| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Institute for Prevention of Cardiovascular Diseases,
Department of Transfusion Medicine and Hemostaseology, University of Munich, Munich;
Department of Vascular Surgery, Klinikum Munich-Schwabing, Munich;
Rudolf-Virchow Center for Experimental Biomedicine, University of Würzburg, Würzburg;
|| Vascular Biology and Hemostasis, Institute of Clinical Chemistry, University of Munich, Munich, Germany; and
Department of Biochemistry and
Department of Hematology, University of Cambridge, Cambridge, UK
1Correspondence: Institut für Prophylaxe und Epidemiologie der Kreislaufkrankheiten, Klinikum Innenstadt, Universität München, Pettenkoferstr. 9, D 80336 München, Germany. E-mail: wsiess{at}med.uni-muenchen.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2ß1 inhibition with 6F1 mAb. Moreover, the initial platelet response, shape change, induced by plaque was blocked by GPVI inhibition but not with 2ß1 antagonists (6F1 mAb or GFOGER-GPP peptide). Pretreatment of plaques with collagenase or anti-collagen type I and anti-collagen type III antibodies abolished plaque-induced platelet activation. Our results indicate that morphologically diverse collagen type I- and collagen type III-containing structures in lipid-rich atherosclerotic plaques stimulate thrombus formation by activating platelet GPVI. This platelet collagen receptor, essential for plaque-induced thrombus formation, presents a promising new anti-thrombotic target for the prevention of ischemic cardiovascular diseases.—Penz, S., Reininger, A. J., Brandl, R., Goyal, P., Rabie, T., Bernlochner, I., Rother, E., Goetz, C., Engelmann, B., Smethurst, P. A., Ouwehand, W. H., Farndale, R., Nieswandt, B., Siess, W. Human atheromatous plaques stimulate thrombus formation by activating platelet glycoprotein VI.
Key Words: atherosclerotic plaques • ischemic stroke • thrombogenicity • arterial thrombus
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
. In particular, the atheromatous core of vulnerable plaques is much more thrombogenic than collagen-rich matrix isolated from stable sclerotic plaques (3
, 4)
. It is widely held that the main prothrombotic stimulus in plaques is tissue factor, which activates blood coagulation (5
6
7
8
9
10)
.
Atherosclerotic plaques consist mainly of extracellular lipid droplets and matrix constituents, foam cells, and necrotic cell debris (11
, 12)
. Within this material numerous platelet-activating molecules are found such as von Willebrand factor (vWf), fibrin/fibrinogen, thrombospondin, vitronectin, fibronectin, various types of collagens, SDF-1, oxidized LDL, and cholesterol sulfate (12
13
14
15
16
17)
. However, the ability of these molecules in mediating platelet activation and thrombus formation by human atherosclerotic plaques is not known.
We previously found that the lipid core and organic extracts of atheromatous plaques induced shape change of washed platelets that is mediated by lysophosphatidic acid accumulating in the plaque core (18
, 19)
. In the present study, whole human atheromatous plaques were exposed to washed platelets, platelet-rich plasma, and whole blood with the aim to identify the nature of the main platelet-activating material in soft, rupture-prone atherosclerotic plaques and to unravel the principal mechanism of plaque-induced arterial thrombus formation.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2ß1 blocking peptide GFOGER-GPP and the GPVI blocking peptide mCRP were synthesized as described before (20
, 21)
. The mAB against murine GPVI (JAQ1) was generated as previously specified (22)
. The scFv antibodies 10B12 (anti-GPVI blocking antibody) and 2D4 (anti-HLA type A2, control antibody) were described recently (23)
. Anti-collagen type I and anti-collagen type III antibodies were obtained from Rockland Immunochemicals (Gilbertsville, PA, USA). Alexa Fluor®488-labeled goat anti-rabbit IgG antibody and Alexa Fluor®546-phalloidin were obtained from Molecular Probes (Eugene, OR, USA). The mAB P1E6 against 2 was from Chemicon (Hampshire, UK). The mAB 6F1 against 2 was a generous gift from Prof. B. Coller (The Rockefeller University, New York, USA) (24)
. The vWf-GPIb blocking mAB 24G10 was a generous gift from Prof. Hans Deckmyn (Laboratory of Thrombosis Research, Leuven, Belgium) (25)
.The following monoclonal antibodies directly conjugated to fluorochromes were purchased from BD Biosciences (Heidelberg, Germany): phycoerythrin (PE) -conjugated anti-CD14 (M
P9, IgG2b), and fluorescin isothiocyanate (FITC) -conjugated anti-CD41a (HIP8, IgG1).
