Mechanisms of NO release by N1-nitrosomelatonin: Nucleophilic attack versus reducing pathways

Article abstract of DOI:10.1021/jo047720z

A new type of physiologically relevant nitrosamines have been recently recognized, the N¹-nitrosoindoles. The possible pathways by which N¹-nitrosomelatonin (NOMel) can react in physiological environments have been studied. Our results show that NOMel slowly decomposes spontaneously in aqueous solution, generating melatonin as the main organic product (k = (3.7 ± 1.1) × 10^-5 s^-1, Tris-HCl (0.2 M) buffer, pH 7.4 at 37°C, anaerobic). This rate is accelerated by acidification (k_{pH 5.8} = (4.5 ± 0.7) × 10^-4 s^-1, k_{pH 8.8} = (3.9 ± 0.6) × 10^-6 s^-1 Tris-HCl (0.2 M) buffer at 37°C), by the presence of O₂ (k _o = (9.8 ± 0.1) × 10^-5 s^-1 pH 7.4, 37°C, [NOMel] = 0.1 mM, P(O₂) = 1 atm), and by the presence of the spin trap TEMPO (2,2,6,6-tetramethylpiperidine 1-oxyl; k_o = (2.0 ± 0.1) × 10^-4 s^-1, pH 7.4, 37°C, [NOMel] = 0.1 mM, [TEMPO] = 9 mM). We also found that NOMel can transnitrosate to L-cysteinate, producing S-nitrosocysteine and melatonin (k = 0.127 ± 0.002 M^-1 s^-1, Tris-HCl (0.2 M) buffer, pH 7.4 at 37°C). The reaction of NOMel with ascorbic acid as a reducing agent has also been studied. This rapid reaction produces nitric oxide and melatonin. The saturation of the observed rate constant (k = (1.08 ± 0.04) × 10^-3 s^-1, Tris-HCl (0.2 M) buffer, pH 7.4 at 37°C) at high ascorbic acid concentration (100-fold with respect to NOMel) and the pH independence of this reaction in the pH range 7-9 indicate that the reactive species are ascorbate and melatonyl radical originated from the reversible homolysis of NOMel. Taking into account kinetic and DFT calculation data, a comprehensive mechanism for the denitrosation of NOMel is proposed. On the basis of our kinetics results, we conclude that under physiological conditions NOMel mainly reacts with endogenous reducing agents (such as ascorbic acid), producing nitric oxide and melatonin.

Full text of DOI:10.1021/jo047720z

Mechanisms of NO Release by N1-Nitrosomelatonin: Nucleophilic
Attack versus Reducing Pathways
Pablo M. De Biase, Adria´n G. Turjanski, Dar´ıo A. Estrin, and Fabio Doctorovich*
Departamento de Quı´mica Inorga´nica, Analı´tica y Quı´mica Fı´sica/INQUIMAE, Facultad de Ciencias
Exactas y Naturales, Universidad de Buenos Aires/CONICET, Ciudad Universitaria, Pab. II, P. 3,
C1428EHA Buenos Aires, Argentina
doctorovich@qi.fcen.uba.ar
Received December 30, 2004
A new type of physiologically relevant nitrosamines have been recently recognized, the N1-
nitrosoindoles. The possible pathways by which N1-nitrosomelatonin (NOMel) can react in
physiological environments have been studied. Our results show that NOMel slowly decomposes
spontaneously in aqueous solution, generating melatonin as the main organic product (k ) (3.7 (
1.1) × 10-5 s-1, Tris-HCl (0.2 M) buffer, pH 7.4 at 37 °C, anaerobic). This rate is accelerated by
acidification (kpH 5.8 ) (4.5 ( 0.7) × 10-4 s-1, kpH 8.8 ) (3.9 ( 0.6) × 10-6 s-1, Tris-HCl (0.2 M) buffer
at 37 °C), by the presence of O2 (ko ) (9.8 ( 0.1) × 10-5 s-1, pH 7.4, 37 °C, [NOMel] ) 0.1 mM,
P(O2) ) 1 atm), and by the presence of the spin trap TEMPO (2,2,6,6-tetramethylpiperidine 1-oxyl;
ko ) (2.0 ( 0.1) × 10-4 s-1, pH 7.4, 37 °C, [NOMel] ) 0.1 mM, [TEMPO] ) 9 mM). We also found
that NOMel can transnitrosate to L-cysteinate, producing S-nitrosocysteine and melatonin (k )
0.127 ( 0.002 M-1 s-1, Tris-HCl (0.2 M) buffer, pH 7.4 at 37 °C). The reaction of NOMel with
ascorbic acid as a reducing agent has also been studied. This rapid reaction produces nitric oxide
and melatonin. The saturation of the observed rate constant (k ) (1.08 ( 0.04) × 10-3 s-1, Tris-
HCl (0.2 M) buffer, pH 7.4 at 37 °C) at high ascorbic acid concentration (100-fold with respect to
NOMel) and the pH independence of this reaction in the pH range 7-9 indicate that the reactive
species are ascorbate and melatonyl radical originated from the reversible homolysis of NOMel.
Taking into account kinetic and DFT calculation data, a comprehensive mechanism for the
denitrosation of NOMel is proposed. On the basis of our kinetics results, we conclude that under
physiological conditions NOMel mainly reacts with endogenous reducing agents (such as ascorbic
acid), producing nitric oxide and melatonin.
Introduction
The endogenous formation of NO plays a key role in
many bioregulatory systems including platelet inhibition,
vasodilation, and smooth muscle relaxation.2,3 Even
though the normal concentration of NO in mammals is
on the order of nanomolar,4 the local concentration can
be increased substantially, facilitating the formation of
Nitric oxide (NO) represents the biologically important
form of the endothelium-derived relaxing factor (EDRF).1
Under physiological conditions, NO directly activates
soluble guanylyl cyclase (sGC) to transform guanosine
triphosphate (GTP) into cyclic guanosine monophosphate
(GMP), followed by kinase-mediated signal transduction.
(2) Garbers, D. L. Methods 1999, 19, 477-484.
(3) Lucas, K. A.; Pitari, G. M.; Kazerounian, S.; Ruiz-Stewart, I.;
Park, J.; Schulz, S.; Chepenik, K. P.; Waldman, S. A. Pharmacol. Rev.
2000, 52 (3), 375-414.
(4) Goldstein, S.; Czapski, G. J. Am. Chem. Soc. 1996, 118 (14),
3419-3425.
* To whom correspondence should be addressed. Fax: 54-11-4576-
3341.
(1) Moncada, S.; Palmer, R. M. J.; Higgs, E. A. Pharmacol. Rev.
1991, 43 (2), 109-142.
10.1021/jo047720z CCC: $30.25 © 2005 American Chemical Society
Published on Web 06/28/2005
5790
J. Org. Chem. 2005, 70, 5790-5798

Mechanisms of NO Release by NOMel
a complete picture of the mechanisms involved in NO
release by this new type of physiologically relevant
nitroso compound.
Methods
Computational Procedures. The calculations were per-
formed at the DFT level using the GAUSSIAN 98 software
package.15 All species where fully optimized at the B3LYP/
6-31G** level. Solvent effects were modeled using the polarized
continuum model (PCM) scheme. The PCM implementation
given in ref 16, in which the self-consistency between the solute
wave function and solvent polarization is achieved during the
self-consistent field cycle, has been employed. The Gibbs free
energy for solvation in water was obtained by making single-
point calculations using PCM at vacuum-optimized geometries.
The thermal contributions at 37 °C to the free energies were
computed using standard statistical mechanics, the harmonic
oscillator, rigid rotor, and ideal gas approximations. The
frequency analysis predicted the absence of imaginary fre-
quencies for all the optimized species.
Experimental Procedures. i. General Comments. Un-
less otherwise noted, all manipulations were performed with
exclusion of oxygen using standard Schlenk procedures or
septa; Ar was used as an inert gas. 1H NMR spectra were
recorded using 500 and 200 MHz spectrometers. Nitric oxide
was detected with a specific electrode.
FIGURE 1. N1-Nitrosomelatonin (NOMel) structure.
N2O3, a potent nitrosating agent.5 The nitrosation reac-
tion introduces a nitroso group into an organic molecule,
leading to the formation of C-nitroso, N-nitroso, O-
nitroso, or S-nitroso derivatives of the parent molecule.
The formation of S-nitrosothiols is now generally believed
to be of high physiological importance in vivo due to the
capability of these compounds to activate sGC.6
Recently, a new type of physiologically relevant nitro-
samines have been recognized, the N1-nitrosoindoles. The
nitrosation of tryptophan has been studied in proteins
and model peptides, and it has been proposed that it can
compete with cysteine for NO.5 In previous works we
have characterized the novel compound N1-nitrosomela-
tonin (NOMel) (Figure 1)sformed by the reaction of NO
with the naturally occurring hormone melatonin (Mel).7-9
The decomposition of NOMel under normoxic solutions
has been described to be a pH-dependent first-order
reaction with a rate constant in aqueous phosphate buffer
ii. Reagents. Distilled acetonitrile and ethanol and recrys-
tallized ascorbic acid were used. TEMPO (2,2,6,6-tetrameth-
ylpiperidine 1-oxyl), NADH, L-cysteine, and deuterated sol-
vents (benzene, acetonitrile, methanol, and water) were used
as received. N1-Nitrosomelatonin was prepared from melatonin
purchased from Drogueria Saporiti SACIFIA, Buenos Aires,
Argentina. The water used in all reactions was MilliQ water
obtained from deionized water.
at pH 7.4 and 25 °C of (7.0 ( 0.5) × 10-5 s-1 10
.
NOMel
and N1-nitroso-N-acetyltryptophan (NOTrp) are able to
release NO in the presence of reducing agents such as
ascorbic acid and NADH (reduced nicotinamide adenine
dinucleotide).5,10 This was also observed for S-nitroso-
glutathione.5,11 On the other hand, transnitrosation reac-
tions of N1-nitrosamines with thiols12,13 and S-nitroso-
thiols with amines14 have been reported. In view of the
physiological significance of these compounds, their
stability and the possible mechanisms of NO release
deserve further investigation.
iii. N1-Nitrosomelatonin Preparations. Method A.
N1-Nitrosomelatonin was generated by bubbling a 0.2 M
melatonin solution in acetonitrile or ethanol with nitric oxide
and oxygen. Nitric oxide was generated by the reaction of
sodium nitrite and ferrous sulfate in acidic medium. The
solution was bubbled for 30 min, adding NO and O2 in
stoichiometric amounts (NO:O2 ) 4:1). Finally, the system was
purged with argon and cooled to 0 °C. After the addition of
cold water the product precipitated. The so-obtained yellow
powder was filtered, washed with water, and dried. The
isolated yield was about 70%. The purity was checked by TLC
1
In the present work we have studied the denitrosation
mechanism of NOMel as a model for N1-nitrosoin-
doles. We have thoroughly characterized both theoreti-
cally and experimentally the spontaneous decomposition
of NOMel, its reaction with reducing agents such as
ascorbic acid, and the transnitrosation reaction between
NOMel and the thiol L-cysteine (Cys). Our results provide
and H NMR (CD3CN) (ppm): 1.81 (s), 1.84 (s), 2.10 (s), 2.12
(s), 2.86 (m), 3.50 (m), 3.81 (s), 3.87 (s), 6.47 (sa), 6.87 (s), 6.96
(dd), 7.08 (dd), 7.18 (dd), 7.63 (s), 7.81 (s), 8.01 (s), 8.05 (s),
8.22 (s), 8.26 (s).
