their reac-
tions with TPQ are thought to be analogous to those of
amine substrates, in which the C5 carbonyl group of
TPQ undergoes nucleophilic attack by the deproto-
nated hydrazino group of the inhibitors to form a co-
valent adduct via a carbinolamine-like intermediate
(Scheme 2) (25, 26). However, the hydrazine deriva-
tives have a nitrogen atom corresponding to the Ca
atom of amines, which prevents abstraction of a
proton at this position by the conserved Asp residue.
Therefore, the enzyme is trapped as a TPQ/hydrazine
adduct (23, 26, 27), which is a tautomeric mixture of
hydrazone and azo forms, structurally analogous to
TPQssb and TPQpsb, respectively.
In this paper, we report X-ray crystal structures of
AGAO complexed with three hydrazine derivatives,
BHZ, 4-hydroxybenzylhydrazine (4-OH-BHZ) and
PHZ, which are structural analogues of PEA, tyramine
and benzylamine, respectively (Fig. 1). The inter-
actions of the amine substrates with the substrate-
binding pocket of AGAO are expected to be similarly
manifested in those of the corresponding hydrazine de-
rivatives. Based on the results obtained, we provide
structural insights into the substrate specificity of
AGAO, which oxidizes PEA and tyramine as efficient
substrates and benzylamine as a very poor substrate.
fer from TPQred to O2 but also by providing the
binding site for reduced dioxygen species (O2
ꢀ
,
HO2ꢀ, H2O2) (6, 10).
In contrast to the virtually identical catalytic mech-
anism of CAOs commonly promoted by the cooper-
ation of the TPQ cofactor, the conserved Asp residue,
and the bound copper ion, substrate specificities differ
considerably depending on the enzyme source. For ex-
ample, the CAOs from Escherichia coli (ECAO) (12)
and Arthrobacter globiformis (AGAO) (13) oxidize
2-phenylethylamine (PEA) and tyramine efficiently;
however, benzylamine and ethylamine are not efficient-
ly oxidized by these two enzymes. The enzyme from
yeast Hansenula polymorpha (HPAO-1) preferentially
acts on small aliphatic amines such as methylamine
and ethylamine (14, 15), whereas another isozyme
from the same yeast (HPAO-2) shows distinct catalytic
activity toward benzylamine (16). On the other hand,
CAOs from bovine serum (BSAO) (17) and a plant
Pisum sativum (18) efficiently catalyse the oxidation
of long-chain aliphatic diamines, including spermidine.
Structural bases for these substrate specificities have
been provided for several CAOs, whose X-ray crystal
structures are available. BSAO has a narrow substrate-
binding pocket with a negatively charged residue at
the entrance that should preferably bind a diamine
substrate (19). A CAO from yeast Pichia pastoris has
a shallow binding pocket suitable for short chain
168

Structure of TPQ—hydrazine complex
Scheme 2 Presumed reaction mechanism of hydrazine derivatives with TPQ.
Fig. 1 Chemical structures of amine substrates and hydrazine derivatives used in this study.
Active-site titration with hydrazine derivatives
Materials and Methods
The reactivities of BHZ and 4-OH-BHZ with AGAO were deter-
mined by a spectrophotometric titration, as described previously
for PHZ (3, 14, 28). Typically, 5 mM AGAO in 500 ml of 50 mM
HEPES buffer, pH 6.8, was titrated by adding 2 ml aliquots of a
0.1 mM freshly prepared solution of each compound. After incuba-
tion at 30ꢂC for at least 3 min to ensure the completion of the adduct
formation, the increase in absorbance at 380 nm and 384 nm for
Materials
AGAO was expressed in E. coli CD03 harbouring the expression
plasmid pEPO-02, purified to homogeneity as a copper-depleted pre-
cursor form, and finally converted to the TPQ-containing mature
form by incubation with excess copper ions, as described previously
(3). Protein concentrations were determined by measuring the ab-
sorbance at 280 nm using an extinction coefficient of 13.2 for 1%
(w/v) solution of the TPQ-containing form (3) and expressed as
molar concentrations of the subunit, unless otherwise stated. PHZ
was purchased from Nacalai Tesque (Kyoto, Japan); BHZ was from
Aldrich (St. Louis, USA); and PEA was from Wako Pure Chemicals
(Osaka, Japan). All other chemicals were of the highest grade com-
mercially available.
BHZ and 4-OH-BHZ, respectively, was monitored with
spectrophotometer.
a
Stopped-flow measurements
Rapid UV/visible spectral changes of the reaction of AGAO with
benzylamine in 50 mM HEPES buffer, pH 6.8, were monitored at
15ꢂC using a stopped-flow spectrophotometer SX.17MV (Applied
Photophysics, Leatherhead, UK). Typically, equal volumes (ꢃ30 ml
each) of the enzyme solution (0.2 mM AGAO in 50 mM HEPES
buffer, pH 6.8) and substrate solution (2 mM benzylamine) were
mixed in a mixing cell (volume, 20 ml) triggered with an N2-gas
piston; the mixing dead time was generally 2.3 ms at an N2-gas pres-
sure of 500 kPa. UV/visible absorption spectra were recorded every
2.56 ms over the wavelength range of 250—800 nm. Spectral data
were analysed using Pro-Kineticist II (Applied Photophysics) to re-
solve the spectra of the reaction intermediates and to calculate the
rate constants for each reaction step. To avoid spectral changes
associated with the oxidative half-reaction, both enzyme and sub-
strate solutions were kept in a vacuum-type glove box SGV-65 V (AS
ONE, Osaka, Japan) filled with 99.999% Ar gas for at least 6 h
before the stopped-flow measurements.
Synthesis of 4-OH-BHZ
To a solution of 4-benzyloxyphenylmethanol (4.70 g, 21.9 mmol) in
CH2Cl2 (50 ml), SOCl2 (2.87 g, 24.1 mmol) was added at room tem-
perature and the solution was refluxed for 1 h. After removal of
excess SOCl2 and the solvent by evaporation, the residue was dis-
solved in CHCl3 (100 ml), and washed with saturated Na2CO3 and
water. The organic layer was dried over Na2SO4 and concentrated
in vacuo to give 4-benzyloxybenzyl chloride (3.47 g, 68%) as a brown
oil; 1H NMR (CDCl3, 270 MHz) d ¼ 4.48 (2H, s, -CH2-Cl),
4.98 (2H, s, -CH2-OPh), 6.87 (2H, d, J ¼ 8.3 Hz, Ar), 7.21—7.35
(7 H, m, Ar) ppm.
