New functions of platelet C3Ginvolvement in TPO-regulation, ischemia-induced angiogenesis and tumor metastasis

Resumen

GTPases are proteins that control a wide variety of cellular processes, such as proliferation, differentiation and apoptosis, among others. These proteins cycle between two states: one active or GTP-bound, and one inactive or GDP-bound. The exchange of GDP for GTP is catalysed by a group of proteins called GEF (Guanine nucleotide Exchange Factors), whereas GAP (GTPases Activating Proteins) participate in the inactivation of the GTPases. C3G is a GEF for several members of the Ras family of GTPases, mainly Rap1, R-Ras and TC21, and for the Rho family GTPase, TC10. Using mice overexpressing C3G (tgC3G) or a mutant form lacking the catalytic domain (tgC3GCat) specifically in megakaryocytes and platelets, we have previously shown that C3G promotes megakaryocyte (Mk) differentiation and regulates hemostatic functions of platelets. In particular, tgC3G platelets show a greater platelet activation and aggregation, which is correlated with lower bleeding times in tgC3G mice and increased thrombus formation in vivo. Moreover, C3G overexpression in platelets alters the release of platelet α-granules, characterized by the retention of vascular endothelial growth factor (VEGF) and thrombospondin-1 (TSP-1) inside the platelet, resulting in a net proangiogenic secretome. As a consequence, tgC3G expression in platelets promotes faster tumor growth in two models of heterotopic tumor cell implantation: 3LL (Lewis Lung carcinoma) cells and B16-F10 melanoma cells. Platelet C3G also promotes pulmonary metastasis of B16-F10 cells. However, transgenic C3G expression does not modify platelet counts in peripheral blood. In my thesis, I have deepened into the role of C3G in megakaryopoiesis, ischemia-induced angiogenesis and tumor metastasis. For that, we developed an additional mouse model, in which C3G is specifically deleted in megakaryocytes (Rapgef1flox/flox; PF4-Cre+/-, hereinafter C3G-KO). C3G-KO mice did not show differences in the number of Mk in the bone marrow (BM) or in platelet count in peripheral blood, similar to what was observed in the tgC3G model. However, ablation of C3G promoted Mk maturation in BM cultured in medium supplemented with thrombopoietin (TPO) and a cocktail of cytokines, suggesting a possible role of C3G in pathological megakaryopoiesis. Since platelet C3G showed no effect on megakaryopoiesis and thrombopoiesis in a physiological context, we analysed its role in two in vivo models of pathological megakaryopoiesis: TPO injection and 5-Fluorouracyl (5-FU)-induced myelosuppression. The intravenous injection of TPO stimulates megakaryopoiesis and increases platelet levels, while 5-FU induces platelet depletion around day 7 after injection, which is followed by a profound increase in platelet numbers, known as platelet rebound, 10-15 days after treatment. Surprisingly, TPO injection produced a more efficient increase in platelet levels in C3G-KO mice than in their controls. Moreover, after reaching the peak, the levels of C3G-KO platelets decreased slower than control platelets, suggesting an impaired downregulation. On the other hand, 5-FU injection resulted in significant lower platelet rebound in C3G-KO mice compared to their control. However, similar to the TPO-stimulation model, while control animals recovered normal platelet levels 20 days after the injection, platelet levels remained elevated in C3G-KO mice, indicating a defective downregulation. Opposite results were obtained in tgC3G mice, showing a greater rebound and a faster downregulation of platelet levels. TPO is the major cytokine involved in megakaryopoiesis and platelet formation. TPO is constitutively produced in the liver and its plasma concentration varies inversely with the number of platelets, which remove it by clearance. In platelets, TPO binds to its receptor, c-Mpl, and induces its endocytosis, recycling and degradation. This mechanism is driven by the E3 ubiquitine ligase c-Cbl, responsible for the ubiquitination of c-Mpl and its degradation by the proteasome and lysosome systems. C3G and c-Cbl interaction has been detected previously in K562 cells, an erythro-megakaryoblastic cell line that acquires Mk markers upon stimulation with phorbol 12-myristate 13-acetate (PMA). In this Thesis we demonstrated that C3G and c-Cbl also interact in platelets and that C3G is a mediator of c-Cbl phosphorylation and activation by kinases of the Src family (SFK). Indeed, C3G ablation in platelets completely impaired c-Cbl phosphorylation in response to TPO, due to a decreased interaction between Src and c-Cbl. This resulted in reduced c-Mpl internalization and degradation. In concordance, c-Mpl ubiquitination (measured by immunofluorescence as colocalization between c-Mpl and ubiquitin signal) was significant decreased in TPO-stimulated C3G-KO platelets. Moreover, total ubiquitinated proteins