Sebesta, Jacob, authorPeebles, Christie A. M., advisorPeers, Graham, committee memberPrasad, Ashok, committee memberReardon, Kenneth, committee member2021-06-072022-06-022021https://hdl.handle.net/10217/232566Metabolic engineering is developing into a field that can change the way we produce a wide variety of valuable chemicals. Many chemicals are already produced in microbial cultures. Metabolic engineering enables us to modify organisms to produce metabolites they don't usually produce, assuming an enzyme can be identified in another organism that catalyzes the formation of that product (or an enzyme can be designed for that task through protein engineering). The distribution of accumulated metabolites can also be altered. There are some cases where metabolites can be accumulated through cultivation practices. Methods of metabolic engineering to overexpress, knockdown, or knockout native enzymes provide additional tools to alter cellular metabolism and drive accumulation of those products. Precise control over gene expression is central to these efforts. To avoid competition with human food crops and the resources need to produce them, cyanobacteria may be utilized for production of valuable chemicals. Through photosynthesis, they can utilize carbon dioxide from geological formations or from industrial waste streams. Since most metabolic engineering has been developed in E. coli and yeast, it was necessary to first adapt the basic methods for use in cyanobacteria. Along with my co-authors Dr. Allison Werner and Dr. Christie Peebles, we reviewed methods for producing genetically modified Synechocystis Sp. PCC6803 (S. 6803). To facilitate the generation of strains with many modifications, we covered the method developed in the Peebles Lab for making markerless selections which remove any antibiotic selection markers. A previous graduate student in the Peebles lab, Stevan Albers, found that strong promoter-ribosome binding site combinations that drove high expression of GFP did not necessarily result in high expression when used to drive expression of a different gene. Therefore, in our work to produce bisabolene in S. 6803 we tested many ribosome binding sites. In addition, we tested five different codon optimizations of the bisabolene synthase to ensure that expression was not prevented by slow translation elongation. We found that the simple measure of the codon adaptation index (CAI) correlated with expression of the five different codon optimizations. Using a thermodynamic model of translation initiation, we designed ten ribosome binding sites to increase bisabolene synthase expression by 10-fold. Only one of those designs actually approached a 10-fold increase, highlighting the need to continue testing several ribosome binding sites to achieve a desired expression level. Since industrial cultivation of cyanobacteria occurs outdoors, subject to natural light:dark cycles, we tested two of the designed strains in light:dark cycles. The strains reached similar bisabolene titers after being exposed to the same amount of total light period as those previously tested in continuous light. Overall, this work increased the highest bisabolene titer reported in cyanobacteria by approximately 10-fold. The need to test many ribosome binding sites limits progress in cyanobacterial metabolic engineering. The research of others suggest that ribosome binding sites interact with coding sequences by forming secondary structures with different free energy of folding. The estimation of the free energy of folding may be inaccurate, and, further, the kinetics of such folding may also be important to translation initiation rates. We tested two different designs to limit the impacts that secondary structures that span either side of the start codon may have on translation initiation rates in both E. coli and S. 6803. Utilization of a 21-nucleotide leader sequence after the start codon to make the sequence context consistent for ribosome binding sites between different coding sequences did not improve the correlation found between the expression of two different reporter genes in either organism. Bicistronic designs use translational coupling between an upstream open reading frame and the gene of interest with a ribosome binding site contained within the upstream open reading frame to re-initiate translation. This design exploits the helicase activity of ribosomes in elongation mode to actively unfold the secondary structure around the start codon of the gene of interest. We expected this activity to reduce the impacts of secondary structure and improve the correlation in expression between two different reporter genes. Intriguingly, the correlation was much improved in E. coli, but not in S. 6803. Together, this dissertation suggests that there are important differences in translation initiation between E. coli and S. 6803. Improved ribosome binding site design for cyanobacteria would facilitate further increases in terpenoid production both by enabling higher expression of heterologous terpenoid synthases and by reducing the number of strains that must be tested to achieve the desired expression level for each enzyme. Future directions suggested by this work include studies of translation initiation mechanisms in cyanobacteria, development of cell-free expression systems to facilitate rapid testing of many different genetic constructs, and further efforts at pathway engineering to increase terpenoid titer and productivity in cyanobacteria.born digitaldoctoral dissertationsengCopyright and other restrictions may apply. User is responsible for compliance with all applicable laws. For information about copyright law, please see https://libguides.colostate.edu/copyright.metabolic engineeringtranslation initiationSynechocystis sp. PCC6803bisaboleneText