Loss of α4A- and ß1-tubulins leads to severe platelet spherocytosis and strongly impairs hemostasis in mice.

Native circulating blood platelets present with a discoid flat morphology maintained by a submembranous peripheral ring of microtubules, named marginal band. The functional importance of this particular shape is still debated, but it was initially hypothesized to facilitate platelet interaction with the injured vessel wall and to contribute to hemostasis. The importance of the platelet discoid morphology has since been questioned on the absence of clear bleeding tendency in mice lacking the platelet-specific ß1-tubulin isotype, which exhibits platelets with a thinner marginal band and an ovoid shape. Here, we generated a mouse model inactivated for ß1-tubulin and α4A-tubulin, an α-tubulin isotype strongly enriched in platelets. These mice present with fully spherical platelets completely devoid of a marginal band. In contrast to the single knockouts, the double deletion resulted in a severe bleeding defect in a tail-clipping assay, which was not corrected by increasing the platelet count to normal values by the thrombopoietin-analog romiplostim. In vivo, thrombus formation was almost abolished in a ferric chloride-injury model, with only a thin layer of loosely packed platelets, and mice were protected against death in a model of thromboembolism. In vitro, platelets adhered less efficiently and formed smaller-sized and loosely assembled aggregates when perfused over von Willebrand factor and collagen matrices. In conclusion, this study shows that blood platelets require 2 unique α- and ß-tubulin isotypes to acquire their characteristic discoid morphology. Lack of these 2 isotypes has a deleterious effect on flow-dependent aggregate formation and stability, leading to a severe bleeding disorder.

Glycoprotein V : the unsolved GPV puzzle.

Glycoprotein V (GPV) is a highly expressed 82 KDa platelet surface transmembrane protein which is loosely attached to the GPIb-IX complex. Despite remaining questions concerning its function, GPV presents several unique features which have repercussions in hematology, atherothrombosis, immunology and transfusion. GPV is specifically expressed in platelets and megakaryocytes and is an ideal marker and reporter gene for the late stages of megakaryopoiesis. The ectodomain of GPV can be released by a number of proteases, namely thrombin, elastase and ADAM10 and 17. Although it was originally proposed as a thrombin receptor, this hypothesis was abandoned since thrombin activation was preserved after blockade of GPV cleavage and in Gp5 knockout mice. The combined potential of GPV to reflect the direct action of thrombin, platelet exposure to strong agonists and inflammatory conditions has led one to evaluate its utility as a marker in the context of atherothrombosis. Increased plasma levels of soluble GPV have notably been recorded in myocardial infarction, stroke and venous thromboembolism. It is also highly valued in transfusion to monitor platelet storage lesions. GPV presents several polymorphisms, which are a possible source of alloantibodies, while autoantibodies have been frequently detected in immune thrombocytopenia. The real biological function of this glycoprotein nevertheless remains an enigma, despite the respectively decreased and increased responses to low concentrations of collagen and thrombin observed in Gp5 knockout mice. Current studies are exploring its role in modulating general or VWF-induced platelet signaling, which could bear relevance in thrombosis and platelet clearance.

Mutations in the most divergent α-tubulin isotype, α8-tubulin, cause defective platelet biogenesis.

BACKGROUND: In the panel of genes commonly associated with inherited macrothrombocytopenia, an important fraction encodes key cytoskeletal proteins such as tubulin isotypes, the building blocks of microtubules. Macrothrombocytopenia-causing mutations have been identified in the TUBB1 and TUBA4A genes, emphasizing their importance in the formation of platelets and their marginal band, a unique microtubule ring-like structure that supports the platelet typical disc-shaped morphology. This raised the hypothesis that other tubulin isotypes normally expressed in platelets could play a similar role in their formation. OBJECTIVES: To assess whether tubulin isotype genes other than TUBA4A and TUBB1 could be implicated in inherited macrothrombocytopenia. METHODS: We used high throughput sequencing to screen a cohort of 448 French blood donors with mild thrombocytopenia for mutations in a panel of selected genes known or suspected to be involved in platelet biogenesis. RESULTS: We identified six distinct novel mutations in TUBA8, which encodes the most-divergent α-tubulin, as the causative determinant of macrothrombocytopenia and platelet marginal band defects. Functionally, all TUBA8 mutations were found to fully or partially inhibit the incorporation of the mutated α8-tubulin in the microtubule network. CONCLUSION: This study provides strong support for a key role of multiple tubulin genes in platelet biogenesis by discovering variants in a tubulin gene that was previously not known to be important for platelets.

Marginal band microtubules are acetylated by αTAT1.

The discoid shape of resting platelets is maintained by a peripheral, circular bundle of microtubules called marginal band. Marginal band microtubules are acetylated on lysine 40 of the alpha-tubulin subunits. We have previously shown that the deacetylase HDAC6 is responsible for tubulin deacetylation in platelets and that the hyperacetylated state of the microtubules in HDAC6KO platelets correlates with faster activation/spreading kinetics, pointing to a regulatory role of this modification. So far, the question about the reverse enzyme, responsible for tubulin acetylation in platelets, has remained unanswered. Several enzymes have been described as having tubulin acetylation activity. Here we identify αTAT1 as the enzyme responsible for the acetylation of marginal band microtubules. We show that αTAT1 deficiency has only minor consequences for platelet production and function. A residual tubulin acetylation level in αTAT1 deficient platelet lysates suggests the presence of an additional tubulin-acetylating enzyme that is unable to acetylate marginal band microtubules.

An essential role for α4A-tubulin in platelet biogenesis.

