Laboratoire de Physiopathologie et Pharmacologie Cardiovasculaires
Expérimentales, Facultés de Médecine et Pharamacie,
Dijon, France
 |
Introduction |
Free
radical production and calcium overload are considered as the two major
events implicated in the development of myocardial ischemia and
reperfusion injury (Hearse and Bolli, 1992
; Maxwell and Lip, 1997;
Piper et al., 1998). The oxidative stress consecutive to an imbalance
between the production of radical species and the protection by several
antioxidant systems can lead to electrophysiological, biochemical, and
mechanical disturbances, dramatically impairing the ability of the
heart to recover from the initial ischemic insult (Hearse and Tosaki,
1987; Bolli, 1991). Among the possible mechanisms that are supposed to
be implicated in this postischemic oxidative burst, the uncoupling of
mitochondrial respiratory chain (Turrens, 1997) and the activation of
enzymes such as xanthine oxidase (Sobey et al., 1992) or NADPH oxidase
(Griendling and Ushio-Fukai, 1997) have been successfully investigated.
However, the interactions between free radical species (e.g.,
superoxide anion, hydroxyl radical, and nitric oxide) are more
difficult to understand in this specific situation.
Nitric oxide (·NO) is a gaseous nitrogen-centered free radical,
released from L-arginine and dioxygen by nitric-oxide
synthases (NOSs). At least three different isoforms of NOS have been
identified to date. The catalytic scheme is shared by the different
isoforms of NOS; however, uncoupled electron transfers have been
described in NOS I (Heinzel et al., 1992; Pou et al., 1992), II (Xia
and Zweier, 1997), and III (Vasquez-Vivar et al., 1998; Xia et al., 1998) under conditions of low concentrations of L-arginine
and/or tetrahydrobiopterin, with oxygen being the acceptor of
the electrons, giving rise to the superoxide anion (O
). Therefore, uncoupled electron transfer in NOS sometimes leads to the
generation of a mixture of ·NO and O, species that can
then react with each other at a near diffusion limited rate (6.7 × 109 M
1 · s1) to produce the peroxynitrite anion
(ONOO), which is considered as a very reactive
and toxic molecule (for review, see Beckman and Koppenol, 1996). All
three isoforms of NOS may be expressed in the heart (for review, see
Shah and MacCarthy, 2000), albeit in a cell-specific manner, and the
numerous physiological effects of ·NO on cardiac function have
been reviewed (Kelly et al., 1996).
During myocardial reperfusion, the role of ·NO in the
development of myocardial injury has been extensively studied in
different experimental models, using either NOS antagonists
(Depré et al., 1995; Naseem et al., 1995; Zweier et al., 1995;
Wang and Zweier, 1996; Brunner et al., 1997; du Toit et al., 1998;
Zhang et al., 2001), L-arginine (Takeuchi et al., 1995;
Engelman et al., 1996; Brunner et al., 1997; Wang et al., 1997; Mizuno
et al., 1998), ·NO donors (Brunner et al., 1997; du Toit et al.,
1998), or NOS-knockout mice models (Flögel et al., 1999; Kanno et
al., 2000). However, there are contradictory results concerning the
possible protective or deleterious role of ·NO during ischemia
and reperfusion. If ·NO is a free radical, its reactivity as a
radical species is low and the toxicity of ·NO is likely to
result from its reaction with O to produce peroxynitrite. In
situations where an increase in ·NO occurs at the same time as
an increase in oxygen radical species production, such as during
myocardial reperfusion, this may be deleterious because of the
increased formation of ONOO. On the other hand,
its role as a sink for superoxide may preserve the cellular environment
from hydrogen peroxide/Fenton-driven oxidative reactions. The
implication of NOS activity as a modulator of oxidative stress during
cardiac ischemia and reperfusion is hence conflicting and deserves more
thorough investigation.
Therefore, the aim of our study was to investigate the role of
NOS modulation on 1) the evolution of functional parameters and the
level of postischemic recovery and 2) the amount of free radical
species released during a sequence of global myocardial ischemia and
reperfusion, using L-arginine as a substrate or
NG-nitro-L-arginine
methyl ester (L-NAME) as an inhibitor, in
comparison with their D-specific enantiomers.
| |
Materials and Methods |
Chemicals.
The spin trap
-phenyl-N-tert-butylnitrone (PBN;
Sigma, Saint Quentin Fallaner, France) was purified by
sublimation under argon gas and stocked at 
80°C in dark vials.
