MELAS (Mitochondrial Encephalomyopathy, Lactic Acidosis and Stroke- like episodes) syndrome is a rare mitochondrial disorder mainly caused by the m.3243A>G mutation in the mitochondrial DNA. This mutation affects mitochondrial proteins translation, causing a defect in the synthesis of mitochondrial respiratory chain components. MELAS patients suffer from a wide variety of symptoms that involve neurodegeneration and there is no curative treatment for the disease. In this thesis, we have studied MELAS syndrome pathophysiology using patient-derived fibroblasts carrying the m.3243A>G mutation. We evaluated several parameters of mitochondrial function, mitochondrial quality control pathways, such as autophagy and mitophagy, and inflammation. Our results show clear phenotypical alterations in MELAS fibroblast, such as decreased mitochondrial bioenergetics, impaired autophagy/mitophagy fluxes and NLRP3 inflammasome activation. Then, we performed a screening of compounds in a yeast model harboring the A14G mutation (equivalent to A3243G). We tested the effect of one of the positive compounds, rosmarinic acid, in MELAS cybrids and patient-derived fibroblasts. Rosmarinic acid was demonstrated to have a moderate effect on MELAS cellular pathophysiology. However, cybrids and fibroblasts are not much vulnerable to energy- dependent defects resulting from mitochondrial dysfunction. For that reason, we generated induced neurons (iNs) by direct reprogramming of fibroblasts carrying the m.3243A>G mutation. The pathophysiological characterization of MELAS iNs indicates that they can be used as a cellular model to elucidate the mechanisms underlying the disease as well as a possible screening platform.