Laser-induced Vascular Injury Research Articles

4D intravital imaging studies identify platelets as the predominant cellular procoagulant surface in a mouse hemostasis model

AbstractInterplay between platelets, coagulation factors, endothelial cells (ECs), and fibrinolytic factors is necessary for effective hemostatic plug formation. This study describes a 4-dimensional (4D) imaging platform to visualize and quantify hemostatic plug components in mice with high spatiotemporal resolution. Fibrin accumulation after laser-induced vascular injury was observed at the platelet plug–EC interface, controlled by the antagonistic balance between fibrin generation and breakdown. We observed less fibrin accumulation in mice expressing low levels of tissue factor or F12−/−mice compared with controls, whereas increased fibrin accumulation, including on the vasculature adjacent to the platelet plug, was observed in plasminogen-deficient mice or wild-type mice treated with tranexamic acid. Phosphatidylserine (PS), a membrane lipid critical for the assembly of coagulation factors, was first detected at the platelet plug–EC interface, followed by exposure across the endothelium. Impaired PS exposure resulted in a significant reduction in fibrin accumulation in cyclophilin D−/−mice. Adoptive transfer studies demonstrated a key role for PS exposure on platelets, and to a lesser degree on ECs, in fibrin accumulation during hemostatic plug formation. Together, these studies suggest that (1) platelets are the functionally dominant procoagulant cellular surface, and (2) plasmin is critical for limiting fibrin accumulation at the site of a forming hemostatic plug.

Molecular docking-assisted screening reveals tannic acid as a natural protein disulphide isomerase inhibitor with antiplatelet and antithrombotic activities.

Protein disulphide isomerase (PDI) promotes platelet activation and constitutes a novel antithrombotic target. In this study, we reported that a PDI‐binding plant polyphenol, tannic acid (TA), inhibits PDI activity, platelet activation and thrombus formation. Molecular docking using plant polyphenols from dietary sources with cardiovascular benefits revealed TA as the most potent binding molecule with PDI active centre. Surface plasmon resonance demonstrated that TA bound PDI with high affinity. Using Di‐eosin‐glutathione disulphide fluorescence assay and PDI assay kit, we showed that TA inhibited PDI activity. In isolated platelets, TA inhibited platelet aggregation stimulated by either GPVI or ITAM pathway agonists. Flow cytometry showed that TA inhibited thrombin‐ or CRP‐stimulated platelet activation, as reflected by reduced granule secretion and integrin activation. TA also reduced platelet spreading on immobilized fibrinogen and platelet adhesion under flow conditions. In a laser‐induced vascular injury mouse model, intraperitoneal injection of TA significantly decreased the size of cremaster arteriole thrombi. No prolongation of mouse jugular vein and tail‐bleeding time was observed after TA administration. Therefore, we identified TA from natural polyphenols as a novel inhibitor of PDI function. TA inhibits platelet activation and thrombus formation, suggesting it as a potential antithrombotic agent.

Increased levels of the megakaryocyte and platelet expressed cysteine proteases stefin A and cystatin A prevent thrombosis

Increased platelet activity occurs in type 2 diabetes mellitus (T2DM) and such platelet dysregulation likely originates from altered megakaryopoiesis. We initiated identification of dysregulated pathways in megakaryocytes in the setting of T2DM. We evaluated through transcriptomic analysis, differential gene expressions in megakaryocytes from leptin receptor-deficient mice (db/db), exhibiting features of human T2DM, and control mice (db/+). Functional gene analysis revealed an upregulation of transcripts related to calcium signaling, coagulation cascade and platelet receptors in diabetic mouse megakaryocytes. We also evidenced an upregulation (7- to 9.7-fold) of genes encoding stefin A (StfA), the human ortholog of Cystatin A (CSTA), inhibitor of cathepsin B, H and L. StfA/CSTA was present in megakaryocytes and platelets and its expression increased during obesity and diabetes in rats and humans. StfA/CSTA was primarily localized at platelet membranes and granules and was released upon agonist stimulation and clot formation through a metalloprotease-dependent mechanism. StfA/CSTA did not affect platelet aggregation, but reduced platelet accumulation on immobilized collagen from flowing whole blood (1200 s−1). In-vivo, upon laser-induced vascular injury, platelet recruitment and thrombus formation were markedly reduced in StfA1-overexpressing mice without affecting bleeding time. The presence of CA-074Me, a cathepsin B specific inhibitor significantly reduced thrombus formation in-vitro and in-vivo in human and mouse, respectively. Our study identifies StfA/CSTA as a key contributor of platelet-dependent thrombus formation in both rodents and humans.

