| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Thrombosis |
From the Division of General Internal Medicine, Department of Internal Medicine, University of Innsbruck, Innsbruck, Austria.
Correspondence to Prof Christian J. Wiedermann, MD, Department of Internal Medicine, University of Innsbruck, Anichstrasse 35, A-6020 Innsbruck, Austria. E-mail Christian.Wiedermann{at}uibk.ac.at
| Abstract |
|---|
|
|
|---|
Key Words: artherosclerosis platelets platelet function inhibitors vascular biology
| Introduction |
|---|
|
|
|---|
ADP release from activated platelets induces further platelet recruitment, followed by cell aggregation via binding to platelet purinergic type (P) P2X-, P2T-, and P2Y-receptors.18 Exposure to ADP results in a rapid increase in Ca2+ influx, Rho-GTPasedependent phopholipase C activation, and inhibition of stimulated adenylate cyclase. ADP causes granule release and thromboxane A2 generation.18 Neutrophils exert locomotive activity in response to ADP and ATP released from activated platelets via activation of P2Y4 and P2Y6 receptors.19 Direct interactions of neutrophils and platelets are mediated through P-selectin (CD62P), the expression of which represents the first step in the formation of a leukocyte-platelet thrombus in vivo.20 An increase of these cell-cell interactions was observed in patients with acute myocardial infarction.21 Simvastatin (SVN) treatment resulted in normalization of altered platelet aggregation and reduced the thromboxane-metabolite excretion.6,7 As clinical benefits of statins are evident even in patients receiving aspirin,6 reduction in thromboxane per se is unlikely to be the explanation for antithrombotic effects. Indirectly, modification of endothelial function may alter platelet activity,12,14 even though some antiplatelet effects of statins are thought to be lipid dependent.2,22 There is also one contradictory study showing no effects of SVN on platelet function.23
Atherosclerosis patients suffer from an increased oxidant tone as reflected by augmented isoprostane formation.24 Prostaglandin F2-like compounds are potent platelet activators and smooth muscle cell mitogens; they can be formed nonenzymatically by free radical attack of arachidonic acid in cell membranes.24 By exerting anti-oxidant effects, statins could reduce platelet activity. In direct interactions with neutrophils, pravastatin or cerivastatin (CVN) had no effects on respiratory burst.16,17 Pravastatin may inhibit NADPH-oxidase activation of oxidative burst in a cell line.25
Resting platelets limit the release of oxygen radicals from chemoattractant-stimulated neutrophils, which may be affected by aspirin or clopidogrel administration, that limit this effect.26 This mechanism may contribute to the prevention of excessive damage of host tissue in the vasculature. In contrast, activated platelets stimulate superoxide anion production and release from neutrophils and monocytes.27 Circulating platelets can be activated by hemostatic (eg, thrombin, ADP) but also by proinflammatory stimuli (eg, interleukin-1, interferon-
), which may thus further augment oxygen radical release from neutrophils, resulting in a positive feedback regulation.28 The role of statins in this leukocyte-platelet interaction is as yet unknown and was, therefore, investigated in the present study. We confirm that thrombin-activated platelets enhance respiratory burst activity of neutrophils and show an increased ADP/ATP release; these two thrombin-induced effects are abolished when platelets are treated with CVN and SVN.
| Methods |
|---|
|
|
|---|
Preparation of Neutrophils
Neutrophils were obtained from peripheral EDTA-anticoagulated blood of healthy volunteers after discontinuous density gradient centrifugation through a layer of Ficoll-Hypaque, followed by hypotonic lysis of contaminating erythrocytes. Cell preparations yielded >95% neutrophils (by morphology in GIEMSA stains) and >99% viability (by trypan dye exclusion).29 After staining with Sarsted Thrombo Plus, no platelets could be found in the neutrophil fraction.
Preparation of Platelets
Healthy volunteer blood donors took no medication for at least two weeks before the study. Blood donors were all normocholesteremic. Human platelets were obtained from peripheral blood anticoagulated with acid citrate dextrose by a two-step centrifugation at 110g and 180g. Platelets were then washed in HBSS and counted with Sarstedt Thrombo Plus and resuspended in RPMI 1640 containing 0.05% bovine serum albumin for measurement of adenosine release. Plastic or siliconized glassware was used for all platelet manipulations.26
Treatment of Neutrophils and Platelets
Neutrophils
Neutrophils were incubated concomitantly with CVN (1 pmol/L to 100 µmol/L) and thrombin (10 µU/mL to 100 mU/mL) for 45 minutes. Preincubation experiments were also performed. These cells were exposed to CVN alone for 30 minutes and washed twice; thereafter, thrombin was added for an additional 15 minutes. After washing cells, superoxide anion release was measured.