Isolation of human atheromatous plaques
Atherosclerotic tissue specimens were obtained from patients who underwent surgery for high-grade carotid artery stenosis. Patient consent was obtained as approved by the Ethics Committee of the Faculty of Medicine of the University of Munich. The carotid plaque tissue was removed by a technique of intraoperative endarterectomy that preserved the plaque structure en bloc (26)
. Specimens containing lipid-rich soft plaques were collected. The atheromatous plaques were carefully dissected from other regions of the atherosclerotic tissue specimens such as the collagen fiber-rich plaque shoulder and the proximal region containing connective tissue, foam cells, and small pools of lipid deposits (26)
. The plaques were characterized by histological analysis. In some atheromatous plaque specimens it was possible to macroscopically excise the lipid-rich core of toothpaste-like consistency (18)
. The plaques were weighed, homogenized with a glass pestle and potter, and stored at –80°C. The atheromatous plaque concentration was 50 mg wet weight/mL corresponding to protein concentrations of 0.5–2.5 mg/mL.
Platelet isolation, platelet shape change, aggregation, and secretion
Informed consent from blood donors was obtained in accordance with the Helsinki protocol. Platelet-rich plasma (PRP) was prepared from citrate (0.31% wt/vol) or hirudin (200U/mL) anticoagulated blood by centrifugation for 20 min at 180 x g. Washed platelets were isolated as described (27)
with the following modifications: after the first wash, platelets were resuspended in the second wash buffer (20 mM HEPES, 138 mM NaCl, 2.9 mM KCl, 1 mM MgCl2, 0.36 mM NaH2PO4, 5 mM glucose, 0.6 U/mL apyrase, pH 6.2), centrifuged, and resuspended in the same buffer but with a pH of 7.4. Platelet shape change and aggregation stimulated by collagen or atheromatous plaque were measured turbidimetrically using stirred PRP (1100 rpm) or platelet suspensions at 37°C in a LABOR® aggregometer (Fa. Fresenius, Bad Homburg, Germany) or lumi-aggregometer (Chronolog, Havertown, PA, USA) (19)
. Secretion of ATP from dense granules was measured in the lumi-aggregometer after addition of luciferin/luciferase. Platelet collagen receptors were inhibited by preincubating platelets with anti-2-integrin antibodies P1E6 or 6F1, the 2ß1 peptide ligand GFOGER-GPP, the monomeric collagen-related peptide that blocks GPVI (21
, 28)
, or anti-GPVI antibody 10B12.
Platelet aggregation and platelet-monocyte aggregate formation in whole blood
Samples of citrated blood were stirred in a LABOR® aggregometer at 1000 rpm at 37°C, collagen or atheromatous plaque were added, and aliquots were removed before and after agonist addition and placed in fixative containing formaldehyde (29)
. Whole blood platelet aggregation in citrated blood was measured as loss of single platelets using a platelet counter (Sysmex® Platelet Counter PL-100, Tao Medical Electronics, Japan). Platelet-monocyte conjugates were measured (as percentage of total) by flow cytometry after labeling platelets with an anti-CD41a-FITC antibody and monocytes with an anti-CD14-PE antibody (29)
. In some experiments, blood was preincubated with 10B12 (200 µg/mL), 2D4 (200 µg/mL), 6F1 mAb (20 µg/mL), or mouse IgG1 (20 µg/mL) for 15 min at 37°C before platelet-monocyte conjugate measurement.
Blood coagulation
Coagulation in citrated plasma, platelet-rich plasma, and blood (0.3 mL) was determined by thrombelastography (roTEG, Dynabyte, Munich, Germany) (30)
. Atheromatous plaque (1.65 mg/mL) or vehicle was added. The coagulation process was started by the addition of CaCl2 (16.6 mM).
Plaque collagen content and localization
Atheromatous plaque material was incubated with 0.5 M acetic acid (pH 3) for 24 h at 4°C with constant stirring for collagen extraction. After centrifugation at 13,000 rpm for 1 h at 4°C, collagen was determined in the supernatant with the Sircol collagen assay.
For immunohistochemical localization of collagen-positive structures in situ, the atheromatous plaques were fixed in 4% formaldehyde (pH 7.0) and embedded in paraffin as described (26)
. Serial paraffin sections (3 µm) were deparaffinized and dehydrated. The specimens were microwaved in citrate buffer (pH 6.0) for 20 min. Endogenous peroxidase was blocked by 3% H2O2, followed by incubation with anti-collagen type I or type III antibodies or the same concentrations of rabbit IgG (control) antibody. Immunostaining used the streptavidin-horseradish peroxidase technique (DakoChem Mate Detection kit). Specific binding of the antibodies was detected by DAB (0.1% 3'3'-diaminobenzidine).
