Year of Award

2023

Document Type

Dissertation

Degree Type

Doctor of Philosophy (PhD)

Degree Name

Biochemistry & Biophysics

Department or School/College

Department of Chemistry and Biochemistry

Committee Chair

Bruce E. Bowler

Commitee Members

Mark L. Grimes, Stephen R. Sprang, Travis J. Wheeler, Travis S. Hughes

Keywords

free energy, protein folding

Publisher

University of Montana

Abstract

Proteins are large, flexible molecules with an extremely large number of potential conformations. Proteins expressed in cells traverse available conformations to reach a consistent, thermodynamically stable, biologically active structure through a process known as protein folding. The atomic composition of the protein, defined by a sequence of amino acid residues encoded in DNA as a gene, determines the protein folding pathway and ultimate native structure of the protein molecule. Understanding the relationship between the sequence of amino acids and the resulting protein structure has been a central challenge in protein research for decades. To fill this knowledge gap, we test the hypothesis that the distribution of conformers observed for a short protein sequence across all known protein structures reflects that sequence's intrinsic structural properties. Qualitative and quantitative predictions based on our model are tested against experimental data for protein stability and folding pathways.

Replica-exchange Monte Carlo simulations, data mining of the Worldwide Protein Data Bank (wwPDB), analysis of published protein stability data, thermodynamic and kinetic folding experiments, and Xray crystallography were used to characterize the structural properties of amino acid sequences. The role of turn sequences in guiding the protein folding process was extensively characterized by the combined methods. Turn composition, structural preferences, and cooperation with neighboring residues determined whether a turn had an active, passive, or counter-active role in a protein's folding process.

Proline-rich turns, NPSNP and KPSDP, from the two-helix bundles found in bacterial type III secretion system needle proteins form native-like structure early in the folding process. Each of these

turns are flanked by sequences with very high helix propensity that, when oriented by the turn, can actively nucleate the hydrophobic core of the protein. The hydrophobic turn, MGYE, from the three-helix bundle UBA(1) also forms native-like structure early in the folding process. This turn structure places the Met (M) and Tyr (Y) residues together, nucleating the hydrophobic core of UBA(1). These two residues can then stabilize the adjacent helices to form a Helix- Turn-Helix structure. The second, proline-containing turn in UBA(1), ASYNNP, forms non-native structure early in the folding process. This turn restructures late in the folding process when the third helix docks to the previous Helix-Turn-Helix structure. Each of the active turns characterized (NPSNP, KPSDP, and MGYE) direct the folding process by nucleating the protein's hydrophobic core.

A general purpose computational method to model the local structural properties of protein sequences was developed from data mined from the wwPDB. Turn mechanisms can be rapidly characterized using the tool, EmCAST, in conjunction with a PDB structure of the protein of interest. The impact of surface mutations on protein stability can also be scored by EmCAST. Models and calculations were extensively validated against experimental data for multiple protein and peptide systems. Calculations for stabilizing mutations at well-structured positions in UBA(1) produced a near perfect correlation with experimental measurements (R2 = 0.97). A user-friendly web interface to the software was developed to share the method with other protein researchers. Our model provides key insights into the protein sequence/structure relationship that can be used to characterize protein surface stability, identify regions with dynamic structure, and predict protein folding intermediates.

Download

Share

COinS
 

© Copyright 2023 Michael Tyler Rothfuss