Platelets
Resting platelets were activated with thrombin (0.1 mU/mL). To examine whether CVN or SVN is able to inhibit activation of platelets and what the subsequent impact on neutrophil respiratory burst activity is, coincubation-experiments were performed with platelets and various concentrations of statins (1 pmol/L to 100 µmol/L) and thrombin (0.1 m U/mL) for 15 minutes at 37°C (humidified atmosphere). The platelet-neutrophil ratio was 50:1 and incubation (interaction) time was 30 minutes. After this period, respiratory burst experiments were performed. For metabolic substitution, MVA (500 µmol/L) was added to the cells.
Respiratory Burst Activity of Neutrophils
Neutrophil respiratory burst activity was detected by an assay with the fluorochrome DCFH-DA. One hundred microliters per well (96-well plate) of 2x105 neutrophils with or without platelets were immersed in a 10 µmol/L solution of DCFH-DA in phenol red-free HBSS with or without fMLP (1 µmol/L) as a triggering agent. Plates were incubated at 37°C/5%CO2. Fluorescence activity was determined at 485±20 nm excitation and 530±25 nm emission wavelengths by using the CytoFluor 4000 fluorescence measurement system (Millipore Corp).
Measurement of Adenine Compounds in Platelet Supernatants by HPLC
Platelets (108 cells/mL) were coincubated with statins at various concentrations (1 pmol/L to 100 µmol/L) and thrombin (0.1 mU/mL) or left untreated. After an incubation period of 20 minutes, the supernatants were harvested by centrifugation. Three hundred microliters of each sample were subjected to HPLC-analysis.
Separation of ATP and ADP30 was performed on a Lichrosorb RP-18 column (200x4.6 mm, 5 µm; Agilent Technologies) connected to a ZORBAX SB-C18-precolumn (12.5x4.6 mm, 5 µm; Agilent Technologies). Signals were detected with a UV-visible detector (HP series 1100, Hewlett-Packard). The chromatographic conditions were as follows: isocratic elution at room temperature with 0.1 mol/L NaH2PO4 (pH 6.0 NaOH)/methanol (96%/4%); flow rate 0.7 mL/min; pressure approximately 122 bar; UV-visible detector wavelength, 254 nm; injection volume, 30 µL; retention time for ATP, 5.43 minutes; retention time for ADP, 6.13 minutes; elution time, 11 minutes.
Western Blot Analysis of Rho-GTPases
Platelets were incubated with thrombin (0.1 mU/mL) and SVN (1 pmol/L to 100 µmol/L) and dissolved in RPMI for 20 minutes. Medium-treated cells served as controls. Cells were lysed in Nonidet P40 (Roche) lysis buffer. Proteins were separated on 7.5% SDS polyacrylamide gels and blotted onto polyvinylidene fluoride membranes. The antibody was then diluted to a final concentration 0.05 µg/mL, and blots were incubated overnight at room temperature. Immunoreactivity was determined by using peroxidase-conjugated goat anti-mouse IgG and the enhanced chemiluminescence reaction (Amersham). Intensity of the Western blot bands was quantified with the Fluor-S MultiImager System and the Quantity One Software (BioRad Laboratories).
Chemotaxis Experiments
Platelets were incubated with thrombin and various concentrations of SVN (1 pmol/L to 100 µmol/L) for 20 minutes. Supernatants were harvested and subjected to chemotaxis experiments, which were performed with modified 48 blind-well microchemotaxis chambers equipped with 5-µm-poresized nitrocellulose filters. The supernatants were in the lower part and the neutrophils in the upper part of the chamber. Neutrophils were allowed to migrate toward the platelet supernatants for 30 minutes. As a control, exogenous ADP (10 µg/mL to 40 µg/mL) with and without SVN (100 µmol/L) was added to the lower part of the chemotaxis chamber. To exclude a potential direct influence of SVN, neutrophils migrated toward SVN (1 pmol/L to 100 µmol/L) with or without ADP (20 µg/mL). Data are expressed as chemotaxis index, which is the ratio between the distance of directed and undirected migration.
