(Circulation. 1995;91:2080-2088.)
© 1995 American Heart Association, Inc.
Articles |
From Laboratorio de Nefrologia-Hipertensión, Instituto de Investigaciones Médicas, Fundación Jiménez Díaz, Facultad de Medicina, Universidad Autónoma, Madrid, Spain.
Correspondence to Carlos Caramelo, MD, Instituto de Investigaciones Médicas, Fundación Jiménez Díaz, Facultad de Medicina, Universidad Autónoma de Madrid, Avda Reyes Católicos, 2, 28040 Madrid, Spain.
| Abstract |
|---|
|
|
|---|
Methods and Results The present study, using platelets and neutrophils from normal subjects, was undertaken to assess the putative effect of aspirin on the neutrophil-mediated, platelet-inactivating effect. Aspirin facilitated the inhibitory effect of neutrophils on platelet activation by thrombin, ADP, or epinephrine. This effect was equally evident in vitro and in blood samples of normal individuals taking aspirin. A significant stimulation of NO-mediated mechanisms in the presence of aspirin was disclosed by different methods, as follows: (1) the increased metabolism of arginine to citrulline, (2) the increase of cGMP in the platelet/neutrophil system, and (3) the inhibitory action of the L-arginine (L-Arg)competitive analogue L-NMMA, which was reversed by L-Arg. The effect of aspirin appeared to be related to cyclooxygenase inhibition, since it was reproduced by using indomethacin. The vasoconstricting peptide endothelin-1 (ET-1) reversed the effect of aspirin through the endogenous production of platelet-activating factor (PAF) by neutrophils, as judged by the marked inhibitory effect of the PAF antagonist BN-52021.
Conclusions Our results show that a significant part of the effect of aspirin on platelet activation involves a neutrophil-mediated, NO/cGMP-dependent mechanism. The presence of ET-1 counterbalances these effects of neutrophils on platelet activation, therefore acting as an indirect proactivating agent. These results add new elements for interpreting the effects of aspirin on the interactions between blood cells, with special reference to high endothelin states (for example, ischemia/reperfusion processes).
Key Words: aspirin platelets endothelin
| Introduction |
|---|
|
|
|---|
With regard to the interactions between platelets and neutrophils affecting platelet aggregation, the results published to date are less than homogeneous. A study by Del Maschio et al12 has shown that neutrophils may favor platelet aggregation, whereas other communications described an antiaggregating effect of neutrophils.13 14 15
Adding more complexity to the interactions between diverse types of cells at the microcirculatory level, activated platelets release vasoactive mediators with potential implications in the pathogenesis of vascular occlusion or vasospasm. In this regard, activating platelets release the vasoconstricting agent thromboxane A2 and induce the production of endothelin-1 (ET-1) by the vascular endothelium.2 ET-1 increases in ischemic processes, but its actual role in the final outcome of the ischemic phenomenon is still incompletely clarified. ET-1 is apparently devoid of any direct effect on platelets2 16 but has, nonetheless, effects on neutrophils, which may account for the interactions between neutrophils, platelets, and endothelial cells. In this regard, ET-1 is involved in neutrophil Ca2+ release,17 superoxide anion generation,18 cGMP production,19 aggregation,20 and adhesion to the endothelium.21
In human pathology, platelet activation is particularly relevant in myocardial ischemia, and several pharmacological strategies have been devised to prevent intravascular platelet activation. Aspirin remains a keystone of these preventive and damage-limiting strategies.22 23 24 Current knowledge maintains that low doses of aspirin decrease in vivo platelet aggregation by a selective inhibitory effect on thromboxane A2 production by platelets with maintenance of prostacyclin production by the endothelium.22 23 24 25 The role of aspirin, however, must be readdressed in light of the new information concerning platelet and neutrophil interactions.
In several studies, antiaggregating effects of aspirin have been reported that cannot be explained only by the above-mentioned action on thromboxane A2 and prostacyclin production. In this regard, Gaspari et al26 found that aspirin prolonged bleeding time in uremia by a mechanism distinct from platelet cyclooxygenase inhibition. Di Gaetano et al27 suggested that the inhibition of cyclooxygenase by aspirin does not fully explain the antithrombotic effects of the drug. The existence of nonplatelet-mediated antithrombotic effects of aspirin was further suggested by Mehta and Mehta.28 It is clear that besides its well-known effects on endothelial cells and platelets, which involve cyclooxygenase inhibition, aspirin affects neutrophil function by mechanisms that are still incompletely understood.27 Especially interesting are the reports showing that in activated neutrophils, aspirin can inhibit the production of superoxide anion29 and antagonize the effects on pulmonary artery contractility.30 Recently, Valles et al31 described that aspirin enhanced the neutrophils' downregulatory effect on thrombin-induced platelet aggregation. These findings, together with the above-mentioned data regarding the cross-talk between platelets and neutrophils by exchanging active metabolites, give additional support to the hypothesis that aspirin may possess a relevant effect on neutrophil/platelet interactions.
Further evidence has been accumulated in recent years about the protective role of aspirin in the control of the complications of myocardial infarction24 30 and in the prevention of coronary bypass closure.32 Of interest, a decreased incidence of nonplatelet-mediated arrhythmias accompanying myocardial infarction, which putatively are a neutrophil-related complication, has been found in aspirin-treated individuals.3 Based on the above-mentioned data, the aim of the present investigation was therefore to examine the role of aspirin in the inactivating effects of neutrophils on activated platelets. In addition, in an effort to further understand the effects of aspirin in pathophysiological conditions, we examined the effects of ET-1, a vasoactive mediator that augments in ischemic circumstances, on the platelet-neutrophil interaction.
| Methods |
|---|
|
|
|---|
The words aspirin and acetylsalicylic acid can be used indistinctly, but preference has been given to aspirin.
Neutrophils and Platelet-Rich Plasma Preparation
All the
experiments were done using human platelet-rich plasma
(PRP) and neutrophils. Rabbit PRP and neutrophils were also used to
examine the validity of the findings in cells of a different species.
Therefore, unless otherwise indicated, the findings reported in the
"Results" section correspond to experiments performed on human
cells. Human or rabbit PRP and neutrophils were obtained from the same
donors for each experiment. With the exception of the experiments with
previous aspirin administration (see below), volunteer donors and
animals did not receive drugs for at least 20 days before the
experiments. All the experiments performed in the present study
were approved by the Clinical Research Committee, and all the subjects
gave written informed consent. Blood extractions in the rabbits were
done by ear venipuncture, after local anesthesia.
Human neutrophils were isolated from peripheral blood by Ficoll/Hypaque centrifugation as previously described.18 19 20 21 Neutrophils (95% pure, 98% viable by trypan blue exclusion) were resuspended in calciumphysiological saline solution (PSS) containing (in mmol/L): 140 NaCl, 4.6 KCl, 2.0 CaCl2, 1.0 MgCl2, 5.0 D-glucose, and 10.0 HEPES, pH 7.4. A similar procedure was followed for isolation of New Zealand rabbit neutrophils.
PRP was prepared as described.33 In brief, whole blood was obtained in 10% (vol/vol) acid-citrate-dextrose and centrifuged at 800g for 15 minutes. PRP was collected and the platelet number counted by a Coulter counter. Platelet number was adjusted with platelet-poor plasma, obtained from the same individual or animal, to 2.5x108 cells/mL plasma.
An additional study was performed to ascertain the in vivo relevance of the in vitro findings. For this purpose, aspirin (200 mg/d, one oral morning dose) was given to human control subjects (n=6) during 4 days, and blood was processed as indicated above. The results of these experiments were compared with those of baseline experiments done in the same individuals before the administration of aspirin. As an estimate of the circulating levels of aspirin, serum salicylate levels were measured by a Du Pont ACA analyzer at 510-nm wavelength.
Platelet Aggregometry
Platelet activation was registered in a
Lumiaggregometer
(Aggrecorder, four channels) by the change in light transmission.
Previously, a platelet-poor sample was used as control for 100% light
transmission. To correct for the possible light absorption induced by
the presence of neutrophil suspension, the platelet-poor sample
contained a number of neutrophils equal to the platelet-rich
sample.
