Reactions of peroxynitrite with uric acid: Formation of reactive intermediates, alkylated products and triuret, and in vivo production of triuret under conditions of oxidative stress

Article abstract of DOI:10.1080/15257770902736400

Hyperuricemia is associated with hypertension, metabolic syndrome, preeclampsia, cardio-vascular disease and renal disease, all conditions associated with oxidative stress. We hypothesized that uric acid, a known antioxidant, might become prooxidative following its reaction with oxidants; and, thereby contribute to the pathogenesis of these diseases. Uric acid and 1,3-15N2-uric acid were reacted with peroxynitrite in different buffers and in the presence of alcohols, antioxidants and in human plasma. The reaction products were identified using liquid chromatography-mass spectrometry (LC-MS) analyses. The reactions generate reactive intermediates that yielded triuret as their final product. We also found that the antioxidant, ascorbate, could partially prevent this reaction. Whereas triuret was preferentially generated by the reactions in aqueous buffers, when uric acid or 1,3-15N2-uric acid was reacted with peroxynitrite in the presence of alcohols, it yielded alkylated alcohols as the final product. By extension, this reaction can alkylate other biomolecules containing OH groups and others containing labile hydrogens. Triuret was also found to be elevated in the urine of subjects with preeclampsia, a pregnancy-specific hypertensive syndrome that is associated with oxidative stress, whereas very little triuret is produced in normal healthy volunteers. We conclude that under conditions of oxidative stress, uric acid can form reactive intermediates, including potential alkylating species, by reacting with peroxynitrite. These reactive intermediates could possibly explain how uric acid contributes to the pathogenesis of diseases such as the metabolic syndrome and hypertension. Copyright Taylor & Francis Group, LLC.

Full text of DOI:10.1080/15257770902736400

This article was downloaded by: [University of Memphis]
On: 07 August 2012, At: 05:23
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Nucleosides, Nucleotides and Nucleic
Acids
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/lncn20
Reactions of Peroxynitrite with
Uric Acid: Formation of Reactive
Intermediates, Alkylated Products and
Triuret, and In Vivo Production of Triuret
Under Conditions of Oxidative Stress
Christine Gersch a , Sergiu P. Palii b , Witcha Imaram c , Kyung Mee
Kim b , S. Ananth Karumanchi d , Alexander Angerhofer c , Richard J.
Johnson a & George N. Henderson a b e
a Division of Nephrology and Hypertension, Department of Medicine,
University of Florida, Gainesville, Florida, USA
b Division of Endocrinology and Metabolism, Department of
Medicine, College of Medicine, University of Florida, Gainesville,
Florida, USA
c Department of Chemistry, College of Liberal Arts and Sciences,
University of Florida, Gainesville, Florida, USA
d Beth Israel Deaconess Medical Center, Harvard Medical School,
Boston, Massachusetts, USA
e General Clinical Research Center, University of Florida,
Gainesville, Florida, USA
Version of record first published: 13 Feb 2009
To cite this article: Christine Gersch, Sergiu P. Palii, Witcha Imaram, Kyung Mee Kim, S. Ananth
Karumanchi, Alexander Angerhofer, Richard J. Johnson & George N. Henderson (2009): Reactions of
Peroxynitrite with Uric Acid: Formation of Reactive Intermediates, Alkylated Products and Triuret,
and In Vivo Production of Triuret Under Conditions of Oxidative Stress, Nucleosides, Nucleotides and
Nucleic Acids, 28:2, 118-149
To link to this article: http://dx.doi.org/10.1080/15257770902736400
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation
that the contents will be complete or accurate or up to date. The accuracy of any
instructions, formulae, and drug doses should be independently verified with primary
sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand, or costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.

Nucleosides, Nucleotides and Nucleic Acids, 28:118–149, 2009
Copyright ꢀC Taylor & Francis Group, LLC
ISSN: 1525-7770 print / 1532-2335 online
DOI: 10.1080/15257770902736400
REACTIONS OF PEROXYNITRITE WITH URIC ACID: FORMATION OF
REACTIVE INTERMEDIATES, ALKYLATED PRODUCTS AND TRIURET,
AND IN VIVO PRODUCTION OF TRIURET UNDER CONDITIONS OF
OXIDATIVE STRESS
Christine Gersch,1 Sergiu P. Palii,2 Witcha Imaram,3 Kyung Mee Kim,2
S. Ananth Karumanchi,4 Alexander Angerhofer,3 Richard J. Johnson,1 and
George N. Henderson1,2,5
1Division of Nephrology and Hypertension, Department of Medicine, University of Florida,
Gainesville, Florida, USA
2Division of Endocrinology and Metabolism, Department of Medicine, College of Medicine,
University of Florida, Gainesville, Florida, USA
3Department of Chemistry, College of Liberal Arts and Sciences, University of Florida,
Gainesville, Florida, USA
4Beth Israel Deaconess Medical Center, Harvard Medical School, Boston,
Massachusetts, USA
5General Clinical Research Center, University of Florida, Gainesville, Florida, USA
Hyperuricemia is associated with hypertension, metabolic syndrome, preeclampsia, cardio-
2
vascular disease and renal disease, all conditions associated with oxidative stress. We hypothesized
that uric acid, a known antioxidant, might become prooxidative following its reaction with
oxidants; and, thereby contribute to the pathogenesis of these diseases. Uric acid and 1,3-15N2-uric
acid were reacted with peroxynitrite in different buffers and in the presence of alcohols, antioxidants
and in human plasma. The reaction products were identified using liquid chromatography-mass
spectrometry (LC-MS) analyses. The reactions generate reactive intermediates that yielded triuret
as their final product. We also found that the antioxidant, ascorbate, could partially prevent this
reaction. Whereas triuret was preferentially generated by the reactions in aqueous buffers, when
uric acid or 1,3-15N2-uric acid was reacted with peroxynitrite in the presence of alcohols, it yielded
alkylated alcohols as the final product. By extension, this reaction can alkylate other biomolecules
containing OH groups and others containing labile hydrogens. Triuret was also found to be elevated
in the urine of subjects with preeclampsia, a pregnancy-specific hypertensive syndrome that is
associated with oxidative stress, whereas very little triuret is produced in normal healthy volunteers.
We conclude that under conditions of oxidative stress, uric acid can form reactive intermediates,
Received 28 April 2008; accepted 11 December 2008.
This work was financially supported by Gatorade and by NIH grants DK-52121, HL-68607 and MO1-
RR00082.
Address correspondence to George N. Henderson, Division of Nephrology, Department of
Medicine, Health Science Center, Box 100224, University of Florida, Gainesville, FL 32610-0224, E-mail:
hende3000@yahoo.com
118

Reactions of Peroxynitrite with Uric Acid
119
including potential alkylating species, by reacting with peroxynitrite. These reactive intermediates
could possibly explain how uric acid contributes to the pathogenesis of diseases such as the metabolic
syndrome and hypertension.
Keywords Uric acid; methyluric acid; peroxynitrite; cardiovascular disease; endothelial
dysfunction; triuret; ascorbate; alkylation; preeclampsia; hypertension; oxidative stress;
reactive intermediates
INTRODUCTION
Uric acid is a purine generated during the degradation of RNA, DNA,
and adenine nucleotides. Its immediate precursor, xanthine, is converted to
uric acid by an enzymatic reaction with either xanthine oxidase or xanthine
dehydrogenase. Xanthine oxidase generates the oxidants superoxide and
hydrogen peroxide during this reaction. In most mammals, uric acid is
further degraded to allantoin by the enzyme uricase; however, this enzyme
was mutated in humans nearly 15 million years ago.[1] Thus, uric acid is
considered an endproduct of purine metabolism in humans.
Uric and ascorbic acids are considered the two most important water
soluble antioxidants in human plasma,[2,3] uric acid plasma levels being
about 6 times that of ascorbic acids plasma levels. Uric acid can react
with a variety of oxidants, including superoxide anion, hydrogen peroxide,
hydroxyl radical, and peroxynitrite. From among these oxidants, uric acid
is known to react preferentially with peroxynitrite.[4,5] These reactions are
thought to be initiated by the donation of an electron by uric acid to
generate the urate radical, followed by its irreversible degradation to a
variety of endproducts.[5,6] In this regard, the mechanism by which urate
acts as an antioxidant is different from that of ascorbate in that ascorbate
generates an ascorbyl radical in a reversible reaction such that ascorbate
can be regenerated.[7]
The importance of the antioxidant properties of uric acid with re-
spect to oxidants such as peroxynitrite has been demonstrated using in
vitro cell culture systems.[4,8] Peroxynitrite is generated by the diffusion
controlled reaction of superoxide and nitric oxide.[7] This reaction may
occur in the vasculature, and particularly the endothelium, due to the
presence of superoxide (generated by NADPH oxidase, xanthine oxidase,
and other systems) and nitric oxide (generated by endothelial nitric oxide
synthase; eNOS). Peroxynitrite has been shown to cause oxidative damage
via the nitration of tissues that has been hypothesized to contribute to the
development of vascular disease.[7] Incubation of endothelial cells with
peroxynitrite rapidly causes oxidative stress, leading to the oxidation of
tetrahydrobiopterin,[7] which is critical in the enzymatic reaction by which
eNOS generates nitric oxide.[8] The addition of uric acid to cells along

