(Received for publication, September 19, 1996, and in revised form, December 20, 1996)
From the Tumor Biology Section, Radiation Biology
Branch, NCI, National Institutes of Health, Bethesda, Maryland 20892, the ¶ Department of Physiology, Louisiana State University Medical
Center, Shreveport, Louisiana 71130, and the
Department of
Chemistry, Grambling State University, Grambling, Louisiana 71245
Thiol-containing proteins are key to numerous
cellular processes, and their functions can be modified by thiol
nitrosation or oxidation. Nitrosation reactions are quenched by
O2, while the oxidation chemistry mediated by
peroxynitrite is quenched by excess flux of either NO or
O
2. A solution of glutathione (GSH), a model
thiol-containing tripeptide, exclusively yielded S-nitrosoglutathione when exposed to the NO donor,
Et2NN(O)NONa. However, when xanthine oxidase was added to
the same mixture, the yield of S-nitrosoglutathione
dramatically decreased as the activity of xanthine oxidase increased,
such that there was a 95% reduction in nitrosation when the fluxes of
NO and O
2 were nearly equivalent. The presence of superoxide
dismutase reversed O
2-mediated inhibition, while catalase
had no effect. Increasing the flux of O
2 yielded oxidized
glutathione (GSSG), peaking when the flux of NO and O
2 were
approximately equivalent. The results suggest that oxidation and
nitrosation of thiols by superoxide and NO are determined by their
relative fluxes and may have physiological significance.
The endogenously formed radicals, NO1
and superoxide (O2), are thought to play key roles in the
regulation of a number of physiological and pathophysiological
mechanisms. This interaction has been suggested to be pivotal in
cytotoxic mechanisms involving the modulation of oxidative stress by
NO. Thiols are critical sites of interaction for NO and O
2 in
biological systems, perhaps explaining why proteins and peptides
containing thiols that undergo nitrosation as well as oxidation
reactions might modulate cellular function. Reactions of chemical
species derived from NO can result in the nitrosation of thiols to form
S-nitrosothiol complexes (1, 2). Oxidation of thiols by
oxidants, some of which are derived from NO (3), results in disulfide
products, affecting cellular metabolism and physiological functions
(4-7).
Recently, we have shown that oxidative and nitrosative chemistry
resulting from a continuous chemical generator of NO, SPER/NO (a
NONOate complex that spontaneously releases NO at physiological pH), in the presence of a source of O2, xanthine oxidase
(XO), is modulated by the relative rate of formation of NO and
superoxide (8, 9). On one hand, the nitrosative chemistry of amines is
quenched by the presence of O
2 (8), while increasing the flux
of O
2 increases oxidation mediated by peroxynitrite (9). On
the other hand, oxidation mediated by peroxynitrite is quenched by
excess formation of either NO or O
2 (9). We have explored the
oxidative and nitrosative chemistry resulting from the interaction of
NO and O
2 with thiols and demonstrate that there is a balance between fluxes of NO and O
2 with respect to oxidative and
nitrosative products of glutathione.
Glutathione (GSH), hypoxanthine (HX), superoxide dismutase
(SOD), catalase, and diethylenetriaminepentaacetic acid (DETAPAC) were
purchased from Sigma. Stock solutions were prepared as follows: 50 mM HX was made in 0.08 mM NaOH, 10 mM DETAPAC in milliQ water, and 10 mM GSH in
milliQ water. These stock solutions were diluted in PBS (pH 7.4) to the
reported final concentration. XO was purchased from Boehringer
Mannheim. DEA/NO was a generous gift from Dr. Joseph Saavedra. Stock
solution of DEA/NO were prepared as described previously (10), and
stock concentrations were determined by UV absorption at 250 nm, using
an extinction coefficient of 8000 M1
cm
1.
A solution containing 0.5 mM HX
and 50 µM DETAPAC in PBS was prepared and referred to as
the HX solution. Typically, GSH or DAN was added to the solutions,
followed by catalase and/or SOD, and when appropriate, XO was followed
by DEA/NO. Samples were incubated for 1 h at 37 °C. At the end
of this time DEA/NO was completely decomposed and 50 µM
allopurinol was added to inhibit the activity of XO. The solutions were
then analyzed as described below. Under these assay conditions, 1 milliunit/ml XO produced a flux of O2 and
H2O2 of approximately 1 nmol/min each as
described previously (9).