A long-acting PAI-1 inhibitor reduces thrombus formation.

Plasminogen activator inhibitor 1 (PAI-1) is the main inhibitor of tissue-type and urokinase-type plasminogen activators (t/uPA) and plays an important role in fibrinolysis. Inhibition of PAI-1 activity prevents thrombosis and accelerates fibrinolysis, indicating that PAI-1 inhibitors may be used as effective antithrombotic agents. We previously designed a PAI-1 inhibitor (PAItrap) which is a variant of inactivated urokinase protease domain. In the present study, we fused PAItrap with human serum albumin (HSA) to develop a long-acting PAI-1 inhibitor. Unfortunately, the fusion protein PAItrap-HSA lost some potency compared to PAItrap (33 nM vs 10 nM). Guided by computational method, we carried out further optimisation to enhance inhibitory potency for PAI-1. The new PAItrap, denominated PAItrap(H37R)-HSA, which was the H37R variant of PAItrap fused to HSA, gave a six-fold improvement of IC50 (5 nM) for human active PAI-1 compared to PAItrap-HSA, and showed much longer plasma half-life (200-fold) compared to PAItrap. We further demonstrated that the PAItrap(H37R)-HSA inhibited exogenous or endogenous PAI-1 to promote fibrinolysis in fibrin-clot lysis assay. PAItrap(H37R)-HSA inhibits murine PAI-1 with IC50 value of 12 nM, allowing the inhibitor to be evaluated in murine models. Using an intravital microscopy, we demonstrated that PAItrap(H37R)-HSA blocks thrombus formation and platelet accumulation in vivo in a laser-induced vascular injury mouse model. Additionally, mouse tail bleeding assay showed that PAItrap(H37R)-HSA did not affect the global haemostasis. These results suggest that PAItrap(H37R)-HSA have the potential benefit to prevent thrombosis and accelerates fibrinolysis.

Defective release of α granule and lysosome contents from platelets in mouse Hermansky-Pudlak syndrome models

AbstractHermansky-Pudlak syndrome (HPS) is characterized by oculocutaneous albinism, bleeding diathesis, and other variable symptoms. The bleeding diathesis has been attributed to δ storage pool deficiency, reflecting the malformation of platelet dense granules. Here, we analyzed agonist-stimulated secretion from other storage granules in platelets from mouse HPS models that lack adaptor protein (AP)-3 or biogenesis of lysosome-related organelles complex (BLOC)-3 or BLOC-1. We show that α granule secretion elicited by low agonist doses is impaired in all 3 HPS models. High agonist doses or supplemental adenosine 5′-diphosphate (ADP) restored normal α granule secretion, suggesting that the impairment is secondary to absent dense granule content release. Intravital microscopy following laser-induced vascular injury showed that defective hemostatic thrombus formation in HPS mice largely reflected reduced total platelet accumulation and affirmed a reduced area of α granule secretion. Agonist-induced lysosome secretion ex vivo was also impaired in all 3 HPS models but was incompletely rescued by high agonist doses or excess ADP. Our results imply that (1) AP-3, BLOC-1, and BLOC-3 facilitate protein sorting to lysosomes to support ultimate secretion; (2) impaired secretion of α granules in HPS, and to some degree of lysosomes, is secondary to impaired dense granule secretion; and (3) diminished α granule and lysosome secretion might contribute to pathology in HPS.