Statistical Analysis
Data are expressed as mean and SEM. Means were compared by Mann-Whitney U-test and Wilcoxon signed rank test after Kruskal-Wallis ANOVA. A P value <0.05 was considered significant. Analyses were performed with the StatView software package (Abacus Concepts).
| Results |
|---|
|
|
|---|
Platelet-Mediated Effects on Respiratory Burst Activity of Neutrophils
An increase of superoxide anion production in fMLP-triggered respiratory burst by activated platelets was observed as was previously demonstrated.27 Resting platelets significantly inhibited fMLP-triggered respiratory burst activity of neutrophils, whereas thrombin-treated (activated) platelets increased respiratory burst of neutrophils. CVN or SVN had no influence on the decrease of superoxide anion production by resting platelet in neutrophils, whereas coincubation of platelets with thrombin (0.1 mU/mL) and the statins (1 pmol/L to 100 µmol/L) led to significant inhibition of the thrombin-dependent increase of oxidative burst activity (Figure 1). The CVN- or SVN-induced extent of inhibition was comparable to the effect caused by resting platelets and was insensitive to MVA (Table 1).
|
|
Respiratory Burst of Neutrophils Affected by Blockade of P2-Receptor
To determine whether inhibition of respiratory burst by statin-treated platelets is due to an inhibition of P2-receptor signaling, experiments with P2 blockade by trypan blue19 were performed. Thrombin activated platelets coincubated with trypan blue, a nonselective P2 antagonist, decreased respiratory burst of neutrophils significantly. Activated platelets incubated with SVN (1 µmol/L) and trypan blue had additive effects in inhibiting superoxide anion release from neutrophils. (Figure 2).
|
Statins Effects on ADP and ATP Release From Platelets
Because it is known that platelets enhance their ADP and ATP secretion on activation, we investigated whether coincubation of the statins and thrombin leads to a decreased release of the nucleotides.
Platelets were incubated with CVN or SVN (1 pmol/L to 100 µmol/L) or thrombin (0.1 mU) or left untreated. CVN and SVN inhibited thrombin-stimulated ATP and ADP release from platelets. Significant effects on ADP secretion were observed with 1 µmol/L and 100 µmol/L CVN, whereas the ATP release was significantly diminished at CVN concentrations of 100 pmol/L to 100 µml/L. SVN inhibited ATP release at all concentrations tested, and ADP release was diminished significantly at concentrations between 100 pmol/L and 100 µmol/L (Figure 3).
|
Statin Effects on Rho-GTPases
To delineate whether treatment of platelets with CVN and SVN inhibits dissociation of Rho-GTPases from its cytoplasmatic dissociation inhibiting protein RhoGDI, we activated platelets with thrombin (0.1 mU/mL) and coincubated them with various concentrations of the statins (1 pmol/L to 100 µmol/L). Quantification of the bound monoclonal antibody to unbound, which means RhoGDI without associated Rho-GTPases, cytoplasmatic RhoGDI showed an increase in thrombin- and ADP-treated platelets and a concentration dependent decrease in statin-treated platelets (Figure 4).
|
Effects of Statin-Treated Platelet Supernatants on Chemotactic Response of Neutrophils
Supernatants from activated platelets induced a significant increase of migration of neutrophils into micropore filters of modified Boyden chambers compared with untreated platelets. SVN treatment of platelets diminished this response dose dependently. In the absence of platelets, neither did SVN itself influence random migration nor did ADP trigger chemotaxis of neutrophils (Figure 5).
|
| Discussion |
|---|
|
|
|---|
Several lines of evidence suggest that platelet hyperactivity might play an important role in enhancing the risk of thrombotic complications of atherosclerosis. Enhanced platelet activation and increased respiratory burst activity of neutrophils lead to endothelial dysfunction and loss of antithrombotic endothelial properties.31 The adhesion of platelets to neutrophils was shown to be mediated by the interaction between P-selectin on platelets and sialyl Lewis(x) (CD15) on neutrophils.32,33 This binding may trigger the respiratory burst in neutrophils.27 According to another hypothesis, ADP metabolism on the surface of neutrophils is the mechanism regulating burst, ie, inhibiting burst by adenosine or stimulating it by ADP.34,35 Because statin treatment is reported to reduce expression of P-selectin on platelets,36 we investigated possible roles of CVN and SVN in neutrophil-platelet interactions and found that thrombin-activated and CVN- or SVN-treated platelets decreased oxidative burst of neutrophils. In this investigation, we confirm that thrombin-activated platelets increase respiratory burst activity of neutrophils. This effect is significantly diminished when platelets are exposed to CVN or SVN at concentrations as low as 100 pmol/L. Statin-treated thrombin-activated platelets decreased their ADP and ATP release in a concentration-dependent manner at concentrations between 1 pmol/L and 100 µmol/L. Plasma levels in therapeutic administration of statins are a 10- to 100-fold higher than concentrations used in this study.