The rate of the decomposition of DEA/NO is typically
determined by the loss of its characteristic absorbance at 250 nm.
However, the HX solution strongly absorbs in this region and does not
allow an accurate determination of the kinetics of decomposition. We have recently described a method used to determine the amount of NO
released from an NO donor, which involves monitoring at 500 nm where HX
solution does not absorb (11). In a solution containing 0.5 mM HX and 50 µM DETAPAC, 0.5 g/100 ml of
sulfanilamide and 0.03 g of
N-(1-naphthyl)ethylenediamine dihydrochloride (NEDD)/100 ml
of PBS was dissolved as reported previously (11). Monitoring the
appearance of the characteristic azo dye band at 500 nm, the rate
constant for decay of DEA/NO was determined by a plot of ln(Abs Absinf) versus time. Since the absorbance
requires 2 molecules of NO for every one of azo dye formed, this rate
was multiplied by two to get the rate of NO release under these
conditions.
GSNO was detected using the fluorescence method described by Cook et al. (12). 200 µl of the sample solution was placed in a 2-ml solution containing 500 µM DAN and 100 µM mercuric chloride in PBS. The samples were allowed to stand at room temperature for 30 min and then read on a Perkin-Elmer fluorometer LS50B with excitation at 375 nm and emission at 450 nm. To standardize the fluorescence reading from naphthaltriazole (NAT) with S-nitrosothiol, known amounts of GSNO (0-10 µM) were treated in the identical manner described above. A standard curve was generated correlating [GSNO] with fluorescence. GSNO concentrations from the experiments were extrapolated from standard curve.
HPLC Method for Determination of GSSG1 µl of the sample of the incubation mixture was taken and 145 µl of 0.6% sulfosalicylic acid and 10 mM N-butylmaleimide. To this, 34 µl of borate (200 mM) was added to adjust the pH to 8.5. The solution was allowed to stand for 5 min, at which time 20 µl of 1 mM 7-methoxycoumarin-3-carboxylic acid, succimidyl ester (Molecular Probes, Eugene; OR) was added, and the solution was incubated for 15 min. 25 µl of the sample was injected onto a Millipore HPLC system with a NOvapacTM C-18 (3.9 × 150 mm) column (Waters Corp., Milford, MA). A gradient was run with buffer A containing 5% acetonitrile, 34 mM sodium perchlorate, 0.25% glacial acetic acid, and buffer B containing 80% acetonitrile, 34 mM sodium perchlorate, 0.25% glacial acetic acid. Fluorescence was determined by excitation at 345 and emission at 410 nm.
Electrochemical Determination of NOThe production of NO via the spontaneous decomposition of DEA/NO at 37 °C in the presence or absence of different fluxes of superoxide and hydrogen peroxide was determined electrochemically using an electrode specific for NO (World Precision Instruments, Sarasota, FL). DEA/NO was introduced into a thermostatted cell containing a 2.5-ml reaction volume to a final concentration of 20 and 50 µM in a solution containing 50 µM DETAPAC, 500 µM HX, and when appropriate, 10 milliunits/ml XO in PBS (pH 7.4) ± SOD (0.1 mg/ml). The current obtained was monitored as a function of time, and the electrode was standardized using acidified NaNO2.
In the presence
of 0.2 mM GSH in PBS, 20 and 50 µM DEA/NO
yielded GSNO (Fig. 1, A and B).
When HX/XO was added with increasing activity, a marked decrease in the
formation of GSNO was observed: less than 0.2 µM GSNO was
formed at 5 and 10 nmol/min XO activity when the concentrations of
DEA/NO were 20 and 50 µM, respectively (Fig. 1). However,
nitrosation of GSH surprisingly increased as XO activity was increased
to 100 milliunits/ml (Fig. 1, A and B), a process
unaffected by 20 µM catalase. The lack of effect of
catalase suggests that O2 and not hydrogen peroxide attenuates nitrosation. Such results lessen the concerns of oxygen depletion by
xanthine oxidase, since catalase will adequately reoxygenate the
solution by converting hydrogen peroxide to oxygen.
