Date of Graduation
Biostatistics, Bioinformatics and Systems Biology
Doctor of Philosophy (PhD)
Gregory S. May
Michael C. Lorenz
Matthew R. Bennett
Synthetic biology aims to build biological devices to understand living systems and explore new applications. Synthetic gene circuits such as genetic switches, oscillators and logic gates are at the core of many synthetic biology applications. These gene circuits often include a sensor/regulator protein capable to detect small molecules and then transduce them into a regulatory signal to generate measurable output. Similar signal transduction networks are also abundant in nature. However, in many natural and engineered scenarios, the output also affects the regulator/sensor protein. How such interactions between the regulator/sensor and the output affect synthetic gene circuit function has not been investigated. In order to address this question, I took advantage of Saccharomyces cerevisiae synthetic gene circuits built previously in our laboratory: Negative Regulation (NR), Negative Feedback (NF) and Positive Feedback (PF). Previous research had characterized the behavior of these gene circuits at various inducer (anhydrotetracycline, ATc) concentrations when they controlled the bifunctional Zeocin Resistance gene (ZeoR) fused to the reporter yEGFP. In these gene circuits, yGFP::ZeoR was a passive target, which did not interact with its upstream transcriptional regulator. In order to study the effect of an active target on gene network dynamics, I replaced yEGFP::ZeoR with PDR5::GFP to create three new gene circuits, NRpump, NFpump and PFpump. The PDR5 gene produces a multidrug resistance pump that belongs to the ATP-binding cassette protein family. Once Pdr5 is expressed, it pumps out various small molecule chemicals including the inducer, altering the activity of its upstream transcriptional regulators, and thereby creating a feedback loop. Therefore, these reconfigured gene circuits enabled the investigation of the question: how the protein pump alters the characteristics of the original NR, NF and PF gene circuits.
In this dissertation, I show that the dose response behavior of the NRpump, NFpump and PFpump gene circuits differs from their non-pump counterparts. Studying gene circuits controlling non-functional PDR5 mutants indicated that the efflux pumping activity of Pdr5 caused loss of linearity in NFpump compared to NF dose-response. However, the dose-response behavior of NRpump and NFpump with the PDR5 mutant still differed from the behavior of the original NR and NF gene circuits. With the help of stochastic models developed by my collaborator, I hypothesized and then proved experimentally that lower expression level of the regulator, TetR, in all NRpump and NFpump strains (both with functional and non-functional Pdr5) compared to NR and NF, should be responsible for the remaining dose response differences. Similar to the other pump-controlling gene circuits, the PFpump gene circuit had a more sensitive dose-response compared to the original PF. Although both gene circuits produced bimodal distributions, the finesses and cellular transition rates between the two subpopulations were different.
Finally, I tested the evolution of non-induced NRpump, NFpump and PFpump strains in a fluconazole-containing environment. While PFpump cells maintained fluconazole sensitivity, NRpump and NFpump cells started to develop fluconazole resistance after 48 hours. Expression of Pdr5 was the cause of resistance. The elevated Pdr5 expression level remained the same after fluconazole removal, suggesting mutational breakdown of these gene circuits. However, bimodal expression patterns evolved in some NRpump and NFpump cell cultures after 256 hours in fluconazole environment suggesting that other mutations might have occurred besides those causing gene circuits’ breakdown.
Synthetic Biology, Synthetic gene circuits, Genetics, Computational Biology, Microbiology