Toluene (high-performance liquid chromatography grade) was purchased
from Fluka (Saint Quentin Fallaner, France). All other chemicals
were purchased from Sigma.
Perfusion Technique and Perfusion Medium.
The investigation
conforms with the Guide for the Care and Use of Laboratory Animals
published by the National Institutes of Health (NIH Publication 85-23, revised 1996). Male Wistar rats (307 ± 2 g) were purchased
at Depré (Saint Doulchard, France). The rats were
anesthetized with sodium thiopental (60 mg/kg i.p.) and heparin was
intravenously injected (500 IU/kg). After 1 min, the hearts were
excised and placed in a cold (4°C) perfusion buffer bath until
contractions ceased. Each heart was then immediately cannulated through
the aorta and perfused by the Langendorff method, at a constant
perfusion pressure equivalent to 80 cm of water (8 kPa). The perfusion
buffer consisted of a modified Krebs-Henseleit bicarbonate buffer (118 mM NaCl, 25 mM NaHCO3, 1.2 mM
MgSO4, 1.2 mM
KH2PO4, 4.7 mM KCl, 5.5 mM
glucose, and 3 mM CaCl2). Before use, all
solutions were filtered through a 0.8-µm filter (Millipore Corporation, Bedford, MA) to remove any particulate contaminants. The
perfusion fluid was gassed with 95% oxygen and 5% carbon dioxide (pH
7.3-7.5 at 37°C). An elastic water-filled latex balloon (no. 4; Hugo
Sachs Electronik, Hugstetten, Germany) was inserted into the left
ventricle through the mitral valve and connected to a pressure
transducer, the output of which was connected to a physiograph. The
filling pressure was individually adjusted to 12 to 18 mm Hg (1.6-2.5
kPa) to achieve a maximal contractile performance. A TA 240 recorder
(Gould, Cleveland, OH) was used to measure heart rate and
intraventricular pressures: left end diastolic ventricular pressure
(LEDVP) and left systolic ventricular pressure. The left ventricular
developed pressure (LVDP) was calculated from left systolic ventricular
pressure LEDVP and rate-pressure product (RPP) was from the
product of LVDP and heart rate. Coronary flow was measured by the timed
collection of the effluent.
Perfusion Protocols.
Twelve groups of hearts were subjected
to different ischemia-reperfusion protocols at 37°C (Fig.
1). After a stabilization phase of 15 min, isolated hearts were perfused aerobically for 15 min (preischemic
control period). Global normothermic ischemia was then induced by
clamping aortic inflow for 30 min, during which a thermoregulated
chamber maintained the heart temperature at 37°C. After ischemia,
aortic inflow was resumed for 30 min (reperfusion period).
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 1.
Perfusion protocols of PBN-free or PBN-treated groups
of isolated rat hearts. Drug perfusion was started 10 min before the
induction of ischemia and continued throughout 30 min of reperfusion.
PBN (3 mM) was perfused as indicated under Materials and
Methods. Arrows ( ) indicate time for coronary sample
collection.
|
|
In a first series of experiments (PBN-free groups), hearts were
infused with 100 µM
NG-nitro-D-arginine
methyl ester (D-NAME; group 1a, n = 7), 100 µM L-NAME (group 2a,
n = 7), 3 mM D-arginine (group
3a, n = 7), or 3 mM L-arginine
(group 4a, n = 7). The compounds were directly dissolved in the perfusion medium and administrated 10 min before ischemia and throughout reperfusion period.
In a second series of experiments (PBN-treated groups), isolated hearts
were administrated the same compounds under the same conditions (100 µM D-NAME, group 1b, n = 14; 100 µM
L-NAME, group 2b, n = 15; 3 mM
D-arginine, group 3b, n = 6; and
3 mM L-arginine, group 4b, n = 6). The amount of free radicals released was measured with spin
trapping ESR as described previously (Vergely et al., 2001b). The spin
trap PBN (3 mM) was infused upstream of the coronary bed 5 min before
the onset of ischemia and during the reperfusion period (15 min from
the beginning and 5 min before the end of the reperfusion period).
Five-microliter aliquots of coronary effluent samples were collected at
different times before ischemia and during reperfusion (Fig. 1,
arrows), immediately extracted with 0.75 ml of
N2-gassed ice-cold toluene, frozen, and kept into liquid nitrogen until ESR measurement.