ML359, a Small Molecule Inhibitor of Protein Disulfide Isomerase That Prevents Thrombus Formation and Inhibits Oxidoreductase but Not Transnitrosylase Activity

Vascular thiol isomerases comprise a family of enzymes including protein disulfide isomerase (PDI), ERp5, and ERp57 that are important in the process of thrombus formation. PDI is secreted at sites of vascular injury, and antibody-mediated PDI inhibition prevents thrombus formation in a mouse laser injury model. Our group has previously reported on the discovery of the small molecule PDI inhibitors quercetin-3-rutinoside and ML359. Identified as part of a high-throughput screen, ML359 is a second-generation PDI inhibitor that selectively blocks PDI oxidoreductase activity with approximately ten-fold the potency of quercetin-3-rutinoside. To better understand the mechanism of allosteric modulation of PDI by small molecules, we evaluated the association of ML359 with isolated domains of PDI, determined the effects of ML359 on a variety of PDI functions, and compared the activity of ML359 to that of quercetin-3-rutinoside. PDI is composed of four thioredoxin-like domains and an x-linker region in the sequence a-b-b’-x-a’. Major substrate binding is thought to occur in the b-b’ region while the a and a’ domains contain catalytically active cysteine motifs (CGHC) that mediate the oxidoreducase, nitrosylase, and thiol isomerase functions of PDI. In order to identify potential binding sites of ML359 on PDI, we constructed and expressed the domain fragments a, ab, abb’, b’xa’, and a’. These fragments were tested in the presence of 10 µM ML359 using an insulin turbidometric assay that measures the oxidoreductase activity of PDI. ML359 demonstrated full inhibition of oxidoreductase activity when full-length PDI and the b’xa’ fragment were used whereas no inhibition was observed with the other fragments assayed. These results are consistent with docking studies showing that ML359 likely binds in a pocket formed at the b’x interface. In contrast, when the same experiment was performed in the presence of 30 µM of quercetin-3-rutinoside, inhibition was only noted with full-length PDI and the abb’ and b’xa’ fragments, suggesting that binding was dependent on the b’ and not the x-linker region. To determine if ML359 has differential effects on the oxidoreductase and nitrosylase functions of PDI, we utilized a platelet-based assay in which fluorescence intensity stemming from the NO-sensitive intracellular dye DAF-FM was measured as an indicator of PDI-mediated translocation of NO from the extracellular surface into the cytosol (transnitrosylation). While quercetin-3-rutinoside potently inhibited PDI-mediated transnitrosylation activity, ML359 had no effect. These results are consistent with the idea that the transnitrosylase and oxidoreducase functions of PDI are separable and inhibition of either is specific to the small molecule used. We evaluated the ability of ML359 to inhibit thrombosis in a mouse laser injury model. Intravital microscopy was used to follow thrombus formation in mouse cremaster arterioles after laser-induced vascular injury. Infusion of ML359 resulted in inhibition of thrombus formation, in contrast to thrombosis seen after infusion of vehicle alone. In summary, ML359 is a second generation small molecule inhibitor of PDI that likely binds at the b’x interface of the enzyme. Furthermore, ML359 is able to selectively inhibit PDI oxidoreductase activity without affecting transnitrosylase activity. ML359 may prove a useful molecular probe to better understand the different functions of PDI relative to thrombus formation in vivo. DisclosuresNo relevant conflicts of interest to declare.

Regulation of Protein Disulfide Isomerase By S-Nitrosylation Controls Its Function during Thrombus Formation

Protein disulfide isomerase (PDI) is an oxidoreductase that is essential for thrombus formation following vascular injury. Clinical trials testing the efficacy and safety of PDI inhibition in the setting of thrombotic disease are currently underway. Yet while preclinical and clinical trials of PDI in thrombosis have progressed rapidly, the mechanisms by which PDI is regulated in the vasculature and how it mediates thrombosis remain unknown. PDI has an a-b-b’-x-a’ domain structure, where the a and a’ domains contain a CGHC motif responsible for cleaving and forming disulfide bonds. The active site cysteines within the catalytic CGHC motif that perform oxidoreductive reactions can also undergo S-nitrosylation. We have evaluated the hypothesis that nitric oxide (NO) converts PDI into a nitrosylase and regulates PDI oxidoreductase activity in the vasculature during thrombus formation. Initial studies demonstrated that incubation of recombinant PDI with the NO donor, SNAP, resulted in an 83±1.4% decrease in its reductase activity. A transnitrosylase assay using the NO indicator DAF-FM showed that S-nitrosylated PDI (SNO-PDI) transferred NO into platelets and inhibited platelet aggregation. To define the molecular determinants of PDI nitrosylation activity, we evaluated mutant PDIs containing Cys -> Ala mutations of the CGHC (a domain)/CGHC (a’ domain) motifs in the platelet-based transnitrosylase assay. Wild-type PDI (CGHC/CGHC) demonstrated full reductase and nitrosylase activity and the enzymatically dead mutant (AGHA/AGHA) showed neither activity. In contrast, the CGHA/CGHA mutant maintained nitrosylase activity (41±0.23%), but had no reductase activity. This observation suggested that reductase and nitrosylase activities were separable. To further evaluate this supposition, we screened a series of PDI mutants in which intervening sequences of the CGHC domain had been modified. The screen identified CGPC/CGPC as a nitrosylase-biased mutant that showed a 59±2.31% decrease in reductase activity, but a 72±1.83% increase in nitrosylase activity compared to wild-type PDI. Another nitrosylase-biased mutant, CGRC/CGRC, showed a similar activity pattern. Since PDI is prothrombotic and SNO-PDI is antithrombotic, we compared the activity of nitrosylase-biased mutants with wild-type PDI in platelet aggregation studies in the presence of physiological concentrations of GSNO. While wild-type PDI had little effect on platelet aggregation, nitrosylase-biased PDIs such as the CGPC/CGPC and CGRC/CGRC mutant completely inhibited platelet aggregation. These studies show that the prothrombotic oxidoreductase activities of PDI are separable from their antithrombotic nitrosylase activities and that nitrosylase-biased PDI mutants have antiplatelet activity. We next evaluated the effect of PDI nitrosylation on thrombus formation in vivo. Infusion of SNO-PDI into mice inhibited thrombus formation following laser-induced vascular injury of cremaster arterioles. Mice deficient in glutathione-S-nitrosyl reductase (GSNOR) were used to assess the role of endogenous NO in thrombus formation. GSNOR enzymatically reduces GSNO, the main storage form of NO in cells. Platelet accumulation and fibrin formation were hardly detectable in GSNOR-/- mice. Infusion of recombinant WT PDI, but not an enzymatically dead PDI, reversed the defect in platelet accumulation and fibrin generation to levels of WT mice. In order to visualize NO during thrombus formation, the NO-sensitive dye DAF-FM was infused into mice and NO signal in endothelium monitored following laser-induced injury. DAF-FM signal decreased rapidly following laser injury of cremaster arterioles, indicating an activation-induced reduction in endothelial NO in vivo. In conclusion, our studies show that oxidoreductase and nitrosylase activities of PDI are separable and support a model whereby high endothelial NO levels maintain vascular quiescence in part by maintaining PDI as a nitrosylase and blocking its prothrombotic PDI activity. We propose that the reduction of NO levels that occurs with vascular injury or endothelial dysfunction contributes to the conversion of PDI from an anti-thrombotic nitrosylase to a prothrombotic reductase. DisclosuresNo relevant conflicts of interest to declare.

Abstract 282: Functional Analysis Of Micrornas In Vascular Disease

microRNAs (miRNA, miR) are emerging as pivotal modulators of vascular development and disease. Research is my lab is focused on elucidating the functional mechanism and exploring therapeutic potential of microRNAs in vascular diseases. Our recent studies have shown that miR-23 and miR-27 in miR-23~27~24 family are required for proper angiogenesis and neovascularization in a laser-induced vascular injury model. Here we extend our study and provide evidence that miR-24 regulates actin dynamics in ECs through targeting multiple members downstream of Rho signaling, including Pak4, Limk2 and Diaph1 proteins. Consistent with the critical role for actin cytoskeleton in cell motility and proliferation, overexpression of miR-24 in ECs blocks stress fiber and lamellipodia formation, represses EC migration, proliferation and tube formation in vitro, as well as angiogenesis in an ex vivo aortic ring assay. Overexpression of miR-24 in transgenic mice represses postnatal retinal vascular development. Moreover, subretinal delivery of miR-24 mimics represses laser-induced CNV in vivo. Mechanistically, knockdown of miR-24 target protein LIMK2 or PAK4 inhibits stress fiber formation and tube formation in vitro, mimicking miR-24 overexpression phenotype in angiogenesis. Taken together, these findings demonstrate that miR-24 represses angiogenesis by simultaneously regulating multiple components in the actin cytoskeleton pathways, suggesting distinct function of miR-23~27~24 family members in angiogenesis. Manipulation of actin cytoskeleton pathways by miR-24 may represent an attractive therapeutic solution for numerous vascular diseases.

Sutton's law and anti-β2GPI antibodies

In this issue of Blood, Arad and colleagues demonstrate that infusion of affinity― purified anti-β 2 ―glycoprotein I (P 2 GPI) antibodies from 3 patients with antiphospholipid sydrome (APS) into mice enhances the development of arterial thrombosis after laser-induced vascular injury. 1 This interesting observation directly demonstrates that human anti-β 2 GPI antibodies are thrombogenic.

Platelets generated from human embryonic stem cells are functional in vitro and in the microcirculation of living mice

Platelets play an essential role in hemostasis and atherothrombosis. Owing to their short storage time, there is constant demand for this life-saving blood component. In this study, we report that it is feasible to generate functional megakaryocytes and platelets from human embryonic stem cells (hESCs) on a large scale. Differential-interference contrast and electron microscopy analyses showed that ultrastructural and morphological features of hESC-derived platelets were indistinguishable from those of normal blood platelets. In functional assays, hESC-derived platelets responded to thrombin stimulation, formed microaggregates, and facilitated clot formation/retraction in vitro. Live cell microscopy demonstrated that hESC-platelets formed lamellipodia and filopodia in response to thrombin activation, and tethered to each other as observed in normal blood. Using real-time intravital imaging with high-speed video microscopy, we have also shown that hESC-derived platelets contribute to developing thrombi at sites of laser-induced vascular injury in mice, providing the first evidence for in vivo functionality of hESC-derived platelets. These results represent an important step toward generating an unlimited supply of platelets for transfusion. Since platelets contain no genetic material, they are ideal candidates for early clinical translation involving human pluripotent stem cells.