In hypercholesterolemic patients, CVN and SVN were shown to reverse increased platelet-derived thrombin generation.36,37 As thrombin is a potent agonist for a number of biological responses that may mediate inflammatory and reparative responses to vascular injury,37 patients with increased thrombin generation are considered to be at increased cardiovascular risk. Our data show that thrombin-activated platelets increase respiratory burst activity of neutrophils, an effect that was previously observed with the use of other platelet-activating agents.38
Thrombin augments ADP and ATP secretion in human platelets through binding to the platelet-activating receptors (PAR) -1, -2 and -4.39 When platelets were stimulated with thrombin, we observed that ADP/ATP release was effectively diminished by coincubation with statins. Because supernatants from activated platelets, which contain higher amounts of ADP and ATP, also induced a strong chemotactic response in neutrophils, which was again diminished after statin treatment of platelets, we suggest that this phenomenon might contribute to the proposed inhibitory effects of statins on platelet-neutrophil interactions. PAR signaling pathways involve the activation of Rho-GTPases that relays signals to the cytoskeleton.39 There is evidence that statins may affect cells via interference with signaling pathway proteins such as nuclear laminin B, ras proto-oncogen, Rho-related proteins, and the
-subunit of heterodimeric GTP-binding proteins.15 Our findings confirm reports on Rho-GTPasedependent actions of statins in pro-inflammatory agents stimulated endothelial and smooth muscle cells. Inhibition of platelet function is also one such pathway.
In the neutrophil, expression of the adhesion proteins and the increased chemotactic response are enhanced by stimulation of P2Y receptors, which are coupled to Rho-GTPases.19 In vivo statins may also interfere with neutrophil-related pathophysiology via this mechanism.15 The inhibition of HMG-CoA reductase decreases the amount of membrane-associated small Rho-GTPases and substitution of geranylgeranyl-pyrophosphate, a product of the mevalonate pathway, and results in a redistribution of these GTPases to the membrane.40 Because statins affect Rho-GTPases by inhibiting isoprenylation,16 it is likely that both P2Y- and PAR-functions in platelets and neutrophils are altered after statin therapy. This hypothesis would be consistent with our observation that thrombin-activated platelets, when coincubated with the statins, decreased levels of Rho-GTPases and failed to increase oxidative burst activity of neutrophils. It may also explain the failure of CVN and SVN to reduce neutrophil respiratory burst in the absence of platelets.16,17 Data of the present study suggest the statins may protect vascular wall cells from excessive superoxide anion release via modification of neutrophil recruitment to the forming thrombus and interaction with platelets by downregulating membrane associated Rho-GTPases.
When platelets were stimulated with thrombin, there was an enhanced release of ADP and ATP. Extracellular ADP and ATP leads to occupation of purinergic receptors on several cells including neutrophils, triggering of mobilization of intracellular free calcium, which is thought to prime cells for enhanced oxidative burst activity.24 Because thrombin activates platelets and triggers their ADP/ATP release, it is most likely that increased generation of adenosine derivatives is responsible for enhanced oxygen radical production of neutrophils when coincubated with thrombin-treated platelets. In our investigations, this effect was diminished when platelets were treated with the statins. However, the autocrine action of ADP and ATP via P2 receptors on platelets leads in turn to higher ADP/ATP plasma levels, which recruit further platelets to form a thrombus and trigger the release of thrombin as a pro-inflammatory mediator.12,18,41 Because both ADP and ATP not only stimulate chemotactic and phagocytotic activity of neutrophils via P2Y-receptors, but also prime cells for enhanced oxygen radical production and degranulation,19 the statin-induced decrease in ADP and ATP release reflects controlled platelet activation. By this mechanism, as also possibly via inhibition of PAR and P2-receptor signaling through an inhibition of small Rho-GTPases, statin treatment may protect the vasculature from an excessive superoxide anion burden. However, there could be another possible pathway for the thrombin effect. Thrombin-activatable fibrinolysis inhibitor is a carboxypeptidase Blike enzyme generated by the complex thrombin/thrombomodulin. This enzyme inhibits fibrinolysis by removing C-terminal residues from partially degraded fibrin and thus decreasing plasminogen binding on the surface of fibrin which still may adhere to neutrophils. Even if this or similar proteolytic activities are weak or specific, they constitute a risk of thrombosis and may also be affected by statins. Effects regarding the effect of statins on this mechanism would be worth further investigation.
In conclusion, by inhibiting platelet activation through a diminished response to thrombin and by decreasing the generation of adenosine derivatives, therapeutic concentrations of statins may inhibit oxygen-free radical release and neutrophil-platelet interaction, which might be of relevance for neutrophil-dependent endothelial dysfunction, including ischemia/reperfusion, atherosclerosis, and acute coronary syndromes. We suggest that patients taking such statins may have a decreased respiratory burst activity of neutrophils and a lower risk for thrombus formation through decreased adenosine nucleotide generation of platelets in vivo. The two-pronged mechanism of action may render CVN and SVN, as two representative lipophilic statins, effective anti-atherosclerotic drugs.
| Acknowledgments |
|---|
Received February 24, 2002; accepted March 22, 2002.
| References |
|---|
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2002 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |