Predição in silico de Proteínas Extracelulares de Kluyveromyces lactis e suas relações com fatores transcricionais

• Ano de defesa: 2008
• Autor(a) principal: Brustolini, Otávio José Bernardes
• Orientador(a): Passos, Flávia Maria Lopes
• Banca de defesa: Ramos, Humberto Josué de Oliveira , Fietto, Juliana Lopes Rangel , Silva, Fabyano Fonseca e , Fontes, Elizabeth Pacheco Batista
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Viçosa
• Programa de Pós-Graduação: Doutorado em Bioquímica Agrícola
• Departamento: Bioquímica e Biologia molecular de plantas; Bioquímica e Biologia molecular animal
• País: BR
• Palavras-chave em Português: RNA; DGE; Software
• Palavras-chave em Inglês: RNA; GIP; software
• Área do conhecimento CNPq: CNPQ::CIENCIAS BIOLOGICAS::BIOQUIMICA::BIOLOGIA MOLECULAR
• Link de acesso: http://locus.ufv.br/handle/123456789/344

In this work we have created an in silico system to address secretion of a desired protein among the genome data of Kluyveromyces lactis. The completed K. lactis genome sequencing has provided a tool to construct such a system. In order to explore a potential K. lactis extracellular secretome, four computational prediction algorithms have been applied: SignalP (presence or absence of an N-terminal signal peptide and clivage site), Phobius (transmembrane topology), big-PI Predictor (GPI modification site) and WolfPsort (subcellular addressing, including extracellular prediction). These algorithms have correctly predicted 95 yeast secreted proteins sought in public databases (NCBI, UNIProt and MIPS). They have also predicted as intracellular the same number (i.e. 95) of random sequences found in K. lactis database. The K. lactis database consists of 5327 sequences (http://cbi.labri.fr/Genolevures). When analyzed by SignalP 3.0, it has pointed out 698 putative proteins with N-terminal signal peptides. In this group, 260 were predicted by Phobius to have no transmembrane domains and 236 were found by the big-PI Predictor to have no GPI modifications site. Finally, the predicted K. lactis secretome was estimated to consist of up to 101 sequences by WolfPSORT which eliminates proteins with subcellular targeting. In order to validate theses analysis, both groups of predicted and annotated extracellular ivproteins were compared by Hotelling s T2 test. The analysis has shown no differences between the mean values of these two groups. The physiological significance of those potential extracellular proteins was similarly investigated by analyzing the relationship between the S. cerevisiae transcriptional regulators ortologues in K. lactis and the putative promoters (i.e. 1 KB upstream) of those extracellular proteins. It was applied the methodology proposed by Yeastract which search for elements such as binding sites that indicates associations between transcriptional factor and target genes. The physiological condition favoring protein expression in extracellular medium was obtained by searching Gene Ontology (http://www.geneontology.org). It has been shown that most of the transcriptional regulators of K. lactis extracellular proteins are related to stress response, especially presence of drugs into the medium. Also pH stress and limiting nitrogen can induce the extracellular proteins.