Georgia Institute of Technology
Residue Interaction Network Approach to Conformational Dynamics and Enzyme Evolution in β-Lactamases and Protein Tyrosine Phosphatases
Abstract
dc:description.abstractEnzyme catalysis is closely connected to conformational dynamics, which can be described in terms of the formation and breaking of non-covalent interactions. Analyzing these interaction networks from molecular dynamics simulations and characterizing their conservation or remodeling across evolutionarily related proteins can yield physical insight relevant to protein engineering and enzyme specificity. This thesis describes the development of tools that enable characterization and further processing of these interaction networks and their application in two enzyme families: class A β-lactamases and protein tyrosine phosphatases (PTPs). Key Interaction Networks (KIN) is a framework for identifying conserved non-covalent interactions across protein families from crystal structures or molecular dynamics trajectories. Combined with Key Interactions Finder (KIF), which prioritizes interactions by their correlation with changes in protein dynamics, this provides a set of tools for systematic comparison of interaction architectures across related enzymes. This combination of tools was used to characterize catalytically relevant dynamic features of class A β-lactamases. In particular, the analysis identified interactions that dictate specificity differences between ancestral enzymes that function as substrate generalists and modern enzymes that act as specialists. These techniques were further integrated with other computational approaches, including enhanced sampling and empirical valence bond theory, to characterize catalytic loop dynamics in archaeal PTPs. This work showed that these enzymes challenge the conventional view of PTP dynamics as primarily regulated by acid loop motion. Instead, the archaeal PTPs feature a flexible P-loop whose dynamics are modulated by the protonation states of active-site residues. The same regulation is observed in ShufPTP, an engineered archaeal PTP generated by sequence shuffling, which combines extreme thermostability, enhanced P-loop flexibility, and mechanistic promiscuity. Interaction network analysis was used to identify key residues responsible for P-loop dynamics and isolate the differentiating hydrophobic network responsible for emergent properties of ShufPTP. Together, these studies demonstrate that interaction network analysis can identify evolutionarily conserved, catalytically relevant patterns across protein families and inform protein engineering strategies.
Degree
thesis:*- Name thesis:degree_name
- Chemistry, PhD
- Department thesis:degree_department
- Chemistry and Biochemistry
- Grantor
- Georgia Institute of Technology
- Year dc:date.issued
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Yehorova, Dariia
- Advisor dc:contributor.advisor
-
- Kamerlin, Lynn
- Committee members dc:contributor.committeemember
-
- Sherrill, David
- Gumbart, JC
- Lieberman, Raquel
- Kretchmer, Joshua
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/1853/81781