Reaction of vitamin e compounds with n-nitrosated tryptophan derivatives and its analytical use

Article abstract of DOI:10.1002/chem.200700395

We recently showed that nitrosated tryptophan residues may act as endogenous nitric oxide storage compounds. Here, a novel reaction of nitrosotryptophan derivatives is described, in the form of the release of nitric oxide from N-nitrosotryptophan derivatives initiated either by a-tocopherol or by its water-soluble form trolox. α-Tocopherol and trolox were found to release stoichiometric amounts of nitric oxide from N-acetylN-nitrosotryptophan as well as from the nitrosotryptophan residue in albumin. The reaction proceeds both in water and in lipophilic solution and reconstitutes the indole moiety of the tryptophan molecule quantitatively. During this reaction, α-tocopherol- and trolox-derived phenoxyl-type radicals were identified as intermediates by ESR spectrometry. The chemical mechanism of the NO-releasing process was established. Since S-nitrosothiols donot react under the applied conditions, it is suggested that the trolox-dependent release of nitric oxide may be utilizable for the detection of N-nitrosotryptophan residues in biological samples. Furthermore, as N-nitrosotryptophan derivatives do not undergo spontaneous decay in lipophilic environments, vitamin E may have the so far unrecognized function of preventing the accumulation of N-nitrosotryptophan residues to toxic concentrations in biological systems.

Full text of DOI:10.1002/chem.200700395

DOI: 10.1002/chem.200700395
Reaction of Vitamin E Compounds with N-Nitrosated Tryptophan
Derivatives and Its Analytical Use
Karsten Müller,[a] Hans-Gert Korth,[b] Herbert de Groot,[a] and Michael Kirsch*[a]
Abstract: We recently showed that ni-
trosated tryptophan residues may act
as endogenous nitric oxide storage
compounds. Here, a novel reaction of
nitrosotryptophan derivatives is de-
scribed, in the form of the release of
nitric oxide from N-nitrosotryptophan
derivatives initiated either by a-toco-
pherol or by its water-soluble form
trolox. a-Tocopherol and trolox were
min. The reaction proceeds both in
water and in lipophilic solution and re-
constitutes the indole moiety of the
tryptophan molecule quantitatively.
During this reaction, a-tocopherol- and
trolox-derived phenoxyl-type radicals
were identified as intermediates by
ESR spectrometry. The chemical mech-
anism of the NO-releasing process was
established. Since S-nitrosothiols do
not react under the applied conditions,
it is suggested that the trolox-depen-
dent release of nitric oxide may be uti-
lizable for the detection of N-nitroso-
tryptophan residues in biological sam-
ples. Furthermore, as N-nitrosotrypto-
phan derivatives do not undergo spon-
taneous
decay
in
lipophilic
environments, vitamin E may have the
so far unrecognized function of pre-
venting the accumulation of N-nitroso-
tryptophan residues to toxic concentra-
tions in biological systems.
found
to
release
stoichiometric
amounts of nitric oxide from N-acetyl-
N-nitrosotryptophan as well as from
the nitrosotryptophan residue in albu-
Keywords: nitric oxide · nitroso-
thiol · tryptophan · vitamins
Introduction
troso residues in albumin[4] have also been investigated in
À
this context. Zhang et al.[5] compared the NO2 /HCl-depen-
In the human body, endogenous nitric oxide reacts rapidly
with a variety of metalloproteins, preferably with heme en-
zymes, and may in this way be inactivated.[1] In order to
maintain its physiological functions under such conditions,
dent nitrosation of bovine serum albumin with the nitrosa-
tion of carboxymethyl–bovine serum albumin (CM-BSA)
and found a regiospecific nitrosation at Trp214 in CM-BSA.
In physiological environments, different modes of nitrosa-
tion have been described in the literature. N2O3, generated
C [6,7]
C
NO is believed to be protected by the formation of so-
C
called “storage compounds”, which may act as NO donors
by autoxidation of NO ,
rapidly nitrosates thiols/thio-
when needed. Accordingly, interest in nitric oxide storage
compounds has strongly increased in recent years.
Various forms of storage compounds that may prevent the
lates,[8] as well as tryptophan derivatives.[9] A peroxynitrite-
driven nitrosation of thiols[10] and N-terminal-blocked tryp-
tophan derivatives has also been reported.[11–13]
C
artificial scavenging of NO in intra- and extracellular mi-
In aqueous solution around pH 7, N-nitroso derivatives
undergo slow hydrolysis to give the parent amine and nitrite
in a proton-catalysed process, followed by nucleophilic
attack of OHÀ/H2O(Scheme 1). [14]
lieux have been proposed. Beside low-molecular-weight S-
nitroso compounds, such as GSNO,
[2,3]
S-nitroso and N-ni-
In accord, the rate of hydrolysis is accelerated in the pres-
ence of various nucleophiles YÀ, which in essence represents
a transnitrosation reaction. Nitroso compounds (YNO) from
halide or pseudohalide nucleophiles are hydrolysed rapid-
ly,[15] but formation of hydrolytically more stable transnitro-
sation products is well known: for example, for n-butanol
(nBuONO),[16] H2O2 (ONOOH)[17] or thiols (RSNO).[18]
[a] K. Müller, Prof. Dr. H. de Groot, Priv.-Doz. Dr. M. Kirsch
Institut für Physiologische Chemie, Universitätsklinikum Essen
Hufelandstrasse 55, 45122 Essen (Germany)
Fax. (+49)201-723-5943
E-mail: michael.kirsch@uni-essen.de
[b] Dr. H.-G. Korth
Institut für Organische Chemie, Universität Duisburg-Essen
Universitätsstrasse 5, 45117 Essen (Germany)
C
However, in cases in which a stabilized radical Y can be
produced, the intermediate YNOmay also undergo facile
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
C
homolysis to release NO [Eq. (1)].
7532
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7532 – 7542

FULL PAPER
ence of a-tocopherol, the decay of NANT was accelerated
by a factor of 4.5 with an apparent rate constant of kobs
=
(8.4Æ0.1)”10À5 sÀ1 (Figure 1), thus confirming an a-toco-
Scheme 1. Proton-catalyzed and nucleophile-accelerated hydrolysis of N-
nitrosamines.
Further, for sufficiently reactive and sterically unhindered
C
radicals Y , the decay of the N-nitroso compound might
C
then be further accelerated by attack of Y on the nitroso
group [Eq. (2)].
Figure 1. Decay of NANT (1 mm) in ethyl acetate, monitored spectropho-
tometrically at 335 nm in the absence (open symbols) and in the presence
(closed symbols) of a-tocopherol (1 mm) at 378C. Each value represents
the average (Æs.d.) of at least three independent runs.
pherol-induced decomposition of NANT. During this reac-
tion, a-tocopheryl radicals and nitric oxide are being pro-
duced, as indicated by ESR experiments (see below).