PRP (500 µL) was incubated at 37°C for 10 minutes in the aggregometer with continuous stirring (500 rpm) and was then stimulated with submaximal concentrations of thrombin (0.025 U/mL). Five minutes before, 100 µL of the neutrophil suspension was added to PRP to reach a final amount of 1.25x108 platelets and 1x106 neutrophils (125:1), which approximates the relative concentrations in normal blood. When required, indomethacin, aspirin, or ET-1 was added to the platelet-neutrophil suspension. In all cases, the comparative baseline measurements were done in the presence of the solvent of aspirin and indomethacin, ethanol, in the same dilution used for the incubations with the cyclooxygenase inhibitors (final ethanol concentration in the assay <0.1%). No significant effects were detected in any case with these concentrations of ethanol. In some experiments, PRP was incubated with the specific platelet-activating factor (PAF) antagonist BN-52021 10 minutes before adding ET-1. In other experiments, neutrophils were preincubated with the L-arginine analogue L-NMMA or with a monoclonal antibody against the CD18 antigen of the ß2-integrins, TS1/18, for 45 or 10 minutes, respectively, before they were added to the assay. For standardizing the measurements, only the values of turbidimetry at 3 minutes were used for doing the calculations. This time period corresponds to the maximal or near-maximal value of the first wave of platelet aggregation. This primary wave is representative of platelet activation rather than platelet aggregation and is partially reversible. In this regard, at the relatively high concentrations of thrombin, ADP, or epinephrine used, the first wave was the only relevant process in the platelet activation phenomenon.
Measurement of Intracellular [Ca2+]
Cytosolic [Ca2+]
([Ca2+]i) was measured as previously
described.17 Platelets were loaded by incubating PRP,
prepared as detailed above, with 2 µmol/L fura-2/AM during 60 minutes
at 37°C at room atmosphere. Platelets were pelleted by centrifugation
(2500 rpm, 10 minutes), washed twice, and resuspended in platelet-poor
plasma obtained from the same individual and adjusted to
2.5x108 cells/mL plasma. Neutrophils
(2x106 cells/mL), aspirin (600 µg/mL), or both
were added when required, allowing 5 minutes of preincubation in all
cases. Measurements were performed with an LS-50 Perkin Elmer
Luminescence Spectrometer in a stirring-warmed cuvette (37°C) as
previously described.17 19 Experiments were done in
the
presence of thrombin (0.025 U/mL), ADP (10-6 mol/L), or
epinephrine (10-5 mol/L). The results are expressed in
nmol/L of [Ca2+]i.
Measurement of cGMP
cGMP concentrations were measured as
described
previously19 in acetylated samples by means of a
radioimmunoassay kit (Amersham). PRP, neutrophils, or both were
preincubated for 10 minutes at 37°C with the phosphodiesterase
inhibitor 3-isobutyl-1-methylxanthine (IBMX; 2 mmol/L). cGMP was
measured in suspensions of platelets, neutrophils, and platelets plus
neutrophils, both in the presence and absence of aspirin. cGMP levels
were measured 3 minutes after the cells were activated with thrombin
(0.025 U/mL) to reproduce the conditions of the platelet aggregation
experiments.
Determination of L-[3H]-Citrulline
Content
As detailed elsewhere,19 neutrophils were
incubated
(30 minutes, 37°C) in PSS-Ca2+ containing
10-4 mol/L L-arginine and 1 µCi/mL
L-[3H]-arginine. Unincorporated
L-[3H]-arginine was washed twice with
PSS-Ca2+ buffer. After
L-[3H]-arginine labeling, neutrophils were
coincubated with PRP in the presence and absence of aspirin (600
µg/mL) or vehicle (ethanol, final concentration <0.1%). Platelets
were activated with thrombin (0.025 U/mL) during 3 minutes at 37°C.
Further control experiments in the absence of thrombin-stimulated
platelets were carried out. After centrifugation, PRP and neutrophils
were lysed with cold methanol, and the supernatant was evaporated to
dryness under N2 at 37°C. As previously
described,19 the extracts were resuspended in 20 mmol/L
HEPES/KOH, pH 5.5, and applied to columns of Dowex AG50WX8
(Na+ form), which were subsequently eluted with water
(L-citrulline fraction) and 0.5 mol/L NaOH
(L-arginine fraction).
L-[3H]-citrulline fraction was quantified by
liquid scintillation counting.
Measurement of PAF Production
PAF production by the
neutrophils was measured as
[3H]-acetate incorporation into PAF.
[3H]-acetatelabeled neutrophils (25
µCi/5x106 neutrophils) were resuspended in
PSS-Ca2+ buffer and incubated at 37°C for 3
minutes in the presence or absence of vehicle (ethanol <0.1), aspirin
(600 µg/mL), or 10-7 mol/L ET-1. Incubation was finished
by the addition of 3 mL 1 mol/L HCl in methanol. Lipids were extracted
from the methanolic phase as described by Bligh and
Dyer.34 Samples were dried under N2 and
separated by thin-layer chromatography on precoated plates of silica
gel 60, using propionic:propanol:water:chloroform (2:2:1:1) as the
mobile phase. The silica was scraped off in narrow bands, based on
their coimigration with a PAF standard.35 Radioactivity
was determined by liquid scintillation counting.