Method B. NOMel was synthesized by adding a 2-fold
excess of ethyl nitrite to a deoxygenated 0.2 M melatonin
solution in ethanol. Finally, the same procedure as in method
A was followed to obtain the purified product. The isolated
yield was about 80%.
iv. Kinetics. Rate measurements involving N1-nitrosome-
latonin were done in spectrophotometric cells at 37 °C,
measuring the disappearance of the absorption at 346 nm (ꢀ
(5) Kirsch, M.; Fuchs, A.; de Groot, H. J. Biol. Chem. 2003, 278 (14),
11931-11936.
(6) Stamler, J. S.; Lamas, S.; Fang, F. C. Cell 2001, 106 (6), 675-
683.
(7) Turjanski, A. G.; Chaia, Z. D.; Doctorovich, F.; Estrin, D.;
Rosenstein, R.; Piro, O. E. Acta Crystallogr., C 2000, 56, 682-683.
(8) Turjanski, A. G.; Leonik, F.; Estrin, D. A.; Rosenstein, R. E.;
Doctorvich, F. J. Am. Chem. Soc. 2000, 122 (42), 10468-10469.
(9) Turjanski, A. G.; Saenz, D. A.; Doctorovich, F.; Estrin, D. A.;
Rosenstein, R. E. J. Pineal Res. 2001, 31 (2), 97-101.
(10) Blanchard-Fillion, B.; Servy, C.; Ducrocq, C. Free Radical Res.
2001, 35 (6), 857-866.
(15) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Strat-
mann, R.; Burant, J.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin,
K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.;
Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochter-
ski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick,
D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski,
J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith,
T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.;
Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.;
Andres, J. L.; Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople,
J. A. Gaussian 98, Rev. A7; Gaussian, Inc.: Pittsburgh, PA, 1998.
(16) Cossi, M.; Barone, V.; Cammi, R.; Tomasi, J. Chem. Phys. Lett.
1996, 255, 327.
(11) Kashiba-Iwatsuki, M.; Yamaguchi, M.; Inoue, M. FEBS Lett.
1996, 389 (2), 149-152.
(12) Hallett, G.; Williams, D. L. H. J. Chem. Soc., Perkin Trans. 2
1980, No. 4, 624-627.
(13) Sonnenschein, K.; de Groot, H.; Kirsch M. J. Biol. Chem. 2004,
279, 45433-45440.
(14) Al-Mustafa, A. H.; Sies, H.; Stahl, W. Toxicology 2001, 163 (2-
3), 127-136.
J. Org. Chem, Vol. 70, No. 15, 2005 5791

De Biase et al.
TABLE 1. Observed Rate Constants (ko) for NOMel
Decomposition in Diverse Media at 37 °C
ko
(10-5 s-1
)
pH
medium
2.2 ( 0.1
6.80a
distilled and degassed water
with 1.5% v/v acetonitrile
0.2 M Tris-HCl buffer in MilliQ
water with 0.8% v/v ethanol
0.2 M phosphate buffer in water
with 0.8% v/v ethanol
3.7 ( 1.1
7.40
7.40
7.22
14.0 ( 0.1
85.2 ( 1.3
1.0 M phosphate buffer in water
with 1.7% v/v acetonitrile
a Initial pH, measured at 25 °C, all others at 37 °C.
) 8616 ( 300 M-1 cm-1). The observed rate constants (kobs
)
FIGURE 2. Effect of pH on the observed rate constant (ko)
for NOMel decay in Tris-HCl buffer at 37 °C. The solution
has less than 1% ethanol to facilitate the dissolution of NOMel
in water.
for the spontaneous decay process, the reaction with TEMPO,
O2, cysteine, and ascorbic acid, were determined by initial rates
or by the integral method, showing that all kinetics were first-
order with respect to NOMel concentration. These reactions
were carried out in Tris-HCl or phosphate aqueous buffered
solutions with 0.8% v/v ethanol or less than 1.7% v/v aceto-
nitrile, respectively.
of NOMel in aqueous solution. However, the main
products of the reaction (HNO2 and Mel) cannot be
explained in terms of this mechanism alone, indicating
that other pathways are also operative.
v. E° Determinations. Voltammetry was carried out at
100 mV/s in water-ethanol (1:1) using potassium nitrate
as the supporting electrolyte, and a three-electrode configu-
ration (Pt working and counter electrodes, Ag/AgCl reference
electrode). The working temperature was 25 °C, and the
NOMel concentration used was about 1.5 mM. The cyclic
voltammograms showed that the reduction was chemically
irreversible.
NOMel spontaneous decomposition was carried out
anaerobically in deuterated water-methanol mixtures
(1:1), methanol, or benzene, avoiding light exposure.
Products were characterized by 1H NMR. In water-
methanol mixtures, NOMel decomposed completely and
only melatonin (Mel) was detected as the organic product
by NMR (Figure SI1 in the Supporting Information). In
contrast, in benzene and methanol, no reaction was
observed after the same time (6 days). This result
indicates that water plays an essential role in the
anaerobic reaction, producing Mel and presumably ni-
trous acid (eq 4). In benzene and in methanol, the only
operative reaction seems to be NOMel reversible N-N
homolytic bond dissociation, so this equilibrium (eq 1) is
displaced to the reactant side.
Results and Discussion
Reaction of NOMel with Water and Other Nu-
cleophiles. It has been reported that 3-substituted
N1-nitrosoindoles decompose spontaneously through a
first-order reaction,5,10,17 in the pH 0-7 range, yielding
nitrous acid and the corresponding indole.17 Moreover,
it was found that in aqueous solution at about pH 7 the
reactant is able to slowly release NO;5 in the presence of
oxygen NOMel decomposes, producing melatonin and
oxidized melatonin derivatives (eq 2).10 These last two
pieces of evidence suggest that the reaction may involve
a N-N homolytic bond dissociation step (eq 1).
NOMel + H2O f Mel + NO2- + H+
(4)
The spontaneous decomposition of NOMel in water was
followed by the decrease in NOMel absorbance at 346 nm
under an argon atmosphere and avoiding light exposure.
