Reaction of peroxynitrite with melatonin: A mechanistic study

Article abstract of DOI:10.1021/tx980243t

The pH profile of the peroxynitrite/melatonin reaction suggests that both peroxynitrous acid (ONOOH) and its anion (ONOO^-) are reactive toward melatonin, but at physiological pH most of the reaction with melatonin involves ONOOH and the activated form of peroxynitrous acid (ONOOH*). The formation of hydroxylated products (mainly 6-hydroxymelatonin) suggests that melatonin also reacts with ONOOH*. The overall peroxynitrite/melatonin reaction is first-order in melatonin and first-order in peroxynitrite, but the hydroxylation of melatonin is presumed to be zero-order in melatonin. Melatonin is metabolized in the liver, mainly to 6-hydroxymelatonin, so we do not think this metabolite is a useful biomarker for melatonin's antioxidant activity; however, 6-hydroxymelatonin is a better chain-breaking antioxidant than melatonin and may contribute to the beneficial effects of melatonin in vivo. As is now well-known, CO₂ modulates the reactions of peroxynitrite. The reaction of peroxynitrite with melatonin in the absence of added bicarbonate produces mainly 6-hydroxymelatonin and 1,2,3,3a,8,8a-hexahydro- 1-acetyl-5-methoxy-8a-hydroxypyrrolo[2,3-b]indole, with some isomeric 1,2,3,3a,8,8a-hexahydro-1-acetyl-5-methoxy-3a-hydroxypyrrolo[2,3-b]indole. In the presence of added bicarbonate, product yields decrease and 6- hydroxymelatonin is not formed. These facts suggest that melatonin scavenges reactive species (such as CO₃(°-) and °NO₂) that are produced from the peroxynitrite/CO₂ reaction. The spectrum of the melatoninyl radical cation is observed both in the absence and in the presence of added bicarbonate, suggesting that the melatoninyl radical cation is the initial product and the hydroxypyrrolo[2,3-b]indole products are derived from it. Unlike tyrosine, where both nitrated and hydroxylated products can be isolated, nitromelatonin is not found in the final products from the melatonin/peroxynitrite reaction in either the absence or presence of added bicarbonate. However, we suggest that 2-hydroxy-3-nitro- and/or 2-hydroxy-3-peroxynitro-2,3-dihydromelatonin are formed as intermediates and subsequently decompose to give 1,2,3,3a,8,8a- hexahydro-1-acetyl-5-methoxy-8a-hydroxypyrrolo[2,3-b]indole. Since peroxynitrite/CO₂ governs the reactions of peroxynitrite in vivo, we suggest that the hydroxypyrrolo[2,3-b]indole products are the main products from the oxidation of melatonin by peroxynitrite-derived species in vivo, and that these products may serve as indexes for melatonin's antioxidant activity.

Full text of DOI:10.1021/tx980243t

526
Chem. Res. Toxicol. 1999, 12, 526-534
Rea ction of P er oxyn itr ite w ith Mela ton in : A Mech a n istic
Stu d y
Houwen Zhang, Giuseppe L. Squadrito,* Rao Uppu, and William A. Pryor*
Biodynamics Institute, Louisiana State University, Baton Rouge, Louisiana 70803-1800
Received November 4, 1998
The pH profile of the peroxynitrite/melatonin reaction suggests that both peroxynitrous acid
(ONOOH) and its anion (ONOO-) are reactive toward melatonin, but at physiological pH most
of the reaction with melatonin involves ONOOH and the activated form of peroxynitrous acid
(ONOOH*). The formation of hydroxylated products (mainly 6-hydroxymelatonin) suggests
that melatonin also reacts with ONOOH*. The overall peroxynitrite/melatonin reaction is first-
order in melatonin and first-order in peroxynitrite, but the hydroxylation of melatonin is
presumed to be zero-order in melatonin. Melatonin is metabolized in the liver, mainly to
6-hydroxymelatonin, so we do not think this metabolite is a useful biomarker for melatonin’s
antioxidant activity; however, 6-hydroxymelatonin is a better chain-breaking antioxidant than
melatonin and may contribute to the beneficial effects of melatonin in vivo. As is now well-
known, CO2 modulates the reactions of peroxynitrite. The reaction of peroxynitrite with
melatonin in the absence of added bicarbonate produces mainly 6-hydroxymelatonin and
1,2,3,3a,8,8a-hexahydro-1-acetyl-5-methoxy-8a-hydroxypyrrolo[2,3-b]indole, with some isomeric
1,2,3,3a,8,8a-hexahydro-1-acetyl-5-methoxy-3a-hydroxypyrrolo[2,3-b]indole. In the presence of
added bicarbonate, product yields decrease and 6-hydroxymelatonin is not formed. These facts
suggest that melatonin scavenges reactive species (such as CO3•- and NO2) that are produced
•
from the peroxynitrite/CO2 reaction. The spectrum of the melatoninyl radical cation is observed
both in the absence and in the presence of added bicarbonate, suggesting that the melatoninyl
radical cation is the initial product and the hydroxypyrrolo[2,3-b]indole products are derived
from it. Unlike tyrosine, where both nitrated and hydroxylated products can be isolated,
nitromelatonin is not found in the final products from the melatonin/peroxynitrite reaction in
either the absence or presence of added bicarbonate. However, we suggest that 2-hydroxy-3-
nitro- and/or 2-hydroxy-3-peroxynitro-2,3-dihydromelatonin are formed as intermediates and
subsequently decompose to give 1,2,3,3a,8,8a-hexahydro-1-acetyl-5-methoxy-8a-hydroxypyrrolo-
[2,3-b]indole. Since peroxynitrite/CO2 governs the reactions of peroxynitrite in vivo, we suggest
that the hydroxypyrrolo[2,3-b]indole products are the main products from the oxidation of
melatonin by peroxynitrite-derived species in vivo, and that these products may serve as indexes
for melatonin’s antioxidant activity.
damage caused by radicals in vivo (15-20). Melatonin
also has been reported to cause a dose-dependent inhibi-
tion of the oxidation of dihydrorhodamine 123 by peroxy-
nitrite (21), suggesting that it is a scavenger of peroxy-
nitrite. Recently, we reported the observation of the
spectrum of the melatoninyl radical cation (MLT•+)
during the peroxynitrite/MLT reaction, providing direct
evidence of an initial one-electron transfer from MLT to
peroxynitrite (22). The reaction of MLT with peroxyni-
trite does not yield nitromelatonin in either the absence
or presence of CO2 (22), despite the fact that CO2 is
known to catalyze nitrations by peroxynitrite (9-11, 23,
24).
In tr od u ction
Peroxynitrite,1 which is formed in vivo via the coupling
of NO and O2•-, is able to oxidize a variety of biomol-
ecules, including thiols, lipids, ascorbate, DNA, sulfides,
and aromatic species (1-7). These reactions can involve
the transfer of either one or two electrons, and can be
either zero- or first-order in substrate (8). However, in
aerobic biological systems, the reactions of peroxynitrite
are generally modulated by CO2 due to the fast reaction
of peroxynitrite with CO2 and the abundance of CO2 (9-
12), and few biomolecules can compete with CO2 for
peroxynitrite (13, 14).
