A novel procedure for generating both nitric oxide and superoxide in situ from chemical sources at any chosen mole ratio. First application: Tyrosine oxidation and a comparison with preformed peroxynitrite

Article abstract of DOI:10.1021/tx0001272

The first method for generating ·NO and O₂^·- at any known, constant ratio has been developed. Spermine NONOate and di(4-carboxybenzyl)hyponitrite decay with first-order kinetics and exactly equal rate constants (half-lives of 80 min) at 37 °C and pH 7.5 to give 200 and 40 mol % ·NO and O₂^·- respectively. Tyrosine oxidation to dityrosine and 3-nitrotyrosine (the major and minor products under the conditions used in these experiments) has been studied (mainly in the presence of CO₂) using various different ratios of the rates of formation of ·NO and O₂^·-. The ·NO/O₂^·- = 1.0 product profiles are very similar to those of the products derived from equal amounts of ·NO and O₂^·- generated at a ·NO/O₂^·- ratio of 1.0 from SIN-1 but are very different from those derived from preformed peroxynitrite. All the experimental results can be explained in terms of free radical chemistry. The product profiles at all the ·NO/O₂^·- ratios could be satisfactorily simulated provided an important group of reactions which lead to the consumption of dityrosine was included.

Full text of DOI:10.1021/tx0001272

Chem. Res. Toxicol. 2000, 13, 1287-1293
1287
A Novel P r oced u r e for Gen er a tin g both Nitr ic Oxid e a n d
Su p er oxid e in Situ fr om Ch em ica l Sou r ces a t An y
Ch osen Mole Ra tio. F ir st Ap p lica tion : Tyr osin e
Oxid a tion a n d a Com p a r ison w ith P r efor m ed
P er oxyn itr ite
George R. Hodges,* J asvir Marwaha, Thomas Paul,* and K. U. Ingold*
Steacie Institute for Molecular Sciences, National Research Council of Canada,
Ottawa, Ontario, Canada K1A 0R6
Received J une 19, 2000
•
•-
The first method for generating NO and O2 at any known, constant ratio has been
developed. Spermine NONOate and di(4-carboxybenzyl)hyponitrite decay with first-order
kinetics and exactly equal rate constants (half-lives of 80 min) at 37 °C and pH 7.5 to give 200
and 40 mol % •NO and O2•-, respectively. Tyrosine oxidation to dityrosine and 3-nitrotyrosine
(the major and minor products under the conditions used in these experiments) has been studied
(mainly in the presence of CO2) using various different ratios of the rates of formation of •NO
•
and O2•- . The NO/O2•- ) 1.0 product profiles are very similar to those of the products derived
•
•-
•
•-
from equal amounts of NO and O2 generated at a NO/O2 ratio of 1.0 from SIN-1 but are
very different from those derived from preformed peroxynitrite. All the experimental results
•
•-
can be explained in terms of free radical chemistry. The product profiles at all the NO/O2
ratios could be satisfactorily simulated provided an important group of reactions which lead
to the consumption of dityrosine was included.
yields dityrosine (di-Tyr)1 and 3-nitrotyrosine (NO2-Tyr)
(9) or that nitrotyrosine has been identified in proteins
at sites of inflammation such as atherosclerotic lesions
(10, 11). The tyrosyl radical intermediate in this reaction
has been observed by ESR spectroscopy (12).
In tr od u ction
Peroxynitrite may be formed in vivo by the diffusion-
controlled coupling of endogenous superoxide and nitric
oxide. Its conjugate acid decomposes rapidly (t1/2 ) 0.6 s
at 25 °C)
+
H
O2•- + •NO f -OONO y z HOONO f
HNO3 + HO• + NO2 (1)
•
giving nitrate together with smaller quantities of the
hydroxyl radical and nitrogen dioxide (1-5). The yield
of the hydroxyl radical is ca. 28% at pH 6.3 (5) and
declines at higher pHs (2, 3). Under physiological condi-
tions, these reactions are superseded by a reaction of the
peroxynitrite anion with carbon dioxide (k ) 3 × 104 M-1
s-1) (6) to give a short-lived adduct which decomposes to
ca. 67% nitrate and ca. 33% carbonate radical anion and
Ex vivo reactions of preformed peroxynitrite with
biological substrates are of doubtful relevance in vivo
•
where both NO and O2•- must also be present, either of
•
NO2 (5, 7, 8).
which may react with the bioradical initially produced
by HO•, NO2 , or CO3•-. The absence of •NO and O2•- can
•
be overcome by using SIN-1 (3-morpholinosydnonimine)
which generates both NO and O2•-. The SIN-1 is first
•
hydrolyzed to SIN-1A which is oxidized by O2 to SIN-1•+
and O2•-, and the SIN-1•+ then spontaneously decom-
poses, releasing •NO (13). The yields of •NO and O2•- from
SIN-1 should be equal, but it seems unlikely that these
two radicals are commonly (or ever) produced in equal
amounts in vivo. It has also been demonstrated (13) that
other electron acceptors can compete with O2 for SIN-
The radicals formed from peroxynitrite are very strong
(HO•) or moderately strong (CO3•- and NO2 ) one-electron
•
oxidizing agents. It is, therefore, not surprising that
treatment of tyrosine (Tyr) with preformed peroxynitrite
* To whom correspondence should be addressed: Steacie Institute
for Molecular Sciences, National Research Council of Canada, 100
Sussex Dr., Ottawa, Ontario, Canada K1A 0R6. Telephone: (613) 990-
0938. Fax: (613) 941-8447. E-mail: george.hodges@avecia.com,
thomas.paul@nrc.ca, and keith.ingold@nrc.ca.
1
Abbreviations: di-Tyr, dityrosine; NO2-Tyr, 3-nitrotyrosine;
SIN-1, 3-morpholinosydnonimine; s-NONOate, spermine NONOate;
XO, xanthine oxidase.
10.1021/tx0001272 CCC: $19.00 Published 2000 by the American Chemical Society
Published on Web 11/23/2000

1288 Chem. Res. Toxicol., Vol. 13, No. 12, 2000
Hodges et al.
