| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Thrombosis |
2 Signaling in Arterial Thrombosis as a Function of Lesion Severity
From INSERM U.311, Etablissement Français du Sang-Alsace, Strasbourg, France.
Correspondence François Lanza or Christian Gachet, INSERM U.311, Etablissement Français du Sang-Alsace, 10 rue Spielmann, BP 36, 67065 Strasbourg Cedex, France. E-mail francois.lanza{at}efs-alsace.fr or christian.gachet@efs-alsace.fr
| Abstract |
|---|
|
|
|---|
2ß1 integrin) acting upstream of phospholipase C (PLC)
2. However, PLCß transduces signals from G
q protein-coupled receptors for soluble agonists (P2y1, TxA2/TP, and thrombin/PAR). A Gi-dependent pathway amplifies most of these responses.
Methods and Results To evaluate the role of adhesion receptors signaling in arterial thrombosis, PLC
2 knockout mice were studied in blood perfusion assays over fibrillar collagen and in a laser-induced mesenteric artery model of thrombosis. In vitro, PLC
2-deficient platelets formed a single layer incapable of generating a thrombus on collagen, whereas G
q-deficient platelets formed reduced size aggregates compared with wild-type cells. In the in vivo model, PLC
2/ mice displayed defective thrombus formation in superficial lesions but productive thrombosis after a more severe laser injury. In contrast, resistance to thrombosis was observed in G
q/ mice in both levels of injury.
Conclusions These results demonstrate that signaling through PLC
2 plays an important role in arterial thrombosis, but that its contribution depends on the severity of the vascular lesion.
This study evaluated the role of adhesion receptors signaling through PLC
2 in arterial thrombosis. PLC
2-deficient mice showed resistance to thrombus formation in superficial lesions but productive thrombosis after a more severe laser injury formation. In contrast, resistance to thrombosis was observed in G
q/ mice in both levels of injury.
Key Words: adhesion mouse model phospholipase C platelet thrombosis
| Introduction |
|---|
|
|
|---|
are the 2 main isotypes involved in PI hydrolysis in platelets. PLCß transduces signals from G
q protein-coupled receptors for soluble agonists such as P2Y1 (ADP), TP
(TxA2), PAR (
-thrombin), and 5-HT2A (serotonin).3 A major role of the G
q/PLCß pathway in platelet activation has been clearly demonstrated in studies using G
q-deficient mouse platelets, which were unresponsive to ADP, TxA2, or thrombin.3,4 PLC
lies downstream of adhesion receptors such as integrins
2ß1 and
IIbß3, the glycoprotein (platelet glycoprotein [GP]) Ib-V-IX complex, and GPVI, the major collagen activation receptor.5 A major role of the PLC
2 isotype in the response to adhesion receptors has been demonstrated in studies of PLC
2-deficient platelets. Impaired activation has been observed to von Willebrand factor through its GPIb-V-IX receptor, to collagen, and to GPVI- and
2ß1 integrin-specific ligands,68 and after integrin
IIbß3 outside-in signaling.9,10 A third pathway triggered by ADP/P2Y12 acting through Gi-PI3K-Rap1b also plays a major role in platelet activation because it amplifies most of the responses triggered by PLC.11,12 Despite demonstration of a role of the 2 main PLC pathways in vitro, their relative contribution to arterial thrombosis is less well-characterized.