A mixture of 4-benzyloxybenzyl chloride
(2.32 g, 10 mmol), N2H4ꢁH2O (5.0 g, 92.4 mmol), dioxane (10 ml)
and ethanol (3 ml) was refluxed for 10 h. After removal of the solv-
ent by evaporation, 10% HCl was added to the residue. The result-
ing precipitate was collected and washed with water.
Recrystallization from 2-propanol yielded colourless needles of
4-benzyloxybenzylhydrazine hydrochloride (1.91 g, 72%); 1H NMR
(DMSO-d6, 270 MHz) d ¼ 4.03 (2H, s, -CH2-NHNH2), 5.00 (2H, s,
-CH2-OPh), 6.94 (2H, d, J ¼ 8.3 Hz, Ar), 7.19—7.44 (7H, m, Ar)
X-ray crystallographic analysis
To complete the adduct formation with PHZ, BHZ and 4-OH-BHZ,
an AGAO solution (final 0.15 mM subunit) was mixed with each
hydrazine compound (final 0.3 mM) and incubated in a micro test
tube at 37ꢂC for 12 h before crystallization. After dialysis to re-
move the unreacted hydrazine, the enzyme was crystallized by the
micro dialysis method as described previously (7). The protein solu-
tion was placed in a 50 -ml dialysis button and dialysed at 16ꢂC
against 1.05 M potassium—sodium tartrate in 25 mM HEPES
buffer, pH 6.8. After sufficient growth of the crystals, the dialysis
buttons were transferred into the new reservoir solution supple-
mented with 45% (v/v) glycerol at 16ꢂC for 24 h as a cryoprotectant.
The crystals were mounted on thin nylon loops (ꢀ, 0.4—0.5 mm) and
ppm.
A solution of 4-benzyloxybenzylhydrazine hydrochloride
(1.10 g, 4.80 mmol) in methanol (50 ml) was hydrogenated over
10% Pd/C (140 mg) at 2 atm overnight. After filtering the catalyst
through Celite, the filtrate was evaporated and the residue was
recrystallized from 2-propanol/ether to give colourless needles of
4-OH-BHZ hydrochloride (0.66 g, 79%); 1H NMR (DMSO-d6,
270 MHz) d ¼ 3.96 (2H, s, -CH2-NHNH2), 6.78 (2H, d, J ¼ 8.3 Hz,
Ar), 7.23 (7H, d, J ¼ 8.3 Hz, Ar) ppm.
169

T. Murakawa et al.
frozen by flash cooling to 100 K in cold N2 gas stream. Before ex-
posure to the X-rays, the UV/visible absorption spectra of the crystals
were analysed by single-crystal micro-spectroscopy at 100 K (7, 9).
The micro-spectrophotometer system consisted of a deuterium
tungsten halogen light DT-MINI (Ocean Optics, FL, USA),
Cassegrainian mirrors (Bunkoh-Keiki Co. Ltd. Tokyo, Japan),
an optical fibre and a linear CCD-array spectrometer SD2000
(Ocean Optics). Absorption spectra were recorded over the wave-
length range of 178—879 nm and corrected for the air blank base-
line and dark reference.
files for refinement used for CNS were generated with XPLO2D in
the X-UTIL package (34). The residue 382 was replaced by a com-
plex of TPQ-BHZ, TPQ-4-OH-BHZ, or TPQ-PHZ, and the con-
formation was adjusted on the basis of the 2Fo — Fc, Fo — Fc, and
omit maps. These models were refined further through several
cycles of B-factor and positional refinements after introducing solv-
ent molecules followed by manual modelling with xfit. The quality of
the structure was validated with MolProbity (35). The details and
statistics of the data processing and crystallographic refinements are
summarized in Table 1. The atomic coordinates and structure factors
for the AGAO/BHZ, AGAO/4-OH-BHZ and AGAO/PHZ have
been deposited in the Protein Data Bank Japan at the Institute for
Protein Research, Osaka University (PDBj; http://www.pdbj.org/)
with the accession codes 2E2V, 2E2U and 2E2T, respectively.
Diffraction data sets were collected at 100 K with the synchrotron
˚
X-radiation (ꢁ ¼ 0.9 A) using an imaging plate, DIP6040 (Bruker
AXS, Yokohama, Japan) in the station BL44XU at the SPring-8
(Hyogo, Japan). The collected data was processed and scaled using
MOSFLM (29) and SCALA in CCP4 (30), respectively. The crystals
of AGAO reacted with BHZ (AGAO/BHZ) and 4-OH-BHZ
(AGAO/4-OH-BHZ) were found to belong to the space group I2,
which is identical to the AGAO crystals determined under the cryo
conditions and contains the dimer molecule in the asymmetric unit
(6) (PDB accession code, 1IU7). The crystals of AGAO reacted with
PHZ (AGAO/PHZ) belonged to the space group C2, which is iden-
tical to those of the precursor AGAO (apo form) determined at an
ambient temperature and contains a single subunit in the asymmetric
unit (31) (1AVK). For the phase determination, molecular replace-
ment was performed using a homo dimer model in AGAO/BHZ and
AGAO/4-OH-BHZ and a monomer model in AGAO/PHZ. The
wild-type AGAO structure (1IU7) without solvent molecules was
used as a search model for molecular replacement for AGAO/
BHZ and AGAO/4-OH-BHZ with TPQ382 being replaced with an
unmodified Tyr residue, whereas the precursor structure (1AVK)
was used without any modification for AGAO/PHZ. The program
used for molecular replacement, refinements, calculation of the
electron-density maps and assignment of solvent molecules was
CNS Version 1.1 (32). Manual rebuilding was performed using an
xfit module in an XtalView software package (33). Following
rigid-body refinement, the initial structures of the AGAO/PHZ,
AGAO/BHZ and AGAO/4-OH-BHZ were further refined through
simulated annealing with a slow cooling protocol from 2500 to 300 K
at 25 K-drop per cycle, and several cycles of B-factor and positional
refinements. In the modelling stage, the models of the TPQ-
hydrazine complexes were built for hydrazone and azo forms using
Insight II (Accelrys, CA, USA), and then topology and parameter
Results
Transient-state kinetic analysis with benzylamine
We previously determined the substrate specificity of
AGAO by using various amine substrates (9, 13). In
general, substrates with an aromatic (or heterocyclic)
ring at an opposite end of the aliphatic amino group
were found to be good substrates for AGAO with large
kcat/Km values. However, benzylamine, which also con-
tains a benzene ring, served as a very poor substrate
(Km ¼ 100 mM, kcat ¼ 0.24 sꢀ1) as compared to PEA
(2.5 mM, 44 sꢀ1) and tyramine (10.4 mM, 35 sꢀ1) when
measured at 15ꢂC (7, 9, 36). Since the first reaction
product aldehyde from these substrates containing
the distal aromatic ring is released in the former reduc-
tive half-reaction and therefore the latter oxidative
half-reaction should proceed irrespectively of the amine
substrates (Scheme 1), the low reactivity of benzyla-
mine, particularly in view of its small kcat value, may
be ascribed to a lowered reaction rate(s) in a step(s) of
the reductive half-reaction. To identify the reaction
step(s) accounting for the markedly decreased
Table I. Statistics of data collection and crystallographic refinement.