1A and, in the presence of such oxidants, SIN-1 becomes
“just” a source of •NO. Although the abilities of CO3•- and
HO• to effect one-electron oxidation of SIN-1A were not
examined (13), there is at least a possibility that these
radicals could function in such a way and cause SIN-1
with preformed peroxynitrite. However, all our results
can be explained in terms of known radical chemistry.
Exp er im en ta l P r oced u r es
Ch em ica ls. SOTS-1 and di-Tyr were synthesized as de-
scribed previously (17, 19); other chemicals were of the highest
purity and used as supplied: Chelex 100, Tyr, and NO (Aldrich),
freshly manufactured s-NONOate (both Alexis Corp. and Mo-
lecular Probes), H2O2 (Anachemia), phosphate (BDH), O2 and
CO2 (Praxair), methanol (EM Science), and NO2-Tyr (Sigma).
Rea ction s w ith P r efor m ed P er oxyn itr ite. Unless stated
otherwise, all reactions were performed at 37 °C, in 50 mM
Chelex 100-treated phosphate buffer (pH 7.5). Reaction mixtures
were first purged with and then the reactions were carried out
under 100% O2 or 90% O2/10% CO2 [resulting in 4 mM CO2(aq)
and 80 mM HCO3-] (20). Purging the solutions with CO2 slightly
reduced their pH which was countered by the addition of sodium
hydroxide, after which the solution was purged for at least a
further 15 min and the pH checked to ensure it was 7.5 before
the reaction was started.
Rea ction s w ith SOTS-1 a n d /or s-NONOa te. Solutions of
SOTS-1 and Tyr (4 mM) were made up at 0 °C and purged for
ca. 30 min with the desired gas. This solution was then added
to solid s-NONOate and heated to 37 °C with gentle rocking in
a water bath while being kept in a sealed system under a slight
positive pressure (provided by a small balloon) of the original
purging gas throughout the experiment. The solution was not
stirred, and gas was not bubbled through it or passed over it
during reaction to avoid displacement of the •NO. Samples were
analyzed with a HPLC system fitted with a Spherisorb ODS-2
5a column (Waters). The eluent was 20 mM phosphate buffer
(pH 3.0) and methanol [97/3 (v/v) between 0 and 10 min followed
by ramping over the course of 2 min to 90/10 (v/v)]. The retention
times of Tyr (followed at 276 nm), di-Tyr (followed by fluores-
cence; λex ) 284 nm, λem ) 410 nm), and NO2-Tyr (followed at
365 nm) were 6.7, 10.6, and 15.0 min, respectively.
•
•-
to generate an excess of NO over O2
.
Another approach has been to combine a chemical
•NO source, spermine NONOate (s-NONOate), with an
enzymatic O2•- source, xanthine oxidase (XO) along with
hypoxanthine (14, 15).
Unfortunately, the XO/hypoxanthine (or other substrate)
•-
reaction produces both O2 and, in comparable but
usually higher yields, H2O2, and the organic products
derived from the substrates themselves react with per-
oxynitrite. Moreover, XO is commonly contaminated with
adventitious iron, so that Fenton and Haber-Weiss
chemistry may also occur. Furthermore, XO generates a
•-
relatively short burst of O2 which cannot easily be
•
matched to the first-order formation of NO from the
NONOate. Since the products derived from tyrosine
depend on the absolute and relative amounts of •NO and
•
O2•- (vide infra), any change in the NO/O2•- ratio during
an experiment will limit the derivable information.
The reaction of tyrosine (2 mM) with equal concentra-
•
•-
tions of NO and O2 has also been studied recently by
the radiolysis of aerated 50 mM phosphate buffer (pH
7.6) containing 0.38 mM CO2, 10 mM nitrite, and 1 M
methanol (16). Unfortunately, this technique does not
permit the generation of •NO and O2•- at ratios other than
Kin etic Sim u la tion s. The products formed after 4.5 h by
the action of SOTS-1 and s-NONOate on Tyr were modeled on
107 molecules using the “Chemical Kinetics Simulator” (CKS,
version 1.01, IBM), a program which uses a stochastic algorithm
to calculate the consequences of a given set of reactions (21). A
few simulations on 108 molecules reduced the stochastic “noise”
but gave results which were not significantly different.
•
•-
1/1. The probable effects of different NO/O2 ratios on
the yield of NO2-Tyr had, therefore, to be estimated by
kinetic simulation.
We have overcome the above-mentioned difficulties by
combining a clean, chemical source of •NO (s-NONOate,
•-
reaction 4) with
a clean, chemical source of O2
Resu lts
(SOTS-1, eq 5) (17, 18).
Ma tch in g •NO a n d O2 F or m a tion Ra tes. Each
•-
•
s-NONOate molecule yields two molecules of NO (22),
O2
9
•-
[-O2C-C6H4-CH2ONd]2
SOTS-1
•-
f f
8 0.4O2
(5)
while 0.4 molar equiv of O2 is produced from O2 by
SOTS-1 (17). The rate of decomposition of SOTS-1 has
1
been measured by H NMR spectroscopy and found to
The great advantage of this new system is that the
decompositions of s-NONOate and SOTS-1 both follow
first-order kinetics with rate constants that are exactly
equal at 37 °C and pH 7.5. This means that the ratio of
be independent of oxygen pressure and pH (data not
shown). In contrast, the rate of NO release depends on
•
pH (22) because protonation of the s-NONOate is neces-
•
sary for NO release. At 37 °C in 50 mM phosphate buffer,
•
•-
the rates of formation of NO and O2 can be set to any
value, not necessarily equal, and this ratio will be
maintained until the s-NONOate and SOTS-1 have
decomposed completely (g4.5 h). Although the rates of
s-NONOate decay, monitored at 252 nm, followed clean
first-order kinetics from pH 7.6 to 7.2, and at pH 7.5,
the half-life was 4900 s (independent of the presence of
SOTS-1, data not shown). This exactly matches the half-
life of SOTS-1 under the same conditions (17) [but is, for
reasons, e.g., pH, ionic strength, etc., which were not
explored, about twice the half-life reported in previous
studies at pH 7.4 (23, 24)]. Since reactions 4 and 5 are
both first-order and have equal rate constants, the ratio
of the rates of formation of •NO and O2•- will be constant
throughout an experiment irrespective of the chosen
•
•-
formation of NO and O2 decline exponentially, this
could be overcome by adding “makeup” SOTS-1 and
s-NONOate every 20-30 min. We did not bother with
this refinement for our first application of this new
system which was tested on tyrosine, and the results
were compared with those obtained using SIN-1 and
preformed peroxynitrite. There are very significant dif-
ferences in product yields as the ratios of the rates of
•NO/O2 ratio. SIN-1, another source of NO and O2
,
•-
•
•-
•
•-
formation of NO and O2 are changed, and even at a
1/1 ratio, the product yields differ from those obtained
decays with a half-life of about 2500 s at 37 °C in
phosphate buffer at pH 7.4 (13).

