Content uploaded by Martin Feelisch
Author content
All content in this area was uploaded by Martin Feelisch
Content may be subject to copyright.
ANTIOXIDANTS & REDOX SIGNALING
Volume 3, Number 2, 2001
Mary Ann Liebert, Inc.
Forum Review
Mechanisms of the Antioxidant Effects of Nitric Oxide
DAVID A. WINK,
1
KATRINA M. MIRANDA,
1
MICHAEL G. ESPEY,
1
RYZARD M. PLUTA,
2
SANDRA J. HEWETT,
3
CAROL COLTON,
4
MICHAEL VITEK,
4
MARTIN FEELISCH,
5
and MATHEW B. GRISHAM
5
ABSTRACT
The Janus face of nitric oxide (NO) has prompted a debate as to whether NO plays a deleterious or protective role
in tissue injury. There are a number of reactive nitrogen oxide species, such as N
2
O
3
and ONOO
2
, that can alter
critical cellular components under high local concentrations of NO. However, NO can also abate the oxidation
chemistry mediated by reactive oxygen species such as H
2
O
2
and O
2
2
that occurs at physiological levels of NO.
In addition to the antioxidant chemistry, NO protects against cell death mediated by H
2
O
2
, alkylhydroperoxides ,
and xanthine oxidase. The attenuation of metal/peroxide oxidative chemistry, as well as lipid peroxidation, ap-
pears to be the major chemical mechanisms by which NO may limit oxidative injury to mammalian cells. In ad-
dition to these chemical and biochemical properties, NO can modulate cellular and physiological processes to limit
oxidative injury, limiting processes such as leukocyte adhesion. This review will address these aspects of the chem-
ical biology of this multifaceted free radical and explore the beneficial effect of NO against oxidative stress. An-
tioxid. Redox Signal. 3, 203–213.
203
INTRODUCTION
E
NDOGENOUS FORMATION of the radical species
nitric oxide (NO) results in an array of bi-
ological effects. Over the last decade, the role
that NO plays in oxidative injury has been de-
bated (2, 8, 23, 55). In the presence of oxygen
(O
2
) or reactive oxygen species (ROS) such as
superoxide (O
2
2
), NO is converted into reac-
tive nitrogen oxide species (RNOS), which can
irreversibly modify a variety of biological mol-
ecules (summarized in 54). Thus, NO is often
considered to be a toxic species. However, NO
has also been shown to abate oxidative injury
in several biological systems. This review in-
cludes a discussion of the diverse chemistry of
NO, which can result in processes as disparate
as abatement of oxidative stress and mediation
of cell death and tissue injury. Also discussed
are the mechanisms by which NO modifies cel-
lular processes that protect cells and tissue
from oxidative damage. Finally, the effects of
NO treatment in various disease models are re-
viewed.
One of the fundamental determinants of the
role of NO
in vivo
is its chemistry. As NO is a
radical, it has a number of potential chemical
pathways in biological systems. As discussed
in previous articles, the complex chemistry of
NO can be divided primarily into the two cat-
1
Tumor Biology Section, Radiation Biology Branch, National Cancer Institute and
2
Surgery Branch, NINDS,
Bethesda, MD.
3
Department of Neuroscience, University of Connecticut Health Center, Farmington, CT 06030-6125.
4
Division of Neurology, Duke University Medical Center, Durham, NC.
5
Department of Molecular and Cellular Physiology, Louisiana State University Medical Center, Shreveport, LA.
egories of direct and indirect effects (54). Direct
effects are comprised of those reactions in
which NO interacts directly with the biological
target. Conversely, indirect effects are medi-
ated by RNOS formed from the reaction of NO
with O
2
or O
2
2
.
One advantage to categorizing the reactions
of NO in this manner is that the concentration
of NO can be used to predict the contribution
of each effect. Direct effects require low con-
centrations of NO, whereas indirect effects oc-
cur at much higher NO concentrations (54).
Toxicity largely occurs under conditions where
NO-derived species are formed via the indirect
route. Thus, deleterious effects derived from
the chemistry of NO are often confined to con-
ditions of high local NO concentrations. The
antioxidant effects of NO are primarily a result
of direct effects at lower NO concentrations.
CHEMISTRY OF THE ANTIOXIDANT
PROPERTIES OF NO
An antioxidant is a substance that prevents
oxidant formation or scavenges oxidants pro-
duced under conditions of oxidative stress. The
primary source of oxidative stress is ROS, such
as O
2
2
and peroxide. These species often pro-
duce oxidants through Fenton-type reactions.
RNOS, such as peroxynitrite (ONOO
2
), nitro-
gen dioxide (NO
2
), and nitroxyl (HNO), are an-
other potential source of oxidative stress, al-
though they are less potent oxidants than those
formed as a result of Fenton-type reactions.
Further, lipid peroxidation perpetuates oxida-
tive stress through formation of a variety of
lipid-oxy and -peroxy adducts. The versatile
chemistry of NO provides antioxidant mecha-
nisms against all three of these types of oxi-
dants predominantly through radical–radical
and ligand–metal interactions.
Fenton chemistry
In general, Fenton-type reactions occur be-
tween peroxide and transition metals and re-
sult in formation of hydroxyl radicals or hy-
pervalent peroxo- or oxo-metal complexes.
Fe
21
1 H
2
O
2
R
Fe
31
1
?
OH, Fe
41
O, FeO
2
(1)
These reactive intermediates can then alter bi-
ological substances such as proteins, nucleic
acids, and lipids, resulting in tissue injury (54).
NO can abate Fenton-mediated oxidative stress
by direct scavenging of oxidants, prevention of
peroxide reaction, and scavenging of reducing
equivalents supplied by O
2
2
(Fig. 1).
Hydroxyl radicals and high valent metal
complexes are scavenged by NO at near diffu-
sion control (
e.g.
, .10
9
M
21
s
21
for NO 1
?
OH;
4). However, reactions between these highly re-
active compounds and substances other than
NO can also proceed with similarly high sec-
ond-order rate constants (21, 22). Therefore, the
determining factor for the antioxidant proper-
ties of NO is the relative concentration of NO
with respect to other potential reaction sites
(
i.e.