AGAO/BHZ
AGAO/4-OH-BHZ
AGAO/PHZ
Data collection
Temperature (K)
100
0.9
I2
100
0.9
I2
100
0.9
C2
˚
Wavelength (A)
Space group
Unit cell dimensions
˚
a, b, c (A),
ꢂ (deg)
158.05, 62.64, 183.95
157.97, 63.07, 184.05
157.64, 63.08, 91.86
112.16
652,680
175,433
3.7
26.5—1.72
99.2
7.1
112.04
720,500
189,031
3.8
26.6—1.68
98.9
5.8
112.16
329,900
87,299
3.8
26.5—1.73
100.0
5.7
No. of observations
No. of unique reflections
Multiplicity
˚
dmax — dmin (A)
Overall completeness (%)
Overall Rmerge (%)a
Refinements
˚
dmax — dmin (A)
No. of solvent atom
26.5—1.80
1205
26.6—1.68
1218
26.5ꢀ2.05
318
RMS deviation from ideal valuesb
˚
Bond lengths (A)
0.006
1.38
21.7
25.2
5
0.005
1.35
20.9
23.6
5
0.005
1.36
20.5
24.3
5
Bond angles (deg)
Residual R (%)c
Free residual R (%)d
Fraction of data in Rfree set (%)
Ramachandran plot
Favoured/allowed/outlier (%)
96.3/3.5/0.2
96.0/3.8/0.2
95.8/4.0/0.2
aRmerge ¼ Æh Æi jIh,i —5Ih4j / Æh Æi Ih,i, where Ih,i is the intensity value of the ith measurement of h and5Ih4is the corresponding mean value
c
of Ih for all i measurements.bRMS deviation from ideal values were calculated with CNS. R ¼ ÆjjFoj — jFcjj / ÆjFoj. dFree residual R is an R
factor of the CNS refinement evaluated for 5% of the reflections that were excluded from the refinement.
170

Structure of TPQ—hydrazine complex
reactivity of benzylamine, we conducted
a
reductive half-reaction with benzylamine proceeds
with multiple intermediates. Although these spectral
changes apparently differ from those observed in the
reactions with PEA (7) and tyramine (36), the same
model has been applied to the oxidation of benzyla-
mine in solving the mechanism by global analysis of
the spectral changes. The multi-wavelength data of
spectral changes with benzylamine were therefore fitted
to the four-step model (Scheme 3) (7, 36), where the
absorption spectrum of apo AGAO (without TPQ)
was used as that of TPQred, assuming that TPQred ex-
hibits no absorption 4300 nm (25). In the data fitting,
the absorption spectra of TPQox and TPQred were fixed,
whereas the spectra of other species and rate constants
at all steps were varied until a reasonable solution was
finally obtained. Shown in Fig. 2D and 2E are the
deconvoluted UV/visible absorption spectra and the
time courses of concentration changes of TPQox,
TPQssb, TPQpsb, TPQred and TPQsq. The absorption
transient-state kinetic analysis of the reductive
half-reaction with benzylamine under single-turnover
conditions.
Upon anaerobic mixing of AGAO with excess ben-
zylamine, the TPQox-derived absorption band at
480 nm slightly decreased with a concomitant peak
shift to 460 nm within 50 ms (Fig. 2A). An additional
shoulder band appeared around 310 nm. In the follow-
ing phase (from 50 ms to 3.0 s), the absorption band at
460 nm increased again and a shoulder band ꢃ340 nm
increased (Fig. 2B). In the final phase (from 3.0 to
16.0 s), the species with a shoulder band ꢃ340 nm
and with a ꢁmax at about 460 nm is gradually converted
to a new species with absorption peaks at 365, 438 and
465 nm, assignable to TPQsq (Fig. 2C) (6, 37).
Isosbestic points of the spectra were observed at 400
and 525 nm in the initial phase (Fig. 2A) and at 540 nm
in the intermediary phase (Fig. 2B), indicating that the
Fig. 2 Spectral changes during the reductive half-reaction of wild-type AGAO reacted with benzylamine. (A) UV/visible absorption spectra were
recorded after mixing AGAO (0.1 mM subunit) with 1.0 mM benzylamine in 50 mM HEPES buffer, pH 6.8, at 15ꢂC under anaerobic conditions.
The spectra obtained at (A) 0, 2.3, 6.4, 12, 17, 22, 27, 32, 37, 42 and 47 ms, (B) 0.1, 0.4, 0.8, 1.2, 1.6, 2.0, 2.6 and 3.0 s, and (C) 3, 4, 6, 8, 10, 12, 14
and 16 s are shown. Arrows indicate the direction of the spectral changes. Deconvoluted absorption spectra (D) and the time course of the
concentration changes of each intermediate (E) are shown for TPQox (black solid), TPQssb (gray solid), TPQpsb, (black dashed), TPQred (gray
dashed) and TPQsq (black dotted).
Scheme 3 Kinetic mechanism of the reductive-half reaction applied for global analysis of spectral changes.
171

T. Murakawa et al.
spectrum obtained for TPQssb, with a broad and weak
absorption 4400 nm, resembles that of a model com-
pound in a non-polar (CH2Cl2) or polar (CH3CN)
aprotic solvent (25). The spectrum obtained for
TPQpsb shows a broad peak at 460 nm with a higher
absorption coefficient. This absorption band consider-
ably red-shifted compared with those formed with
PEA and tyramine (ꢁmax ¼ 410 mm), probably due to
the wide resonance system extended from the TPQ ring
to the benzene ring of the substrate (cf. Scheme 1, D,
where R ¼ C6H5). The calculated spectrum for TPQsq
with absorption peaks at 360, 438 and 467 nm agrees
well with those obtained for the reductive half-reaction
with PEA and tyramine as substrates. The substrate-
independent spectrum of TPQsq is reasonable, because
TPQsq is formed after the hydrolysis of TPQpsb and the
release of the first product aldehyde in the reductive
half-reaction (Scheme 1).
Rate constants for the reaction with benzylamine ob-
tained by global analysis are summarized in Table 2,
together with those determined previously for the re-
actions with PEA (7) and tyramine (36). Significant
differences were observed for all rate constants except
for kþ4 and kꢀ4. In particular, rate constants of the pro-
ton abstraction step (kþ2, TPQssb ! TPQpsb) and its
reverse (kꢀ2), and also the step of TPQpsb hydrolysis
(kþ3, TPQpsb ! TPQred) and its reverse (kꢀ3) are
102—103 orders of magnitude smaller than those for
PEA and tyramine, indicating that the decreases in
these rate constants are presumably associated with
the low reactivity of benzylamine. The two steps appear
to be rate-limiting for the overall reaction with benzy-
lamine as the steady-state kcat value (0.24 sꢀ1) (9) is
comparable with the kþ2 and kþ3 values. The rate con-
stant of the TPQox ! TPQssb step (kþ1) is also 10- to
15-fold smaller in the reaction with benzylamine than
the reactions using PEA and tyramine as the sub-
strates. In contrast, the rate constant of the
TPQred ! TPQsq step (kþ4) is similar between the
three substrates, and this observation is again consist-
ent with this step proceeding after the release of the
product aldehyde and being independent of amine sub-
strate used.
X-ray crystal structure of AGAO reacted with
hydrazine derivatives
To elucidate the interactions of the active-site residues
with the hydrazine adducts, and thereby decipher those
with the reaction intermediates formed from the cor-
responding amine substrates, X-ray crystal struc-
tures of the complexes thoroughly reacted with the
hydrazine derivatives were determined. Three hydra-
zine derivatives, BHZ, 4-OH-BHZ and PHZ, which
are regarded as structural analogues for PEA, tyram-
ine and benzylamine, respectively, were used in this
study (Fig. 1). Prior to crystallization, the reactivities
of these hydrazine derivatives with AGAO were exam-
ined in solution by spectrophotometric titration; the
TPQ/hydrazine adducts are known to exhibit intense
and characteristic absorption bands in the visible
region (21—24). Previously, the titration of AGAO
with PHZ gave 0.63 mol of PHZ reacted with 1 mol
of the AGAO subunit (7). Similar results were ob-
tained for the titration with BHZ (0.64 mol/mol) and
4-OH-BHZ (0.78 mol/mol) in the present study.
Reaction of 51 mol of hydrazine derivatives per mol
of subunit has often been observed for many CAOs
and ascribed to a half-of-the-site reactivity of the
enzyme dimer (22, 38). However, structural evidence
for the active-site cooperativity between the two sub-
units of the dimer has not been obtained for any CAOs
including AGAO. It is possible that the less-than-
stoichiometric reactivities of hydrazine derivatives are
derived from the conformational flexibility of the TPQ
cofactor fluctuating between the productive (active)
‘off-copper’ and non-productive (inactive) ‘on-copper’
forms (15, 26, 39—42).
Crystals of AGAO complexed with BHZ, 4-OH-
BHZ and PHZ exhibited a deep yellow colour with
absorption maxima at 387, 375 and 435 nm, respect-
ively (Fig. 3), which are essentially identical with those
measured in solution as described above. The crystals
were then subjected to X-ray diffraction analyses. As
summarized in Table 1, all crystals of the AGAO/
BHZ, AGAO/4-OH-BHZ and AGAO/PHZ adducts
˚
diffracted to ꢃ1.7 -A resolution and therefore the dif-
fraction data were further processed to structure
refinements.
After several cycles of initial refinements of the
AGAO/BHZ structure, the Fo — Fc omit map around
residue 382 (TPQ) clearly showed an extra electron
density connected to the C5 position of TPQ, indicat-
ing the presence of a covalently-linked hydrazine de-
rivative. The structure of the hydrazine adduct was
modelled on the basis of the 2Fo — Fc and Fo — Fc
maps and also the Fo — Fc omit map. It has been re-
ported that hydrazine derivatives form a covalent
adduct with TPQ in either a hydrazone (N2—N1 ¼
C5; Fig. 4A) or azo (N2 ¼ N1—C5; Fig. 4B) form
(25, 26). To identify the predominant form in the
AGAO/BHZ structure, both azo and hydrazone
models were examined for fitting to the electron dens-
ity map. After several cycles of independent refine-
ments using the two models, R and Rfree values
decreased to essentially identical values (21.7% and
25.2%, respectively) for both models (Table 1).
Although the two models similarly fitted to the
Table II. Kinetic constants of each step of the reductive half-reaction
determined at 15ꢂCa.
Benzylamineb
PEAc
Tyramined
k
k
k
k
k
k
k
k
(sꢀ1
(sꢀ1
(sꢀ1
(sꢀ1
(sꢀ1
(sꢀ1
(sꢀ1
(sꢀ1
)
)
)
)
)
)
)
)
48
32
699 ꢄ 75
201 ꢄ 26
411 ꢄ 6.0
111 ꢄ 9.4
197 ꢄ 18
282 ꢄ 106
93 ꢄ 120
43 ꢄ 28
601 ꢄ 4.7
165 ꢄ 7.1
250 ꢄ 8.0
57 ꢄ 4.3
108 ꢄ 0.58
47 ꢄ 11
þ1
ꢀ1
þ2
ꢀ2
þ3
ꢀ3
þ4
ꢀ4
0.405
0.089
0.31
0.074
71
83 ꢄ 17
24
43 ꢄ 8.1
aReactions were analysed according to Scheme 3. bSD values for all
kinetic constants were 50.005 sꢀ1 after the Levenberg—Marquardt
minimization. cAdopted from Chiu et al. (7). dAdopted from
Murakawa et al. (36).
172

Structure of TPQ—hydrazine complex
0.2
0.15
0.1
0.05
0
0.2
0.15
0.1
A
B
C
0.15
0.1
0.05
0
0.05
0
300
400
500
600
300
400
500
600
300
400
500
600
Wavelength(nm)
Wavelength(nm)
Wavelength(nm)
Fig. 3 Absorption spectra of crystals of AGAO/hydrazine adducts. UV/visible absorption spectra were measured for the single crystals
(ꢃ0.2ꢅ0.4ꢅ0.1 mm) of the (A) AGAO/BHZ, (B) AGAO/4-OH-BHZ and (C) AGAO/PHZ adducts.
Fig. 4 X-ray crystal structure of the TPQ/BHZ adduct. Stereo diagrams of the refined models of the hydrazone form (A) and the azo form (B) of
the TPQ/BHZ adduct are shown with an annealed Fo — Fc omit map contoured at 4.4 s. Upper and lower panels are the views from two different
directions by a 90ꢂ rotation. Chemical structures for the hydrazone and azo forms are depicted on the right.
electron density map, close inspection of the aromatic
ring of BHZ in a view parallel to the TPQ ring revealed
that the hydrazone model fits better to the Fo — Fc omit
map than the azo model (Fig. 4A and B, lower panels).
The C10, N1 and N2 atoms are also better
accommodated to the electron density in the hydra-
zone model than in the azo model. Collectively, the
structure refinement strongly suggests that the hydra-
zone form is predominant in the crystal of the AGAO/
BHZ adduct.