Tyrosine Oxidation by Superoxide and Nitric Oxide
Chem. Res. Toxicol., Vol. 13, No. 12, 2000 1289
F igu r e 1. Buildup of di-Tyr during the reaction of SOTS-1 (1.25
•
mM ≡ 0.5 mM O2•-) and s-NONO (0.25 mM ≡ 0.5 mM NO)
with Tyr (4 mM) in 50 mM phosphate buffer (pH 7.5) at 37 °C
under an atmosphere of a 9/1 O2/CO2 mixture. The inset shows
formation of di-Tyr after 4.5 h when SOTS-1 (0.5 mM ≡ 0.2 mM
•
O2•-) and s-NONOate (0.1 mM ≡ 0.2 mM NO) react with Tyr
(0.4-4.0 mM) under the same conditions: (bars) experimental
results (errors of less than (15%) and (O and b) simulated yields
(black and white circles refer to k-8 values of 4 × 105 and 2 ×
105 M-1 s-1, respectively; see the text for details). Connecting
lines are only a visual aid.
Rea ction of Tyr w ith O2•-. Di-Tyr was not detected
after the incubation (4.5 h) of oxygenated SOTS-1 (1.25
•
F igu r e 2. Effect of NO concentration on di-Tyr and NO2-Tyr
•-
mM ≡ 0.5 mM O2 in total) with Tyr (4 mM) in the
•-
(inset) yields after 4.5 h from SOTS-1 [1.25 mM ≡ 0.5 mM O2
•-
presence or absence of CO2.
(top) and 2.5 mM ≡ 1.0 mM O2 (bottom)] and Tyr (4 mM) in
Rea ction of Tyr w ith •NO. Incubation (4.5 h) of
50 mM phosphate buffer (pH 7.5) at 37 °C under
a 9/1
atmosphere of O2 and CO2: (bars) experimental results (errors
of less than (15%) and (b) simulated yields. Connecting lines
are only a visual aid.
•
s-NONOate (0.1 mM ≡ 0.2 mM NO total) with Tyr (4
mM) under an O2/CO2 atmosphere (9/1) did not yield
detectable quantities of NO2-Tyr or di-Tyr. However, at
higher concentrations of s-NONOate and after a lag
phase, both di-Tyr and NO2-Tyr were formed in steadily
increasing quantities. The yield of NO2-Tyr went through
a maximum of 1.6 µM at 5 mM s-NONOate and that of
di-Tyr through a maximum of 8.8 µM at 2.5 mM s-
NONOate. The formation of NO2-Tyr and di-Tyr required
O2, and the lag phase can be attributed to the slow
oxidation of •NO to NO2• (the one-electron oxidant of Tyr
in such systems). Addition of 100 µL of buffer which had
•
been purged with N2 and then saturated with NO to 0.5
mL of Tyr (4 mM, saturated with a 9/1 O2/CO2 atmo-
sphere) also gave NO2-Tyr (1.2 µM) and di-Tyr (0.8 µM).
•
Rea ction of Tyr w ith NO a n d O2•-. Incubation (4.5
h) of SIN-1 (0.5 mM) with Tyr (4 mM) under an O2/CO2
atmosphere (9/1) gave 23.9 µM di-Tyr and 1.2 µM NO2-
Tyr, while under 100% O2, the yield of di-Tyr fell to 3.5
µM and only traces of NO2-Tyr were produced. Incubation
•
(4.5 h) of s-NONOate (0.25 mM ≡ a total of 0.5 mM NO)
and SOTS-1 (1.25 mM ≡ a total of 0.5 mM O2•-) with
Tyr (4 mM) gave 26.5 µM di-Tyr and 0.56 µM NO2-Tyr
under an O2/CO2 atmosphere, while under 100% O2, the
yield of di-Tyr fell to 2.9 µM and only trace quantities of
NO2-Tyr could be detected. The yield of di-Tyr increased
with the initial Tyr concentration (0.4-4.0 mM, inset of
Figure 1), and therefore, all subsequent reactions were
performed using 4 mM Tyr (close to its solubility limit).
The buildup of di-Tyr follows first-order kinetics which
match the decay kinetics of s-NONOate and SOTS-1
(half-life of 80 min); see Figure 1. The effects of varying
F igu r e 3. Effect of O2•- concentration on di-Tyr and NO2-Tyr
(inset) yields after 4.5 h from s-NONOate [0.05 mM ≡ 0.1 mM
•
•NO (top), 0.2 mM ≡ 0.4 mM NO (middle), and 0.4 mM ≡ 0.8
•
mM NO (bottom)] and Tyr (4 mM) in 50 mM phosphate buffer
(pH 7.5) at 37 °C under a 9/1 atmosphere of O2 and CO2: (bars)
experimental results (errors of less than (15%) and (b) simu-
lated yields. Connecting lines are only a visual aid. White circles
in the middle panel represent simulated yields of di-Tyr, divided
by 4.8, using the reaction mechanism described by Goldstein et
al. (16).
•
•-
the ratio of the rates of formation of NO and O2 are
shown in Figures 2 and 3, respectively.