, the pseudo first-order rate constants). As
discussed below, lipid peroxidation is abated
through a radical–radical reaction mechanism
at biologically relevant concentrations of NO
(48). Conversely, the reaction products of the
Fenton reaction generally must be produced in
close proximity to a macromolecule to induce
DNA strand breaks or protein oxidation.
Therefore, diffusion of NO to these regions
would likely not yield high enough effective
NO concentrations to out-compete the delete-
rious reactions between oxidants and biomole-
cules. Under these conditions, NO would not
WINK ET AL.
204
FIG. 1. Antioxidant mechanism of NO in the Fenton-
type reactions.
be a good scavenger of oxidants and probably
would not have an antioxidant role.
The antioxidant properties of NO also in-
clude prevention of oxidant formation. For in-
stance, formation of a metal nitrosyl complex
can prevent addition of peroxide to the metal
coordination site. Metals such as copper and
nickel have been invoked as mediators of ox-
idative stress in some cases, however, iron com-
plexes are the predominant catalyst for this re-
action
in vivo
, generally through Fenton-type
reactions. To understand better the relationship
between iron nitrosyl complex formation and
prevention of oxidant production, it is useful
to consider the environment of the interacting
metal and ligands.
Hard–soft acid–base rules state that hard li-
gands, which donate their electrons less read-
ily than soft ligands, tend to associate more
strongly with hard, low polarizable metals. For
instance, ferric iron, which is a hard metal, will
bind more tightly to the hard ligand hydroxide
than will the softer ferrous ion. However, bind-
ing of soft ligands can result in softening of
hard metals. The ability of ligands to alter the
metal environment affects the reactivity of
metal-catalyzed reactions, which depend on
the oxidation state of the metal as well as the
ligand field.
For instance, hydroxyl radical is the primary
product of the reaction between peroxide and
soft ferrous iron surrounded by a hard ligand
field. In softer ligand fields, such as provided
by heme complexes, both the ferric and ferrous
states principally produce oxidizing species
such as Fe
41
and Fe
51
oxo complexes rather
than hydroxyl radicals. As NO is a soft ligand,
it will readily bind to ferrous complexes to form
a metal-nitrosyl species, but as a general rule,
if peroxide can react with the metal site, NO
can also bind. The resulting nitrosyl complex
inhibits the reaction between peroxide and the
metal, thereby preventing ROS production.
As described above, formation of hydroxyl
radicals or metal-oxo species often requires
Fe
21
. However, iron is more stable in the ferric
valence state
in vivo
. Reduction of Fe
31
to Fe
21
(Haber–Weiss chemistry) can involve O
2
2
,
which can provide the electrons to facilitate cat-
alytic oxidation of biological compounds (20).
NO and O
2
2
react at diffusion control to form
ONOO
2
, which rapidly rearranges to nitrate in
the absence of other reactive species (45). Shunt-
ing of O
2
2
to nitrate thus inhibits reduction of
ferric iron and prevents catalytic formation of
ROS. This may be an important mechanism for
abatement of Fenton-type reaction-mediated
oxidative stress by NO.
NO/O
2
2
reaction
Although scavenging of O
2
2
by NO can pre-
vent production of ROS via the Fenton reac-
tion, ONOO
2
is itself an oxidant. Exposure to
synthetic ONOO
2
has been shown to induce
tissue injury through reactions such as DNA
damage and lipid peroxidation (46). Therefore,
formation of ONOO
2
from the reaction be-
tween O
2
2
and NO has been speculated to in-
duce oxidative stress
in vivo
(2, 46). However,
the chemistry of the NO/O
2
2
reaction can of-
ten give different results from exposure to syn-
thetically generated ONOO
2
.
The oxidative chemistry of the NO/O
2
2
re-
action was found to depend on the relative rate
of production of the two radicals (41, 59). For
instance, in the presence of a constant O
2
2
flux
from xanthine oxide/hypoxanthine (XO), oxi-
dation of dihydrorhodamine or glutathione
(GSH) increased with increasing NO until the
rates of formation of these radicals were equiv-
alent. Further increases in NO concentration,
produced from a class of compounds known as
NONOates, which release NO in a controlled
manner over specific time periods (31), resulted
in marked decreases in dihydrorhodamine or
GSH oxidation. These data demonstrate that, in
the presence of excess NO, the reactive inter-
mediates formed during decomposition of
ONOO
2
are scavenged by NO. This reaction
ultimately results in production of the ni-
trosating species N
2
O
3
(Fig. 2; 59). These find-
ings indicate that the oxidative chemistry of the
NO/O
2
2
reaction is confined to a small range
of fluxes because excess NO can convert ox-
idative RNOS to nitrosating species.
More recently, carbon dioxide (CO
2
) was
shown to react with ONOO
2
to form a potent
adduct, CO
2
OONO
2
, which has oxidative prop-
erties similar to those of ONOO
2
(17, 36, 38).
However, CO
2
activation of ONOO
2
is pH-in-
dependent, unlike formation of the reactive
species HOONO in the absence of CO
2
. Excess
NO also quenches the oxidation chemistry of this
MECHANISMS OF THE ANTIOXIDANT EFFECTS OF NO
205
adduct (28). The CO
2
adduct has been proposed
to decompose to the carbonate radical (CO
3
2
)
and NO
2
(38). Reaction of excess NO with NO
2
formed in this manner would again result in pro-
duction of N
2
O
3
(Fig. 2; 18, 28, 54). These stud-
ies suggest that the spatial positioning as well as
timing of formation of these two radicals will de-
termine the extent of RNOS-mediated oxidation
and that oxidative stress and nitrosative stress
are balanced within the NO/O
2
2
reaction.
Therefore, nitrosative stress may provide an op-
timal antioxidant environment.
Lipid peroxidation
The process of lipid peroxidation results in
formation of a variety of lipid-oxy and -peroxy
adducts (21). Perpetuation of lipid oxidation by
these species can result in cell membrane com-
promise (Fig. 3). Reaction of NO with these per-
oxy and oxy radicals terminates lipid peroxi-
dation via Eq. 2 and results in protection
against ROS (45):
LOO
?