This has recently been demonstrated to be the case for
YÀ =ascorbate and certain phenolic compounds (that is, cat-
echol derivatives).[19,20]
Since a-tocopherol is a lipophilic compound, the influence
of the polarity of the environment on the rate of the a-toco-
pherol–NANT reaction (100 mm each) was analysed by em-
ploying various organic solvents with a range of dielectric
constants. However, the decay rate of NANT was only
weakly affected by the polarity (dielectric constant) of the
solvent, varying inversely by just a factor of two between
ethyl acetate (e=6.4) and dimethyl sulfoxide (e=46.7)
(Table S1, Supporting Information). This observation clearly
argues against the formation of charged intermediates in the
a-tocopherol-mediated decomposition of NANT.
Among the naturally occurring phenolic compounds, vita-
min E plays a major role. a-Tocopherol is the major constit-
uent of natural vitamin E and is the most important lipid-
soluble antioxidant. It is essential for the prevention of lipid
peroxidation[21,22] (termination of radical chain reactions)
and for cellular signalling.[23] It has been reported that vita-
min E additionally inhibits smooth muscle cell prolifera-
tion,[24,25] decreases protein kinase C activity[23] and increases
[26]
C
NO levels through regulation of eNOS.
Here we demonstrate that a-tocopherol and its water-
In order to investigate the foregoing reaction in aqueous
solution, we employed the water-soluble a-tocopherol ana-
logue trolox [(R)-6-methoxy-2,5,7,8-tetramethylchromane-2-
carboxylic acid]. As mentioned in the Introduction, NANT
undergoes slow, proton-catalysed hydrolysis in aqueous solu-
tion to give exclusively NAT and nitrite. In previous
papers,[17,19,20,27–30] this process was reported to be of first
order, and from the data of Meyer et al.[14] an apparent
room-temperature rate constant of kobs =(6.6Æ0.1)”10À5 sÀ1
at pH 7.0 can be extrapolated. This value was reproduced
well in this study [kobs =(6.1Æ0.1)”10À5 sÀ1]. However, close
inspection of the decay traces revealed that the decomposi-
tion of NANT—both in the absence and in the presence of
trolox—could not satisfactorily be described by a first-order
rate law, nor by a second-order rate law (Figure 2). Excel-
lent empirical fits (r2 ꢀ0.998 in all cases) were achieved by
assuming a (minor) contribution of a linear (apparent
zeroth order) change in absorbance to a first-order decay. In
the absence of trolox, the kinetic analysis gave a ca. 50%
lower first-order rate constant for the dominant first-order
C
soluble analogue trolox release NO quantitatively from N-
terminal blocked N-nitrosotryptophan derivatives, not only
from the isolated molecules but also from N-nitrosotrypto-
phan residues in proteins. This reaction is not catalysed by
heavy metal ions and occurs both in aqueous and in nonaqu-
eous solution.
Results
Kinetics of the a-tocopherol/trolox–NANT reaction: In
order to corroborate the existence of a reaction between a-
tocopherol and N-acetyl-N-nitrosotryptophan (NANT), the
decay of NANT (1 mm) in lipophilic medium (ethyl acetate),
was monitored spectrophotometrically at 335 nm and 378C
in the absence and in the presence of a-tocopherol (1 mm).
In the absence of a-tocopherol, the concentration of NANT
decreased by 7% within 60 min, corresponding to a first-
order rate constant of kobs =(1.9Æ0.1)”10À5 sÀ1. In the pres-
Chem. Eur. J. 2007, 13, 7532 – 7542
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7533

M. Kirsch et al.
Figure 2. Decay of NANT (100 mm) in phosphate buffer at pH 7.4 and 11,
monitored spectrophotometrically at 335 nm in the absence (open sym-
bols) and in the presence (closed symbols) of trolox (100 mm) at 25 and
378C, respectively. Each value represents the average (Æs.d.) of at least
three independent runs. Continuous lines are empirical fits to a superpo-
sition of a first- and zeroth-order rate law.
Figure 3. Dependence of the apparent first-order rate constants for decay
of NANT (100 mm) on the concentration of trolox in phosphate buffer at
pH 7.4 and 11, T=258C. Each value represents the average (Æs.d.) of at
least three independent experiments. The continuous lines are empirical
fits to a hyperbolic function.
pected, preformed products. Inspection of Figure 4a,b (only
carbonyl region shown) reveals the formation of the trolox-
298 K
app
process at pH 7.4 and 258C [k
=(3.1Æ0.6)”10À5 sÀ1]. At
310 K
app
378C, the decay is about two times faster [k
=(6.5Æ
0.6)”10À5 sÀ1]. In accord with the reported proton catalysis
of hydrolysis,[14,27] NANT was found to be more stable at
pH 11, its concentration decreasing by only 5% within
310 K
app
60 min with a rate constant of k
=(8.0Æ1.1)”10À6 sÀ1 at
378C (Figure 2), in line with our previously reported data.[30]
(At this temperature, evaluation in terms solely of a first-
order process gave a slightly higher kobs of (1.2Æ0.1)”
10À5 sÀ1.) At pH 7.4 and 258C, addition of trolox (100 mm)
298 K
app
accelerated the decay of NANT by a factor of five [k
=
(1.5Æ0.1)”10À4 sÀ1]. At these concentrations, the decay at
310 K
app
378C was about 1.5 times faster [k
=(2.1Æ0.3)”
10À4 sÀ1]. Notably, and differently to what was observed in
the absence of trolox, decay of NANT in the presence of an
equimolar amount of trolox was further increased at pH 11
=(5.3Æ0.1”10À4 sÀ1], to give a final concentration of
310 K
app
[k
25.2Æ5.1 mm after 60 min (Figure 2).
For a fixed concentration of NANT (100 mm), the depen-
dence of the decay rate constants on the concentration of
trolox (0–200 mm) was determined at 258C at pH values 7.4
and 11 (Figure 3). (The data for Figure 3 were evaluated by
taking a minor contribution from a zeroth-order decay into
account. However, evaluation in terms solely of a first-order
process gave very similar values, with less than 20% differ-
ence.) As is evident from Figure 3, the decay rate constants
increased nonlinearly (apparently hyperbolically) with in-
creasing trolox concentration, to approach limiting values of
about 3.5”10À4 sÀ1 at pH 11 and 1.5”10À4 sÀ1 at pH 7.4 for
trolox concentrations greater than 200 mm. Hence, excess
trolox accelerates the decay of NANT about 50-fold at
pH 11, but only an about five times rate enhancement is ob-
served at pH 7.4.
