Engenharia de proteínas com base em evolução: interconversão funcional entre 5-hidroxiisourato hidrolases e transtirretinas

• Ano de defesa: 2022
• Autor(a) principal: Rafael Pereira Lemos
• 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: Transtirretina; Modelagem computacional; Engenharia de proteínas; Bioinformática; Pré-Albumina; Engenharia de Proteínas
• Link de acesso: http://hdl.handle.net/1843/55408

Transthyretins are a protein family that exhibits binding and distribution of thyroid hormones (T3 and T4) as primary functions. Despite being exclusive to vertebrates, a major part of the evolutionary process of these proteins occurred before the emergence of such organisms, originating via duplication of purine catabolic pathway genes. Those encode the 5- hydroxyisourate hydrolase (HIUase) enzyme, which acts in the conversion of 5- hydroxyisourate (HIU) to 2-oxo-4-hydroxy-4-carboxy-5-ureidoimidazoline (OHCU). Differently from transthyretins, the HIUase subfamily is ubiquitous in both prokaryotes and eukaryotes, with hominids being an important exception. In vivo, both subfamilies are found in a homotetrameric form, with each monomer being formed by eight beta-strands connected by seven loops, and one alpha-helix. Tetramers are made stable by hydrophobic interactions between each dimer pair, leading to the formation of an internal charged catalytic cavity. Considering these proteins' intricate evolutionary history, as well as their high primary and quaternary structural similarity, we hypothesized that specific in silico substitutions would be able to switch their functions. We then engineered two putative protein sequences, where one subfamily representative sequence was substituted in specific and correlated locally conserved positions by the other, and vice-versa. Applying computational modeling, we aimed to better understand the neofunctionalization process of these paralogues. Using Modeller and AlphaFold, we generated 3D homology structural models, while also employing a chimeric manual alignment, between the reference and engineered proteins, to further improve the results. The best models were refined, validated, and then their cavities geometries, volumes and electrostatic potentials were analyzed using CAVER and APBS software suites. Our results indicate that, as expected, the volumes and geometries differ from one another, due to size and physicochemical differences between their ligands. The observed changes in polarity and large amino acid residue side chains fit the subfamilies’ structural conservation.