1 NO
R
LOONO (2)
where LOO
?
is the lipid peroxy radical. Chain
termination also prevents oxidation of low
density lipoproteins in both endothelial cells
(52) and macrophages (26). Reduction in oxi-
dized cholesterol is thought to reduce initiation
of atheroscleroses mediated by foaming mac-
rophages. Other inflammatory processes such
as production of leukotrienes are also effected
by NO (54, and references therein). Lipoxyge-
nase, which mediates a variety of lipid oxida-
tion, is inhibited by NO (54, and references
therein). Thus, termination of lipid peroxida-
tion may be one of the most important antiox-
idant properties of NO.
CHEMICAL TOXICOLOGY OF
NO AND ROS
Although NO often acts as an antioxidant,
the chemistry of NO can also affect cellular pro-
cesses such that cells or tissue becomes more
susceptible to oxidative stress. This complexity
of potential NO reactions within the cell has led
to differing opinions on the role of NO in ox-
idative stress (23). The effect of NO on the tox-
icity of hydrogen peroxide (H
2
O
2
), alkylhy-
droperoxide, and O
2
2
at the cellular level was
therefore examined by clonogenic assay, which
is the primary means for determining the cy-
totoxicity of chemical substances as it accounts
for both necrotic and apoptotic death.
Cytotoxic effects of NO/H
2
O
2
H
2
O
2
mediates oxidation of biological mole-
cules, which can result in tissue damage. Al-
though NO does
not
react directly with H
2
O
2
(55), it can protect cells against H
2
O
2
-mediated
toxicity (19, 55, 56, 57). Exposure of lung fibro-
WINK ET AL.
206
FIG. 2. Chemistry of reaction between NO and O
2
2
.
FIG. 3. Mechanism for termination of lipid peroxida-
tion by NO.
blasts to increasing concentrations of H
2
O
2
in-
duced marked increases in cytotoxicity (55).
Addition of NONOate compounds surpris-
ingly resulted in protection against the cyto-
toxicity of H
2
O
2
(55). Pre- or posttreatment
with these NO donor complexes did not result
in protection; in fact, the by-product of the de-
composition of NO, nitrite, increased the cy-
totoxicity of H
2
O
2
. Similar observations were
made in neuronal (55), hepatoma (56), and en-
dothelial cells (6, 19).
These protective effects of NO were not re-
stricted to NONOates, as endogenous forma-
tion of NO in endothelial cells was also sug-
gested to be involved in protection against
damage to vascular smooth muscle mediated
by H
2
O
2
(37). Further, the
S
-nitroso-containing
compounds
S
-nitrosothiolglutathione (GSNO)
and
S
-nitroso-
N
-acetylpenicillamine (SNAP)
also protected against H
2
O
2
-mediated toxicity
(58). However, nitrovasodilators commonly
used in the clinic, such as 3-morpholinosyd-
nonimine (SIN-1) and sodium nitroprusside
(SNP), increased the toxicity of H
2
O
2
(14, 58).
Angeli’s salt (AS; Na
2
N
2
O
3
), which is struc-
turally similar to the NONOate compounds but
donates nitroxyl (NO
2
) instead of NO, signifi-
cantly potentiated the toxicity of H
2
O
2
(58).
These results demonstrate that the common
putative NO donors modulate the toxicity of
H
2
O
2
differently, and so caution is required in
the interpretation of experimental results.
The diverse effects exhibited by the NO
donors examined may be explained by vari-
ance in donor effect on cellular antioxidant de-
fenses and in amount and flux of NO produced.
One of the major cellular defenses against H
2
O
2
is consumption of this oxidant by the enzymes
glutathione peroxidase and catalase (21, 22).
When the kinetics for the disappearance of
H
2
O
2
were examined in the presence of the dif-
ferent NO donors, cellular consumption of
H
2
O
2
was noted to be inhibited to varying de-
grees by several of the donors. For instance, the
amount of time required to decompose 0.75
m
M
H
2
O
2
in the presence of SNP, the
NONOate Et
2
NN(O)NO (DEA/NO), AS, and
SNAP increased by 30–200% (58). Conversely,
SIN-1 and GSNO retarded H
2
O
2
consumption
by as much as 400%. Thus, enhancement of
H
2
O
2
-mediated toxicity by AS and SIN-1 might
be partially explained by inhibition of H
2
O
2
consumption. However, this cannot be the sole
mechanism by which NO enhances or protects
against H
2
O
2
, because GSNO, SNAP, and
DEA/NO also decreased the rate of decompo-
sition of H
2
O
2
, but proved to be cytoprotective.
Furthermore, cellular exposure to the NO
donors resulted in varied reduction of intra-
cellular levels of GSH. Exposure of V79 cells to
1 m
M
nitrite, SNAP, SIN-1, GSNO, DEA/NO,
or AS for 1 h resulted in varying degrees of de-
pletion of intracellular GSH (60). For instance,
SNAP, GSNO, and DEA/NO resulted in only
modest GSH decreases (,30%), which were re-
covered rapidly. Conversely, SIN-1 and AS de-
creased intracellular GSH levels by as much as
85%, whereas nitrite decreased GSH levels by
50% in these cells. This reduction in cellular
GSH may be the mechanism by which SIN-1,
AS, and nitrite enhance H
2
O
2
toxicity.
The observed differences in protective effects
among the various chemical NO donors may
also be a reflection of the actual flux of NO pro-
duced by each compound. The temporal pro-
files of NO release by the different compounds,
as assessed with an NO-selective electrode,
demonstrate that the amount of NO released
over time is quite variant (58). Both the
NONOates and
S
-nitroso complexes, which
protected against H
2
O
2
toxicity, released NO
over the entire time course of exposure to H
2
O
2
(1 h). However, SIN-1, SNP, and AS did not
produce measurable NO (,1 m
M
) under these
experimental conditions, coincident with a lack
of protection against H
2
O
2
.
SNP appears to increase the toxicity of ROS
by yet another mechanism. Chemical interaction
with SNP can result in formation of not only NO,
but also cyanide (CN
2
) and free iron. The iron
chelator desferrioxamine (DF) completely pro-
tected cells from H
2
O
2
, yet only partially pro-
tected against the toxicity of H
2
O
2
combined
with SNP (58). This discrepancy may be ac-
counted for by enhanced release of CN
2
from
SNP. Monocytes and polymorphonuclear leuko-
cytes have also been shown to facilitate release
of CN
2
from SNP, which is a phenomenon be-
lieved to be mediated by H
2
O
2
(5). A transition
metal complex with a labile ligand was sug-
gested to oxidize substrates further via Fenton-
type catalysis (58). Further evidence supporting
MECHANISMS OF THE ANTIOXIDANT EFFECTS OF NO
207
this hypothesis comes from Imlay and col-
leagues, who showed that bacteria became more
sensitive to H
2
O
2
in the presence of CN
2
(27).