173

T. Murakawa et al.
Fig. 5 X-ray crystal structures of the TPQ/4-OH-BHZ and TPQ/PHZ adducts. Stereo diagrams of the refined models of the hydrazone forms of
the TPQ/4-OH-BHZ adduct (A) and the TPQ/PHZ adduct (B) are shown with an annealed Fo — Fc omit map contoured at 4.4 s. Upper and
lower panels are the views from two different directions by a 90ꢂ rotation. Chemical structures for the hydrazone form are depicted on the right.
Similarly, structure refinements for the AGAO/
4-OH-BHZ adduct have revealed that the hydrazone
form is the major tautomer (Fig. 5A). Comparison of
the structures of AGAO/BHZ and AGAO/4-OH-BHZ
adducts indicated that the aromatic rings of both in-
hibitors occupy the same position in the active site of
AGAO, as described later. In contrast, in the crystal
structure of the AGAO/PHZ adduct, the electron dens-
ity in the Fo — Fc omit map was rather poor for the
aromatic ring of the inhibitor (Fig. 5B, upper panel).
If the predominant tautomer is an azo form (analogous
to TPQpsb), the inhibitor/quinone moiety has a fully
conjugated system and the aromatic ring of PHZ and
the quinone ring of TPQ should adopt a coplanar con-
formation (43). However, the electron density corres-
ponding to the aromatic ring of BHZ tilts upward
about 45ꢂ against the remaining part of the electron
density containing the quinone ring of TPQ (Fig. 5B,
lower panel), indicating that the coplanar azo form is
less feasible. Thus, it was concluded that the hydrazone
form is also predominant in the crystal structure of the
AGAO/PHZ adduct. It is noteworthy that in the
best-fit model of the hydrazone form of the AGAO/
PHZ adduct, the hydrogen at the N2 atom but not at
the N1 atom is hydrogen-bonding with the C4 oxygen
atom of the TPQ ring to form a six-member ring
(Fig. 5B). Finally, the overall structures of the
AGAO/PHZ, AGAO/BHZ and AGAO/4-OH-BHZ
adducts all refined in the hydrazone forms are essen-
tially identical with the structure of the wild-type,
ligand-free AGAO (PDB: 1IU7) (6), with root mean
square deviations for the main-chain atoms of 0.43,
˚
0.19 and 0.14 A, respectively.
Discussion
In this study, we have determined the crystal structures
of AGAO complexed with three hydrazine derivatives,
BHZ, 4-OH-BHZ and PHZ. The X-ray structural data
suggest that all the AGAO/hydrazine adducts occur
predominantly in a hydrazone form that is structurally
analogous to TPQssb. The crystal structure of the
AGAO/BHZ adduct has already been reported by an-
other group (PDB: 1W5Z), but without particular
174

Structure of TPQ—hydrazine complex
discussion with regard to the substrate specificity of
AGAO (44). The previously and presently reported
structures of the AGAO/BHZ adduct are virtually
identical with each other despite significant differences
in the preparations of recombinant AGAO (strep-
tagged or non-tagged), the conditions for adduct for-
mation (crystal soaking or pre-reacted in solution),
and the space groups of the crystals obtained (C2 or
I2). Furthermore, the crystal structure of ECAO com-
plexed with 2-hydrazinopyridine (2-HP) has been re-
ported, in which the hydrazone tautomer is also
predominant (26). In the crystal structure of the
ECAO/2-HP adduct, the hydrazone form appeared
to be stabilized by two hydrogen-bonding interactions
between the TPQ/2-HP moiety and two active-site resi-
dues: the catalytic base (Asp383, corresponding to
Asp298 in AGAO) and the conserved tyrosine residue
(Tyr369, corresponding to Tyr284 in AGAO) (26). The
latter hydrogen bond is also found in the structures of
the AGAO/hydrazine adducts (Fig. 6). However, there
is no hydrogen-bonding interaction between the aro-
matic ring of the hydrazine derivatives and the catalyt-
ic base (Asp298) of AGAO. Instead, hydrophobic
interactions between the aromatic ring of the bound
inhibitors and the amino acid residues constituting
the substrate-binding pocket presumably play an im-
portant role in stabilizing the hydrazone form, as
described below.
with the side-chains of Tyr302, Leu137, Tyr307,
Phe105, Pro136, Ala135, Leu358, Trp359, Trp168
and Phe407 (Leu358 and Trp359 are derived from
the neighboring subunit of the dimer) (in the order
shown in Fig. 6, clockwise from Tyr302), and held
nearly perpendicular against the TPQ ring with a di-
hedral angle of ꢃ85ꢂ for the planes formed by the
N1-N2-C10-C20 atoms (see also Figs 4A and 5A).
Furthermore, the binding of the aromatic ring of
4-OH-BHZ is stabilized by hydrogen bonds between
the 4-OH group and two water molecules (Wat2 and
Wat3) contained in the hydrophobic cavity, where
Wat2 is also hydrogen bonding with Wat1 which is
further interacting with the carboxyl group of
Asp357 (Fig. 6). Wat1 and Wat2 have also been
observed in the ligand-free AGAO structure (PDB:
1IU7) (6). These hydrophobic and hydrogen-bonding
interactions are essentially identical with those
observed in the intermediates (TPQssb
)
of
a
catalytic-base mutant (D298A) of AGAO formed in
the reaction with PEA and tyramine (7, 36).
Altogether, the aromatic rings of both BZH and
4-OH-BHZ are accommodated inside the hydrophobic
cavity of the substrate-binding pocket in a well-defined
conformation (Fig. 7). The substrate-binding pocket is
located at the bottom of the funnel-shaped substrate
channel (31, 45). In marked contrast to the aromatic
rings of BHZ and 4-OH-BHZ, hydrophobic inter-
actions of the aromatic ring of PHZ appear to be
only partially at one side and near the end of the
substrate-binding pocket in the structure of the
AGAO/PHZ adduct (Fig. 7). The poor electron dens-
ity in the Fo — Fc omit map (Fig. 5B) also suggests an
undefined conformation of the aromatic ring of PHZ,
The active-site structures of the ligand-free AGAO
and three AGAO/hydrazine adducts (AGAO/PHZ,
AGAO/BHZ and AGAO/4-OH-BHZ) are compared
in Fig. 6. It is clearly seen that the aromatic ring of
BZH and 4-OH-BHZ is tightly bound in the
substrate-binding pocket by hydrophobic interactions
Fig. 6 Comparison of the active-site structures. Active-site structures of the ligand-free AGAO (pale gray), AGAO/PHZ (black), AGAO/BHZ
(dark gray) and AGAO/4-OH-BHZ (white) adducts are superimposed in stick models. Broken arrows represent the distance between an oxygen
atom of the carboxyl group of Asp298 and the N2 atoms of PHZ, BHZ and 4-OH-BHZ. Three water molecules (Wat1, Wat2 and Wat3)
identified in the substrate-binding pocket of the AGAO/4-OH-BHZ adduct are shown as small spheres. Dotted lines represent hydrogen bonds.