Identification of a Novel Par1 inhibitor Using a Chemical Genetic Screen

AbstractAbstract 2018The anucleate platelet is not amenable to traditional forward genetic analysis, but is a good candidate for a chemical genetic screen. In collaboration with the Broad Institute Probe Development Center, we have performed a high throughput screen using a modified luciferin-luciferase detection system to identify inhibitors of Par1-mediated dense granule secretion. For primary screening, over 300,000 compounds were assayed in duplicate using freshly outdated platelet-rich plasma supplied by several blood banks across the United States. Computational analyses of the primary data demonstrated that approximately 0.2% of compounds showed ≥50% inhibition relative to maximally inhibitory concentrations of the known antiplatelet agent, cilostazol. Secondary screening using 8-point dose response curves and counterscreening to exclude luciferase inhibitors identified 137 small molecules that inhibited Par1-mediated ATP/ADP release without significant inhibition of luciferase. Twenty eight compounds were selected for further testing based on their IC50s in confirmatory assays (<10 μM), lack of activity in unrelated bioassays, and chemical structure. Known platelet inhibitors were excluded. Of the 28 compounds, 16 compounds potently inhibited SFLLRN-induced α-granule release from washed platelets, as monitored by P-selectin expression. IC50s for these compounds ranged from <0.3 to 1 μM. None of the compounds that failed to inhibit α-granule release demonstrated significant inhibition of SFLLRN-induced 14C-serotonin release at 10 μM. Of the 16 confirmed inhibitory compounds, 12 demonstrated >50% stimulation of cAMP levels compared to platelets exposed to PGE1 and were not further analyzed. One of the 4 remaining compounds lost activity upon resynthesis. The three compounds that demonstrated activity upon resynthesis included a diaminophenyl, an acylhydrazone, and a cyanopyridone. A series of chemical analogs for each of the three series was synthesized and/or obtained from commercial vendors. Structure activity relationships were determined by testing these compounds in the luciferin-luciferase assay. The most potent analog of each series was then further analyzed. The representative compounds (10 μM) failed to block platelet granule release induced by the Par4 activating peptide AYPGFK, PMA, calcium ionophore, or collagen, suggesting that the compounds act at or near Par1. However, only one of the three compounds, termed DAP, significantly inhibited thrombin-induced platelet activation. Determination of DAP activity in cultured cells expressing Par1 confirmed DAP-mediated inhibition of Par1 activity. Analysis of dose-response curves at multiple inhibitor concentrations were consistent with non-competitive inhibition of Par1-mediated granule secretion. In functional studies, DAP inhibited platelet aggregation and secretion, but not platelet shape change. In contrast, the orthosteric Par1 inhibitor Sch79797 blocked aggregation, secretion, and shape change. Despite its inability to inhibit activation through human Par4, DAP blocked activation through mouse Par4, demonstrating complete inhibition of AYPGFK-mediated aggregation of mouse platelets at 5 μM. Platelets obtained from mice 45 minutes after infusion of 10 mg/kg DAP failed to aggregate in response to AYPGFK. The effect of DAP on thrombus formation was evaluated using intravital microscopy. These studies showed that infusion of 10 mg/kg DAP potently inhibited platelet accumulation following laser-induced vascular injury of cremaster arterioles in mice. These studies demonstrate the feasibility of using blood-banked platelets in a large-scale chemical genetic screen. Compounds that block platelet function via elevation of cAMP are common, indicating a need to include cAMP assays early in the screening funnel. The novel antithrombotic DAP identified in this screen appears to block Par1 via an allosteric mechanism. The ability of DAP to inhibit mouse Par4, but not human Par4, indicates that DAP acts outside the ligand binding pocket of Par1. This compound will provide a useful probe to study the effect of small molecule modulation of Par1 on receptor conformation and downstream signaling. Disclosures:No relevant conflicts of interest to declare.

Conservation of Hemostatic System Component Function Between Zebrafish and Mammals.

Abstract 3165Poster Board III-102Extensive DNA sequence similarity has previously been demonstrated for most known components of the mammalian hemostatic system and their orthologous zebrafish genes, and vascular occlusion has been observed in zebrafish following laser-induced vascular injury (Jagadeeswaran P, et al. Methods Mol Med 2006).To test potential conservation of function across species, activated bovine thrombin (Sigma-Aldrich), over a range of concentrations was injected into the orbital vascular plexus of zebrafish embryos at 72 hours post fertilization (hpf). The severity of thrombus formation was relative to the thrombin concentration, with thrombosis consistently observed in 72 hpf wild type (WT) embryos at 30-90 ng of thrombin (1-3 nl of a 30mg/ml solution diluted in 0.9% NaCl, with 0.25% phenol red as a tracer). With injection of higher doses (90 ng), complete occlusion of the circulation was generally observed within 30 seconds, with a large thrombus visible in the heart. Results were scored as zero when no thrombosis was observed, grade 1 when thrombosis was seen in the vessels but blood flow was preserved, and grade 2 for total occlusion of the circulation. Thirty of 30 WT embryos with grade 1 thrombosis showed resolution of the thrombus by 5 minutes post injection, consistent with the activity of an endogenous fibrinolytic system in zebrafish, compared to only 22 of 30 among WT embryos scored as grade 2. No thrombosis was observed with injection of BSA (N=30) or thrombin inactivated by incubation at room temperature for 24 hours (N=30). One cell stage zebrafish embryos were injected with antisense morpholino oligonucleotides (MOs) designed to target the mRNAs transcribed from the zebrafish genes for the fibrinogen alpha-chain, von Willebrand Factor (VWF), or ADAMTS13. Each resulting embryo (only embryos without morphologic defects were used), along with a matched non-MO-treated control embryo, was then injected at 72 hpf with 30-60 ng bovine thrombin, and thrombus formation scored blindly by 2 independent observers at 30 seconds and 2 minutes post injection. An ∼70% reduction in mean thrombosis score was observed for embryos treated with the fibrinogen alpha-chain MO (N=29) compared to controls (p<0.01). A similar ∼60% reduction in mean thrombosis score was observed for VWF MO-treated embryos (N=27) compared to controls (p< 0.01). In contrast, injection of an ADAMTS13 MO (N=31) resulted in an ∼2.5 fold increase in mean thrombosis score (p< 0.01). Co-injection of the VWF and ADAMTS13 MOs (N=15) resulted in a mean thrombosis score that no longer differed significantly from control (p =0.36). In conclusion, bovine thrombin injection induces prompt thrombosis in zebrafish, which appears to be dependent on zebrafish fibrinogen and VWF function, with subsequent spontaneous thrombus dissolution suggesting endogenous zebrafish fibrinolytic activity. Inhibition of ADAMTS13 appears to enhance thrombosis in this system, in a VWF-dependent manner. These results suggest that the structure and function of key hemostatic components is tightly conserved between zebrafish and mammals and that the injection of bovine thrombin into the zebrafish could serve as the basis for future high throughput mutagenic and chemical screens. DisclosuresNo relevant conflicts of interest to declare.

Endobrevin/VAMP-8–dependent dense granule release mediates thrombus formation in vivo

Individuals whose platelets lack dense or alpha-granules suffer various degrees of abnormal bleeding, implying that granule cargo contributes to hemostasis. Despite these clinical observations, little is known regarding the effects of impaired platelet granule secretion on thrombus formation in vivo. In platelets, SNARE proteins mediate the membrane fusion events required for granule cargo release. Endobrevin/VAMP-8 is the primary vesicle-SNARE (v-SNARE) responsible for efficient release of dense and alpha-granule contents; thus, VAMP-8(-/-) mice are a useful model to evaluate the importance of platelet granule secretion in thrombus formation. Thrombus formation, after laser-induced vascular injury, in these mice is delayed and decreased, but not absent. In contrast, thrombus formation is almost completely abolished in the mouse model of Hermansky-Pudlak syndrome, ruby-eye, which lacks dense granules. Evaluation of aggregation of VAMP-8(-/-) and ruby-eye platelets indicates that defective ADP release is the primary abnormality leading to impaired aggregation. These results demonstrate the importance of dense granule release even in the earliest phases of thrombus formation and validate the distal platelet secretory machinery as a potential target for antiplatelet therapies.

Endobrevin/VAMP-8-Mediated Dense Granule Release Is Required for Efficient Thrombus Formation in Vivo.

Individuals whose platelets lack dense core or alpha-granules suffer varying degrees of abnormal bleeding, implying that granule cargo contributes to hemostasis. Despite these clinical observations, little is known regarding the effects of impaired platelet granule secretion on thrombus formation in vivo. The release of cargo from platelet granules requires a group of membrane proteins called SNAREs (Soluble NSF Attachment Protein Receptors) that mediate fusion of granule membranes to the plasma membrane and open canalicular system. Endobrevin/VAMP-8 is the primary vesicular-SNARE (v-SNARE) responsible for efficient release of dense core and a-granule contents. To evaluate the importance of VAMP-8-mediated secretion on the kinetics of thrombus formation in vivo, we measured platelet accumulation following laser-induced vascular injury in VAMP-8−/− mice. Three different phases of thrombus formation - initiation, maximal accumulation, and stabilized platelet accumulation - were tested. Analysis of initial thrombus formation from wild-type and VAMP-8−/− mice showed that average platelet accumulation in VAMP- 8−/− mice was 23% of accumulation in wild-type mice (P=0.009) at 30 sec following injury. There was a trend towards smaller maximal thrombus size in VAMP-8−/− mice, but the difference was not statistically significant (P=0.1). Average stabilized platelet accumulation at 180 sec in VAMP-8−/− mice was 40% of wild-type mice (P=0.05). Thus, thrombus formation is delayed and decreased in VAMP-8−/− mice, but not absent. Dense granule release occurs more rapidly than alpha-granule release, which does not occur for 2–3 min following laser-induced vascular injury. Agonist-induced dense granule release from VAMP-8−/− platelets is defective. To directly evaluate the role of dense granule release on the kinetics of thrombus formation, we assessed thrombus formation in the mouse model of Hermansky-Pudlak syndrome, ruby-eye, which lack dense granules. Thrombus formation following laser-induced vascular injury was nearly abolished in ruby-eye mice such that maximal platelet accumulation was 15% that of wild-type mice. In vitro, the thrombin doses required to induce irreversible aggregation in wild-type, VAMP-8−/−, and ruby-eye platelets were 25 mU, 50 mU, and 150 mU, respectively. Incubation with apyrase had little effect on thrombin-induced aggregation of VAMP-8−/− or ruby-eye platelets. In contrast, incubation of wild-type platelets with apyrase reduced their thrombin sensitivity compared to that of ruby-eye platelets. Supplementation with a substimulatory ADP concentration reversed the thrombin-induced aggregation defect in VAMP-8−/− and ruby-eye mice. Thus, defective ADP release is the primary abnormality leading to impaired aggregation in VAMP-8−/− and ruby-eye mice. Tail bleeding times were assessed in VAMP- 8−/− mice to evaluate the role of VAMP-8 in hemostasis. In contrast to ruby-eye mice, which have a markedly prolonged bleeding time, tail bleeding times in VAMP-8−/− mice were not significantly prolonged compared to those in wild-type mice. These results demonstrate the importance of VAMP-8 and dense granule release in the initial phases of thrombus formation and validate the distal platelet secretory machinery as a potential target for anti-platelet therapies.

Knockdown of prothrombin in zebrafish

Thrombin is a serine protease generated from its zymogen, prothrombin, and plays a central role in the coagulation cascade. It is also important for mammalian development. The zebrafish has now been established as an excellent genetic model for studies on mammalian hemostasis and development. In this report, we used prothrombin-specific antisense morpholinos to knock down the levels of prothrombin to characterize the effects of prothrombin deficiency in the zebrafish embryo. Prothrombin morpholino-injected zebrafish embryos yielded an early phenotype exhibiting severe abnormalities that later showed occasional bleeding. In a second late phenotype, the embryos had no observable morphological abnormalities in early stages, but showed occasional bleeding at later stages. These phenotypes resembled characteristics shown by prothrombin knockout mice. Laser-induced vascular injury on some of the normal appearing phenotypic larvae showed a prolonged time to occlusion, and recombinant zebrafish prothrombin injected into these larvae restored a normal time to occlusion thus showing the specificity of the morpholino effect. The system developed here should be useful for investigation of the role of thrombin in vertebrate development.