Rea ction of Tyr w ith P r efor m ed P er oxyn itr ite a t
24 °C. Peroxynitrite was synthesized according to a
literature method (25). This preformed peroxynitrite (0.5
mL, 11 mM in N2 saturated 25 mM NaOH) was syringe
pumped into Tyr [5 mL, 4 mM in a 9/1 O2/CO2 saturated
200 mM phosphate buffer (pH 7.5, increasing to 7.6
during the addition)] which was being rapidly vortexed
under a 9/1 O2/CO2 atmosphere. Yields of di-Tyr (0.1 mM)
and NO2-Tyr (0.1 mM) were independent of the per-

1290 Chem. Res. Toxicol., Vol. 13, No. 12, 2000
Hodges et al.
oxynitrite addition rate (equal volume from 13 s to 136
min) and represent individual yields for both products
of 10% based on peroxynitrite (in the absence of CO2, the
product yields did depend on the addition rate; unpub-
lished results). Preformed alkaline peroxynitrite (0.2 mL,
2 mM and O2-saturated) was also mixed with Tyr [0.2
mL, 4 mM in O2-saturated 100 mM phosphate buffer (pH
4) to which 6 mM Na2CO3 had been added 2 min prior to
use] in a rapid mixing (1 ms) apparatus. The resulting
solution (pH 7.5) contained di-Tyr (50 µM, 5% yield) and
NO2-Tyr (170 µM, 17% yield). Interestingly, the percent-
age yields in these very rapid mixing experiments were
independent of the peroxynitrite concentration (0.1-1.7
mM, after mixing).
respectively, and in the presence of CO2, the yield of NO2-
Tyr increased using preformed peroxynitrite (by ca. 70%)
but was unchanged using in situ-generated peroxynitrite3
(15). This led to the suggestion that peroxynitrite formed
in situ was “different” from “authentic” (i.e., preformed)
peroxynitrite. This is, of course, highly unlikely. However,
in its reaction with Tyr, peroxynitrite formed slowly in
situ will appear to be different from a bolus addition of
an equal quantity of preformed peroxynitrite (both in the
presence and in the absence of CO2) because of the
different rates of radical formation.4 Thus, with a bolus
addition, particularly in the presence of CO2, the initial
concentrations of Tyr-O• and NO2 will be very high,
•
•
making their combination very favorable [k(NO2 +Gly-
Tyr-O•) ) 3 × 109 M-1 s-1] (26) and resulting in a NO2-
Tyr yield of ∼17% (∼5% di-Tyr). However, the slow (4.5
h) in situ formation of peroxynitrite in the presence of
CO2 involves only very low radical concentrations, and
Discu ssion
•
Tyr is not oxidized to di-Tyr or NO2-Tyr by either NO
•-
•
•
or O2 (although it is oxidized slowly by NO and O2
most of the NO2 that formed will react with Tyr to
•
•
generate Tyr-O• [k(NO2 +Gly-Tyr) ) 3 × 105 M-1 s-1] (26).
following a lag phase during which some of the NO is
converted into NO2 ) (16). Tyr is oxidized by preformed
•
As a consequence, the yield of NO2-Tyr is drastically
reduced (to 0.1-0.3%). Under “intermediate” conditions,
the yield of NO2-Tyr falls between these extremes, e.g.,
10% for syringe pump addition of preformed peroxyni-
trite. A less important reason for product differences
between in situ-generated and preformed peroxynitrite
peroxynitrite and by peroxynitrite formed in situ from
•NO and O2•-, the principle products being di-Tyr and
NO2-Tyr. The yields of Tyr oxidation products from
peroxynitrite generated in situ from equal quantities of
•-
•NO (0.5 mM from s-NONOate) and O2 (0.5 mM from
•-
is that in the former case reactions of Tyr-O• with O2
SOTS-1) were in excellent agreement with the yields
obtained using 0.5 mM SIN-1 (which also yields 0.5 mM
•NO and 0.5 mM O2•-). Thus, in the presence of CO2, the
di-Tyr and NO2-Tyr yields were 5.3 and 0.1% (s-NON-
Oate/SOTS-1) and 4.8 and 0.2% (SIN-1), respectively,
while in the absence of CO2, the yields of di-Tyr fell to
0.6% (s-NONOate/SOTS-1) and 0.7% (SIN-1) with only
traces of NO2-Tyr being formed in either system. The
essentially equal yields of products from the s-NONOate/
SOTS-1 and SIN-1 experiments which were matched in
terms of total •NO and O2•- production demonstrate that
none of the organic radicals generated from SOTS-1 (17)
•
and NO become more probable and may contribute to
lower NO2-Tyr and di-Tyr yields (vide infra).
Tyr oxidation by in situ-generated peroxynitrite gave
maximum yields of di-Tyr and a local maximum of NO2-
•
•-
Tyr at a ratio of the rates of formation of NO to O2 of
1.05 (see Figures 2 and 3), the yield of di-Tyr decreasing
as the total quantities of NO and O2 increased; viz.,
•
•-
[•NO]total ) [O2
]
) 0.1, 0.2, 0.4, 0.5, 0.8, and 1 mM
•-
total
resulted in di-Tyr yields of 6.0, 6.0, 5.7, 5.3, 4.2, and 3.8%,
respectively, while NO2-Tyr yields remained at ca. 0.1%.
This decrease in di-Tyr yield is caused by further reac-
•
is trapped by the NO formed from the s-NONOate, a
•
tions of di-Tyr (see below). Even at equal total NO and
conclusion consistent with our studies on the absolute
•-
O2 concentrations of 0.2 mM, the further reactions of
•-
kinetics of each step in the process leading to O2 (18).
di-Tyr resulted in the di-Tyr yield increasing almost
linearly with the initial Tyr concentration (up to its
solubility limit, 4 mM; see the inset of Figure 1).
It also proves that SIN-1 provides equal quantities of
•NO and O2 under the reaction conditions that were
•-
employed.
Product profiles for the CO2-mediated reaction (Figures
2 and 3) are easy to understand qualitatively. Thus, for
When Tyr is oxidized by preformed peroxynitrite under
an O2/CO2 atmosphere (9/1), overall product yields are
greater and their distribution differs. Thus, syringe pump
addition of preformed peroxynitrite (in total 1 mM) to a
rapidly vortexed solution of Tyr (4 mM) gave di-Tyr and
NO2-Tyr, each in yields of 10% at times from 13 s to 136
min which indicates that the peroxynitrite anion reacts
at the alkaline peroxynitrite: CO2-buffer interface and
not in the bulk solution. In contrast, addition in 1 ms
gave di-Tyr and NO2-Tyr in yields of ca. 5 and 17%,
respectively, which match yields reported by others for
the rapid flow mixing of preformed peroxynitrite with Tyr
in the presence of O2 and CO2, e.g., 3% di-Tyr and 17%
NO2-Tyr (9). Under these fast mixing conditions, reaction
occurs in the bulk solution.
Pfeiffer and Mayer (15) first reported that the reaction
of Tyr with peroxynitrite formed in situ (both from SIN-1
and from s-NONOate and XO/hypoxanthine) gave sub-
stantially lower yields of NO2-Tyr (yields of di-Tyr were
not given) than in its reaction with preformed peroxyni-
trite.2 Under CO2-free conditions, 4 and ∼0.07% NO2-Tyr
were obtained using preformed and in situ peroxynitrite,
•-
a constant total yield of O2 (Figure 2), di-Tyr was not
•
detected in the absence of NO, but in the presence of
•NO, peroxynitrite is formed and Tyr is converted to di-
Tyr and NO2-Tyr which reach maxima at equal total
•NO and O2 concentrations. However, as the quantity
•-
•
of NO is further increased, it traps Tyr-O• causing the
•
2 The maximum yield of NO2-Tyr was obtained with NO alone (with
O2) and decreased upon addition of O2•- (15). The explanation for this
result is unclear.
3
However, in our experiments, CO2 increased the yields of NO2-
Tyr from in situ-generated peroxynitrite.
4
γ-Radiolysis provides slow, constant, and equal rates of formation
•
•-
of NO and O2 (16). In the presence of CO2, the yields of NO2-Tyr
were reduced from 0.6 to 0.2% as the radiation intensity was decreased.
Yields of di-Tyr were not reported.
5
•
The oxidation of dihydrorhodamine by separately generated NO
•-
and O2 has also been reported to be at a maximum when the •NO/
•-
O2 ratio equals 1.0, and it was suggested “that excess production of
one radical over the other may act as an endogenous modulator of
ONOO-...” (14). However, the postulated •NO-induced and O2•--induced
decompositions of ONOOH appear to be unknown reactions, and they
certainly are not required to explain the maximum in the yields of
di-Tyr and NO2-Tyr at a •NO/O2 ratio of 1.0, vide infra.
•-

Tyrosine Oxidation by Superoxide and Nitric Oxide
Chem. Res. Toxicol., Vol. 13, No. 12, 2000 1291
Ta ble 1. Gold stein et a l.’s P r op osed Mech a n ism for th e
sults, one rate constant was changed and new reactions
which consume di-Tyr were included (see Table 2).
Goldstein et al. only modeled NO2-Tyr yields but noted
that in γ-radiolysis reactions the di-Tyr yield was not
linear, suggesting subsequent reactions of di-Tyr to form
higher derivatives.
•-
Action of NO• a n d O2 on Tyr in th e P r esen ce of CO2
ka,b
O2•- + NO• f -OONO
-OONO + CO2 -O2COONO
4.3-6.7 × 109
f
3 × 104
-O2COONO f NO3- + CO2
>1 × 103 c
-O2COONO f NO2• + CO3
Tyr + CO3•- f Tyr-O•
Tyr + NO2• f Tyr-O•
>2 × 103 c
•-
•
The reaction between Tyr-O• and NO (last two reac-
4.5 × 107
3.2 × 105
tions in Table 1) is complex and not well understood. It
appears to be reversible (28, 29), and C- and O-nitroso-
tyrosine may interconvert with the latter hydrolyzing to
Tyr and nitrite (30). Reactions 6 and 7 were used for
modeling purposes.
Tyr-O• + O2•- f products
Tyr-O• + NO2• f NO2-Tyr
Tyr-O• + NO2• f other products
2Tyr-O• f di-Tyr
1.5 × 109
1.3 × 109
1.7 × 109
2.3 × 108 d
2O2•- (+2H+) f H2O2
2 × 105
NO• + NO2• h N2O3
N2O3 (+H2O) f 2NO2
1.1 × 109 (f), 8.4 × 104 (r)
2 × 103, 1.9 × 106[HCO3-],
8 × 105[phosphate]
4.5 × 109 (f), 1.1 (r)
1.3
-
Tyr-O• + NO• h TyrONO
TyrONO f products
(6)
(7)
NO2• + O2•- h -O2NO2
-
-O2NO2 f O2 + NO2
2NO2• h N2O4
4.5 × 108 (f), 6.9 × 103 (r)
N2O4 (+H2O) f NO3- + NO2
1 × 103
-
Goldstein et al. used a k6 of 1 × 109 M-1 s-1, a k-6 of
1000 s-1 (actual value suggested to be <1000 s-1), and a
k7 of 0.18 s-1 [actual value suggested to be >k-6/(5 × 103)].
However, with these values, the simulation overesti-
•
2NO• + O2 f 2NO2
2.9 × 106
Tyr-O• + NO• h Tyr-ONO
Tyr-ONO f products
1 × 109 (f), 1 × 103 (r)
0.18e
a
Rate constants in units of s-1, M-1 s-1, and M-2 s-1 for first-,
•-
mated experimental di-Tyr yields in the absence of O2
b
second-, and third-order reactions, respectively. These rate con-
and therefore gave a false product profile as the total
O2•-/•NO concentration ratio was raised from 0 to 1; for
an example, see Figure 3 (middle). A much better fit to
stants were taken from ref 16 and relate to room temperature.
We have not attempted to “adjust” them to 37 °C. c -O2COONO
-
•
d
yields of 67% (NO3 + CO2) and 33% (NO2 + CO3•-). In our
simulations, the yield of di-Tyr in this reaction was assumed to
be 35%. e In our simulations, this rate constant was increased to
experiment was achieved using a k7 of 2 s-1
.
The coupling of para-substituted phenoxyl radicals
yields a large number of products (31), e.g., 10 products
in the oxidation of p-cresol with ferric chloride (32), and
its oxidation by horseradish peroxidase and H2O2 gave 8
times as much of the ortho-para coupled product as of
the ortho-ortho coupled analogue of di-Tyr (33). Some
oxygen-carbon coupled products are also expected. Our
di-Tyr yields were best simulated assuming two Tyr-O•
radicals combine to give a 35% yield of di-Tyr and that
(unlike di-Tyr, vide infra) the other 65% of the products
do not further react.