Platelet adhesion and immunofluorescence microscopy
Coverslips were coated with 0.1 mL collagen (10 µg/mL) or atheromatous plaque (5 mg/mL) dissolved in PBS containing 15 mM fatty acid-free albumin and incubated for 16 h at 4°C in PBS-albumin (75 mM) to prevent nonspecific binding. Coverslips were rinsed carefully with PBS, then incubated for 1 h at room temperature with anti-collagen type I or type III antibodies (1:20 in PBS) or the same concentrations of rabbit IgG (control) and washed with PBS. Secondary antibody Alexa-Fluor®488 goat anti-rabbit antibody (1:200 in PBS) was added for 45 min at RT. Coverslips were rinsed again with PBS; 0.25 mL of a 4 x 105/µL washed platelet suspension was added to each and incubated for 30 min at 37°C. Coverslips were washed with PBS and fixed in 0.16% formaldehyde in PHEM buffer for 10 min at RT. After a PBS wash, the platelets were permeabilized using 0.2% Triton X-100 in PBS for 10 min at RT. F-actin of platelets was stained with rhodamine-phalloidin (50-fold dilution). Fluorescence microscopy was performed using a Zeiss confocal fluorescence microscope (LSM510 META) and software.
Specific binding of recombinant 2 I-domain and GPVI to plaque
Coating of Immulon-2 96-well plates with plaque material for recombinant protein adhesion was performed as described above using 50 µL of diluted plaque per well. Coating of Immulon-2 96-well plates with GFOGER or CRP has been described (31)
. Binding of proteins to plaques was performed as described for recombinant GPVI and 2 I-domain to collagen (23
, 31)
.
Collagenase digestion of collagen and atheromatous plaque material
Collagen(20 µg) or 0.02 mL atheromatous plaque were incubated with 0.6 µg collagenase in 50 µL TESCA buffer (50 mM TES, 0.36 mM CaCl2) containing additional 9 mM CaCl2 for 24 h at 37°C. Control incubations were with BSA (1 µg) instead of collagenase.
Analysis of platelet adhesion and thrombus formation in flowing whole blood
Atheromatous plaque immobilized onto glass coverslips and fluorescently stained with anti-collagen antibodies was inserted into a parallel plate flow chamber. Perfusion at a wall shear rate of 1500 s–1 was obtained by aspiration of the blood through the chamber via a syringe pump (Harvard Apparatus, Holliston, MA. USA). The perfusion chamber was mounted on the stage of an upright microscope (Axioscope 2 plus, Carl Zeiss, Germany) for real-time visualization of platelet adhesion and aggregation in flowing blood. The platelets required no labeling, but were observed via bright field (BF), fluorescence, and reflection interference contrast microscopy (RICM) (32)
. The latter allows us to visualize the interface between the membranes of adherent and spreading platelets and immobilized substrates on the glass coverslips. Combination of BF and fluorescence microscopy allowed us to localize platelet adhesion and aggregation with respect to the collagen type I- and III-positive structures. Platelet deposition was recorded continuously on S-VHS videotape at an acquisition rate of 25 frames/s using a Pulnix CCD camera connected to a VCR (Panasonic NV-HS930, Japan). For inhibition of platelet GPVI or integrin 2ß1, blood was preincubated for 15 min at 37°C with 50 µg/mL 10B12 and 2D4 (control) or 20 µg/mL 6F1 and mouse IgG1 (control). Platelets were labeled by addition of 10 µM mepacrine before start of flow. Digitizing and analyzing images was performed off-line using the Matrox Inspector software package (Matrox Electronix Systems Ltd., Canada). During the last 2 min of each 10 min flow experiment, the superfused area was scanned and 15 images (visual fields) of the surface were obtained (40x objective, zoom factor 0.4). In each field the areas of plaques and of respective aggregates attached to them were measured and related as percentage "plaque covered with platelet aggregates." Single platelet adhesion was easily discernible and was not included in the measurements. The video clips (see attached CD) were prepared with Adobe Premiere (Adobe Systems, San Jose, CA, USA).
