COMPUTATIONAL REDESIGN OF EZYMES AND CHANNEL PROTEINS

Open Access
- Author:
- Ghaffari, Soodabeh
- Graduate Program:
- Chemical Engineering
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- August 02, 2021
- Committee Members:
- Costas D Maranas, Thesis Advisor/Co-Advisor
Scott Thomas Milner, Committee Member
Phillip Savage, Program Head/Chair
Esther Gomez, Committee Member - Keywords:
- computational protein designs
channel proteins
Rosetta
IPRO
enzyme engineering
optmization - Abstract:
- Proteins are biological macromolecules ubiquitous in nature and have diverse functions, including inducing immune responses through antibodies, regulating intracellular ion concentrations, and catalyzing chemical reactions using enzymes. The enzyme's application scope is restricted by the range of reactions that may be catalyzed and by performance parameters, such as stability and kinetics. Protein engineering offers a way to obtain enzymes that are better at catalyzing non-natural reactions. Unfortunately, while purely experimental approaches to protein engineering are widely available, they are either arduous or infeasible for more complicated systems. Computational protein design is proving to be a highly efficient method for designing proteins for specific biotechnological purposes. When applied to existing enzymes, computational redesign enables orders of magnitude increase in catalytic activity toward a novel substrate. Additionally, computational tools allow the construction of entirely novel active sites capable of catalyzing processes not before observed in biological systems. This thesis describes the use of computational protein design tools, Iterative Protein Redesign and Optimization (IPRO) and Rosetta software suite to redesign a) channel proteins (E.coli Outer membrane protein F) for maximizing water vapor transport rates while eliminating chemical warfare agents (CWAs) passage to generate CWA-protective fabrics, b) trans-enoyl-CoA reductase which involves in fatty acid biosynthesis to alter their specificity for medium-chain fatty acids production. Additionally, this thesis presents a novel version of IPRO in which destabilizing mutations are excluded from selection during protein redesign. IPRO is a computational protein engineering tool that incorporates random backbone perturbations with deterministic amino acid selection procedures, followed by several steps to refine the protein-ligand structure, to improve or reduce binding for a given target or multiple targets. Rosetta is a unified software package for protein structure prediction and functional design. It explores the relevant conformational/sequence space, primarily using Metropolis Monte Carlo sampling approaches. It then assesses the generated structural models using a scoring system that balances the physics and knowledge-based energy terms.