The fact that DF completely protected against
the toxicity of CN
2
suggests that metal–perox-
ide reactions are required to initiate cytotoxic-
ity. Thus, the DF-insensitive enhancement of
H
2
O
2
-mediated toxicity by SNP could be attrib-
uted to an iron complex, which cannot be bound
by DF, and so could catalyze the Fenton oxida-
tion chemistry of cellular molecules.
Cytotoxic effects of NO/alkylperoxides
NO was also found to act as an antioxidant
against XO-mediated lipid peroxidation (49). We
examined the effect of NO on organic hy-
droperoxide-mediated toxicity, which is thought
to be mediated by oxidation of lipophilic mem-
branes (57). Our studies further illustrate the im-
portance NO production for the duration of ex-
posure to oxidants. DEA/NO, which is a
NONOate with a half-life of
,
2 min at 37°C, pro-
vided minimal protection against either
tert-
butyl hydroperoxide or cumene hydroperoxide,
whereas the longer lived PAPA/NO (half-life of
15 min at 37°C) exhibited marked protection. The
different effects of these two NONOates on cy-
totoxicity can be attributed to the timing of de-
livery of NO as exposure to organic peroxides
usually requires up to 2 h to induce observable
toxicity (57). As alkylhydroperoxides require
longer times to penetrate cells and thereby exert
their damage, longer sustained fluxes of NO are
more protective.
Several potential mechanisms may be in-
volved in the protection by NO against organic
hydroperoxide-mediated toxicity. Intracellular
metalloproteins, such as those containing heme
moieties, react quickly with organic peroxides
to form hypervalent complexes (22). Upon de-
composition, these complexes release intracel-
lular iron, which in turn can catalyze damage
to macromolecules such as DNA. NO can react
with these hypervalent metalloproteins at near
diffusion control, which may restore these ox-
idized species to the ferric form (16, 50). Re-
duction of these metal-oxo proteins prevents
both their oxidative chemistry and their release
of free iron (16, 29, 56), thus limiting intracel-
lular damage mediated by oxidative stress.
Interestingly, although NO can protect
against the toxicity of H
2
O
2
in mammalian cells,
the opposite effect is observed in
E. coli
. Deliv-
ery of H
2
O
2
either as a bolus or through the en-
zymatic activity of XO, exhibited only modest
bactericidal activity (44). However, simultane-
ous exposure to both H
2
O
2
and NO, delivered
either as gas or by a NONOate complex, in-
creased bactericidal activity by four orders of
magnitude. Addition of catalase abated toxicity,
whereas exposure to superoxide dismutase
(SOD) had no effect. These results demonstrate
that the synergistic action of NO and H
2
O
2
rather than O
2
2
was responsible for this bacte-
ricidal activity. Thus, the combination of NO
and H
2
O
2
may be ideally suited to combat
E. coli
infections due to the additional protective effect
of NO for the host. This mechanism may also
apply to other species of bacteria, albeit with dif-
ferent kinetics. For instance, the cytotoxicity of
O
2
2
in
Staphylococcus
was abrogated by NO at
early time points, yet NO helped sustain toxicity
at longer time intervals. The maximal effect was
dependent upon the timing of exposure to NO,
H
2
O
2
and O
2
2
(30). These findings may explain
why NO and ROS are produced by immune ef-
fector cells at different times following exposure
to different pathogens.
The diametric responses of mammalian cells
and prokaryotes to the combination of
NO/H
2
O
2
may reflect their different cellular
structures and complement of metalloproteins.
Bacteria utilize iron-sulfur clusters to a greater
extent than do mammalian cells, and these
types of proteins are especially susceptible to
degradation mediated by NO or RNOS (12, 24).
In
E. coli
, decomposition of iron complexes oc-
curs in the periplasmic space, which is in close
proximity to the cytoplasm. This relative lack of
compartmentalization may allow iron to bind
to and oxidize DNA. However, due to the or-
ganelle structure of mammalian cells, metal la-
bilization may be limited to the cytoplasm and
mitochondria. In such a cellular arrangement,
metals would be required to travel a large dis-
tance to reach the nucleus and bind to DNA.
Cytotoxic effects of NO/O
2
2
Treatment of cells with ONOO
2
results in
cell death in both bacterial (62) and mammalian
WINK ET AL.
208
systems (reviewed in 46). However, treatment
of lung fibroblasts and neurons with O
2
2
and
NO did not exhibit appreciable toxicity (55).
Other studies showed that ovarian carcinoma
cells exposed to 5 m
M
SIN-1, which is a si-
multaneous generator of NO/O
2
2
, did not re-
sult in appreciable toxicity (14). In fact, cells
treated concomitantly with O
2
2
and NO re-
leasing compounds resulted in protection
against O
2
2
-mediated toxicity and
did not
dis-
play appreciable toxicity due to ONOO
2
for-
mation (58). These results suggest that there
is a distinct difference between treating cells
with bolus concentrations of synthetic ONOO
2
(millimolar) and generating ONOO
2
with
NO/O
2
2
systems.
Part of the discrepancy between bolus ad-
ministration and
de novo
synthesis of ONOO
2
can be explained in terms of reactant concen-
trations. Beckman and co-workers described
that penetration into cells by bolus treatment
required high concentrations of ONOO
2
be-
cause the cell membrane forms a formidable
barrier for ONOO
2
penetration to intracellular
targets (62). Although extracellular generation
of NO and O
2
2
results in ONOO
2
formation,
the short lifetime of this species in solution does
not allow accumulation of high enough con-
centrations to penetrate the cell. Therefore, the
amount of ONOO
2
that could cross the cellu-
lar membrane under more biologically relevant
conditions than bolus administration, despite
production of stoichiometrically high amounts
over a prolonged time period, is dramatically
reduced. The cell membrane thus limits the
contribution of extracellular ONOO
2
to toxi-
cological mechanisms.