175

T. Murakawa et al.
BHZ
PHZ
4-OH-BHZ
No ligand
Fig. 7 Comparison of the surfaces of the hydrophobic cavity of the substrate-binding pocket. The surfaces of the hydrophobic cavity of the
substrate-binding pocket are shown in stereo for the AGAO/BHZ (BHZ), AGAO/4-OH-BHZ (4-OH-BHZ), AGAO/PHZ (PHZ) adducts and
the ligand-free AGAO (No ligand). Bound inhibitors are shown by stick models.
even though the hydrazone form has been assigned as
the major tautomer. It is interesting to note that in the
structures of the AGAO/hydrazine adducts, several
residues constituting the substrate-binding pocket
have conformations different from those in the struc-
ture of ligand-free enzyme (Fig. 6). Namely, in the
AGAO/BHZ and AGAO/4-OH-BHZ structures, the
side chains of Tyr302, Tyr307, Phe105 and Leu358
moved closer to the aromatic ring of the inhibitors and
as a result the hydrophobic cavity became narrower
than that of the ligand-free enzyme (Fig. 7). Vice
versa, the residues Pro136, Phe105 and Leu137 in the
AGAO/PHZ structure have moved to make the cavity
even wider than the ligand-free enzyme (Fig. 7). These
findings strongly suggest that the substrate-binding
pocket of AGAO is formed, at least partially, by an
induced-fit movement of those residues upon binding
of the substrates (inhibitors).
N2 atom of the TPQ/hydrazine adducts corresponds to
the Ca atom of TPQssb. Thus, the distances between
the carboxyl group of Asp298 and the N2 atom of
hydrazone would represent those between the proton
donor (TPQssb) and acceptor (the carboxyl group of
Asp298) in the catalytic intermediate and may be cor-
related with the catalytic efficiencies of the substrate
amines.
As determined by transient-state kinetic analysis,
the rate constant of the proton abstraction step
(k
þ2
,
TPQssb
!
TPQpsb) in the reaction with
benzylamine is slower by ꢃ103-fold than those with
PEA and tyramine (Table 2). We previously reported
that the proton abstraction step is driven by quantum
mechanical proton tunnelling in the reductive
half-reaction of AGAO with both PEA and tyramine
(36). As the tunnelling is highly dependent on the
width and height of the energy barrier, which is closely
correlated with the proton donor-to-acceptor distance,
it is predicted that the contribution of the tunnelling in
the reaction with benzylamine, with a longer proton
donor-to-acceptor distance, is very small as compared
with those of PEA and tyramine. Although the extent
of the contribution of proton tunnelling in the reaction
with benzylamine has not been determined, the signifi-
cantly small rate constant (kþ2) at the proton abstrac-
tion step (TPQssb ! TPQpsb) is most likely derived
from the longer distance between the carboxyl group
of Asp298 and the Ca atom of TPQssb (N2 atom of
hydrazone).
Structural comparison of the AGAO/hydrazine ad-
ducts provides another important feature of the active
site. As the result of the binding of the inhibitor aro-
matic rings to the hydrophobic pocket near the TPQ
cofactor, distances between the side-chain carboxyl
group of the catalytic base (Asp298) and the N2
atom of the TPQ/hydrazine adducts are also fixed;
˚
they were estimated to be 2.86, 3.06 and 5.76 A in
the AGAO/BHZ, AGAO/4-OH-BHZ and AGAO/
PHZ adducts, respectively (Fig. 6). The distances of
˚
2.86 and 3.06 A suggest the presence of hydrogen
bonds between the deprotonated carboxyl group of
Asp298 and the N2 hydrogen atom in the hydrazine
forms of the AGAO/BHZ and AGAO/4-OH-BHZ ad-
ducts. Clearly, the distance in the AGAO/PHZ adduct
is about twice as long when compared with the dis-
tances in the AGAO/BHZ and AGAO/4-OH-BHZ ad-
ducts. As described above, the TPQ/hydrazone can be
regarded as a structural analogue of TPQssb formed in
the reaction with the corresponding amine substrates,
which is difficult to capture crystallographically unless
a catalytic-base mutant (D298A) is used (7, 36). The
The hydrolysis rate of the benzylamine-derived
TPQpsb (kþ3, TPQpsb ! TPQred) is also slower by sev-
eral hundred-fold than those derived from PEA and
tyramine (Table 2). We previously suggested that the
proton-abstracting catalytic base (Asp298) also partici-
pates in the hydrolysis of TPQpsb by raising the bas-
icity of the water molecule attacking the imine carbon
atom (Scheme 1, D0) (7). Thus, a longer distance be-
tween the carboxyl group of Asp298 and the imine
carbon atom of the presumed benzylamine-derived
176

Structure of TPQ—hydrazine complex
TPQpsb could be a reason for the greatly decreased rate
constant (kþ3) in this step; although the water molecule
involved in the hydrolysis of TPQpsb has not been iden-
tified crystallographically. Finally, the moderately
7. Chiu, Y.C., Okajima, T., Murakawa, T., Uchida, M.,
Taki, M., Hirota, S., Kim, M., Yamaguchi, H.,
Kawano, Y., Kamiya, N., Kuroda, S., Hayashi, H.,
Yamamoto, Y., and Tanizawa, K. (2006) Kinetic and
structural studies on the catalytic role of the aspartic
acid residue conserved in copper amine oxidase.
Biochemistry 45, 4105—4120
8. Suzuki, S., Okajima, T., Tanizawa, K., and Mure, M.
(2009) Cofactors of amine oxidases: Copper ion and its
substitution and the 2,4,5-trihydroxylphenylalanine
quinone, in Copper Amine Oxidases. Structures,
Catalytic Mechanisms, and Role in Pathophysiology
(Floris, G. and Mondovi, B., eds.), pp. 19—37, CRC
Press, Taylor & Francis Group, Boca Raton, FL
decreased rate constant in the step of TPQox
!
TPQssb in the reaction with benzylamine (kþ1, Table
2) is not interpretable in the absence of the crystal
structure of the Michaelis complex.