For mice experiments, heparinized whole blood (20U/mL) from wild-type, FcR
–/–, or GPVI-depleted mice was perfused over plaque core-coated coverslips. To induce GPVI depletion, mice were injected with JAQ1 (100 µg/mouse i.p.) and blood was analyzed on day 5 postinjection (33)
. Perfusion was carried out at room temperature using a pulse-free pump at high shear stress (4 min, flow rate of 7.53 mL/h, equivalent to a wall shear rate of 1000 s–1). At the end of the experiment, phase-contrast images were recorded from at least five different microscopic fields (63x objectives) and analyzed offline using Metamorph software (Visitron, Munich, Germany).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We previously observed that the lipid core and the lipid fraction of human atheromatous plaques induced rapid shape change in washed platelets without subsequent aggregation (Fig. 1
a, left tracing) (18
, 19)
. This could be attributed to lysophosphatidic acid accumulated in the plaque core (18
, 19)
. However, when washed platelets were exposed to the whole atheromatous plaque (AP), a different platelet response was observed: after a lag, shape change occurred, followed by aggregation (Fig. 1a
, right tracing). This aggregation response was dose dependent and, in plaques from most of the patients, reached a maximum at a concentration of 1–2.5 mg wet weight/mL plaque.
|
Atheromatous plaque also induced platelet shape change, aggregation, and dense granule secretion in PRP (Fig. 1b
). The plaque-induced platelet response showed kinetics similar to that after collagen stimulation of PRP. Plaques from most of the patients (56 of 60) induced maximal or submaximal aggregation of PRP (82±39%, mean±SD, n=56; Fig. 1c
). However, atheromatous plaques from four patients (corresponding to 7% of plaques studied) were inactive. They induced neither aggregation nor secretion in PRP or washed platelet suspensions (Fig. 1b
and data not shown).
The atheromatous plaques were also active when exposed to whole anticoagulated blood. They induced dose-dependently irreversible aggregation with similar kinetics as collagen (Fig. 1d
, left panel).
Plaques stimulate platelet-dependent clot formation
Atheromatous plaque elicited platelet-monocyte aggregation in whole blood in a dose-dependent manner (Fig. 1d
, right panel). Since platelet-monocyte interaction leads to the activation of intravascular tissue factor and initiation of blood coagulation, we asked whether atheromatous plaque might promote platelet-dependent fibrin generation and clot formation. We found that 35 of 37 plaques stimulated coagulation, and 15 of these required the presence of platelets to induce clot formation. The rate of fibrin generation in PRP induced by these plaques was 953 ± 278 s (mean±SD), and absent in PPP (>1500 s; Fig. 1e
). The other 20 plaques stimulated coagulation in recalcified PPP (rate of fibrin generation: 680±20 s; mean±SD) due to the presence of tissue factor in the plaques (data not shown). The rate of fibrin generation induced by these plaques was shorter in PRP (452±175 s, mean±SD; P<0.0001), indicating acceleration of coagulation by the platelet-stimulating effect of the plaques. These results show that the plaque-induced coagulation either depended on or was enhanced by the activation of platelets.
Plaques contain morphologically diverse collagenous structures that stimulate directly platelet adhesion and aggregation
The similar kinetics of plaque- and collagen-induced platelet responses prompted us to ask whether this activation is mediated through collagen receptors. Indeed, when the two main platelet collagen receptors, integrin 2ß1 and glycoprotein (GP) VI (34
, 35)
, were blocked, plaque-induced aggregation and dense granule secretion of human platelets were inhibited (Fig. 2
a). Plaque-induced shape change and aggregation of mouse platelets were abrogated in the presence of the blocking anti-mouse GPVI antibody JAQ1 (Fig. 2b
) (22)
. These results indicate that atheromatous plaques stimulate platelets through activation of collagen receptors, particularly GPVI, and suggest that collagen-like material was responsible for the plaque-induced platelet aggregation.
|
To explore whether atheromatous plaque contains collagen, immunostaining of atheromatous plaque sections with specific collagen antibodies revealed the presence of collagen type I- and type III-positive structures in the cap as well as the core of plaques in situ (Fig. 3
a). Plaques coated to glass coverslips and stained with anti-collagen antibodies showed collagen type I- and type III-positive structures exhibiting bright fluorescence by confocal immunofluorescence microscopy (Fig. 3b
). By comparison with collagen fibers, the collagenous material found in the plaques was morphologically very heterogeneous in size and shape. Mostly they did not show a regular fibrillar structure, but rather appeared in the form of small or as large fragments showing interconnected patches or a spongiform shape. Double staining often indicated a colocalization of collagen type I and type III structures (Fig. 3b
).
|
When human platelets were allowed to adhere to atheromatous plaque material under static conditions, they attached to both small and large collagen-containing structures. The platelets changed shape and spread on the collagen type I- or type III-positive structures, indicating they were activated (Fig. 3c
, left and data not shown). Similarly, platelets adhered to and spread on collagen fibers. Platelet spreading, but not shape change and attachment to the collagenous structures, was inhibited after preincubating platelets with RGDS, which blocks the integrin IIbß3 (data not shown). After inhibition of the integrin 2ß1 and GPVI, platelet adhesion to plaques and collagen was abolished. Rarely, single platelets could be observed that were discoid, i.e., nonactivated (Fig. 3c
, right). These results show that platelets adhered directly to the collagenous plaque components via their collagen receptors, not to other plaque components.