Another factor to consider with respect to tox-
icity mediated directly by ONOO
2
chemistry is
that the reaction between NO and ONOO
2
forms NO
2
, as was discussed above. Competi-
tion for O
2
2
by cellular components such as SOD
and redox proteins increases the amount of NO
required to form ONOO
2
. As the NO flux then
exceeds the O
2
2
flux, ONOO
2
is converted to
potent nitrosating agents. Hence, the chemistry
of extracellular formation of ONOO
2
by excess
NO converts ONOO
2
to nitrite. Direct necrotic
cell death mediated by oxidative chemistry of
ONOO
2
from exposure of simultaneous
NO/O
2
2
derived from NADPH is thus unlikely.
CELLULAR AND PHYSIOLOGICAL
EFFECTS OF NO AND
OXIDATIVE STRESS
In addition to abating oxidative stress chem-
ically, NO can protect against oxidative stress
at the cellular level. For example, pretreatment
with NO several hours prior to exposure can
confer protection against H
2
O
2
(43). In bacte-
ria, treatment with NO resulted in up-regula-
tion of the SOXR and subsequent expression of
protective proteins against ROS (43). Hepato-
cytes also became resistant to peroxide injury
following treatment with NO (33). Expression
of hemoxygenase as well as other enzymes may
participate in NO-induced protection against
peroxide (33). In addition to stimulating ex-
pression of protective proteins, NO has a num-
ber of effects on cellular functions that may
influence oxidative stress. For instance, the
presence of NO may limit ROS production by
preventing assembly of NADPH oxidase,
which is one of the major contributors to ox-
idative stress in various immune responses (9).
In addition to limiting interfering with ox-
idative stress, NO has been shown to prevent
the induction of some genes that are induced
from ROS. For instance, oxidative stress can in-
duce early growth response-1, which through
the extracellular signal–regulated kinase path-
way can activate a number of adhesion mole-
cules. It has been shown that NO derived from
a donor or endothelial NO synthase (eNOS)
abates the up-regulation of this system (7).
NO can also inhibit oxidative stress pro-
cesses at physiological levels, which can result
in protection against tissue injury. NO has been
shown to attenuate ischemia–reperfusion in-
jury in numerous organs such as heart, intes-
tine, liver, kidney, and lung. To illustrate, NO,
either from donors or produced endogenously,
was shown to protect against myocardial isch-
emia–reperfusion injury (39, 51). Endogenous
NO can also attenuate lung damage after in-
testinal ischemia. These reports suggest that
under different conditions where ischemia–
reperfusion is involved NO can abate tissue
damage.
Neutrophil and leukocyte adhesion is also
inhibited by NO (34, 39). Leukocyte adhesion
and extravasation to the endothelium is one of
MECHANISMS OF THE ANTIOXIDANT EFFECTS OF NO
209
the primary mechanisms by which tissue dam-
age is mediated in ischemia–reperfusion. Pre-
venting an increase in activated leukocytes in
these regions is thought to protect tissue from
ischemia–reperfusion injury (34, 39). Increased
leukocyte adhesion and infiltration involve a
number of proinflammatory molecules, in-
cluding vascular cell adhesion molecule, in-
tercellular adhesion molecule (ICAM), and
CD11/CD18 expression (1, 11, 32, 53), and are
induced by ROS formation such as superfused
XO in the rat mesentery (15). This response
could be inhibited by treatment with either
SOD or NO donors, which scavenged O
2
2
, thus
preventing activation of the cell adhesion mol-
ecule, P-selectin. It also has been shown that
monocyte chemotactic protein-1 induced by
oxidative stress is inhibited by NO derived ei-
ther from an NO donor or from NOS-3 (53). In
summary, oxidative stress-induced leukocyte
adhesion mechanisms are inhibited by the
presence of NO.
NO also abates oxidative chemistry from
ROS, other than O
2
2
, which can lead to leuko-
cyte adhesion. Oxidants from the Fenton reac-
tion enhance production of the proinflamma-
tory molecules platelet-activating factor and
leukotriene B4, as well as increase the adhesion
molecules P-selectin, E-selectin, and ICAM-1
(35). Oxidant activation of platelet activating
factor and leukotriene B4 can activate nuclear
factor-kB, which can result in synthesis of in-
terleukin-8 and other adhesion molecules. In-
hibition by NO of the oxidative chemistry in an
analogous manner as described above for pro-
tection against peroxide-mediated toxicity also
prevents the cascade of events that leads to in-
creased leukocyte activation. Thus, the ability of
NO to limit ROS chemistry prevents expression
of different proinflammatory molecules and
subsequently limits leukocyte activation and ox-
idative stress in ischemia–reperfusion-mediated
tissue injury.
The role of NO in ischemia–reperfusion in-
jury in the brain has been controversial (13).
ONOO
2
toxicity has been invoked as the
causative factor responsible for reperfusion-
mediated tissue damage (2, 10). However, as
discussed above, NO can be a powerful anti-
oxidant. A recent study assessed the value of
administering NO donor complexes at the
time of reperfusion in both a focal and global
brain ischemia–reperfusion injury in rat mod-
els (40). The major problem with systemic ad-
ministration of most NO donors is that the re-
sulting dramatic reduction of blood pressure
places the patient at risk. This problem can be
overcome by delivering NONOates with very
short half-lives directly to the brain through
intracarotid injection.
Infusion of DEA/NO in this manner, at the
time of reperfusion subsequent to 5–20 min of
ischemia, resulted in a dramatic decrease of the
infarct area (40). A concurrent dramatic reduc-
tion in salicylate oxidation was also observed.
Electrochemical monitoring of the brain NO
levels showed that an initial NO burst (from
eNOS) declined with time in the absence of NO
donors. This reduction of NO production cor-
responded to an increase in salicylate oxida-
tion. In the presence of the NO donor, NO
levels were maintained throughout the experi-
ment, and salicylate oxidation was reduced.
From these data it was concluded that the pri-
mary function of NO in ischemia–reperfusion
injury is as an antioxidant, which in part was
responsible for reduction in infarct size. There-
fore, NO donors have two benefits: restoration
of blood flow and limiting oxidative stress in
the injured tissue. Endogenous production of
NO by eNOS limits infarct volume as do other
antioxidants, such as Tempol (40, 47). These
studies suggest that NO has a powerful anti-
oxidant role in ischemia–reperfusion injury in
the brain. Thus, for stroke, NO administration
following treatment with antithrombotic
agents may be beneficial.
