Substrate-assisted O2 activation in a cofactor-independent dioxygenase

Article abstract of DOI:10.1016/j.chembiol.2013.11.013

In contrast to the majority of O₂-activating enzymes, which depend on an organic cofactor or a metal ion for catalysis, a particular group of structurally unrelated oxygenases is functional without any cofactor. In this study, we characterized the mechanism of O₂ activation in the reaction pathway of a cofactor-independent dioxygenase with an α/β-hydrolase fold, which catalyzes the oxygenolytic cleavage of 2-alkyl-3-hydroxy-4(1H)-quinolones. Chemical analysis and electron paramagnetic resonance spectroscopic data revealed that O₂ activation in the enzyme's active site is substrate-assisted, relying on single electron transfer from the bound substrate anion to O₂ to form a radical pair, which recombines to a C2-peroxide intermediate. Thus, an oxygenase can function without a cofactor, if the organic substrate itself, after activation to a (carb)anion by an active-site base, is intrinsically reactive toward molecular oxygen.

Full text of DOI:10.1016/j.chembiol.2013.11.013

Chemistry & Biology
Article
Substrate-Assisted O2 Activation
in a Cofactor-Independent Dioxygenase
Sven Thierbach,1 Nguyen Bui,2 Josef Zapp,3 Siri Ram Chhabra,4 Reinhard Kappl,2 and Susanne Fetzner1,
*
1Institute of Molecular Microbiology and Biotechnology, University of Mu¨ nster, Corrensstrasse 3, 48149 Mu¨ nster, Germany
2Department of Biophysics, School of Medicine, Saarland University, Clinical Center, Building 76, 66421 Homburg, Germany
3Pharmaceutical Biology, Saarland University, Campus, Building C2, 66123 Saarbru¨ cken, Germany
4School of Life Sciences, Centre for Biomolecular Sciences, University of Nottingham, University Park, Nottingham, NG7 2RD, UK
*Correspondence: fetzner@uni-muenster.de
http://dx.doi.org/10.1016/j.chembiol.2013.11.013
SUMMARY
bacter sp. Rue61a, an isolate from the biological wastewater
treatment plant of a coal tar refinery (Niewerth et al., 2012).
In contrast to the majority of O2-activating enzymes, Hod catalyzes the oxygenolytic cleavage of two C-C bonds of
its nitrogen-heteroaromatic substrates (Figure 1). Together with
genes coding for other enzymes of the degradation pathway,
which depend on an organic cofactor or a metal ion
for catalysis, a particular group of structurally unre-
lated oxygenases is functional without any cofactor.
In this study, we characterized the mechanism of
O2 activation in the reaction pathway of a cofactor-
independent dioxygenase with an a/b-hydrolase
fold, which catalyzes the oxygenolytic cleavage of
2-alkyl-3-hydroxy-4(1H)-quinolones. Chemical anal-
ysis and electron paramagnetic resonance spec-
troscopic data revealed that O2 activation in the
enzyme’s active site is substrate-assisted, relying
on single electron transfer from the bound substrate
anion to O2 to form a radical pair, which recombines
to a C2-peroxide intermediate. Thus, an oxygenase
can function without a cofactor, if the organic
substrate itself, after activation to a (carb)anion by
an active-site base, is intrinsically reactive toward
molecular oxygen.
the hod gene is part of a catabolic operon whose transcription
is upregulated when Arthrobacter sp. Rue61a is growing on qui-
naldine or on aromatic compounds downstream in the catabolic
pathway (Parschat et al., 2007). However, Hod also is able to
oxygenate and thus inactivate 2-heptyl-3-hydroxy-4(1H)-quino-
lone (a.k.a. the Pseudomonas quinolone signal [PQS]), a quorum
sensing signaling molecule of Pseudomonas aeruginosa that is
involved in the regulation of many virulence factors (Pustelny
et al., 2009).
Hod is devoid of any cofactor (Bauer et al., 1996) and belongs
to the a/b-hydrolase fold superfamily of proteins (Fischer et al.,
1999; Steiner et al., 2010). a/b-Hydrolases usually employ a
catalytic triad consisting of a nucleophilic residue, a strictly
conserved histidine, and an acidic residue for catalysis (Bugg,
2004; Nardini and Dijkstra, 1999). Remarkably, such a triad is
structurally conserved in Hod, comprising the residues S101,
H251, and D126 (Steiner et al., 2010) (Figure 1). Residues H251
and D126, which form a charge-relay tandem, have a key role
in catalysis. Steady-state kinetic and UV spectroscopic data
indicated that upon binding of the physiological substrate
3-hydroxy-2-methyl-4(1H)-quinolone (MHOQ) to the enzyme,
INTRODUCTION
Oxygenases catalyze the incorporation of oxygen atoms from H251 acts as a general base to form an MHOQ anion (Frer-
molecular oxygen into their organic substrates. Oxygen is ichs-Deeken et al., 2004). From the crystal structure of Hod in
chemically unreactive in its triplet ground state, because its complex with MHOQ, it was deduced that deprotonation by
direct reaction with a singlet species to form singlet products the His/Asp dyad selectively occurs at the 3-OH group of the
is ‘‘spin-forbidden.’’ Enzymes that use O2 as substrate have to substrate (Steiner et al., 2010). However, the hypothesis of
overcome this barrier to reaction. One possible strategy is initial a general base mechanism does not address the key question
activation of the organic substrate to a radical, which can react of how the enzyme satisfies the quantum rule.
with O2 in a spin-allowed process. Another strategy is to activate
Two possible mechanisms have been proposed for cofactor-
O2, either by orbital overlap with a suitable transition metal ion, or independent enzymatic O2 activation. For urate oxidase, activa-
by an electron transfer from a potent electron donor to form a tion of the enzyme-bound urate dianion to a urate monoanion
d
reduced oxygen species. Reduced flavin or pterin cofactors radical and enzyme radical (EH ) occurs without participation
and some metal centers can effectively catalyze such electron of dioxygen. Subsequent O2 activation is thought to involve
d
transfer reactions (Bugg, 2003). The vast majority of oxygenases electron transfer from EH to form the superoxide anion radical
indeed depend on an organic cofactor or transition metal ion (Gabison et al., 2011). For Hod, we have suggested that it
for catalysis. In recent years, however, several oxygenases and activates O2 by electron transfer from the enzyme-bound sub-
oxidases have been identified that neither contain nor require strate anion to form a [substrate radical – superoxide anion
any cofactor (reviewed in Fetzner and Steiner, 2010), which radical] pair, which recombines to a peroxide intermediate
raises the question of how these enzymes work.
(Frerichs-Deeken et al., 2004; Steiner et al., 2010). Substrate-
1H-3-hydroxy-4-oxoquinaldine 2,4-dioxygenase (Hod) is in- mediated O2 activation has also been discussed as catalytic
volved in quinaldine (2-methylquinoline) utilization by Arthro- strategy of 3,5-dihydroxyphenylacetyl-CoA 1,2-dioxygenase
Chemistry & Biology 21, 217–225, February 20, 2014 ª2014 Elsevier Ltd All rights reserved 217

Chemistry & Biology
Substrate-Assisted Enzymatic O2 Activation
Table 1. Apparent Steady-State Kinetic Parameters of HodC and
HodC-W160A
K
M (mM)
MHOQ
2.7 (0.3)a
kcat (sÀ1
)
Protein
HodC
BHOQ
MHOQ
BHOQ
29.4 (5.3)a 38.4 (0.3)a 41.2 (2.5)a
HodC-W160A 272.1 (27.7) 167.6 (22.4) 0.93 (0.05) 0.12 (0.01)
The average of at least three independent experiments is given with the
SD in parentheses. See also Figure S1.
aData from Steiner et al. (2010).