Next, N-nitrosation was examined as a function of increasing
XO activity. A solution of 0.5 mM DAN and 20-50
µM DEA/NO in PBS results in the formation of NAT. Unlike
the formation of GSNO, increasing XO activity had little effect on
nitrosation until XO activity was 5-10 milliunits, at which point the
yield of NAT was dramatically reduced (Fig. 2, A and
B). Analogously to the formation of GSNO,
nitrosation of DAN increased as the flux of O2 and
H2O2 was increased to 100 nmol/min each, though
not to the same extent as that observed prior to addition of XO (Fig. 2, A and B).
Increasing amounts of SOD, and not catalase, increased the nitrosation
of GSH resulting from 50 µM DEA/NO in the presence of 20 milliunits of XO (Fig. 3A)., suggesting that
O2 and not hydrogen peroxide may attenuate nitrosation.
Nitrosation of DAN was inversely proportional to the flux of
O
2 in this case as well (Fig. 3B). Since SOD would
not affect the rate of oxygen consumption by XO but the availability of
O
2, reduction of GSNO and NAT formation by XO does not result
from oxygen depletion but rather implicates O
2 as being
the responsible species for reducing nitrosation.
Formation of GSSG
Peroxynitrite can oxidize GSH to GSSG. The
nearly diffusion controlled reaction between NO and O2 to form
peroxynitrite (13) suggests this reaction could affect the redox status
of intracellular and extracellular thiols. GSH can be oxidized in the
presence of hydrogen peroxide (3); therefore, catalase was present in these experiments to prevent this side reaction, since the nitrosation of GSH was not effected by the presence of catalase. When 20 or 50 µM DEA/NO was exposed to a solution of HX containing 0.2 mM GSH, variation of XO activity increased formation of
GSSG (Fig. 4). The levels of GSH decreased in both
cases.
Detection of NO
An electrode sensitive for NO was used to
determine the role of this diatomic molecule in the reactions described
above. 50 µM DEA/NO was introduced to an aqueous solution
as described under "Materials and Methods." The concentration of NO
initially rose to ~6 µM and then fell to near zero
(Fig. 5). This result is consistent with NO being
released by DEA/NO, followed by consumption via the auto-oxidation
reaction. When 10 milliunits of XO were present, NO was undetectable by
this method. When 0.1 mg/ml SOD was present with 50 µM
DEA/NO and 10 milliunits of XO, NO was detected, suggesting that
O2 was responsible for the decrease in NO released from DEA/NO.
The formation of reactive intermediates derived from oxygen and NO
has been invoked in mechanisms for the degradation of
biomacromolecules, with attendant pathophysiological consequences. The
near diffusion-controlled reaction between NO and O2 to form
peroxynitrite (13) has suggested that this RNOS may be central in the
toxic and regulatory mechanisms of NO and oxygen radicals (14, 15). One
class of reaction products of NO is the S-nitrosothiols,
which are formed in vivo and which have been suggested to
regulate a variety of physiological functions (7, 16). It has been
proposed that proteins modified by NO to contain
S-nitrosothiol groups may regulate physiological functions
such as vascular tension (16-18) as well as modulating the
transcription of genes. Conversely, S-nitrosation of some proteins inhibits key enzymes (19, 20), including those which contain
zinc finger motifs (19, 21, 22). Furthermore, NO via RNOS reacts with
metallothionein to form an S-nitrosothiol (21), which either
represents a detoxification pathway (23) or results in liberation of
toxic metals such as cadmium (24).
In an attempt to examine the mechanisms by which such diverse actions
are mediated, we have begun to examine the chemistry resulting from the
interaction of NO and O2 with respect to nitrosation and
oxidation (8, 9). Previously, we showed that the nitrosation of DAN was
suppressed by the presence of O
2 (8). As seen in Figs. 1 and
3, increasing concentrations of O
2 dramatically reduced the
nitrosation of both GSH and DAN. As seen in Fig. 3, SOD reversed this
inhibition, suggesting that the reduction in nitrosation is due to the
presence of O
2. Since NO or resultant intermediates from NO
donor complexes do not inhibit the activity of XO (25), this inhibition
of nitrosation by O
2 can be explained by the following
mechanism. In the absence of XO, nitrosation is mediated by
intermediates formed in the NO/O2 reaction (i.e.