To be certain that the measured material corresponded to [3H]-PAF, selected samples were analyzed by high-performance liquid chromatography using a dual Kontron model 420, as previously described.20 The retention time of [3H]-PAF obtained from neutrophils was identical to a synthetic labeled PAF (C16-C18 mixture, Amersham).
Statistical Methods
Results are expressed as mean±SEM.
Unless otherwise stated,
each value corresponds to a minimum of six experiments done in
triplicate. Comparisons were done by ANOVA or paired and unpaired
Student's t test when appropriate. Bonferroni's
correction
for multiple comparisons was used to determine the level of
significance of the P value.
| Results |
|---|
|
|
|---|
The data
represented in Fig 1
were taken at
3 minutes after the stimulus, corresponding to maximal platelet
activation. However, the turbidimetry curves in the presence or absence
of aspirin were already different when measured at 1 minute from the
start of the experiment (% light transmission: -aspirin, 26±2,
+aspirin, 16±2; P<.01). The effect of aspirin was
concentration dependent (Fig 1
), showing the maximal inhibition
with
600 µg/mL. This latter concentration was used for most of the
following studies. In the absence of aspirin, no significant effect of
neutrophils on platelet activation was detected (Fig 1
). To
examine
whether the effect of aspirin was due to its cyclooxygenase-inhibiting
action, similar experiments were done in the presence of a different
cyclooxygenase inhibitor, namely, indomethacin. A 38±6% inhibitory
effect of platelet activation by neutrophils was observed in the
presence of 10-6 mol/L indomethacin (n=4,
P<.01 with respect to controls). This effect was not
different from that obtained with aspirin (P=NS).
|
In
the absence of neutrophils, aspirin alone (600 µg/mL) did not
result in changes of thrombin-induced platelet activation (Fig
1
). A
significant inhibition of thrombin-induced platelet activation was
still present when only the neutrophils were previously treated
with aspirin (% inhibition, 20±3, P<.05), albeit with
lesser intensity than that observed when both platelets and neutrophils
were incubated with aspirin.
A concentration-dependent, aspirin-related
inhibition of platelet
activation by neutrophils was also observed when ADP or epinephrine was
used as platelet activator (Fig 2A
and 2B
). An
effect of
aspirin per se was detected on the epinephrine-induced activation curve
(Fig 2B
); also, a small but significant effect of aspirin on
ADP-stimulated platelet activation was observed (Fig 2A
).
|
The spontaneous platelet activation (% light transmission <5%) was not changed by the incubation of PRP with neutrophils and aspirin. Moreover, no differences were observed in the above-mentioned effects of aspirin when similar aggregometry experiments were done using rabbit PRP and neutrophils (data not shown).
Measurements of
[Ca2+]i transients in
platelets were performed as a second method to examine platelet
activation in the presence or absence of aspirin and neutrophils. As
shown in Table 1
, thrombin-induced
[Ca2+]i transients in platelets were
not modified by neutrophils or aspirin (600 µg/mL) alone. In the
presence of aspirin, neutrophils markedly decreased the
thrombin-stimulated [Ca2+]i transients
in platelets (Table 1
). A neutrophil-mediated,
aspirin-dependent
inhibition of [Ca2+]i transients in
platelets was also observed when ADP (10-6 mol/L) or
epinephrine (10-5 mol/L) was used as platelet activator
(% decrease of [Ca2+]i peak, 62±7
and 68±6, respectively, both P<.01). An inhibitory effect
induced by aspirin on ADP- or epinephrine-stimulated
[Ca2+]i transients was also observed
in the absence of neutrophils, although with lesser intensity than that
shown when neutrophils were present (% decrease of
[Ca2+]i peak by aspirin in the absence
of neutrophils: ADP, 10±3; epinephrine, 18±6;
P<.05).
|
Baseline [Ca2+]i in platelets (110±8 nmol/L) was not modified by the incubation of PRP with neutrophils or aspirin.
The inhibitory effect of neutrophils in the presence of aspirin on the [Ca2+]i transients in either thrombin-, ADP-, or epinephrine-stimulated platelets was reverted by the preincubation of neutrophils with the L-argininecompetitive analogue L-NMMA (10-5 mol/L) (% inhibition of the aspirin effect, 88±8, 82±5, and 84±3, respectively, P<.01).