ESR Spin Trapping.
Toluene extracts were thawed and
bubbled with N2 for 20 s. ESR spectra were
recorded at 293°K with an ESP 300E-X band spectrometer (Bruker,
Wissenbourg, France) using a TM110 cavity and an
aqueous flat cell. The following parameters were selected for optimal detection of PBN spin adducts: microwave power, 20 mW; microwave frequency, 9.74 GHz; modulation amplitude, 1.6 G; modulation frequency, 100 kHz; gain, 1.6 to 3.2 × 106; scan rate,
0.95 G s1; time constant, 163.84 ms; and
conversion time, 82 ms.
The signal intensity, which is proportional to the concentration of
spin adducts, was measured directly from the field scan and expressed
as spin adduct concentration (nM) by double integration of the
experimental spectra using 2,2,6,6 tetramethylpiperidine-N-oxyl nitroxide as an integration
standard. The spin adduct release rate (pmol/min/g of heart) at each
perfusion time was obtained by multiplying the adduct concentration by
the respective coronary flow.
Determination of Oxygen Radical Absorbance Capacity.
The
potential antioxidant properties of D-arginine,
L-arginine, D-NAME, and L-NAME were
evaluated as oxygen radical absorbance capacity (ORAC) according to a
modified method of Cao et al. (1993). Briefly, the reaction mixture
contained a final concentration of 3.75 × 108 M 
-allophycocyanin in 75 mM phosphate
buffer, pH 7.0, at 37°C in the presence or the absence of Trolox (1 µM) or of the compounds (10 µM-10 mM). The reaction was initiated
by the introduction of 3 × 103 M
2,2'-azobis(2-amidinopropane)-4-hydrochloride and followed spectrophotometrically by the decrease in fluorescence at 598-nm excitation and 615-nm emission. Trolox was used as a reference antioxidant for calculating the ORAC values, with one ORAC unit defined
as the net protection area provided by 1 µM final concentration of Trolox.
Statistical Analysis.
All data are presented as means ± S.E.M. Statistical analysis was performed with a t test,
determining differences between L- or
D-compound-treated hearts, at each time of the
perfusion protocol.
| |
Results |
Myocardial Functional Parameters during Experimental
Ischemia-Reperfusion: PBN-Free Hearts
Coronary Flow.
At the beginning of the preischemic
perfusion period, the coronary flow of isolated perfused hearts (Fig.
2) was stable around the value of 14 ml/min (16 ml/min/g of heart tissue).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 2.
Evolution of coronary flow of isolated rat hearts
treated with 100 µM NAME (A) or 3 mM arginine (B), before, during,
and after 30 min of global total ischemia. The time for the
administration of compounds is shown with the arrow. Results are
presented as means ± S.E.M.  , 100 µM
D-NAME-treated hearts (n = 7);  , 100 µM L-NAME-treated hearts (n = 7);
 , 3 mM D-arginine-treated hearts (n = 7);  , 3 mM L-arginine-treated hearts
(n = 7). Significantly different from the
D-analog group:  , P < 0.05; ,
P < 0.01; , P < 0.001.
|
|
The administration of 100 µM D-NAME (Fig. 2A), started 5 min thereafter, did not significantly influence the evolution of coronary flow, despite a small and progressive increase (12%). Conversely, when 100 µM L-NAME was infused upstream of
the coronary bed, the increase of vascular resistances led to a
significant decrease in coronary flow: just before the induction of
ischemia, the coronary flow of L-NAME-treated hearts was
46% lower than the control group treated with D-NAME.
After 30 min of no-flow ischemia, reperfusion allowed only a partial
recovery of coronary flow, reaching only 50% of its preischemic value.
The administration of L-NAME during reperfusion led to an
even lower postischemic coronary flow, with significant intergroup
differences observed from the beginning of reperfusion.
The administration of 3 mM D- or
L-arginine (Fig. 2B) during the preischemic perfusion
period slightly decreased (11%) the level of coronary flow. During
postischemic reperfusion, the coronary flow was dramatically impaired
and recovered to ~25% of its initial value. No differences in the
evolution of coronary flow were seen between these
two groups.
Left End Diastolic Ventricular Pressure.
During the
preischemic perfusion period, the LEDVP, which was initially set
between 12 and 18 mm Hg, was stable and not modified by the treatments
(Fig. 3).