Since collagenous structures in the plaques were found to be responsible for the induction of platelet adhesion and aggregation, we assessed whether the amount of collagen determines the platelet reactivity of plaques. We found, however, no correlation between collagen content and the ability of atheromatous plaques to activate platelets (Fig. 1b
, Fig. 4
). We further observed no difference in the quantity or morphology of the collagen-containing structures of active and inactive plaques as judged by confocal immunofluorescence microscopy (data not shown).
|
Collagen type I- and type III-containing structures are essential for platelet activation by plaques
To explore whether only collagenous plaque constituents were responsible for platelet activation, additional experiments were performed. First, degradation of the collagenous material by pretreatment of plaques with collagenase abolished plaque-induced aggregation of PRP (Fig. 5
a). Preincubation of plaques with anti-collagen type I and anti-collagen type III antibodies inhibited plaque-induced shape change and aggregation of PRP (Fig. 5b
). Each anti-collagen antibody alone inhibited only partly plaque-induced aggregation, even at much higher concentrations. Rabbit IgG added as control was inactive (data not shown). To completely block plaque-induced aggregation, preincubation of plaques with anti-collagen type I and anti-collagen type III antibodies was necessary. These data indicate that collagen type I- and collagen type III-containing structures are both essential for plaque-induced platelet activation.
|
The collagenous plaque constituents initiate platelet activation
We performed additional experiments to explore whether other adhesive proteins in atherosclerotic plaques such as vWf, fibrinogen/fibrin, fibronectin, and vitronectin might activate platelets. Plaque vWf or other adhesive proteins bound to the collagenous structures might initiate platelet activation by binding to GPIb and/or the integrin IIbß3 and induce platelet signaling possibly leading to collagen receptor activation and secondary binding of platelets to the collagenous material.
Platelets were therefore preincubated with a monoclonal anti-GPIb antibody (24G10, 5 µg/mL for 3 min), which blocks the binding of vWf to GPIb (25)
. Plaque-induced aggregation was 62 ± 15% in control and 60 ± 24% in PRP preincubated with 24G10 (mean±SD; n=6 plaques), which abolished ristocetin-induced aggregation. Blockade of the integrin IIbß3 with RGDS, known to bind many adhesive proteins (fibrin, fibrinogen, fibronectin, vitronectin), including vWf, inhibited plaque and collagen-induced platelet aggregation, but not shape change (Fig. 6
). Shape change was almost completely blocked by the GPVI antibody 10B12. These results indicate that collagenous structures and not other adhesive proteins of plaques initiate platelet activation.
|
Relative roles of integrin 2ß1 and GPVI in the initial phase of plaque-induced platelet activation
In the experiments shown in Fig. 3c
, blockade of the integrin 2ß1 and GPVI inhibited completely platelet adhesion to collagen or collagenous plaque constituents. To explore the contribution of each of the two collagen receptors for plaque-induced platelet activation in more detail, binding studies with the recombinant collagen receptors were first carried out. Mg-dependent specific binding of recombinant 2 I-domain to 6 plaques was 0.212 ± 0.03 (mean±SE) compared with binding to GFOGER (=1.00), the integrin 2ß1 peptide ligand. The 10B12-inhibitable (25 µg/mL) binding of recombinant GPVI ectodomain to the same 6 plaques was 0.3317 ± 0.036 (mean±SE) compared with its binding to CRP (=1.00), the GPVI ligand. Therefore, plaques are able to specifically bind both platelet collagen receptors.
To examine the role of the two collagen receptors in plaque-induced platelet shape change, the initial platelet response, studies with 2ß1 antagonists and GPVI antibody were performed in the presence of RGDS to prevent fibrinogen-mediated platelet aggregation. Preincubation of platelets with 6F1 mAb or GFOGER-GPP peptide slightly delayed and reduced collagen-induced shape change of hirudin-anticoagulated PRP but did not inhibit plaque-induced shape change (Fig. 6)
. Plaque-induced shape change in the presence of 6F1 mAb was 98 ± 7% of control (mean±SD, n=3) and not different than in the presence of mouse IgG1 (Fig. 6b
). 10B12 reduced collagen-induced shape change and almost completely blocked plaque-induced shape change (9±12% of control, mean±SD, n=3; Fig. 6
), indicating a central role of GPVI in initiating plaque-induced platelet activation.
Studies with 10B12 were carried out in the absence of RGDS to allow plaque-induced aggregation. Plaque-induced platelet aggregation was completely inhibited by 10B12 (data not shown). These results demonstrate that the collagenous plaque constituents trigger mainly via binding to GPVI platelet signaling, leading to shape change and subsequent platelet aggregation.
Plaques trigger thrombus formation in anticoagulated blood under arterial flow conditions by activating platelet GPVI in mice and humans
To explore whether plaques stimulate platelet adhesion and aggregation under conditions resembling those in vivo, whole human blood anticoagulated with recombinant hirudin was perfused at arterial wall shear rates (1500 s–1) over immobilized plaque material. Bright field and fluorescence microscopy (Fl) demonstrated single platelet attachment at the downstream margin of collagen-like structures of the atheromatous plaque (see arrowheads in Fig. 7
a and supplemental video on http://www.fasebj.org). Within the first 2 min after start of flow, single platelets adhered transiently to these structures. After 3 min, platelet adhesion became stable and more platelets were recruited. Subsequently, platelet aggregates were formed and platelet thrombi grew rapidly. RICM showed areas of spread platelets at the bottom of the large aggregates after 5 min but not 3 min (dark gray area, arrows), indicating that the spread platelets firmly anchored the aggregates on the substrate. Both the small and large collagen-positive structures were able to activate platelets in flowing blood. The confocal immunofluorescence micrograph taken after perfusion demonstrated that platelets had changed their shape and had spread and aggregated on the collagenous structures of the plaque (Fig. 7a
, IF).
|
To define the role of the two collagen receptors in plaque-induced platelet adhesion and thrombus formation under flow conditions, we used blood from integrin 2-deficient mice (36)
or mice in which GPVI had been blocked or depleted by using the GPVI antibody JAQ1 (33)
. At high shear flow, platelets from wild-type and 2–/– mice rapidly adhered and within 4 min formed large aggregates (Fig. 7b
and data not shown), whereas inhibition of GPVI on wild-type platelets with JAQ1 blocked adhesion and thrombus formation. Consistent with this, platelets from GPVI-depleted mice failed to adhere to and form thrombi on plaque core; the same result was obtained with blood from FcR–/– mice, which lack GPVI (Fig. 7b
) (22)
. These results demonstrate that mouse platelet adhesion and thrombus formation on human atheromatous plaque strictly depend on GPVI.
To examine the role of the two collagen receptors in plaque-induced thrombus formation of human blood under flow conditions, studies with antibodies against 2ß1 (6F1) and GPVI (10B12) were performed. Videomicroscopy of mepacrine-stained platelets demonstrated that 10B12 and 6F1 allowed platelet adhesion. However, in the presence of 10B12 adhesion was only transient, with arrested platelets rapidly detaching, whereas with 6F1 platelet adhesion was permanent as in control. Moreover, 10B12 completely blocked platelet aggregation but 6F1 was not inhibitory (Fig. 8
). The control antibodies 2D4 and mouse IgG1 were without effect on platelet adhesion and aggregate formation. These results show that human thrombus formation on atheromatous plaque is strictly dependent on GPVI.