Phenoxyl radicals very rapidly abstract phenolic
hydrogen atoms (34, 35) and the TyrO•/di-Tyr reaction
is expected to be fast with a rate constant in the range
106-108 M-1 s-1. Reaction 8 will be reversible with k8 >
k-8 because the O-H bond in di-Tyr is weaker than in
Tyr (DFT calculations predict a difference of 1.2 kcal
mol-1).6
2 s-1
.
Ta ble 2. Ad d ition a l Rea ction s Used in Ou r Sim u la tion of
•-
th e Action of NO• a n d O2 on Tyr in th e P r esen ce of CO2
ka
•-
SOTS-1 f 0.4O2
1.5 × 10-4
1.5 × 10-4
b
b
s-NONO f 2NO•
Tyr-O• + di-Tyr h Tyr + di-Tyr-O•
di/tri-Tyr + CO3•- f di/tri-Tyr-O•
di/tri-Tyr + NO2• f di/tri-Tyr-O•
1 × 107 (f),c 4 × 105 (r)c
1.1 × 109 c
8 × 106 c
di/tri-Tyr-O• + NO2• f NO2-(di/tri-Tyr) 1.3 × 109 d
di/tri-Tyr-O• + O2•- f products
di/tri-Tyr-O• + NO• h di/tri-Tyr-ONO
di/tri-Tyr-ONO f products
1.5 × 109 d
1 × 109 (f),d 1 × 103 (r)d
2d
di-Tyr-O• + Tyr-O• f tri-Tyr
2di/tri-Tyr-O• f poly-Tyr
2.3 × 108 d,e
2.3 × 108 d
2.3 × 108 d
tri-Tyr-O• + Tyr-O• f poly-Tyr
See footnote a in Table 1. See the text. c Arbitrary value; see
the text. Reaction assumed to have the same rate constant as
for Tyr-O•. e See footnote d in Table 1, di-Tyr being replaced with
tri-Tyr.
a
b
d
Tyr-O• + di-Tyr h Tyr + di-Tyr-O•
(8)
yields of di-Tyr and NO2-Tyr to decrease. The di-Tyr yield
does not decrease to zero but to the steady background
level produced using s-NONOate alone. The NO2-Tyr
yield decreases to a minimum and then increases (as
Since reaction 8 is fast, the equilibrium between Tyr-O•
and di-Tyr-O• is much more important for simulation
than the absolute rate constants. Consequently, k8 was
arbitrarily set at 107 M-1 s-1 and a value of 4 × 105 M-1
2•NO + O2 f 2NO2 ) to a second maximum, which can
•
•-
•
s-1 was used for k-8.7 The CO3 and NO2 radicals will
also oxidize di-Tyr faster than Tyr. Under the simulated
conditions, effectively all CO3•- reacts with either Tyr or
di-Tyr, and since their phenoxyl radicals are in equilib-
exceed the first maximum, and then decreases again (as
•NO + NO2• f N2O3). Similarly, at a constant total yield
of •NO (Figure 3), the di-Tyr and NO2-Tyr yields increase
on addition of O2•- to a maximum at equal total O2•- and
•NO concentrations, and then decrease eventually reach-
ing zero when the total concentration of O2•- far exceeds
•-
rium, the relative rates of the CO3 reaction are unim-
portant. (The same simulated results were obtained
•-
that of •NO, where all Tyr-O• radicals are trapped by O2
•-
assuming di-Tyr reacted with CO3 100 times faster or
(k ) 1.5 × 109 M-1 s-1) (27).
slower than Tyr!) The absolute rate of the reaction of di-
•
•
Tyr with NO2 will influence the NO2 steady-state
concentration and, therefore, the yield of NO2-Tyr. On
the basis of reaction 8, the CO3•- and NO2• reactions with
Sim u la tion s. The action of SOTS-1 and s-NONOate
on Tyr (4 mM) in the presence of O2 (1 mM) and a CO2/
HCO3- mixture (4 and 80 mM, respectively) for 4.5 h was
initially modeled using the reaction mechanism of Gold-
stein et al. (16) (Table 1) and the first two reactions in
Table 2 which reflect the exponential decay of SOTS-1
and s-NONOate. However, to fit our experimental re-
6
E. J ohnson and J . S. Wright, private communication.
To fit the experimental yields from Tyr (4 mM) at progressively
7
higher equal total •NO and O2•- concentrations, k-8 was optimized only
to the nearest 1 × 105 M-1 s-1
.

1292 Chem. Res. Toxicol., Vol. 13, No. 12, 2000
Hodges et al.
di-Tyr were arbitrarily assumed to be 25 times faster
than their reactions with Tyr. In radical-radical reac-
tions, di-Tyr-O• was assumed to react at the same rate
as Tyr-O•. The combination of Tyr-O• and di-Tyr-O• was
simulated to yield 35% tri-Tyr, the reactivity of which
was assumed to be the same as that of di-Tyr. Higher
combinations of “poly-Tyr” were formed in such low yield
that their further reactions were not considered.
Con clu sion s
We consider the overall quality of the fit of simulated
and experimental yields over such a wide range of total
•NO/O2•- concentration ratios to be a most gratifying proof
that the products of the reaction of Tyr with peroxynitrite
(preformed or generated in situ) in the presence of CO2
can be qualitatively and quantitatively accounted for by
known free radical chemistry. We therefore conclude that
the products formed in the Tyr reactions are formed solely
by free radical reactions.
The close agreement7 between the simulated and
experimental di-Tyr yields shown in Figures 2 and 3
(inset in Figure 1 discussed below) indicates that the
kinetic model fulfills its purpose; i.e., it accounts for the
experimental results. Experimental and modeled NO2-
Tyr yields show the same trend, and there is a reasonable
Ack n ow led gm en t. We most sincerely thank Profes-
sor Sara Goldstein and genuinely appreciate her kind,
thorough, and invaluable comments on an earlier version
of this paper. We are also indebted to two referees for
some very helpful comments and suggestions.
