Modeling Metabolism and Regulation to Facilitate Strain Characterization and Design

Open Access
Author:
Mueller, Thomas J
Graduate Program:
Chemical Engineering
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
September 23, 2016
Committee Members:
  • Costas D Maranas, Dissertation Advisor
  • Costas D Maranas, Committee Chair
  • Phillip E Savage, Committee Member
  • Manish Kumar, Committee Member
  • James Gregory Ferry, Outside Member
Keywords:
  • Optimization
  • Cyanobacteria
  • Systems Biology
  • Computational Biology
  • Metabolic Modeling
Abstract:
Computational models of metabolism and regulation can provide insights into not only existing cellular processes, but also serve as a platform for developing strain design strategies to achieve a desired goal. The work discussed here focuses on the development of models for organisms with native capabilities that make them appealing candidates for industrial application, most notably cyanobacteria. Cyanobacteria are photosynthetic prokaryotes that have been explored for the production of a wide variety of compounds, but suffer from several drawbacks as compared to existing industrial strains including a limited knowledgebase and slower growth rates. The first part of this dissertation focuses on expanding that knowledgebase through the development of genome-scale metabolic models for several model cyanobacteria and the introduction of a semi-automated model generation workflow to expedite accurate model development. This workflow is then applied to Synechococcus UTEX 2973, a fast growing cyanobacterium, in order to better understand possible causes for its fast-growth phenotype. The regulatory response of an organism to its environment is an important factor in understanding an organism. The second part of this dissertation targets the incorporation of gene expression data into metabolic models to refine flux predictions, as well as the development of regulatory networks for several cyanobacteria over the diurnal cycle. The CoreReg method applies additional constraints based on gene expression data on a genome-scale metabolic model, and allows for the identification of specific reactions, the effects of whose regulation propagates throughout the larger network. These reactions can be targeted to alleviate the effects of a stressor on the organism’s metabolism. The creation of regulatory networks over the complete diurnal cycle facilitates the identification of transcription factors native to an organism that can be coopted to control other processes. This was done to select several transcription factors to control processes in the model cyanobacterium Synechocystis 6803 to accommodate the introduction of nitrogen fixation.