During platelet biogenesis, microtubules (MTs) are arranged into submembranous structures (the marginal band) that encircle the cell in a single plane. This unique MT array has no equivalent in any other mammalian cell, and the mechanisms responsible for this particular mode of assembly are not fully understood. One possibility is that platelet MTs are composed of a particular set of tubulin isotypes that carry specific posttranslational modifications. Although ß1-tubulin is known to be essential, no equivalent roles of α-tubulin isotypes in platelet formation or function have so far been reported. Here, we identify α4A-tubulin as a predominant α-tubulin isotype in platelets. Similar to ß1-tubulin, α4A-tubulin expression is up-regulated during the late stages of megakaryocyte differentiation. Missense mutations in the α4A-tubulin gene cause macrothrombocytopenia in mice and humans. Defects in α4A-tubulin lead to changes in tubulin tyrosination status of the platelet tubulin pool. Ultrastructural defects include reduced numbers and misarranged MT coils in the platelet marginal band. We further observed defects in megakaryocyte maturation and proplatelet formation in Tuba4a-mutant mice. We have, thus, discovered an α-tubulin isotype with specific and essential roles in platelet biogenesis.

Hirudin and heparin enable efficient megakaryocyte differentiation of mouse bone marrow progenitors.

Hematopoietic progenitors from murine fetal liver efficiently differentiate in culture into proplatelet-producing megakaryocytes and have proved valuable to study platelet biogenesis. In contrast, megakaryocyte maturation is far less efficient in cultured bone marrow progenitors, which hampers studies in adult animals. It is shown here that addition of hirudin to media containing thrombopoietin and serum yielded a proportion of proplatelet-forming megakaryocytes similar to that in fetal liver cultures (approximately 50%) with well developed extensions and increased the release of platelet particles in the media. The effect of hirudin was maximal at 100 U/ml, and was more pronounced when it was added in the early stages of differentiation. Hirugen, which targets the thrombin anion binding exosite I, and argatroban, a selective active site blocker, also promoted proplatelet formation albeit less efficiently than hirudin. Heparin, an indirect thrombin blocker, and OTR1500, a stable heparin-like synthetic glycosaminoglycan generated proplatelets at levels comparable to hirudin. Heparin with low affinity for antithrombin was equally as effective as standard heparin, which indicates antithrombin independent effects. Use of hirudin and heparin compounds should lead to improved culture conditions and facilitate studies of platelet biogenesis in adult mice.

Biosynthesis of platelet glycoprotein V expressed as a single subunit or in association with GPIb-IX.

Glycoprotein (GP) V is noncovalently linked to GPIbalpha, GPIbbeta and GPIX within the platelet GPIb-V-IX complex, a receptor for von Willebrand factor and thrombin. Two functions have been ascribed to GPV, namely, the modulation of thrombin- and collagen-dependent platelet responses. The biosynthesis of this molecule was investigated in pulse-chase metabolic labelling experiments performed in CHO cell lines transfected with GPV, alone or in the presence of GPIb-IX. GPV could not be detected at the surface of cells expressing the single subunit but was found instead as a soluble form in the culture medium. In pulse-chase studies, an immature 70 kDa protein was detected in cell lysates, whereas a fully processed 80-82 kDa form was only observed in the culture supernatants at later chase times. Immature GPV was N-glycosylated and retained before the medial Golgi while the secreted molecule contained complex sialylated sugars. The mature soluble form of GPV was produced by an enzymatic cleavage which was not affected by inhibitors of proteasome, calpain or metalloproteinases. When GPV was cotransfected with GPIb-IX, the former was no longer found in the culture supernatant but was retained in the cell membrane as shown by fluorescence-activated cell sorting and confocal microscopy analyses. Surface expressed GPV was processed from an immature 70 kDa form to produce a mature 80 kDa protein, processing similar to the intracellular trafficking of GPIbalpha. These results indicate that correct biosynthesis and surface expression of GPV in platelets requires the presence of the other subunits of the GPIb-V-IX complex.

Role of the intracellular domains of GPIb in controlling the adhesive properties of the platelet GPIb/V/IX complex.

Glycoprotein (GP) Ib/V/IX complex-dependent platelet adhesion to von Willebrand factor (VWF) is supported by the 45-kd N-terminal extracellular domain of the GPIb alpha subunit. Recent results with an adhesion blocking antibody (RAM.1) against GPIb beta, which is disulfide linked to GPIb alpha, have suggested a novel function of this subunit in regulating VWF-mediated platelet adhesion, possibly involving its intracellular face. A putative cooperation between the GPIb alpha and GPIb beta cytoplasmic domains was investigated by measuring the adhesion under flow to immobilized VWF of K562 and Chinese hamster ovary (CHO) cells transfected with GPIb/(V)/IX containing mutations in this region. Adhesion of cells carrying a glycine substitution of the GPIb beta Ser166 phosphorylation site was 50% lower than normal and became insensitive to inhibition by RAM.1. In contrast, forskolin or PGE(1) treatment increased both the phosphorylation of GPIb beta and adhesion of control cells, both effects being reversed by RAM.1, but had no influence on cells expressing the Ser166Gly mutation. A role of the GPIb alpha intracellular domain was also apparent as the VWF-dependent adhesion of cells containing deletions of the entire (Delta 518-610) or portions (Delta 535-568, Delta 569-610) of the GPIb alpha cytoplasmic tail was insensitive to RAM.1 inhibition. Cells carrying progressive 11 amino acid deletions spanning the GPIb alpha 535-590 region were equally unresponsive to RAM.1, with the exception of those containing GPIb alpha Delta 569-579, which behaved like control cells. These findings support a role of the GPIb beta intracellular domain in controlling the adhesive properties of the GPIb/V/IX complex through phosphorylation of GPIb beta Ser166 and point to the existence of cross-talk between the GPIb beta and GPIb alpha intracellular domains.