Figure 4. 13C NMR spectra (carbonyl region) from: A) reaction of trolox
(100 mm) with NANT (100 mm), and B) reaction of trolox (100 mm) with
Fremyꢁs salt (100 mm). Spectra were recorded in a mixture (80:20 v/v) of
[D6]DMSOand phosphate buffer (50 m m, pH 7.4) after a reaction period
of 18 h at room temperature. Spectral assignments: a) trolox quinone,
b) trolox, c) NAT, and d) NANT.
Product analysis: The trolox–NANT reaction mixtures were
analysed by 13C NMR spectrometry and comparison with ex-
7534
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7532 – 7542

Vitamin E Compounds
FULL PAPER
derived benzoquinone (d=188.3, 187.6, 179.3 ppm) and of
NAT (d=175.2, 171.5 ppm) as final products from the
trolox–NANT reaction. No other products could be detected
(see Table S2, Supporting Information for a complete listing
of the resonances).
The yields of NAT and the fraction of unconverted
NANT from the NANT–trolox (90 and 100 mm, respectively)
reaction were quantified in independent experiments. As
demonstrated in Figure 5, NAT was stoichiometrically pro-
porarily) stationary concentrations. This assumption was
confirmed by monitoring the reactions between NANT and
trolox or a-tocopherol by ESR spectrometry. The character-
istic ESR patterns of the trolox radical and the a-tocopheryl
radical could be instantaneously detected after mixing of the
reactants (Figures 6, 7).
Figure 6. ESR spectrum of the trolox phenoxyl radical produced from the
reaction between trolox (7.5 mm) and N-acetyl-N-nitrosotryptophan
(7.5 mm) in deoxygenated phosphate buffer (50 mm, pH 7.4, 188C):
a) prior to addition of NANT, b) 10 min, and c) 24 min after addition of
NANT. All spectra were recorded with identical instrument settings.
Figure 5. Decomposition of NANT and production of NAT during the
trolox–NANT reaction. Decomposition of NANT (90 mm) was read at
335 nm after addition of trolox (100 mm) in phosphate buffer (50 mm,
pH 11, 258C). NAT production was fluorimetrically determined at lem
=
355 nm (lexc =270 nm) from a 10-fold diluted aliquot of the reaction mix-
ture. Each value represents the mean (Æs.d.) of at least three indepen-
dent experiments.
duced at the expense of NANT, because the sum of NAT
and NANT during the reaction was constant at 88.8Æ
1.0 mm. After a reaction period of 60 min, 59.9Æ2.7 mm
NANT had been decomposed, thereby yielding 62.4Æ4.7 mm
NAT. Thus, NANT was quantitatively converted into NAT.
ESR measurements: As depicted in Equation (1), it was
proposed that the transnitrosation products YNO(Y
=
C
trolox, a-tocopherol) existed in equilibrium with NO and
the corresponding phenoxyl-type radicals. Since the latter
radicals are known to decay by disproportionation to the
parent phenol and the related p-quinone and o-quinone me-
thide, respectively, with rate constants of k3 =1180mÀ1 sÀ1 at
378C[31] (Scheme 2), they would be expected to attain (tem-
Figure 7. ESR spectrum of the a-tocopheryl radical produced from the
reaction between a-tocopherol (7.5 mm) and N-acetyl-N-nitrosotrypto-
phan (7.5 mm) in deoxygenated ethyl acetate at 188C: a) 2 min, and
b) 27 min after mixing of the reactants. c) Spectrum from (b) after 1 min
purging of the sample with argon. All spectra were recorded with identi-
cal instrument settings.
The good hyperfine resolu-
tions of the initial spectra are
increasingly lost with reaction
duration (compare Figures 6b,c
and 7a,b). This observation can
easily be attributed to the accu-
mulation of released nitric
oxide, the spin exchange inter-
action of which with the phe-
noxyl-type radicals leads to the
well known phenomenon of
Scheme 2. Disproportionation of the phenoxyl-type trolox radical in aqueous solution (top) and of the a-toco-
pheryl radical in nonaqueous solution (bottom).
line broadening.[32] This inter-
Chem. Eur. J. 2007, 13, 7532 – 7542
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7535

M. Kirsch et al.
pretation was confirmed by the recording of a well resolved
C
spectrum of the a-tocopheryl radical after removal of NO
by briefly flushing the sample with argon (Figure 7c). Be-
C
cause of the continued production of NO , this spectrum was
again subject to line broadening within the next four mi-
nutes (not shown). Note that the integrated intensities of
the spectra shown in Figure 6b,c and 7a–c were found to be
identical within the error limits of the ESR experiment, con-
firming the above proposal of the attainment of steady-state
levels of the radicals.
Nitric oxide release: The release of nitric oxide from the
C
trolox–NANT reaction was first monitored with a NO -sensi-
tive electrode. These measurements were carried out in an
Figure 9. Steady-state concentrations of nitric oxide from the reactions
between N-acetyl-N-nitrosotryptophan and trolox (closed circles) or a-to-
C
open system, so the NO level after mixing of the reactants
readily advances to a steady state, controlled by the rate of
C
copherol (open circles). NO steady-state concentrations were measured
C
NO production versus diffusion into the gas phase and
C
at 378C with a NO electrode after addition of NANT (5–100 mm) to
decay by autoxidation (Figure 8).
fixed concentrations either of trolox or of a-tocopherol (100 mm each) in
phosphate buffer (50 mm, pH 7.4) or 50% ethanol/phosphate buffer.
Each value represents the mean (Æs.d.) of at least three independent ex-
periments.
C
Figure 8. Trolox-induced NO -release from N-acetyl-N-nitrosotryptophan.
C
The steady state concentration of liberated NO was monitored with a
C
NO -sensitive electrode after addition of NANT (10 mm) to trolox
(100 mm) in phosphate buffer (50 mm, pH 7.4, 258C).
Figure 10. Quantification of nitric oxide release from the reaction be-
tween N-acetyl-N-nitrosotryptophan and trolox (100 mm each) in oxygen-
At a fixed trolox concentration of 100 mm, the steady-state
C
free phosphate buffer (50 mm, pH 7.4, 378C). NO was trapped by
C
NO concentration increased apparently exponentially with
FNOCT-4 (100 mm) and FNOCT-4-NOH was quantified fluorimetrically
at lem =460 nm (lexc =320 nm). Simultaneously, residual NANT was
monitored spectrophotometrically at 335 nm. Each value represents the
mean (Æs.d.) of at least three independent experiments.
increasing NANT concentration, to level off at about 2.5 mm
(Figure 9).
