Saccharomyces cerevisiae as a tool for the modeling of variants and the functional analysis of genes associated with mitochondrial diseases

Mitochondrial pathologies are multisystemic disorders caused by mutations in both nuclear and mitochondrial genes, resulting in an impairment of the OXPHOS activity or, more in general, of the mitochondrial metabolism. Altogether, these pathologies show an overlap of the affected tissues and of the symptoms, even if the primary causes are mutations in different genes. On the contrary, it is often observed that mutations in the same gene can lead to different pathological phenotypes, depending on different factors which in several cases are still unknown. Over the last 40 years, knowledge of the molecular basis of mitochondrial diseases has deeply increased, thanks to a multidisciplinary approach that involved both clinicians and researchers. Before the development of next-generation sequencing (NGS), novel mutations and novel genes were discovered at low rates according to the “function first”: although the “validation”, that is the demonstration of the causative role of the identified mutation(s), was essential, few novel mutations have a clinical uncertainty. However, the advent of NGS, thanks also to the continuous reduction of the costs for performing whole exome sequencing (WES) and whole genome sequencing (WGS) as well as the improvement of the technology, has led to a paradigm shift towards “genetics first”. Several novel mutations, often private, have been identified, most of which are classified as variants of uncertain significance (VUS). Although in several cases the phenotype of the patients is coherent with the mutations identified, validation is still necessary, to confirm, or exclude, the pathogenicity of the mutations identified. The validation is mandatory when mutations have been identified in a novel candidate gene known to be associated with mitochondrial metabolism but never reported as associated with mitochondrial diseases, and, even more, so if the mutations have been identified in a gene whose function is unknown or not related to mitochondrial function. For the reasons reported above, confirmation by in vitro and/or in vivo analysis is a fundamental prerequisite for demonstrating an underlying defect. In this thesis work, the yeast Saccharomyces cerevisiae was used as a model system to contribute to reaching this aim. The work focused on the validation of 10 mutations found in patients suffering from different pathologies in six different genes. The activities carried out by the proteins encoded by the genes under study vary throughout mitochondrial metabolism, from the replication, repair, and recombination activity of POLG, to the involvement in the oxidative phosphorylation of SDHA, BCS1L, and CYCS, up to the modification of mitochondrial tRNAs of ELAC2 and the mitochondrial aminoacyl tRNA synthetase activity of LARS2. Eight of these new variants were confirmed to affect the mitochondrial metabolism in yeast, though with different mechanisms and to different extents, strongly suggesting a pathogenic role of the equivalent human mutation. A single homozygous mutation found in LARS2 was demonstrated to be neutral in yeast, inducing our collaborators who identified the mutation to search for another mutation that could be explicative of the pathological phenotype, which was identified in linkage disequilibrium with the LARS2 mutation in another gene, CCR2. Besides validation, we aimed at identifying a putative role of SNF8 in mitochondrial metabolism, since patients recently found to harbor compound mutations in such gene have symptoms typical of a mitochondrial disease We discovered that in yeast, the lack of Snf8 resulted in a general defect of the mitochondrial function, affecting OXPHOS activity, stability of the mtDNA and mitochondrial morphology, the latter typical of mutants with defects in mitochondrial fusion or fission. This thesis also involved the search for beneficial molecules capable of recovering the defective mitochondrial phenotype of a strain mutated in MIP1, the ortholog of POLG, one of the nuclear genes mainly associated with a plethora of mitochondrial diseases, called POLG-related diseases, and characterized by mtDNA depletion or multiple mtDNA deletions. Through an approach mainly focused on drug repurposing and based on the screening of thousands of drugs, we identified five molecules whose administration to yeast resulted in a simultaneous improvement in the stability of the mtDNA, in the levels of mtDNA, and the increase of the oxygen consumption rate. In future, these molecules will be tested by collaborators in ad-hoc homozygous and heterozygous zebrafish polg mutants and, eventually, in fibroblasts from patients affected by mutations in POLG. As reported for other drugs identified using the same strategy, these molecules could also have a positive effect on other pathologies caused by mutations associated with mtDNA instability, becoming a new example of therapy for more diseases based not on targeting the primary cause, that is the nuclear mutation, but the secondary cause, that is the depletion and/or the deletions of the mtDNA.