729
Biochem. J. (1999) 339, 729–736 (Printed in Great Britain)
Peroxynitrite induces haem oxygenase-1 in vascular endothelial cells :
a link to apoptosis
Roberta FORESTI1, Padmini SARATHCHANDRA, James E. CLARK, Colin J. GREEN and Roberto MOTTERLINI
Vascular Biology Unit, Department of Surgical Research, Northwick Park Institute for Medical Research, Harrow, Middlesex HA1 3UJ, U.K.
Peroxynitrite (ONOO−) is a potent oxidizing agent generated by
the interaction of nitric oxide (NO) and the superoxide anion. In
physiological solution, ONOO− rapidly decomposes to a hydroxyl radical, one of the most reactive free radicals, and nitrogen
dioxide, another species able to cause oxidative damage. In the
present study we investigated the effect of ONOO− on the
expression of haem oxygenase-1 (HO-1), an inducible protein
that is highly up-regulated by oxidative stress. Exposure of
bovine aortic endothelial cells to ONOO− (250–1000 µM) produced a concentration-dependent increase in haem oxygenase
activity and HO-1 protein expression. This effect was completely
abolished by the ONOO− scavengers uric acid and N-acetylcysteine, and partly attenuated by 1,3-dimethyl-2-thiourea, a
scavenger of hydroxyl radicals. ONOO− also produced a concentration-dependent increase in apoptosis and cytotoxicity, which
were considerably decreased by uric acid and N-acetylcysteine. A
70 % decrease in apoptosis was observed when cells were exposed
to ONOO− in the presence of 10 µM tin protoporphyrin IX
(SnPPIX), an inhibitor of haem oxygenase activity. When
SnPPIX was added 5 min after ONOO−, apoptosis decreased by
only 40 %, which suggests that an interaction between ONOO−
and the protoporphyrin occurs in our system. Increased haem
oxygenase activity by pretreatment of cells with haemin resulted
in elevated bilirubin production and was associated with a
substantial decrease (35 %) in ONOO−-mediated apoptosis.
These results indicate the ability of ONOO− to modulate the
expression of the stress protein HO-1 and suggest that the haem
oxygenase pathway contributes to protection against the cytotoxic action of ONOO−.
INTRODUCTION
in studies in itro, ONOO− has been shown to be highly
bactericidal toward Escherichia coli, to cause thiol oxidation and
tyrosine nitration, as well as inducing apoptosis in tumour cell
lines and human endothelial cells [19–21].
Apoptosis is a highly regulated process of cell death and seems
to be an essential and critical mechanism used by biological
systems to maintain homoeostasis. Severe oxidative stress is
thought to be a major mediator of apoptosis in several cellular
systems. For example, H O can induce apoptosis in murine
# #
peritoneal macrophages when applied extracellularly [22]. Increased O −d production due to the down-regulation of Cu}Zn
#
superoxide dismutase has also been shown to induce apoptosis
[23]. Accordingly, apoptosis can be prevented by compounds
with antioxidant properties [24].
We have shown that NO donors induce HO-1 expression in
vascular endothelial cells and that thiol compounds diminish the
increase in haem oxygenase activity and HO-1 protein expression
mediated by NO [25,26]. Moreover, we suggested that both O −d
#
and ONOO− partly contribute to the increased activity of this
protein [26]. To extend the findings of our previous work, we
sought in the present study to investigate whether ONOO−
affects HO-1 expression and causes apoptosis in bovine aortic
endothelial cells. In addition, we attempted to define the specific
role of HO-1 induction in ONOO−-mediated apoptosis.
Haem oxygenase is a widely distributed enzyme in mammalian
tissues ; its main function is associated with the degradation of
haem to iron, CO and biliverdin [1]. The subsequent step in haem
breakdown is accomplished by biliverdin reductase, which converts biliverdin into bilirubin. Two distinct isoforms of haem
oxygenase have been extensively studied : HO-2 is the constitutive
isoenzyme and HO-1 (heat shock protein 32, HSP 32), the
inducible form, is a stress protein [2]. The expression of HO-1 is
elicited by many conditions and agents that produce an imbalance
in cellular homoeostasis. A variety of factors, including heavy
metal ions [1,3], various haemoglobins [4–6], endotoxins [7] and
oxidative stress [8–10], are capable of up-regulating the gene for
HO-1 in different tissues. The induction of HO-1 by stress
situations might have physiological importance because the end
products of haem catabolism, biliverdin and bilirubin, possess
antioxidant properties [11] and HO-1-derived CO has been
implicated in vasoregulation [12,13] and signal transduction [14].
Recent experimental evidence showed that HO-1-deficient cells
and mice are susceptible to the accumulation of free radicals and
to oxidative injury after endotoxin administration and established
that HO-1 is an important enzymic antioxidant system [15].
Among oxidant molecules, peroxynitrite (ONOO−) is very
powerful. It derives from the equimolar reaction of NO and
superoxide (O −d) and decomposes at physiological pH to gen#
erate a strong oxidant with a reactivity similar to the hydroxyl
radical, and nitrogen dioxide [16,17]. ONOO− has recently been
considered to be a mediator of cellular injury in many systems
and its production has been implicated in the pathophysiology of
diseases such as acute endotoxaemia, neurological disorders,
atherosclerosis and ischaemia}reperfusion [18,19]. Furthermore,
Key words : bilirubin, HSP32, stress response, tin protoporphyrin
IX.