G
q/ mice are protected against platelet-dependent thromboembolism,3 and perfusion of G
q/ blood over collagen results in formation of smaller aggregates.13 However, resistance to arterial thrombosis of G
q/ mice has not yet been clearly documented. Studies of receptors converging on G
q/PLCß nevertheless support an important contribution of this pathway to arterial thrombosis. In particular, mice lacking the P2Y1 ADP receptor or lacking either of the major thrombin receptors, PAR3 or PAR4, demonstrated resistance to arterial thrombosis.1416
Unlike its well-studied functions in intracellular signaling and platelet activation in vitro, the involvement of PLC
2 in arterial thrombosis is largely unexplored. An important contribution to thrombus formation has been suggested from perfusion of PLC
2/ blood over collagen matrices, which demonstrated a decreased surface coverage by platelets apparently incapable of forming thrombi.7 Although in vivo thrombosis in PLC
2/ mice has not yet been reported, a role of this isotype may be postulated from studies of mice depleted of receptors acting upstream of PLC
2.1719 Among these, GPVI linked to the Fc receptor (FcR)
-chain is a strong promoter of PLC
2 activation.20 Mice genetically deficient in GPVI were generated recently, and studies in a collagen flow assay showed normal platelet adhesion but no thrombus formation.21 In vivo thrombosis evaluated in a GPVI-immune depleted model showed prevention or decreased thrombus formation in carotid arteries.18 These results suggested that blockade of the GPVI/FcR
/PLC
2 pathway could represent an interesting antithrombotic strategy. PLC
2 has also been implicated in GPIb-V-IX and
2ß1-dependent activation.6,7 However, the role of GPIb-V-IXdependent signaling is difficult to assess in the absence of a suitable animal model, and there is controversy concerning arterial thrombosis protection in
2ß1-deficient mice.17,22
An understanding of the role of the 2 PLC pathways that channel signals from adhesive proteins and soluble agonists, respectively, and their relative contribution to arterial thrombosis would be of importance in view of future drug development. Therefore, the aim of this work was to evaluate the contribution of the PLC
2 signaling pathway to thrombus formation, in relation with the role played by PLCß-dependent signaling. In the absence of PLCß-deficient mice, a direct insight into its role can be gained from studies of G
q-deficient animals. PLC
2-deficient and G
q-deficient mice were examined using an in vitro model of high shear blood perfusion over a collagen-coated surface and an in vivo thrombosis model with laser-induced mesenteric arteriole injury.
| Materials and Methods |
|---|
|
|
|---|
Mouse Strains
G
q/, G
q+/+, PLC
2/, and PLC
2+/+ mice were provided by Professor Stefan Offermanns (Universität Heidelberg, Germany)3 and Professor J. Ihle (St. Jude Childrens Research Hospital, Memphis, Tenn),23 respectively. For detailed methods, please see http://atvb.ahajournals.org.
In Vitro Model of Platelet Adhesion to Immobilized Collagen in a Flow System
Hirudin anticoagulated blood from wild-type and knockout mice was perfused through collagen-coated capillaries and platelet thrombus formation was visualized under a fluorescence microscope. Analysis of surface coverage was performed off-line from numeric photographs of the capillary surface. Thrombus volumes were determined using a fluorescence microscope equipped with a confocal scanner and SlideBook software (Intelligent Imaging Innovations, Denver, Colo).
In Vivo Model of Thrombosis
Localized injury of the luminal surface of a mesenteric arteriole was induced with a pulsed nitrogen dye laser (440 nm) applied through the objective of an inverted Leica DMIRB microscope with a Micropoint system (Photonics Instruments, St Charles, Ill).24,25 Thrombus formation was analyzed by wide-field and fluorescent light microscopy using a charge-coupled device camera and a SlideBook software.
Statistical Analyses
Statistical analyses of differences in thrombus formation between 2 mouse strains or after treatment were performed by analyzing mean thrombus volume (perfusion over collagen) or area under curve (arterial thrombosis) and a nonparametric MannWhitney test. P<0.05 was considered to be significant. All tests were performed using PrismM (GraphPad Software, San Diego, Calif).
| Results |
|---|
|
|
|---|
2/ Blood and Reduced Thrombus Formation in G
q/ Blood During Perfusion Over a Collagen Surface at High Shear
2- and G
q-dependent signaling pathways for thrombus formation was first studied in vitro using a flow assay. Whole blood anticoagulated with the direct thrombin inhibitor hirudin was perfused under arterial flow conditions (3000 s1) over a fibrillar collagen matrix. Under these conditions, platelets from wild-type mice initially adhered as single cells and progressively formed aggregates, which increased rapidly in size along the collagen fibers (Figure 1a). After 90 seconds of perfusion,
30% of the total surface was covered by platelets, predominantly (80%) in the form of large thrombi (Figure 1b, black bars). Perfusion of G
q/ blood showed that individual platelets adhered with kinetics comparable to those of platelets in matched G
q+/+ blood (Figure 1a). Thrombus growth during the 2 minutes of perfusion was nevertheless abnormal, with an increased number of smaller aggregates. However, image analysis revealed only a slight decrease in the percentage of the surface covered by platelet aggregates (90.7%), which was comparable to the wild-type (Figure 1b). This minimal defect in surface coverage prompted us to investigate the thrombus volume. After 90 seconds of perfusion, the surface was analyzed by confocal scanning microscopy with 3-µm sections and the platelet thrombi were reconstructed using computer-assisted image analysis software. The thrombus volume was significantly decreased in G
q/ blood, reaching only 41% of that in wild-type blood, largely because of a reduced thrombus height (Figure 1c).
|
A more important defect was observed in perfusion studies of PLC
2-deficient blood. Although PLC
2/ platelets adhered normally during the first 20 seconds of perfusion, the first layer of adherent platelets was virtually incapable of capturing additional platelets from flowing blood or of forming aggregates (Figure 1a). After 2 minutes of perfusion, single platelets represented the majority of elements (85.6%) bound to the collagen surface, unlike in wild-type or G
q/ blood (Figure 1b). This lack of thrombus formation was observed equally well at high (3000 s1) as it did at moderate (1500 s1) shear rates (Figure I, available online at http://atvb.ahajournals.org). In the absence of aggregates, the surface continuously captured individual platelets and the number of adherent cells steadily increased during the 2 minutes of perfusion (Figure I). Confocal microscopy 3-dimensional analysis confirmed that PLC
2-deficient platelets adhered as a single layer incapable of forming aggregates (Figure 1c).
Characterization of a Laser-Induced Mesenteric Arterial Thrombosis Model of Increasing Severity
In vitro flow studies on collagen surfaces provide a tool to evaluate thrombus formation and dissect the role of individual partners, but conversely they do not reproduce the complexity of the blood vessel. Experiments were therefore performed in an in vivo model of laser-induced localized thrombosis in mesenteric arterioles, which was first established in control C57BL/6J mice. Reproducible superficial and severe injuries were induced by adjusting the laser intensity and number of pulses as described in the Methods section.
In superficial injuries, as already described,25 thrombus formation evolved biphasically, with platelets quickly accumulating at the site of endothelial desquamation to form a parietal thrombus peaking at 53 seconds. This was followed by progressive erosion, resulting in an 82% decrease in thrombus surface area after 2 minutes (Figure 2a). In severe injuries, a parietal thrombus formed progressively during the first 90 seconds and reached a size 12-times larger than in superficial lesions, leading to near occlusion of the vessel lumen (Figure 2b). Contrary to the thrombi in the less severe injury, the thrombus size did not significantly decrease during the next 3 minutes.
|
Both models were evaluated for their responses to known antiplatelet drugs and thrombin blockade. Thrombus formation was abolished in the superficial lesion by injection of the GPIIb-IIIa antagonist eptifibatide and was profoundly decreased in mice treated with the P2Y12 ADP-receptor inhibitor clopidogrel, but it was not affected by treatment with hirudin or with aspirin (Figure 2a). In the severe lesion, thrombus formation was greatly reduced by injection of eptifibatide (80% reduction in size at 2 minutes) and by treatment with clopidogrel (65% reduction in size at 2 minutes), and was not affected by aspirin treatment (Figure 2b). However, and contrary to the superficial lesion, thrombus formation was severely decreased after treatment with hirudin, indicating thrombin formation.