O2 binding, but probably affects binding and orientation of
the substrate and possibly also has an impact on later steps in
catalysis. In order to enlarge the substrate cavity, we replaced
W160 with alanine by site-directed mutagenesis of the hodC69S
gene and purified the HodC-W160A protein (see the Supple-
mental Experimental Procedures available online). Compared
to HodC, HodC-W160A showed a 41-fold decrease in the
apparent kcat and a 100-fold higher apparent KM for the physio-
logical substrate MHOQ (Table 1). However, its apparent KM for
O2 (879 214 mM) was similar to that of HodC (805 146 mM),
suggesting that O2 binding is not significantly perturbed by the
amino acid substitution. The apparent KM of HodC for 2-butyl-
3-hydroxy-4(1H)-quinolone (BHOQ) is ꢀ11-fold higher than for
Figure 1. 1H-3-hydroxy-4-oxoquinaldine 2,4-dioxygenase
(A) Cartoon representation of HodC, with the a/b-hydrolase fold core domain
and cap domain in light and dark gray, respectively. Red dots indicate the
positions of the triad residues (S101, D126, H251).
(B) Active site of HodC complexed with MHOQ; the residues of the conserved
triad, the cap domain residue W160, and MHOQ are shown in stick repre-
sentation. The cyan mesh indicates the putative O2 position.
(C) Reaction catalyzed by Hod. R = H, n-alkyl (C1 to C9). (A) and (B) were
generated from Protein Data Bank (PDB) codes 2WJ3, 2WJ4, and 2WM2,
respectively, using the PyMOL Molecular Graphics System, Version 1.3
¨
(Schrodinger).
DpgC (Tseng et al., 2004; Widboom et al., 2007), the quinone- MHOQ, whereas the apparent kcat values are similar (Table 1),
forming monooxygenases SnoaB and ActVA-orf6 (Grocholski indicating that the extended alkyl moiety affects substrate bind-
et al., 2010; Sciara et al., 2003), the two-component monooxyge- ing rather than turnover. In case of the HodC-W160A protein, a
nase SnoaW/SnoaL2 that catalyzes a cosubstrate-dependent 1.6-fold decrease of the apparent KM for BHOQ (compared to
but cofactor-independent hydroxylation of an anthracycline its KM for MHOQ) and an 8-fold lower apparent kcat for BHOQ
substrate in nogalamycin biosynthesis (Siitonen et al., 2012), suggest a somewhat stronger, but less productive binding of
and Renilla luciferase that is an imidazolopyrazinone monooxy- the substrate with the longer alkyl chain. Thus, positioning of
genase (Loening et al., 2006; Woo et al., 2008). Among the the organic substrates, especially of BHOQ, within the active
cofactorless oxidases, electron transfer from enzyme-bound site seems to be less accurate or more flexible in HodC-
(deprotonated) substrate to O2 has been proposed for Aln6, W160A than in HodC. The replacement also might increase
which is involved in the biosynthesis of the aromatic polyketide conformational flexibility of the protein and thus weaken the
alnumycin (Oja et al., 2012), and for PqqC that catalyzes the final shielding of the active site from solvent.
step of pyrroloquinoline quinone biosynthesis (Bonnot et al.,
To verify that HodC-W160A catalyzes the same reaction as
2013; RoseFigura et al., 2011). Such an activation of O2 by the the wild-type enzyme, the extracted organic product of MHOQ
organic substrate itself is analogous to the reaction of reduced conversion was identified by HPLC and comparison with the
flavins with dioxygen (Bruice, 1984; Chaiyen et al., 2012; authentic reference compound as N-acetylanthranilic acid (see
Massey, 1994; Palfey et al., 1995), however, direct proof for Figure S1); as expected, mass spectrometric analysis of the
the hypothesis that this mechanism is effective in cofactorless product showed a molecular ion peak [M+H]+ at mass-to-charge
oxygenases is lacking. In this study, we combined chemical ratio (m/z) 180 and fragments at m/z 162 (15%) [C9H7NO2+H]+,
and spectroscopic methods to unravel the reaction pathway 138 (10%) [C7H7NO2+H]+, and 120 (100%) [C7H5NO+H]+.
of Hod and to provide evidence that cofactor-independent Electrospray ionization-mass spectrometry (ESI-MS) analysis
oxygenases mediate substrate-assisted O2 activation.
of the product of BHOQ conversion by HodC or HodC-W160A
revealed a peak of m/z 222, corresponding to the [M+H]+ ion
of N-pentanoylanthranilic acid. Fragmentation resulted in ions
with m/z 204 (4%) [C12H13NO2+H]+, 138 (19%) [C7H7NO2+H]+,
and 120 (100%) [C7H5NO+H]+. Carbon monoxide released
RESULTS
Catalytic Properties of HodC and HodC-W160A
The crystal structure of HodC (the C69S mutant of Hod) sug- was detected spectrophotometrically as CO-hemoglobin, as
gested that the cap domain residue W160 plays an important described previously (Pustelny et al., 2009). The possibility of a
role in shaping the catalytic pocket, imposing an inward curva- 2,3-dioxygenolytic side reaction to form an a-oxo acid, and
ture on the surface of the active site cavity (Steiner et al., subsequent hydrolysis to formate and the corresponding
2010). Because enzyme-bound MHOQ is sandwiched between N-acylanthranilic acid, was examined with formate dehydro-
W160 and a putative O2 cavity at a basin of the active site pocket genase. In the assay, whose sensitivity would have enabled
(Steiner et al., 2010) (see also Figure 1B), we assumed that the detection of %0.5% of side product from the organic
substitution of W160 by a small residue does not interfere with substrates, we did not observe any activity, confirming that
218 Chemistry & Biology 21, 217–225, February 20, 2014 ª2014 Elsevier Ltd All rights reserved

Chemistry & Biology
Substrate-Assisted Enzymatic O2 Activation
concentration than that reported for an analogous experiment
with urate oxidase (Gabison et al., 2011). No EPR signal was
observed at room temperature, 70 K, or 20 K for anoxic solutions
of up to 2 mM enzyme and 5 mM MHOQ, even after 4 hr of
spectra accumulation, suggesting that interaction of MHOQ
with Hod does not generate radical species, differently from
the reaction in urate oxidase.