In conclusion, X-ray crystal structures of AGAO
complexed with three different hydrazine derivatives
have provided novel structural insights into the sub-
strate specificity of AGAO. Based on the structures of
the TPQ/hydrazine adducts, the reactivities of PEA
and tyramine as preferred substrates and also of ben-
zylamine as a poor substrate for AGAO have been
discussed. Furthermore, an induced-fit movement of
substrate (inhibitor)-binding residues has been identi-
fied for the first time in CAOs.
9. Taki, M., Murakawa, T., Nakamoto, T., Uchida, M.,
Hayashi, H., Tanizawa, K., Yamamoto, Y., and
Okajima, T. (2008) Further insight into the mechanism
of stereoselective proton abstraction by bacterial copper
amine oxidase. Biochemistry 47, 7726—7733
10. Shepard, E.M., Okonski, K.M., and Dooley, D.M.
(2008) Kinetics and spectroscopic evidence that the
Cu(I)-semiquinone intermediate reduces molecular
oxygen in the oxidative half-reaction of Arthrobacter glo-
biformis amine oxidase. Biochemistry 47, 13907—13920
Acknowledgements
The X-ray diffraction data were obtained at the beam line BL44XU
in SPring-8 with the approval of the Institute for Protein Research,
Osaka University.
11. Mure, M. and Klinman, J.P. (1995) Model studies of
topaquinone-dependent
Characterization of reaction intermediates and mechan-
amine
oxidases.
2.
Funding
ism. J. Am. Chem. Soc. 117, 8707—8718
This research is supported by Grants-in-Aid for Scientific Research
from Japan Society for the Promotion of Science: Young Scientists
(B) to T.M. (#11023838) and M.T. (#19750139); and Scientific
Research (B) to Y.Y. (#19350084) and T.O. (#18350085).
12. Roh, J.H., Suzuki, H., Azakami, H., Yamashita, M.,
Murooka, Y., and Kumagai, H. (1994) Purification,
characterization, and crystallization of monoamine oxi-
dase from Escherichia coli K-12. Biosci. Biotech.
Bbiochem 58, 1652—1656
13. Shimizu, E., Ohta, K., Takayama, S., Kitagaki, Y.,
Tanizawa, K., and Yorifuji, T. (1997) Purification and
properties of phenylethylamine oxidase of Arthrobacter
globiformis. Biosci. Biotech. Biochem. 61, 501—505
Conflict of interest
None declared.
References
14. Cai, D. and Klinman, J.P. (1994) Copper amine oxidase:
heterologous expression, purification, and characteriza-
tion of an active enzyme in Saccharomyces cerevisiae.
Biochemistry 33, 7647—7653
1. Knowles, P.F. and Dooley, D.M. (1994) Amine oxidases,
in Metal Ions in Biological Systems (Sigel, H. and Sigel,
A., eds.) Vol. 30, pp. 361—403, Marcel Dekker,
New York
15. Hevel, J.M., Mills, S.A., and Klinman, J.P. (1999)
Mutation of a strictly conserved, active-site residue
alters substrate specificity and cofactor biogenesis in a
copper amine oxidase. Biochemistry 38, 3683—3693
2. Floris, G. and Mondovi, B. (eds.) (2009) Copper Amine
Oxidases. Structures, Catalytic Mechanisms, and Role in
Pathophysiology, CRC Press, Taylor & Francis Group,
Boca Raton, FL
16. Chang, C.M., Klema, V.J., Johnson, B.J., Mure, M.,
Klinman, J.P., and Wilmot, C.M. (2010) Kinetic and
structural analysis of substrate specificity in two copper
3. Matsuzaki, R., Fukui, T., Sato, H., Ozaki, Y., and
Tanizawa, K. (1994) Generation of the topa quinone co-
factor in bacterial monoamine oxidase by cupric
ion-dependent autooxidation of a specific tyrosyl residue.
FEBS Lett. 351, 360—364
amine
oxidases
from
Hansenula
polymorpha.
Biochemistry 49, 2540—2550
17. Agostinelli, E., De Matteis, G., Mondovı, B., and
Morpurgo, L. (1998) Reconstitution of Cu2þ-depleted
bovine serum amine oxidase with Co2þ. Biochem. J.
330, 383—387
4. Cai, D. and Klinman, J.P. (1994) Evidence of a
self-catalytic mechanism of 2,4,5-trihydroxyphenyl-
alanine quinone biogenesis in yeast copper amine
oxidase. J. Biol. Chem. 269, 32039—32042
18. Pietrangeli, P., Federico, R., Mondovı, B., and
Morpurgo, L. (2007) Substrate specificity of copper-
containing plant amine oxidases. J. Inorg. Biochem.
101, 997—1004
19. Lunelli, M., Di Paolo, M.L., Biadene, M., Calderone, V.,
Battistutta, R., Scarpa, M., Rigo, A., and Zanotti, G.
(2005) Crystal structure of amine oxidase from bovine
serum, J. Mol. Biol. 346, 991—1004
20. Duff, A.P., Cohen, A.E., Ellis, P.J., Kuchar, J.A,
Langley, D.B., Shepard, E.M., Dooley, D.M.,
Freeman, H.C., and Guss, J.M. (2003) The crystal struc-
ture of Pichia pastoris lysyl oxidase. Biochemistry 42,
15148—15157
5. Choi, Y.H., Matsuzaki, R., Fukui, T., Shimizu, E.,
Yorifuji, T., Sato, H., Ozaki, Y., and Tanizawa, K.
(1995) Copper/topa quinone-containing histamine oxi-
dase from Arthrobacter globiformis. Molecular cloning
and sequencing, overproduction of precursor enzyme,
and generation of topa quinone cofactor. J. Biol.
Chem. 270, 4712—4720
6. Kishishita, S., Okajima, T., Kim, M., Yamaguchi, H.,
Hirota, S., Suzuki, S., Kuroda, S., Tanizawa, K., and
Mure, M. (2003) Role of copper ion in bacterial copper
amine oxidase: spectroscopic and crystallographic stu-
dies of metal-substituted enzymes. J. Am. Chem. Soc.
125, 1041—1055
177

T. Murakawa et al.
for macromolecular structure determination. Acta
Crystallogr. D54, 905—921
33. McRee, D.E. (1999) XtalView/Xfit—A versatile program
for manipulating atomic coordinates and electron dens-
ity. J. Struct. Biol. 125, 156—165
21. Janes, S.M. and Klinman, J.P. (1991) An investigation of
bovine serum amine oxidase active site stoichiometry:
evidence for an aminotransferase mechanism involving
two carbonyl cofactors per enzyme dimer. Biochemistry
30, 4599—4605
34. Kleywegt, G.J. and Jones, T.A. (1997) Model building
and refinement practice. Methods Enzymol. 277, 208—230
35. Chen, V.B., Arendall, W.B. III, Headd, J.J., Keedy,
D.A., Immormino, R.M., Kapral, G.J., Murray, L.W.,
Richardson, J.S., and Richardson, D.C. (2010)
MolProbity: all-atom structure validation for macromol-
ecular crystallography. Acta Crystallogr. D66, 12—21
36. Murakawa, T., Okajima, T., Kuroda, S., Nakamoto, T.,
Taki, M., Yamamoto, Y., Hayashi, H., and Tanizawa,
K. (2006) Quantum mechanical hydrogen tunneling in
bacterial copper amine oxidase reaction. Biochem.