Melatonin (MLT2), the main secretory product of the
pineal gland, has been proposed to protect against
Herein we report kinetic studies of the reaction of MLT
with peroxynitrite in the absence and presence of carbon
dioxide. The possible mechanisms and the biological
relevance of these reactions also are discussed in terms
of the reaction kinetics and product analyses.
* To whom correspondence should be addressed: G.L.S. at
gsquadr@LSU.edu or W.A.P. at wpryor@LSU.edu or to either author
at Biodynamics Institute, 711 Choppin Hall, Louisiana State Univer-
sity, Baton Rouge, LA 70803-1800. Phone: (225) 388-2063. Fax: (225)
388-4936.
The sum of ONOOH and ONOO- is termed peroxynitrite, and the
1
Exp er im en ta l P r oced u r es
2-
sum of CO2, H2CO3, HCO3-, and CO3 is termed bicarbonate unless
otherwise indicated.
Abbreviations: MLT, melatonin; 3-HO-MLT, 3-hydroxymelatonin;
6-HO-MLT, 6-hydroxymelatonin.
2
Ma ter ia ls. MLT (97%) and C18 TLC plates (10 cm × 10 cm)
were purchased from Aldrich (Milwaukee, WI), and sodium
10.1021/tx980243t CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/26/1999

Reaction of Melatonin with Peroxynitrite
Chem. Res. Toxicol., Vol. 12, No. 6, 1999 527
azide and melatonin were from Sigma (St. Louis, MO). Deionized
water (>15 MΩ/cm) was used in the preparation of buffers and
reagents.
In str u m en ta tion . The stopped-flow instrument that was
used for the measurement of reaction rates was from On-Line
Instrument Systems (Bogart, GA), and it was equipped with a
UV/vis rapid scan spectrophotometer and a 2 cm cell. The
Sander 200 ozonizer was from Erwin Sander (Uetze-Eltze,
Germany). The HP8451A diode array UV/vis spectrophotometer
with a 1 cm cell was from Hewlett-Packard (Willmington, DE).
The GC/MS model HP5890/5973 apparatus was from Hewlett-
Packard. The column that was used was a SGE BPX5 column
(25 m); the temperatures of the injection port and detector were
250 and 280 °C, respectively. The column was initially set at
200 °C (5 min), and then the temperature was increased to 230
°C (10 min) and finally to 250 °C (5 min) at speeds of 10 °C/
min.
F igu r e 1. Typical kinetic trace for the reaction of peroxynitrite
with melatonin (MLT). The reaction was carried out using a
stopped-flow apparatus at 25.0 ( 0.1 °C by mixing equal
volumes of peroxynitrite solution with a buffer solution contain-
ing MLT. After the mixing but before the reaction, the mixture
contained 0.125 mM peroxynitrite and 4.9 mM MLT. The pH
measured at the outlet was 7.0. The curve was fitted to the first-
order formation equation to calculate kobs values (see the inset
of Figure 2). Data are shown at 400 nm. (Inset) Data for the
same reaction as observed with longer times.
Syn th esis of P er oxyn itr ite. Peroxynitrite was synthesized
by the ozonation of an aqueous solution of sodium azide (25,
26). Briefly, 0.25 g of sodium azide was dissolved in 25 mL of
water preadjusted to pH 11 using 1 N sodium hydroxide. After
cooling to <5 °C, the solution was ozonized by feeding oxygen
to the ozonator at a flow rate of 100 mL/min. The concentration
of peroxynitrite was monitored by measuring the A302nm of
samples diluted 100 times using pH 11 water. It requires about
60 min for the absorbance to reach the maximum (ca. 1.1), and
the ozonation was stopped after the absorbance decreased to
ca. 0.9 (25, 26). The resulting solution was purged for 10 min
with nitrogen and then stored at -12 °C until it was used.
Sch em e 1
Kin etics. Peroxynitrite solutions (pH 11) were mixed on the
stopped-flow spectrophotometer with equal volumes of 50 mM
phosphate buffer containing varying amounts of MLT (ionic
strength adjusted to 0.3 M using NaCl). The pH values were
measured both before and after mixing of the reactants. A small
pH jump (ca. 0.15) was usually observed on mixing, and the
final pH was reported unless otherwise stated. The temperature
was maintained to within (0.1 °C.
For pseudo-first-order reactions, the molar ratio of MLT to
peroxynitrite was greater than 4:1. The decay of peroxynitrite
in the absence of MLT is usually analyzed at 302 nm, but since
MLT absorbs at this wavelength, the peroxynitrite/MLT reaction
was analyzed at 400 nm where the interference from both
peroxynitrite and MLT is minimal. Data are the average of at
least three runs.
P r ep a r a tion of Mela ton in Solu tion s. Melatonin has a
relatively low solubility. To prepare 25 mL of a 10 mM aqueous
MLT solution, MLT (58 mg) was dissolved in 1.5 mL of 5 N HCl
and the resulting solution diluted with 0.05 M phosphate buffer
and then adjusted to the desired pH using 50% NaOH. The
samples were monitored by 1H NMR, and it was found that MLT
is stable during this procedure. However, to minimize autoxi-
dation, only freshly prepared solutions of MLT were used in
these investigations.
Resu lts
Kin etics. The stopped-flow reaction kinetics analyzed
at 400 nm suggest that the peroxynitrite/MLT reaction
occurs in two stages. In the first stage, the absorbance
increases rapidly to a maximum, revealing the formation
of intermediates derived from melatonin (Figure 1). In
the second stage, the absorbance decreases slowly, show-
ing the decay of these intermediates (inset of Figure 1).
As we determined previously (22), the melatoninyl radical
cation is the main intermediate; however, the reaction
of melatonin with peroxynitrite is complex, and we cannot
rule out small contributions of secondary intermediates,
such as unstable melatonin adducts (Scheme 1), to the
absorbance.
P r od u ct An a lysis. Aliquots of peroxynitrite (1 mL each)
were flow-mixed with an equal volume of 0.05 M phosphate
buffer (pH 6.85-7.20) containing various amounts of MLT and
sodium bicarbonate. The reaction mixtures were incubated for
20 min and then dried on a rotary evaporator at 60 °C. The
residue was extracted with methanol, and the methanolic
extracts were analyzed by GC/MS and/or 1H NMR.
Rea ction s a t P r ep a r a tive Sca le. Stock solutions of peroxy-
nitrite (ca. 50 mM) were added dropwise to 2 mL of 0.05 M
phosphate buffer (pH 7.2) containing 4 mM MLT and saturated
sodium bicarbonate until an overall molar ratio of peroxynitrite
to MLT of 5:1 was reached. Throughout the course of addition
of peroxynitrite, the contents were constantly stirred using a
vortex mixer. The reaction mixture were dried on a rotary
evaporator at 60 °C; the residue was extracted into methanol,
and the products were separated by reversed phase TLC (C18
silica developed with water/methanol in a ratio of 80:20).