In an attempt to detect oxidizing intermediates that might
occur during the Hod reaction, the cyclic hydroxylamine
spin probe 1-hydroxy-3-methoxycarbonyl-2,2,5,5-tetramethyl-
pyrrolidine (CMH) was added to the reaction mixtures. CMH is
a one-electron reductant that reacts very fast with the superox-
ide anion radical (k ꢀ104 MÀ1
s
À1) to the CM-nitroxide radical
(CM ), but it is also known to be susceptible to other oxidative
d
d
species (Dikalov et al., 2011; Ouari et al., 2011). CM is very
stable and elicits a typical 3-line EPR spectrum whose temporal
d
evolution was recorded (Figure S2). CM formation during
conversion of MHOQ or BHOQ by wild-type HodC protein in
the presence of O2 was very slow and followed linear kinetics
(Figure 2, solid red traces). In contrast, significant amounts of
d
CM were formed in a nonlinear kinetics when CMH was present
during the HodC-W160A catalyzed reaction (Figure 2, solid black
traces). The decrease in the CMH oxidation rate can be attrib-
uted to the decelerating enzymatic reaction due to substrate
depletion, as illustrated by the kinetics of the signal of the
tetrathiatriarylmethyl radical (TAM) oxygen probe that is also
monitored by EPR (Figure 2, dashed black lines). The initial rates
of O2 consumption observed in these assays approximately
corresponded to the rates of substrate conversion, measured
spectrophotometrically under conditions identical to those of
the EPR experiments. In case of the HodC-W160A-catalyzed
conversion of BHOQ, for example, the initial rates of O2
consumption (Figure 2B, black dashed trace) and BHOQ con-
version were ꢀ11 mM minÀ1 and ꢀ12 mM minÀ1, respectively.
CMH oxidation during the HodC-W160A reaction was not
quenched by catalase or superoxide dismutase (SOD), and
d
Figure 2. Kinetics of CM Formation and O2 Consumption during
the Enzyme-Catalyzed Conversion of MHOQ and BHOQ
d
Kinetics of CM formation and O2 consumption during the enzyme-catalyzed
conversion of MHOQ (A) and BHOQ (B). The contents of the assay mixtures
d
are indicated in the insets. Solid lines, formation of CM ; dashed lines, con-
sumption of O2, as deduced from the time course of the TAM signal. Constant
intensities of the TAM signal from control experiments were defined to
correspond to 100% of dissolved O2. The stable minimum of the TAM intensity
after conversion of MHOQ and BHOQ by HodC or HodC-W160A, respectively,
was assumed to represent 0% O2. Traces are the means of at least three
replicates. During oxygenolysis of BHOQ by HodC-W160A, CMH oxidation
proceeded with an initial rate of 2.11 ( 0.05) mM minÀ1. The initial CMH
oxidation rate was ꢀ4.5-fold lower when MHOQ was converted. The already
d
addition of H2O2 did not significantly increase CM formation
(data not shown). The additional presence of triphenylphosphine
(TPP), a reagent selective for organic hydroperoxides (see
below), during the enzymatic BHOQ oxidation did not signifi-
cantly interfere with CMH oxidation (Figure 2B, gray traces).
In oxic protein solutions without an organic substrate, forma-
d
high content of CM at start time is explained by the delay of 1–2 min between
sample preparation and measurement of the first EPR spectrum. The CMH
oxidation rate finally reached a value of 0.03 ( 0.04) mM minÀ1 (0.06 [ 0.02] mM
minÀ1 in case of the MHOQ oxidation), which is similar to the constant rate of
CMH oxidation in substrate reference solutions (green traces).
See also Figure S2.
d
tion of CM was negligible (Figure 2, blue and pink traces), dis-
proving the possibility that the CMH oxidation observed during
the reaction is due to a reactive oxygen species that is formed
by direct O2 activation by the enzyme. The observation that
anoxic mixtures of HodC-W160A and substrate likewise showed
d
HodC-W160A catalyzes the 2,4-dioxygenolytic cleavage of
MHOQ and BHOQ.
only background formation of CM (comparable to that observed
in oxic substrate reference solutions) argues against the possibil-
ity that the CMH oxidizing species derives from direct substrate
activation to a radical by the enzyme. Taken together, the results
clearly indicate that formation of the reactive CMH-oxidizing
Oxidizing Species Are Formed during the Hod-Catalyzed
Reaction
A KD value of 2.66 0.26 mM was determined for the anoxic species by HodC-W160A was strictly dependent on the pres-
complex of HodC and MHOQ by fluorescence titration, indi- ence of both the organic substrate and O2.
cating tight interaction of the substrate with the enzyme in the
When HodC-H251A, which still binds MHOQ but due to its
absence of O2. To detect radicals possibly generated in Hod inability to deprotonate the substrate is virtually inactive, or
upon substrate binding, we performed electron paramagnetic HodC-W160A that was inactivated by covalent histidine modi-
resonance (EPR) experiments with anoxic mixtures of HodC fication using diethyl pyrocarbonate, were tested for CMH
and its natural substrate MHOQ, using an even higher enzyme oxidation in the presence of organic substrate and O2, the rates
Chemistry & Biology 21, 217–225, February 20, 2014 ª2014 Elsevier Ltd All rights reserved 219

Chemistry & Biology
Substrate-Assisted Enzymatic O2 Activation
with a sufficiently low potential to drive the reduction of O2 to
form the postulated [S – O2 À] pair. CMH shows two peak
d
d
d
pairs in CV, corresponding to the transitions CMH 4 CM and
d
CM 4 CMox (Figure S3A). The pH dependence of the first
À
d
transition is similar to that of MHOQ (S /S ), but the midpoint
potential is lower by ꢀ70 mV (Figure S3B). This implies that
CMH over a wide pH range is able to reduce a substrate radical
dÀ
formed at the reaction site. Of course, CMH can reduce O2
because the midpoint potential of the O2 À/O2
2À
d
couple
(+0.74 V versus Ag/AgCl) is much more positive.
Considering the redox behavior of the organic substrate
and CMH, the observed stoichiometry of one H2O2 and two
d
CM formed during BHOQ oxidation by HodC-W160A can be
explained by the reaction of CMH with substrate radical as well
as superoxide, according to Equation 1:
d
Figure 3. Concentrations of CM and H2O2 Formed during the HodC-
W160A Catalyzed Oxidation of Different Concentrations of BHOQ
All reactions were incubated until full conversion of the organic substrate. Final
d
CM concentrations were corrected for background oxidation, and the
ꢁ
ꢁ
CMH + S /CM + SÀ + H+
corresponding H2O2 concentrations were corrected for the spontaneous
decay of H2O2 as deduced from control experiments. The bars represent the
means of at least three independent replicates with their SEM.
See also Figure S3.
ꢁ
CMH + Oꢁ2 À/CM + HOÀ2 :
(Equation 1)
d
Alternatively, after reduction of S by CMH to regain the sub-
strate anion, an electron can be transferred from SÀ, whose
reduction potential is sufficiently low, to nearby O2 À to produce
d
d
of CM formation were similar to those observed in oxic sub-
2À
O2 and again the substrate radical, which in turn is reduced
by a second CMH molecule:
strate reference solutions (data not shown), confirming that
CMH oxidation during the HodC-W160A reaction is associated
with turnover.
CMH + S /CM + SÀ + H+
ꢁ
ꢁ
When the nitrone spin traps 5,5-dimethyl-1-pyrroline N-oxide
(DMPO; 200 mM) or a-(4-pyridyl N-oxide)-N-tert-butylnitrone
(POBN; 200 mM), were used instead of CMH under the same
assay conditions, no EPR signal was observed even after pro-
longed spectra accumulation. Therefore, a direct identification
of carbon- or oxygen-centered radical species from the EPR
signatures of spin adducts was not possible. The absence
of any adduct signals also shows that radical intermediates,
if present, are not released into the solvent.