120
C. Gersch et al.
with the peroxynitrite prevents this oxidative damage and partially restores
nitric oxide generation. Complete restoration of nitric oxide production
by eNOS occurs when both uric acid and ascorbate are added along with
the peroxynitrite to cells.[4] Additionally, uric acid may help to maintain
extracellular levels of the antioxidant enzyme superoxide dismutase, an
important enzyme in protecting the extracellular milieu from oxidative
stress.[9] Taken together, these studies suggest that uric acid is beneficial.
This has led to the conclusion that elevated uric acid levels in cardiovascular
disease may represent a beneficial host response to the oxidative stress
known to occur in this disease process.[2,10]
Elevated serum uric acid levels are strongly associated with hypertension,
obesity, the metabolic syndrome, renal disease, and cardiovascular disease.
As discussed above, most authorities have viewed the elevation of uric acid
as a secondary or compensatory phenomenon. This assumption has evolved
due to the difficulty in demonstrating uric acid to be an independent risk
factor in cardiovascular disease.[11] However, the increasing evidence that
uric acid elevations often precede the development of hypertension, obesity,
and metabolic syndrome[12–14] have raised questions as to the validity of
this assumption. However, the increasing evidence that uric acid elevations
often precede the development of hypertension, preeclampsia, obesity, and
metabolic syndrome[12–15] have raised questions as to the validity of this
assumption.
In recent years, our group has demonstrated in a series of cell culture,
animal, and human studies that uric acid may be a true risk factor for
hypertension, renal disease, and the metabolic syndrome.[16–22] Raising uric
acid levels in rats with an uricase inhibitor rapidly results in the development
of hypertension and renal disease.[16–18] We have also found that nearly 90%
of early onset essential hypertension in adolescents have elevated serum
uric acid levels (>5.5 mg/dL). In contrast, none of the control subjects in
this study had elevated serum uric acid levels. The serum uric acid levels
reported in this study were independent of obesity and renal function.[19]
Preliminary trials in humans have also reported that lowering uric acid levels
can lower blood pressure.[20,23] We have also reported that lowering serum
uric acid levels in a rat model of the metabolic syndrome corrected most of
the features of the metabolic syndrome.[21]
In this regard, several studies have reported that uric acid can become
a prooxidant under the right conditions.[24–27] This can be explained by
the formation of a urate radical under conditions of oxidative stress.[28–30]
These data,[24–30] when taken together with the studies from our group,
have led us to hypothesize that uric acid may combine with reactive oxygen
species and nitrogen species in vivo to produce oxidation products that
mediate deleterious effects on the body and contribute to the pathogenesis
of hypertension, the metabolic syndrome, preeclampsia, renal disease, and
cardiovascular disease.

Reactions of Peroxynitrite with Uric Acid
121
A number of groups have investigated the reaction of uric acid with
peroxynitrite and the biological implications for this reaction. Kuzkaya et
al.[4] have reported that the scavenging of peroxynitrite by uric acid is
important in preventing the oxidation of tetrahydrobiopterin, which would
lead to the uncoupling of nitric oxide synthase. Skinner et al.[31] have
reported a unique uric acid derivative which was detected in peroxynitrite
reactions. Mass spectroscopic analysis indicated that this product was a
nitrated uric acid derivative 2-nitrito-4-amino-5-hydroxyimidazoline. This
unique product of the uric acid/ONOO- reaction was reported to result in
endothelium-independent vasorelaxation of rat thoracic aorta. Thus, it was
proposed that the uric acid nitration/nitrosation product may play a pivotal
role in human pathophysiology by releasing ·NO, which could decrease
vascular tone, increase tissue blood flow, and thereby constitute a role for
uric acid not previously described. Robinson et al.[5] have reported triuret
as a product of the chemical reaction of uric acid with peroxynitrite. Santos
et al.[28] have reported the production of an aminocarbonyl radical in the
reaction of uric acid with peroxynitrite. They also demonstrated that this
radical is likely to be the species responsible for the effects of urate in
amplifying peroxynitrite-mediated oxidation of liposomes and LDL. We also
have identified by ESR two carbon-based radicals in the reaction of uric acid
with peroxynitrite.[32] The quantitative significance of the radicals detected
to final products of the reaction of uric acid with peroxynitrite is unknown
and it is also unknown if other pathways are important for the reactions of
uric acid with peroxynitrite. Given the background that these studies pro-
vide, our group chose to characterize the products and intermediates of
the reaction of uric acid with peroxynitrite under a variety of conditions
using LC-MS and LC-MS/MS as the primary analytical tools. LC-MS was
used because it permitted the determination of the distribution of uric
acid among the various products (determine the quantitative significance
of various pathways) as well as their characterization. We demonstrate for
the first time that uric acid can generate alklyating (substitution of the OH
hydrogen of alcohols by dehydroxyallantoin group) species.
The principal objective of this study was to define the reactions of UA
with peroxynitrite under a variety of experimental conditions with the view
to identify products and intermediates.
MATERIALS AND METHODS
Chemicals, Preparation of Reagents and General Experimental
Conditions
All chemicals and reagents, unless otherwise specified, were obtained
from Sigma. LiOH·1H2O and 1-methyl uric acid were purchased from
Fluka. Distilled, deionized water (metal ion free), and EDTA (500 mM)

122
C. Gersch et al.
were purchased from Gibco (Langley, OK, USA). 1,3-15N2-Uric acid, d3-
methanol and d4-methanol were purchased from Cambridge Isotopes
(Andover, MA, USA). N-tert-butyl-α-phenylnitrone (PBN) was purchased
from Alexis Biochemicals (San Diego, CA, USA). Diethylenetriaminepenta-
acetic acid (DTPA) was purchased from Sigma. Peroxynitrite was pur-
chased from Cayman Chemical (Ann Arbor, MI, USA). The distilled water
purchased from Gibco was used in solution preparation. Uric acid
(100 mM), 1,3-15N2-uric acid (100 mM) and L-tyrosine were prepared
by dissolving these compounds in KOH (300 mM). Lithium urate
(30 mM) and lithium 1,3-15N2-urate (30 mM) solutions were prepared
by dissolving uric acid and 15N2-uric acid, respectively, in a LiOH so-
lution (35 mM, pH = 8.05), ascorbate (100 mM), tetrahydrobiopterin
(100 mM), glutathione (100 mM), cysteine (100 mM), and LiOH·1H2O
(35 mM) were prepared by dissolving these compounds in distilled water.
EDTA (500 mM) was diluted to 10 mM using distilled water. Potassium
phosphate buffer (300 mM) was prepared by dissolving 16.00 g monobasic
potassium phosphate and 31.70 g dibasic potassium phosphate in distilled
water. The pH of the resulting final solution was adjusted to pH = 7.4
or 9.0 using KOH (300 mM) or 1:1 phosphoric acid. The final volume of
the potassium phosphate buffer was 1.0 L. In all experiments peroxynitrite
was shielded from light and kept on ice and reactions were performed
protected from light. The human studies were approved by the Institutional
Review Board at the University of Florida. The reactions are summarized in
Figure 1.
Reactions in aqueous phosphate buffers
Reaction 1
Lithium urate reactions
Reaction 2
Skinner et. al. Reactions
Reaction 3
O
6
Reactions in the presence of PBN
Reaction 4
Reaction 5
H
N
Reactions in ammonium acetate buffer
HN
7
O
O
5
4
1
2
8
HO
N
O
Reactions in MeOH, EtOH & i-PrOH
+
Reaction 6
Reaction 7
Reaction 8
9
3
NH
Reactions of 1-methyluric acid in MeOH, EtOH & i-PrOH
O
NH
Uric acid
Reactions of products of Reaction 1 with MeOH & EtOH
Reactions in Human Plasma
Reaction 9
Reactions in the presence of ascorbate
Reaction 10
FIGURE 1 The summary of the reaction of peroxynitrite with uric acid and 1,3-15N2-uric acid. Reactions
1–8 were conducted as a pair with labeled and unlabeled uric acid to aid in the mass spectrometry
analysis and identification of products. The effect of multiple pHs (7.4 and 9.0) were also investigated
in Reactions 1 and 2 and these did not alter the formation products determined by mass spectrometry.
Reaction 7 employed 1-methyluric acid instead of uric acid. For the structures of products see Figure 3.