Although the role of NO in ischemia–reper-
fusion injury represents acute tissue damage,
atherosclerosis depicts a chronic disease that is
mediated by oxidative stress. Although ather-
osclerosis has been proposed to be improved
by the presence of NO, the positive role of NO
in such diseases may be multifaceted, exerting
protective effects chemically, biochemically,
and physiologically (25, 61). Chemically, NO
can prevent oxidative damage and limit lipid
peroxidative chain propagation (25, 48, 49).
This is similar to the factors that limit alkylhy-
droperoxide toxicity as discussed above. Ter-
WINK ET AL.
210
mination of lipid peroxidation by NO may de-
crease low-density lipid oxidation whose prod-
ucts promote foaming macrophage and in-
crease plaques. Furthermore, low-density lipid
oxidation can lead to mast cell degranulation,
which can induce further inflammation. Bio-
logically, NO can inhibit platelet aggregation,
prevent mast cell degranulation and leukocyte
adhesion, and thereby is beneficial with respect
to diseases that involve lipid oxidation (42).
Thus, NO may have multiple effects in main-
taining the health of the vascular wall, which
may play a critical role in a number of acute
and chronic injuries to tissue.
CONCLUSION
The ability of NO to abate oxidative stress
plays an important role in both physiological
and pathophysiological mechanisms involving
properties at the chemical, cellular, and physi-
ological levels. The relatively low concentra-
tion of NO required to be an antioxidant sug-
gests that in addition to its involvement with
cyclic GMP, this radical molecule serves to
counterbalance oxidative stress. This balance
between NO and oxidative stress provides an
important regulatory mechanism in numerous
physiological effects. Imbalance in this redox
symbiotic relationship can lead to different
pathophysiological conditions.
ABBREVIATIONS
AS, Angeli’s salt (Na
2
N
2
O
3
); CN
2
, cyanide;
CO
2
, carbon dioxide; DEA/NO, Et
2
NN(O)NO
2
;
DF, desferrioxamine; eNOS, endothelial nitric
oxide synthase; GSH, glutathione; GSNO,
S
-
nitrosothiolglutathione; H
2
O
2
, hydrogen per-
oxide; ICAM, intercellular adhesion molecule;
LOO
?
, lipid peroxy radical; NO, nitric oxide;
NO
2
, nitrogen dioxide; O
2
2
, superoxide;
ONOO
2
, peroxynitrite; RNOS, reactive nitro-
gen oxide species; ROS, reactive oxygen species;
SIN-1, 3-morpholinosydnonimine; SNAP,
S
-ni-
troso-
N
-acetylpenicillamine; SNP, sodium ni-
troprusside; SOD, superoxide dismutase; XO,
xanthine oxide/hypoxanthine.
REFERENCES
1. Bao X, Lu C, and Frangos JA. Temporal gradient in
shear but not steady shear stress induces PDGF-A and
MCP-1 expression in endothelial cells: role of NO,
NFkB, and egr-1. Arterioscler Thromb Vasc Biol 19: 996–
1003, 1999.
2. Beckman JS. Ischemic injury mediator. Nature 345:
27–28, 1990.
3. Beckman JS. The double-edged role of nitric oxide in
brain function and superoxide-mediated injury. J Dev
Physiol 15: 53–59, 1991.
4. Buxton GV, Greenstock CL, Phillip W, and Ross AB.
Critical review of rate constants for reactions of hy-
drated electrons, hydrogen atoms and hydroxyl rad-
icals in aqueous solution J Phys Chem Ref Data 17:
513–886, 1988.
5. Campbell JM, McCrae F, Reglinski J, Wilson R, Smith
WE, and Sturrock RD. The interaction of sodium ni-
troprusside with peripheral white blood cells in vitro:
a rationale for cyanide release in vivo. Biochim Biophys
Acta 1156: 327–333, 1993.
6. Chang J, Rao NV, Markewitz BA, Hoidal JR, and
Michael JR. Nitric oxide donor prevents hydrogen
peroxide-mediated endothelial cell injury. Am J Phys-
iol 270: L931–L940, 1996.
7. Chiu JJ, Wung BS, Hsieh HJ, Lo LW, and Wang DL.
Nitric oxide regulates shear stress–induced early
growth response-1 expression via the extracellular
signal–regulated kinase pathway in endothelial cells.
Circ Res 85: 238–246, 1999.
8. Choi DW. Nitric oxide: foe or friend to the injury of
brain. Proc Natl Acad Sci U S A 53: 9741–9743, 1993.
9. Clancy RM, Leszczynska-Piziak J, and Abramson SB.
Nitric oxide, an endothelial cell relaxation factor, in-
hibits neutrophil superoxide anion production via a
direct action on the NADPH oxidase. J Clin Invest 90:
1116–1121, 1992.
10. Dawson DA. Nitric oxide and focal cerebral ischemia:
Multiplicity of actions and diverse outcome. Cere-
brovasc Brain Metab Rev 6: 299–324, 1994.
11. De Caterina R, Libby P, Peng HB, Thannickal VJ,
Rajavashisth TB, Gimbrone MA Jr, Shin WS, and
Liao JK. Nitric oxide decreases cytokine-induced en-
dothelial activation. Nitric oxide selectively reduces
endothelial expression of adhesion molecules and
proinflammatory cytokines. J Clin Invest 96: 60–68,
1995.
12. Drapier J-C, and Bouton C. Modulation by nitric ox-
ide of metalloprotein regulatory activities. Bioessays
18: 1–8, 1996.
13. Espey MG, Miranda KM, Feelisch M, Fukuto J,
Grisham MB, Vitek MP, and Wink DA. Mechanisms
of cell death governed by the balance between ni-
trosative and oxidative stress. Ann N Y Acad Sci 899:
209–221, 2000.
14. Farias-Eisner R, Chaudhuri G, Aeberhard E, and
Fukuto JM. The chemistry and tumoricidal activity of
nitric oxide/hydrogen peroxide and the implications
MECHANISMS OF THE ANTIOXIDANT EFFECTS OF NO
211
to cell resistance susceptibility. J Biol Chem 271:
6144–6151, 1996.
15. Gaboury J, Woodman RC, Granger DN, Reinhardt P,
and Kubes P. Nitric oxide prevents leukocyte adher-
ence: role of superoxide. Am J Physiol 265: H862–H867,
1993.
16. Gorbunov NV, Osipov AN, Day BW, Zayas-Rivera B,
Kagan VE, and Elsayed NM. Reduction of ferryl-myo-
globin and ferrylhemoglobin by nitric oxide: a pro-
tective mechanism against ferryl hemoprotein-in-
duced oxidations. Biochemistry 34: 6689–6699, 1995.
17. Gow A, Duran D, Thom SR, and Ischiropoulos H.
Carbon dioxide enhancement of peroxynitrite-medi-
ated protein tyrosine nitration. Arch Biochem Biophys
333: 42–48, 1996.
18. Grisham MB, Jourd’Heuil D, and Wink DA. Nitric ox-
ide. I. Physiological chemistry of nitric oxide and its
metabolites: implications in inflammation. Am J Phys-
iol 276: G315–G321, 1999.
19. Gupta MP, Evanoff V, and Hart CM. Nitric oxide at-
tenuates hydrogen peroxide-mediated injury to por-
cine pulmonary artery endothelial cells. Am J Physiol
272: L1133–L1141, 1997.
20. Haber F, and Weiss JJ. The catalytic decomposition of
hydrogen peroxide by iron salts. Proc R Soc Lond [A]
147: 332–336, 1934.
21. Halliwell B. Reactive oxygen species in living sys-
tems: source, biochemistry, and role in human dis-
ease. Am J Med 91: 14S–22S, 1991.
22. Halliwell B, and Gutteridge JMC. Free radicals, aging
and disease. Free Radic Biol Med 416–509, 1989.
23. Halliwell B, Zhao K, and Whiteman M. Nitric oxide
and peroxynitrite. The ugly, the uglier and the not so
good: a personal view of recent controversies. Free
Radic Res 31: 651–669, 1999.
24. Hentze MW, and Kuhn LC. Molecular control of ver-
tebrate iron metabolism: mRNA-based regulatory cir-
cuits operated by iron, nitric oxide, and oxidative
stress. Proc Natl Acad Sci U S A 93: 8175–8182, 1996.
25. Hogg N, Kalyanaraman B, Joseph J, Struck A, and
Parthasarathy S. Inhibition of low-density lipoprotein
oxidation by nitric oxide. Potential role in atherogen-
esis. FEBS Lett. 334: 170–174, 1993.
26. Hogg N, Struck A, Goss SP, Santanam N, Joseph J,
Parthasarathy S, and Kalyanaraman B. Inhibition of
macrophage-dependent low density lipoprotein oxi-
dation by nitric oxide donors. J Lipid Res 36: 1756–
1762, 1995.
27. Imlay JA, Chin SM, and Linn S. Toxic DNA damage
by hydrogen peroxide through the Fenton reaction in
vivo and in vitro. Science 240: 640–642, 1988.
28. Jourd’heuil D, Miranda KM, Kim SM, Espey MG,
Vodovotz Y, Laroux S, Mai CT, Miles AM, Grisham
MB, and Wink DA. The oxidative and nitrosative
chemistry of the nitric oxide/superoxide reaction in
the presence of bicarbonate. Arch Biochem Biophys 365:
92–100, 1999.
29. Kanner J, Harel S, and Granit R. Nitric oxide as an
antioxidant. Arch Biochem Biophys 289: 130–136, 1991.
30. Kaplan SS, Lancaster JR, Basford RE, and Simmons
RL. Effect of nitric oxide on staphylococcal killing and
interactive effect with superoxide. Infect Immun 64:
69–76, 1996.
31. Keefer LK, Nims RW, Davies KW, and Wink DA.
NONOates (diazenolate-2-oxides) as nitric oxide dos-
age forms. Methods Enzymol 268: 281–294, 1996.
32. Khan BV, Harrison DG, Olbrych MT, and Alexander
RW. Nitric oxide regulates vascular cell adhesion
molecule 1 gene expression and redox-sensitive tran-
scriptional events in human vascular endothelial cells.
Proc Natl Acad Sci U S A 93: 9114–9119, 1996.
33. Kim YM, Bergonia H, and Lancaster JR Jr. Nitrogen
oxide-induced autoprotection in isolated rat hepato-
cytes. FEBS Lett 374: 228–232, 1995.
34. Kubes P, Suzuki M, and Granger DN. Nitric oxide: an
endogenous modulator of leukocyte adhesion. Proc
Natl Acad Sci U S A 88: 4651–4655, 1991.
35. Kurose I, Wolf R, Grisham MB, Aw TY, Specian RD,
and Granger DN. Microvascular responses to inhibi-
tion of nitric oxide production: role of active oxidants.
Circ Res 76: 30–39, 1995.
36. Lemercier JN, Padmaja S, Cueto R, Squadrito GL,
Uppu RM, and Pryor WA. Carbon dioxide modula-
tion of hydroxylation and nitration of phenol by per-
oxynitrite. Arch Biochem Biophys 345: 160–170, 1997.
37. Linas SL, and Repine JE. Endothelial cells protect vas-
cular smooth muscle cells from H
2
O
2
attack. Am J
Physiol 272: F767–F773, 1997.
38. Lymar SV, and Hurst K. Rapid reaction between per-
oxynitrite ion and carbon dioxide: implications for bi-
ological activity. J Am Chem Soc 290: 52–57, 1995.
39. Ma X-L, Weyrich AS, Lefer DJ, and Lefer AM. Di-
minished basal nitric oxide release after mycardial
ischemia and reperfusion promotes neutrophil ad-
herence to coronary endothelium. Circ Res 72:
403–412, 1993.
40. Mason R, Pluta RM, Walbridge S, Wink DA, Oldfield
EH, and Boock RJ. The protective effect of nitric ox-
ide on reactive oxygen species production following
reperfusion injury: in vitro and in vivo. J Neurosurg 93:
99–107, 2000.
41. Miles AM, Bohle DS, Glassbrenner PA, Hansert B,
Wink DA, and Grisham MB. Modulation of superox-
ide-dependent oxidation and hydroxylation reactions
by nitric oxide. J Biol Chem 271: 40–47, 1996.
42. Moncada S, Palmer RMJ, and Higgs EA. Nitric oxide:
physiology, pathophysiology, and pharmacology.
Pharmacol Rev 43: 109–142, 1991.
43. Nunoshiba T, DeRojas-Walker T, Wishnok JS, Tan-
nenbaum SR, and Demple B. Activation by nitric ox-
ide of an oxidative-stress response that defends Esch-
erichia coli against macrophages. Proc Natl Acad Sci
U S A 90: 9993–9997, 1993.
44. Pacelli R, Wink DA, Cook JA, Krishna MC, DeGraff
W, Friedman N, Tsokos M, Samuni A, and Mitchell
JB. Nitric oxide potentiates hydrogen peroxide-
induced killing of Escherichia coli. J Exp Med 182: 1469–
1479, 1995.
45. Padmaja S, and Huie RE. The reaction of nitric oxide
with organic peroxyl radicals. Biochem Biophys Res
Commun 195: 539–544, 1993.
46. Pryor WA, and Squadrito GL. The chemistry of per-
WINK ET AL.
212
oxynitrite and peroxynitrous acid: Products from the
reaction of nitric oxide with superoxide. Am Phys J
268: L699–L721, 1996.
47. Rak R, Chao DL, Pluta RM, Mitchell JB, Oldfield EH,
and Watson JC. Neuroprotection by the stable ni-
troxide Tempol during reperfusion in a rat model
of transient focal ischemia. J Neurosurg 92: 646–651,
2000.
48. Rubbo H, Radi R, Trujillo M, Telleri R, Kalyanaraman
B, Barnes S, Kirk M, and Freeman BA. Nitric oxide
regulation of superoxide and peroxynitrite dependent
lipid peroxidation: formation of novel nitrogen con-
taining oxidized lipid derivatives. J Biol Chem 269:
26066–26075, 1994.
49. Rubbo H, Parthasarathy S, Barnes S, Kirk M, Kalya-
naraman B, and Freeman BA. Nitric oxide inhibition
of lipoxygenase-dependent liposome and low-density
lipoprotein oxidation: termination of radical chain
propagation reactions and formation of nitrogen-con-
taining oxidized lipid derivatives. Arch Biochem Bio-
phys 324: 15–25, 1995.
50. Shvedova AA, Tyurina YY, Gorbunov NV, Tyurin
VA, Castranova V, Kommineni C, Ojimba J, Gandley
R, McLaughlin MK, and Kagan VE. tert-Butyl hy-
droperoxide/hemoglobin-induced oxidative stress
and damage to vascular smooth muscle cells: differ-
ent effects of nitric oxide and nitrosothiols. Biochem
Pharmacol 57: 989–1001, 1999.
51. Siegfried MR, Erhardt J, Rider T, Ma XL, and Lefer
AM. Cardioprotection and attenuation of endothelial
dysfunction by organic nitric oxide donors in myo-
cardial ischemia–reperfusion. J Pharm Exp Ther 260:
668–675, 1992.
52. Struck AT, Hogg N, Thomas JP, and Kalyanaraman
B. Nitric oxide donor compounds inhibit the toxicity
of oxidized low-density lipoprotein to endothelial
cells. FEBS Lett 361: 291–294, 1995.
53. Tsao PS, Wang B-Y, Buitrago R, Shyy J-Y, and Cooke
JP. Nitric oxide regulates monocyte chemotactic pro-
tein-1. Circulation 96: 934–940, 1997.
54. Wink DA, and Mitchell JB. The chemical biology of
nitric oxide: insights into regulatory, cytotoxic and cy-
toprotective mechanisms of nitric oxide. Free Radic
Biol Med 25: 434–456, 1998.
55. Wink DA, Hanbauer I, Krishna MC, DeGraff W, Gam-
son J, and Mitchell JB. Nitric oxide protects against
cellular damage and cytotoxicity from reactive oxy-
gen species. Proc Natl Acad Sci U S A 90: 9813–9817,
1993.
56. Wink DA, Hanbauer I, Laval F, Cook JA, Krishna MC,
and Mitchell JB. Nitric oxide protects against the cy-
totoxic effects of reactive oxygen species. Ann NY
Acad Sci 738: 265–278, 1994.
57. Wink D, Cook JA, Krishna MC, Hanbauer I, DeGraff
W, Gamson J, and Mitchell JB. Nitric oxide protects
against alkyl peroxide-mediated cytotoxicity: further
insights into the role nitric oxide plays in oxidative
stress. Arch Biochem Biophys 319: 402–407, 1995.
58. Wink DA, Cook J, Pacelli R, DeGraff W, Gamson J,
Liebmann J, Krishna M, and Mitchell JB. Effect of var-
ious nitric oxide-donor agents on peroxide mediated
toxicity. A direct correlation between nitric oxide for-
mation and protection. Arch Biochem Biophys 331:
241–248, 1996.
59. Wink DA, Cook JA, Kim S, Vodovotz Y, Pacelli R,
Krishna MC, Russo A, Mitchell JB, Jourd’heuil D,
Miles AM, and Grisham MB. Superoxide modulates
the oxidation and nitrosation of thiols by nitric oxide
derived reactive intermediates. J Biol Chem 272: 11147–
11151, 1997.
60. Wink DA, Vodovotz Y, DeGraff W, Cook JA, Pacelli
R, Krishna MC, and Mitchell JB. Protective effects of
NO against oxidative injury. In: Nitric Oxide and In-
fection, edited by Fang F. New York: Plenum Press,
2000, pp. 259–272.
61. Yates MT, Lambert LE, Whitten JP, McDonald I, Mano
M, Ku G, and Mao SJT. A protective role for nitric ox-
ide in the oxidative modification of low density
lipoproteins by mouse macrophage. FEBS Lett 309:
135–138, 1992.
62. Zhu L, Gunn C, and Beckman JS. Bactericidal activ-
ity of peroxynitrite. Arch Biochem Biophys 298: 452–457,
1992.
Address reprint requests to:
David A. Wink, Ph.D.
Radiation Biology Branch
National Institutes of Health/National Cancer
Institute
Building 10, Room B3-B69
Bethesda, MD 20892
E-mail:
Wink@mail.nih.com
Received for publication October 2, 2000; ac-
cepted November 20, 2000.
MECHANISMS OF THE ANTIOXIDANT EFFECTS OF NO
213