|
Platelet-monocyte aggregate formation in blood stimulated by plaque (69±9%) was inhibited by 10B12 (21±14%) but not 6F1 (63±16%, mean±SD, n=4), showing a crucial role of GPVI in mediating plaque-induced attachment of platelets to monocytes.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
6
7
8
9
10)
. This implies that coagulation is activated first and that platelets are only stimulated in a second step through thrombin generated by the coagulation system. In this scenario, anticoagulation appears to be the adequate anti-thrombotic strategy. In contrast, our study demonstrates that atheromatous plaques directly activate platelets triggering thrombus formation under arterial flow conditions. Plaque-induced thrombus formation was observed in blood maximally anticoagulated with the thrombin-inhibitor hirudin. These results indicate that plaque-induced platelet thrombus formation occurred in the absence of coagulation evoked by plaque-derived or blood-borne tissue factor. Furthermore, anticoagulation would not be the adequate strategy to prevent and treat arterial thrombosis after plaque rupture in patients. We observed plaque-induced platelet activation in 93% of the 60 patients studied with lipid-rich atheromatous plaques. This type of plaque is known to be vulnerable and prone to rupture. Of the many platelet-activating molecules present in atherosclerotic plaques, we identified morphologically diverse collagenous type I and type III structures in the plaques as the material responsible for triggering platelet shape change and platelet adhesion as well as stimulating subsequent platelet spreading, aggregation, and secretion. Platelet activation was not only observed in washed platelets, but also in plasma and whole blood. Enzymatic degradation of plaque collagenous material, inhibition of plaque collagen type I and III, as well as blockade of platelet collagen receptors abolished plaque-induced platelet shape change, adhesion, and aggregation. These results clearly establish that collagen receptor activation by plaque collagen type I and III molecules triggers and sustains plaque-induced platelet activation. We show that plaques stimulated platelet-monocyte aggregate formation and indirectly fostered blood coagulation through the activation of platelets. Finally, the collagenous structures present in the plaque were able to induce thrombus formation in blood under arterial flow conditions
Why is the collagen-rich matrix of sclerotic plaques, the media of normal arteries that are packed with collagen fibers, and the subendothelium of normal arteries much less thrombogenic than the collagenous material in atheromatous plaques (3
, 5
, 37
, 38)
? We found that the quantity of collagen in the atheromatous plaque did not determine platelet reactivity. Plaques from some patients were inactive, although their collagen content was not decreased. Since we observed a lack of the typical fibrillar structure of collagen in the plaques, we hypothesize that degradation of the collagen matrix might increase its capacity to interact with and to activate platelets. Collagen molecules in plaques are known to be proteolytically cleaved by collagenases, predominantly by specific matrix metalloproteinases overexpressed in inflamed atherosclerotic lesions and by lysosomal enzymes released from activated monocytes (10
, 14
, 39)
. These enzymes might cleave collagens to small diffuse cross-linked fragments, which, although in amorphous polymeric state, retain the capacity to activate GPVI. This collagenous material may be analogous to cross-linked synthetic collagen-related peptide, which, in contrast to monomeric collagen-related peptide, potently activates platelets, mainly through activation of GPVI (28)
. Such an interpretation is supported by inhibition of plaque-induced platelet activation with the antibody JAQ1, which inhibits the binding site of collagen-related peptide on GPVI (40)
. A possible explanation for the lack of plaque-induced platelet activation observed in some patients might then be a decreased content of cross-linked collagen fragments.
Our results demonstrate that GPVI is more important than integrin 2ß1 in triggering platelet activation and inducing thrombus formation under flow onto the atheromatous plaques. Integrin 2ß1 antagonists delayed and reduced shape change by fibrous collagen (Fig. 6a
) and, in a recent study, inhibited single platelet adhesion to collagen fibers and reduced subsequent Ca2+ signaling, PS exposure and aggregate size (41)
. In contrast, plaque-induced shape change as well as permanent platelet adhesion and thrombus formation onto the plaque was not affected by the 2ß1 antagonists, but blocked by the GPVI antibody. Moreover, mice platelets lacking the collagen receptor GPVI or showing an impaired GPVI signaling were unable to adhere to atheromatous plaque or form thrombi. Since we found that active plaques specifically bound recombinant 2 I-domain and GPVI, we conclude that platelet GPVI ligation by the plaque collagenous components induces more efficient platelet signaling than integrin 2ß1 ligation.
In conclusion, our results suggest that platelet activation is the initial and central event leading to atherothrombosis after rupture of atheromatous plaques. Specifically, we identified collagen type I- and collagen type III-containing plaque constituents and platelet GPVI as essential for mediating plaque-induced thrombus formation. Binding of these collagenous plaque components to GPVI induces intracellular signaling leading to the initial phase of platelet activation (i.e., platelet shape change and adhesion) and, subsequently, via fibrinogen receptor activation and secretion of positive feedback mediators (ADP, thromboxane), to platelet spreading, aggregation, and thrombus formation (Fig. 9
). Our study suggests a novel anti-thrombotic strategy to prevent atherothrombosis in patients at cardiovascular risk. It has recently been described that platelet adhesion and aggregation on the injured vessel wall in mice could be blocked by the infusion of a soluble dimeric form of GPVI. There was no major effect on bleeding time from this treatment (42)
. Future in vivo experimental studies are needed to clarify possible beneficial effects of GPVI inhibition in preventing atherothrombosis after plaque rupture.
|
|
ACKNOWLEDGMENTS |
|---|
Received for publication August 14, 2004. Accepted for publication February 18, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. J. Reininger, I. Bernlochner, S. M. Penz, C. Ravanat, P. Smethurst, R. W. Farndale, C. Gachet, R. Brandl, and W. Siess A 2-Step Mechanism of Arterial Thrombus Formation Induced by Human Atherosclerotic Plaques. J. Am. Coll. Cardiol., March 16, 2010; 55(11): 1147 - 1158. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. I. Jones, S. Bray, S. F. Garner, J. Stephens, B. de Bono, W. G. J. Angenent, D. Bentley, P. Burns, A. Coffey, P. Deloukas, et al. A functional genomics approach reveals novel quantitative trait loci associated with platelet signaling pathways Blood, August 13, 2009; 114(7): 1405 - 1416. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Consoli, E. Martelli, M. D'Adamo, R. Menghini, D. Arcelli, O. Porzio, A. Pandolfi, G. R. Pistolese, A. Consoli, R. Lauro, et al. Insulin Resistance Affects Gene Expression in Endothelium Arterioscler Thromb Vasc Biol, February 1, 2008; 28(2): e7 - e9. [Full Text] [PDF]
|
|
|
|
 |
|
 E. Calvo, F. Tokumasu, O. Marinotti, J.-L. Villeval, J. M. C. Ribeiro, and I. M. B. Francischetti Aegyptin, a Novel Mosquito Salivary Gland Protein, Specifically Binds to Collagen and Prevents Its Interaction with Platelet Glycoprotein VI, Integrin {alpha}2beta1, and von Willebrand Factor J. Biol. Chem., September 14, 2007; 282(37): 26928 - 26938. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Rabie, D. Varga-Szabo, M. Bender, R. Pozgaj, F. Lanza, T. Saito, S. P. Watson, and B. Nieswandt Diverging signaling events control the pathway of GPVI down-regulation in vivo Blood, July 15, 2007; 110(2): 529 - 535. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. M. Ruggeri and G. L. Mendolicchio Adhesion Mechanisms in Platelet Function Circ. Res., June 22, 2007; 100(12): 1673 - 1685. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Grothusen, S. Umbreen, I. Konrad, K. Stellos, C. Schulz, B. Schmidt, E. Kremmer, O. Teebken, S. Massberg, M. Luchtefeld, et al. EXP3179 Inhibits Collagen-Dependent Platelet Activation via Glycoprotein Receptor-VI Independent of AT1-Receptor Antagonism: Potential Impact on Atherothrombosis Arterioscler Thromb Vasc Biol, May 1, 2007; 27(5): 1184 - 1190. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. J.H. Sachs and B. Nieswandt In Vivo Thrombus Formation in Murine Models Circ. Res., April 13, 2007; 100(7): 979 - 991. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Gross, P. Tilly, D. Hentsch, J.-L. Vonesch, and J.-E. Fabre Vascular wall-produced prostaglandin E2 exacerbates arterial thrombosis and atherothrombosis through platelet EP3 receptors J. Exp. Med., February 19, 2007; 204(2): 311 - 320. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. N. O'Connor, P. A. Smethurst, L. W. Davies, L. Joutsi-Korhonen, D. J. Onley, A. B. Herr, R. W. Farndale, and W. H. Ouwehand Selective Blockade of Glycoprotein VI Clustering on Collagen Helices J. Biol. Chem., November 3, 2006; 281(44): 33505 - 33510. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Bigalke, S. Lindemann, R. Ehlers, P. Seizer, K. Daub, H. Langer, T. Schonberger, E. Kremmer, D. Siegel-Axel, A. E. May, et al. Expression of platelet collagen receptor glycoprotein VI is associated with acute coronary syndrome Eur. Heart J., September 2, 2006; 27(18): 2165 - 2169. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. L. Kahn GPVI: no magic bullet for thrombosis Blood, June 1, 2006; 107(11): 4199 - 4199. [Full Text] [PDF]
|
|
|
|
|
|
A. J. Reininger, H. F. G. Heijnen, H. Schumann, H. M. Specht, W. Schramm, and Z. M. Ruggeri Mechanism of platelet adhesion to von Willebrand factor and microparticle formation under high shear stress Blood, May 1, 2006; 107(9): 3537 - 3545. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
10 nm), which are present at t = 5 min but not at 3 min. By changing the focal plane, formation of large aggregates was observed above the areas of spread platelets (Fl and BF) after 5 min. Bar = 40 µm. The experiment is representative of 11 experiments with 5 different atheromatous plaques. IF, confocal fluorescence micrographs of platelets adhering to plaque core at the end of the perfusion period. Green, plaque material stained with anti-collagen type III antibody. Red, platelets stained intracellularly for F-actin by rhodamine-coupled phalloidin. Bar: 10 µm. A video showing the formation of platelet thrombus onto plaque core in real time: supplemental video on http://www.fasebj.org. b) Heparinized whole blood (20U/mL) from wild-type mice ± JAQ1 (100 µg/mL), from GPVI-depleted mice, or from FcR