Reactions of Peroxynitrite with Uric Acid
123
Determination of Peroxynitrite Concentration
In preliminary studies, we have used PN prepared using two different
published methods[33,34] and two commercial products (Cayman Chemical
and Upstate Biotechnology, Lake Placid, NY). All produced the same
products. After the initial testing, we used the peroxynitrite from Cayman
Chemical for all the studies reported here.
Peroxynitrite concentration was determined prior to its use (as per
the instructions from Cayman Chemical) in each experiment as follows:
Peroxynitrite (Cayman) was placed on ice to thaw. A 25 µL aliquot of the
peroxynitrite stock solution was diluted with 975 uL of KOH (300 mM).
The absorbance of the resulting solution at 302 nm was then measured,
using KOH (300 mM) as a blank. The concentration of peroxynitrite
was then calculated using the molar extinction coefficient of 1670/M·cm.
The volume of peroxynitrite needed to yield the desired concentration of
peroxynitrite for a given experiment was then calculated.
Reactions of Uric Acid with Peroxynitrite in Aqueous Buffers
[Reaction 1]
Uric acid stock solution (300 µL of a uric acid stock solution) was
added to potassium phosphate buffer (pH 7.4) containing EDTA. One
equivalent of peroxynitrite (in some experiments both 1 and 3 equivalents
were used) was then added to the reaction mixture. The stock peroxynitrite
concentration varied in each lot of commercially available peroxynitrite.
As a result, the volumes of potassium phosphate buffer and peroxynitrite
would vary by experiment, but the total reaction mixture volume was held
constant at 3.00 mL, the added volume of uric acid stock solution was
always 300 µL (concentration of the uric acid stock solutions was varied
such that this volume of added uric acid remained constant). The final
concentrations of uric acid varied from 300 µM (physiological) to 10 mM
and peroxynitrite varied from 300 µM 10 mM, and these reagents were used
in a 1:1 molar ratio unless otherwise specified. Final concentration of EDTA
was 0.1 mM. Reactions in aqueous buffer were conducted at pHs 7.4 and 9.0.
The reaction mixture was vortexed for 10 seconds and allowed to incubate at
room temperature for 3 minutes. After 3 minutes at room temperature, the
reaction mixtures were placed in ice and transferred to LC-MS vials on ice
via Pasteur pipettes. The vials were immediately placed at −80◦C until mass
spectroscopic analysis could be performed. Please note that Reaction 1 was
our “model reaction” upon which all other reaction procedures were based.
Reactions of Lithium Urate with Peroxynitrite [Reaction 2]
Lithium urate (300 µL of a prepared stock solution) reactions with
peroxynitrite were conducted using conditions and volumes of reagents as
in Reaction 1. Final concentrations of Li urate ranged from 0.3 to 10 mM.

124
C. Gersch et al.
Final concentrations of peroxynitrite ranged from 0.3 to 10 mM. Lithium
urate were conducted at pHs 7.4 and 9.0.
Skinner Reactions [Reaction 3]
We have duplicated some of the reactions reported by Skinner
et al.[31] with view to determine the product formed under their reaction
conditions. Uric acid stock solutions were prepared in 100 mM potassium
phosphate buffer containing 0.1 mM DTPA. Peroxynitrite (0, 0.25, 0.5,
0.75, or 1.0 mM) was then added to potassium phosphate buffer solution
and allowed to incubate by itself in the buffer solution for 15 minutes.
The total volume of the peroxynitrite/potassium phosphate buffer mixture
was 2.70 mL. The volume of the potassium phosphate buffer was varied
to account for variances in the peroxynitrite stock solution concentration.
Next, 300 µL of one of the uric acid stock solutions was added to each of
the peroxynitrite/potassium phosphate buffer solutions. This was repeated
using three different stock solutions of uric acid such that at a final volume
of 3.0 mL, the final concentrations of uric acid were 0.1, 0.2, and 0.3 mM.
These reaction mixtures were allowed to incubate at room temperature for
another 15 minutes, and were then placed in LC-MS vials and stored at
−80◦C until LC-MS analysis.
Reactions of Uric Acid with Peroxynitrite in the Presence of PBN
[Reaction 4]
Uric acid (300 µL of 100 m M stock solution) was added to potassium
phosphate buffer containing DTPA (0.1 mM final concentration) and the
spin trapping agent, PBN. PBN concentrations used were 10 or 50 mM.
Final reaction volumes were held constant at 3 mL (see Reaction 1). One
equivalent (final concentration of 10 mM) of peroxynitrite was then added
to the reaction mixture. The reaction mixture was transferred to LC-MS vials
and stored at −80◦C to await LC-MS analysis.
Reactions of Uric Acid with Peroxynitrite in Ammonium Acetate
Buffer [Reaction 5]
Uric acid was weighed out into a vial (5.04 mg) to which a mixture of
50% ammonium acetate and 50% acetonitrile was added. The volume of the
ammonium acetate:acetonitrile mixture and the volume of peroxynitrite to
be added to the reaction mixture was varied to account for variances in the
peroxynitrite stock solution concentration. To the heterogeneous mixture
of uric acid in ammonium acetate:acetonitrile was added 1 equivalent of
peroxynitrite, such that the final concentrations of uric acid and peroxyni-
trite were each 10 mM. The mixture was vortexed, and allowed to sit for

Reactions of Peroxynitrite with Uric Acid
125
3 min at room temperature. It was then transferred to LC-MS vials to await
analysis.
Reactions of Uric Acid with Peroxynitrite in Methanol, Ethanol,
and Isopropanol [Reaction 6]
Uric acid powder (5.04 mg) was added to methanol, d4-methanol
[CD3OD; some reactions were also conducted with d3-methanol (CD3OH)],
ethanol or isopropanol and mixed. Uric acid only partially dissolved under
these conditions. One equivalent of peroxynitrite (diluted to 1.5 ml) was
then added to the reaction mixture (final volume was 3 ml of 50:50
alcohol:water). The final concentrations of uric acid and peroxynitrite were
each 10 mM. The reaction mixture was vortexed and allowed to incubate
at room temperature for 3 minutes. Uric acid dissolved completely as the
reaction proceeded. After 3 minutes, at room temperature, the reaction
mixtures were placed in ice and transferred to LC-MS vials on ice and stored
at −80◦C to await LC-MS analysis.
Reactions of 1-Methyluric Acid with Peroxynitrite in Methanol,
Ethanol, and Isopropanol [Reaction 7]
The above series of reactions (see Reaction 6) were carried out in the
same manner using 1-methyluric acid in the place of uric acid in order to
elucidate the mechanism of reaction of peroxynitrite with uric acid.
Reaction of Products of Reaction 1 with Methanol and Ethanol
[Reaction 8]
To an ice-cold solution of the reaction products from Reaction 1
(aqueous buffers), two times the volumes of either methanol or ethanol was
added and vortexed. This cold solution was immediately analyzed by LC-MS.
Reactions of Uric Acid with Peroxynitrite in Human Plasma
[Reaction 9]
15N2-uric acid was also reacted with peroxynitrite using human plasma
diluted (1:3 plasma:buffer) with potassium phosphate buffer (pH 7.4)
rather than in potassium phosphate buffer alone. This set of reactions was
run following the procedure for Reaction 1 with the following exceptions:
15N2-uric acid, at one of three different concentrations, was added to a 1:3
mixture of human plasma and potassium phosphate buffer. Final concentra-
tions of 15N2-uric acid were 0.3 mM (normal physiological concentration),
1.0 mM or 10 mM. Final concentrations of peroxynitrite were 1, 3 and 30
mM, respectively. The volume of the plasma:buffer mixture and the volume

126
C. Gersch et al.
of peroxynitrite to be added to the reaction mixture was varied (as above)
to account for variances in the peroxynitrite stock solution concentration.
All final reaction volumes were 3 mL. Once the reactions were complete,
3.0 mL of ice-cold 20% TCA was added to each of the reaction mixtures in
order to deproteinize the mixture prior to LC-MS analyses. The resulting
reaction mixture:TCA slurries were centrifuged (20 minutes at 4◦C) in
Eppendorf tubes (1.5 mL volume). The supernate was pipetted off of
the centrifuged mixtures and was filtered using TCA-pretreated 0.2 µM
centrifuge filters. The supernate was filtered for 10 min at 4◦C, and was
then transferred to LC-MS vials and stored at −80◦C to await LC-MS
analysis.
Reactions of Uric Acid with Peroxynitrite in the Presence of
Ascorbate [Reaction 10]
Uric acid (300 µL) was added to potassium phosphate buffer (pH 7.4)
containing EDTA. One equivalent of ascorbate (300 µL, see “Preparation
of Reagents”) was then added to the reaction mixtures. Next, 1 and 2
equivalents (separate reactions) of peroxynitrite were added to the reaction
mixtures. The reaction mixtures were then vortexed for 10 seconds and
allowed to incubate at room temperature for 3 minutes. Final reaction
volume was 3.0 mL in each case. Final concentrations of uric acid and
ascorbate were 10 mM in each case. After 3 minutes at room temperature,
the reaction mixtures were placed in ice and transferred to LC-MS vials on
ice. The vials were immediately placed at −80◦C until LC-MS analysis could
be performed.
The Preparation of 4,5-Dimethoxy-4,5-dihydrouric Acid (3K)
Compound 3K was prepared based on the method of Biltz and Heyn[35]
by passing a rapid stream of chlorine (Matheson Tri-Gas) into a suspension
of uric acid in MeOH, the temperature being kept below 15◦C.
LC-MS Analyses of Reaction Mixtures
The reaction mixtures were analyzed by LC-MS in the fullscan (both
positive and negative), tandem mass spectrometry (MS/MS) and single
reaction monitoring (SRM) modes using atmospheric pressure chemical
ionization (APCI) method. In addition, some samples also were further
analyzed using the electrospray ionization (ESI) method. This enabled us
to detect both positive and negative ions present as well as the unique
fragmentation pattern of each intermediates/products present. The LC-
MS analyses were performed using a Finnigan triple quadrupole mass
spectrometer model TSQ 7000 and an Agilent quaternary pump HPLC

Reactions of Peroxynitrite with Uric Acid
127
model 1100 equipped with a Phenominex Luna C18 column (either 150
or 250 × 4.6 mm). Samples were kept cold until analysis to minimize the
decomposition of labile intermediates. The mobile phases used in the LC
separation of the uric acid reaction products included NH4OAc/AcOH and
methanol (as gradient), formic acid/methanol gradient or NH4OAc/AcOH
/acetonitrile gradient. Typical HPLC analyses were carried out in a gradient
elution mode using an aqueous mobile phase A (5 mmol/L ammonium
acetate and 0.1% by volume of AcOH in water) and MeOH as organic
mobile phase B. Mobile phase flow was 0.6 mL/min. Two typical HPLC
gradients used are as follows: Gradient 1 (10 minutes, HPLC run): The
gradient began at 90% A. The composition was linearly ramped to 75% A
over the next 9 minutes, it remained constant for 0.5 minutes, and then
it was reversed to the original composition of 90% A over 0.5 minutes.
Gradient 2 (16 minutes HPLC run): The gradient began at 95% A. The
composition was linearly ramped to 80% A over the next 9 minutes and
to 10% A over the next 3 minutes, it remained constant for 2 minutes,
and then it was reversed to the original composition of 95% A over 1
minute, after which it was kept constant for 1 minute to re-equilibrate the
column.
Evaluation of the molecular weight and fragmentation patterns of the
intermediates and products was performed using the mass spectrometer
after the compounds had been separated on the LC column. The mobile
phase flow was 0.6 mL min−1, and the injection volume was 20 µL. Uric acid
concentrations were determined in SRM mode using APCI source in the
negative mode (reactions monitored for uric acid were m/z 166.9→95.9 at
25V and 166.9→123.9 at 25V and for 15N2-uric acid were m/z 168.9→96.9
at 25V and 168.9→124.9 at 25V). Methylallantoin, allantoin and triuret were
quantitated in the APCI positive mode using their respective SRM ions. In
the TSQ 7000 instrument, nitrogen was used as both the sheath (80 psi)
and the auxiliary (10 units) gases. For APCI mode, the vaporizer was kept
at 500◦C, and the heated capillary temperature was maintained at 200◦C.
The corona current was set to 3 kA by applying approximately 4 kV to the
corona needle. The second quadrupole was used as a collision chamber
with argon as a collision gas at a pressure in vicinity of 2.5 × 10−3 Torr.
The LC-MS/MS ran on Xcalibur software, which is a flexible Windows NT
PC-based data acquisition system that allowed complete instrument control.
The operations of the LC-MS and data analyses were performed using the
ThermoFinnigan Xcalibur software.
We have developed validated LC-MS/MS methods for the quantitation
of uric acid and its metabolites triuret, allantoin, and 6-mainouracil. The
draft of these methods has been submitted elsewhere for publication.
The calibrations for the quantitations were performed using commercial
standards of the metabolites and UA. The correlation coefficients were
0.991∼0.999, from the calibration curve. The intra- and interday precision

128
C. Gersch et al.
(RSD.%) of the metabolites ranged from 0.5% to 13.4% and 2.5% to
12.2%, respectively. For UA, the correlation coefficient was 0.9998∼1.0000
and intra- and interday precision ranged 1.30∼3.07 and 0.61∼
3.68.
Determination of Triuret Levels in Healthy Volunteers, Patients
with Normal Pregnancy, and Patients with Preeclampsia
The triuret concentration in the urine samples determined by LC-
MS/MS developed specifically for the determination of uric acid metabo-
lites in urine. Urine samples from volunteers or patients were stored at
−80◦C until analysis in cryo-vials. Before analysis the samples were diluted
with water and mixed with water and filtered through 0.2 µM filter. The
LC-MS/MS analyses were carried out with APCI interface operated in
positive-ion mode. Liquid chromatography analyses were performed in a
˚
gradient elution mode using Phenomenex Luna 5µ C18 (2) 100A (150 mm
× 4.6 mm) column and ammonium acetate/acetic acid and methanol as
mobile phases. The mobile phase flow was 0.6 mL min−1, and the injection
volume was 20 µL. In mass spectrometry, nitrogen was used as both the
sheath (80 psi) and the auxiliary (10 units) gases. The vaporizer was kept at
500◦C, and the heated capillary temperature was maintained at 200◦C. The
corona current was set to 3 kA by applying approximately 4 kV to the corona
needle. The ions (parent M+1 ion at m/z 147) selected for SRM were m/z
130, 104. and 87.1. at 25V. The quantitation was achieved by using external
calibration standard. The calibration range was 0.004 to 10.0 mg/dL and
it was linear with the R2 value of 0.9964. The operation of the LC-MS
and data analysis was performed using the ThermoFinnigan Xcalibur 1.4
software.
RESULTS
In the past, the purine uric acid was considered to be one of the
endproducts generated during the degradation of RNA, DNA, and adenine
nucleotides. We found, however, that urate reacts with peroxynitrite to
yield reactive intermediates that further react with alcohol to yield uric
acid derived products containing the alkoxyl groups. When the reaction
mixture was left overnight (in about 60 minutes most of the uric acid
had reacted) at ambient temperature before LC-MS analysis, triuret was
identified as the sole final product of the reaction. Formation of triuret
occurs in plasma and in vivo under pathological conditions. Reactions
conducted in the presence of MeOH, EtOH, and i-PrOH resulted in
the alkylation of uric acid intermediates to yield O-alkylated allantoin.
To our knowledge, this alkylation to yield O-alkylallantoin has not been
described in the literature in uric acid’s reactions with peroxynitrite. This

Reactions of Peroxynitrite with Uric Acid
129
alkylation potentially could be an alternate pathway by which peroxynitrite-
activated uric acid could exert biological effects under pathological con-
ditions. The peroxynitrite reactions with uric acid are summarized in
Figure 1.
Reactions of Uric Acid with Peroxynitrite in Aqueous Buffers
A series of reactions of uric acid with peroxynitrite were conducted
with both unlabeled and labeled uric acid and analyzed by LC-MS. The
parallel use of labeled and unlabeled uric acid permitted the identification
of uric acid derived products by LC-MS and LC-MS/MS. Uric acid reactions
were conducted typically at pH 7.4. Some of the reactions were conducted
simultaneously at pHs 7.4 and 9.0 and the pH had no effect on the products
suggesting that the reaction occurs with the mono anion of uric acid.
The final concentrations of uric acid varied from 300 µM (the high end
of the human physiological concentration) to 10 mM and peroxynitrite
concentrations were 1 and 3 equivalents to uric acid. In the ranges studied,
the uric acid and peroxynitrite concentrations had no effect on the products
formed. Treatment with one equivalent of peroxynitrite resulted in the
complete or nearly complete reaction of uric acid.
Figure 2 shows the full scan LC-MS total ion trace of one such reaction
in which kinetic studies were conducted at refrigeration temperature (4◦C).
The reaction initially produced two intermediates: one with a [M+NH4]+
or [M+H]+ (see discussions later) ion at m/z 216 {2B [ 3H (or 3R);
Figure 3]; RT = 4.93 minutes} and second with a [M+H]+ ion at m/z
231 [2D (3K); RT = 6.01 minutes]. At 0 minutes (3 minutes reaction
at ambient temperature, followed by refrigeration), 41% uric acid had
reacted to yield 21% 2B and 20% 2D as determined by LC-MS. At 30
minutes, 60% uric acid had reacted to yield 31% 2B, 23% 2D, and 6%
triuret. By 1hr at room temperature 75% of uric acid had reacted to
produce 3% 2B, 6% 2D, 33% triuret and 33% 2E (3O, m/z 173, RT 3.74
minutes). After 16 hours at room temperature, all the starting materials
and intermediates were converted to triuret (2C). From several studies
conducted by us in aqueous medium, it is evident that the formation of
triuret is a signature reaction of peroxynitrite with uric acid in aqueous
medium. The LC-MS/MS identification of triuret is shown in Figure 4.
Robinson et al.[5] had identified triuret as a product of the reaction of
peroxynitrite with uric acid. Our studies indicate the formation of labile
intermediates that undergoes further reaction with HPLC eluent (MeOH)
to yield alkylated methoxy compounds. When allowed to further react in
aqueous solution (without exposure to MeOH before reaction is complete),
these intermediates undergo conversion to triuret in the reaction mixture.
These intermediates could react with biomolecules with suitable functional
groups to form products other than triuret. No significant quantities of

b u n A e d a n e l R a t i v
b u n A e d a n e l R a t i v
b u n A e d a n e l R a t i v
b u n A e d a n e l R a t i v
130

Reactions of Peroxynitrite with Uric Acid
131
O
O
-2e- -2H+
O
Path 1
Path 2
N
O
O
OH
H
N
H
N
O
OH
O
6
H
N
N
N
1
7
5
+H2
O
HN
HN
HN
O-
*
*
HN
*
O
O
*
O
8
2
N
O
*
N
*
4
O-
N
O-
N
*
NH
9
O
O
O
N
O
O
O-
N
*
NH
5-Hydroxyisourate
+H2
3D
Dehydrourate
3C
3
HO
3A
Urate
O
O
3B
+H2
O
N
O
OH
O
O-
H
N
N
O
-H2
O
Path 3
O
O
H
NH2
N
*
+MeOH
(+ROH)
O
HN
O
CH3 (R)
*
O
-CO2
O
*
Medium
N
N
NH
H
N
O
3G
O-
N
O
HO
NH2
*
H
N
N
+MeOH
H
N
-
+H2
O
H
-H2
O
C-
3F
*
O
HN
HN
(+ROH)
O-
*
*
OH
O
O-
H
N
C
3E
*
NH
N
-2e- -2H+
*
O-
N
*
O
NH
N
+H+
NH2
O
N
3I
*
Methoxyurate M+1 201
O
3H
-MeOH
Medium
O
+H
CH
3(R)
H
N
+MeOH
(-ROH)
+H2
O
*
NH
O
O
NH
OH
-2e- -H+
H
N
(+ROH)
3J
NH2
O
*
O-
NH2
H
N
N
O
O
-CO2
CH
3(R)
O
H
N
O
*
NH
*
NH
O
O
O
O
NH2
HN
H
N
*
3M Allantoin M+1 = 161
O-
*
O
*
N
-CO2
O
NH
HN
*
3L
*
NH
+H+
O-
*
NH
OH
N
O
NH
O
3N
3R
+OH-
*
N
CH
O
3(R)
O
NH
O
O-
3K
O
NH2
NH2
Dimethoxydihydrourate
M+1 = 233
*
CH
3(R)
H
N
NH2
*
*
O
O
NH
NH
OH
-CO2
O
3P
-2e- -H+
NH2
+
CH3
O
O
O
*
NH
O
NH
N
3O
O
H
H
N
O-Methylallantoin M+1 = 175
N
O
NH2
*
HN
HN
*
NH4
O-
*
O
+
*
N* = 15N Label
O
N
N
O
*
N
*
N
NH
O
NH
O
3H
3S
M+1 ion values are for labeled M+1 ions
OH
Triuret
3Q
M+1 =149
H
M+NH4+ = 218
H
R =Et, i-Pr or alkyl
3H1
FIGURE 3 The schematic diagram proposed for the reaction of 1,3-15N2-uric acid with peroxynitrite
in aqueous solutions (Reactions 1 and 2) and in the presence of methanol (Reaction 6; ethanol also
produces similar products with the Et group substituting for Me group). The reactions with unlabeled
uric acid (not shown) produced the corresponding unlabeled products. Reactions conducted in the
presence of MeOH or when reaction product mixture in the cold was treated with MeOH, products 3K
and 3O were produced. Two alternate pathways from uric acid are proposed. One involves the addition of
peroxynitrite or the elements peroxynitrite across the C4-C5 double bond (path 2, Figure 3) of uric acid
to form intermediate 3B and the second one involves the the addition of ·ONO radical to yield 3E, 3F or
(and) 3R. Both pathways explain the formation of m/z 116. Based on several reports of the formation
·
of ONO radicals from peroxynitrite, the pathway leading to 3E is the logical next step in its reaction
with urate. One possibility is that 3B forms from a complex radical mediated process or from partially or
completely formed radicals still enclosed in the solvent shell. In aqueous buffers triuret (3Q) is the final
product with some reactions producing small quantities (1–2%) of allantoin (3M). Reactions in MeOH-
H2O mixtures produced O-methylated allantoin as the dominant final product (96%). This reaction
indicates that uric acid activated by peroxynitrite can act as alklyating agent for biological molecules
containing OH groups, and possibly other groups such as –SH, –COOH, –NH2 and >NH.
alloxan or parabanic acid were detected in the reactions, although small
quantities (1–2%) of allantoin were observed in some reactions.
MS/MS experiments showed that the peak at m/z 216 produced ions at
m/z 199, 167, 156, and 124 at a collision energy of 25 V.
When 1,3-15N2-uric acid was used, all the products observed, including
triuret were doubly labeled suggesting that they all retained nitrogens N1
and N3 of the six membered ring of uric acid.

c e a n u b n d A e v i t l a e R
n c e d a u b n A i v e l a e t R
c e a n u b n d v i t e l A a e R
e c n d a u n b A v i e t a l e R
6 0 9 .
5 7 5 .
4 6 5 .
3 6 5 .
3 0 5 .
3 3 5 .
2 2 5 .
0 2 5 .
1 2 5 .
1 1 5 .
4 8 3 .
n c e d a b u n A i v e l a e t R
e
n c d a b u n A i v e l a e t R
e c n d a u n b A v i e t a l e R
e
n c d a u n b A v i e t a l e R
132

Reactions of Peroxynitrite with Uric Acid
133
Uric Acid Reactions with Peroxynitrite in Aqueous Buffers in the
Presence of PBN
A series of reactions of uric acid with peroxynitrite were conducted with
both unlabeled and labeled uric acid in the presence of excess PBN (5
times that of uric acid and peroxynitrite) and studied by LC-MS its effect
on the products in Reaction 1. The idea behind the experiments was to
see if trapping of potential urate radical by PBN would alter the products
of reactions outlined in Figure 3. Analysis of the reaction products showed
that there were no significant differences in the products between reactions
conducted with and without PBN. The usual uric acid products analyzed by
LC-MS (see Figure 2) were well separated from PBN and hence we were
able to determine the effect of PBN on their formation. However, if small
quantities of PBN adducts were formed under these conditions and they
came under the PBN peak in LC-MS analysis, the existing method will not
be able to determine them. Since these experiments were designed to test
the effect of added PBN on the products in Figure 2, it posed no problem.
Uric Acid Reaction with Peroxynitrite in Ammonium
Acetate Buffer
Reactions of uric acid with peroxynitrite were also conducted in ammo-
nium acetate buffer and a 50:50 mixture of ammonium acetate buffer and
acetonitrile. These reactions produced mainly triuret.
Skinner et al.[31] Reaction
We have repeated some of the reactions of uric acid and peroxynitrite re-
ported by Skinner et al.[31] They have reported the product of the reaction
to be 2-nitrito-4-amino-5-hydroxyimidazoline; whereas, we observed triuret
as the only identifiable product in the LC-MS analyses of these reaction
products (see Figure 4 for the tandem mass spectrometric identification of
triuret).
Uric Acid Reaction Products with Peroxynitrite in Alcohol (MeOH,
EtOH or i-PrOH)-Water (50:50)
Subsequent analyses of Reaction 1 product mixtures showed that the
products 2B[3H or (3R)] and 2D(3O) were formed in the HPLC when
MeOH was used as one of the mobile phases. Use of other common LC-
MS solvents such as acetonitrile did not provide good separation of the
reaction products, and thus were not pursued further. To further explore
the mechanism of this reaction, we chose to expose urate to peroxynitrite
in methanol containing aqueous buffer (from peroxynitrite solutions). This
resulted in the initial formation of products, 3H (or 3R) (15%), unreacted