Similar experiments were performed with a-tocopherol in
an ethanol/phosphate buffer mixture (1:1). Here, a steady-
state concentration of about 1.2 mm was achieved at equimo-
lar concentrations of the reactants (Figure 9): that is, only
about 50% relative to the trolox–NANT reaction in pure
buffer. In any case, nitric oxide was unequivocally confirmed
as a product in aqueous solution.
each) at 378C yielded FNOCT-4-NOH (57.9Æ6.6 mm) after
a reaction period of 60 min. A similar amount of NANT
(62.0Æ11.2 mm) was consumed during the reaction. The sum
C
of residual NANT and produced NO amounted to about
In order to quantify the total yield of nitric oxide from
the trolox/a-tocopherol–NANT reaction, the FNOCT-4
assay was applied.[33] In contrast with the electrochemical
(NO-electrode) method, which only measures the momenta-
100Æ5 mm for the whole reaction period. Thus, the trolox–
C
NANT reaction produces NAT and NO in a stoichiometric
1:1 ratio. It is worth noting that no increase in the fluores-
cence of FNOCT-4-NOH exceeding the basal decay[33] was
observed from NANT in the absence of trolox. This obser-
vation clearly rules out the possibility that decomposition of
NANT in the presence of reductants proceeds through ini-
C
ry concentration of liberated NO , the FNOCT assay is an
integrating method (i.e., it scavenges all nitric oxide and
yields, after a rapid reduction step, stoichiometric amounts
of the fluorescent dye FNOCT-4-NOH). As shown in
Figure 10, the reaction between NANT and trolox (100 mm
À
tial homolysis of the N NObond, as has been proposed by
de Biase et al.[29] According to the characteristics of the
7536
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7532 – 7542

Vitamin E Compounds
FULL PAPER
C
Fischer–Ingold persistent radical effect (i.e., transient and
persistent radicals from the same precursor dominantly yield
the cross-reaction product),[34] decay of NANT through ini-
Quantification of the NO production from the trolox–
NO-albumin reaction was carried out by the FNOCT
method by addition of varying amounts of NO-albumin (0–
100 mm) and trolox (400 mm) to a solution of FNOCT-4
(100 mm) in phosphate buffer (pH 7.4, 258C). The excess of
À
tial N NObond homolysis should not be observed in the
C
absence of an efficient radical trap for NO (persistent radi-
C
cal) and/or the simultaneously produced NANT aminyl radi-
cal (transient radical). Therefore, stoichiometric and essen-
tially quantitative production of both nitric oxide and NAT,
trolox was employed not only to enhance the rate of NO re-
lease, but also to ensure rapid and complete reduction of
the initial FNOCT-4-nitric oxide adduct (a nitroxide radical)
to the fluorescent dye FNOCT-4-NOH.[33] The build-up of
the FNOCT-4-NOH fluorescence followed a clean first-
À
as observed here, contradicts homolytic cleavage of the N
NObond of NANT as the initial step.
C
In addition, the apparent first-order rate constant of
NANT decomposition [kapp =(2.1Æ0.3)”10À4 sÀ1] (see
above) corresponds well with the pseudo-first-order rate
order process; initial rates of NO production were evaluat-
ed from the first 300 s of fluorescence growth. Constant
fluorescence intensities of FNOCT-4-NOH were achieved
after a reaction period of 90 min, indicating completion of
constant of NO production [kNO =(3.3Æ0.3)”10À4 sÀ1], as
C
C
C
extracted from the trace of Figure 10. These data exclude
NO release. The final NO concentration, as well as the ini-
C
C
any significant NO -consuming pathways, differently to what
tial rates of NO production, both showed a linear depend-
has been observed for the ascorbate–NANT reaction.[20]
ence on the nitrosoalbumin concentration (Figure 12), con-
firming that the reaction is of first order in nitrosoalbumin.
With allowance for the 82.5% degree of nitrosation of albu-
a-Tocopherol–nitrosoalbumin reaction: In order to demon-
C
C
strate that the trolox-/a-tocopherol-mediated release of NO
min, a yield of 89.5Æ7.2% of NO was evaluated from the
is not restricted to the model compound NANT, additional
experiments were performed with N-nitrosated protein albu-
min. Only one tryptophan residue is present in albumin,
thus allowing straightforward quantification. The degree of
nitrosation of albumin was determined spectrophotometri-
cally to be 82.5%. Varying concentrations of trolox (5–
50 mm) were added to nitrosated albumin (100 mm) in phos-
regression line.
C
phate buffer (pH 7.4, 378C), and the release of NO was
C
monitored with the NO electrode. Trolox-induced liberation
C
of NO from the N-nitrosated tryptophan residue of the pro-
tein indeed took place (Figure 11). In analogy with the data
obtained from NANT, the steady-state concentration of re-
C
leased NO approached a constant value with increasing
C
concentrations of trolox (ca. 1.1 mm NO at ꢀ50 mm trolox).
C
Figure 12. Quantification and initial rates of NO release from the reac-
tion between trolox and N-nitrosated-albumin. N-Nitroso-albumin (25–
100 mm; 82.5% nitrosation) was added to
a mixture of FNOCT-4
(100 mm) and trolox (400 mm) in phosphate buffer (50 mm, pH 7.4, 258C),
and the fluorescence of FNOCT-4-NOH was monitored at lem =460 nm
C
(lexc =320 nm). NO production was evaluated from the constant fluores-
cence intensity after 90 min. Each value represents the mean (Æs.d.) of
at least three independent experiments.
Although the trolox–N-nitrosoalbumin reaction was
slightly less effective than the trolox–NANT reaction in re-
C
leasing NO , the foregoing data clearly demonstrate that the
reaction proceeds with high efficiency for N-nitroso-trypto-
phan residues in proteins.
Figure 11. Release of nitric oxide from the reaction between trolox and
C
N-nitrosated albumin. Steady-state concentrations of NO were measured
C
with the NO -sensitive electrode after addition of trolox (5–50 mm) to N-
C
Influence of NO on the NANT–trolox reaction: The above
nitrosated albumin (100 mm) in phosphate buffer (50 mm, pH 7.4, 378C).