The reaction was studied in different buffer media to test
the influence of the media over the reaction rate (Table
1). We found that in phosphate buffer the reaction is
almost 4 times faster than in Tris-HCl buffer under the
same conditions. This kind of catalysis by phosphate was
also observed for the hydrolysis of N2O3.18 We determined
the observed rate constant (ko) between pH 6 and pH 9
in Tris-HCl-buffered solutions at 37 °C (Figure 2). The
reaction rate increases at low pH, indicating that protons
play a role in NOMel decomposition. Similar results were
seen for N1-nitroso-N-acetyltryptophan (NOTrp) between
pH 0 and pH 7.17
NOMel h NO + Mel•
(1)
Mel• + O2 f oxidized melatonin derivatives (2)
In Tris aqueous buffer, the spin trap compound TEMPO
accelerated the observed rate constant (ko) for NOMel
decay (eq 3) (ko ) (2.0 ( 0.1) × 10-4 s-1, pH 7.4, 37 °C,
ν ) ko[NOMel]
(3)
[NOMel] ) 0.1 mM, [TEMPO] ) 9 mM) as compared with
the rate constant without TEMPO (Table 1). The decay
rate in the presence of molecular oxygen was also
accelerated (ko ) (9.8 ( 0.1) × 10-5 s-1, pH 7.4, 37 °C,
[NOMel] ) 0.1 mM, P(O2) ) 1 atm). This rate was not
significantly affected by raising the pH to 9. The above
results support the intermediacy of free radicals in the
reaction and therefore the homolytic bond dissociation
Furthermore, Meyer et al. studied the effect of nucleo-
philes (Nu-’s) on the NOTrp decay rate at pH 6.17 They
observed that I-, N3-, SCN-, Br-, and Cl-, in decreasing
order of effectiveness, catalyze NOTrp decomposition. For
the more powerful nucleophiles, the first-order depen-
(17) Meyer, T. A.; Williams, D. L. H.; Bonnett, R.; Ooi, S. L. J. Chem.
Soc., Perkin Trans. 2 1982, No. 11, 1383-1387.
(18) Lewis, R. S.; Tannenbaum, S. R.; Deen, W. M. J. Am. Chem.
Soc. 1995, 117 (14), 3933-3939.
5792 J. Org. Chem., Vol. 70, No. 15, 2005

Mechanisms of NO Release by NOMel
SCHEME 1. NOTrp Denitrosation Mechanism Proposed by Castro et al.19
TABLE 2. Optimized Bond Distances and Mulliken
Charges Obtained from DFT Calculations
the basis of the ∆G° of protonation (water, 37 °C), the
most plausible intermediate is 3 (3-NO-1-H-MMIndole+).
distance
distance
N-O (Å)
X
X-NO (Å)
δa
1-NO-MMIndole + H3O+ ∆G°8
H-NO-MMIndole+ + H2O (5)
NO
1.159
1.073
1.279
1.220
1.145
1.199
1.134
1.135
1.129
1.227
1.340
0
NO+
+1
NO-
1-NOMMIndole
-1
N1
N1
N1
C3
C2
N1
N1
N1
1.363
1.685
1.417
2.146
2.169
2.277
1.308
1.272
-0.227
+0.275
-0.082
+0.276
+0.258
+0.311
-0.086
-0.205
1
2
3
4
5
6
7
A direct protonation does not explain the previous
results in which it was observed that a limiting step prior
to protonation exists.17,19 Moreover, the DFT-calculated
pKa of intermediate 3 (-9.3) does not seem to be compat-
ible with the observed results. In the range of pH in
which we worked (pH 5.8-9.0), we found that the species
most susceptible to protonation is the indolic radical Mel•
(Table 4), suggesting that the previous limiting step could
be homolytic N-N bond dissociation of NOMel to Mel•
and NO, followed by protonation of the indolic radical.
On the basis of these results, a new mechanism for the
denitrosation of NOMel, which in principle could be
extended to other 3-substituted-nitrosoindoles, can be
proposed (Scheme 2). In Figure 4, the free energy profile
for the proposed mechanism is compared with that
corresponding to mechanisms b′ and c′ proposed by
Castro et al.19 While they proposed NO translocation
followed by protonation of the indole, our suggested
pathway involves indole protonation prior to NO recom-
bination. The formation of their proposed intermediate,
3-NO-MMIndole, is thermodynamically disfavored. Con-
sidering its equilibrium concentration, this adduct would
be undetectable. On the other hand, the equilibrium
concentration of MMIndole• is considerably higher (0.1
µM for 0.1 mM nitrosoindole). Energetic values for each
step are shown in Table SI1 in the Supporting Informa-
tion.
Nucleophilic attack would take place in step d. The
energetic value of this step was calculated considering
water as the nucleophile. At a pH greater than 5, NO+
carriers undergo a change to nitrite, averting the deni-
trosation reversibility (Scheme 2, step g). This new
mechanism seems to be compatible with our results and
with those reported previously.
NOMel Transnitrosation to L-Cysteine (Cys). The
kinetics for the reaction of NOMel and Cys was measured
in Tris-HCl (0.2 M) buffer, pH 7.4 at 37 °C, under an
excess of Cys ([Cys] ) 1.9 mM/10 mM) with respect to
NOMel ([NOMel] ) 100 µM). We found a global second-
order reaction, first-order in NOMel and in Cys with a
rate constant kr ) 0.127 ( 0.002 M-1 s-1 at 37 °C. The
first-order in Cys result was also verified under true
second-order conditions, measuring initial rates at pH
9.0 where NOMel spontaneous decay is less significant
([Cys] ) 0.19 mM/0.38 mM, [NOMel] ) 0.1 mM, 37 °C,
kr ) 0.91 ( 0.03 M-1 s-1 determined using eq 7).
a δ ) Mulliken net charge over the NO group in water.
dence is soon lost when the nucleophile concentration is
increased, and ko tends to the same limiting value for
each nucleophile. This indicates that an existing previous
step limits the nucleophilic attack. Moreover, they ob-
served that between pH 2 and pH 4 there is little change
in ko. This saturation was attributed to a protonation of
NOTrp with a pKa value of ca. 5.5;17 however, this pKa is
much larger than expected for any possible protonation
site in NOTrp that can lead to denitrosation. In another
work, Castro et al. explained this effect, proposing a
translocation of the NO moiety from position 1 to position
3 of the indolic ring (Scheme 1).19 The 3-nitrosoindole so
formed was suggested to be more susceptible to proto-
nation than its precursor, the 1-nitrosoindole. In that
way, the limiting step responsible for the saturation
would be translocation. Finally, this protonated adduct
loses NO+, in a pathway catalyzed by Nu-’s.