The spontaneous decomposition of peroxynitrite cannot
be followed in the presence of MLT due to the overlap of
absorption of MLT and peroxynitrite (see Experimental
Procedures), and it is only possible to measure the rate
of formation of products at 400 nm. Since the reactions
were performed under pseudo-first-order conditions, the
observed rate constant, kobs, was calculated by fitting the
rising phase of the kinetic traces (the first stage) to an
equation that is first-order in peroxynitrite. At a given

528 Chem. Res. Toxicol., Vol. 12, No. 6, 1999
Zhang et al.
Ta ble 1. Ra te a n d Equ ilibr iu m Con sta n ts
constant
value
comment
k1
1.34 s-1
rate constant for the spontaneous
decomposition of peroxynitrite (31)
k2
159 ( 5 M-1 s-1 rate constant for the ONOOH/MLT
reaction
5 ( 5 M-1 s-1 rate constant for the ONOO-/MLT
reaction
k2′
pKPN
6.5
acidity constant of ONOOH based
on the kapp for the
peroxynitrite/MLT reaction
F igu r e 2. Influence of pH on the peroxynitrite/melatonin
reaction. The plotted values of kapp are calculated from the slopes
of the lines shown in the inset. Nonlinear fitting of kapp to eq 6
gives the value of the second-order rate constants and the acidity
constant of peroxynitrite (k2 ) 159 ( 5 M-1 s-1, k2′ ) 5 ( 5 M-1
s-1, and pKPN ) 6.5). (Inset) The reactions were performed on
the stopped-flow instrument (see the legend of Figure 1). After
the mixing but before the reaction, the mixture contained 0.25
mM peroxynitrite and 0, 1.25, 2.0, 3.5, or 5.0 mM melatonin.
The pH values as measured at the outlet were 5.10, 5.54, 5.94,
6.25, 6.73, 7.35, and 8.00. For other reaction conditions, see the
legend of Figure 1.
pH, kobs increases linearly with the concentration of MLT,
indicating that the reaction also is first-order in MLT
(inset of Figure 2). The reactions were performed using
relatively high concentrations of MLT to minimize the
influence of adventitious CO2. [CO2 modulates the reac-
tions of peroxynitrite (13, 14, 23, 24).] Equation 1 depicts
the acid dissociation of peroxynitrous acid (pKPN ) 6.8),
and eq 2 depicts the overall spontaneous decomposition
of peroxynitrite.
F igu r e 3. Influence of pH on product formation. The circles
represent the values of ∆Amax (the difference between the initial
absorbance and the absorbance at the maximum in a plot like
that shown in Figure 1) which are shown for the reaction of
0.25 mM peroxynitrite with 2.0 mM MLT performed on the
stopped-flow instrument. The dotted line represents the loss of
MLT (∆[MLT]) calculated using eq 8, with a pKPN of 6.5, a k1 of
1.34 s-1, a k2 of 159 M-1 s-1, and a k2′ of 5 M-1 s-1
.
The rate and equilibrium constants discussed here and
in later sections are summarized in Table 1.
KPN
In flu en ce of p H on P r od u ct Yield s. Surprisingly,
the value of ∆Amax (∆Amax ) Amax - At)0; see Figures 1
and 3) increases with pH, as shown in Figure 3 (circles)
despite the fact kapp decreases with pH. This reflects the
complexity of the peroxynitrite/MLT reaction and sug-
gests the peroxynitrite anion also reacts at basic pH. A
mechanism that can explain these data is given below.
Activa tion P a r a m eter s. Activation parameters were
obtained using the Eyring equation (eq 7) applied to kapp
at pH 6.0, and the study was conducted by stopped flow
by varying the temperature from 7 to 45 °C (Figure 4).
ONOOH y z H+ + ONOO-
(1)
(2)
k1
9
ONOOH
8 H+ + NO3
-
Thus, the observed rate constant for the formation of
products described in eqs 3 and 4 can be written as shown
in eq 5, where the constant C reflects the zero-order
component due to the reaction of ONOOH* with MLT
(see below).
k2
ONOOH + MLT
ONOO- + MLT
9
8 products
(3)
(4)
[H+] + KPN
h
kBT
∆Sq ∆Hq
k2′8 products
ln kapp
)
-
(7)
[H+]
(
)
9
R
RT
k2[H+] + k2′K
[H+] + KPN
PN[MLT] + C
(5)
At pH 6, in eq 6, the term k2′KPN is small relative to k2-
[H+], and can be neglected. Thus, in eq 7, kapp([H+] + KPN)/
[H+] is approximately k2, and the activation parameters
obtained at pH 6 approximate those for the reaction of
ONOOH with MLT. However, it must be kept in mind
that kapp is a composite that includes several reactions,
and the activation parameters obtained from kapp will
have contributions from all these processes. In eq 7, the
value of KPN is 6.5 at 25 °C and varies with temperature
with a ∆H° of 2.7 ( 1.0 kcal/mol (27) [other investigators
have confirmed that the enthalpy of dissociation of
ONOOH is small and positive (28)] and h and kB are the
Planck and Boltzmann constants, respectively. The val-
ues for the activation enthalpy, ∆Hq, and entropy, ∆Sq,
are calculated to be 8.1 ( 0.6 kcal/mol and -20 ( 2 eu,
respectively.
kobs
)
For the sake of convenience, we define an apparent
second-order rate constant, kapp, in eq 6.
k2[H+] + k2′KPN
kapp
)
(6)
[H+] + KPN
The value of kapp is pH-dependent and was calculated
from the slopes of the plots such as those shown in the
inset of Figure 2. In eqs 3 and 4, it is assumed that both
ONOOH and ONOO- react with MLT; see below. The
second-order rate constant, k2 (for the ONOOH/MLT
reaction), k2′ (for the ONOO-/MLT reaction), and the
acidity constant of peroxynitrite, pKPN, are determined
to be 159 M-1 s-1, 5 M-1 s-1 and 6.5, respectively (Figure
2), by nonlinear fitting of kapp to eq 6 at different pHs.
P er oxyn itr ite/Mela ton in Rea ction in th e P r es-
en ce of Ad d ed Bica r bon a te. Upon mixing of peroxy-
nitrite with MLT and bicarbonate in the stopped-flow

Reaction of Melatonin with Peroxynitrite
Chem. Res. Toxicol., Vol. 12, No. 6, 1999 529
F igu r e 4. Eyring plot for the reaction of peroxynitrite with
melatonin. The data points are the slopes of the plots shown in
the inset vs 1/T. The temperatures were 7.0, 16.5, 24.1, 37.1,
and 45.2 ( 0.1 °C. The activation parameters are as follows:
∆Hq ) 8.1 ( 0.6 kcal/mol and ∆Sq ) -22 ( 2 eu. (Inset) The
reactions were conducted on the stopped-flow instrument and
were monitored at 400 nm. After the mixing but before the
reaction, the mixture contained 0.25 mM peroxynitrite and 0,
1.22, 2.0, 3.5, or 5.0 mM MLT. The pH as measured at the outlet
was 6.02 ( 0.06. For the other reaction conditions, see Figure
1.
F igu r e 6. Effect of CO2 on kobs. The values of kobs were
calculated by fitting kinetic traces similar to the one shown in
Figure 5 to a first-order equation. The reactions were performed
on the stopped-flow instrument at 25.0 ( 0.1 °C (data shown at
400 nm). After mixing but before the reaction, the mixture
contained 0.2 mM peroxynitrite, 1.0 mM MLT, and 0, 0.1, 0.2,
0.4, 1.4, 1.6, or 2.0 mM bicarbonate. The pH as measured at
the outlet was 7.01 ( 0.03. For the other reaction conditions,
see Figure 1.