134
C. Gersch et al.
uric acid (12%), triuret (detected), 3K (23%), and 3O (50%). When this
mixture was left at ambient temperature overnight, all of the unreacted
uric acid and products (other than 3O) were converted to a mixture
containing 96% 3O and 4% triuret. Reaction with labeled uric acid gave the
same correspondingly labeled products. The tandem mass spectrometric
identification of alkylallantoins is shown in Figure 5. Even though 3O
contains the dehydroxyallantoin group, it is not derived form allantoin, but
rather from methoxylated (alkoxylated) precursor 3H.
Uric acid reactions with peroxynitrite were also conducted in the
presence of labeled MeOH (both d3- and d4-methanol gave same products).
Reaction of unlabeled uric acid with peroxynitrite in d4-MeOH produced
primarily d3-labeled 3O (6A; Figure 6) (M+1 = 176) (∼88% of products)
and a mixture of d3- and d6- [∼12% of products; ratio of d3 (M+1 = 234)
and d6(M+1 = 237) = 1.0:0.45] labeled 3K (6B and 6C), corresponding
to the incorporation of the d3 substituted Me group to form d3- and
d6-dimethyl-3K. Small amounts of 3H (3R)(M+1 = 216) were also present.
Reactions of 15N2-uric acid with peroxynitrite in d4-methanol produced
primarily the corresponding d3- and 15N2-labeled 3O (6D) (M+1 = 178)
and d3 and d6 and 15N2-[ratio of d3 (M+1 = 236) and d6 (M+1 = 239)
= 1.0: 0.40]-labeled 3K (6E and 6F), corresponding to the incorporation of
the d3 substituted Me group to form d3- and d6-dimethyl-3K. Small amounts
of corresponding 3H (or 3R) isomer (M+1 = 218) were also present.
We also explored the reaction of uric acid with peroxynitrite in EtOH
and i-PrOH that have greater degree of steric hindrance than methanol. LC-
MS analysis of the EtOH reaction products showed a single peak (RT = 3.92
minutes) with an M+1 ion of m/z 187, corresponding to the O-Et-allantoin
(3O with Et group instead of Me group). Similar LC-MS analysis of the
reaction mixture in i-PrOH produced a dominant peak (8.07 minutes at 12
minutes run; 95%) at m/z 201, corresponding to the isopropyl substituted
3O. No detectable amount of triuret was formed in reactions conducted in
the presence of EtOH and i-PrOH, even though these mixtures contained
50% (by volume) of water. When the ice-cold reaction mixtures of uric
acid and peroxynitrite in aqueous buffer containing preformed reactive
intermediates from previously completed Reaction 1 (3 minutes reaction at
ambient temperature, followed by cooling in ice-water mixture) was treated
with MeOH or EtOH, it resulted in the formation of the corresponding
methylated or ethylated allantoin (3O) as determined by LC-MS analyses.
We also studied the reaction of 1-methyluric acid with peroxynitrite in
MeOH to further confirm our identification of the products and to test the
generality of the reaction of uric acids. LC-MS analysis showed mainly two
peaks (3.60 minutes (26%), 7.05 minutes (74%); 15 minutes analytical run).
The peak at 3.6 minutes had a [M+H]+ ion at m/z 187, corresponding to
the monomethylated 3O and the peak at 7.05 minutes had a [M+H]+ ion

135

136
C. Gersch et al.
2H
2H
2H
2H
2H
2H
2H
2H
2H
O
O
O
H
N
O
H
N
H
N
O
HN
NH2
HN
O
O
O
NH
NH
O
NH
NH
O
NH
O
d
NH
O
O
2H
2H
6A M+1=176
3-Methylallantoin
H
2H
O-d
H
H
6C M+1=237
6B M+1=234
3-Dimethoxydihydrouric acid
d
6-Dimethoxydihydrouric acid
2H
2H
2H
2H
2H
2H
2H
2H
2H
O
O
O
O
H
N
H
N
H
N
H15
N
O
H15
N
15NH2
O
O
O
NH
15NH
NH
O
15NH
O
15NH
NH
O
O
O
2H
2H
H
6D M+1=178
3-Methylallantoin
15N2-O-d
2H
H
H
6E M+1=236
3-Dimethoxydihydrouric acid
6F M+1=239
6-Dimethoxydihydrouric acid
1,3-15N2-d
1,3-15N2-d
FIGURE 6 Structures of labeled 4,5-dimethoxy-4,5-dihydrouric acid and O-methylallantoin.
at m/z 251, corresponding to the monomethyl 3K. When the same reaction
was conducted in d4-methanol, the peak at 3.6 minutes had an [M+H]+ ion
of m/z 190, corresponding to the incorporation of the d3 substituted Me
group to form d3-methyl-3O and the peak at 7.05 minutes had an [M+H]+
ion of m/z 251, corresponding to the incorporation of two d3-Me groups
from solvent to form N-methyl-3K.
The unlabeled compound 3K was synthesized using the reaction of
chlorine[35] with uric acid in MeOH. It had the same mass spectrum,
retention time and fragmentation pattern as 3K in LC-MS analysis. This
provided further confirmation of the structure of 3K.
Reactions of Lithium Urate with Peroxynitrite
We also investigated the reaction of lithium urate (more soluble at
physiological pH). When urate was exposed to peroxynitrite under these
conditions, the reaction proceeded in the same fashion as it did when the
urate had been dissolved in KOH and then added to the phosphate buffer
prior to being exposed to peroxynitrite.

Reactions of Peroxynitrite with Uric Acid
137
RT 0.00 -
RT 0.00 -
RT:
RT:
MA:
MA:
100
95
90
85
80
75
70
65
60
55
50
100
95
90
85
RT:
MA:
PN+UA+ASC=1:1:1
PN+UA+ASC=2:1:1
Triuret
80
75
70
65
60
55
50
45
40
PN=peroxynitrite
UA=Uric Acid
ASC=Ascorbate
45
40
Uric acid
35
30
35
30
Uric Acid
14.9
14.7
25
20
25
20
14.4
13.9
14.3
14.0
1
1
15
10
13.7
13.5
13.6
13.3
12.7
13.0
3.5
6.2
6
5
5
0
0
0
2
4
6
8
1
1
1
0
2
4
8
1
12
14
Time
Time
FIGURE 7 LC-MS analysis (total ion current) of the reaction of uric acid in the presence of 1 molar
equivalent of ascorbate, and 1 (left) or 2 (right) molar equivalents of peroxynitrite. In a 1:1:1 molar
ratio of ascorbate:urate:peroxynitrite, ascorbate appears to completely block the reaction of urate with
peroxynitrite, as unreacted uric acid is observable, but the product triuret is not. In increasing the
concentration of peroxynitrite, such that we have a 1:1:2 molar ratio of ascorbate:urate:peroxynitrite,
the ascorbate is still able to partially block the reaction of urate with peroxynitrite, as some unreacted
uric acid is observable, as well as the product triuret.
Reaction of Urate with Peroxynitrite in the Presence of
antioxidants
We next decided to test the possibility that the reaction of urate with
peroxynitrite could be blocked by ascorbate. To this end, we added one
equivalent of ascorbate to the phosphate buffer containing urate prior
to the addition of the peroxynitrite. In reactions conducted using a 1:1:1
molar ratio of ascorbate:urate:peroxynitrite, urate was well protected from
oxidation by peroxynitrite by ascorbate. As illustrated in the full scan LC-
MS total ion trace (Figure 7), neither the intermediates nor triuret were
observable when the 1:1:1 molar ration of reagents was used. When a
1:1:2 molar ratio of ascorbate:urate:peroxynitrite was used, urate was only
partially (70%) oxidized to triuret (Figure 7), and none of the reactive
intermediates were observable.
Reaction of Urate with Peroxynitrite in Human Plasma
We added 15N2-urate to human plasma (in order to distinguish any
reaction products from endogenous reaction products of urate) and then

