Deciphering the architecture/function relationship in complex bacterial promoters through Synthetic Biology approaches

• Autor(a) principal: Monteiro, Lummy Maria Oliveira
• Data de Publicação: 2020
• Tipo de documento: Tese
• Idioma: eng
• Título da fonte: Biblioteca Digital de Teses e Dissertações da USP
• Texto Completo: https://www.teses.usp.br/teses/disponiveis/17/17131/tde-08022021-151242/
• topic: Complex promoters; Design de circuitos; Design of circuits; Engenharia de proteínas; Fatores de transcrição; Gene regulation; Machine learning; Machine learning; Promotores complexos; Protein engineering; Regulação gênica; Transcription factors

Gene regulation has been studied extensively, however the complexity of the regulatory mechanisms still remains unknown. Understanding how gene regulation occurs is important not only to better understand the complexity of an organism but to postulate new rules, characterize new biological parts and then allow new design of biological circuits, for example. A possible strategy to unravel the mechanisms of action and complexity of bacterial promoters would be to combine the knowledge of gene regulation with the use of approaches from synthetic biology and bioinformatics, which, in turn, allow to design and build new functions in biological systems. Progress in synthetic biology is often made possible by powerful bioinformatics tools that allow the integration of the design, construction and testing stages of the biological engineering cycle. Consequently, the development of new bioinformatics tools is useful and important for scientists working on the design, development and testing of parts to extend or modify the behavior of organisms and design them to perform new tasks. In this context, the present thesis described (i) the existence of emergent properties in complex synthetic promoters in Escherichia coli, which could be extrapolated to naturally occurring regulatory systems and would significantly impact the engineering of synthetic biological circuits in bacteria. Taken together, these data demonstrate how small changes in the architecture of bacterial promoters could result in drastic changes in the final regulatory logic of the system, with important implications for the understanding of natural complex promoters in bacteria and their engineering for novel applications; (ii) the inducer recognition mechanism of two AraC/XylS regulators from Pseudomonas putida (BenR and XylS) for creating a novel expression system responsive to acetyl salicylate (i.e. Aspirin). Using protein homology modeling and molecular docking with the cognate inducer benzoate and a suite of chemical analogues, we identified the conserved binding pocket of these two proteins. As a result, a collection of engineered transcription factors (TFs) was generated with enhanced response to a well characterized and largely innocuous molecule with a potential for eliciting heterologous expression of bacterial genes in animal carriers; (iii) the complexity of transcription factors in environmental communities. We created one bacterial transcription factor database (BacTFDB) that was used to train a deep learning model to predict novel TFs and their families in metagenomics and metranscriptomics samples (PredicTF). PredicTF provides the first tool to profile TFs in yet-to be cultured bacteria and it opens the potential to evaluate regulatory networks in complex microbial communities. PredicTF is a flexible, open source pipeline able to predict and annotate TFs in genomes and metagenomes. PredicTF is avaliable at https://github.com/mdsufz/PredicTF.