Maly, Dustin JRose, John Christopher2017-08-112017-08-112017-08-112017-06Rose_washington_0250E_17285.pdfhttp://hdl.handle.net/1773/40263Thesis (Ph.D.)--University of Washington, 2017-06Cells continuously sense both their external and internal environments, integrate diverse and often conflicting information, and respond. The potential responses are remarkably varied: it could be to kill an invading pathogen, repair damaged DNA, proliferate to heal a wound, or undergo programmed cell death. The process of sensing, integrating, and responding to information is carried about by complex biochemical networks within the cell. These cellular networks dictate a cell’s behavior, function, and identity. Moreover, dysregulation of these networks is a hallmark of diseases as prevalent and damaging as cancer and diabetes. It is imperative that we advance our understanding of these cellular networks in order to improve treatments for when these networks go awry. Unfortunately, our ability to adequately investigate these networks is hampered by their sheer complexity. Inherent to cellular networks are multiple layers of complexity: they are complex in time, in space, and in their architecture (i.e. the arrangement of the interactions that comprise the network). Recently, high-throughput methods have greatly improved our ability to observe these networks, allowing the characterization of thousands of genes or proteins in a single experiment. Yet, our ability to interrogate cellular networks has not kept pace with our ability to observe them. As a result, studies often yield conclusions which are largely phenomenological with limited mechanistic insight. Mechanistic insights are critical to developing novel therapies or better utilizing existing ones. To gain such insights would require tools that confer tunable control of individual network components with spatial and temporal precision, allowing systematic dissection of a network. Up to this point, the development of such tools, generally engineered proteins that are controlled by small molecules or light, has been hampered by the reliance on empirical protein engineering strategies that are inefficient, arduous, and costly. To address this outstanding problem, we developed a computational framework for the systematic design of small-molecule controlled proteins. This framework utilizes protein design tools developed by David Baker’s lab, and we show it greatly expedites the development process. Using this framework, we engineered Chemically Inducible Activator of RAS (CIAR), which enables tunable, spatiotemporally precise control of RAS activation. RAS is frequently hyperactivated in human cancers, and using CIAR we characterized dynamic features or RAS biology which were inaccessible to previous methods. For instance, we found that RAS signaling kinetics differ between cell lines, which may reflect the divergent propensities of different cell types to develop RAS-driven cancers. We also show that RAS signaling can be rewired by small molecule inhibitors currently used to treat melanoma, which sheds light on a phenomenon where this drug actually promotes growth of secondary cancers. More recently, we have demonstrated that we can control RAS activation at different subcellular locations, such as the Golgi. It has been suggested that the subcellular location of RAS activation can dictate phenotypic outcomes, including whether a developing T-cell proliferates or undergoes apoptosis. CIAR will enable unprecedented examinations of such phenomena. In parallel, we took a similar approach to study the processes involved in genome engineering. The development of CRISPR/Cas9 for genome editing has led to intense interest in both its potential research and therapeutic applications. Yet, little is known regarding the dynamics of Cas9-mediated DNA cleavage and subsequent DNA repair. Using an approach analogous to that used to develop CIAR, we engineered a rapidly inducible Cas9 variant, chemically-inducible Cas9. Additionally, we developed the first assay for quantitative, temporally-resolved monitoring of double-strand breaks (DSBs), DSB-ddPCR. Using these two technologies, we conducted a first-ever examination of Cas9-mediated DNA cleavage and repair dynamics. We found that Cas9 cleavage is rapid, and that both cleavage and repair kinetics differ between loci. We envision these technologies will enable in-depth examinations of this heretofore unexplored region of CRISPR/Cas9 biology, with potential implications for its applications in research and the clinic. Taken together, the technologies developed in the course of this dissertation demonstrate the utility of precision tools for the study of complex cellular networks. CIAR and ciCas9 have already begun to yield insights into intracellular signaling and genome engineering and are the foundation for several ongoing investigations. Beyond these specific tools, the design methods and engineering strategies we have devised may have greater impact. It is our hope that they will aid the development of future technologies to probe the inner workings of our cells, deepening our understanding of the networks which drive human health and disease.application/pdfen-USnoneChemical BiologyCRISPR/Cas9Genome EngineeringIntracellular SignalingProtein DesignRASMolecular biologyBiochemistryCellular biologyMolecular and cellular biologyThesis