Role of platelet C3G in vesicle trafficking and spreading. Involvement in hemostasis and platelet-mediated inflammation

Resumen

C3G is a guanine nucleotide exchange factor (GEF) for several members of the Ras family of GTPases, such as Rap1, R-Ras, or TC21, and of the Rho family, such as TC10. GTPases are proteins that act as molecular switches, cycling between a GTP-bound state (active conformation) and a GDP-bound state (inactive conformation), resulting in rapid switching of signaling pathways. This cycle is regulated by GEF proteins, which promote the exchange of GDP for GTP, triggering the activation of the GTPase, and by GAP proteins, which catalyze the hydrolysis of GTP into GDP. C3G participates in most cellular processes, such as proliferation, adhesion and migration, apoptosis or cell-cell interaction, among others. Previous results from the group have revealed a role for C3G in platelet function, through its GEF activity. Thus, C3G regulates Rap1b activation, induced by most platelet agonists, through the PKC-Src, ERK-Shp2, P2Y12-PI3K, and TXA2 pathways. C3G also modulates pathologic megakaryopoiesis in response to thrombopoietin (TPO) or 5-fluorouracil treatment, by regulating the TPO-Mpl pathway in megakaryocytes. Finally, a series of evidences indicated that C3G could control the release of platelet ¿-granule content, since C3G regulates the secretion of angiogenic factors. C3G overexpression in platelets causes retention of VEGF, SDF-1, and TSP-1 within them, leading to a proangiogenic secretome that promotes tumor growth and metastasis. Likewise, the absence of C3G triggers the release of VEGF-1 and SDF-1 and the retention of TSP-1, which also results in a secretome that promotes angiogenesis. In this Thesis work, we have studied the role of C3G in (i) hemostasis, (ii) platelet spreading on substrate, (iii) secretion of platelet granules and (iv) platelet-mediated inflammation. For this, we have used two transgenic mouse models that overexpress C3G (tgC3G) or a mutant form in which C3G lacks the catalytic domain (tgC3G¿Cat), specifically in megakaryocytes (MKs) and platelets, and a C3G knockout mouse model (C3G-KO), in which C3G is specifically deleted in these cell types. In the first place, we have corroborated in C3G-KO platelets, the participation of C3G in the PKC-Rap1 pathway, which regulates the second wave of thrombin-induced Rap1 activation. In addition, we have determined that PKC¿ would be the PKC isoform that controls C3G in this system. Thus, C3G-KO platelets showed defective inside-out signaling, presenting alterations in the activation of the integrin ¿IIbß3, responsible for platelet aggregation, and in the exposure of P-selectin on the surface. C3G would also modulate secondary hemostasis, regulating phosphatidylserine (PS) translocation. Indeed, C3G-KO platelets displayed defective PS translocation, resulting in decreased thrombin generation upon tissue factor (TF) stimulation. However, the absence of C3G promoted a procoagulant secretome that compensated for the faulty translocation of PS, generating normal thrombin levels. In contrast, C3G overexpression induced an anticoagulant secretome and did not affect PS translocation, suggesting that low levels of C3G are necessary but sufficient to induce PS translocation. In the second part of this work, we have examined the role of C3G in outside-in signaling, specifically, in platelet spreading and clot retraction. Previous results of the group have described that C3G positively modulates platelet spreading on poly-L-lysine, independently of its GEF function. In this work we describe that the role of C3G in spreading is controlled by PKC and Src and is substrate dependent. Specifically, C3G would modulate platelet spread on poly-L-lysine, collagen/CRP, fibronectin, and vitronectin, but not on fibrinogen, laminin, or osteopontin, independently of its GEF domain. These differences can be partially explained by the defects in the expression of ¿V and ß1 integrins observed in C3G-KO platelets. C3G regulates platelet spreading by modulating remodelling of the actin cytoskeleton, but not of microtubules. In addition, we have shown that C3G is associated with the formation of lamellipodia, but not filopodia, through the activation of Rac1 GTPases. In addition, C3G interacted with proteins related to actin cytoskeleton dynamics, such as ß-actin, the Arp2/3 complex, VASP, and Abi-1. The C3G¿Cat mutant also interacted with Arp2 and ß-actin and induced Rac1 activation, which is consistent with the spreading results and suggests that the role of C3G in actin cytoskeleton remodeling is independent of Rap1. Thus, C3G would modulate the formation of lamellipodia through the control of the Rac1/WAVE/Arp2/3 pathway. C3G is also involved in the formation of focal adhesion complexes (FA) on platelets, through its interaction with talin, p130Cas, c-Cbl, vinculin, and FAK, in response to thrombin. C3G regulated the number of FA during platelet spreading on CRP and fibronectin; however, both overexpression and deletion of C3G induced increased FA formation, indicating that C3G would up- and down-regulate different components during FA formation. In fact, C3G differentially modulated the phosphorylation of paxillin, c-Cbl and p130Cas on fibronectin and CRP. In addition, C3G-KO platelets exhibited a significant delay in clot retraction, further supporting a role for C3G in outside-in signaling. In the third part of the Thesis, we have explored the role of C3G in the secretion of platelet granules. Previous results from the group demonstrated that C3G controls the release of angiogenic factors, such as VEGF, SDF-1 and TSP-1, which are contained in ¿-granules. Indeed, deletion of C3G in platelets resulted in increased cargo release. The role of C3G in granule secretion appears to be specific to ¿-granules, not affecting ¿-granules or lysosomes. This increased secretion of ¿-granules observed in C3G-KO platelets was not