Each value represents the mean (Æs.d.) of at least three independent ex-
periments. The solid line is an empirical fit to the sum of two exponen-
tials.
experiments demonstrated that nitric oxide is released in
the trolox/a-tocopherol-mediated decay of NANT as depict-
ed in Equation (1). With regard to possible reversibility of
Chem. Eur. J. 2007, 13, 7532 – 7542
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7537

M. Kirsch et al.
the NANT–trolox/a-tocopherol transnitrosation reaction, an
external supply of nitric oxide would be expected to exert a
retarding effect on the decay of NANT. We therefore exam-
of nitric oxide from NANT and N-nitrosoalbumin described
above suggest that this process might be useful for selective
detection of N-nitrosated tryptophan residues in proteins,
because in the absence of Cu+ ions S-nitrosothiols do not
react in a similar manner. Figure 14 demonstrates the high
C
ined the effect of additional NO by monitoring the decay of
C
NANT in the presence of the NO -releasing compound
MAHMA/NO. In order to exclude possible nitrosation of
C
NAT by N2O3 (formed by autoxidation of NO ), these ex-
periments were performed under hypoxic conditions (1–
2 mm residual O2). In the absence of trolox, decomposition
C
of NANT was not significantly inhibited by NO , because
after addition of MAHMA/NO(100 mm) to NANT (100 mm,
phosphate buffer pH 7.4, 378C) a residual NANT concentra-
tion of 97.4Æ0.2 mm was monitored after a reaction period
310 K
app
of 900 s [k
=(1.4Æ0.4)”10À4 sÀ1] (Figure 13). This value
Figure 14. Stepwise release of nitric oxide from an N-acetyl-N-nitroso-
tryptophan/GSNOmixture. A mixture of GSNO(25 mm) and NANT
(25 mm) was treated consecutively with trolox and CuSO4 (500 mm each)
in phosphate buffer (50 mm, pH 11, 378C), and release of nitric oxide was
C
monitored with the NO -sensitive electrode. The trace shown is represen-
tative of three independent measurements.
C
Figure 13. Influence of NO on the decay of N-acetyl-N-nitrosotrypto-
phan. The decay of NANT (100 mm) was followed spectrophotometrically
in deoxygenated phosphate buffer (50 mm, pH 7.4, 378C) in the absence
of any additives (open circles), in the presence of trolox (100 mm; closed
C
selectivity of trolox in the liberation of NO only from N-ni-
trosated tryptophan compounds. A solution containing both
GSNO( S-nitrosoglutathione) and NANT (25 mm each) was
supplemented with trolox (500 mm), with the effect of a slow
release of NO resembling the time profile of NO produc-
tion from low concentrations of NANT only. After the
C
squares), in the presence of the NO donor MAHMA/NO(100 mm; open
diamonds), and in the presence of trolox and MAHMA/NO(100 mm
each; closed circles). Each value represents the mean (Æs.d.) of at least
three independent experiments.
C
C
C
NANT-dependent NO equilibrium level had been achieved,
addition of CuSO4 (500 mm) led to an instantaneous burst of
C
compares well with the 96.4Æ0.5 mm concentration of
NO , indicating its rapid liberation from GSNO. A reduced
NANT detected in the absence of MAHMA/NOunder oth-
production rate of nitric oxide was observed when only
150 mm CuSO4 were added, and at Cu2+ concentrations
lower than the applied concentration of EDTA (100 mm),
310 K
erwise identical conditions [k
=(1.8Æ0.3)”10À4 sÀ1]. The
app
above results again rule out the proposal of de Biase et al.[29]
À
C
mentioned above that N NObond homolysis of NANT is
the initial process of NANT decomposition. In the presence
liberation of NO from GSNOcould no longer be detected,
due to complete complexation of the copper ions (data not
shown). These observations suggested that it should be pos-
sible to quantify both types of nitroso compounds in nitro-
sated proteins by use of just one reducing agent: namely
trolox. To validate this hypothesis, additional experiments
with fully N- and S-nitrosated albumin (150 mm) were per-
formed. N- and S-Nitrosoalbumin was first mixed with
of trolox (100 mm), NANT (9.2Æ1.4 mm) was decomposed
C
within 900 s in the absence of the NO donor compound
310 K
[k
=(2.1Æ0.1)”10À4 sÀ1]; see above. In the presence of
app
both trolox and MAHMA/NO(100 mm each), the decay was
reduced by about 50% (4.8Æ0.3 mm consumption of NANT;
Figure 13). Note, however, that in the latter case the time
C
dependence of NANT decay for the first 15 min is strictly
trolox (150 mm) to release NO from the nitrosotryptophan
310 K
linear [k
=(5.1Æ0.1)”10À2 msÀ1], whereas in the other
residues, and Cu2+ (150 mm) was then added to decompose
the nitrosocysteine residues (Figure 15). In fact, as is evident
from Figure 14, behaviour similar to that seen for the
GSNO–NANT mixture (see above) could be observed when
both posttranslational modifications were present in the pro-
tein.
app
cases the superposition of a linear and an exponential decay
mentioned above (see Figure 2) is followed.
Use of trolox/a-tocopherol for analytical purposes: The
characteristics of the trolox-/a-tocopherol-mediated release
7538
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7532 – 7542

Vitamin E Compounds
FULL PAPER
aryl nitrite should therefore exist in equilibrium with the
parent phenol and the nitrosotryptophan derivative, as well
as with the phenoxyl-type radical and nitric oxide
(Scheme 3).
In fact, no products other than nitric oxide, a-phenoxyl-
type radicals, and the para-quinone of trolox were observed
in the course of the reaction (Figures 4, 6–11). Quantum
chemical calculations with 1,4-dihydroxy-2,3,5,6-tetramethyl-
benzene as a model for trolox, performed at the CBS-QB3
level of theory, predict that the deprotonated (phenolate)
form of trolox/a-tocopherol should be the reacting entity in
aqueous solution (data not shown). This prediction is strong-
ly supported by the pH dependence of the decay of NANT
as depicted in Figure 2. Scheme 3 also explains why the
overall kinetics of the trolox-/a-tocopherol–NANT reaction
do not obey a second-order rate law as expected for a rate-
limiting first step (Figure 3). This is due to the regeneration
of one trolox/a-tocopherol molecule from the bimolecular
disproportionation of the corresponding phenoxyl radical.
The bimolecular self-decay of the vitamin E radical is a rela-
tively slow reaction (k2 =1180mÀ1 sÀ1),[31] so that the overall
kinetics depend both on the recombination of the vita-
min E-type radical with nitric oxide and on the aryl nitrite-
dependent transnitrosation of the N-terminal blocked tryp-
tophan molecule. Facilitated decomposition of the N-nitro-
sotryptophan derivative by a radical chain mechanism, as
has recently been established for the ascorbyl radical,[20]
seems to be largely prevented by steric hindrance of the
trolox/a-tocopheryl radical. Thus, the a-tocopherol-/trolox-
mediated decomposition of N-nitrosotryptophan derivatives
can be fully explained by Scheme 3.
C
Figure 15. Consecutive release of NO from N-nitrosotryptophan and S-
nitrosocysteine in N- and S-nitrosated albumin. trolox and Cu2+ ions
(150 mm each) were added to N- and S-nitrosoalbumin (150 mm, phos-
C
phate buffer, pH 7.4, 378C), and NO release was determined with the
C
NO -sensitive electrode.