To get additional insight about the spontaneous de-
composition mechanism, we performed DFT calculations
on 1-nitroso-3-methyl-5-methoxyindole (1-NO-MMIndole)
as a model compound. The Z-isomer and E-isomer of
NOMMIndole are almost isoenergetic (data not shown).
We searched for the existence of a protonated nitrosoin-
dolic intermediate, optimizing various initial structures;
the most relevant ones are depicted in Figure 3. In Table
2 there are some selected bond distances for these
intermediates.
Following the changes in the Mulliken populations over
the NO moiety and the N-O distance, as the X-NO
distance increases the NO moiety acquires more NO+
character (X ) N, C; see Table 2). The intermediates 3-5
have the NO+ bound to the aromatic electron cloud as a
sort of π-complex, as has already been proposed.19
The stability of these intermediates was determined
by estimating the free energy changes at 37 °C for the
protonation process (eq 5, Table 3). As expected, solvation
effects are very important for the charged species. On
S-Nitrosocysteine (NOCys) was identified as a reaction
intermediatesfinally producing cystineswith a maxi-
mum in the absorption UV-vis spectra at 544 nm
(19) Castro, A.; Iglesias, E.; Leis, J. R.; Pena, M. E.; Tato, J. V.;
Williams, D. L. H. J. Chem. Soc., Perkin Trans. 2 1986, No. 8, 1165-
1168.
J. Org. Chem, Vol. 70, No. 15, 2005 5793

De Biase et al.
FIGURE 3. Optimized structures of the protonated intermediates: 1, 1-NO-3-H-MMIndole+; 2, 1-NO-2-H-MMIndole+; 3, 3-NO-
1-H-MMIndole+; 4, 2-NO-1-H-MMIndole+; 5, 1-NO-1-H-MMIndole+; 6, 1-NO-N-H-MMIndole+; 7, 1-NO-O-H-MMIndole+.
TABLE 3. Computed DFT Energy and Free Energy
Changes (kcal/mol) of the Protonation Process
posed with a NOMel absorption band. On the other hand,
NOMel is completely converted to melatonin as measured
by 1H NMR (Figure SI2 in the Supporting Information).
EDTA must be added to the reaction vessel to avoid
∆E° + ZPE
(vacuum,
0 K)
∆G°
(vacuum,
37 °C)
∆G°
(water,
37 °C)
HNOMMIndole+
NOCys fast decomposition catalyzed by Cu2+ 23
. When the
1
2
3
4
5
6
7
-33.4
-40.7
-46.9
-47.3
-43.5
-34.7
-45.9
-33.5
-41.1
-47.0
-47.8
-43.4
-33.7
-44.9
25.2
13.0
10.7
11.6
16.8
26.4
19.8
EDTA concentration is decreased, a faster decay of the
NOCys absorption band can be observed (Figure SI3 in
the Supporting Information). At higher thiol concentra-
tions, the formed NOCys quickly disappears due to a
secondary reaction in which it reacts with cysteine,
producing cystine (Figure SI2) and various nitrogenated
products (N2O, NH2OH, and NH3).24,25
Considering our results, we propose that the reaction
involves transnitrosation between NOMel and Cys as has
recently been studied by Sonnenschein et al.:13
TABLE 4. Experimental pKa Values for Some Indole
Radicals and Calculated pKa Value for MMIndole•
radical
indole
melatonin
pKa
radical
pKa
4.6a
tryptophan
MMIndole
4.3a
NOMel + Cys f NOCys + Mel
(6)
4.5a
4.0b
a Experimental values from refs 20 and 21. b Calculated DFT
Cysteine has five different protonation species that can
exist in aqueous solution depending on the pH (Scheme
3).26 It has been proposed that the transnitrosation
reaction between NOCys and Cys involves nucleophilic
attack by the thiolate species 10 and 12.24,26 To find out
value from ∆G° of Table SI1, step b for MMIndole.
(ꢀ544 nm ) 14.9 M-1 cm-1 22
(Figure 5). The reaction was
)
carried out in Tris-HCl buffer at pH 6.4. Although
NOCys has another maximum at 335 nm, it is superim-
(23) Williams, D. L. H. Acc. Chem. Res. 1999, 32 (10), 869-876.
(24) Dicks, A. P.; Li, E.; Munro, A. P.; Swift, H. R.; Williams, D. L.
H. Can. J. Chem. 1998, 76 (6), 789-794.
(20) Mahal, H. S.; Sharma, H. S.; Mukherjee, T. Free Radical Biol.
Med. 1999, 26 (5-6), 557-565.
(21) Jovanovic, S. V.; Steenken, S. J. Phys. Chem. 1992, 96 (16),
6674-6679.
(25) Wong, P. S. Y.; Hyun, J.; Fukuto, J. M.; Shirota, F. N.;
DeMaster, E. G.; Shoeman, D. W.; Nagasawa, H. T. Biochemistry 1998,
37 (16), 5362-5371.
(22) Feelisch, M.; Stamler J. Preparation and detection of S-
nitrosothiols. In Methods in Nitric Oxide Research; Feelisch, M.,
Stamler J., Eds.; Wiley: Chichester, New York, 1996; pp 521-539.