In an effort to address the actual in
vivo importance of the
above-mentioned findings, the role of neutrophils in platelet
activation was examined in normal subjects in samples obtained before
and after the oral intake of aspirin. In all cases, each individual was
used as his or her own control before and after the intake of aspirin;
moreover, in all cases, both platelets and neutrophils were obtained
from the same donor. In these experiments, neutrophils obtained before
aspirin treatment did not significantly influence platelet activation
induced by thrombin (Fig 3
). However, after oral aspirin
administration, a significant inhibition of thrombin-induced platelet
activation by neutrophils was observed (Fig 3
). The serum
salicylate
level in these individuals was 49.4±0.4 µg/mL, which corresponds to
68.6±0.5 µg/mL of aspirin on a molar basis.
|
In the absence
of neutrophils, platelet activation in PRP taken from
the human donors before or after oral aspirin administration was
similar (Fig 3
).
Mechanisms of the Effect of Aspirin on Platelet/Neutrophil
Interactions: Role of Nitric Oxide Synthesis Inhibition
Based on the
data of Salvemini et al,13 the
hypothesis was raised that NO would be a relevant mediator of the
aspirin-induced effect of neutrophils on platelet activation.
Subsequently, to examine the implication of NO in the above-described
effects, neutrophils were preincubated with L-NMMA (10-5
mol/L). Incubation of neutrophils alone with L-NMMA blocked the
aspirin-sensitive inhibition of platelet activation by neutrophils (%
inhibition, 92±6; P<.05). Preincubation of neutrophils
with L-NMMA produced a concentration-dependent inhibition of the
antiaggregating effect in the presence of aspirin (% inhibition by the
threshold concentration 10-8 mol/L, 7±1;
P<.05). The inhibitory effect of neutrophils on
thrombin-stimulated platelet activation in the presence of lower
concentrations of aspirin was also blocked by L-NMMA (data not shown).
However, no effect of L-NMMA was detected on thrombin-induced platelet
activation in the absence of neutrophils (% light transmission in
L-NMMApreincubated PRP stimulated by thrombin, 78±9; n=3;
P=NS). Like in the in vitro studies, L-NMMA also suppressed
the neutrophil-induced inhibition of platelet activation in samples
obtained from aspirin-treated donors (Fig 3
).
More data supporting the putative role of NO in the effect of neutrophils on platelet activation were obtained by preincubating neutrophils in the presence of excess L-arginine, the substrate for NO formation. In this regard, the addition of L-arginine (10-4 mol/L) to 10-5 mol/L L-NMMApreincubated neutrophils restored up to 86±4% of the inhibitory effect of neutrophils in the presence of aspirin. In the absence of aspirin and L-NMMA, L-arginineloaded neutrophils showed a small but significant inactivating effect (% inhibition, 11±3; n=3; P<.05).
Additional experiments were done to examine the role of oxygen free radicals on the platelet-inactivating effect of neutrophils observed in the presence of aspirin. In this regard, a small but significant effect of the superoxide anion scavenger superoxide dismutase (60 U/mL) was detected on the platelet-inactivating effect of neutrophils in the presence of aspirin (% inhibition of platelet activation by neutrophils in the presence of 300 µg/mL of aspirin, 37±4% with and 28±3% without superoxide dismutase; P<.05). Furthermore, a small effect of superoxide dismutase by itself was detected on platelet activation in the presence of neutrophils incubated without aspirin (% inhibition, 10±2; n=3; P<.05). Of note, aspirin concentrations with submaximal effect were used in these experiments to facilitate the observation of any putative interaction. No effect was detected on either spontaneous activation or thrombin-induced platelet activation by the incubation of platelets alone with superoxide dismutase (data not shown).
Formation of Nitric Oxide by Neutrophils
To determine whether
aspirin favored NO generation by neutrophils,
we studied the accumulation of
L-[3H]-citrulline in
L-[3H]-arginineloaded neutrophils. To
reproduce the conditions of the platelet activation experiments,
platelets and thrombin were also added during the assay, with a time
and concentration profile similar to that used in the experiments
referred to above. In the presence of aspirin, the generation of
L-[3H]-citrulline on neutrophils increased
significantly (Table 2
), indicating that the
aspirin-sensitive inhibition of platelet activation mediated by
neutrophils is coupled to an increased NO production. The stimulatory
effect of aspirin on NO production did not need the presence of
activated platelets, since it occurred also when neutrophils alone were
incubated with aspirin (Table 2
). Thrombin did not change NO
production
by neutrophils in the absence of platelets (Table 2
).
|
Effect of Aspirin on cGMP Levels
The inhibition of
thrombin-stimulated platelet activation mediated
by aspirin and neutrophils was accompanied by a potentiated increase of
cGMP levels (Fig 4
). On the other hand, when the cGMP
levels were measured in the mixed cellular suspensions (platelets plus
neutrophils) in the absence of aspirin, they increased only in an
additive manner (Fig 4
).
|
The presence of aspirin did not
alter platelet baseline cGMP levels.
