Mitochondria are double membrane–enclosed organelles that play a crucial role in ATP production, metabolism, regulation of cellular signaling and amplification of programmed cell death (Wasilewski and Scorrano, 2009). In the process of apoptosis mitochondria release cytochrome c and other cofactors that are required to amplify cell death (Li et al., 1997). The complete release of cytochrome c depends on the changes in the shape and in the ultrastructure of the organelle, since during these processes mitochondrial network undergoes fragmentation, that is accompanied by cristae remodeling and widening of cristae junctions (Frank et al., 2001; Scorrano et al., 2002). Of note, deregulation of apoptosis represents a typical hallmark of cancer, since cancer cells exploit the inhibition of the mitochondrial arm of apoptosis to acquire the malignant phenotype (Thompson, 1995). Mitochondria are dynamic organelles, and all processes that impinge on the changes in the shape and in the ultrastructure of the organelle are controlled by a regulated action of mitochondria shaping proteins, which represent large GTPases that share structural homology with the dynamin protein family (Dimmer and Scorrano, 2006). Mitochondrial shape in the steady state is a result of the balanced action of fission and fusion events (Griparic and van der Bliek, 2001). The process of mitochondrial fission is controlled by a synchronized action of a cytosolic protein Drp1 (Dynamin – related protein 1) (Cereghetti et al., 2008), that is recruited to the outer mitochondrial membrane where it binds its adaptors Fis1 (Fission – 1), MFF (Mitochondrial fission factor), Mid49 and Mid51 (Mitochondrial division), and participates in the division of mitochondria (Palmer et al., 2011). Mitochondrial fusion, on the other hand, is a process controlled by mitofusins (Mfn1 and Mfn2), proteins located in the outer mitochondrial membrane, together with the only inner membrane GTPase - Optic Atrophy 1 (Opa1) (Santel and Fuller, 2001; Chen et al., 2003; Cipolat et al., 2004). In humans, alternative splicing of Opa1 gives rise to 8 mRNA splice variants which further get processed by proteolytic proteases giving rise to 2 long and 3 short forms of Opa1 (Olichon et al., 2007; Duvezin-Caubet et al., 2007). Opa1 is a multifunctional protein: apart from its function in promoting mitochondrial fusion (Cipolat et al., 2004), it also plays a role in the control of apoptosis by keeping in check the cristae remodeling pathway, by forming multimeric complexes at the cristae junctions, keeping in shape the size of these junctions (Frezza et al., 2006; Cipolat et al., 2006). Another important role of Opa1 is in the control of mitochondrial metabolism, because Opa1 favors the superassembly of respiratory chain complexes into supercomplexes, increasing the efficiency of oxidative phosphorilation (Cogliati et al., 2013). All these functions concur to determine the beneficial outcome of its mild overexpression in vivo, which protects from heart and brain ischaemia, denervation-induced muscular atrophy and fulminant hepatitis (Varanita et al., 2015). Furthermore, it corrects mouse models of primary mitochondrial dysfunction caused by defects in components of the respiratory chain (Civiletto et al., 2015). However, all these beneficial effects come with a counterpart, since a handful of studies reported that Opa1 is overexpressed in several human cancers where high levels of Opa1 correlated with a worst prognosis and an impaired response to therapy (Fang et al., 2012), while blocking its expression was associated with an induction of the mitochondria - associated apoptotic pathway in the cancer cell and a better clinical outcome (Zhao et al., 2013). In this Thesis we set out to understand what role does Opa1 play in the acquisition and maintenance of the cancer phenotype, both in cellular and animal models, while reasoning that a possible explanation why we don’t have constitutively high Opa1 levels is the fact that the trade off of Opa1 overexpression could be an increased susceptibility to cancer development/progression. Well established cell lines, initially deriving from patients diagnosed with diffuse large B cell lymphoma (DLBCL) served as our in vitro model system. DLBCLs are one of the most common adult non-Hodgkin lymphoid malignancies today (Lohr et al., 2012). They are a genetically heterogeneous group of tumors that can be further divided in several subsets, identified by their distinct molecular signatures (Alizadeh et al., 2000). Genome wide arrays and multiple clustering algorithms defined a B cell receptor (BCR)/proliferation cluster (BCR–DLBCL), which displays upregulation of genes encoding BCR signaling components, and an OxPhos cluster (OxPhos–DLBCL) which is enriched in genes involved in mitochondrial oxidative phosphorylation. The OxPhos subset lacks an intact BCR signaling network, suggesting dependence on alternative survival mechanisms, which are not yet defined (Monti et al., 2005; Caro et al., 2012). Since a proteomic approach, aimed at carefully dissecting components of the mitochondrial proteome in the BCR versus OxPhos cell group, identified increased levels of Opa1 in the OxPhos (Danial N, manuscript in preparation), we wished to elucidate what role does Opa1 play in these cancer cell subsets. In order to test whether Opa1 overexpression contributes to the development and progression of cancer in vivo, we reached out to an already established mouse lymphoma model, the Eµ-myc transgenic mouse (Adams et al., 1985), that we further crossed with a mouse model of controlled Opa1 overexpression that was recently generated in our lab (Cogliati et al., 2013), and the net result of this cross gave rise to the mouse model we used in our study. In this Thesis we present evidence that Opa1 is increasingly processed in the BCR subset of diffuse large B cell lymphoma, and that mitochondrial morphology, metabolism, and ultrastructure are different between the BCR and the OxPhos DLBCL subsets that display different levels of Opa1. Furthermore, we also show evidence of a marked synergy between Opa1 and c-Myc in doubly transgenic mouse models, where Opa1 overexpression is contributing to the development of, and aggravating cancer in Eμ-Myc transgenic animals. The work performed in this thesis highlights a role for Opa1 in DLBCL features, and tumor progression in vivo. Thus, our data indicate that Opa1 displays oncogenic features and it can be taken into consideration as a novel therapeutic target for cancer treatment.