ꢁ
SÀ + Oꢁ2 À/S + O22À
CMH + S /CM + SÀ + H+ :
(Equation 2)
ꢁ
ꢁ
In Equation 2, an independent approach of two CMH molecules
to the substrate radical and to superoxide is not required. Rather,
the substrate radical is used as a relay to shuttle an electron from
the first CMH to nearby superoxide, avoiding possible structural
restrictions for the second CMH to access the superoxide that
may be shielded by the bound substrate species (Steiner et al.,
2010).
Stoichiometry of the Products of CMH Oxidation
Whereas neither the bulky SOD nor the nitrone spin traps appear
to be able to sufficiently approach the active site pocket for
reaction, CMH seems to come close enough for allowing an
electron transfer. To find out whether CMH reduces superoxide
Reduction of the substrate radical by CMH, as suggested in
both equations 1 and 2, regenerates the substrate. This should
result in an apparently lower rate of substrate consumption in
the enzymatic reaction if CMH is present. The effect of CMH
(600 mM) on BHOQ turnover by HodC-W160A was determined
in a spectrophotometric assay that monitors BHOQ consump-
2À
in the active site to form O2 or HO2À, the H2O2 content after
BHOQ conversion by HodC-W160A in the presence of CMH
d
was measured and compared with the amount of CM . Interest-
tion. The initial rate of BHOQ conversion (13.56
0.14 mM
ingly, the oxidation of CMH during the conversion of BHOQ
by HodC-W160A yielded one equivalent of H2O2 from two
equivalents of CMH, independent of the amount of substrate
converted by the enzyme (Figure 3).
minÀ1) was reduced by ꢀ12% (11.91 0.18 mM minÀ1), support-
ing the proposal of substrate recycling in the presence of CMH.
For a more detailed understanding of the electron transfer Release of an Organic Hydroperoxide from the
processes involved, the redox behavior of MHOQ, BHOQ and Hod-Catalyzed Reaction
CMH was analyzed by cyclic voltammetry (CV) (see Figure S3). Following formation of the [S – O2 À] pair, the next step in
d
d
À
d
The midpoint potential of S /S (MHOQ) linearly decreases the proposed Hod reaction pathway involves formation of a
as a function of pH, and for pH values above 12, becomes C2- (or C4-) peroxide anion or C2- (or C4-) hydroperoxide
sufficiently negative to thermodynamically favor the transfer of (Frerichs-Deeken et al., 2004; Steiner et al., 2010). Organic
an electron from SÀ to O2 (Figure S3B). According to pKa calcu- hydroperoxides (but not cyclic peroxides or H2O2) convert TPP
lations, substrate monoanions are prevailing between pH 11 and to triphenylphosphine oxide (TPPO) (Bou et al., 2008); control
13.5 (Figure S3C). A local pH above ꢀpH 12 in the protein’s experiments verified that H2O2 does not oxidize TPP significantly
active site would produce a monoanionic substrate species under our assay conditions. Whereas the HodC-catalyzed
220 Chemistry & Biology 21, 217–225, February 20, 2014 ª2014 Elsevier Ltd All rights reserved

Chemistry & Biology
Substrate-Assisted Enzymatic O2 Activation
178 (ꢀ2%) for [C9H7NO3+H]+, 160 (ꢀ1%) for [C9H5NO2+H]+,
146 (13%) for [C9H7NO+H]+, and 132 (<1%) for [C8H5NO+H]+.
Except for the fragment ion with m/z 216, which probably results
from preferred water elimination from the proposed 2-butyl-2-
hydroxy-1,2-dihydroquinoline-3,4-dione, a similar fragmentation
pattern was reported for 3-alkyl-2,3-dihydroxy-4(1H)-quinolones
´
by Lepine et al. (2004). To analyze the structure and determine
the position of the OH-group in the proposed compound, 1D
and 2D nuclear magnetic resonance (NMR) methods were
applied. In the 1H NMR spectrum the resonances for the n-butyl
group were found between 0.7 and 1.7 ppm (H-9–H-12) showing
the expected multiplicities. The assignments followed directly
from the correlations in the proton-proton shift correlation
spectroscopy (HHCOSY). In the low field part of the spectrum
four aromatic protons (H-5–H-8) appeared between 7.0 and
7.7 ppm, indicating a 1,2-disubstituted benzene moiety by its
characteristic coupling pattern. Two singlets at 10.74 and 5.65
ppm suggested OH and NH protons. Apart from signals for
the n-butyl and the 1,2-disubstituted benzene moiety the 13C
NMR showed additional resonances for the carbonyls C-3
(d 196.1) and C-4 (d 172.89) and one tertiary hydroxyl group
(d 81.69), which therefore must be located at C-2. In this con-
Figure 4. TPPO Formation in the Enzyme-Catalyzed Conversion of
MHOQ and BHOQ
The inset describes the contents of the assays besides TPP (100 mM) and
organic substrate (1 mM of MHOQ or 100 mM of BHOQ). The concentration of
CMH in the sample of HodC-W160A and BHOQ was 600 mM. Due to autoxi-
dation of TPP in all cases, the above TPPO concentrations represent the
differences of substrate-containing samples and reference samples without
the respective substrate. The bars represent the means of at least three
independent replicates with their SEM.
oxidation of MHOQ did not result in significant TPPO formation, figuration, the adjacent C-9 protons have shifts below 2 ppm,
oxidation of BHOQ by HodC yielded ꢀ2.4-fold more TPPO than which in case of a more planar configuration are expected
the control reaction (Figure 4, light gray bars). Oxidation of both ꢀ2.5 ppm. The 2D NMR HHCOSY, heteronuclear single
substrates by the mutant protein HodC-W160A resulted in much quantum coherence (HSQC), and heteronuclear multiple bond
higher amounts of TPPO (Figure 4, meshed bars). The additional correlation (HMBC) data corroborated the given assignments
presence of SOD during the enzymatic reactions did not affect (Table 2; Figure S5). From the characteristic resonances and
TPPO formation. Interestingly, presence of CMH during BHOQ the overall good agreement between the experimental and
conversion by HodC-W160A significantly quenched TPP oxida- predicted shifts and couplings (Table 2), we conclude that the
tion (Figure 4). This observation, together with the finding that isolated compound is indeed 2-butyl-2-hydroxy-1,2-dihydroqui-
TPP does not affect CMH oxidation, indicates that the CMH noline-3,4-dione.
oxidizing species are formed earlier in the reaction pathway
than the organic hydroperoxide.
DISCUSSION
Structure of the Product of Hydroperoxide Reduction
From the available crystal structures of cofactorless oxygenases
Thin layer chromatography (TLC) separation of extracts of the (Grocholski et al., 2010; Loening et al., 2007; Sciara et al., 2003;
products of the HodC-W160A-catalyzed conversion of BHOQ, Steiner et al., 2010; Widboom et al., 2007) and oxidases (Carlson
performed in the presence or in the absence of TPP, revealed et al., 2008; Colloc’h et al., 1997, 2008; Juan et al., 2008; Lee
an unknown compound, which showed yellow fluorescence et al., 2005; Schwarzenbacher et al., 2004), it has become
under UV light (366 nm) (Figure S4). CMH interfered with forma- evident that different protein folds can provide the catalytic scaf-
tion of this compound (Figure S4), suggesting that it derives from fold for cofactor-independent O2 activation. Nevertheless, the
a ‘‘late’’ intermediate, such as the organic hydroperoxide. enzymes appear to share common mechanistic features. For
Reduction by TPP of the proposed C2-hydroperoxide of several members, formation of a (carb)anion intermediate was
BHOQ would yield 2-butyl-2-hydroxy-1,2-dihydroquinoline- proposed to occur as initial catalytic step (Frerichs-Deeken
3,4-dione. Its formation in the absence of TPP might be due to et al., 2004; Grocholski et al., 2010; Imhoff et al., 2003; Sciara
elimination of H2O2 from the hydroperoxide to form 2-butylqui- et al., 2003; Tseng et al., 2004; Widboom et al., 2007). Another
noline-3,4-dione, followed by addition of water. This reaction aspect that seems to be common to most reactions is the ability
would be analogous to the conversion of 5-hydroperoxyisourate of the deprotonated substrates to form resonance-stabilized
via dehydrourate to 5-hydroxyisourate by urate oxidase (Kahn radicals. For MHOQ and also for anthrones, which are substrates
and Tipton, 1998; Sarma and Tipton, 2000). The hypothesis of quinone-forming monooxygenases, radical formation from the
of H2O2 elimination is supported by the detection of H2O2 respective anion was observed in chemical model reactions
concentrations similar to those of TPPO in samples contain- (Frerichs-Deeken et al., 2004; Mu¨ ller, 1996).
ing HodC-W160A and BHOQ, incubated in the absence of
reductants.
In order to detect possible radical species formed during the
Hod reaction, we used the spin probe CMH. Significant amounts
d
MS analysis of the unknown compound revealed an m/z of of CM were formed during the reaction when the mutant protein
234 for [M+H]+, suggesting the molecular formula C13H15NO3. HodC-W160A was the catalyst, but not during HodC-catalyzed
Fragmentation resulted in ions with m/z 216 (100%) for conversions, suggesting that wild-type protein efficiently shields
[C13H13NO2+H]+, 188 (3%) for [C11H9NO2+H]+ or [C12H13NO+H]+, its active site, whereas the mutant protein may have a more
Chemistry & Biology 21, 217–225, February 20, 2014 ª2014 Elsevier Ltd All rights reserved 221

Chemistry & Biology
Substrate-Assisted Enzymatic O2 Activation
H2O2 would result in up to 1 H2O2 from oxidation of 1 CMH, pre-
sumably with an ill-defined stoichiometry. The observed exact
stoichiometry of 1 H2O2 formed from the oxidation of 2 CMH
does hardly conform to a possible reaction pathway via a peroxyl
radical intermediate, but can be explained by a 1:1 formation of
Table 2. Experimental and Predicted Chemical Shifts and
J-Couplings of the BHOQ Hydroperoxide Decomposition Product
Experimentala
Predictiona
Position
dC
dHb (J)
dC
dHb (J)
1
—
5.65 s
—
4.67 s
dÀ
d
substrate radical S and superoxide radical O2
reaction.
in the Hod
2
81.69
—
86.44
197.75
183.94
124.57
124.97
123.63
134.13
114.40
149.12
34.90
27.01
23.77
14.10
—
—
The [S – O2 À] pair could either result from initial formation
d
d
3
196.1
172.89
119.02
126.74
122.54
135.93
116.14
141.38
28.99
24.71
22.60
13.77
—
—
—
d
d
4
—
—
of S and enzyme-H and subsequent single electron transfer
+
from the enzyme radical to form enzyme-H , S , and O2 À, as
discussed for the urate oxidase reaction. Again, the absence of
any radicals in anoxic mixtures of HodC and substrate and the
CMH oxidation data argue against this pathway. We therefore
propose that the radical pair is formed by direct electron transfer
from the enzyme-bound substrate anion to dioxygen (Figure 5).
This view is supported by the CV results that indicate that the
substrate monoanions have sufficiently low potentials to thermo-
dynamically allow electron transfer to oxygen. In the protein, the
substrate anion, deprotonated at its 3-OH by H251, is hydrogen
bonded with its N1 proton to the carbonyl oxygen of W36
(Steiner et al., 2010), which additionally facilitates the delivery
of an electron from substrate anion to dioxygen. Considering
the close proximity of substrate to oxygen in the active site
d
d
4a
5
—
—
7.71 dd (8/1.5)
7.10 ddd (8/8/1)
7.59 ddd (8/8/1.5)
7.06 brd (8)
—
7.71 d (8)
6.99 dd (8/8)
7.42 dd (8/8)
6.67 d (8)
—
6
7
8
8a
9
1.71 m /1.65 m
1.22 m (2H)
1.15 m (2H)
0.76 t (7;3H)
10.74 brs
1.84 / 1.68
1.50 (2H)
1.50 (2H)
0.90 (7;3H)
8.77
10
11
12
2-OH
Proposed structure
O
˚
pocket of Hod (as suggested by distances of 2.51 and 2.63 A
H
C
5
of a chloride mimic to the substrate’s C2 and C4 atoms, respec-
tively; Steiner et al., 2010), the transfer appears to be kinetically
feasible.
O
4a
8a
4
HC 6
HC7
3
2
OH
Recombination of the [S – O2 À] pair should yield an organic
peroxide intermediate, which presumably is stabilized by the
oxyanion hole of the a/b-hydrolase fold of Hod (Steiner et al.,
2010). When using the nonphysiological substrate BHOQ, trace
amounts of organic hydroperoxide indeed were released from
the Hod reaction. Identification of the alcohol produced from
the reaction of the hydroperoxide with TPP as 2-butyl-2-
hydroxy-1,2-dihydroquinoline-3,4-dione verified that hydroper-
oxide formation had occurred at C-2 of the substrate. The
reaction products carbon monoxide and N-acylanthranilic acid
of the Hod reaction probably result from intramolecular attack
8
1
d
d
C
H
N
H
C9
H2
C
H2
10
C11
CH3
H2
12
See also Figures S4 and S5.
aThe chemical shifts are given in ppm, the J-couplings in Hz.
bMultiplicity: brs, broad singlet; s, singlet; brd, broad doublet; d, doublet;
m, multiplet; t, triplet.
open or more flexible active-site pocket. The observation that of the C2-peroxide anion at the C4 carbonyl and subsequent
CMH oxidation only occurs during the ongoing Hod reaction collapse of an endoperoxide intermediate (Figure 5).
indicated that the enzyme is not able to directly activate O2, or
In summary, our data indicate that a major role of Hod, besides
its organic substrate. This is supported by EPR studies on the activating the substrate by general base catalysis and stabilizing
anoxic complex of HodC and MHOQ, which—in contrast to the transition states along the reaction pathway, is to provide an
anoxic urate oxidase-urate complex (Gabison et al., 2011)— environment that facilitates electron transfer from the substrate
did not reveal any radical species. The reaction of urate oxidase anion to O2. Thus, the enzyme utilizes the intrinsic reactivity of
has been studied in great detail (Busi et al., 2007; Colloc’h et al., its enzyme-bound substrate anion for O2 activation. Structural
1997, 2008; Gabison et al., 2011; Kahn and Tipton, 1998; Sarma and biochemical data on Hod and other cofactor-independent
and Tipton, 2000). In this enzyme, O2 activation was proposed oxygenases support the notion that these enzymes share the
to occur via single electron transfer from an enzyme radical to common catalytic steps of initial base catalysis, substrate-assis-
form the superoxide anion radical. However, as in our study, ted O2 activation, and radical recombination to a peroxide
spin trapping of superoxide or other oxygen radicals failed, intermediate.
suggesting that the activated oxygen species is not released
from the enzyme (Busi et al., 2007; Gabison et al., 2011).