Therefore, the 2 models respond to the more potent antiplatelet agents, a GPIIb-IIIa blocker and clopidogrel, are insensitive to aspirin and differ in their response to thrombin blockade.
PLC
2/ Mice Display a Decreased Tendency to Thrombosis in Superficial Lesions
A severe impairment of thrombus formation was observed in PLC
2/ mice after superficial laser-induced lesion of the mesenteric arteries. Maximum thrombus surface area represented only 35% of that in matched PLC
2+/+ mice, and after 2 minutes almost no residual platelets could be visualized on the damaged area, unlike in PLC
2+/+ animals (Figure 3a). A different picture emerged in the more severe injury model, in which thrombosis developed actively in PLC
2/ mice. The initial thrombus growth was comparable to that in PLC
2+/+ mice, and the maximal thrombus size and stability were not significantly decreased (Figure 3b). These results indicate that PLC
2-dependent platelet signaling is needed for parietal arterial thrombus formation in a mild lesion of the vessel wall, but that its deficiency cannot prevent thrombosis in a more severe lesion exposing deeper subendothelial structures.
|
G
q/ Mice Are Protected Against Thrombosis in a Model of Severe Arterial Injury
Defective arterial thrombus formation was also observed in G
q/ mice, but the defect differed from that of PLC
2/ animals in that it was also observed in a severe lesion. In a superficial injury, the thrombus peak was diminished by 52% as compared with G
q+/+ mice (Figure 4a). Stepwise decreases were noticed in the descending portion of the curve, indicating a tendency to embolization of platelet aggregates. Unlike in PLC
2/ animals, a residual layer of platelets remained at later times. The thrombus instability in G
q/ blood was even more apparent in the severe injury model (Figure 4b). In these lesions, although the thrombus initially developed normally during the first minute, this was followed by a steep decrease in thrombus size because of stepwise detachment of large platelet emboli. This behavior, clearly different from that observed in PLC
2/ mice (Figure 3b), indicates that the G
q-dependent signaling pathway plays a major role in thrombus growth and stability but is less essential for initial thrombus formation.
|
| Discussion |
|---|
|
|
|---|
2.2,3,9,26 The present study established that signaling through PLC
2 and G
q/PLCß plays an important role in arterial thrombosis, but that the contributions of these 2 pathways differ depending on the severity of the vascular lesion.
In vitro perfusion of blood over a fibrillar collagen surface under high (3000 s1) or intermediate (1500 s1) shear conditions revealed defective thrombus formation in the PLC
2- and G
q-deficient mice. In these experiments, the initial capture of single platelets, in which the GPIb-V-IX complex plays a major role, was not affected by the absence of PLC
2 or G
q, confirming that these adhesive events occur independently of PLC-dependent inside-out signaling.9,10 The ensuing aggregate formation and thrombus extension were, however, completely prevented in PLC
2 deficiency, indicating that the signals originating from adhesion receptors were absolutely required to initiate thrombus growth in this flow system. A comparable defect in collagen flow experiments has been reported for mice deficient in GPVI or FcR
-chain,7,13,21 or after blockade of GPVI in human platelets.27 This is in agreement with the major involvement of GPVI/FcR-
in collagen-induced activation and its critical requirement for the PLC
2 pathway.26 Previous flow studies of PLC
2-deficient blood performed at lower shear rates (800 s1) showed a decreased platelet coverage of the collagen surface but provided no information on the single or aggregated nature of the adherent platelets.7
Although an analysis of the platelet surface coverage pointed to efficient aggregate formation in G
q-deficient blood, confocal scanning analyses revealed a significant decrease in the total thrombus volume with aggregates individually of smaller size than in the wild-type, suggesting a defect in thrombus growth and/or stability. Because our assay was performed in the presence of hirudin, G
q-dependent inside-out activation would be expected to depend mostly on ADP and TxA2. In G
q deficiency, Gi and G13 signaling remain normal3 and thus provide alternate routes for aggregate formation. Clopidogrel treatment completely abolished aggregation in G
q/ perfusion experiments, thereby supporting a major role of the P2Y12/Gi pathway (data not shown).