However, aspirin increased neutrophil cGMP content (Fig 4
),
although in
a small proportion.
Effect of Endothelin-1 on Aspirin-Dependent Modulation of Platelet
Activation by Neutrophils
An additional set of experiments was
performed to examine
whether the above-mentioned findings were valid in pathological
conditions. In particular and based on recent
data18 19 20 21
showing its effects on neutrophil activation, we studied the effect of
ET-1, a mediator that increases during ischemia. Experiments were
therefore conducted to determine whether the presence of ET-1 could
modify the interaction between neutrophils and platelets. Table
3
summarizes the effects of ET-1 (10-7
mol/L) on the platelet/neutrophil interaction in the presence of
aspirin. ET-1 blocked the aspirin-dependent inhibition of
thrombin-stimulated platelet activation mediated by neutrophils.
Moreover, in the absence of aspirin, the addition of ET-1 to the
platelet-neutrophil mixture elicited a small but significant increase
in spontaneous platelet activation (% light transmission, 12±2;
P<.05 with respect to spontaneous activation without ET-1).
This effect of ET-1 appeared to involve mostly the neutrophil side of
the interaction, since in the absence of neutrophils, ET-1 did not
stimulate platelet activation (% light transmission, 4±1;
P=NS) or change thrombin-induced activation (% light
transmission, 77±5; P=NS). The concentration of ET-1
was
chosen based on our previous studies, in which 10-7 mol/L
ET-1 consistently provoked a maximum peak of
[Ca2+]i on neutrophils, which
correlated with neutrophil activation and neutrophil adhesion to the
endothelial
surface.17 19 20 21
Simultaneous experiments were
done based on the previous evidence showing that ET-1 provokes
neutrophil activation20 in order to rule out the
possibility that the blocking action of ET-1 on the
platelet-antiaggregating effect of neutrophils could be due to the
ET-1induced activation of neutrophils. These additional experiments
were performed by preincubating the neutrophils with a monoclonal
antibody against the CD18 antigen TS1/18 to block the
neutrophil-neutrophil aggregating interactions induced by ET-1, as
previously observed in our laboratory (López-Farré et al,
unpublished observations). The inhibitory effect of ET-1 was not
dependent on the increase of neutrophil activation, since it persisted
in TS1/18-preincubated neutrophils (Table 3
). No effect of the
TS1/18
antibody by itself was detected on the platelet activation curve either
in the presence or in the absence of neutrophils (P=NS, data
not shown).
|
Thereafter, since ET-1 was devoid of direct actions on
platelets, we
examined the hypothesis that ET-1 induces the release of a mediator
from neutrophils that blocks the aspirin-dependent inhibition of
platelet activation. In this regard, we have found previously that ET-1
stimulates PAF release from neutrophils.20 Therefore, we
investigated the role of PAF as a possible mediator of the ET-1 effects
described above by preincubating the platelets during 10 minutes with
the specific PAF antagonist BN-52021 (5x10-5 mol/L) and
then adding the neutrophils and ET-1. As shown in Table 3
, the
blockade
by ET-1 of the platelet-antiaggregating effect of ET-1 was reverted by
BN-52021. No effect of BN-52021 by itself was detected on
thrombin-induced platelet activation in the absence of neutrophils
(Table 3
). To further ascertain whether aspirin would interfere
with
ET-1induced PAF production by neutrophils, additional experiments to
examine PAF production by neutrophils were carried out in the presence
or absence of aspirin. These experiments confirmed that ET-1 induced
PAF production by neutrophils and disclosed no effect of aspirin on
this phenomenon (Table 4
).
|
| Discussion |
|---|
|
|
|---|
The existence of platelet inactivation by neutrophils was demonstrated in 1989 by Salvemini et al.13 These authors have also found that this effect was mediated by an NO-dependent, cGMP-mediated mechanism. On the other hand, Del Maschio et al12 described a proactivating effect of neutrophils. In part, the apparent disagreement of the results from these two groups could be attributed, in light of the data obtained in the present study, to the presence13 or absence12 of aspirin in the platelet incubation media and to relevant differences in the experimental conditions, namely, use of washed platelets12 or PRP, prestimulation of the neutrophils with cytochalasin B,13 and aequorin loading.12 The results by Salvemini et al13 are in agreement with ours and with those of other authors, including other communications from Del Maschio's group.12 15 The new information in our experiments includes the finding of the aspirin dependence of the inactivating effect of neutrophils, the relation of aspirin and NO production, and the potential role of ET-1 through a PAF-related pathway.
The effect of aspirin was evident within a wide range of concentrations. Moreover, in the presence of aspirin, neutrophils not only counteracted the effect of thrombin but also those of ADP and epinephrine, therefore suggesting that platelet activation was blocked at a postreceptor level. Importantly, the concentrations of aspirin used are within the in vivo therapeutic range of the drug.23 24 25 36 37 38 In this regard, the present experiments using blood from subjects taking aspirin confirm the in vivo relevance of the described in vitro effects and the similarity of the NO-dependent mechanisms.
Different types of experiments were performed to address the mechanisms involved in the effect of aspirin on neutrophils. First, the identity of the effects of both aspirin and indomethacin strongly suggested that the main effect of aspirin in the neutrophil-mediated platelet inactivation was due to cyclooxygenase inhibition. The inhibition of the effect of neutrophils plus aspirin on platelet aggregation by the L-arginine competitive analogue L-NMMA supported the NO-mediated nature of the inactivating phenomenon. The specificity of this inhibition was established by the reversal of the L-NMMA effect by excess L-arginine. The experiments using separate incubations with aspirin of either neutrophils alone or neutrophils with platelets suggested that the effect of aspirin on platelet activation and L-[3H]-citrulline production occurred preferentially on the neutrophil side of the neutrophil-platelet association. Therefore, this effect appears to involve either the existence of a mediator that would appear in conditions of cyclooxygenase inhibition and activate NO production from neutrophils or the interference by aspirin with an inhibitory effect on NO formation.
Further insight into the characteristics of the aspirin effect was obtained by studying the [Ca2+]i response, which allowed a very sensitive assessment of the early events of platelet activation. In our study, in the presence of aspirin, neutrophils interfered with the [Ca2+]i signaling pathway by an L-NMMAinhibitable mechanism. Therefore, it appears that NO can inhibit platelet activation through the blockade of calcium-stimulated pathways. The fact that thrombin-mediated as well as ADP- and epinephrine-mediated responses were affected showed that aspirin blocked a postreceptor mechanism involving [Ca2+]i mobilization. An explanation for these results could be found in the increase of cGMP observed when platelets were incubated with aspirin-treated neutrophils, since cGMP is a potent blocker of calcium-related activation in numerous cellular types, including platelets.39
The present experimental model did not allow specific identification of the putative agent(s) responsible for the aspirin effect on NO production. However, as suggested by preliminary observations in our laboratory,40 a role by arachidonic acid or noncyclooxygenase arachidonic acid metabolites is suspected. Arachidonic acid or its metabolites can be expected to increase after cyclooxygenase inhibition by aspirin. Arachidonic acid fulfills several conditions to be considered a possible mediator, since it can activate NO release from endothelial cells41 and has been proposed as a direct modulator of neutrophil function.42 Despite the possible effects of aspirin on the inhibition of superoxide anion production,27 28 which may have consequences on the rate of NO degradation,43 44 our results support the existence of an increased NO formation induced by the presence of aspirin. In this regard, the absence of a relevant effect of superoxide dismutase in the present experimental conditions diminishes the potential importance of a superoxide anionrelated mechanism. However, an effect of aspirin on the platelet side of the association can be suspected from the observation that, even though the increase in L-[3H]-citrulline production was already at maximum when neutrophils were incubated with aspirin in the absence of platelets, the platelet antiactivating effect of the neutrophils was more intense when both platelets and neutrophils were simultaneously incubated with aspirin.
Several recent findings indicated that aspirin might contribute to increased NO levels or NO effects. In this sense, an antagonistic effect of aspirin has been observed on the vasoconstrictor properties of the L-argininecompetitive analogue L-NMMA45 and hypoxic vasoconstriction.46 Even closer to our results, aspirin might alter the contractile effects of neutrophils on the pulmonary artery.30 In addition, salicylate, a part of the aspirin molecule, has been shown recently to interfere with leukocyte adhesion,47 a property modulated by NO.48 Finally, as a direct demonstration, our experiments showed that aspirin increased NO in neutrophils.