138
C. Gersch et al.
RT: 4.76
148.8
MA: 8302674
100
95
100
95
Panel A
Panel B
90
90
85
80
75
85
80
75
165.9
70
65
70
65
60
M+1 ion
M+
Labeled
Triuret (total
Ion current of
m/z 148.8)
Ammo-
nium
ion
60
55
50
55
50
45
40
45
40
35
30
25
35
30
25
20
15
20
15
10
5
10
5
180.8
166.9
168.1
149.9
4.44
147.8
132.9
5.86
115.7
129.8
3.87
6.17
12.99
105.9
162.3
186.8
6.93
8
12.77
190.7
202.8
8.71
1.01
0
0
0
2
4
6
10
12
14
120
140
160
180
200
Time (min)
m/z
FIGURE 8 LC-MS identification of 15N2-triuret from plasma reactions of 15N2-uric acid with peroxyni-
trite: A) the total ion chromatogram of the plasma extract; B) plot of m/z 149; and C) the ions of triuret
peak at 4.76 minutes (panel A). m/z 149 is the M+1 ion of labeled triuret and m/z 166 is the ammonium
adduct of labeled triuret.
exposed the labeled urate to peroxynitrite. Figure 8 shows the identifi-
cation of labeled triuret from plasma reactions of labeled uric acid with
peroxynitrite by full scan LC-MS total ion trace. This was also confirmed by
SRM analysis. The formation of triuret from the reaction of uric acid from
peroxynitrite, even in the presence of plasma components suggests that
measuring triuret concentration can indicate the extent of peroxynitrite
reaction in vitro and in vivo.
Determination of Triuret Levels in Healthy Volunteers, Patients
with Normal Pregnancy, and Patients with Preeclampsia
We have conducted a preliminary investigation with a limited number of
subjects who are normal healthy volunteers, patients with normal pregnancy
and patients with preeclampsia. The triuret concentration data from the

Reactions of Peroxynitrite with Uric Acid
139
FIGURE 9 Triuret concentrations in the urine samples from normal healthy volunteers (N = 6),
patients with normal pregnancy (N = 7), and patients with pre-eclampsia (N = 9).
LC-MS/MS analysis of urine samples is shown in Figure 9. As is evident there
are significant differences in the amount of triuret in healthy volunteers and
patients.
DISCUSSION
Peroxynitrite reacts with lipids, DNA and proteins via direct oxidative
reactions or indirect radical-mediated mechanisms.[36] In this paper we
report the formation of reactive intermediates and alkylated products from
the reaction of peroxynitrite with uric acid. These intermediates further
react in aqueous medium to form triuret or react with molecules with
functional groups such as OH to form alkylated products. The study scheme
is summarized in Figure 1.
While the reaction of uric acid with peroxynitrite might be beneficial
as a mechanism for removing peroxynitrite, in this paper we demonstrate
that reaction also leads to the formation of unique reactive intermediates
that further react with alcohols to give alkylated products or react with
aqueous medium to give triuret. The first labile product (2B) observed had
an m/z 216 (218 for the labeled uric acid). This indicates that the molecular
weight of the intermediate is 215, demonstrating that the intermediate has
gained a nitrogen (a mass of 215 indicates odd number of nitrogens).
The mass difference of 47 units (between uric acid and 2B) indicates a
gain of elements of HONO group. We considered structures 3R and 3S
shown in Figure 3 as potential compound 2B. MS/MS experiments at 25
V produced ions at m/z 199, 167, 156 and 124, representing the loss of NH3

140
C. Gersch et al.
from 216, MeOH from 199, HN=C=O from 199 and HN=C=O from 167,
respectively. This could be potentially could be the ammonium adduct of 3H
(structure 3H1) (3H itself was not identified in the reaction mixtures). The
formation of 3H can lead to the formation of additional products 3K and
3O that were observed in the reaction. Thus, the ammonium adduct 3H1
can be formed in the mass spectrometer (LC-MS analysis were conducted in
ammonium acetate/acetic acid buffer) and the formation of such adducts
are common in LC-MS analyses.[37] There is some doubt about 2B being
3H1, as 3H itself has not been identified and under the conditions we
employed for LC-MS analysis, ammonium adduct formation, if present is
a minor pathway. There is a possibility that the intermediate 2B is a mixture
(unresolved peak) of 3R and the ammonium adduct of 3H (both will
have the same M = 1 ion). Whether it is 3R or ammonium adduct of
3H, this intermediate is unstable and leads to triuret or alkylated allantoin
3O (depending on the medium). As stated, the attempt to isolate 2B by
preparative HPLC resulted in its complete conversion to triuret (the isolated
fraction contained only triuret).
One can visualize the addition of peroxynitrite or the elements perox-
ynitrite across the C4-C5 double bond (path 2, Figure 3) of uric acid to form
intermediate 3B and 3B can lose the elements of water (from C4 and N9)
to give the intermediate 3F.
3 F also can be formed directly by the addition of ·ONO radical to urate
followed by the loss of H· to the medium (path 3) or the abstraction of
hydrogen by the resulting radical from the medium to give 3R. 3F can also
be generated from 3R by oxidation. 3F can exchange its ONO group with
water or MeOH to yield 3D or 3H, respectively. These then can lead to
other observed products. It has been reported[38–45] that peroxynitrite can
dissociate either partially or completely to yield HO· and ·ONO radicals, in
the following manner:
NO· +· O−2 → ONOO−
ONOO− + H+
ONOOH
→
←
ONOOH → [·OH- - - -NO2· ] → ·OH + NO2·
·
In addition, others have reported or predicted the formation of NO2
radical from peroxynitrite or equivalent compounds.[46–48]
One possibility is that 3B forms from a complex radical mediated process
or from partially or completely formed radicals still enclosed in the solvent
shell.
For oxidation reactions of uric acid, other authors[5,6] had proposed
a pathway involving two electron oxidations (path 1, Figure 3) to yield
dehydrourate 3C which can further react with water to yield allantoin and

Reactions of Peroxynitrite with Uric Acid
141
triuret. This mechanism may not be oper·ative in peroxynitrite reactions
and it is possible that path 3 (addition of ONO radical) or possibly path
2 could be operative as evidenced by the following. Methyl substitution at
position N7 in methylated uric acids should prevent the reaction to form 3C
equivalent because the double bond cannot be formed between C5 and N7.
This may mean that the reaction may not proceed. But, our investigations
(data not shown here) of the peroxynitrite reactions of a series of methyl-
substituted uric acids (substituted at the nitrogen positions) showed that
the reaction with peroxynitrite proceeds to produce peroxynitrite reaction
products including the dimethoxy compound equivalent to 3K. We believe
that the primary unstable intermediates formed in our reactions are 3E and
3F (path 3) as a result of the addition of the ·ONO radical to uric acid and
they then lead to the products observed. In addition, it is possible that 3R is
formed as an intermediate.
Therefore, we are proposing that 3H as a possible intermediate for 3K
and 3O. As mentioned this intermediate is labile and has been identified
only in ice-cold reaction mixtures. The attempted isolation of peak 2B by
preparative HPLC resulted in its conversion to triuret (isolated fraction
contained only triuret). In some reactions small quantities (1–2%) of
allantoin was formed, likely through 3G and 3K that leads to triuret. Kahn
and coworkers[49] identified 3G and 3J by NMR in urate oxidations by
uricase. The pathway for the formation of triuret follows that of Robinson
et al.[5]
The reactive intermediates further react to generate the end product
of this reaction, triuret, which now appears to be a signature end product.
The reaction occurs with 1 equivalent of peroxynitrite. Use of higher
concentrations of peroxynitrite (up to 3 equivalents) did not alter the
products observed by LC-MS. Within the range of pHs tested (pH 7.4 and
9), pH had no effect on the products suggesting that it is the mono anion
of urate that reacts with peroxynitrite. In addition, the cations present (Li,
K, or Na) had no effect on the products observed. While the observation
that triuret is a product of the peroxynitrite and uric acid reaction had been
previously shown,[5] to our knowledge this reaction product had not been
documented in any biological system.
There is a possibility that nitric oxide can be generated from perox-
ynitrite and that this can further react with UA. As reported in our recent
publication,[53] nitric oxide reactions with UA produce 6-aminouracil as the
sole product. No 6-aminouracil was observed in any of the reactions of UA
with peroxynitrite, thus excluding the possible generation of any significant
amounts of nitric oxide.
The presence of MeOH used either as eluent or intentionally added to
investigate the alkylation potential of uric acid activated by peroxynitrite,
results in the partial conversion of 3H to 3K, the dimethoxy adduct of
uric acid (4,5-dimethoxy-4,5-dihydrouric acid). Under these conditions, the