Discussion
a-Tocopherol, the vitamin E derivative with the highest bio-
logical activity, acts in human beings as a supremely effec-
tive lipid-soluble antioxidant. It is believed that a-tocopher-
ol functions as a chain-breaking antioxidant[35] and scaveng-
es reactive nitrogen oxide species,[36] thereby preventing the
formation of nitroso compounds.[37] To the best of our
knowledge, a novel chemical capability of a-tocopherol and
of its water-soluble derivative trolox is introduced here.
Both compounds react with N-terminal blocked N-nitroso-
tryptophan derivatives through restoration of the indole ring
of the tryptophan moiety and release of stoichiometric
amounts of nitric oxide.
The addition of a-tocopherol/trolox for analytical purposes:
Thanks to the capability of the vitamin E derivatives a-toco-
pherol and trolox to liberate nitric oxide selectively and stoi-
chiometrically from N-nitrosated tryptophan compounds
both in aqueous solution and in lipophilic solvents, one may
suggest the use of the trolox–NANT reaction for the deter-
Chemical mechanism: Our data are in accordance with the
view that the a-tocopherol/trolox denitrosated N-nitroso-
tryptophan derivatives by a stepwise mechanism. The first
reaction should be the transfer
of the nitroso function from
NANT to a-tocopherol/trolox
to yield the corresponding aryl
nitrite (Scheme 3).
Similar O-nitrosation reac-
tions between NANT and either
H2O2,[30]
ascorbate,[20]
cate-
chols[19] or n-butanol,[16] as well
as between N-nitrosoamides
and alcohols,[38] are known.
Generally, aryl nitrites are un-
stable compounds, undergoing
rearrangement and transnitrosa-
tion.[15] O-Nitrosophenols have
short lifetimes as they exist in
C
equilibrium with NO and the
corresponding phenoxyl-type
radical.[39–41] The intermediary Scheme 3. Suggested mechanism showing O-nitrosated a-tocopherol derivatives as key intermediates.
Chem. Eur. J. 2007, 13, 7532 – 7542
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7539

M. Kirsch et al.
mination of N-nitrosotryptophan residues in proteins. Since
the trolox–NANT reaction proceeds even under oxygen-free
conditions, the liberated nitric oxide can easily be detected
the accumulation of N-nitrosotryptophan residues to toxic
levels in lipophilic environments.
Since human skin is very lipophilic and both nitrosoamine
and a-tocopherol concentrations are relatively high,[47,49] it
appears reasonable that this reaction between a-tocopherol
and N-nitrosotryptophan residues may play a major role in
this tissue.
C
with a NO -sensitive electrode as demonstrated here, but it
C
should additionally be possible to transport NO into the gas
phase, where it can be detected by the sensitive ozone
chemiluminescence method.[15] Although these methods can
detect low amounts of nitric oxide, it is often impossible to
extrapolate the concentration of the nitric oxide-donating
compound (in our case N-nitrosotryptophan residues in pep-
tides and proteins) from the detected amount of gaseous
Experimental Section
C
nitric oxide. For quantitative NO monitoring, the cheletrop-
Materials: N-Acetyl-d,l-tryptophan, trolox, bovine serum albumin
(BSA), EDTA and Chelex 100 were purchased from Sigma (Taufkirchen,
Germany). NaNO2, Me2SO, acetonitrile and ethyl acetate were obtained
from Merck (Darmstadt, Germany). a-Tocopherol was from Fluka
Chemie (Steinheim, Germany), ethanol from J.T. Baker (Deventer, Neth-
erlands) and MAHMA/NOfrom Situs (Düsseldorf, Germany). N-Acetyl-
N-nitrosotryptophan, the fluorescent nitric oxide cheletropic trap
(FNOCT-4) and NO-albumin were prepared as described in refs. [30,33]
Stock solutions were freshly prepared on a daily basis and their concen-
trations were spectrophotometrically determined.
ic compound FNOCT-4 is best applied, because of its ability
to accumulate the released NO in a reaction product that
C
can be easily detected fluorimetrically. In addition to the ca-
pability to detect small amounts of N-nitroso residues, it is
additionally possible to differentiate between N-nitroso and
S-nitroso residues. In accord with our findings, trolox does
not react with S-nitroso compounds, but trolox does reduce
Cu2+ ions to cuprous ions, which are well known to catalyse
the release of nitric oxide from S-nitroso derivatives such as
Experimental system: As nitrosation reactions are highly sensitive to the
presence of metal ions, phosphate buffer solutions (50 mm) were treated
with the heavy metal scavenger resin Chelex 100 (0.5 g in 15 mL). The so-
lution was gently shaken, stored overnight and then carefully decanted
from the resin. Afterwards, the pHs of all solutions were readjusted to
pH 7.4Æ0.1 by addition of either H3PO4 (50 mm) or K3PO4 (50 mm). In
order further to avoid any interference from traces of transition metals,
EDTA (100 mm) was added to all reaction mixtures.
[42]
GSNO. Consecutive and selective trolox-mediated nitric
oxide release either from N-nitrosotryptophan residues in
the absence of Cu2+ or from S-nitrosocysteine residues in
the presence of Cu2+ ions is therefore possible (Figures 14
and 15).
C
Determination of nitric oxide with a NO -sensitive electrode: Nitric oxide
Physiological implications: Various physiological aspects can
be related to the function of a-tocopherol. Although Desid-
eri et al. noticed significant increases in nitric oxide produc-
tion in hypercholesterolemic patients mediated by vitamin E
supplementation,[43] its benefit in vasodilatation and preven-
tion of myocardial infarction is disputed.[44,45]
C
formation was determined by use of a NO sensitive electrode (ISO-NO;
World Precision Instruments, Sarasota, Florida), as described in Ref. [50]
The reaction mixtures were continuously stirred throughout the measure-
ments, and temperatures were maintained at 25Æ1 and 37Æ18C. The
C
electrode was calibrated daily and NO production was quantified accord-
ing to the manufacturerꢁs instructions, by employing potassium iodide
(100 mm) in H2SO4 (0.1m) as a calibration solution to which various
amounts of NaNO2 (0.5 mm) were added. The concentrations of a-toco-
pherol/trolox, as well as of NANT and NO-albumin, were varied.
Rubbo et al. noted that the reaction between nitric oxide
and a-tocopherol inhibited membrane lipid peroxidation in
a more potent manner than the one-electron a-tocopherol
plus ascorbate redox cycle system.[36] Since nitric oxide is
known to protect cultured cells against reactive oxygen spe-
cies,[46] it is suggested (and very likely) that nitric oxide pref-
erentially scavenges lipid hydroperoxyl radicals irreversibly,
rather than scavenging the a-tocopheroxyl radical in a rever-
sible manner. Although the a-tocopheroxyl radical has a
limited capability to propagate radical chain reactions,[47] the
dismutation of this radical in part regenerates the parent a-
tocopherol (Scheme 2). The reaction between vitamin E de-
rivatives and N-nitrosotryptophan residues therefore in-
creases the general antioxidative power of vitamin E to-
wards oxidizing radicals.