The results of the experiments using ET-1 illustrate the possible in vivo outcome of the platelet/neutrophil/aspirin interactions in conditions of increased ET-1. The inhibitory effect of ET-1 on the neutrophil-dependent platelet inactivation adds more insight into the possible platelet/neutrophil interactions in high ET-1 states, namely, ischemia/reperfusion or endothelial damage of different causes. As observed by other authors16 and confirmed in the present study, ET-1 appears to be devoid of a direct effect on platelets. Our results, however, showed for the first time that ET-1 may indirectly modulate platelet activation by interfering with the antiaggregating effect of neutrophils. Actually, in the presence of ET-1, the neutrophils appear to promote platelet activation rather than inactivation. Since this effect was not inhibited by the TS1/18 antibody, it can be considered to be independent of the neutrophil-aggregating effect of endothelin.20 A role of P-selectin,47 48 49 which is involved in the platelet-neutrophil physical interaction,49 50 51 is theoretically possible and cannot be ruled out from our results. However, recent data from Rinder et al52 suggested that aspirin does not affect P-selectin surface expression and therefore do not support the hypothesis of a P-selectinrelated effect.
The inhibition of the effects of ET-1 on the platelet-neutrophil association by the PAF antagonist BN-52021 strongly suggests that it is due to PAF-induced activation. Moreover, PAF production by neutrophils was enhanced in the presence of ET-1, an effect that was not blocked by aspirin (see Reference 20 and the present study). This effect may account for the observed neutrophil-dependent platelet activation in the presence of ET-1. ET-1 may, therefore, act as an in vivo negative modulator of the platelet-antiaggregating effect of neutrophils through a PAF-mediated mechanism. This pathway may be potentially relevant in pathological conditions associated with increased local levels of ET-1.
The observed effects of aspirin are of potential interest in the setting of ischemia/reperfusion, particularly in the myocardium. In circumstances of coronary occlusion and subsequent reperfusion, there exists a deep perturbation of endothelial function, with decreased formation of vasodilating/antiaggregating mediators, for example, NO and prostacyclin, and blunted endothelium-dependent relaxation.53 54 The role of neutrophils, which adhere to the damaged endothelium, is not completely understood in this setting.55 Of further interest, recent results obtained in our laboratory indicate the existence of neutrophil-related effects of NO on red blood cells, which may be relevant for the regulation of microcirculatory hemorheology and may expand the scope of the possible blood cell interactions beyond those mentioned in the present study.56
Conclusions
The present results raise the possibility that
aspirin
modifies the way in which neutrophils and platelets interact. These
findings may, therefore, add a novel interpretation to the protective
effect of aspirin in the control of ischemic damage.
| Acknowledgments |
|---|
Received October 12, 1994; accepted November 26, 1994.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. Patrono, C. Baigent, J. Hirsh, and G. Roth Antiplatelet Drugs: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition) Chest, June 1, 2008; 133(6_suppl): 199S - 233S. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. A. Meadows and D. L. Bhatt Clinical Aspects of Platelet Inhibitors and Thrombus Formation Circ. Res., May 11, 2007; 100(9): 1261 - 1275. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. G.L. Biondi-Zoccai, M. Lotrionte, P. Agostoni, A. Abbate, M. Fusaro, F. Burzotta, L. Testa, I. Sheiban, and G. Sangiorgi A systematic review and meta-analysis on the hazards of discontinuing or not adhering to aspirin among 50 279 patients at risk for coronary artery disease Eur. Heart J., November 2, 2006; 27(22): 2667 - 2674. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Carrasco-Martin, S. Alonso-Orgaz, J. C De la Pinta, M. Marques, C. Macaya, A. Barrientos, M. M Gonzalez, A. Garcia-Mendez, P. J. Mateos-Caceres, J. C Porres, et al. Endothelial hypoxic preconditioning in rat hypoxic isolated aortic segments Exp Physiol, July 1, 2005; 90(4): 557 - 569. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Patrono, B. Coller, G. A. FitzGerald, J. Hirsh, and G. Roth Platelet-Active Drugs: The Relationships Among Dose, Effectiveness, and Side Effects: The Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy Chest, September 1, 2004; 126(3_suppl): 234S - 264S. [Abstract] [Full Text] [PDF] |
||||
![]() |
|