N2O3) (Equations 1-3).
![]() |
(Eq. 1) |
![]() |
(Eq. 2) |
![]() |
(Eq. 3) |
![]() |
(Eq. 4) |
![]() |
(Eq. 5) |
![]() |
(Eq. 6) |
The relative fluxes of NO and O2 are an important aspect of
the foregoing chemical reactions. As described above, maximal oxidation
and minimal nitrosation occurred when the activity of XO was 5-20
milliunits. Since the rate of formation of NO from DEA/NO was 5 × 10
3 s
1, the flux of NO generated by 20 and
50 µM DEA/NO would be predicted to be initially
approximately 10 and 25 µM per min. This suggests that
the maximal oxidation and minimal nitrosation occurs when the flux of
NO and O
2 were nearly equivalent. These results are consistent
with our previous finding that maximal oxidation of dihydrorhodamine occurred when the rate of formation of O
2
was nearly equivalent to that of NO (9).
The effects of XO on nitrosation of GSH and DAN in Figs. 1 and 3 were
distinctly different, suggesting a difference in the effects on these
two substrates of increasing the flux of O2. As XO increased
from 1 to 10 milliunits, the formation of GSNO decreased, yet
nitrosation of DAN did not decrease with increasing XO until XO
activity was 5-10 milliunits. One reason for this discrepancy is the
formation of ONOO
, which reacts sequentially with another
2 mol of NO to form N2O3, via the intermediate
formation of NO2 (Equations 7 and 8) (27, 28), nitrosating
DAN or GSH (Equations 1 and 2).
![]() |
(Eq. 7) |
![]() |
(Eq. 8) |
![]() |
(Eq. 9) |
These data suggest that nitrosative and oxidative stresses are in a
delicate balance, which depends on the relative fluxes of NO and
O2. At a 1:1 ratio of formation of these radicals, the
formation of OONO
is favored, which then results in
maximal metal free oxidation and hydroxylation (9). Nitrosation occurs
when the fluxes of NO and O
2 are not balanced. These competing
processes suggest a mechanism whereby nitrosative and oxidative
stresses may be created, with effects in vivo. The formation
of nitrosylated thiols versus disulfide adducts could be
important in the regulation of gene transcription as well as the
modulation of ion channels (7). Furthermore, GSH may be critical in
attenuating oxidative and nitrosative stresses mediated by RNOS.
Formation of O
2 and subsequent conversion of NO to
ONOO
in the presence of GSH may in fact be a
detoxification reaction, since GSH scavenges ONOO
anion
prior to the formation of the potent oxidant HOONO. Such a mechanism
would prevent the formation of both N2O3 and
oxidation mediated by ONOO
. Finally, SOD may attenuate
nitrosation and oxidation reactions, since this enzyme reversed the
inhibition of nitrosation observed at activities of XO (Fig. 3).
One of the intriguing aspects of this work is that the fluxes of NO in
these experiments are as low as 1-5 µM, yet can bring about either nitrosative or oxidative chemistry. In Fig. 5, the 50 µM DEA/NO generated less than 6 µM NO for
about 30 min. The presence of O2 vitally eliminated the
detection of NO (Fig. 5). This result suggests that concentrations of
NO could be in the nanomolar range, at which point nitrosation and
oxidation by RNOS can occur. Although the production of RNOS is well
controlled in vivo, their production could be
dysregulated at sites near a cellular or organellar source of
O
2 in vivo. Superoxide would not be expected to
migrate far from its source, while NO can migrate over many cell
lengths (29). When NO enters such a site of superoxide formation, we
hypothesize that a gradient of concentrations would be established,
with attendant high fluxes of nitrosation. At similar fluxes,
ONOO
could be formed and react chemically. Unlike
O
2, ONOO
can migrate and then either be
consumed by intracellular thiols such as GSH or react with NO or
superoxide to form NO2 (Scheme 1).
Additionally, extracellular SOD bound to the endothelium may be
critical in attenuating O
2 in the extracellular space. Our data suggest that SOD may also modulate nitrosation, thereby
attenuating oxidative and nitrosative stresses.