Communications to the Editor
J. Am. Chem. Soc., Vol. 122, No. 45, 2000 11253
a catalytic component of the reaction, but H2O2 production is
abrogated. DTT is not unique in its effect on the urate oxidase
reaction; cysteine was shown to block H2O2 production and reduce
5-hydroxyisourate as well.
Recycling of urate in the presence of DTT was clearly
demonstrated by monitoring the time course of a reaction. In the
presence of 30 µM urate and 25 mM DTT the reaction required
approximately 5 h to reach completion and over 2 mM oxidized
DTT was produced.6 Under similar conditions in the absence of
DTT the urate was consumed in 5 min. The reaction in the
presence of DTT does not continue indefinitely because 5-hy-
droxyisourate is somewhat unstable; nonenzymatic decomposition
of 5-hydroxyisourate leads to the formation of allantoin (5).1
Clear evidence that DTT reacts directly with urate hydroper-
oxide comes from isotope labeling studies. Earlier work has
7
demonstrated that H2O2 arises from O2 and that the hydroxyl
group at C5 of hydroxyisourate is derived from solvent1 in the
normal catalytic reaction. Scheme 1 outlines the labeling patterns
in allantoin that are predicted to arise from reduction of urate
hydroperoxide and from 5-hydroxyisourate generated by the
normal urate oxidase reaction. The critical difference is that
reduction of urate hydroperoxide traps the oxygen from O2 in
5-hydroxyisourate where it will appear ultimately as the C4
carbonyl oxygen in allantoin. In the normal catalytic reaction the
C4 carbonyl oxygen is introduced from solvent by hydration of
the presumptive dehydrourate intermediate (4). Thus, the urate
oxidase reaction was conducted in H218O in the presence of DTT,
and the allantoin produced was isolated and analyzed by elec-
trospray mass spectrometry (Figure 2). Unlabeled and [18O]-
allantoin were observed as M + H ions at m/z 159 and 161,
respectively, and were produced in a ratio of 1:2. Since the H216O
content of the reaction mixture was less than 5% (H216O was
introduced by the addition of the enzyme) the significant fraction
of unlabeled allantoin that was observed must have arisen through
reduction of urate hydroperoxide. Collisionally activated dis-
sociation of the ion giving rise to the m/z 161 peak yielded a
fragmentation pattern that is consistent with the expected position
of the 18O label in allantoin; the corresponding fragmentation
pattern of the m/z 159 peak confirmed the production of unlabeled
allantoin. A control experiment in which [4-18O]allantoin was
generated by conducting the urate oxidase reaction in H218O in
the absence of DTT demonstrated that the 18O label did not
exchange with solvent during sample isolation.
Figure 2. ESI mass spectra of allantoin produced in the urate oxidase
reaction in H218O in the presence of DTT. Top: Fragmentation of the
m/z 159 ion. Bottom: Fragmentation of the m/z 161 ion. The reaction
was conducted in a total volume of 0.2 mL and contained 0.5 mM urate
and 50 mM DTT, buffered at pH 7.4 with potassium phosphate. The
reaction was initiated by the addition of 10 µg of urate oxidase (in a
volume of 0.01 mL). The reaction was incubated at room temperature
overnight and allantoin was isolated by HPLC as described in ref 6. The
buffer was removed by adsorbing the sample to Dowex AG1-x8 resin in
the formate form; allantoin was eluted with 0.1% formic acid, concentrated
to a volume of 0.3 mL and analyzed by postive ion ESI mass
spectrometry.
architecture of the protein. The crystal structure of Aspergillus
flaVus urate oxidase reveals that the active site is located in a
shallow cleft on the surface of the protein.9 Rapid-mixing chemical
quench experiments are currently in progress to examine the
kinetics of the appearance and disappearance of urate hydroper-
oxide and other intermediates in the catalytic reaction.
The trapping experiments reported here provide independent
evidence that a urate hydroperoxide species, presumably formed
by reaction between the urate dianion and O2, is an intermediate
in the urate oxidase reaction. We are aware of only a single report
of analogous trapping of the flavin hydroperoxide intermediate
in an enzymatic reaction.8
The facility with which external reagents can access the active
site of urate oxidase presumably is a consequence of the
Acknowledgment. This work was supported by USDA grant 98-
35305-6548.
(6) 30 µM urate and 25 mM DTT were combined in 100 mM potassium
phosphate, pH 7.4, in a volume of 1 mL. After addition of 40 µg of urate
oxidase, aliquots were removed periodically and analyzed by HPLC. Depro-
teinized aliquots were injected onto a Sperisorb ODS-2 column and eluted at
1.5 mL/min with 5 mM ammonium phosphate, pH 3.39. Elution times were
as follows: allantoin, 2.7 min (void volume); urate, 8.8 min; reduced DTT,
18 min; oxidized DTT, 30 min.
Supporting Information Available: Figures of H2O2 production and
oxygen consumption in the presence and absence of DTT (PDF). This
JA002829J
(7) Bentley, R.; Neuberger, A. Biochem. J. 1952, 52, 694-699.
(9) Colloc’h, N.; El Hajji, M.; Bachet, B.; L’Hermite, G.; Schiltz, M.;
Prange, T.; Castro, B.; Mornon, J. Nature Struct. Biol. 1997, 4, 947-952.
(8) McCapra, F.; Hart, R. J. Chem. Soc. Chem. Commun. 1976, 273-274.