Date of Award
Doctor of Philosophy (PhD)
Ryan T. Gill
Stephanie J. Bryant
Corrella S. Detweiler
Robin D. Dowell
Many critical societal issues could be alleviated via advances in biological engineering, specifically those bolstering our ability to design complex phenotypes like bacterial tolerance, resistance, and chemical production. Complex phenotypes are controlled through overlapping regulatory mechanisms encompassing multiple systems, including metabolism, cellular structure, energy consumption, and more. Moreover, immense heterogeneity within and across bacterial populations makes it challenging to discern key genes and pathways contributing to phenotypes of interest. Thus, to enable engineering of complex bacterial phenotypes, it is essential to consider unconventional perspectives from which to identify targets.
Here, we describe efforts concentrated on non-genetic data to ascertain key genes, pathways, and behaviors underlying Escherichia coli resistance, tolerance, and production. We perform adaptation experiments in laboratory strain E. coli to identify gene expression signatures causal to adaptive resistance, a transient response that enables stepwise tolerance increases. Adaptation experiments are designed to generate divergence and provide a diverse set of populations in which common trends point to general adaptive resistance. We find that differential gene expression variability enables identification of significant disparities between transcriptome profiles in unadapted and adapted populations, with lower variability genes having more impact on adaptation. We investigate the role of unknown genes displaying significant variability shifts and perturb the expression of identified genes to demonstrate the presence of synergistic and antagonistic interactions in combination with antibiotic treatment. Next, we investigate the non-genetic basis for partial carbapenem-resistance detected in a clinically isolated multidrug-resistant E. coli. mRNA, antisense RNA, and small RNA are recognized that distinguish the response to carbapenems with differing susceptibility. Then, we build a database of hundreds of tolerance-associated transcriptome studies to facilitate learning from prior research efforts. We leverage the database to identify universal gene expression signatures promoting tolerance, and to locate genes and regulators associated with specific stress conditions. Finally, we perform genome-scale metabolic network analysis to explore the impact of variability in gene expression on metabolism and overproduction of ethylene from a heterologous pathway. In all, this work exhibits the power of non-genetic data for providing insight into complex bacterial phenotypes.
Erickson, Keesha Elizabeth, "Novel Approaches to Bioengineering Target Identification: A Focus on Non-Genetic Contributions to Complex Bacterial Phenotypes" (2017). Chemical & Biological Engineering Graduate Theses & Dissertations. 109.