(26) Patel, H. M. S.; Williams, D. L. H. J. Chem. Soc., Perkin Trans.
2 1990, No. 1, 37-42.
5794 J. Org. Chem., Vol. 70, No. 15, 2005

Mechanisms of NO Release by NOMel
SCHEME 2. Proposed Denitrosation Mechanism for 3-Substituted 1-Nitrosoindoles
which thiol species are involved in the reaction, we
studied the rate of reaction 6 as a function of pH. Cys
solutions were prepared in phosphate-buffered solutions
at pH 5.0, 7.4, 8.3, 9.9, and 12.4 at 37 °C; the NOMel:
Cys ratio was about 1:600.
As under these conditions the spontaneous decomposi-
tion of NOMel competes with transnitrosation, we also
measured the spontaneous decomposition of NOMel
under the same conditions without Cys (Figure SI4 in
the Supporting Information). The second-order rate
constant for the reaction was established as follows:
spontaneous decomposition constant at the same condi-
tions as kobs, and [Cys] corresponds to the total cysteine
concentration. The second-order rate constants (kr) are
plotted as a function of pH in Figure 6.
The reaction rate constant (kr) correlates with the
thiolate concentration [RS-] (Figure 6), consistently with
a reaction mechanism which involves nucleophilic attack
of the thiolate on the nitroso moiety of NOMel. This is
in agreement with previous results for S-transnitrosation
reactions between S-nitrosothiols and thiols26 in which
an intermediate formed by these two reactants has been
recently reported.29
The activation barrier for the reaction of Cys with
NOMel was determined by measuring kr at 28, 37, and
46.5 °C in 1 M phosphate buffer, pH 7.4. The activation
energy was found to be 22 ( 1 kcal/mol. We have obtained
similar results for the NOCys/Cys transnitrosation reac-
tion.29 This value is 6 kcal/mol higher than that reported
for NOTrp.13
kr ) (ko - kd)/[Cys]
(7)
where kr is the transnitrosation rate constant, kobs is the
observed pseudo-first-order rate constant, kd is the
It has been observed that, at a concentration of thiol
38-fold higher than 2 mM, kr decreases from 0.127 to
0.033 M-1 s-1 (37 °C, pH 7.4). We have tested the effect
of phosphate, Tris, and Cu2+ on this reaction, disregard-
ing all of them as responsible for this effect. Comparing
our kr with that obtained by Sonnenschein et al.,13 we
observe this same trend. In less concentrated Cys solu-
tions (20-fold lower) the kr determined for NOMel and
Cys was 57-fold higher (kr ) 7.2 ( 2.4 M-1 s-1, [Cys] )
100 µM, [NOMel] ) 100 µM).13 This rate constant must
be overestimated because NOMel spontaneous decay was
not considered significantly appreciable at this condition.
These results show that at increasing cysteine concentra-
tion the observed rate constant (ko ) kr[Cys] + kd) tends
to a limiting value. Although it could be proposed that
cysteine could be behaving as the above-mentioned
(27) Benesch, R. E.; Benesch, R. J. Am. Chem. Soc. 1955, 77 (22),
5877-5881.
(28) Reuben, D. M. E.; Bruice, T. C. J. Am. Chem. Soc. 1976, 98 (1),
114-121.
(29) Perissinotti, L. L.; Turjanski, A. G.; Estrin, D.; Doctorovich, F.
J. Am. Chem. Soc. 2005, 127 (2), 486-467.
FIGURE 4. Free energy profile in water at 37 °C from DFT
calculations.
J. Org. Chem, Vol. 70, No. 15, 2005 5795

De Biase et al.
FIGURE 5. UV-vis detection of S-nitrosocysteine at 544 nm (ꢀ544 nm ) 14.9 M-1 cm-1). Conditions: 32 °C, 0.15 M Tris-HCl
buffer, pH 6.4, [EDTA] ) 55.1 mM (Cu2+ chelant), [Cys] ) 13.0 mM, [NOMel] ) 9.0 mM. Inset: The visible spectrum for the
reaction (from 510 to 590 nm) is shown at three different times (122, 1232, and 2702 s).
SCHEME 3. Cysteine Protonation Species26
nucleophiles (Cl-, Br-, etc.), this cannot be the main
pathway since the maximum observed rate constant (ko)
for the NOMel-Cys reaction was found to be higher than
that for the first limiting step of the nucleophile pathway
(NOMel homolytic rupture, Scheme 2, step a; k1, eq 11
below) and also because at alkaline pH the reaction does
not decrease its rate as expected (Scheme 2, steps b-d).
Thus, cysteine could be exerting a direct nucleophilic
attack on NOMel in a reversible and nonconcerted way.
FIGURE 6. Relative concentration of cysteine thiolate species
(rct) and second-order rate constant values of the reaction of
NOMel with cysteine (kr) as a function of pH. rct ) [RS-]/[RS-]
+ [RSH]. [RS-] and [RSH] were obtained from the correspond-
ing equilibrium constants (see the Supporting Information).
Notes: a, ref 27; b, ref 28; c, fitted to data.
NOMel + Cys- + H+ T MelNOCys- +
H+ T MelHNOCys f NOCys + MelH
It has been proposed that indoles can compete with
thiols for NO, thus forming N-nitrosoindoles.5 Moreover,
we propose that N-nitrosoindoles can transnitrosate to
thiols as the common reaction between S-nitrosothiols
and thiols. In this sense N-nitrosoindoles could act as NO
carriers as well as S-nitrosothiols.
Reaction of NOMel with Reducing Agents. Previ-
ous studies suggest that the reaction of NOMel with
reducing agents releases NO.5,10 This reaction could have
great biological relevance at the time of explaining
guanilyl cyclase activation when a nitroso compound is
introduced into a biological system. At first, the reaction
of NOMel with NADH was tested, and NO concentration
was monitored with a specific electrode (Figure SI5 in
Therefore, the observed saturation could be attributed
to a later limiting step just as the break off of a formed
hypothetic NOMel-cysteine adduct, similar to the one
we have found for nitrosocysteine.29 We consider that this
issue should be further studied.