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 3.
Evolution of LEDVP of isolated rat hearts treated
with 100 µM NAME (A) or 3 mM arginine (B), before during and after 30 min of global total ischemia. The time for the administration of
compounds is shown with the arrow. Results are presented as means ± S.E.M. , 100 µM D-NAME-treated hearts
(n = 7); , 100 µM L-NAME-treated
hearts (n = 7); , 3 mM
D-arginine-treated hearts (n = 7); and
, 3 mM L-arginine-treated hearts (n = 7). Significantly different from the D-analog group: ,
P < 0.05.
|
|
The induction of ischemia was rapidly followed by the cessation of
myocardial contractions and by a gradual increase in intraventricular pressure, representative of the phenomenon of ischemic contracture. After 20 min of ischemia, the intraventricular pressure reached a peak
value close to 80 mm Hg and then decreased slowly. The pressure inside
the left ventricle was shown to be statistically lower at the end of
ischemia for hearts administered with 100 µM D-NAME.
Upon the relief from 30 min of global ischemia, LEDVP increased rapidly
and reached a maximum 3 min after the onset of reperfusion, a feature
characteristic of postischemic contracture. This peak of LEDVP was
between 95 and 100 mm Hg for NAME-treated hearts and between 108 and
120 mm Hg for arginine-treated hearts, with a significantly higher peak
level for L-arginine. Then LDVP decreased slowly but
remained at a high level during the whole reperfusion period. No
differences in the recovery of LEDVP were observed between
L- or D-groups. However, LEDVP was at least 20 mm Hg lower at the end of the reperfusion period for hearts treated
with 100 µM D- or L-NAME than with
D- or L-arginine.
Rate-Pressure Product.
RPP corresponds to the product of LVDP
(left end systolic ventricular pressure LEDVP) and heart rate
and is usually considered as a good index of myocardial contractile
efficiency (Fig. 4). The RPP of isolated
perfused rat hearts was initially close to 40 × 103 beats · mm Hg/min.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 4.
Evolution of RPP (LVDP × heart rate) of
isolated rat hearts treated with 100 µM NAME (A) or 3 mM arginine (B)
before, during, and after 30 min of global total ischemia. The time for
the administration of compounds is shown with the arrow. Results are
presented as means ± S.E.M. , 100 µM
D-NAME-treated hearts (n = 7); , 100 µM L-NAME-treated hearts (n = 7);
, 3 mM D-arginine-treated hearts (n = 7); and , 3 mM L-arginine-treated hearts
(n = 7). Significantly different from the
D-analog group: , P < 0.05.
|
|
The administration of 100 µM L-NAME induced a progressive
30% decrease of RPP (P < 0.05), consecutive to the
impairment of LVDP without changes in heart rate. Conversely, 3 mM
L-arginine did not modify the evolution of RPP,
compared with D-arginine.
During ischemia, contractions ceased, and LVDP and RPP were reduced to
zero. With reperfusion, RPP was only slightly restored and remained at
a low level at the end of the reperfusion period. The final RPP of
NAME- or arginine-treated hearts corresponded to only 30 and 10% of
its initial value, respectively, without intergroup differences.
Compared with the preischemic level of RPP, the recovery of hearts
infused with D-NAME was 29 ± 11 versus 40 ± 11% with L-NAME.
Rhythm Disturbances.
Rhythm abnormalities (Fig.
5) are frequently observed after 30 min
of a global normothermic ischemia and are mostly represented by
ventricular tachycardia and fibrillation. The average duration of these
rhythm disturbances was not modified by the treatment with the active
compound.
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 5.
Mean duration of rhythm disturbances after 30 min of global total ischemia and 30 min of reperfusion of isolated rat
hearts treated with 100 µM D-NAME () or 100 µM
L-NAME () (A) or 3 mM D-arginine () or 3 mM L-arginine () (B). Results are presented as
means ± S.E.M.
|
|
Myocardial Functional Parameters of Control Hearts during
Experimental Ischemia-Reperfusion.
The evolution of functional
parameters of isolated rat hearts treated neither with any of the
arginine analogs nor with PBN is presented as an indication in Table
1. The recovery of coronary flow LVDP and
RPP was 43 ± 4, 31 ± 14, and 21 ± 14%, respectively. Because this control group was not processed at the same moment as the
arginine analogs, they are not directly comparable.