caused by alterations in the number of granules, indicating that C3G would act as a secretion brake. In platelets, granule secretion is controlled by SNARE proteins. Under resting conditions, C3G-KO platelets showed lower levels of P-selectin, as well as v-SNAREs (vesicle SNAREs) VAMP-7 and VAMP-8, all of them located in ¿-granules. However, after thrombin stimulation the levels (measured as fluorescence signal) increased, suggesting that the absence of C3G could promote granule fusion, prior to exocytosis (compound exocytosis). In fact, a high percentage of VAMP-7 ¿-granules from C3G-KO platelets showed a peripheral distribution pattern, that is, granules close to the plasma membrane (PM), which was not observed in C3G-wt platelets, where the granules preferentially concentrated in the center of the platelet. This abnormal alpha granule distribution observed in C3G-KO platelets is possibly due to the actin cytoskeleton defects detected in these platelets. C3G regulates ¿-granule secretion through RalA activation, which is dependent on C3G GEF activity. However, we did not observe any effect of C3G on Rab27b activation, consistent with its lack of involvement in the regulation of ¿-granules. Additionally, C3G interacted with VAMP-8, Syntaxin 11, SNAP23 and Munc18-b, all of which involved in the association between v-SNAREs and t-SNAREs to form the trans-SNARE complex, but not with components of the exocystic complex. These interactions were not detected in tgC3G¿Cat platelets, suggesting that they would be mediated by the C3G catalytic domain (GEF). C3G overexpression caused deficient formation of the trans-SNARE complex, while the opposite phenotype was observed in C3G-KO platelets. C3G would modulate the formation of the trans-SNARE complex through its positive role in the formation of the VARP/VAMP-7/Arp2/3 complex, which prevents the uncontrolled secretion of ¿-granules and the remodeling of the actin cytoskeleton. All of this explains why the absence of C3G in platelets promoted a kiss-and-run phenotype, in which platelets show a decrease in the exposure of P-selectin on the surface, due to a lack of incorporation of the ¿-granule membrane into the PM, accompanied by a marked increase in secretion, after platelet stimulation. Finally, we have shown that the role of C3G in granule secretion is not platelet-specific, but that C3G also modulated granule secretion in the PC12 cell line. Overexpression of an active C3G mutant induced slower NPY secretion, whereas absence of C3G triggered the opposite phenotype. However, C3G positively regulated the number of vesicles docked to the PM, as well as the number of exocytic events. Thus, C3G mutants showed the opposite phenotype in PC12 cells to that of platelets; ie, overexpression of the active C3G mutant caused increased secretion, whereas C3G silencing promoted decreased secretion. This contradictory phenotype could be explained by the different functionality of the VARP-VAMP-7 interaction in platelets and in neurons. In neurons, VARP positively regulates VAMP-7, inducing the trans-SNARE complex, while, in platelets, the VAMP-7-VARP interaction promotes the opposite effect. All these results support the notion that C3G regulates granule secretion through the modulation of the trans-SNARE complex formation and Ral activation, both processes dependent on its GEF activity. In the last part of the Thesis, we have examined the role of C3G in platelet-mediated inflammation. Platelets are important regulators of the immune response through their role in leukocyte activation, mediated by the formation of platelet-leukocyte aggregates (PLA) and the secretion of inflammatory factors contained in their ¿-granules. C3G modulated PLA formation, following thrombin stimulation, through regulation of P-selectin and CD40L exposure on the platelet surface. Specifically, C3G promoted the interaction between platelets and neutrophils (NE) and B lymphocytes, but not with T lymphocytes or monocytes. In addition to favoring platelet-NE interaction, C3G also modulated NE activation, finding a higher NETosis in tgC3G platelets, after thrombin stimulation, and a lower NETosis in C3G-KO platelets. In addition, the absence of C3G resulted in increased secretion of PF4, IL-1¿, CX3CL1, among others, while C3G overexpression showed the opposite phenotype, that is, retention of inflammatory factors, consistent with what was observed in the secretion of angiogenic and coagulation factors. However, it seems that the platelet secretome did not influence the formation of NETosis, with P-selectin exposure being the most important factor. C3G also modulated in vivo the progression of colitis induced by DSS (Dextran Sodium Sulfate) treatment. Indeed, the absence of C3G significantly accelerated DSS-induced colitis symptoms, in contrast to in vitro experiments. C3G-KO mice suffered significant weight loss accompanied by diarrhea and bloody stools and increased bacterial dissemination. Consistently, the absence of C3G also induced increased PLA formation, specifically with NE, but not with monocytes or lymphocytes, accompanied by increased NE counts in peripheral blood. These results suggest that platelet C3G could act as a protector against ulcerative colitis. In conclusion, in this work we present evidence of a role for C3G in the modulation of platelet spreading through the regulation of lamellipodia formation via its participation in the Rac1/WAVE2/Arp2/3 pathway, which is independent of its GEF function. In addition, C3G would regulate the secretion of ¿-granules by modulating the formation of the trans-SNARE and the VARP/VAMP-7/Arp2/3 complexes, as well as the activation of Ral, in a Rap1-dependent manner. Finally, we present preliminary results, both in vitro and in vivo, on a novel role of the C3G protein in the platelet-mediated inflammatory response, through the regulation of PLA formation, with NE and B lymphocytes, and the release of inflammatory factor.