With the purpose of studying the thermodynamic
feasibility of this reaction, DFT calculations were made
at the B3LYP/6-31G** level. We employed two structural
analogues for NOMel and Cys, 1-nitroso-3-methyl-5-
methoxyindole and L-cysteine ethyl ester, respectively.
We found a standard free energy value at 37 °C in water
solution of -4.6 and -22.5 kcal/mol for the reaction with
the thiol and the thiolate, respectively, accordingly with
a previous work on sulfur-to-sulfur transnitrosation.30
(30) Wang, K.; Wen, Z.; Zhang, W.; Xian, M.; Cheng, J. P.; Wang,
P. G. Bioorg. Med. Chem. Lett. 2001, 11 (3), 433-436.
5796 J. Org. Chem., Vol. 70, No. 15, 2005

Mechanisms of NO Release by NOMel
TABLE 5. Effect of pH on the Observed Rate Constant
(kobs) for the Reaction of NOMel with Asc in 0.2 M
Tris-HCl-Buffered Solutions at 37 °Ca
pH
kobs (10-3 s-1
)
pH
kobs (10-3 s-1
)
7.0
7.4
7.7
1.27
1.12
1.06
8.4
9.2
1.05
0.85
a [NOMel]i ) 0.1 mM, [Asc]i ) 10 mM.
TABLE 6. Experimental Activation Parameters
Corresponding to the Homolytic Rupture of NOMel
∆Gq (kcal mol-1
∆Hq (kcal mol-1
∆Sq (cal mol-1 K-1
)
)
22.51 ( 0.14
27.56 ( 0.02
16.49 ( 0.07
)
FIGURE 7. Dependence of ascorbate concentration on the
observed rate constant (kobs) for the reaction of NOMel with
Asc ([NOMel]i ) 0.13 mM, 37 °C, pH 7.4): points, experimental
data; curve, values fitted according to eqs 11 and 12.
that derives from a constant concentration of a NOMel
ethanolic solution in equilibrium with NO; as in methanol
the decomposition rate in ethanol is not appreciable.
Equation 11 adjusts reasonably well to the experimen-
tal data (Figure 7), with k1 ) (1.08 ( 0.04) × 10-3 s-1
and a ) (1.0 ( 0.1) × 10-4 M. To support this considera-
tion, we studied the pH dependence of the reaction at
the saturation level, measuring kobs between pH 7 and
pH 9, finding that kobs has a very small pH dependency
(Table 5).
This information suggests that an Asc species whose
concentration does not change with pH (between pH 7
and pH 9) could be participating. Asc pKa values31 show
that ascorbate (Asc-) is the most relevant and the only
pH-independent species of all Asc species (Asc, Asc-,
Asc2-) at these pHs. So we can modify eq 10 likewise, as
depicted in Scheme 2, path e (DAsc•-‚ ) unprotonated
monodehydroascorbate radical):
the Supporting Information). This result showed that the
reaction quickly releases NO. Similar results were ob-
tained with another reducing agent, ascorbic acid (Asc),
following first-order kinetics with respect to NOMel (eq
8, where v0 is the initial rate). Figure 7 shows kobs
kobs ) v0/[NOMel]
(8)
dependence on [Asc]T (total ascorbic acid molar concen-
tration), not observed in a previous study.10 It can be seen
that, starting from [Asc]T ≈ 10-3 M (100:1 ratio with
respect to [NOMel]), kobs remains practically constant,
suggesting that NOMel does not react directly with Asc;
otherwise the reaction would not be zeroth-order in [Asc]T
at high concentrations.
A mechanism that could explain the observed depen-
dence involves homolytic rupture of NOMel as the first
step (eq 9, as suggested above), and subsequent reaction
of Asc with Mel• (eq 10)
Mel• + Asc- f MelH + DAsc•-
(10′)
To verify that NOMel homolytic rupture is the limiting
step, we measured kobs under saturation conditions
([NOMel] ) 0.1 mM, [Asc]T ) 4.3 mM) at different
temperatures (26, 31, 37, and 42 °C). The experimental
activation parameters obtained from these results are
shown in Table 6. The experimental activation enthalpy
for this reaction agrees very well with the DFT-estimated
value of 27.6 kcal mol-1 for NOMMIndole. This result is
consistent with the fact that NOMel homolytic rupture
is the limiting step of this reaction.
With the purpose of analyzing the thermodynamical
feasibility of the reaction, we performed DFT calculations
for the global reaction, using 3-methyl-5-methoxyindole
as a model compound (1-NO-MMIndole ) 1-nitroso-3-
methyl-5-methoxyindole, 1-H-MMIndole ) 3-methyl-5-
methoxyindole):
NOMel T NO + Mel•
(9)
Mel• + Asc f MelH + DAscH•
(10)
(DAscH• ) monodehydroascorbate radical). DAscH• could
further react with Mel•, producing melatonin and dehy-
droascorbate (DAsc). Anyhow, the destiny of this inter-
mediate will not be considered, since its concentration is
low with respect to [Asc]T at the initial reaction rates
measured.
Considering that other decomposition pathways are
very slow compared with this reaction (a valid assump-
tion taking into account [Asc]T), and using the steady-
state approximation for [Mel•], we can obtain the follow-
ing expressions:
2(1-NO-MMIndole) + Asc- + H+ f
2NO + 2(1-H-MMIndole) + DAsc (13)
kobs ) k1/(1 + a/[Asc]T)
a ) k-1[NO]i/k2
(11)
(12)
The 2:1 stoichiometry of the global reaction was deter-
mined experimentally by measuring NOMel and Asc
concentrations spectrophotometrically. The free energy
value for this reaction (eq 13) in water at 37 °C (-11.9
kcal/mol) points out that NOMel reduction at this condi-
where k1 and k-1 are the rate constants for the direct
and reverse step 9, respectively, k2 is the rate constant
for step 10, and [NO]i is the nitric oxide concentration at
the initial state of the reaction. We assume this amount
to be constant because it corresponds to the NO amount
(31) Lide, D. R. HandBook of Chemistry and Physics (1913-1995),
75th ed.; CRC Press Inc.: Boca Raton, FL, 1994.