F igu r e 7. Effect of CO2 on ∆Amax. The values of ∆Amax (see
Figure 3 for definition) are plotted vs the concentration of
bicarbonate for the reactions shown in Figure 6.
F igu r e 5. Typical kinetic trace for the reaction of peroxynitrite
with melatonin in the presence of added bicarbonate. The
reactions were performed on the stopped-flow instrument and
were observed at 400 nm and 25.0 ( 0.1 °C. After mixing but
before the reaction, the mixture contained 0.2 mM peroxynitrite,
1.0 mM melatonin, and 2.0 mM sodium bicarbonate. (Inset) The
same reaction as observed at a longer period of time. For the
other reaction conditions, see Figure 1.
instrument, the absorbance at 400 nm initially increases
rapidly to reach a maximum, and then decreases at a
much slower rate (inset of Figure 5). The rising phase
of the absorption trace is fitted to a first-order equation
to calculate the value of kobs
.
At a given concentration of MLT, the value of kobs
increases nonlinearly with the concentration of bicarbon-
ate (Figure 6). However, ∆Amax, which reflects the amount
of the intermediate(s) formed in the reaction, decreases
with added bicarbonate (Figure 7). At a given concentra-
tion of bicarbonate, increasing the concentration of MLT
causes the value of kobs to decrease (Figure 8) but ∆Amax
to increase (Figure 9). These observations can be ex-
plained in terms of the competition between MLT and
CO2 for peroxynitrite, and the ability of MLT to capture
intermediates and interfere with the catalytic cycle of
CO2. We previously observed that tyrosine also decreases
the fraction of CO2 that undergoes a true catalytic cycle
(29).
F igu r e 8. Effect of melatonin on the kobs of the peroxynitrite/
CO2 reaction. The values of kobs were calculated by fitting kinetic
traces similar to the one shown in Figure 5 to a first-order
equation. The reactions were performed on the stopped-flow
instrument at 25.0 ( 0.1 °C (data shown at 400 nm). After
mixing but before the reaction, the mixture contained 0.5 mM
peroxynitrite, 5.0 mM bicarbonate, and 0, 0.25, 0.5, 0.75 1.0, or
1.25 mM MLT. The pH as measured at the outlet was 7.11 (
0.02. For the other reaction conditions, see Figure 1.
the absence of MLT from pH 5.0 to 8.0. Also, the value
of pKPN (6.5; see Table 1) obtained from Figure 2 is in
reasonable agreement with the value of 6.75 previously
determined using other methods (8, 30). These facts
suggest that ONOOH is the reactive species toward MLT
(Figure 2 and eq 3).
Strikingly, the product yield (Figure 3) increases with
pH. This can be explained by suggesting that the peroxy-
nitrite anion (ONOO-) also reacts with melatonin to yield
oxidation products at basic pH (eq 4). The rate constant
for the ONOO-/MLT reaction (k2′ ) 5 M-1 s-1, eq 4) is
smaller than that for the ONOOH/MLT reaction (k2 )
Discu ssion
Rea ctive Sp ecies. In the absence of added bicarbon-
ate, the pH profile of kapp for product appearance in the
peroxynitrite/MLT reaction (Figure 2) closely resembles
the pH profile for the decomposition of peroxynitrite in

530 Chem. Res. Toxicol., Vol. 12, No. 6, 1999
Zhang et al.
Ta ble 2. Activa tion P a r a m eter s for Som e P er oxyn itr ite
Rea ction s In volvin g ONOOH
reaction
∆Hq (kcal/mol) ∆Sq (eu)
ref
ONOOH/MLT
8.1 ( 0.6
9.3 ( 0.5
9.1 ( 0.3
4.6 ( 0.4
7.7 ( 0.4
-20 ( 2 this worka
-16 ( 2 46
ONOOH/ascorbate
ONOOH/tryptophan
ONOOH/I-
-19 ( 1 31
-23 ( 1 33
-22 ( 2 30
ONOOH/methionine
a
The activation parameters for melatonin were calculated using
a ∆H° of 2.7 ( 1.0 kcal/mol for the acid dissociation of ONOOH,
as we reported previously (27). All the other values do not take
into account this enthalpy of dissociation, but the correction is
small. For example, the uncorrected values for melatonin are 7.4
( 0.6 kcal/mol for ∆Hq and 23 ( 2 eu for ∆Sq.
F igu r e 9. Effect of melatonin on ∆Amax. Values of ∆Amax (see
Figure 3 for definition) are plotted vs the concentration of MLT
for the reactions shown in Figure 8.
groups (MLT bears a methoxy substituent on C-5, while
tryptophan bears a hydrogen atom on the same position)
is substantiated by a Hammett study of 5-substituted
indole-3-carboxylic acids, where σm was used and electron-
donating groups had only small effects on the pK of the
carboxylic acids (32). These facts taken together support
a mechanism in which an electron is transferred from
MLT to ONOOH, and the reaction center is the 3-position
of MLT. This mechanism will be discussed below in more
detail.
159 M-1 s-1, eq 3), and under physiological conditions,
the reaction of MLT with ONOO- is about 10 times
slower than with ONOOH.
We derived3 eq 8 to calculate the loss of MLT (∆[MLT])
and explain the pH dependence we observe for the yields
of products (Figure 3).
∆[MLT]
k1[H+] × ln 1 -
(
)
[MLT]0
k2[H+] + k2′KPN
-
+ ∆[MLT] )
[peroxynitrite]0 (8)
The values of ∆Hq and ∆Sq for the peroxynitrite/MLT
reaction are in good agreement with those of other
reactions of peroxynitrite (Table 2). The large, negative
∆Sq value (-20 ( 2 eu) suggests a restricted configura-
tion of ONOOH, and MLT is required for the reaction to
occur. For example, a hydrogen bond between peroxyni-
trite and the indolic hydrogen has been suggested for the
peroxynitrite/tryptophan reaction (31), and association
with the involvement of some bond making has been
proposed for the peroxynitrite/iodide reaction (33).
P r otection of Mela ton in by CO2 fr om P er oxyn i-
tr ite-Med ia ted Oxid a tion . Carbon dioxide provides
partial protection of some biomolecules, such as glu-
tathione (34) and selenomethionine (35), against peroxy-
nitrite. Therefore, we investigated whether CO2 also
protects MLT. Figures 6 and 7 show kobs, the observed
rate constant for the formation of the products from MLT,
and ∆Amax, a measure of the yield of products form the
peroxynitrite/MLT reaction in the presence of bicarbonate
at pH 7.0, respectively. Values of kobs increase with
increasing concentrations of bicarbonate (Figure 6),
showing that the reaction of peroxynitrite with CO2 is
the rate-limiting step. However, the yields of the products
from MLT decrease with increasing concentrations of
bicarbonate (Figure 7), indicating protection of MLT by
CO2. Nevertheless, CO2 does not provide complete protec-
tion of MLT, as discussed below.