If (despite the failure to detect a radical in anoxic ES com-
SIGNIFICANCE
plexes) the Hod reaction nevertheless would involve initial acti- In chemistry and biology, reactions of organic compounds
d
vation of the bound substrate anion to form a radical S that with O2 are restricted by the spin conservation rule.
d
rapidly reacts with O2 to an organic peroxyl radical S-OO , reac- 1H-3-hydroxy-4-oxoquinaldine 2,4-dioxygenase (Hod) is
d
tion of the latter with CMH would form CM and S-OOH. Possible one of a few oxygenases that are functional without any
(more or less efficient) decomposition of the latter to release cofactor, raising the question of how O2 is activated. It has
222 Chemistry & Biology 21, 217–225, February 20, 2014 ª2014 Elsevier Ltd All rights reserved

Chemistry & Biology
Substrate-Assisted Enzymatic O2 Activation
Figure 5. Proposed Reaction Pathway of the Hod-Catalyzed 2,4-Dioxygenolysis of 2-Alkyl-3-hydroxy-4(1H)-quinolones and Side Reactions
of Intermediates
R = H, n-alkyl (C1 to C9). Brackets indicate enzyme-bound reaction intermediates. Side reactions are indicated by dashed arrows.
See also Figure S3.
EXPERIMENTAL PROCEDURES
been proposed that cofactorless oxygenases use base
catalysis to activate their organic substrate, followed by
Chemicals
electron transfer from the substrate (carb)anion to O2
MHOQ was synthesized from 2-methyl-4(1H)-quinolone according to
to form a [substrate radical – superoxide radical] pair, how-
Cornforth and James (1956) and Eiden et al. (1978). BHOQ was synthesized
ever, this hypothesis is largely untested. Alternatively, acti-
essentially by the procedure reported by Ilangovan et al. (2013); for details,
vation of the substrate to a radical may be mediated by the
enzyme, followed by electron transfer from the enzyme
radical to O2.
see the Supplemental Experimental Procedures. Suppliers and handling of
other chemicals is also described in the Supplemental Experimental
Procedures.
Hod catalyzes the oxygenation of 2-n-alkyl-3-hydroxy-
4(1H)-quinolones and thus is able to inactivate the Pseudo-
monas aeruginosa signaling molecule PQS. It uses a protein
architecture that is generally employed to catalyze hydrolyt-
ic reactions. Previous biochemical and structural data
revealed a general base mechanism as an initial step in
catalysis. In this study, we combined chemical and spectro-
scopic methods to characterize the mechanism of O2 acti-
vation by Hod and to identify intermediates formed along
the reaction pathway. Electron paramagnetic resonance
studies indicated that Hod is not able to directly activate
its organic substrate, or O2, to a radical species. However,
spin probe oxidation during substrate conversion by a Hod
variant yielded H2O2 in a stoichiometry that is consistent
with formation of substrate radical and superoxide anion
radical, resulting from single electron transfer from sub-
strate anion to O2. Cyclic voltammetry studies confirmed
that the proposed electron transfer processes are thermo-
dynamically favorable. Further along the Hod reaction,
a C2-hydroperoxide intermediate is produced, indicating
radical recombination at the C-2 position of the heterocycle.
Our study provides evidence for the long-standing but
poorly supported hypothesis that reactions catalyzed by
cofactor-independent oxygenases are substrate-assisted,
utilizing the intrinsic reactivity of the enzyme-bound sub-
strate anion toward O2.
Protein Expression and Purification, Enzyme Assay,
and Site-Directed Mutagenesis of hod
Wild-type Hod under oxic conditions forms dimers due to formation of an
intermolecular disulfide bridge. Therefore, previous functional and structural
studies were performed with Hod carrying a substitution of Cys69 by serine
(Beermann et al., 2007; Frerichs-Deeken et al., 2004; Steiner et al., 2010).
The Hod-C69S variant, termed HodC, shows the same catalytic activity
as Hod (Frerichs-Deeken et al., 2004) and is also used as ‘‘wild-type’’ protein
in this study. Growth of recombinant Escherichia coli M15 (pREP4, pQE30-
hodC69S) as well as E. coli harboring pQE30-hodC69S-W160A or pQE30-
hodC69S-H251A, and purification of His6-tagged Hod proteins were
performed as described in Beermann et al. (2007).
The catalytic activity of Hod proteins was determined spectrophoto-
metrically by measuring substrate consumption as described previously
(Frerichs-Deeken et al., 2004). For details on the enzyme assay, see the
Supplemental Experimental Procedures. Site-directed mutagenesis of the
hod-C69S gene to replace W160 of HodC by alanine was performed according
to the protocol of the QuikChange site-directed mutagenesis kit (Stratagene);
for primers, see the Supplemental Experimental Procedures.
Identification of Reaction Products and Side Products
The organic products of the reactions catalyzed by HodC and HodC-W160A
were prepared by chromatography as described previously (Pustelny et al.,
2009) and identified by HPLC and mass spectrometry. The possibility
of formate formation in a side reaction was assessed with formate dehy-
drogenase (see the Supplemental Experimental Procedures for details).
Organic hydroperoxides released from the enzymatic reactions were deter-
mined based on their specific reaction with TPP to TPPO, and quantification
of TPPO by HPLC (see the Supplemental Experimental Procedures).
Chemistry & Biology 21, 217–225, February 20, 2014 ª2014 Elsevier Ltd All rights reserved 223

Chemistry & Biology
Substrate-Assisted Enzymatic O2 Activation
2-Butyl-2-hydroxy-1,2-dihydroquinoline-3,4-dione was isolated by prepara-
tive TLC and identified by mass spectrometry and high field NMR correlation
spectroscopy.
Bugg, T.D.H. (2003). Dioxygenase enzymes: catalytic mechanisms and
chemical models. Tetrahedron 59, 7075–7101.
Bugg, T.D.H. (2004). Diverse catalytic activities in the alphabeta-hydrolase
family of enzymes: activation of H2O, HCN, H2O2, and O2. Bioorg. Chem.
32, 367–375.
EPR Studies
The solution EPR experiments were performed at 25ꢂC with a Bruker ESP300e
spectrometer. The cyclic hydroxylamine spin probe CMH was used to monitor
the formation of oxidizing species during the Hod reactions; for details, see
the Supplemental Experimental Procedures. Consumption of O2 was followed
with the oximetry probe TAM, a tetrathiatriarylmethyl radical. To monitor
the kinetic behavior, time-resolved spectra were analyzed by calculating the
Busi, E., Sinicropi, A., Terzuoli, L., Marinello, E., and Basosi, E. (2007).