The dramatic decrease in thrombus development in PLC
2-deficient blood in the perfusion system suggested an essential role of this pathway in initiating thrombus formation with a more modest involvement of G
q/PLCß. However, the results generated in vivo led to reconsider the relative contribution of the two PLC pathways. In fact, protection against thrombosis was observed in both PLC
2/ and G
q/ mice but varied depending on the degree of vessel wall injury.
In a mild injury, the results were in accordance with the collagen flow assay with an important role of the PLC
2 pathway (Figure 3a). Defective arterial thrombosis indicated the importance of platelet responses to von Willebrand factor and collagen in this model. Existence of a residual thrombus suggested, however, the existence of alternate routes of activation possibly through another PLC
isotype, such as PLC
1,7 or caused by traces of soluble agonists such as ADP.11 Blockade of this residual response by clopidogrel treatment (data not shown) indicated its requirement for ADP. Indication that thrombus formation in the superficial lesion depends on combined signaling through adhesion receptors and ADP/P2Y12 is further supported by its incomplete blockade in G
q/ mice (Figure 4a).
Resistance to thrombosis was no longer observed in PLC
2/ mice after a more severe injury, suggesting an inverse relationship between the importance of PLC
2 and severity of the arterial lesion. This hypothesis was further supported by an increased defect after a milder lesion (23% decrease in laser intensity) in the superficial injury model (Figure II, available online at http://atvb.ahajournals.org). One likely explanation for the lack of protection against thrombosis in PLC
2/ mice after a severe injury is its strong dependency to thrombin as shown by its sensitivity to hirudin. Adhesion functions being preserved in PLC
2/ (Figure 1), thrombin generated locally, and subsequent ADP release would then be sufficient for full platelet activation and thrombus growth. Therefore, PLC
2-dependent activation triggered by adhesive receptors seems to play a minor role in thrombus extension in a thrombin-dependent arterial injury. The origin of thrombin in the severe injury model has not been formally identified but could result from tissue factor (TF) exposure or, maybe less likely, from turbulent flow conditions. Analysis of TF-deficient (TF+/) or low-TF mice28 or in situ detection of TF by real-time imaging24 could be used to address this question.
Blockade of the GPVI/FcR
-chain receptor for collagen was recently proposed as a promising antithrombotic strategy on the grounds of decreased collagen-induced thrombus formation in human studies27 and of reduced ex vivo and in vivo thrombosis in GPVI immune-depleted and FcR
-chaindeficient mice.18 In fact, the responses in different flow studies ranged from severely defective thrombus formation to modest consequences.21,29 The present finding of normal thrombosis in PLC
2-deficient mice in a severe arterial lesion raises questions as to the usefulness of targeting this pathway and the upstream adhesive receptors.
A role of PLC
2-dependent signaling in superficial injuries could still have some relevance to the clinical situation. Both superficial lesions and rupture of atherosclerotic plaques are considered to be important in precipitating arterial thrombus formation in patients with coronary disease.30,31 In the former situation, in which the role of thrombin might be less predominant, the underlying matrix and its richness in collagen fibers could play a more important role in thrombus formation
Resistance to thrombosis was observed in both types of injuries in G
q/ mice, confirming the essential role of ADP- and thrombin-dependent signaling for arterial thrombosis.11,16 Decreased thrombosis in the superficial lesion in G
q/ mice is most probably caused by lack of P2Y1 signaling because this model is not affected by aspirin and hirudin treatments. Contrary to PLC
2/, resistance to thrombosis was also observed in the more severe lesion in G
q/ and was in the form of a decreased stability of the forming thrombus.