MATERIALS AND METHODS
Materials
Bovine aortic endothelial cells were obtained from the European
Collection of Animal Cell Culture (Salisbury, Wilts., U.K.).
ONOO− was purchased from Alexis Corp. (Bingham, Notting-
Abbreviations used : HO-1, haem oxygenase-1 ; SnPPIX, tin protoporphyrin IX.
1
To whom correspondence should be addressed (e-mail r.foresti!ic.ac.uk).
# 1999 Biochemical Society
730
R. Foresti and others
(v}v) fetal bovine serum, 2 mM -glutamine, 100 i.u.}ml penicillin
and 0.1 mg}ml streptomycin. For the haem oxygenase activity
assay and Western blot analysis for HO-1, cells were cultured in
75 cm# flasks to reach confluence. For electron microscopy
studies, cells were grown in 25 cm# flasks or in tissue-culture Petri
dishes (60 mm in diameter) ; for the cytotoxicity assay, cells were
subcultured in 24-well tissue culture plates.
Experimental protocol
Figure 1 Effect of ONOO− on haem oxygenase activity and HO-1 expression
in endothelial cells
(A) Haem oxygenase activity was measured 6 h after exposure of endothelial cells to various
concentrations of ONOO− (250–1000µM) added as a bolus to the culture medium. Control
experiments were performed by exposing cells to complete medium alone. Each bar represents
the mean³S.E.M. for five or six independent experiments. *P ! 0.05 compared with control.
(B) HO-1 protein expression was analysed by the Western immunoblot technique (as described
in the Materials and methods section) in samples of cells treated as above. The antibody
complex was detected with an ExtrAvidin alkaline phosphatase kit. Results are representative
of three independent experiments.
ham, Notts., U.K.). ONOO− was supplied in a 0.3 M NaOH
solution ; on arrival it was immediately stored at ®80 °C. The
ONOO− concentration was monitored each time before the
experiment by measuring the absorbance at 302 nm (molar
absorption coefficient 1670 M−"[cm−") after the addition of
25 µl of ONOO− stock solution to 975 µl of ice-cold 0.3 M
NaOH. ONOO− was found to be stable for at least 1 month. Decomposed ONOO− was prepared by reacting fresh ONOO− with
an equal volume of 0.3 M HCl ; 1 h after the reaction, the absorbance was read at 302 nm and it was verified that the ONOO−
concentration had decreased to less than 5 % of that in the original stock solution. All chemicals used for the fixation and preparation of cells for electron microscopy studies were purchased
from Agar Scientific (Stansted, Essex, U.K.). Tin protoporphyrin
IX (SnPPIX) and haemin were from Porphyrin Products
(Logan, UT, U.S.A.). N-Acetylcysteine, uric acid, 1,3-dimethyl2-thiourea and all other chemicals were obtained from Sigma.
Culture of endothelial cells
Bovine aortic endothelial cells were grown in 75 cm# flasks with
Iscove ’s modified Dulbecco ’s medium supplemented with 10 %
# 1999 Biochemical Society
Confluent endothelial cells were washed with phosphate buffer
and fresh medium was added to the cultures. A single bolus of
ONOO− (250, 500, 750 or 1000 µM final concentration) was
rapidly added against the edge of the culture dish (or flask) and
mixed for few seconds to distribute ONOO− across the dish. In
some experiments, 1 mM uric acid or N-acetylcysteine, two
agents that directly scavenge ONOO−, were added to the medium
before the addition of ONOO− (750 µM). Because ONOO−
decomposes at physiological pH to generate nitrogen dioxide
and the harmful hydroxyl radical, in another group of experiments ONOO− (750 µM) was added to culture medium already
supplemented with 1 mM 1,3-dimethyl-2-thiourea, a compound
that scavenges hydroxyl radicals. Endothelial cells were also
incubated for 2, 5 and 10 min in phosphate buffer to which
750 µM ONOO− was added as a bolus ; at the end of ONOO−
exposure, the buffer was removed and replaced with fresh
complete medium for 6 h. Additional experiments were performed by exposing cells to ONOO−-treated medium, which was
obtained by mixing ONOO− (750 µM final concentration) for
5 min in a tube containing complete medium. SnPPIX (10, 25 or
50 µM), an inhibitor of haem oxygenase activity, was added to
the cell culture medium either before or 5 min after ONOO−
addition. Finally, endothelial cells were pretreated with haemin
(100 µM) in complete medium for 2 h followed by 16 h of
incubation in medium alone. The haemin-preconditioned cells
were then exposed to 1000 µM ONOO− added as a bolus to the
medium.
Assay for endothelial haem oxygenase activity
Endothelial haem oxygenase activity was measured 6 h after
exposure to ONOO− or 16 h after haem treatment as described
previously [25,26]. In brief, microsomes from cells exposed to
various treatments were added to a reaction mixture containing
the following : NADPH (0.8 mM), glucose 6-phosphate (2 mM),
glucose-6-phosphate dehydrogenase (0.2 unit), 3 mg of rat liver
cytosol prepared from a 105000 g supernatant fraction as a
source of biliverdin reductase, MgCl (0.2 mM), PBS (100 mM,
#
pH 7.4) and haemin (20 µM). The reaction was conducted in the
dark for 1 h and terminated by the addition of 1 ml of chloroform ; the extracted bilirubin was then calculated by the difference
in absorbance between 464 and 530 nm (ε 40 mM−"[cm−"). Haem
oxygenase activity was expressed in pmol of bilirubin}h per mg
of cell protein. The total protein content of confluent cells was
determined with a Bio-Rad DC protein assay (Bio-Rad, Hemel
Hempstead, Herts., U.K.) by comparison with a standard curve
obtained with BSA.
Western blot analysis for HO-1
Samples of endothelial cells treated for the haem oxygenase assay
were also analysed by the Western immunoblot technique, as
described previously [26]. In brief, cells were lysed and an equal
amount of protein (30 µg) from each sample was boiled in
Laemmli buffer and separated by SDS}PAGE. Proteins were
Peroxynitrite-mediated induction of endothelial haem oxygenase-1
Table 1
731
Haem oxygenase activity and bilirubin production in endothelial cells treated with haemin or ONOO−
Endothelial cells were exposed to medium alone (control) or medium containing 100 µM haemin for 2 h followed by an additional 16 h of incubation in complete medium. ONOO− (750 µM) was
added as a single bolus to endothelial cells in PBS for 2, 5 or 10 min followed by exposure of cells to complete medium for a total incubation period of 6 h. Additional experiments were performed
by exposing cells to ONOO−-treated medium, which was obtained by mixing ONOO− (750 µM final concentration) for 5 min in a tube containing complete medium. SnPPIX (10 µM), an inhibitor
of haem oxygenase activity, was added to cells in culture medium 5 min after the addition of ONOO− (750 µM) as a single bolus. Haem oxygenase activity was determined by using a
spectrophotometric assay that measures the formation of bilirubin as described in the Materials and methods section. Bilirubin released into the culture medium was determined after extraction
with benzene as described in the Materials and methods section. The results are expressed as means³S.E.M. for three to five independent experiments. *P ! 0.05 compared with control.
Treatment
Haem oxygenase activity
(pmol of bilirubin/h per mg of protein)
Bilirubin production
(nmol/h per flask)
Control
Haemin
ONOO− in PBS for 2 min
ONOO− in PBS for 5 min
ONOO− in PBS for 10 min
ONOO−-treated medium
ONOO−SnPPIX added 5 min later
347³13
2738³124*
751³13*
893³17*
886³13*
456³19*
117³5*
9³0.7
18³1.4*
–
–
–
–
–
then transferred to nitrocellulose membrane and probed with
polyclonal rabbit anti-(HO-1) antibodies (Stressgen, Victoria,
Canada). The antibody complex was detected with an ExtrAvidin
alkaline phosphatase kit (Sigma).
Determination of apoptosis in endothelial cells by morphological
analysis
Although different methods for detecting apoptosis have been
developed, such as the detection of DNA fragmentation by
electrophoresis and assays in itro for endonucleases and the
cytoplasmic release of histones, electron microscopy still provides
the most reliable method for recognizing apoptotic cells [27].
Apoptosis was assessed 6 h after each treatment as described
above. Cells were washed with 0.1 M phosphate buffer and fixed
in 3 % (v}v) glutaraldehyde in 0.1 M phosphate buffer for 2 h.
Samples were washed twice with phosphate buffer and the
secondary fixation was performed with 1 % (w}v) osmium
tetroxide in 0.1 M phosphate buffer for 1 h at room temperature.
Cell monolayers were then washed twice with distilled water,
block-stained with 4 % (w}v) aqueous uranyl acetate for 1 h in
the dark, washed with distilled water and scraped into Eppendorf
tubes. Specimens were dehydrated through increasing acetone
series, and infiltrated with acetone}Araldite CY212 resin (1 : 1,
v}v) overnight. Samples were finally embedded in Araldite CY212
resin and blocks were polymerized at 60 °C for 18 h. Sections
were cut with a microtome (Reichert-Jung Ultracut E), floated
on distilled water, collected on Formvar-coated copper grids and
stained with 2 % (w}v) uranyl acetate for 30 min and in 0.13 M
lead citrate for 5 min. Stained sections were examined with a
1200CX electron microscope (Jeol, Tokyo, Japan). The apoptotic
index was calculated by considering the maximal apoptosis
caused by 1000 µM ONOO− as 100 % and expressing all results
from the various protocols as a percentage of this value.
Cytotoxicity assay
A colorimetric assay kit from Promega (Madison, WI, U.S.A.)
was used to assess cell viability 6 h after the various treatments.
The reaction is based on the reduction of a tetrazolium compound
[3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4sulphophenyl)-2H-tetrazolium salt] (MTS) into a coloured formazan product that is soluble in tissue culture medium. The
conversion is accomplished by NADPH or NADH produced by
dehydrogenase enzymes in metabolically active cells. At the end
of the incubation, cells were washed and the assay was performed
by adding new medium containing the tetrazolium compound (at
a ratio of 20 µl of reagent}100 µl of medium) and incubating for
2 h at 37 °C. The amount of formazan product was measured
with a microplate reader (MR 7000 ; Dynatech) at 490 nm. The
measured absorbance is directly proportional to the number of
living cells in culture.
Determination of bilirubin in cell culture medium
Haem oxygenase-derived bilirubin was measured in the cell
culture medium by using a modification of a method described
recently [28]. In brief, endothelial cells pretreated with haemin
for 2 h were exposed to fresh culture medium for an additional
16 h. Control experiments were represented by cells exposed to
medium alone for 18 h. At the end of the incubation period,
0.5 ml of culture supernatant was collected and 250 mg of BaCl
#
was added. After vortex-mixing (10–15 s), 0.75 ml of benzene
was added and tubes were vortex-mixed again vigorously for
60 s. The benzene phase containing the extracted bilirubin was
separated from the aqueous phase by centrifugation at 13000 g
for 30 min. Bilirubin was measured spectrophotometrically
as a difference in absorbance between 464 and 530 nm
(ε 27.3 mM−"[cm−") and the rate of bilirubin excreted into the
culture medium was expressed as nM}h per flask.
Statistical analysis
Differences in the data between the groups were analysed by
using one-way analysis of variance combined with the Bonferroni
test. Values were expressed as means³S.E.M. and differences
between groups were considered to be significant at P ! 0.05.
RESULTS
Effect of ONOO− on haem oxygenase activity and HO-1
expression in endothelial cells
When endothelial cells were treated with increasing concentrations of ONOO− added as a single bolus to the cell culture
medium, a concentration-dependent increase in haem oxygenase
activity, as well as HO-1 expression, was observed at 6 h (Figure
# 1999 Biochemical Society
732
R. Foresti and others
scavengers in preventing the increase in haem oxygenase activity
and HO-1 expression by ONOO− (Figure 2). In contrast, when
the hydroxyl radical scavenger 1,3-dimethyl-2-thiourea was used,
only a partial attenuation of the enzyme activity and HO-1
expression was seen. Taken together, these results suggest a
primary role for ONOO− itself and a contribution of hydroxyl
radicals to up-regulating HO-1 in endothelial cells. This was
further sustained by the finding that decomposed ONOO− did
not produce any significant change in the levels of haem
oxygenase (Figure 2).
Effect of ONOO− on apoptosis and cell viability in endothelial
cells
Figure 2 Effect of uric acid and N-acetylcysteine on ONOO−-mediated
increase in haem oxygenase activity and HO-1 expression in endothelial
cells
(A) Haem oxygenase activity was measured 6 h after exposure of endothelial cells to ONOO−
(750 µM) alone or to ONOO− in the presence of 1 mM 1,3-dimethyl-2-thiourea (DMTU), 1 mM
uric acid (urate) or 1 mM N-acetylcysteine (NAC). Decomposed ONOO− was also tested at
750 µM. Control experiments were performed by exposing cells to complete medium alone.
Each bar represents the mean³S.E.M. for five or six independent experiments. *P ! 0.05
compared with control ; †P ! 0.05 compared with ONOO− alone. (B) HO-1 protein expression
was analysed by the Western immunoblot technique (as described in the Materials and methods
section) in samples of cells treated as above.
1). Exposure of cells to ONOO− (750 µM) in phosphate buffer
for 2, 5 or 10 min followed by incubation with fresh medium also
resulted in a significant increase in haem oxygenase activity at 6 h
(Table 1). Control experiments performed by exposing cells to
phosphate buffer alone for 10 min (but not 2 and 5 min) showed
only a slight increase in haem oxygenase activity (results not
shown). Incubation of cells with ONOO−-treated medium also
produced a minor change in enzyme activity (Table 1). In an
additional set of experiments, the stock solution of ONOO− was
treated with MnO to remove the possible residual H O . At
#
# #
500 µM ONOO−, endothelial haem oxygenase activity was
785³25 pmol bilirubin}h per mg of protein with ONOO−
pretreated with MnO , compared with 805³40 pmol bilirubin}h
#
per mg of protein with ONOO− (not significant).
Uric acid and N-acetylcysteine, a glutathione precursor with
antioxidant properties, are commonly used as ONOO− scavengers. In our experiments we observed a direct effect of both
# 1999 Biochemical Society
As shown in Figure 3(A), treatment of endothelial cells with
ONOO− resulted in apoptotic cell death. Cells showed morphological characteristics typical of the apoptotic event, such as
chromatin condensation and segregation, and condensation of
the cytoplasm with preservation of the integrity of the organelles ;
in some experiments, a careful examination of the fixed tissues
also revealed the presence of some cell death by necrosis at the
highest concentrations of ONOO− used. ONOO−-mediated apoptosis was concentration-dependent, with a considerable number
of dead cells at the concentration of 1000 µM (Figure 3B). The
possibility that in our experiments apoptosis was induced by
NaOH, the vehicle in which ONOO− is solubilized, can be
excluded. In fact, endothelial cells appeared perfectly viable
when, instead of ONOO−, the same amount of 0.3 M NaOH was
added to the culture medium (results not shown). The apoptotic
response to ONOO− was, however, markedly attenuated by
exposing cells to ONOO− in the presence of uric acid or Nacetylcysteine, as shown in Table 2. Uric acid or N-acetylcysteine
at 1 mM were capable of decreasing ONOO−-mediated apoptosis
by approx. 80 % when compared with ONOO− alone.
As shown in Figure 4(A), exposure of endothelial cells to
ONOO− resulted in a concentration-dependent decrease in cell
viability. Pronounced damage was observed at 1000 µM ONOO−,
with approx. 55 % of cells being metabolically inactive. Similarly
to our observations with apoptosis, we found that uric acid and
N-acetylcysteine prevented the ONOO−-mediated cytotoxic
effect, restoring cell viability to 92 % and 82 % of control
respectively (Figure 4B).
Effect of a haem oxygenase inhibitor (SnPPIX) on ONOO−mediated apoptosis and cell viability
Because ONOO− induced endothelial HO-1 expression and
caused apoptosis in a concentration-dependent manner, we
wished to test whether haem oxygenase had a role in the process
of apoptotic cell death produced by the powerful oxidant. When
SnPPIX (10 or 25 µM), an inhibitor of haem oxygenase activity,
was added to the culture medium just before ONOO− (1000 µM),
cells showed a significant decrease in programmed cell death
(Table 2). Similarly, 10 µM SnPPIX prevented ONOO−-mediated
cell injury, recovering cell viability to 96 % of that of the control
(Figure 4B). We also found that addition of SnPPIX (10 or
25 µM) 5 min after ONOO− (1000 µM) decreased apoptosis by
only 40 %, as opposed to a decrease by 70 % and 55 % when 10
and 25 µM SnPPIX respectively were added before ONOO−
(Table 2). At higher concentrations (50 µM), SnPPIX added
before ONOO− did not provide any significant protection against
cell damage mediated by the oxidant (results not shown) and
caused a 50 % increase in apoptosis compared with ONOO−
alone. At this concentration, SnPPIX has been reported to partly
inhibit the activity of constitutive NO synthase and guanylate
Peroxynitrite-mediated induction of endothelial haem oxygenase-1
733
A
B
Figure 3
Effect of ONOO− on apoptosis in endothelial cells
(A) Representative pictures of endothelial cells exposed to complete medium alone (control) or 1000 µM ONOO− added as a bolus to the culture medium. After 6 h incubation, cells were fixed
and stained, and apoptosis was assessed by electron microscopy as described in the Materials and Methods section. Morphological changes typical of apoptosis, such as chromatin condensation
and endoplasmic reticulum alignment (arrowed), are visible in cells treated with ONOO−. Magnification ¬3696 for the control and ¬5676 for ONOO−. (B) Apoptosis was assessed 6 h after treatment
with ONOO− as described in the the Materials and methods section. Apoptotic index was calculated by considering the maximal apoptosis caused by 1000 µM ONOO− as 100 % and expressing
all results from the various protocols as a percentage of this value. Each bar represents the mean³S.E.M. for three independent experiments.
Table 2
Effect of various compounds on ONOO−-mediated apoptosis
ONOO− (1000 µM) was added as a single bolus to endothelial cells in complete medium ; apoptosis was assessed after 6 h as described in the Materials and methods section. SnPPIX, an inhibitor
of haem oxygenase activity, was added to the cell culture medium either before or 5 min after ONOO−. N-Acetylcysteine and uric acid were added before ONOO−. Haemin pretreatment was performed
by exposing cells to culture medium containing 100 µM haemin for 2 h followed by an additional 16 h of incubation in medium alone. Apoptotic index was calculated by considering the maximal
apoptosis caused by ONOO− as 100 % and expressing all results as a percentage of this value. The results are expressed as means³S.E.M. for three to five independent experiments. *P !
0.05 compared with ONOO−.
Compound added before ONOO−
Apoptotic index ( %)
Compound added 5 min after ONOO−
Apoptotic index ( %)
ONOO−
SnPPIX (10 µM)ONOO−
SnPPIX (25 µM)ONOO−
N-Acetylcysteine (1 mM)ONOO−
Uric acid (1 mM)ONOO−
Haemin pretreatmentONOO−
100
30³7*
46³1*
20³8*
21³9*
65³4*
ONOO−
ONOO−SnPPIX (10 µM)
ONOO−SnPPIX (25 µM)
100
60³4*
60³12*
# 1999 Biochemical Society
734
R. Foresti and others
added as a bolus to the cell culture resulted in a decrease in
apoptosis by 35 % (Table 2). These results suggest that stimulation of the haem oxygenase}bilirubin pathway protects against
apoptosis mediated by ONOO−. No apoptosis was observed in
cells pretreated with haemin alone (results not shown).
DISCUSSION
Figure 4 Effect of ONOO− on cell viability and effect of antioxidants and
SnPPIX on ONOO−-mediated cell injury
(A) Viability was assessed in endothelial cells 6 h after exposure to various concentrations of
ONOO− (250–1000 µM) added as a bolus to the culture medium. Each bar represents the
mean³S.E.M. for four independent experiments. *P ! 0.05 compared with control. (B) Cell
viability after exposure to ONOO− alone (1000 µM) or ONOO− in the presence of 1 mM Nacetylcysteine (NAC), 1 mM uric acid (urate) or 10 µM SnPPIX. The absorbance at 490 nm
is directly proportional to the number of living cells. Each bar represents the mean³S.E.M.
for four independent experiments. *P ! 0.05 compared with ONOO− alone.
cyclase in itro [29,30]. Therefore the experiments using 50 µM
SnPPIX cannot be interpreted solely on the basis of haem
oxygenase activity blockade.
Effect of haemin pretreatment on endothelial haem oxygenase
activity, bilirubin production and ONOO−-mediated apoptosis
To define whether increased haem oxygenase activity is protective
against apoptosis caused by ONOO−, endothelial cells were
pretreated for 2 h with 100 µM haemin, a well-known HO-1
inducer [26], followed by an additional 16 h incubation in
complete medium. Endothelial haem oxygenase activity after this
treatment was significantly higher than in the control (Table 1).
We also observed that the rate of bilirubin released in the culture
medium of haemin-pretreated cells was considerably enhanced
(see Table 1). Interestingly, compared with untreated cells,
exposure of haemin-preconditioned cells to 1000 µM ONOO−
# 1999 Biochemical Society
Strong evidence supports a major role for ONOO− in the cellular
injury and death that was once attributed entirely to nitric oxide
(NO) [21,31]. In fact, ONOO−, formed from the reaction of NO
and O −d [17], is a powerful oxidant that has been implicated in
#
the pathophysiology of conditions such as ischaemia}
reperfusion, atherosclerosis and shock [18]. ONOO− decomposes
at pH 7.4 to generate a hydroxyl radical and nitrogen dioxide
[16], two highly reactive species that cause oxidative stress.
ONOO− itself also reacts directly with biological targets including
thiols, iron–sulphur centres and zinc fingers and is responsible
for the production of nitrotyrosine [19]. Indeed, nitrotyrosine
has been detected in human atherosclerotic lesions and as a free
amino acid in plasma from patients with rheumatoid arthritis
[32,33]. Another important characteristic of ONOO− is its ability
to promote nitrosative stress [34]. The hypothesis that ONOO−
might also be involved in the induction of the tissue stress
response has not, however, been explored directly. In previous
studies we have shown that NO donors are able to modulate the
activity [25] and expression of the stress protein HO-1 in vascular
endothelial cells [26] and other cell types [35]. Furthermore, we
have demonstrated that thiols interact with NO to regulate HO1 expression, and suggested that O −d and ONOO− might have a
#
role in the stimulation of this protein [26].
In the present study we report that ONOO− is capable of
inducing HO-1 protein expression and increasing haem oxygenase activity in a concentration-dependent manner in cultured
endothelial cells. This effect is attributable to ONOO− itself
rather than to an oxidized or nitrated component of the medium,
because a brief exposure of cells (2–10 min) to ONOO− in
phosphate buffer also resulted in a marked increase in activity.
Furthermore, ONOO−-treated medium produced only a minor
effect on haem oxygenase activity. Uric acid and N-acetylcysteine,
two antioxidants known to scavenge ONOO−, completely abolished the increase in haem oxygenase activity and HO-1 protein.
Although the concentrations of ONOO− used in our experiments
might seem high, the amount of ONOO− reacting with cells is
considerably lower than the actual concentration used because of
the interaction of ONOO− with components of the culture
medium [20]. In addition, the net exposure of cells to ONOO− is
brief because ONOO− decomposes with a half life of less than 1 s
at pH 7.4 [36]. Ischiropoulos et al. [37] have also shown that in
activated murine macrophages the formation of ONOO− might
be as high as 0.11 nmol}min per 10' cells, which might result
in local maximal rates of as much as 1 mM}min.
ONOO−-mediated HO-1 induction might occur via at least
two possible pathways. First, ONOO− itself could, by virtue of its
powerful oxidant and}or nitrosative characteristics, participate
directly in triggering cellular signals responsible for the induction
of the protein. Alternatively, hydroxyl radicals and nitrogen
dioxide, the decomposition products of ONOO−, could be the
mediators of the process. Moreover, the possibility that both
mechanisms might take place simultaneously cannot be excluded.
Our results showing that 1,3-dimethyl-2-thiourea partly attenuated HO-1 induction by ONOO− indicate a contribution of
hydroxyl radicals to the observed effect. Evidence for the
involvement of hydroxyl radicals in the induction of the haem
oxygenase gene has already been reported in human skin
Peroxynitrite-mediated induction of endothelial haem oxygenase-1
fibroblasts by UV-A radiation [38] and in the rat retina by
intense visible light [39]. It has to be noted that previous reports
suggested the possibility that DMTU could also have a direct
scavenging effect towards ONOO− [40,41]. Because scavengers
that effectively intercept nitrogen dioxide, such as β-carotene,
also react with ONOO− [42], it was more difficult in the present
study to determine a potential contribution of nitrogen dioxide
to the modulation of the expression of this stress protein.
We also observed a concentration-dependent increase in apoptosis in endothelial cells by ONOO−. ONOO−-mediated apoptosis
has been described in cultures in itro of PC12 cells [43], HL-60
cells [20] and human endothelial cells [21]. In the last study
mentioned, it was shown that extracellular glutathione protected
human endothelial cells against ONOO−-induced cytotoxicity.
Recently, uric acid has also been reported to prevent apoptosis
produced by ONOO− in neural cell lines [44]. In our experiments,
N-acetylcysteine, a precursor of glutathione biosynthesis, and
uric acid significantly diminished the number of apoptotic cells
observed with ONOO− alone. In line with our findings on
ONOO−-mediated apoptosis, we observed a loss of cell viability
after exposure of endothelial cells to increasing concentrations of
ONOO−. Once again, uric acid and N-acetylcysteine significantly
prevented the cytotoxic action of ONOO−.
Of major interest in this study is the effect of SnPPIX, an
inhibitor of haem oxygenase activity, on ONOO−-mediated
apoptotic cell death and cytotoxicity. Our results show that
10 µM SnPPIX, a concentration of inhibitor that has been shown
in itro to block haem oxygenase activity specifically [29],
effectively protected against cell injury and death when added
before ONOO−. This would suggest that haem oxygenase, and in
particular HO-1 induction, is associated with increased apoptosis.
This is rather surprising, considering the substantial body of
evidence pointing to HO-1 up-regulation as an important intracellular protective system against stress stimuli [3,6,15]. Moreover, recent findings have demonstrated that endothelial cells
transfected with HO-1 are resistant to tumour-necrosis-factor-αmediated apoptosis [45]. This apparent contradiction could be
partly explained if SnPPIX were to react directly with ONOO− in
our experimental system. Interestingly, a group of metalloporphyrins capable of catalysing ONOO− decomposition to nitrate
has recently been identified [31,46,47]. It was therefore plausible
to postulate that SnPPIX, a protoporphyrin that contains tin as
a metal centre, could also act as a catalyst for ONOO− decomposition. Our hypothesis is corroborated by the results of
this study showing that the addition of SnPPIX (10 or 25 µM)
5 min after ONOO− decreased apoptosis by only 40 %, as
opposed to decreases by 70 % and 55 % when 10 and 25 µM
SnPPIX respectively were added before ONOO−. These results
are best explained by a direct interaction of SnPPIX with ONOO−
when both compounds are present simultaneously in the culture
medium. We still cannot explain why 10 µM SnPPIX added
5 min after ONOO− resulted in decreased apoptosis compared
with ONOO− alone. It is possible that SnPPIX might have effects
other than haem oxygenase inhibition and interaction with
ONOO−, potentially related to the cascade of events leading to
apoptosis.
The contribution of HO-1 induction against ONOO−-mediated
damage was sustained by our findings showing that cells preconditioned with haemin, a potent inducer of HO-1, were less
susceptible to apoptosis caused by ONOO−. We also observed
that the rate of release of the antioxidant bilirubin, the end
product of haem degradation by haem oxygenase, was significantly higher in the culture medium of haemin-treated cells,
suggesting that bilirubin could participate in the mechanism of
protection. In agreement with this concept, it has been recently
735
demonstrated that bilirubin is effective in preventing ONOO−mediated protein oxidation in human blood plasma [48]. However, we cannot exclude the possibility that in our experiments
other important products of haem catabolism, such as biliverdin
and CO, might have a defensive role against ONOO−.
In conclusion, our results demonstrate that ONOO− can
stimulate the tissue stress response by modulating HO-1 protein
expression and that exogenously applied antioxidants prevent
this effect. Our findings also provide evidence that induction of
the HO-1 pathway is cytoprotective against cell death caused by
ONOO−.
This work was supported by a Grant from the National Heart Research Fund, Leeds,
U.K. (R. M.). R. M. and J. E. C. are supported by the R. A. F. T. Institute of Plastic
Surgery.
REFERENCES
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Maines, M. D. (1988) FASEB J. 2, 2557–2568
Maines, M. D., Trakshel, G. M. and Kutty, R. K. (1986) J. Biol. Chem. 261, 411–419
Abraham, N. G., Drummond, G. S., Lutton, J. D. and Kappas, A. (1996) Cell. Physiol.
Biochem. 6, 129–168
Balla, J., Jacob, H. S., Balla, G., Nath, K., Eaton, J. W. and Vercellotti, G. M. (1993)
Proc. Natl. Acad. Sci. U.S.A. 90, 9285–9289
Motterlini, R., Foresti, R., Intaglietta, M., Vandegriff, K. and Winslow, R. M. (1995)
Am. J. Physiol. 269, H648–H655
Otterbein, L., Sylvester, S. L. and Choi, A. M. K. (1995) Am. J. Respir. Cell Mol. Biol.
13, 595–601
Rizzardini, M., Carelli, M., Cabello Porras, M. R. and Cantoni, L. (1994) Biochem. J.
304, 477–483
Keyse, S. M. and Tyrrell, R. M. (1989) Proc. Natl. Acad. Sci. U.S.A. 86, 99–103
Vile, G. F., Basu-Modak, S., Waltner, C. and Tyrrell, R. M. (1994) Proc. Natl. Acad.
Sci. U.S.A. 91, 2607–2610
Choi, A. M. K. and Alam, J. (1996) Am. J. Respir. Cell Mol. Biol. 15, 9–19
Stocker, R., Yamamoto, Y., McDonagh, A. F., Glazer, A. N. and Ames, B. N. (1987)
Science 235, 1043–1046
Motterlini, R., Gonzales, A., Foresti, R., Clark, J. E., Green, C. J. and Winslow, R. M.
(1998) Circ. Res. 83, 568–577
Sammut, I. A., Foresti, R., Clark, J. E., Exon, D. J., Vesely, M. J. J., Sarathchandra,
P., Green, C. J. and Motterlini, R. (1998) Br. J. Pharmacol. 125, 1437–1444
Maines, M. D. (1997) Annu. Rev. Pharmacol. Toxicol. 37, 517–554
Poss, K. D. and Tonegawa, S. (1997) Proc. Natl. Acad. Sci. U.S.A. 94, 10925–10930
Beckman, J. S., Beckman, T. W., Chen, J., Marshall, P. A. and Freeman, B. A. (1990)
Proc. Natl. Acad. Sci. U.S.A. 87, 1620–1624
Pryor, W. A. and Squadrito, G. L. (1995) Am. J. Physiol. 268, L699–L722
Beckman, J. S. and Crow, J. P. (1993) Biochem. Soc. Trans. 21, 330–334
Beckman, J. S. and Koppenol, W. H. (1996) Am. J. Physiol. 271, C1424–C1437
Lin, K. T., Xue, J. Y., Nomen, M., Spur, B. and Wong, P. Y. K. (1995) J. Biol. Chem.
270, 16487–16490
Szabo, C., Cuzzocrea, S., Zingarelli, B., O’Connor, M. and Salzman, A. L. (1997)
J. Clin. Invest. 100, 723–735
Albina, J. E., Cui, S., Mateo, R. B. and Reichner, S. (1993) J. Immunol. 150,
5080–5085
Troy, C. M. and Shelanski, M. L. (1994) Proc. Natl. Acad. Sci. U.S.A. 91,
6384–6387
Ratan, R. R., Murphy, T. H. and Baraban, J. M. (1994) J. Neurochem. 62, 376–379
Motterlini, R., Foresti, R., Intaglietta, M. and Winslow, R. M. (1996) Am. J. Physiol.
270, H107–H114
Foresti, R., Clark, J. E., Green, C. J. and Motterlini, R. (1997) J. Biol. Chem. 272,
18411–18417
Kerr, J. F., Gobe, G. C., Winterford, C. M. and Harmon, B. V. (1995) Methods Cell
Biol. 46, 1–27
Turcanu, V., Dhouib, M. and Poindron, P. (1998) Anal. Biochem. 263, 251–253
Zakhary, R., Gaine, S. P., Dinerman, J. L., Ruat, M., Flavahan, N. A. and Snyder,
S. H. (1996) Proc. Natl. Acad. Sci. U.S.A. 93, 795–798
Ignarro, L. J., Barrot, B. and Wood, K. S. (1984) J. Biol. Chem. 259, 6201–6207
Misko, T. P., Highkin, M. K., Veenhuizen, A. W., Manning, P. T., Stern, M. K., Currie,
M. G. and Salvemini, D. (1998) J. Biol. Chem. 273, 15646–15653
Beckman, J. S., Ye, Y. Z., Anderson, P., Chen, J., Accavetti, M. A., Tarpey, M. M. and
White, C. R. (1994) Biol. Chem. Hoppe-Seyler 375, 81–88
Kaur, H. and Halliwell, B. (1994) FEBS Lett. 350, 9–12
Stamler, J. S., Toone, E. J., Lipton, S. A. and Sucher, N. J. (1997) Neuron 18,
691–696
# 1999 Biochemical Society
736
R. Foresti and others
35 Vesely, M. J. J., Exon, D. J., Clark, J. E., Foresti, R., Green, C. J. and Motterlini, R.
(1998) Am. J. Physiol. 275, C1087–C1094
36 Zhu, L., Gunn, C. and Beckman, J. S. (1992) Arch. Biochem. Biophys. 298, 452–457
37 Ischiropoulos, H., Zhu, L. and Beckman, J. S. (1992) Arch. Biochem. Biophys. 298,
446–451
38 Keyse, S. M. and Tyrrell, R. M. (1989) Carcinogenesis 11, 787–791
39 Kutty, R. K., Kutty, G., Wiggert, B., Chader, G. J., Darrow, R. M. and Organisciak,
D. T. (1995) Proc. Natl. Acad. Sci. U.S.A. 92, 1177–1181
40 Whiteman, M. and Halliwell, B. (1997) Free radicals Biol. Med. 22, 1309–1312
41 Wasil, M., Halliwell, B., Grootveld, M., Moorhouse, C. P., Hutchinson, D. C. S. and
Baum, H. (1987) Biochem. J. 243, 867–870
42 Kikugawa, K., Hiramoto, K., Tomiyama, S. and Asano, Y. (1997) FEBS Lett. 404,
175–178
Received 24 June 1998/21 January 1999 ; accepted 17 February 1999
# 1999 Biochemical Society
43 Estevez, A. G., Radi, R., Barbeito, L., Shin, J. T., Thompson, J. A. and Beckman, J. S.
(1995) J. Neurochem. 65, 1543–1550
44 Keller, J. N., Kindy, M. S., Holtsberg, F. W., StClair, D. K., Yen, H. C., Germeyer, A.,
Steiner, S. M., BruceKeller, A. J., Hutchins, J. B. and Mattson, M. P. (1998)
J. Neurosci. 18, 687–697
45 Soares, M. P., Lin, Y., Anrather, J., Csizmadia, E., Takigami, K., Sato, K., Grey, S. T.,
Colvin, R. P., Choi, A. M., Poss, K. D. and Bach, F. H. (1998) Nat. Med. (Tokyo) 4,
1073–1077
46 Salvemini, D., Wang, Z. Q., Stern, M. K., Currie, M. G. and Misko, T. P. (1998)
Proc. Natl. Acad. Sci. U.S.A. 95, 2659–2663
47 Lee, J., Hunt, J. A. and Groves, J. T. (1998) J. Am. Chem. Soc. 120, 6053–6061
48 Minetti, M., Mallozzi, C., Di Stasi, A. M. M. and Pietraforte, D. (1998) Arch. Biochem.
Biophys. 352, 165–174