On the other hand, the reaction of vitamin E with N-ni-
trosotryptophan residues (Scheme 3) would limit the con-
centration of such kinds of nitroso compounds in a lipophilic
environment. N-Nitrosotryptophan derivatives do not spon-
taneously decay in hydrophobic solutions,[18,30] so their con-
centration would continuously increase in the absence of vit-
amin E. Noticeably, NANT is a mutagenic nitrosamine at
high concentrations (>120 mm).[48] We thus believe that vita-
min E has the previously unnoticed function of preventing
C
NO measurement with the fluorescent nitric oxide cheletropic trap
(FNOCT-4): The amounts of NO from the a-tocopherol/trolox–NANT
reaction, as well as from the corresponding one with NO-albumin, were
quantified with FNOCT-4, which directly traps NO to yield the fluores-
cence dye FNOCT-4-NOH (lem =460Æ5 nm, lex =320Æ5 nm). A two-
point calibration of FNOCT-4 (100 mm) was performed by mixing with a-
tocopherol (200 mm) in the absence and in the presence of MAHMA/NO
(200 mm) and reading the fluorescence intensity after an incubation
period of 30 min at 37Æ18C. The stability of the nitric oxide–FNOCT-4
reaction product, FNOCT-4-NOH, was verified for a period of 4 h.
C
C
Kinetic experiments: The a-tocopherol- and trolox-induced decay of
NANT was monitored on a SPECORD S100 spectrophotometer from
Analytic Jena (Jena, Germany), with use of an absorption coefficient of
e335 =6100mÀ1 cmÀ1 for NANT.[51] The reaction between trolox and
NANT in aqueous solution was monitored for 60–120 min by taking re-
cordings every 5 min. Measurements were performed in phosphate buffer
at pH 7.4 and pH 11 and at 25 and 378C. The effect of a-tocopherol on
the decomposition of NANT (100 mm) was also studied in dimethyl sulf-
oxide (DMSO), ethanol, ethyl acetate and acetonitrile.
NAT determination: The formation of NAT and the decay of NANT
were simultaneously monitored during the reaction between NANT and
trolox (100 mm each) in phosphate buffer at pH 11 and room temperature.
The formation of NAT was determined by tenfold dilution of the sample
and subsequent recording of the fluorescence (lexc =270Æ5 nm, lem
=
355Æ5 nm). The decomposition of NANT was determined by reading the
optical density at 335 nm.
7540
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7532 – 7542

Vitamin E Compounds
FULL PAPER
ESR measurements: The trolox radical was identified by ESR spectrome-
try. ESR spectra were recorded at 188C on a Bruker ESP-300E X-band
spectrometer (Bruker, Rheinstetten, Germany) equipped with a TM110
wide-bore cavity. Solutions were prepared from the buffer solution
(pH 7.4, 1 mL) containing trolox and NANT (7.5 mm each) under nor-
moxic conditions. Additional recordings in order to visualize the a-toco-
pherol radical were performed in ethyl acetate with a-tocopherol and
NANT (75 mm each) under both normoxic and hypoxic conditions. Re-
cording conditions were as follows: microwave frequency, 9.79 GHz;
modulation, 0.04 mT; signal gain, 5”105; sweep range, 10 mT; microwave
power, 2 mW; sweep time, 2.8 min. Spectra simulation was performed
with the WinSim program.[52]
[20] A. Kytzia, H.-G. Korth, R. Sustmann, H. de Groot, M. Kirsch,
Chem. Eur. J. 2006, 12, 8786–8797.
[21] G. W. Burton, A. Joyce, K. U. Ingold, Arch. Biochem. Biophys. 1983,
221, 281–290.
[22] K. U. Ingold, A. C. Webb, D. Witter, G. W. Burton, T. A. Metcalfe,
D. P. R. Muller, Arch. Biochem. Biophys. 1987, 259, 224–225.
[23] R. Ricciarelli, A. Tasinato, S. Clement, N. K. Ozer, D. Boscoboinik,
A. Azzi, Biochem. J. 1998, 334, 243–249.
[24] D. Boscoboinik, A. Szewczyk, A. Azzi, Arch. Biochem. Biophys.
1991, 286, 264–269.
[25] A. Azzi, E. Aratri, D. Boscoboinik, S. Clement, N. K. ꢂzer, R. Ric-
ciarelli, S. Spycher, Biofactors 1998, 7, 3–14.
[26] S. ꢃlker, P. P. McKeown, U. Bayraktutan, Hypertension 2003, 41,
534–539.
[27] S. S. Al-Kaabi, G. Hallet, T. A. Meyer, D. L. H. Williams, J. Chem.
Soc. Perkin Trans. 2 1984, 1803–1807.
[28] B. Blanchard-Fillion, C. Servy, C. Ducrocq, Free Radical Res. 2001,
35, 857–866.
13C NMR measurements: NMR experiments were performed on
a
Bruker ADVANCE DRX 500 instrument (Bruker Biospin, Rheinstetten,
Germany) at 125.71 MHz. Chemical shifts (d) are given in ppm relative
to TMS (d=0) as external standard. 13C NMR spectra were obtained
from NANT (100 mm) added to trolox (100 mm), as well as from NANT
(100 mm) mixed with Fremiꢁs salt (100 mm) in [D6]DMSO/phosphate
buffer pH 7.4 (8:2). The oxidation of 2,3,5,6-tetraalkylphenols such as
trolox with Fremyꢁs salt represents a clean synthetic method for the prep-
aration of the corresponding p-benzoquinones.[53] In addition, NANT
(100 mm) or trolox (100 mm) were monitored in the absence of any other
reactants. All samples were measured after a reaction period of 18 h.
[29] P. M. de Biase, A. G. Turjanski, D. A. Estrin, F. Doctorovich, J. Org.
Chem. 2005, 70, 5790–5798.
[30] M. Kirsch, H. de Groot, J. Pineal Res. 2005, 38, 247–253.
[31] V. W. Bowry, K. U. Ingold, J. Org. Chem. 1995, 60, 5456–5467.
[32] R. S. Alger, Electron Paramagnetic Resonance, Wiley-Interscience,
New York, 1968, p. 289–462.
