Entendendo os mecanismos moleculares de mutações que causam esclerose lateral amiotrófica

• Ano de defesa: 2019
• Autor(a) principal: Amanda Tábita da Silva Albanaz
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; Brasil; ICB - INSTITUTO DE CIÊNCIAS BIOLOGICAS; Programa de Pós-Graduação em Bioinformatica; UFMG
• Palavras-chave em Português: ALS; Mutações missense; Mecanismo molecular; Fenótipo; Esclerose amiotrófica lateral; Mutação de sentido incorreto
• Link de acesso: http://hdl.handle.net/1843/35341

Amyotrophic Lateral Sclerosis (ALS) is an age-dependent rare disease, characterized by neurodegenerative effects on motor activity, causing paralysis and death within two to four years. It is mainly caused by missense mutation affecting protein structure and function and SOD1, TDP-43 and FUS/TLS have particular importance for the disease molecular mechanism. Mutations in these proteins can disturb protein structure and function, leading to severe phenotypes, even though their main molecular mechanisms remain unclear. It has been previously reported that protein stability plays an important role in the pathogenicity mechanism of the disease, especially for SOD1, however we hypothesize this effect only partially describes the repertoire of molecular effects leading to different phenotypes. In this work, we investigate mutation properties and their predicted effects in an effort to better understand molecular mechanisms of mutations in ALS. To achieve this, we developed a new and expanded relational database describing ALS mutations and patient clinical data, DynAMISM, which also encompasses structural and sequence features describing these mutations and, in the future, assignments of putative structure-based molecular mechanisms. The database represents an increase in 56.6% of total clinical cases in comparison with a well established database. By analyzing SOD1 mutations, we have identified a strong correlation between molecular properties such as flexibility and effects in protein-protein affinity with clinical outcomes, which might indicate new potential mutation molecular mechanisms. The predictive power of these findings were, then, assessed using a regression tree and a Pearson’s correlation of up to 0.7 was achieved. As future work, we intend to made the database available as a user-friendly web interface, validate the predictive models using blind tests and extend them to other proteins. Understanding the molecular mechanisms of pathogenicity in ALS is an important step to guide better patient management and the development of more effective and personalized treatments, which we believe this work contributes to.