levels dropped drastically in resting, thrombin- and TPO-stimulated C3G deficient platelets. Opposite results were obtained in tgC3G platelets exhibiting greater p-c-Cbl levels, correlated with a faster degradation and a greater ubiquitination of the c-Mpl receptor. In addition, in vitro TPO uptake was significantly decreased in C3G-KO platelets after 30 minutes incubation. In summary, C3G-KO platelets are unable to regulate TPO plasma levels owing to a defective Src-mediated c-Cbl phosphorylation, which results in defective TPO-c-Mpl internalization and degradation. This could explain the sustained elevated platelet levels found in the pathological models. All these results suggest the participation of C3G in the Src-c-Cbl pathway to regulate TPO levels and, hence platelet levels. In the second part of the Thesis, we have studied the involvement of C3G in platelet-mediated angiogenesis. For that, we have used two models of ischemia-induced angiogenesis: tumor implantation and hind-limb ischemia. In response to hypoxia, VEGF, released by platelets and endothelial cells, induces platelet secretion of SDF-1 (stromal-derived factor, also known as CXCL12). VEGF and SDF-1 facilitate the recruitment of bone marrow-derived proangiogenic progenitor cells (BMDC) through interaction with their respective receptors, VEGFR1 and CXCR4. Then, these BMDC, also known as hemangiocytes, release angiogenic factors at the ischemic site that promote the incorporation and assembly of endothelial progenitor cells and the stabilization of new blood vessels. We found that C3G-KO mice showed a significant increase in the recruitment of hemangiocytes, in both ischemic models. In addition, C3G-KO mice exhibited larger and more vascularized tumors after 15 days of the injection of 3LL (Lewis Lung Carcinoma) cells, and a slightly faster recovery of the blood flow after hind-limb ischemia, as compared to their control siblings. In contrast, lower number of hemangiocytes were detected in tgC3G mice in both ischemia models. In concordance with these in vivo results, C3G ablation resulted in increased SDF-1 secretion in vitro, whereas C3G overexpression resulted in an impaired SDF-1 release. In addition, C3G ablation increased VEGF release and TSP-1 retention, which collectively results in a proangiogenic secretome and could explain the increased angiogenesis observed in the in vivo studies. These results suggest that platelet C3G has a role in hypoxia-induced BMDC recruitment through the modulation of SDF-1, VEGF and TSP-1 release. In the third part of the Thesis, we have explored the role of C3G in platelet-mediated tumor metastasis. Previously, we described that transgenic expression of C3G in platelets promotes long-term metastasis of B16-F10 cells in the lungs. To further analyse this C3G function we developed a short-term metastasis model consisting in the injection of B16-F10 cells, expressing EGFP (green fluorescent protein), into the retroorbital sinus of the mouse and the count of EGFP-expressing cells in the lungs, by flow cytometry, 1 hour after the injection. TgC3G mice showed increased number of melanoma cells in the lungs, while C3G platelet ablation resulted in fewer short-term metastases. In vitro adhesion assays in poli-L-Lysine coverslips revealed increased adhesion of melanoma cells when they were incubated with tgC3G platelets. In contrast, melanoma cells incubated with C3G-KO platelets showed decreased adhesion. These results suggest that C3G enhances platelet ability to promote adhesion of tumor cells to the metastatic niche. Since the metastatic potential of tumor cells in vivo correlates with their ability to aggregate platelets, we next studied the contribution of C3G to TCIPA (tumor cell induced platelet aggregation). TgC3G platelets incubated with B16-F10 cells showed significantly greater aggregation than control platelets, which was accompanied by a higher activation of integrin αIIbβ3 and correlated with an increased Rap1 activation. We did not find significant differences in TCIPA between C3G-KO platelets and their controls, in correlation with a similar activation of integrin αIIbβ3, although Rap1 activation was impaired in these platelets, as described in response to other stimuli. All these results support the notion that C3G contributes to platelet-mediated tumor growth and metastasis, including cell adhesion to the metastatic niche and platelet-tumor cell communication, with a direct impact on metastases. In conclusion, in this work we present evidence of a new role of C3G in the modulation of platelet levels through the regulation of c-Cbl-dependent c-Mpl ubiquitination and degradation. Furthermore, platelet C3G participates in hypoxia-induced angiogenesis by modulating hemangiocyte recruitment through the regulation of SDF-1 and VEGF release, as well as regulating TSP-1 release. Finally, we reveal a new function of C3G in TCIPA, facilitating metastatic cell homing and adhesion.