(A value of pKPN of 6.5, and the values of k2 and k2′ given
above were used for this calculation since these values
were obtained directly from fitting the data in Figure 2.)
In eq 8, [MLT]0 and [peroxynitrite]0 are the initial
concentrations of the two reactants. This equation only
approximately predicts ∆[MLT] since it does not take into
account minor processes such as the spontaneous decom-
position of the peroxynitrite anion or reaction of peroxy-
nitrite with traces of adventitious CO2. Despite its
simplicity, eq 8 predicts, as shown in Figure 3 (dotted
lines), that the yield of MLT oxidation will increase with
pH, in agreement with the experimental data (Figure 3,
solid circles).
Kin etics of th e P er oxyn itr ite/Mela ton in Rea c-
tion . The second-order rate constant for the reaction of
MLT with ONOOH is relatively small (159 M-1 s-1) and
very close to that of tryptophan (160 M-1 s-1) (31),
indicating that the methoxy group at C-5 does not
activate MLT toward peroxynitrite. The insensitivity of
the peroxynitrite/melatonin reaction to electron-donating
3
Equation 8 is derived from eq iii which, in turn, is obtained by
dividing the differential rate laws for the disappearance of MLT (eq i)
and peroxynitrite (eq ii).
Ca ta lytic Action of CO2 a n d th e F or m a tion of th e
F r ee Ra d ica ls. Carbon dioxide reacts with peroxynitrite
initially to form 1 (Scheme 1) (9, 10, 13, 14). We have
shown that ca. 80% of 1 in the presence of a substrate
such as tyrosine is converted to 2; 2 then rapidly
decomposes to regenerate CO2 (29). (The remaining 20%
(i)
•
of 1 becomes free CO3•- and NO2, and these radicals are
(ii)
trapped by the substrate.) In the absence of a substrate,
>90% of 1 is converted to 2 and subsequently regenerates
CO2 (36). The regeneration of CO2 in this pathway is fast,
so CO2 is a true catalyst for the decomposition of
peroxynitrite (29, 34, 36). This also means that in the
presence of CO2 only about 20% of the peroxynitrite is
available to oxidize a substrate such as melatonin. In
contrast, in the absence of CO2, and for a substrate that
(iii)

Reaction of Melatonin with Peroxynitrite
Chem. Res. Toxicol., Vol. 12, No. 6, 1999 531
Sch em e 2
Sch em e 3
follows second-order kinetics, all of the peroxynitrite that
undergoes this reaction is capable of oxidizing the
substrate.
lower kobs values (Scheme 1; MLT and other substrates
block the reaction shown as a dotted arrow). If MLT were
able to compete with CO2 for peroxynitrite, kobs would
increase rather than decrease with increasing concentra-
tions of MLT. In fact, melatonin reacts much more slowly
than CO2 with peroxynitrite. The reaction rate constants
for the ONOO-/CO2 and ONOOH/MLT reactions are
30 000 (9) and 159 M-1 s-1, respectively; these values
predict that <1% of the peroxynitrite reacts directly with
MLT under our reaction conditions, as described in the
legend of Figure 8. However, we expect as much as 20%
of the peroxynitrite reacts indirectly with MLT via the
peroxynitrite/CO2 reaction pathway (29). Figure 9 shows
that ∆Amax (a measure of the yield of products) increases
in a MLT concentration-dependent manner, resulting
When peroxynitrite reacts with CO2, intermediate
•
species such as CO3•- and NO2 are formed (9, 10, 13, 14,
36), as shown in Scheme 1. We therefore evaluated the
ability of melatonin to react with the radicals derived
from the peroxynitrite/CO2 reaction. We previously sug-
•-
gested glutathione (34) and tyrosine (29) scavenge CO3
•
and NO2. The rate constant for the reaction of MLT with
•-
CO3 (eq 9) is unknown, but probably is similar to that
for the reaction of tryptamine with CO3•-, 1.3 × 109 M-1
s-1 (37).
CO3•- + MLT f CO3- + MLT•+
(9)
•-
•
from the scavenging of the free radicals CO3 and NO2
by MLT.
Recently, we reported the formation of melatoninyl
radical cation (MLT•+) in a MLT/peroxynitrite/CO2 sys-
P ossible Mech a n ism s for th e F or m a tion of P r od -
u cts. Here we analyze the pathways (see Schemes 2 and
3) that lead to the main products of the reaction, the
hydroxypyrrolo[2,3-b]indoles 1,2,3,3a,8,8a-hexahydro-1-
acetyl-5-methoxy-8a-hydroxypyrrolo[2,3-b]indole (VI/VI′)
and 1,2,3,3a,8,8a-hexahydro-1-acetyl-5-methoxy-3a-hy-
•-
tem due to the reaction of MLT with CO3 (22). Figure
8 shows that with limiting CO2, kobs decreases with
increasing concentrations of MLT; this results since any
CO3•- that reacts with MLT cannot be converted back to
CO2, resulting in a decrease in the CO2 concentration and

532 Chem. Res. Toxicol., Vol. 12, No. 6, 1999
Zhang et al.
droxypyrrolo[2,3-b]indole (VII/VII′), and 6-hydroxy-
melatonin (6-HO-MLT).
tophan instead of melatonin.) The formation of hydroxy-
pyrrolo[2,3-b]indoles is emerging as an important reac-
tion pathway during the oxidation of tryptophan and
tryptophan metabolites. Moreover, Kato et al. (43) also
found peroxynitrite does not nitrate tryptophan.
(1) F or m a tion of VI/VI′ a n d VII/VII′. The peroxyni-
trite/MLT reaction yields 6-HO-MLT and the hydroxy-
pyrrolo[2,3-b]indoles VI/VI′ and VII/VII′ as the major
final products, but only the hydroxypyrrolo[2,3-b]indoles
are formed if CO2 also is present (22). The melatoninyl
radical cation is observed as an intermediate both in the
presence and in the absence of CO2 (22). Therefore, we
assume that MLT•+ leads to the hydroxypyrrolo[2,3-b]-
indoles VI/VI′ and VII/VII′ and not to 6-HO-MLT.
Since the reaction depicted in eq 9 is responsible for
the formation of MLT•+ in the presence of CO2, we
suggest that the ground state of peroxynitrous acid
(ONOOH) is responsible for the formation of MLT•+ in
the absence of CO2 (eq 10). This suggestion is supported
by the fact that the peroxynitrite/MLT reaction is first-
order in peroxynitrite and first-order in MLT, and the
high one-electron reduction potential for ONOOH/•NO2
(1.6 V at pH 7) (38), 0.4 V higher than that of most indolic
compounds (39). In line with this explanation, an electron
transfer pathway has been suggested for the very similar
reaction of peroxynitrite with tryptophan (31).