The different responses of PLC
2-deficient mice in superficial and severe lesions are a warning against conclusions as to a lack or presence of protection against thrombosis solely on the basis of a single model. The possibility of misinterpretation was recently illustrated by differences in terms of embolization and vessel occlusion, in mice with a deletion in the fibrinogen-
chain, according to whether analyses were performed in FeCl3-injured mesenteric arterioles or carotid arteries.3234 An advantage of the laser-induced model used in the present study over the FeCl3 models is the possibility of covering a controlled range of lesions in a limited area of the arterial wall. In a model of FeCl3-induced injury of carotid arteries similar to that used by Jirouskova et al,32 we only observed a mild thrombosis defect in PLC
2-deficient mice with a normal time to first occlusion (Figure III, available online at http://atvb.ahajournals.org). In the same model, thrombosis was severely affected in G
q-deficient mice or in mice treated with hirudin. These results are in line with those in the severe laser-induced lesion and further question the importance of adhesion receptor-dependent signaling in acute arterial thrombosis.
| Acknowledgments |
|---|
Received January 19, 2005; accepted March 3, 2005.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. Leon, A. Eckly, B. Hechler, B. Aleil, M. Freund, C. Ravanat, M. Jourdain, C. Nonne, J. Weber, R. Tiedt, et al. Megakaryocyte-restricted MYH9 inactivation dramatically affects hemostasis while preserving platelet aggregation and secretion Blood, November 1, 2007; 110(9): 3183 - 3191. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. J. Westrick, M. E. Winn, and D. T. Eitzman Murine Models of Vascular Thrombosis Arterioscler. Thromb. Vasc. Biol., October 1, 2007; 27(10): 2079 - 2093. [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] |
||||
![]() |
C. V. Denis and D. D. Wagner Platelet Adhesion Receptors and Their Ligands in Mouse Models of Thrombosis Arterioscler. Thromb. Vasc. Biol., April 1, 2007; 27(4): 728 - 739. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. C. Hughan, C. E. Hughes, O. J.T. McCarty, E. Schweighoffer, I. Soultanova, J. Ware, V. L.J. Tybulewicz, and S. P. Watson GPVI Potentiation of Platelet Activation by Thrombin and Adhesion Molecules Independent of Src Kinases and Syk Arterioscler. Thromb. Vasc. Biol., February 1, 2007; 27(2): 422 - 429. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. R. Vandendries, J. R. Hamilton, S. R. Coughlin, B. Furie, and B. C. Furie Par4 is required for platelet thrombus propagation but not fibrin generation in a mouse model of thrombosis PNAS, January 2, 2007; 104(1): 288 - 292. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Strassel, C. Nonne, A. Eckly, T. David, C. Leon, M. Freund, J.-P. Cazenave, C. Gachet, and F. Lanza Decreased Thrombotic Tendency in Mouse Models of the Bernard-Soulier Syndrome Arterioscler. Thromb. Vasc. Biol., January 1, 2007; 27(1): 241 - 247. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Mangin, C. L. Yap, C. Nonne, S. A. Sturgeon, I. Goncalves, Y. Yuan, S. M. Schoenwaelder, C. E. Wright, F. Lanza, and S. P. Jackson Thrombin overcomes the thrombosis defect associated with platelet GPVI/FcR{gamma} deficiency Blood, June 1, 2006; 107(11): 4346 - 4353. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Dubois, L. Panicot-Dubois, G. Merrill-Skoloff, B. Furie, and B. C. Furie Glycoprotein VI-dependent and -independent pathways of thrombus formation in vivo Blood, May 15, 2006; 107(10): 3902 - 3906. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Hechler, C. Nonne, E. J. Roh, M. Cattaneo, J.-P. Cazenave, F. Lanza, K. A. Jacobson, and C. Gachet MRS2500 [2-Iodo-N6-methyl-(N)-methanocarba-2'-deoxyadenosine-3',5'-bisphosphate], a Potent, Selective, and Stable Antagonist of the Platelet P2Y1 Receptor with Strong Antithrombotic Activity in Mice J. Pharmacol. Exp. Ther., February 1, 2006; 316(2): 556 - 563. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2005 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |