- Main
Green Cells & Virus Hells: Building Tools for Protein Engineering in Bioproduction and Pandemic-Screening Applications
- Garimella, Shruthi S
- Advisor(s): Shah, Priya S
Abstract
Proteins are the primary molecules to facilitate biological processes at the cellular level. Every cell uses a vast array of proteins in a carefully coordinated concert to orchestrate every function to continue their existence. Proteins very rarely work independently and instead rely on cooperative protein interactions for proper control of chemical reactions and larger biological processes. My dissertation focuses on integrating computational and experimental approaches to study these protein interactions and engineer synthetic tools for bioproduction and public health applications. Engineering tools for protein-based applications have become critical for nearly all biological sciences including research, medical, and agricultural endeavors. Chapters 1-3 discuss inducible gene-expression systems, how we optimize a light-inducible system for mammalian gene expression, and how these systems can be applied to industrial bioproduction applications. Chapters 4-6 focus on understanding the viral characteristics of pandemic-potential, developing a computational pipeline screening virus-host protein-protein interactions, and proposing an experimental workflow to validate the interactions.Artificial control of proteins using light-inducible genetic systems is a relatively new feat of biological engineering. These optogenetic systems use light-responsive proteins to control gene expression, ion channels, protein localization, and signaling with a simple “flip of a switch”. One such tool is the light activated CRISPR effector (LACE) system. Its ability to regulate gene expression in a tunable, reversible, and spatially resolved manner makes it attractive for many applications. We optimized LACE to reduce the number of plasmids needed to deliver all four components from four to two, thus two-plasmid LACE (2pLACE). The behavior of 2pLACE was rigorously tested using optogenetic control of enhanced green fluorescent protein (eGFP) expression as a reporter for gene activation. Using human HEK293T cells, we optimized the ratio of the two plasmids, measured activation as a function of light intensity, and determined the frequency of the light to activate the maximum fluorescence. Overall, the 2pLACE system showed a similar dynamic range, tunability, and activation kinetics as well as less variability as the original LACE system. This simplified optogenetics system will be more amenable to biotechnology applications where variability and resources need to be minimized in low-resource environments or tissue engineering.We also considered how different inducible gene-expression systems might be applied in biomanufacturing. Optogenetic tools, like 2pLACE, can offer improved throughput and protein yield in industrial bioproduction, but it is important to consider how these tools can be applied at scale. Previous work simulated how light propagates in a bioreactor at varying geometries to assess how light-inducible mammalian cell cultures utilize light and activate the system. We experimentally tested these simulations with a bench-top photobioreactor model filled with ExpiCHO cells to measure how light scatters through the culture. We observed little change in light intensity within the center of the system, and only some attenuation of light at the system boundaries. Overall, the experiments supported the results of the light propagation experiments which highlight how optogenetics can be applied in industrial bioproduction.With practical bioproduction in mind we then explored the feasibility of how inducible gene expression systems can be implemented in tissue engineering and cultivated meat applications. However, instead of light, the simple molecule glucose was used as the inducer. A glucose-inducible gene expression system relies on the glucose already present in cell-culture media to activate gene expression using glucose responsive elements in the promoter region. While this promoter design did not strongly activate expression, a similar system could be useful in situations where photobioreactors are not available.Viruses, like living organisms, rely on protein interactions to facilitate their replication. Most viral processes are done by hijacking the host’s proteins through physical interaction, allowing viruses to infect and cause disease. Of particular concern are coronaviruses, which have recently caused three major pandemics, including the COVID-19 pandemic. The size of this virus family and the frequency of coronavirus outbreaks underscore the urgent need to identify emerging strains with the potential for human infection. A critical step in the emergence of a virus is the ability of the viral protein to bind a host receptor and mediate cell entry. For two pandemic-causing coronaviruses (SARS-CoV-1 and SARS-CoV-2), this interaction occurs between the Spike S1 subunit and the human ACE2 receptor. We predicted how 28 emerging coronaviruses may bind to humans using structure predictions, docking programs, and molecular dynamics. After generating the S1 domains for these viruses, we predicted how these proteins might bind ACE2 using pandemic-causing viruses as controls. Contact analysis quantified the predicted binding which highlighted an emerging coronavirus, Zhejiang2013, for which there is no evidence regarding its ability to bind human ACE2. We used molecular dynamics to further validate the docking predictions and model the interaction over time.Overall, our results indicated that Zhejiang2013 exhibits dynamic interaction patterns consistent with ACE2 binding. Combining structure predictions with protein docking and molecular dynamics enables us to screen uninvestigated proteins is a rapid and consistent manner before experimentation. We then proposed how to express and purify these proteins to test their binding affinity to ACE2 using biolayer interferometry. Given that two ACE2-binding coronaviruses have caused global pandemics within the past two decades, identifying and testing potential ACE2 binders is critical for early warning and pandemic preparedness.Together, the culmination of this work highlights the power of studying and manipulating protein interactions to both facilitate desired interactions and inhibit detrimental ones.