142
C. Gersch et al.
final stable product formed is the methylated allantoin (3O), likely formed
through the intermediate 3H. We have not identified any direct reaction
product of 3K. The final product of its reaction appears to be 3O formed
through 3H. We have multiple levels of confirmation of the products 3O
and 3K using studies with labeled uric acid, methyluric acid, d4-MeOH and
reactions with other alcohols.
Since products 2B(3H), 2D(3K) and 2E(3O) shown in Figure 2 are
likely to be generated by the reaction of MeOH from eluent (reactions
conducted in ammonium acetate produced only triuret; see Reaction 5)
with reactive intermediates formed before the HPLC analysis, the extent of
their formation at a given time can be used as a approximate measure of
the reactive intermediates in the solution that was injected in the HPLC.
These intermediates could be 3B, 3C, 3C, or other unknown species that
contain the elements of uric acid (all products contain all or part of uric
acid). Thus, based on the above reasoning, one can estimate that 41% of
reactive intermediate(s) remained in solution at T = 0 (3 min at room
temperature), 54% at T = 30 minutes, 42% at T = 1 hour, and 0% at T
= 16 hours.
In the reactions conducted with 1,3-15N2-uric acid, the corresponding
dinitrogen labeled products H (or 3R), 3K, 3O and 3Q] were formed in all
cases. This shows that the identified products all retained the nitrogens 1
and 3 of uric acid. The fact that the labels were retained also lends support
to the proposed reaction scheme. The structure of 3K was confirmed by the
preparation of the compound and comparison of its LC-MS data with that
of the prepared compound. LC-MS/MS studies were conducted to further
confirm the structures of labeled and unlabeled O-methylallantoin (3O).
The fragmentation patterns are shown in Figure 5. The major fragmentation
pattern arises from the loss of HN=CO group from the 5 membered ring
(the lost HN=CO group does not contain any of the nitrogen labels) and
other characteristic peaks arise from the loss of MeOH as well as urea. The
mass spectrometry data lend support to the proposed structures. This is
also indicative of the type of alkylation products that could be formed with
biomolecules.
When the cold reaction mixtures of uric acid and peroxynitrite in
aqueous buffer (from Reaction 1)(reaction time 3 min at ambient tem-
perature, followed cooling in ice-water mixture) were treated with MeOH
or EtOH, it resulted in the formation of the corresponding methylated
or ethylated allantoin (3O), as determined by LC-MS analysis after the
addition of MeOH or EtOH. This is significant because it suggests that
the preformed intermediates in solution were able to alkylate alcohols and
that the intermediates live long enough in solution to participate in further
reactions.
As indicated, the reaction with alcohol is highly preferred even in the
presence of excess water. It occurs even in the presence of low levels of

Reactions of Peroxynitrite with Uric Acid
143
alcohols, as seen in the LC-MS studies using alcohol as one of the mobile
phases (low organic content). We believe that compounds containing
labile hydrogen such as OH can get alkylated once UA is activated by
PN. There are numerous such groups in biological molecules. While it is
true pure alcohols are not abundantly present in the biological milieu, the
observation that uric acid can alkylate chemical species containing hydroxyl
groups suggests that alkylation could be a novel mechanism by which uric
acid acts. Further studies are required to determine which types of biological
compounds might be targeted.
We also examined the products formed from the reaction of uric acid
and peroxynitrite that had been reported by Skinner et al.[31] We observed
only triuret when replicating the same reaction conditions, where the
authors reported 2-nitrito-4-amino-5-hydroxyimidazoline as the product of
the reaction. Our results are consistent with that of Robinson et al.[5]
We selected alcohols as model compounds for biomolecules containing
OH and other active groups and investigated the reactions of uric acid with
peroxynitrite in their presence. Reactions of uric acid with 1 equivalent
of peroxynitrite conducted in methanol initially produced a mixture of
3H (or 3R) (14.6%), unreacted uric acid (11.6%), triuret (detected), 3K
(23.1%), and 3O(50.6%). When this mixture was left at ambient tem-
perature overnight, all of the unreacted uric acid and products (other
than 3O) were converted to a final mixture containing 96% 3O and 4%
triuret. The reactions produced 96% of the alkylation product 3O, when
the medium contained about 50% (by volume; approximately 64% by
mole concentration) water (from buffer) and 50% methanol (by volume;
approximately 36% by mole concentration) and this suggest a high degree
of affinity for the alkylation reactions versus the formation of triuret via the
reaction with water.
Reactions of unlabeled uric acid with d4-MeOH produced primarily
d3-labeled 3O (4A) (M+1 = 176) (∼88% of products) and a mixture of d3-
and d6- (∼12% of products)(ratio of d3- (M+1 = 234; 4B) and d6-(M+1
= 237; 4C) = 1.0:0.45) labeled 3K and small quantities of (less than 1%)
3H (or 3R) (M+1 = 216). The fact that 4B and 4C are present at a ratio of
about 2:1 rather than 4C alone suggests that one of the methoxyl groups in
4C is labile and that it partially exchanges with the methanol in the HPLC
eluent. Such an exchange may go through the formation of monomethoxyl
3H. The presence of only small quantities of 3H (or 3R) suggests that it has
already further reacted with the d4-methanol to give 3K or (and) 3O. Reac-
tions of 15N2-uric acid with peroxynitrite in d4-methanol produced similar
results.
We also investigated the reaction of 1-methyluric acid with peroxynitrite
in MeOH and it further confirmed our identification of the products (the
corresponding methyluric acid products were formed) and it showed the
generality of the reaction of uric acids with peroxynitrite.

144
C. Gersch et al.
When lithium urate was exposed to peroxynitrite under conditions of
Reaction 1, the reactions produced the same products as when the urate
had been dissolved in KOH and then added to the phosphate buffer prior
to being exposed to peroxynitrite. This implies that the counter ion had no
effect on the products of the reaction and that Li can be used when solubility
of urate is a concern.
During the course of our investigations, we wondered whether the
reaction of uric acid with peroxynitrite could be blocked by ascorbate. In re-
actions conducted using a 1:1:1 molar ratio of ascorbate:urate:peroxynitrite,
urate was well protected from oxidation by peroxynitrite by ascorbate. As
illustrated in the full scan LC-MS total ion trace (Figure 7), neither the
intermediates nor triuret were observable when the 1:1:1 molar ratio of
reagents was used. When a 1:1:2 molar ratio of ascorbate:urate:peroxynitrite
was used, urate was partially (70%) oxidized to triuret (Figure 7), and none
of the reactive intermediates were observable. It is unclear from our studies
whether ascorbate degraded peroxynitrite, making it unavailable for further
reaction or whether it preferentially reacted with peroxynitrite.
In order to determine whether or not the reaction of urate with
peroxynitrite could occur in a biological milieu, we added 15N2-urate to
human plasma and then exposed the labeled urate to peroxynitrite. Figure
8 shows the identification of labeled triuret from plasma reactions of
labeled uric acid with peroxynitrite by full scan LC-MS total ion trace.
The formation of triuret from the reaction of uric acid from peroxynitrite,
even in the presence of plasma components suggests that measuring triuret
concentration can indicate the extent of peroxynitrite reaction in vitro and
in vivo.
Preeclampsia is a pregnancy-specific disorder characterized by hyper-
tension and proteinuria that is associated with increased oxidative stress
and reduced antioxidant defenses.[50] As part of an ongoing study we
have conducted the LC-MS/MS analysis of triuret in the urine samples
from preeclampsia patients (Figure 9). In the limited subjects investigated,
the triuret levels were: normal healthy volunteers ꢁ patients with normal
pregnancy ꢁ preeclampsia patients. Normal pregnancy is associated with
some oxidative stress,[51,52] which is amplified in preeclampsia patients. No
one has previously reported the presence of triuret in any human, animal
or cell culture studies. This data is indicative of the potential biological
relevance of triuret pathway. A recent preliminary study by us also indicates
that triuret levels are elevated in smokes compared to non smokers. More
detailed studies with more subjects are needed to confirm and extend these
findings.
When generated in vivo, the intermediates generated by the reaction of
peroxynitrite with uric acid are potentially capable of alkylating molecules
with suitable functional groups in their surroundings. The potential for
such an alkylation of –OH, –SH, –NH2, >NH and –COOH groups is

Reactions of Peroxynitrite with Uric Acid
145
R
O
R
S
H
N
H
N
NH2
O
NH2
O
NH
O
NH
10A
NH
O
NH
10B
PN + ROH
PN + RSH
O
R
H
N
O
HN
O
PN + RCOOH
H
N
O
NH2
NH
O
NH
O
Uric acid
NH
O
NH
R2
NH
R1
PN +
PN + RNH2
10D
R
R1
R2
H
N
H
N
HN
NH2
N
NH2
O
O
NH
NH
O
NH
O
NH
10F
10E
FIGURE 10 Proposed scheme for the potential peroxynitrite activated uric acid’s alkylations of
functionalized biomolecules.
proposed in Figure 10. This could be of particular biological significance, as
enzymes, DNA, and RNA, as well as other biomolecules could be potential
targets of these intermediates. This alkylation potential of biomolecules
is undergoing further study in our group. Taken together, these results
lead us to conclude that under conditions of oxidative stress, uric acid
can form reactive intermediates. These reactive intermediates might be
important in explaining how uric acid contributes to the pathogenesis of
diseases such as hypertension, metabolic syndrome, preeclampsia, renal
disease, and cardiovascular disease; as well as how it might contribute to
the characteristic endothelial dysfunction associated with these conditions.
A possible scenario is that the reactive intermediates can further react with
components of endothelial cells and cause damage and disruption to their
normal function. Perhaps part of this damage, such as alkylation, could be
irreversible and cumulative.