Cellular metabolism is one of the most fundamental systems for any living organisms, involving thousands of metabolites and reactions that forms large interconnected metabolic networks. Proper and comprehensive understanding of the metabolism in human cells has been a field of research for a long time. One of the key parameters in understanding the metabolism are the metabolic fluxes, which are the rates of conversion of metabolic intermediates. Currently, one of the main approaches for determining these fluxes is metabolic flux analysis (MFA), in which isotope-labelled compounds are introduced into the system and measured. Mathematical models are then used to calculate a prediction of the systems flux configuration. However, the current paradigm of MFA lack established methods for validating that a model can accurately predict quantities for which there are no experimental data. In this study, a model for the central human metabolism was created and evaluated with regards to the model’s ability to predict a validation dataset. Further, an uncertainty analysis of these predictions were performed with a prediction profile likelihood analysis. This study has conclusively shown that MFA models can be validated against experimental data that the model has never seen before. Additionally, such model predictions were shown to be observable with a well determined prediction uncertainty. These results shows that a systematic validation of MFA models is possible. This in turn allows for a greater trust to be placed in the models, and in any conclusions that are based on such models.

The effects of climate change have made the need to develop sustainable production practices for biofuels and other chemicals imminent. The development of the green economy has also led to many industries voluntarily improving the sustainability of the products they produce. The microbial production of fatty acid-derived chemicals allows for the opportunity to reduce petroleum-based chemicals in the marketplace. However, for microbial produced chemicals to be industrially competitive, significant work is needed to improve the production capacity of industrial strains. There are a number of bottlenecks and challenges related to the production of various fatty acid derivatives that need to be addressed. One of these key challenges relates to the source of the fermentation feedstock. While sources such as corn or sugar cane are currently common, these feedstocks compete with food supply and require nutrient-rich soils. The use of lignocellulosic feedstocks is preferred to combat this issue, however these feedstocks present their own unique challenges. Pretreatment is required to release fermentable sugars, and this process also results in various fermentation inhibitors released into the solution. A better understanding of how engineered strains utilize these fermentable sugars as well as improving resistance to the inhibitors will help to improve the chemical production capacity of these chemical products. This work will focus on describing key bottlenecks related to fatty acid-derived products, while also evaluating proposed solutions to these bottlenecks. / Master of Science / Currently, many common household products and plastics are developed using petroleum-based components. From plastic bottles to common cosmetics, these contain ingredients that are derived from petroleum. In order to combat our reliance on petroleum for these every day products, it is essential to develop alternate sources for these materials. A potential source involve using plant material and by-products to produce these same compounds that we are able to produce from petroleum. While there has been significant research to produce useful products such as bioethanol from corn, this is not an ideal crop. Corn requires more water and space than other crops such as grasses. In addition, these grasses can grow in soil that food crops are unable to grow in, so we don’t utilize valuable land to develop common household products. However, these grasses are much more difficult to treat and process in order to form these basic chemical ingredients. In order to use grass-based crops, it is possible to engineer organisms such as yeast to process the raw material into valuable chemical precursor. This work aims to genetically engineer yeast in order to produce some of these chemical precursors from a grass-like feedstock. In addition, this work also analyzes how physical characteristics of yeast affect the final product formation. Finally, a model was developed to show how yeast ferments corn-like and grass-like feedstocks differently.

Le contexte de ce projet de thèse était d’approfondir la compréhension du métabolisme de Trypanosoma brucei. Les trypanosomes utilisent différents types de sources de carbone, des hydrates de carbone ainsi que des acides aminés pour alimenter leurs besoins énergétiques et biosynthétiques (conditions imitant réellement l'environnement dans la mouche tse-tse). Les différences de thioesters d'acyl-CoA sont encore inconnues dans ces conditions. Une telle élucidation est essentielle pour comprendre les adaptations métaboliques de l'organisme au cours de son cycle de vie. Cet objectif pourrait être complété par une combinaison d'analyses sensibles de divers groupes de métabolites, de délétions dirigées de gènes ou de régulations négatives. Ces derniers développements intègrent un flux de travail complet d'analyse des flux métaboliques par 13C à l’état-instationnaire. Ce flux de travail combine les méthodes existantes pour la collecte d'échantillons, la métabolomique quantitative basée sur MS et l'analyse isotopique d'acides organiques, d'acides aminés, de composés phosphorylés en plus des thioesters d'acyl Coenzyme A (acyl-CoAs), qui représentent un point central entre le métabolisme central du carbone et les voies anaboliques. Ce flux de travail a d'abord été évalué et validé sur l'organisme modèle Escherichia coli et a fourni de nouvelles idées sur son fonctionnement métabolique. Par la suite, ce flux de travail a ensuite été exploité pour étudier le métabolisme de T. brucei, pour lequel les résultats préliminaires sont décrits et discutés dans cette thèse. / Unusual metabolism of protozoan parasite causing deadly sleeping sickness, Trypanosoma brucei, has been enigmatic for many years. In the past decades, targeted genetic perturbations combined with metabolic analysis have advanced the view on complex compartmentalized metabolism of this organism, but acyl-CoA metabolism on the crossroad between catabolic and anabolic pathways, remains largely uncharacterized. Present work aims at clarifying mitochondrial operation and topology of acyl-CoA network of T. brucei, as well as its interconnections with the rest of metabolism. This has required the development of a complete framework for investigation of acyl-CoA metabolism in T. brucei integrating isotope labeling experiments with metabolite quantification. Sensitive LC-MS method for identification and quantification of acyl-CoAs based on high-resolution mass spectrometry (HRMS) with LTQ-OrbiTrap has been established and applied to investigate acyl-CoA metabolism in the protozoan parasite, as well as in the model organism in systems and synthetic biology, Escherichia coli. Complete workflow from cell cultivation, measurement of extracellular fluxes and analysis of isotopic profile which is result of enzyme-specific incorporation of isotopic tracer allowed modelling of metabolic network and calculation of metabolic fluxes. The entire workflow has been biologically validated and has clarified the link between acyl-CoA and central carbon metabolism in E. coli. The proposed framework has been adapted to T. brucei, for which several sample collection methods have been evaluated thoroughly. It was possible to extract, identify and quantify main acyl-CoA species produced from glucose catabolism. This optimised setup for acyl-CoA analysis will allow collection of data for NMR-based analysis of metabolic end products as well as collection of intracellular metabolites from same sample.