Biophys. Res. Commun. 342, 414—423
37. Hirota, S., Iwamoto, T., Kishishita, S., Okajima, T.,
Yamauchi, O., and Tanizawa, K. (2001) Spectroscopic
observation of intermediates formed during the oxidative
half-reaction of copper/topa quinone-containing pheny-
lethylamine oxidase. Biochemistry 40, 15789—15796
22. Morpurgo, L., Agostinelli, E., Mondovi, B., Avigliano,
L., Silvestri, R., Stefancich, G., and Artico, M. (1992)
Bovine serum amine oxidase: half-site reactivity with
phenylhydrazine, semicarbazide, and aromatic hydra-
zides. Biochemistry 31, 2615—2621
23. Medda, R., Padiglia, A., Pedersen, J.Z., Rotilio, G.,
Finazzi Agro, A., and Floris, G. (1995) The reaction
mechanism of copper amine oxidase: detection of inter-
mediates by the use of substrates and inhibitors.
Biochemistry 34, 16375—16381
24. Bellelli, A., Morpurgo, L., Mondovı, B., and Agostinelli,
E. (2000) The oxidation and reduction reactions of
bovine serum amine oxidase. A kinetic study. Eur. J.
Biochem. 267, 3264—3269
25. Mure, M. and Klinman, J.P. (1993) Synthesis and spec-
troscopic characterization of model compounds for the
active site cofactor in copper amine oxidases. J. Am.
Chem. Soc. 115, 7117—7127
38. Frebort, I., Toyama, H., Matsushita, K., and Adachi, O.
´
(1995) Half-site reactivity with p-nitrophenylhydrazine
and subunit separation of the dimeric copper-containing
amine oxidase from Aspergillus niger. Biochem. Mol.
Biol. Int. 36, 1207—1216
26. Mure, M., Brown, D.E., Saysell, C., Rogers, M.S.,
Wilmot, C.M., Kurtis, C.R., McPherson, M.J., Phillips,
S.E.V., Knowles, P.F., and Dooley, D.M. (2005) Role of
the interactions between the active site base and the sub-
strate Schiff base in amine oxidase catalysis. Evidence
from structural and spectroscopic studies of the
2-hydrazinopyridine adduct of Escherichia coli amine
oxidase. Biochemistry 44, 1568—1582
27. Mure, M., Kurtis, C.R., Brown, D.E., Rogers, M.S.,
Tambyrajah, W.S., Saysell, C., Wilmot, C.M., Phillips,
S.E.V., Knowles, P.F., Dooley, D.M., and McPherson,
M.J. (2005) Active site rearrangement of the
2-hydrazinopyridine adduct in Escherichia coli amine
oxidase to an azo copper(II) chelate form: a key role
for tyrosine 369 in controlling the mobility of the
TPQ-2HP adduct. Biochemistry 44, 1583—1594
28. Mills, S.A. and Klinman, J.P. (2000) Evidence against
reduction of Cu2þ to Cuþ during dioxygen activation
in a copper amine oxidase from yeast. J. Am. Chem.
Soc. 122, 9897—9904
29. Leslie, A.G.W. (1992) Joint CCP4 and EESF-EACMB
Newsletter on Protein Crystallography. SERC
Daresbury Laboratory, Warrington, UK
39. Schwartz, B., Green, E.L., Sanders-Loehr, J., and
Klinman, J.P. (1998) Relationship between conserved
consensus site residues and the productive conformation
for the TPQ cofactor in a copper-containing amine oxi-
dase from yeast. Biochemistry 37, 16591—16600
40. Plastino, J., Green, E.L., Sanders-Loehr, J., and
Klinman, J.P. (1999) An unexpected role for the active
site base in cofactor orientation and flexibility in the
copper amine oxidase from Hansenula polymorpha.
Biochemistry 38, 8204—8216
41. Green, E.L., Nakamura, N., Dooley, D.M., Klinman,
J.P., and Sanders-Loehr, J. (2002) Rates of oxygen and
hydrogen exchange as indicators of TPQ cofactor orien-
tation in amine oxidases. Biochemistry 41, 687—696
42. Mure, M., Mills, S.A., and Klinman, J.P. (2002)
Catalytic mechanism of the topa quinone containing
copper amine oxidases. Biochemistry 41, 9269—9278
43. Wilmot, C.M., Murray, J.M., Alton, G., Parsons, M.R.,
Convery, M.A, Blakeley, V., Corner, A.S., Palcic, M.M.,
Knowles, P.F., McPherson, M.J., and Phillips, S.E.V.
(1997) Catalytic mechanism of the quinoenzyme amine
oxidase from Escherichia coli: Exploring the reductive
half-reaction. Biochemistry 36, 1608—1620
44. Langley, D.B., Trambaiolo, D.M., Duff, A.P., Dooley,
D.M., Freeman, H.C., and Guss, J.M. (2008) Complexes
of the copper-containing amine oxidase from
Arthrobacter globiformis with the inhibitors benzylhydra-
zine and tranylcypromine. Acta Crystallogr F64,
577—583
30. Collaborative Computational Project Number 4 (1994)
The CCP 4 suite: programs for protein crystallography.
Acta Crystallogr. D50, 760—763
31. Wilce, M.C., Dooley, D.M., Freeman, H.C., Guss, J.M.,
Matsunami, H., McIntire, W.S., Ruggiero, C.E.,
Tanizawa, K., and Yamaguchi, H. (1997) Crystal struc-
tures of the copper-containing amine oxidase from
Arthrobacter globiformis in the holo and apo forms: im-
plications for the biogenesis of topaquinone.
Biochemistry 36, 16116—16133
32. Brunger, A.T., Adams, P.D., Clore, G.M., DeLano,
¨
45. Matsuzaki, R. and Tanizawa, K. (1998) Exploring a
channel to the active site of copper/topaquinone-
containing phenylethylamine oxidase by chemical modi-
fication and site-specific mutagenesis. Biochemistry 37,
13947—13957
W.L., Gros, P., Grosse-Kunstleve, R.W., Jiang, J.S.,
Kuszewski, J., Nilges, M., Pannu, N.S., Read, R.J.,
Rice, L.M., Simonson, T., and Warren, G.L. (1998)
Crystallography & NMR system: a new software suite
178