Les réarrangements du cytosquelette sont essentiels pour la libération de plaquettes. Les RHO-GTPase, en tant que régulateurs du cytosquelette d'actine, joue un rôle important dans la formation de proplaquettes. Dans plusieurs processus cellulaires, CDC42 joue un rôle opposé à RHO/ROCK, qui régule négativement la formation de proplaquettes. Nous montrons que l'activité CDC42 augmenté pendant la différenciation des mégacaryocytes (MK). L’inhibition chimique via la CASIN aussi que l’utilisation d'une forme constitutionnellement active et d’une dominante négative de CDC42 ont montré que le CDC42 régule positivement la formation de proplaquettes. Ensuite nous avons investigué les effecteurs de CDC42 pour déterminer lequel est impliqués dans la formation de proplaquettes. Nous nous sommes d'abord concentrés sur PAK2, qui récemment chez la souris a été montré être activé par CDC42 dans la formation de proplaquettes. En utilisant des inhibiteurs chimiques et un shRNA contre PAK2, nous avons déterminé que, chez l'homme, PAK2 a un effet marginal sur la différenciation terminal des MKs. Donc nous avons étudié la famille WASP. Nous avons déterminé que N-WASP, mais pas WASP, était impliqué dans la formation de proplaquettes. L’ablation de N-WASP, entraîne une diminution marquée de la formation de proplaquettes pour un défaut dans le système de membrane de démarcation (DMS). Ceci a été associé à l'activation de RHOA, et a une concomitante augmentation de la phosphorylation de MLC2 sur Ser19. L’activation de N-WASP augmente pendant la différenciation MK. L'inhibition de la phosphorylation de N-WASP par deux inhibiteurs de la famille de Src kinase familiale, PP2 et Dasatinib, entraine la diminution de la formation de proplaquettes. Nous concluons que N-WASP, mais pas WASP régule positivement le développement du DMS et la formation de proplaquettes et que les kinases de la famille Src en association avec CDC42 régulent la formation de proplaquettes via N-WASP. / Cytoskeleton rearrangements are essential in platelet release. The RHO small GTPase family, as regulators of the actin cytoskeleton, plays an important function in proplatelet formation. In many cellular processes, CDC42 plays an opposite role to RHO/ROCK, which negatively regulates proplatelet formation. We show that CDC42 activity increased during megakaryocyte (MK) differentiation. Use of a chemical inhibitor CASIN or of an active or a dominant-negative form of CDC42 demonstrated that CDC42 positively regulates proplatelets formation. We determined which CDC42 effectors were involved in proplatelets formation. We first focused on PAK2, which was recently shown in mouse to be activated by CDC42 and regulate proplatelets formation. Using chemical inhibitors and an shRNA against PAK2, we determined that in human, PAK2 only has a marginal effect on terminal MK differentiation. Therefore, we switched our study towards the WASP family. We determined that N-WASP but not WASP was involved in proplatelets formation. N-WASP knock-down, led to a marked decrease in proplatelets formation due a defect in the demarcation membrane system (DMS). This was associated with RHOA activation, and a concomitant augmentation in the phosphorylation of MLC2 on Ser19. N-WASP phosphorylation, a primed form of N-WASP, increased during MK differentiation. Phosphorylation inhibition by two Src family kinase inhibitors, PP2 and Dasatinib, decreased proplatelets formation. We conclude that N-WASP, but not WASP positively regulates the DMS development and proplatelets formation and that the Src family kinases in association with CDC42 regulate proplatelets formation through N-WASP.

Thrombocythaemia is defined as a circulating platelet count above 450x10$^9$/L in humans. The major cause of thrombocythaemia is reactive $(secondary)$ thrombocythaemia which occurs secondary to many conditions such as infection, cancer and inflammation. However, acquired clonal mutations in mainly Janus Kinas 2 $(JAK2)$, CALR and MPL cause essential thrombocythaemia $(ET)$. ET is a rare disease that leads to an increased risk of cardiovascular thrombotic events. Current treatment of ET uses combination of low dose aspirin to decrease platelet function and cytoreductive agents to decrease thrombopoiesis. The most commonly used cytoreductive agents are hydroxyurea, anagrelide and interferon-$alpha$ and all have unwanted side effects. Cytokine receptor-like factor 3 $(CRLF3)$ is a 2.4kb gene that is ubiquitously expressed throughout the haematopoietic system. Very little is known about the function of CRLF3, with only one peer reviewed journal article in the literature which shows that CRLF3 may negatively regulate the cell cycle at the G0/G1 phase. However, nothing is known about the role of CRLF3 in platelet biology. Using a Crlf3 knockout mouse $(Crlf3-/-)$ developed by the Wellcome Trust Sanger Institute we show CRLF3’s role in platelet biogenesis and how it could be used as a novel therapeutic target to treat ET. Crlf3-/- mice have an isolated and sustained 25-40$\%$ decrease in platelet count compared to wildtype $(WT)$ controls. Platelet function is unaffected as demonstrated in a range in a range of in vitro assays. The thrombocytopenia is a consequence of abnormalities in hematopoietic cells, as shown by bone marrow transplantations. Megakaryopoiesis is upregulated in Crlf3-/- mice and proplatelet morphology is unaffected, suggesting the thrombocytopenia is due to increased platelet clearance. Indeed, splenectomised Crlf3-/- mice show normalised platelet counts within 7 days, showing rapid splenic removal of platelets is responsible for the thrombocytopenia. Abnormal large platelet structures that resemble proplatelets shafts $(preplatelets)$ are abnormally present in the circulation of elderly Crlf3/- mice. Immunohistochemistry showed increased and aberrant tubulin expression in Crlf3-/- platelets compared to WT controls, especially in the preplatelet forms. Cold induced depolymerisation of microtubules was decreased in Crlf3-/- platelets, suggestive of increased tubulin stability, however, the ratio of detyrosinated to tyrosinated tubulin was not altered. We then crossbred Crlf3-/- mice with JAK2 V617F ET mice, to determine the effect of Crlf3 ablation of thrombocythaemia. Crossbred mice showed restoration of platelet counts to WT values without grossly affecting platelet function or other blood lineages, providing the rational for CRLF3 as a novel therapeutic target for treatment of ET. Finally, we aimed to resolve the crustal structure of CRLF3 and discover its interactome. To this end, we were able to resolve the crystal structure of a C-terminal portion of the full length protein containing the predicted fibronectin type III domain. To shed light on the interactome of CRLF3, endogenous CRLF3 was tagged with a tandem affinity purification $(TAP)$ tag using CRISPR/Cas9 technology in induced pluripotent stem cells $(iPSCs)$. We have been able to produce megakaryocytes from these TAP-tagged iPSCs by forward programming. However, as yet we have not been able to generate enough MKs to have adequate material to perform immunoprecipitation assays. Therefore, the interactome of CRLF3 in MKs remains unknown. In conclusion, we identified a mechanism by which Crlf3 controls platelet biogenesis. Slowed maturation of Crlf3-/- preplatelets in the peripheral circulation potentially due to increased structural stability leads to rapid removal of these immature forms by the spleen and therefore a decrease in platelet count. The isolated effect on platelet numbers and normalisation of platelet count in ET mice deficient of Crlf3 provides the rational for further study on CRLF3 drug targeting as a novel therapeutic strategy for ET.