View this table:
[in this window]
[in a new window]
|
TABLE 1
Evolution of myocardial functional parameters of a control group
treated neither with NAME nor arginine nor PBN
Data are presented as means ± S.E.M. and correspond to the
initial preischemic values (t = 0), the values taken
just before the induction of ischemia (t = 14 min) and
at the end of the reperfusion period (t = 75 min).
|
|
Myocardial Functional Parameters and Spin Adduct Release Rate of
PBN-Treated Hearts during Experimental Ischemia-Reperfusion
Functional Parameters.
The evolution of functional parameters
of PBN-treated hearts is presented in Table
2. As already observed without PBN,
L-NAME perfusion induced a significant diminution of
coronary flow, LVDP, and RPP under preischemic control perfusion
conditions. The administration of 3 mM L-arginine before
ischemia was not responsible for any modifications of functional
parameters, compared with D-arginine. At the end of
reperfusion, myocardial parameters were not different between hearts
treated with D- or L-NAME. For the hearts that were perfused in the presence of 3 mM D- or
L-arginine, all parameters were comparable at the end of
the postischemic period, except for LEDVP, which was significantly
higher in the L-arginine group.
View this table:
[in this window]
[in a new window]
|
TABLE 2
Evolution of myocardial functional parameters of NAME (100 µM)- or
arginine (3 mM)-treated hearts for the groups receiving PBN (3 mM) as a
spin trap
Data are presented as means ± S.E.M. and correspond to the
initial preischemic values (t = 0), the values taken
just before the induction of ischemia (t = 14 min) and
at the end of the reperfusion period (t = 75 min).
|
|
ESR Spin Trapping.
Experiments performed with ESR on
coronary effluents showed the presence of a sextuplet signal
(aN = 13.5 G; aH = 2.1 G;
g = 2.012) with coupling constants that could be attributed to
alkyl/alkoxyl spin adducts, but also of a triplet
(aN = 7.9 G; g = 2.013) that could be
attributed to an oxidation product of PBN,
benzoyl-tert-butyl nitroxide (Fig.
6). The concentration of these free
radical species in the coronary effluent was evaluated, and the spin
adduct release rate was calculated by taking into account the level of
coronary flow (Fig. 7). During normoxic
preischemic perfusion, a low spin adduct release rate was observed in
coronary effluents, which was larger in the groups treated with
D- or L-NAME. After
reperfusion of the heart, a large release of alkyl/alkoxyl species
occurred with spin adduct release rate reaching 4 times its preischemic level and remaining at a high level during 30 min of reperfusion. However, the treatment with the active (L-) or
the inactive (D-) compound did not modify the
release of spin adducts, before and after the induction of ischemia.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 6.
EPR representative spectra of coronary effluent
extracts obtained before ischemia and 5 min after starting reperfusion
for hearts treated with 100 µM D- or L-NAME.
Spectral analysis showed the presence of two distinct signals:  ,
aN = 13.5 G, aH = 2.1 G, g = 2.012; , aN = 7.9 G, g = 2.013.
|
|
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 7.
Free radical release rate before and after 30 min of
global total ischemia for 100 µM NAME- (A) or 3 mM arginine
(B)-treated hearts. Results are presented as means ± S.E.M. ,
100 µM D-NAME-treated hearts (n = 14); , 100 µM L-NAME-treated hearts
(n = 15); , 3 mM D-arginine-treated
hearts (n = 6); and , 3 mM
L-arginine-treated hearts (n = 6).
|
|
Antioxidant Activity of Compounds.
The antioxidant capacity of
D- and L-NAME and of D- and
L-arginine was evaluated as their ability to protect a
fluorescent protein (allophycocyanin) toward alkyl/alkoxyl
radical-induced oxidation (Fig. 8).
Antioxidant properties were observed only with D- or
L-NAME at the highest concentration (10 mM), with ORAC values reaching 0.91 ± 0.07 and 1.43 ± 0.13, respectively.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 8.
Antioxidant properties of D-NAME (),
L-NAME (), D-arginine (), and
L-arginine () evaluated as ORAC at 0.1, 1, and 10 mM
concentrations.
|
|
| |
Discussion |
NOS Modulation and Myocardial Function during an
Ischemia-Reperfusion Sequence.
During ischemia and reperfusion
·NO may have dual effects, beneficial in preserving endothelial
function, preventing leukocyte adhesion, and promoting vasodilation,
although harmful because it combines with O to generate
ONOO. Peroxynitrite irreversibly inhibits both
contractility and respiration, in contrast to the reversible effect of
·NO (Radi et al., 1994; Xie et al., 1998), which may represent an aggravating factor for the development of postischemic cellular injury. The balance between beneficial and deleterious effects of
·NO is of key importance with respect to its pathophysiological role as a modulator of cardiac function.
The 100 µM L-NAME infusion induced a rapid 46% decrease
of coronary flow compared with D-NAME infusion. This
increase in coronary vascular resistance can be directly attributed to
the inhibition of NOS activity and the subsequent impairment of
·NO-induced vasorelaxation, as was described previously in a
similar model (Pabla and Curtis, 1995). During L-NAME
administration, the progressive decrease in RPP might be related to the
impairment of oxygen and substrate supplies due to vasoconstriction.
During postischemic reperfusion, coronary flow of
L-NAME-treated hearts was still lower than the
D-NAME group, but when the data were expressed as recovery
of preischemic values, L-NAME groups showed higher coronary
flow recovery than D-NAME (64 ± 4 versus 51 ± 3%, respectively). The evolution of contractility was not different between L- or D-NAME-treated hearts during
reperfusion, but the recovery of preischemic RPP was slightly higher
for the L-NAME (40 ± 11%) than for the
D-NAME groups (29 ± 11%). Various competitive inhibitors of NOS have been shown to reduce reperfusion injury in
various settings (Depré et al., 1995; Naseem et al., 1995; Zweier
et al., 1995; Wang and Zweier, 1996; du Toit et al., 1998; Zhang et
al., 2001). In contrast to the above-mentioned studies, some
experiments reported that NOS antagonists can enhance postischemic myocardial injury (Brunner et al., 1997; du Toit et al., 1998). The
differences may be due to the experimental models, but also to the
concentration of NOS antagonists, because nonvasoactive concentrations
seem to protect the heart against postischemic injury (Depré et
al., 1995; du Toit et al., 1998), whereas concentrations above 100 µM
would be detrimental. Moreover, when using vasoactive concentrations of
NOS inhibitors, the result is considered as beneficial in terms of
postischemic recovery (Zweier et al., 1995; Wang and Zweier, 1996), or
deleterious when expressed in net values (Brunner et al., 1997; du Toit
et al., 1998), as was experienced in our study. Therefore, because
L-NAME treatment does not modify the duration of
postischemic rhythm disturbances, we propose that the administration of
100 µM L-NAME before 30 min of global ischemia and
throughout reperfusion does not afford protection to the heart against
postischemic injury.
Under our experimental conditions, it was not possible to determine
differences in cardiac function between D- or
L-arginine-treated hearts, despite a slight increase in
postischemic peak LEDVP and a decrease in heart rate. Many authors have
described beneficial effects of L-arginine treatment on the
recovery of myocardial function during reperfusion (Engelman et al.,
1996; Brunner et al., 1997; Wang et al., 1997; Mizuno et al., 1998;
Padilla et al., 2000), whereas others have shown a harmful effect of
L-arginine administration (Takeuchi et al., 1995; Mori et
al., 1998). These discrepancies can be attributed to different
experimental models but also to the critical timing of this amino acid
administration. In fact, L-arginine seems even more
powerful as a pretreatment (Engelman et al., 1996; Wang et al., 1997;
Mizuno et al., 1998), and sometimes deleterious during reperfusion
(Takeuchi et al., 1995; Engelman et al., 1996). Therefore, the absence
of functional parameters' variations between D- or
L-arginine observed under our experimental conditions might
be the addition of both beneficial and detrimental effects of
L-arginine administration.
The effects of NAME or arginine administration were not compared in our
study because the experiments were not processed exactly at the same
moment. Therefore, in our experimental conditions, the addition of
either the substrate or a competitive inhibitor of NOS did not modify
significantly the ability of the heart to recover from the ischemic insult.
NOS Modulation and Postischemic Oxidative Stress.
Using the
technique of electron spin resonance associated with PBN spin trapping,
we have investigated the effect of modulating NOS activity on the
release of radical species. This method allows direct identification of
secondary free radicals released by the heart during postischemic
reperfusion (Bolli et al., 1988; Blasig et al., 1994). It is noteworthy
that PBN in itself can modify the evolution of cardiovascular
parameters (Vergely et al., 2001b), but this spin trap is a valuable
tool when investigating the release of free radical species in
biological systems. In the present work, our first observation was that
the preischemic period was associated with the presence, at a low
level, of alkyl/alkoxyl radicals trapped in coronary effluents. After
30 min of global ischemia, a 3- to 5-fold increase in the initial spin
adduct release rate was observed, which was maintained at a high level
throughout reperfusion. Alkyl/alkoxyl radicals are secondary free
radicals, formed by the reaction of primary species such as O or hydroxyl radical (HO·) on different molecules in their
environment. The presence of such secondary species in the coronary
effluent of perfused hearts is likely to originate not only from the
cell components exposed to an oxidative stress but also from the
oxidation of perfusion buffer constituents (Kalyanaraman et al., 1994).
The postischemic free radical release kinetics is consistent with a
previous study (Vergely et al., 2001b) and shows almost the same
pattern among arginine- or NAME-treated hearts. However, preischemic
and postischemic levels are lower in arginine treatment. The same
variability in preischemic spin adduct signals have been observed for
other protocols in our laboratory (Vergely et al., 2001a) and might
originate from using different batches of high-purity PBN. Because the
production of secondary radical adducts is influenced not only by the
level of oxidative stress but also by the presence of other trapping
molecules in the environment, it is possible that a 3 mM concentration
of arginine is an appropriate tool to prevent free radicals from
reacting either with cell constituents or with the spin trap. Indeed,
antioxidant properties of D- and L-arginine
(Rehman et al., 1997; Walner et al., 2001) and of D- and
L-NAME (Rehman et al., 1997) have been described on
HO· (Rehman et al., 1997) or against lipoprotein oxidation
(Walner et al., 2001). However, by using the ORAC test, we observed
that 100 µM D- and L-arginine or 3 mM
D- or L-NAME afforded low antioxidant properties against an alkyl/alkoxyl-generating system.
Another important observation from our work is that the spin adduct
release rate during reperfusion is not modified by the treatment with
the active enantiomers of arginine or NAME. Therefore, the positive or
negative modulation of NOS activity does not seem to influence the
level of postischemic oxidative stress. Several authors have provided
evidence that uncoupled electron transfer can occur in NOS, leading to
the production of superoxide. In physiological conditions, cells
probably maintain sufficient levels of L-arginine for NOS
to catalyze NO synthesis, because L-arginine is found in
mammalian cells at concentrations by far exceeding the
KM values of these enzymes. However,
in conditions such as ischemia and reperfusion, where a leak in amino
acids is observed (Brunner et al., 1997; Song et al., 1998), the
availability of L-arginine and
tetrahydrobiopterin could be impaired and would participate in
NOS-induced generation of O. Therefore, the administration
of either L-arginine or
L-NAME should reduce uncoupled generation of
reactive oxygen species by NOS. On the other hand, as a trap for
O, ·NO produced in cardiac cells during reperfusion
may preserve the tissues from reactive oxygen species-driven reactions,
and the administration of L-arginine would lower,
whereas L-NAME would enhance postischemic
oxidative stress.
However, in our experimental conditions, no modification in the level
of postischemic release of secondary free radicals was observed when
the hearts were perfused in the presence of 3 mM L-arginine
or 100 µM L-NAME in comparison with the inactive
D-enantiomer treatment. Because there was no additional
benefit in terms of postischemic insult, we support the hypothesis that
NOS uncoupling and NOS-driven reactions might not play a major role in
the etiology of the oxidative stress that occurs after 30 min of global
ischemia. This assertion cannot be extrapolated to in vivo conditions
where the presence of circulating immunoinflammatory cells are a
supplementary source of nitric oxide and of oxidative stress.
In conclusion, using either L-NAME as a competitive
inhibitor or L-arginine as a substrate of NOS, in
comparison with specific enantiomers, our study failed to demonstrate
either a positive or negative role of NOS modulation during ischemia
and reperfusion, not only upon functional recovery but also on
postischemic oxidative stress. These results indicate that, in our
experimental conditions, cardiac ·NO synthases might not be
major components implicated in the oxidative burst that follows a
global myocardial ischemia.
We thank Dr. Paul Walker for help.
This work was supported with financial support from the French
Ministry of Research and from the Conseil Régional de Bourgogne.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