Identification and structural characterization of a transient radical species in
the uricase reaction mechanism. Appl. Magn. Reson. 31, 471–482.
Carlson, B.L., Ballister, E.R., Skordalakes, E., King, D.S., Breidenbach, M.A.,
Gilmore, S.A., Berger, J.M., and Bertozzi, C.R. (2008). Function and structure
of a prokaryotic formylglycine-generating enzyme. J. Biol. Chem. 283, 20117–
20125.
d
line’s peak-to-peak intensity of both TAM and CM (Figure S2). The low
temperature (70 K and 20 K) EPR measurements were done with an ELEXSYS
E680 spectrometer (Bruker). Sample preparation for anoxic measurements
was done in an anaerobic glove box. To quantify H2O2, an assay was devel-
oped using CMH as the electron donor for the catalytic reduction of H2O2 by
horse radish peroxidase (see the Supplemental Experimental Procedures
for details).
Chaiyen, P., Fraaije, M.W., and Mattevi, A. (2012). The enigmatic reaction of
flavins with oxygen. Trends Biochem. Sci. 37, 373–380.
Colloc’h, N., el Hajji, M., Bachet, B., L’Hermite, G., Schiltz, M., Prange´ , T.,
Castro, B., and Mornon, J.P. (1997). Crystal structure of the protein drug urate
oxidase-inhibitor complex at 2.05 A resolution. Nat. Struct. Biol. 4, 947–952.
Colloc’h, N., Gabison, L., Monard, G., Altarsha, M., Chiadmi, M., Marassio, G.,
Sopkova-de Oliveira Santos, J., El Hajji, M., Castro, B., Abraini, J.H., and
Prange´ , T. (2008). Oxygen pressurized X-ray crystallography: probing the
dioxygen binding site in cofactorless urate oxidase and implications for its
catalytic mechanism. Biophys. J. 95, 2415–2422.
Cyclic Voltammetry
Cyclic voltammetric (CV) measurements were performed at 25 1ꢂC under
quiescent condition by using
a three-electrode configuration in a one-
compartment cell. A glassy carbon was used as the working electrode,
a saturated silver chloride Ag/AgCl as the reference electrode and a platinum
wire as the auxiliary electrode.
Cornforth, J.W., and James, A.T. (1956). Structure of a naturally occurring
antagonist of dihydrostreptomycin. Biochem. J. 63, 124–130.
SUPPLEMENTAL INFORMATION
Dikalov, S.I., Kirilyuk, I.A., Voinov, M., and Grigor’ev, I.A. (2011). EPR detection
of cellular and mitochondrial superoxide using cyclic hydroxylamines. Free
Radic. Res. 45, 417–430.
Supplemental Information includes Supplemental Experimental Procedures
and five figures and can be found with this article online at http://dx.doi.org/
10.1016/j.chembiol.2013.11.013.
Eiden, F., Wendt, R., and Fenner, H. (1978). Pyrones and pyridones. 74.
Quinolylidene derivatives. Arch. Pharm. (Weinheim) 311, 561–568.
ACKNOWLEDGMENTS
Fetzner, S., and Steiner, R.A. (2010). Cofactor-independent oxidases and
oxygenases. Appl. Microbiol. Biotechnol. 86, 791–804.
We thank B. Philipp for access to the HPLC system, A. Steinbu¨ chel for access
to the Finnigan mass spectrometer, M. Schu¨ rmann and I. Plugge for help with
the MS analyses, P. Weyrauch for construction of the pQE30-hodC69S-
W160A plasmid, and A. Kappius for excellent technical assistance. This
work was supported by the Deutsche Forschungsgemeinschaft (grants FE
383/15-1 and KA 1242/1-1 to S.F. and R.K., respectively).
Fischer, F., Ku¨ nne, S., and Fetzner, S. (1999). Bacterial 2,4-dioxygenases:
new members of the a/b hydrolase-fold superfamily of enzymes functionally
related to serine hydrolases. J. Bacteriol. 181, 5725–5733.
Frerichs-Deeken, U., Ranguelova, K., Kappl, R., Hu¨ ttermann, J., and Fetzner,
S. (2004). Dioxygenases without requirement for cofactors and their chemical
model reaction: compulsory order ternary complex mechanism of 1H-3-
hydroxy-4-oxoquinaldine 2,4-dioxygenase involving general base catalysis
by histidine 251 and single-electron oxidation of the substrate dianion.
Biochemistry 43, 14485–14499.
Received: October 14, 2013
Revised: November 22, 2013
Accepted: November 27, 2013
Published: January 2, 2014
Gabison, L., Chopard, C., Colloc’h, N., Peyrot, F., Castro, B., El Hajji, M.,
´
Altarsha, M., Monard, G., Chiadmi, M., and Prange, T. (2011). X-ray, ESR,
REFERENCES
and quantum mechanics studies unravel a spin well in the cofactor-less urate
oxidase. Proteins 79, 1964–1976.
Bauer, I., Max, N., Fetzner, S., and Lingens, F. (1996). 2,4-dioxygenases
catalyzing N-heterocyclic-ring cleavage and formation of carbon monoxide.
Purification and some properties of 1H-3-hydroxy-4-oxoquinaldine 2,4-
dioxygenase from Arthrobacter sp. Ru¨ 61a and comparison with 1H-3-
hydroxy-4-oxoquinoline 2,4-dioxygenase from Pseudomonas putida 33/1.
Eur. J. Biochem. 240, 576–583.
¨
¨ ¨
Grocholski, T., Koskiniemi, H., Lindqvist, Y., Mantsala, P., Niemi, J., and
Schneider, G. (2010). Crystal structure of the cofactor-independent monoox-
ygenase SnoaB from Streptomyces nogalater: implications for the reaction
mechanism. Biochemistry 49, 934–944.
Ilangovan, A., Fletcher, M., Rampioni, G., Pustelny, C., Rumbaugh, K., Heeb,
´
S., Camara, M., Truman, A., Chhabra, S.R., Emsley, J., and Williams, P. (2013).
Beermann, B., Guddorf, J., Boehm, K., Albers, A., Kolkenbrock, S., Fetzner, S.,
and Hinz, H.-J. (2007). Stability, unfolding, and structural changes of cofactor-
free 1H-3-hydroxy-4-oxoquinaldine 2,4-dioxygenase. Biochemistry 46, 4241–
4249.
Structural basis for native agonist and synthetic inhibitor recognition by the
Pseudomonas aeruginosa quorum sensing regulator PqsR (MvfR). PLoS
Pathog. 9, e1003508.
Imhoff, R.D., Power, N.P., Borrok, M.J., and Tipton, P.A. (2003). General base
catalysis in the urate oxidase reaction: evidence for a novel Thr-Lys catalytic
diad. Biochemistry 42, 4094–4100.
Bonnot, F., Iavarone, A.T., and Klinman, J.P. (2013). Multistep, eight-electron
oxidation catalyzed by the cofactorless oxidase, PqqC: Identification of chem-
ical intermediates and their dependence on molecular oxygen. Biochemistry
52, 4667–4675.
Juan, E.C.M., Hoque, M.M., Shimizu, S., Hossain, M.T., Yamamoto, T.,
Imamura, S., Suzuki, K., Tsunoda, M., Amano, H., Sekiguchi, T., and
Bou, R., Codony, R., Tres, A., Decker, E.A., and Guardiola, F. (2008).
Determination of hydroperoxides in foods and biological samples by the
´
Takenaka, A. (2008). Structures of Arthrobacter globiformis urate oxidase-
ligand complexes. Acta Crystallogr. D Biol. Crystallogr. D64, 815–822.
ferrous oxidation-xylenol orange method:
a review of the factors that
influence the method’s performance. Anal. Biochem. 377, 1–15.
Bruice, T.C. (1984). Oxygen-flavin chemistry. Isr. J. Chem. 24, 54–61.
Kahn, K., and Tipton, P.A. (1998). Spectroscopic characterization of inter-
mediates in the urate oxidase reaction. Biochemistry 37, 11651–11659.
224 Chemistry & Biology 21, 217–225, February 20, 2014 ª2014 Elsevier Ltd All rights reserved

Chemistry & Biology
Substrate-Assisted Enzymatic O2 Activation
´
´
´
´
Lee, D.S., Flachsova, E., Bodnarova, M., Demeler, B., Martasek, P., and
Raman, C.S. (2005). Structural basis of hereditary coproporphyria. Proc.
Natl. Acad. Sci. USA 102, 14232–14237.
transcriptional analysis of genes involved in quinaldine degradation.
J. Bacteriol. 189, 3855–3867.
Pustelny, C., Albers, A., Bu¨ ldt-Karentzopoulos, K., Parschat, K., Chhabra,
S.R., Ca´ mara, M., Williams, P., and Fetzner, S. (2009). Dioxygenase-mediated
quenching of quinolone-dependent quorum sensing in Pseudomonas
aeruginosa. Chem. Biol. 16, 1259–1267.
´
´
Lepine, F., Milot, S., Deziel, E., He, J., and Rahme, L.G. (2004). Electrospray/
mass spectrometric identification and analysis of 4-hydroxy-2-alkylquinolines
(HAQs) produced by Pseudomonas aeruginosa. J. Am. Soc. Mass Spectrom.
15, 862–869.
RoseFigura, J.M., Puehringer, S., Schwarzenbacher, R., Toyama, H., and
Klinman, J.P. (2011). Characterization of a protein-generated O2 binding
pocket in PqqC, a cofactorless oxidase catalyzing the final step in PQQ
production. Biochemistry 50, 1556–1566.
Loening, A.M., Fenn, T.D., Wu, A.M., and Gambhir, S.S. (2006). Consensus
guided mutagenesis of Renilla luciferase yields enhanced stability and light
output. Protein Eng. Des. Sel. 19, 391–400.
Loening, A.M., Fenn, T.D., and Gambhir, S.S. (2007). Crystal structures of the
luciferase and green fluorescent protein from Renilla reniformis. J. Mol. Biol.
374, 1017–1028.
Sarma, A.D., and Tipton, P.A. (2000). Evidence for urate hydroperoxide as an
intermediate in the urate oxidase reaction. J. Am. Chem. Soc. 122, 11252–
11253.
Massey, V. (1994). Activation of molecular oxygen by flavins and flavoproteins.
Schwarzenbacher, R., Stenner-Liewen, F., Liewen, H., Reed, J.C., and
Liddington, R.C. (2004). Crystal structure of PqqC from Klebsiella pneumoniae
at 2.1 A resolution. Proteins 56, 401–403.
J. Biol. Chem. 269, 22459–22462.
Mu¨ ller, K. (1996). Antipsoriatic anthrones: aspects of oxygen radical formation,
challenges and prospects. Gen. Pharmacol. 27, 1325–1335.
Sciara, G., Kendrew, S.G., Miele, A.E., Marsh, N.G., Federici, L., Malatesta, F.,
Schimperna, G., Savino, C., and Vallone, B. (2003). The structure of ActVA-
Orf6, a novel type of monooxygenase involved in actinorhodin biosynthesis.
EMBO J. 22, 205–215.
Nardini, M., and Dijkstra, B.W. (1999). a/b hydrolase fold enzymes: the family
keeps growing. Curr. Opin. Struct. Biol. 9, 732–737.
Niewerth, H., Schuldes, J., Parschat, K., Kiefer, P., Vorholt, J.A., Daniel, R.,
and Fetzner, S. (2012). Complete genome sequence and metabolic potential
of the quinaldine-degrading bacterium Arthrobacter sp. Rue61a. BMC
Genomics 13, 534.
¨
Siitonen, V., Blauenburg, B., Kallio, P., Mantsala, P., and Metsa-Ketela, M.
¨ ¨
¨
¨
(2012). Discovery of a two-component monooxygenase SnoaW/SnoaL2
involved in nogalamycin biosynthesis. Chem. Biol. 19, 638–646.
¨ ¨ ¨
Oja, T., Klika, K.D., Appassamy, L., Sinkkonen, J., Mantsala, P., Niemi, J., and
Steiner, R.A., Janssen, H.J., Roversi, P., Oakley, A.J., and Fetzner, S. (2010).
Structural basis for cofactor-independent dioxygenation of N-heteroaromatic
compounds at the a/b-hydrolase fold. Proc. Natl. Acad. Sci. USA 107,
657–662.
¨
¨
Metsa-Ketela, M. (2012). Biosynthetic pathway toward carbohydrate-like
moieties of alnumycins contains unusual steps for C-C bond formation and
cleavage. Proc. Natl. Acad. Sci. USA 109, 6024–6029.
Ouari, O., Hardy, M., Karoui, H., and Tordo, P. (2011). Recent developments
and applications of the coupled EPR/Spin trapping technique (EPR/ST).
Electron Paramag. Reson. 22, 1–40.
Tseng, C.C., Vaillancourt, F.H., Bruner, S.D., and Walsh, C.T. (2004). DpgC is
a metal- and cofactor-free 3,5-dihydroxyphenylacetyl-CoA 1,2-dioxygenase
in the vancomycin biosynthetic pathway. Chem. Biol. 11, 1195–1203.
Palfey, B.A., Ballou, D.P., and Massey, V. (1995). Oxygen activation by flavins
and pterins. In Active Oxygen in Biochemistry, J.S. Valentine, C. Foote, and
J.F. Liebman, eds. (Dordrecht, the Netherlands: Kluwer Academic
Publishers), pp. 37–83.
Widboom, P.F., Fielding, E.N., Liu, Y., and Bruner, S.D. (2007). Structural
basis for cofactor-independent dioxygenation in vancomycin biosynthesis.
Nature 447, 342–345.
Parschat, K., Overhage, J., Strittmatter, A.W., Henne, A., Gottschalk, G., and
Fetzner, S. (2007). Complete nucleotide sequence of the 113-kilobase
linear catabolic plasmid pAL1 of Arthrobacter nitroguajacolicus Ru¨ 61a and
Woo, J., Howell, M.H., and von Arnim, A.G. (2008). Structure-function studies
on the active site of the coelenterazine-dependent luciferase from Renilla.
Protein Sci. 17, 725–735.
Chemistry & Biology 21, 217–225, February 20, 2014 ª2014 Elsevier Ltd All rights reserved 225