Quantum-chemical calculations: Complete basis set (CBS-QB3) compu-
tations were carried out with the Gaussian 03 (revision AM64 L-G03
RevC0.2) suite of programs.[54] Gibbs free energies of solvation for water
were estimated for the optimized gas-phase geometries by use of the
CPCM-UAHF procedure incorporated in Gaussian 03. Both the CBS-
QB3 and CPCM/(U)HF/6-31+G (d) methodologies are known to provide
thermochemical estimates within “chemical accuracy” (Æ1 kcal
molÀ1).[55,56]
[33] P. Meineke, U. Rauen, H. de Groot, H.-G. Korth, R. Sustmann,
Chem. Eur. J. 1999, 5, 1738–1747.
[34] H. Fischer, Chem. Rev. 2001, 101, 3581–3610.
[35] R. Brigelius-Flohe, M. G. Traber, FASEB J. 1999, 13, 1145–1155.
[36] H. Rubbo, R. Radi, D. Anselmi, M. Kirk, S. Barnes, J. Butler, J. P.
Eiserich, B. A. Freeman, J. Biol. Chem. 2000, 275, 10812–10818.
[37] R. V. Cooney, A. A. Franke, P. J. Harwood, V. Hatch-Pigott, L. J.
Custer, L. J. Mordan, Proc. Natl. Acad. Sci. USA 1993, 90, 1771–
1775.
[38] R. W. Darbeau, N. Alvarez, R. N. Aymond, J. Org. Chem. 2005, 70,
3225–3230.
[39] S. Goldstein, G. Czapski, J. Lind, G. Merenyi, J. Biol. Chem. 2000,
275, 3031–3036.
[40] F. G. Janzen, A. L. Wilcox, V. Manoharan, J. Org. Chem. 1993, 58,
3597–3599.
[41] F. Berho, F. Carlap, M.-T. Rayez, R. Lesclaux, J. Phys. Chem. 1998,
102, 1–8.
[42] D. L. H. Williams, Chem. Commun. 1996, 1085–1091.
[43] G. Desideri, M. C. Marinucci, G. Tomassoni, P. G. Masci, A. Santuc-
ci, C. Ferri, J. Clin. Endocrinol. Metab. 2002, 87, 2940–2945.
[44] B. Halliwell, Lancet 2000, 355, 1179–1180.
[1] M. Hoshino, K. Ozawa, H. Seki, P. C. Ford, J. Am. Chem. Soc. 1993,
115, 9568–9575.
[2] D. Nikitovic, A. Holmgren, J. Biol. Chem. 1996, 271, 19180–19185.
[3] Y. C. Hou, Z. M. Guo, J. Li, P. G. Wang, Biochem. Biophys. Res.
Commun. 1996, 228, 88–93.
[4] K. Harohalli, C. E. Petersen, C.-E. Ha, J. B. Feix, N. V. Bhagavan, J.
Biomed. Sci. 2002, 9, 47–58.
[5] Y. Y. Zhang, A.-M. Xu, M. Nomen, M. Walsh, J. F. Keaney, Jr., J.
Loscalzo, J. Biol. Chem. 1996, 271, 14271–14279.
[6] S. Goldstein, G. Czapski, J. Am. Chem. Soc. 1995, 117, 12078–
12084.
[45] N. Stephens, Lancet 1997, 349, 1710–1711.
[46] A. Wink, J. C. Cook, R. Pacelli, J. Leibmann, M. C. Krishna, J. B.
Mitchell, Toxicol. Lett. 1995, 82–83, 221–226.
[7] M. Grätzel, S. Taniguchi, A. Henglein, Ber. Bunsen-Ges. Phys.
Chem. 1970, 74, 488–492.
[47] M. Walke, D. Beckert, J. Lasch, Photochem. Photobiol. 1998, 68,
502–510.
[8] V. Kharitonov, A. Sundquist, V. Sharma, J. Biol. Chem. 1995, 270,
28158–28164.
[48] S. Venitt, C. Crofton-Sleigh, S. L. Ooi, R. Bonnett, Carcinogenesis
1980, 1, 523–532.
[9] M. Kirsch, A. Fuchs, H. de Groot, J. Biol. Chem. 2003, 278, 11931–
11936.
[49] A. N. Paunel, A. Dejam, S. Thelen, M. Kirsch, M. Horstjann, P.
Gharini, M. Mürtz, M. Kelm, H. de Groot, V. Kolb-Bachofen, C. V.
Suschek, Free Radical Biol. Med. 2005, 38, 606–615.
[50] Z. Taha, F. Kiechle, T. Malinski, Biochem. Biophys. Res. Commun.
1992, 188, 734–739.
[51] R. Bonnett, R. Holleyhead, J. Chem. Soc. Perkin Trans. 1 1974, 962–
964.
[52] D. R. Duling, J. Magn. Reson. Ser. B 1994, 104, 105–110.
[53] H. Zimmer, D. C. Lankin, S. W. Horgan, Chem. Rev. 1971, 71, 229–
246.
[54] Gaussian 03W, Revision C.02, M. J. Frisch, G. W. Trucks, H. B.
Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Mont-
gomery Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
[10] A. van der Vliet, P. A. C. Hoen, P. S.-Y. Wong, A. Bast, C. E. Cross,
J. Biol. Chem. 1998, 273, 30255–30262.
[11] B. Blanchard, D. Pompon, C. Ducrocq, J. Pineal Res. 2000, 29, 184–
192.
[12] M. Kirsch, H. de Groot, J. Pineal Res. 2006, 40, 10–17.
[13] T. Suzuki, H. F. Mower, M. D. Friesen, I. Gilbert, T. Sawa, H. Ohshi-
ma, Free Radical Biol. Med. 2004, 37, 671–681.
[14] T. A. Meyer, D. L. H. Williams, R. Bonnett, S. L. Ooi, J. Chem. Soc.
Perkin Trans. 2 1982, 1383–1387.
[15] D. L. H. Williams, Nitrosation reactions and the chemistryof nitric
oxide, Elsevier, Amsterdam, 2004, p.
[16] R. Bonnett, P. Nicolaidou, Heterocycles 1977, 7, 637- 659.
[17] M. Kirsch, M. Lehnig, Org. Biomol. Chem. 2005, 3, 2085–2090.
[18] K. Sonnenschein, H. de Groot, M. Kirsch, J. Biol. Chem. 2004, 279,
45433–45440.
[19] A. Kytzia, H.-G. Korth, H. de Groot, M. Kirsch, Org. Biomol.
Chem. 2006, 4, 257–267.
Chem. Eur. J. 2007, 13, 7532 – 7542
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7541

M. Kirsch et al.
J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O.
Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Fores-
man, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Na-
nayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian Inc., Pittsburgh PA,
2003.
[55] V. Barone, M. Cossi, J. Phys. Chem. 1998, 102, 1995–2001.
[56] V. Barone, M. Cossi, J. Tomasi, J. Chem. Phys. 1997, 107, 3210–
3221.
Received: March 9, 2007
Published online: July 4, 2007
7542
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7532 – 7542