J. Org. Chem, Vol. 70, No. 15, 2005 5797

De Biase et al.
TABLE 7. Redox Free Energies and Potentials of Some Relevant Species at pH 7 vs NHE (Normal Hydrogen Electrode)
hemireaction
∆G° (kcal/mol)
E° (mV)
hemireaction
∆G° (kcal/mol)
E° (mV)
NOMel/NOMel-
NOIndole/NOIndole-
cystine/cysteine
NAD+/NADH
15.7
13.5
7.8
14.8
-3.7
-680a
-587b
-340
-320
+80
Cu2+/Cu+
-3.7
-13.0
-7.5
+161
+281
+327b
+360
O2/H2O2
indole•/indole-
NO2-/NO
-8.3
DAsc/Asc
a This work. b Reference 32 (measured in acetonitrile). All others from refs 33-37.
TABLE 8. Estimated Half-Lives for the Reactions of
concentrations (Table 7). However, the indole radical has
a high enough redox free energy to accept an electron
from any of these reducing agents. Considering the above-
mentioned results, homolytic rupture of NOMel followed
by reduction (Scheme 2, paths a and e) seems to be the
most likely pathway for NO release by reaction with
reducing agents.
Under conditions compatible with biological media, it
is likely that NOMel will mainly react with reducing
agents, releasing NO (Table 8). Which of these reactions
predominate will depend on the relative local concentra-
tion of these reactants. Taking into account the known
concentrations of Asc34 and cysteine19 in plasma, the
reaction with Asc would be the predominant pathway.
A comprehensive mechanism for NO release by NOMel
in physiological media involves initial homolytic rupture
of NOMel (Scheme 2, path a) followed by reduction (path
e), oxidation (path f), or proton-aided NO translocation
followed by nucleophilic attack (paths b-d). The sug-
gested mechanism explains all our experimental and
theoretical results, and could be in principle extended to
other N-nitrosoindoles such as NOTrp.
NOMel with Various Agents at pH 7.4 and 37 °C
plasma
concn
(µM)
expected rate NOMel
constant
half-life
(min)
NOMel reaction
(kobs)a (s-1
)
reaction with ascorbate
spontaneous decomposition
reaction with cysteine
50b
3.6 × 10-4
32
312
9096
3.7 × 10-5
10 (ref 13) 1.27 × 10-6 c
a Considering that ascorbate and cysteine remain constant
during the reaction. b Although ascorbic acid circulates in human
plasma at approximately 30-50 µM, it accumulates in millimolar
concentrations in host defense cells. Mononuclear leukocytes, for
example, may have intracellular ascorbic acid concentrations of
3.5-6 mM.38 In those cases kobs is almost saturated, and the half-
life is limited to 11 min. c Calculated using our second-order rate
constant kr ) 0.127 ( 0.002 M-1 s-1 at 37 °C, pH 7.4.
tion is thermodynamically feasible (see the Supporting
Information). Other reducing agents such as H2O2 (Table
SI2 in the Supporting Information) and Cu+ (data not
shown) also react with NOMel, although the mechanism
of these reactions might be different than in the case of
Asc.
Acknowledgment. Thanks are due to ANPCyT,
Antorchas, CONICET, and UBA for financial support.
Thanks are also due to Laura L. Perissinotti for the
provided data.
Conclusions
We consider that, under the experimental conditions
of our work, direct NO+ transfer from NOMel was not
compatible with our results except for the case of the
catalytic hydrolysis of NOMel by phosphate where NO+
is probably transferred directly to phosphate. In the other
reactions we discarded the direct NO+ transfer by NOMel
because it would involve the formation of melatonin anion
(Mel-) that is significantly unstable at the pH conditions
used (pKa(Indol-) ) 20.9).32 In the case of NOMel-
phosphate, concerted exchange of NO+ and H+ may be
involved, avoiding the formation of Mel-. The nitrosyl
phosphate formed would be rapidly hydrolyzed.
Supporting Information Available: DFT energetic val-
ues, selected NMR spectra, kinetic data, rate law derivation,
and geometries. This material is available free of charge via
the Internet at http://pubs.acs.org.
JO047720Z
(33) Stryer, L. Biochemistry, 3rd ed.; Reverte: Barcelona, 1988.
(34) Brock, T. D.; Madigan, M. T.; Martinko, J. M. Biology of
Microorganisms, 7th ed.; Prentice Hall: Englewood Cliffs, NJ, 1994.
(35) Harris. Quantitative Chemical Analysis, 4th ed.; W. H. Freeman
& Company: San Francisco, CA, 1995.
(36) Nakamura, K.; Aizawa, M.; Miyawaki, O. Electroenzymology.
Coenzyme regeneration; Springer-Verlag: Berlin, 1998; Vol. 4.
(37) Solomon, E. I.; Baldwin, M. J.; Lowery, M. D. Chem. Rev. 1992,
92 (4), 521-542.
(38) Carcamo, J. M.; Pedraza, A.; Borquez-Ojeda, O.; Golde, D. W.
Biochemistry 2002, 41 (43), 12995-13002.
Direct electron transfer from NADH, Asc, H2O2, or Cu+
to NOMel is not thermodynamically feasible at standard
(32) Zhu, X. Q.; He, J. Q.; Li, Q.; Xian, M.; Lu, J. M.; Cheng, J. P.
J. Org. Chem. 2000, 65 (20), 6729-6735.
5798 J. Org. Chem., Vol. 70, No. 15, 2005