•-
•
agreement in absolute yields at low total O2 to NO
concentration ratios. As this ratio approaches 1, the
model overestimates the experimental yield and the data
could be better modeled using a lower yield of NO2-Tyr
from Tyr-O• + NO2• than the 45% proposed by Goldstein
et al. (16) and used in our simulations.8 However, it
seems more likely that the model underestimates the
competition for NO2• which is quite extraordinary; of its
33% yield from peroxynitrite (reaction 2b), 32.9% experi-
mentally (32.8% by simulation) “disappears” and only
0.1% (0.2%) ends up as NO2-Tyr!
Refer en ces
(1) Mahoney, L. R. (1970) Evidence for the formation of hydroxyl
radicals in the isomerisation of peroxynitrous acid to nitric acid
in aqueous solution. J . Am. Chem. Soc. 92, 5262-5263.
(2) Gatti, R. M., Alvarez, B., Vasquez-Vivar, J ., Radi, R., and Augusto,
O. (1998) Formation of spin adducts during the decomposition of
peroxynitrite. Arch. Biochem. Biophys. 349, 36-46.
(3) Coddington, J . W., Hurst, J . K., and Lymar, S. (1999) Hydroxyl
radical formation during peroxynitrous acid decomposition. J . Am.
Chem. Soc. 121, 2438-2443.
(4) Gerasimov, O. V., and Lymar, S. V. (1999) The yield of hydroxyl
radical from the decomposition of peroxynitrous acid. Inorg. Chem.
38, 4317-4321.
The experimental di-Tyr yields from Tyr (4 mM)
•-
decreased at progressively higher equal total O2 and
•NO concentrations. Even when [O2
]
) [•NO]total
)
•-
total
0.2 mM, the di-Tyr yield increased with the initial Tyr
concentration; see the inset in Figure 1. These facts might
have suggested that not all the oxidizing radicals were
trapped by Tyr. However, the simulations clearly show
that the di-Tyr yield is dependent on the equilibrium
between Tyr-O•/di-Tyr and Tyr/di-Tyr-O• (reaction 8).
Reducing the Tyr concentration or raising the di-Tyr
concentration shifts the equilibrium in favor of di-Tyr-
O• which lowers the final di-Tyr yield. The effect of the
initial Tyr concentration on di-Tyr yield could be simu-
lated perfectly by reducing k-8 from 4 × 105 to 2 × 105
M-1 s-1 (see the inset in Figure 1), but the yields of di-
(5) Hodges, G. R., and Ingold, K. U. (1999) Cage escape of geminate
radical pairs can produce peroxynitrate from peroxynitrite under
a variety of experimental conditions. J . Am. Chem. Soc. 121,
10695-10701.
(6) Lymar, S. V., and Hurst, J . K. (1995) Rapid reaction between
peroxynitrite ion and carbon dioxide: Implications for biological
activity. J . Am. Chem. Soc. 117, 8867-8868.
(7) Goldstein, S., and Czapski, G. (1997) The effect of bicarbonate
on oxidation by peroxynitrite: Implications for its biological
activity. Inorg. Chem. 36, 5113-5117.
(8) Lymar, S. V., and Hurst, J . K. (1998) CO2-catalyzed one-electron
oxidations by peroxynitrite: Properties of the reactive intermedi-
ate. Inorg. Chem. 37, 294-301.
(9) Lymar, S. V., J iang, Q., and Hurst, J . K. (1996) Mechanism of
carbon dioxide-catalyzed oxidation of tyrosine by peroxynitrite.
Biochemistry 35, 7855-7861.
(10) Beckman, J . S., Ye, Y. Z., Anderson, G., Chen, J ., Accavitti, M.
A., Tarpey, M. M., and White, C. R. (1994) Extensive nitration of
protein tyrosines in human atherosclerosis detected by immuno-
histochemistry. Biol. Chem. Hoppe-Seyler 375, 81-88.
(11) Leeuwenburgh, C., Hardy, M. M., Hazen, S. L., Wagner, P.,
Ohishi, S., Steinbrecher, U. P., and Heinecke, J . W. (1997)
Reactive nitrogen intermediates promote low-density lipoprotein
oxidation in human atherosclerotic intima. J . Biol. Chem. 272,
1433-1436.
(12) Santos, C. X. C., Bonini, M. G., and Augusto, O. (2000) Role of
the carbonate radical anion in tyrosine nitration and hydroxyla-
tion by peroxynitrite. Arch. Biochem. Biophys. 377, 146-152.
•-
Tyr from 4 mM Tyr and increasing equal total O2 and
•NO concentrations were then more poorly simulated.
Possibly undescribed reactions play a minor role in
product formation, or some rate constants are not quite
optimal.
It should be noted that all Tyr product-forming reac-
tions in the simulation involve free radicals. It is there-
fore unnecessary to involve nonradical Tyr nitrating
-
species such as O2NOCO2 (36). Furthermore, although
a report that sulfhydryls inhibit CO2-free peroxynitrite-
mediated Tyr nitration (37) has been used as evidence
•-
•-
(13) Singh, R. J ., Hogg, N., J oseph, J ., Konorev, E., and Kalyanaraman,
B. (1999) The peroxynitrite generator, SIN-1, becomes a nitric
oxide donor in the presence of electron acceptors. Arch. Biochem.
Biophys. 361, 331-339.
(14) Miles, A. M., Bohle, D. S., Glassbrenner, P. A., Hansert, B., Wink,
D. A., and Grisham, M. B. (1996) Modulation of superoxide-
dependent oxidation and hydroxylation reactions by nitric oxide.
J . Biol. Chem. 271, 40-47.
against nitration via CO3 [because CO3 reacts more
rapidly with Tyr (Table 1) than with cysteine, k ) 3.2 ×
105 M-1 s-1] (39), this report appears to be in error. In
our hands, cysteine protected Tyr from nitration by
preformed peroxynitrite and CO2 and the level of this
protection was 3 times greater than that calculated
assuming that the cysteine acted solely as a competitive
scavenger for CO3•-. This is due to cysteine’s facile
reduction of phenoxyl radicals (40) leading to a diminu-
tion of all Tyr-O•-derived products.
(15) Pfeiffer, S., and Mayer, B. (1998) Lack of tyrosine nitration by
peroxynitrite generated at physiological pH. J . Biol. Chem. 273,
27280-27285.
(16) Goldstein, S., Czapski, G., Lind, J ., and Mere´nyi, G. (2000)
•
•-
Tyrosine nitration by simultaneous generation of NO and O2
under physiological conditions. HOW THE RADICALS DO THE
8
J OB. J . Biol. Chem. 275, 3031-3036.
The true yield may be slightly higher than 45% since preformed
(17) Ingold, K. U., Paul, T., Young, M. J ., and Doiron, L. (1997)
Invention of the first azo compound to serve as a superoxide
peroxynitrite yields 17% NO2-Tyr, more than half the 33% theoretical
maximum yield (reactions 2b and 3a).

Tyrosine Oxidation by Superoxide and Nitric Oxide
Chem. Res. Toxicol., Vol. 13, No. 12, 2000 1293
thermal source under physiological conditions: Concept, synthe-
sis, and chemical properties. J . Am. Chem. Soc. 119, 12364-
12365.
(29) Gunther, M. R., Hsi, L. C., Curtis, J . F., Gierse, J . K., Marnett,
L. J ., Eling, T. E., and Mason, R. P. (1997) Nitric oxide trapping
of the tyrosyl radical of prostaglandin H synthase-2 leads to
tyrosine iminoxyl radical and nitrotyrosine formation. J . Biol.
Chem. 272, 17086-17090.
(30) Beckman, J . S. (1996) Oxidative damage and tyrosine nitration
from peroxynitrite. Chem. Res. Toxicol. 9, 836-844.
(31) Mihailovic, M. L. J ., and Cekovic, Z. (1971) in The Chemistry of
Functional Groups: the Chemistry of the Hydroxyl Group, Part
1, Chapter 10 (Patai, S., Ed.) Wiley, London.
(32) Chen, C.-L., Connors, W. J ., and Shinker, W. M. (1969) New
carbonyl compounds from dehydrogenation of p-cresol. J . Org.
Chem. 34, 2966-2971.
(33) Pietika¨inen, P., and Adlercreutz, P. (1990) Influence of the
reaction medium on the product distribution of peroxidase-
catalysed oxidation of p-cresol. Appl. Microbiol. Biotechnol. 33,
455-458.
(34) Foti, M., Ingold, K. U., and Lusztyk, J . (1994) The surprisingly
high reactivity of phenoxyl radicals. J . Am. Chem. Soc. 116, 9440-
9447.
(35) Lucarini, M., Pedrielli, P., Pedulli, G. F., Cabiddu, S., and
Fattuoni, C. (1996) Bond dissociation energies of O-H bonds in
substituted phenols from equilibrium studies. J . Org. Chem. 61,
9259-9263.
(36) Uppu, R. M., Squadrito, G. L., and Pryor, W. A. (1996) Accelera-
tion of peroxynitrite oxidations by carbon dioxide. Arch. Biochem.
Biophys. 327, 335-343.
(37) Gow, A., Duran, D., Thom, S. R., and Ischiropoulos, H. (1996)
Carbon dioxide enhancement of peroxynitrite-mediated protein
tyrosine nitration. Arch. Biochem. Biophys. 333, 42-48.
(38) Koppenol, W. H. (1998) Peroxynitrite uncloaked? Chem. Res.
Toxicol. 11, 716-717.
(18) Konya, K. G., Paul, T., Lin, S., Lusztyk, J ., and Ingold, K. U.
(2000) Laser flash photolysis studies on the first superoxide
thermal source. First direct measurements of the rates of solvent-
assisted 1,2-hydrogen atom shifts and a proposed new mechanism
for this unusual rearrangement. J . Am. Chem. Soc. 122, 7518-
7527.
(19) Amado, R., Aeschbach, R., and Neukom, H. (1984) Dityrosine: In
vitro production and characterization. Methods Enzymol. 107,
377-388.
(20) Crovetto, R. (1991) Evaluation of solubility data of the system
CO2-H2O from 273 K to the critical point of water. J . Phys. Chem.
Ref. Data 20, 575-589.
(21) www.almaden.ibm.com.
(22) Hrabie, J . A., Klose, J . R., Wink, D. A., and Keefer, L. K. (1993)
New nitric oxide-releasing zwitterions derived from polyamines.
J . Org. Chem. 58, 1472-1476.
(23) Maragos, C. M., Morley, D., Wink, D. A., Dunams, T. M.,
Saavedra, J . E., Hoffman, A., Bove, A. A., Isaac, L., Hrabie, J .
A., and Keefer, L. K. (1991) Complexes of •NO with nucleophiles
as agents for the controlled biological release of nitric oxide.
Vasorelaxant effects. J . Med. Chem. 34, 3242-3247.
(24) Ramamarthi, A., and Lewis, R. S. (1997) Measurement and
modeling of nitric oxide release rates for nitric oxide donors.
Chem. Res. Toxicol, 10, 408-413.
(25) Gleu, K., and Roell, E. (1929) Die Einwirkung von Ozon auf
Alkaliazid. Z. Anorg. Allg. Chem. 179, 233-266.
(26) Pru¨tz, W. A., Monig, H., Butler, J ., and Land, E. J . (1985)
Reactions of nitrogen dioxide in aqueous model systems: Oxida-
tion of tyrosine units in peptides and proteins. Arch. Biochem.
Biophys. 243, 125-134.
(27) J in, F., Leitich, J ., and von Sonntag, C. (1993) The superoxide
radical reacts with tyrosine-derived phenoxyl radicals by addition
rather than by electron transfer. J . Chem. Soc., Perkin Trans. 2,
1583-1588.
(28) J anzen, E. G., Wilcox, A. L., and Manoharan, V. (1993) Reactions
of nitric oxide with phenolic antioxidants and phenoxyl radicals.
J . Org. Chem. 58, 3597-3599.
(39) Chen, S., and Hoffman, M. Z. (1973) Rate constants for the
reaction of the carbonate radical with compounds of biochemical
interest in neutral aqueous solution. Radiat. Res. 56, 40-47.
(40) Pru¨tz, W. A., Butler, J ., and Land, E. J . (1983) Phenol coupling
initiated by one-electron oxidation of tyrosine units in peptides
and histone. Int. J . Radiat. Biol. 44, 183-196.
TX0001272