(2) F or m a t ion of 6-H yd r oxym ela t on in (6-H O-
MLT). This major product is formed only in the absence
of CO2 and thus is not derived from MLT•+. Its formation
probably involves the reaction of MLT with ONOOH* (eq
11), an activated form of peroxynitrite (8, 44). This species
also has been suggested to be responsible for the hy-
droxylation of phenol, which is first-order in peroxynitrite
but zero-order in phenol. [The rate-determining step is
the formation of ONOOH* (23).] Thus, the hydroxylation
of MLT also should be zero-order in MLT; this also is
suggested by the fact that the extrapolated y-intercept
of traces of kobs versus MLT concentration (inset of Figure
2) is larger than kobs in the absence of MLT. The reactions
of peroxynitrite with methionine, ascorbate, and tryp-
tophan also exhibit a similar pattern (3, 7, 31), and this
kinetic behavior was attributed to the reactions of
ONOOH* that can be detected at low substrate concen-
trations. However, these reactions of ONOOH*, including
the formation of 6-HO-MLT, are unimportant in vivo.
This is true because nearly all the peroxynitrite formed
in vivo will not form ONOOH* because ground state
peroxynitrite (ONOOH and ONOO-) will undergo fast
bimolecular reactions, generally with CO2 or with heme-
containing proteins such as hemoglobin (14), before it can
form ONOOH*. Nevertheless, melatonin is metabolized
by the liver mainly to 6-HO-MLT (45) in a peroxynitrite-
independent pathway.
ET
ONOOH + MLT 8 MLT•+ + •NO2 + OH- (10)
Scheme 2 shows a pathway in which MLT•+ is con-
verted to VI/VI′. In this mechanism, MLT•+ captures
•NO2, which is formed both in the absence and in the
presence of CO2 (Scheme 1 and eq 10), to form the
unstable adduct 2-hydroxy-3-nitro-2,3-dihydromelatonin
(II). Alternatively, MLT•+ can first couple with O2 and
then with •NO2 to form 2-hydroxy-3-peroxynitro-2,3-
dihydromelatonin (II′). Under strongly acidic conditions,
the NO2 or OONO2 groups at C-3 could migrate to C-2
in MLT (40), but at neutral pH, addition of hydroxide
ion is more likely, resulting in III/III′. Because of the
ONOOH* + MLT f f MLT-OH
(11)
Su m m a r y a n d Con clu sion s
The kinetics of the reaction of melatonin with peroxy-
nitrite has a first-order component (involving ONOOH),
and probably also a zero-order component (involving
ONOOH*), as previously observed for methionine, ascor-
bate, and tryptophan (3, 7, 31). Melatonin also reacts
with ONOO- with first-order kinetics; however, the rate
constant for this reaction is much smaller than that for
the melatonin/ONOOH reaction, and the reaction is
unimportant at physiological pH. The combination of
these reactions of melatonin with peroxynitrite gives rise
to a pH profile in which the yields of the products from
melatonin increase with pH. A mechanism is presented
that is in agreement with these observations.
-
acidity of the proton at C-2 in III/III′, NO2 (or O2 and
NO2-) is eliminated, to yield an oxindole derivative (V/
V′), which undergoes ring closure to give the final product
VI/VI′. A related mechanism for the formation of oxindole
from 3-chloroindole has been previously suggested (41).
The NMR spectra reveal that two isomers are formed in
an approximately 10:1 ratio.4 We suggest the other
isomer is a positional isomer in which the hydroxyl group
is on C-3a rather than on C-8a (VII/VII′ in Scheme 3).
In this regard, 1,2,3,3a,8,8a-hexahydro-3a-hydroxypyr-
rolo[2,3-b]-indole-2-carboxylic acid has been reported
together with 2-hydroxytryptophan during the oxidation
of tryptophan with hydrogen peroxide (42), and the tert-
butoxycarbonyl derivatives of these products had been
reported during the oxidation of (tert-butoxycarbonyl)-
tryptophan with peroxynitrite (43). (2-Hydroxytryp-
tophan is the analogue of IV, the open form of VI and
VI′ in Scheme 2, when the original substrate is tryp-
The peroxynitrite/melatonin reaction produces 6-
hydroxymelatonin, 1,2,3,3a,8,8a-hexahydro-1-acetyl-5-
methoxy-8a-hydroxypyrrolo[2,3-b]indole (VI/VI′), and
1,2,3,3a,8,8a-hexahydro-1-acetyl-5-methoxy-3a-hydroxy-
pyrrolo[2,3-b]indole (VII/VII′), but in the presence of
added bicarbonate, 6-hydroxymelatonin is not formed.
The melatoninyl radical cation (MLT•+) is formed when
the peroxynitrite/melatonin reaction occurs in either the
presence or absence of added bicarbonate (22). These
observations indicate that the hydroxypyrrolo[2,3-b]in-
dole products are formed from the secondary reactions
of MLT•+. Carbon dioxide lowers the yield of oxidation of
melatonin and also markedly lowers the yield of 6-hy-
droxymelatonin, but the peroxynitrite/CO2 reaction also
4
The 1H NMR data (in CDCl3) of the major isomer have been
reported previously (20). The major isomer was named 5-methoxy-2-
hydropyrroloindole in ref 20; the current name, 1,2,3,3a,8,8a-hexahy-
dro-1-acetyl-5-methoxy-8a-hydroxypyrrolo[2,3-b]indole, is consistent
with IUPAC and CAS nomenclature. The NMR data for the minor
isomer, 1,2,3,3a,8,8a-hexahydro-1-acetyl-5-methoxy-3a-hydroxypyrrolo-
[2,3-b]indole, are as follows: (δ in parts per million and J values in
hertz) δ 6.900 (1H, d, J ) 2.5), 6.806 (1H, dd, J 1 ) 2.5, J 2 ) 8.6), 6.651
(1H, d, J ) 8.6), 3.775 (3H, s), 3.736-3.652 (2H, m), 3.299 (1H, dd, J 1
) 7.1, J 2 ) 9.6), 2.560-2.339 (2H, m), 2.153 (3H, s). Note: When
resonances of the minor and major isomers overlap, the chemical shifts
of the major isomer are used.
gives CO3 and •NO2 that can oxidize melatonin to
•-
hydroxypyrrolo[2,3-b]indole products.

Reaction of Melatonin with Peroxynitrite
Chem. Res. Toxicol., Vol. 12, No. 6, 1999 533
The peroxynitrite/melatonin reaction does not produce
a nitrated product that is stable and can be isolated,
whether bicarbonate is present. Nitration probably occurs
both in the presence and in the absence of added
bicarbonate, but we suggest that the nitroproduct is
converted to an oxindole that then rearranges to give
1,2,3,3a,8,8a-hexahydro-1-acetyl-5-methoxy-8a-hydroxy-
pyrrolo[2,3-b]indole (VI/VI′). The hydroxylation of mela-
tonin by peroxynitrite to give 6-hydroxymelatonin, on the
other hand, likely occurs by reaction of melatonin with
the activated form of peroxynitrous acid, ONOOH*, but
this reaction is unimportant in vivo because nearly all
peroxynitrite formed in vivo does not form ONOOH* (14).
It is unlikely that the major biological function of
melatonin is to scavenge peroxynitrite or peroxynitrite-
derived species, since the physiological concentrations of
melatonin are very low, much lower than those of other
antioxidants and potential scavengers, such as ascorbate,
glutathione, and urate that react rapidly with peroxyni-
trite and/or peroxynitrite-derived species. Thus, most
peroxynitrite and peroxynitrite-derived species will react
with other antioxidants and not with melatonin. Never-
theless, melatonin is expected to react with the carbonate
radical5 produced from the reaction of peroxynitrite with
CO2 at a near diffusion-controlled rate, as do many other
tryptophan and tryptamine derivatives. Since the per-
oxynitrite/CO2 reaction governs the reactions of peroxy-
nitrite in vivo (13, 14), we suggest the hydroxypyrrolo[2,3-
b]indoles VI/VI′ and VII/VII′ are the main product from
the oxidation of melatonin by peroxynitrite-derived spe-
cies in vivo.
Melatonin is metabolized in the liver mainly to 6-hy-
droxymelatonin, but we do not think this metabolite is a
useful biomarker for melatonin’s antioxidant activity.
However, 6-hydroxymelatonin may be a better chain-
breaking antioxidant than melatonin, because of its
readily transferable phenoxyl hydrogen analogous to that
in R-tocopherol, and this metabolite may contribute to
the beneficial effects of melatonin in vivo. We suggest
that the formation of 1,2,3,3a,8,8a-hexahydro-1-acetyl-
5-methoxy-8a-hydroxypyrrolo[2,3-b]indole (VI/VI′) and/
or its interconverting forms (IV, V, and V′ in Scheme 2)
and the isomeric 1,2,3,3a,8,8a-hexahydro-1-acetyl-5-
methoxy-3a-hydroxypyrrolo[2,3-b]indole (VII/VII′) may
serve as indices for melatonin’s antioxidant activity, and
for evidence of its role as a scavenger of radicals derived
from the reaction of peroxynitrite with carbon dioxide.
To our knowledge, no attempt has been made to measure
these compounds in vivo, and further research in this
area is needed.
Ack n ow led gm en t. This publication was made pos-
sible in part by Grant ES-06754 from the National
Institute of Environmental Health Sciences, NIH. Its
contents are solely the responsibility of the authors and
do not necessarily represent the official views of the
NIEHS, NIH. We thank Dr. Wim Koppenol for the gift
of the stopped-flow instrument.
Refer en ces
(1) Radi, R., Beckman, J . S., Bush, K. M., and Freeman, B. A. (1991)
Peroxynitrite oxidation of sulfhydryls: the cytotoxic potential of
superoxide and nitric oxide. J . Biol. Chem. 266, 4244-4250.
(2) Salgo, M. G., Bermu´dez, E., Squadrito, G. L., and Pryor, W. A.
(1995) Peroxynitrite causes DNA damage and oxidation of thiols
in rat thymocytes. Arch. Biochem. Biophys. 322, 500-505.
(3) Squadrito, G. L., J in, X., and Pryor, W. A. (1995) Stopped-flow
kinetic study of the reaction of ascorbic acid with peroxynitrite.
Arch. Biochem. Biophys. 322, 53-59.
(4) Radi, R., Beckman, J . S., Bush, K. M., and Freeman, B. A. (1991)
Peroxynitrite-induced membrane lipid peroxidation: the cytotoxic
potential of superoxide and nitric oxide. Arch. Biochem. Biophys.
288, 481-487.
(5) Salgo, M. G., and Pryor, W. A. (1996) Trolox inhibits peroxynitrite-
mediated oxidative stress and apoptosis in rat thymocytes. Arch.
Biochem. Biophys. 333, 482-488.
(6) Van Der Vliet, A., O’Neill, C. A., Halliwell, B., Cross, C. E., and
Kaur, H. (1994) Aromatic hydroxylation and nitration of phenyl-
alanine and tyrosine by peroxynitrite. evidence for hydroxyl
radical production from peroxynitrite. FEBS Lett. 339, 89-92.
(7) Pryor, W. A., J in, X., and Squadrito, G. L. (1994) One- and two-
electron oxidations of methionine by peroxynitrite. Proc. Natl.
Acad. Sci. U.S.A. 91, 11173-11177.
(8) Pryor, W. A., and Squadrito, G. L. (1995) The chemistry of
peroxynitrite: a product from the reaction of nitric oxide with
superoxide. Am. J . Physiol. 268, L699-L722.
(9) Lymar, S. V., and Hurst, J . K. (1995) Rapid reaction between
peroxonitrite ion and carbon dioxide: implications for biological
activity. J . Am. Chem. Soc. 117, 8867-8868.
(10) 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.
(11) Denicola, A., Freeman, B. A., Trujillo, M., and Radi, R. (1996)
Peroxynitrite reaction with carbon dioxide/bicarbonate: kinetics
and influence on peroxynitrite mediated oxidations. Arch. Bio-
chem. Biophys. 333, 49-58.
(12) Lentner, C. (1984) Physical Chemistry, Composition of Blood,
Hematology, Somatometric Data. Geigy Scientific Tables, Vol. 3,
pp 71-73, 98, 132, Ciba-Geigy, West Caldwell, NJ .
(13) Squadrito, G. L., and Pryor, W. A. (1998) Oxidative chemistry of
nitric oxide: The roles of superoxide, peroxynitrite, and carbon
dioxide. Free Radical Biol. Med. 25, 392-403.
(14) Squadrito, G. L., and Pryor, W. A. (1998) The nature of reactive
species in systems that produce peroxynitrite. Chem. Res. Toxicol.
11, 718-719.
(15) Antunes, F., Barclay, L. R. C., Ingold, K. U., King, M., Norris, J .
Q., Scaiano, J . C., and Xi, F. (1999) On the antioxidant activity
of melatonin. Free Radical Biol. Med. 26, 117-128.
(16) Reiter, R. J ., Menendez-Pelaez, A., Poeggeler, B., Tan, D.-X.,
Pablos, M. I., and Acun˜a-Castroviejo, D. (1994) The role of
melatonin in the pathophysiology of oxygen radical damage. In
Advances in Pineal Research (Moller, M., and Pe´vet, P., Eds.) pp
403-412, J ohn Libbey & Co., London.
(17) Marshall, K.-A., Reiter, R. J ., Poeggeler, B., Aruoma, O. I., and
Halliwell, B. (1996) Evaluation of the antioxidant activity of the
melatonin in vitro. Free Radical Biol. Med. 21, 307-315.
(18) Cuzzocrea, S., Zingarelli, B., Gilad, E., Hake, P., Salzman, A. L.,
and Szabo´, C. (1997) Protective effect of melatonin in carrageenan-
induced models of local inflammation: Relationship to its inhibi-
tory effect on nitric oxide production and its peroxynitrite
scavenging activity. J . Pineal Res. 23, 106-116.
(19) Reiter, R., Tang, L., Garcia, J . J ., and Munoz-Hoyos, A. (1997)
Pharmacological actions of melatonin in oxygen radical patho-
physiology. Life Sci. 60, 2255-2271.
(20) Longoni, B., Salgo, M. G., Pryor, W. A., and Marchiafava, P. (1998)
Effects of melatonin on lipid peroxidation induced by oxygen
radicals. Life Sci. 62, 853-859.
5 Melatonin is predicted to react with CO3•- produced by the reaction
between peroxynitrite and CO2 despite the low concentration of
melatonin in vivo. This can be demonstrated by using the known rate
constants for the reaction of CO3 with glutathione (5.3 × 106 M-1
•-
s-1) and ascorbate (1.4 × 109 M-1 s-1) and assuming the rate constant
for the reaction of CO3•- with melatonin, which is unknown, is similar
to that of tryptamine (1.3 × 109 M-1 s-1), an analogue and precursor
of melatonin. When the physiological concentrations of ascorbate (61
µM), glutathione (1.1 mM), and melatonin (211 nM) are considered
(12), one can calculate the corresponding pseudo-first-order rate
constants for ascorbate (8.5 × 104 s-1), glutathione (5.8 × 103 s-1), and
melatonin (270 s-1). Since the pseudo-first-order rate constants are
proportional to the amount of these antioxidants that will react with
•-
CO3•-, if one assumes 20 µM CO3 derived from peroxynitrite is
•-
produced in this environment, one can calculate that CO3 reacts
•-
(21) Gilad, E., Cuzzocrea, S., Zingarelli, B., Salzman, A. L., and Szabo´,
C. (1997) Melatonin is a scavenger of peroxynitrite. Life Sci. 60,
169-174.
(22) Zhang, H., Squadrito, G. L., and Pryor, W. A. (1998) The reaction
of melatonin with peroxynitrite: formation of melatonin radical
mostly with ascorbate (74%), but also that enough CO3 reacts with
melatonin to consume 28% of its initial concentration. A more rigorous
simulation using the Bio/chemical Kinetics Simulator software package
GEPASI (Pedro Mendes, University of Wales Aberystwyth) affords
similar results.

534 Chem. Res. Toxicol., Vol. 12, No. 6, 1999
Zhang et al.
cation and absence of stable nitrated products. Biochem. Biophys.
Res. Commun. 251, 83-87.
(23) Lemercier, J .-N., Padmaja, S., Cueto, R., Squadrito, G. L., Uppu,
R. M., and Pryor, W. A. (1997) Carbon dioxide modulation of
hydroxylation and nitration of phenol by peroxynitrite. Arch.
Biochem. Biophys. 345, 160-170.
(24) 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.
(25) Uppu, R. M., Squadrito, G. L., Cueto, R., and Pryor, W. A. (1996)
Synthesis of peroxynitrite by the azide-ozone reaction. Methods
Enzymol. 269, 311-321.
(26) Pryor, W. A., Cueto, R., J in, X., Koppenol, W. H., Ngu-Schwemlein,
M., Squadrito, G. L., Uppu, P. L., and Uppu, R. M. (1995) A
practical method for preparing peroxynitrite solutions of low ionic
strength and free of hydrogen peroxide. Free Radical Biol. Med.
18, 75-83.
(27) Zhang, H., Squadrito, G. L., and Pryor, W. A. (1996) The
contribution of ionization enthalpies to the apparent activation
energy of the reaction of peroxynitrite with thiols. 211th American
Chemical Society National Meeting, New Orleans, LA, March 24-
28, 1996.
(28) Koppenol, W. H., and Kissner, R. (1998) Can OdNOOH undergo
homolysis? Chem. Res. Toxicol. 11, 87-90.
(29) Zhang, H., Squadrito, G. L., and Pryor, W. A. (1997) The
mechanism of the peroxynitrite-carbon dioxide reaction probed
using tyrosine. Nitric Oxide 1, 301-307.
(30) J in, X. (1994) Nitric Oxide Toxicity Mediated by Peroxynitrite.
Ph.D. Dissertation, Louisiana State University, Baton Rouge, LA.
(31) Alvarez, B., Rubbo, H., Kirk, M., Barnes, S., Freeman, B. A., and
Radi, R. (1996) Peroxynitrite dependent tryptophan nitration.
Chem. Res. Toxicol. 9, 390-396.
(35) Padmaja, S., Squadrito, G. L., Lemercier, J .-N., Cueto, R., and
Pryor, W. A. (1997) Peroxynitrite-mediated oxidation of D,L-
selenomethionine: kinetics, mechanism and the role of carbon
dioxide. Free Radical Biol. Med. 23, 917-926.
(36) Pryor, W. A., Lemercier, J .-N., Zhang, H., Uppu, R. M., and
Squadrito, G. L. (1997) The catalytic role of carbon dioxide in the
decomposition of peroxynitrite. Free Radical Biol. Med. 23, 331-
338.
(37) Ross, A. B., Bielski, B. H. J ., Buxton, G. V., Cabelli, D. E., Helman,
W. P., Huie, R. E., Grodkowski, J ., Neta, P., Mulazzani, Q. G.,
and Wilkinson, F. (1998) NDRL-NIST Solution Kinetics Database
3.0, NDRL-NIST, Gaithersburg, MD.
(38) Koppenol, W. H., and Kissner, R. (1998) Can OdNOOH undergo
homolysis? Chem. Res. Toxicol. 11, 87-90.
(39) J ovanovic, S. V., and Steenken, S. (1992) Substituent effects on
the spectral, acid-base, and redox properties of indolyl radicals:
a pulse radiolysis study. J . Phys. Chem. 96, 6674-6679.
(40) Phillips, R. S., and Cohen, L. A. (1988) Synthesis of 2-nitro-L-
tryptophan. J . Hetrocycl. Chem. 25, 191-192.
(41) Sundberg, R. J . (1970) The Chemistry of Indoles. In Organic
Chemistry (Blomquist, A., Ed.) pp 14-17, Academic Press, New
York.
(42) Simat, T. J ., and Steinhart, H. (1998) Oxidation of free tryptophan
and tryptophan residues in peptides and proteins. J . Agric. Food
Chem. 46, 490-498.
(43) Kato, Y., Kawakishi, S., Aoki, T., Itakura, K., and Osawa, T.
(1997) Oxidative modification of tryptophan residues exposed to
peroxynitrite. Biochem. Biophys. Res. Commun. 234, 82-84.
(44) Goldstein, S., Squadrito, G. L., Pryor, W. A., and Czapski, G.
(1996) Direct and indirect oxidations by peroxynitrite, neither
involving the hydroxyl radical. Free Radical Biol. Med. 21, 965-
974.
(32) Melzer, M. S. (1962) Applicability of the Hammett equation to
the indole system: acidity of indole-3-carboxylic acids. J . Org.
Chem. 27, 496-499.
(33) Goldstein, S., Meyerstein, D., van Eldik, R., and Czapski, G. (1997)
Spontaneous reactions and reduction by iodide of peroxynitrite
and peroxynitrate: mechanistic insight from activation param-
eters. J . Phys. Chem. A 101, 7114-7118.
(34) Zhang, H., Squadrito, G. L., Uppu, R. M., Lemercier, J .-N., Cueto,
R., and Pryor, W. A. (1997) Inhibition of peroxynitrite-mediated
oxidation of glutathione by carbon dioxide. Arch. Biochem. Bio-
phys. 339, 183-189.
(45) Markey, S. P., Higa, S., Shih, M., Danforth, D. N., and Tamarkin,
L. (1985) The correlation between human plasma melatonin levels
and urinary 6-hydroxymelatonin excretion. Clin. Chim. Acta 150,
221-225.
(46) Bartlett, D., Church, D. F., Bounds, P. L., and Koppenol, W. H.
(1995) The kinetics of the oxidation of L-ascorbic acid by peroxy-
nitrite. Free Radical Biol. Med. 18, 85-92.
TX980243T