Fragile protein folds: Sequence and environmental factors affecting the equilibrium of two interconverting, stably folded protein conformations.

Xingjian Xu,Matthew R Evans,Igor Dikiy,Kevin H Gardner,Leandro P Marcelino

doi:10.5194/mr-2-63-2021

Abstract

Recent research on fold-switching metamorphic proteins has revealed some notable exceptions to Anfinsen's hypothesis of protein folding. We have previously described how a single point mutation can enable a well-folded protein domain, one of the two PAS (Per-ARNT-Sim) domains of the human ARNT (aryl hydrocarbon receptor nuclear translocator) protein, to interconvert between two conformers related by a slip of an internal β-strand. Using this protein as a test case, we advance the concept of a "fragile fold," a protein fold that can reversibly rearrange into another fold that differs by a substantial number of hydrogen bonds, entailing reorganization of single secondary structure elements to more drastic changes seen in metamorphic proteins. Here we use a battery of biophysical tests to examine several factors affecting the equilibrium between the two conformations of the switching ARNT PAS-B Y456T protein. Of note, we find that factors which impact the HI loop preceding the shifted Iβ-strand affect both the equilibrium levels of the two conformers and the denatured state which links them in the interconversion process. Finally, we describe small molecules that selectively bind to and stabilize the wildtype conformation of ARNT PAS-B. These studies form a toolkit for studying fragile protein folds and could enable ways to modulate the biological functions of such fragile folds, both in natural and engineered proteins.

Highlights

Anfinsen’s hypothesis, which states that a protein’s primary sequence encodes a unique fold or conformation, has dominated the study of protein folding for almost 50 years (Anfinsen, 1973)
While designing point mutants to disrupt the interactions between the ARNT PAS-B domain and other binding partners, we fortuitously discovered that the Y456T variant existed in a slow conformational equilibrium between two populated conformations (Evans et al, 2009)
We show that several features of the HI loop region, which precedes the shifting Iβ strand, can influence the WT / SLIP ratio

Summary

Introduction

Anfinsen’s hypothesis, which states that a protein’s primary sequence encodes a unique fold or conformation, has dominated the study of protein folding for almost 50 years (Anfinsen, 1973). One counterexample is the intrinsic disorder found in a significant number of functional proteins and protein regions These intrinsically disordered regions do not adopt a stable three-dimensional structure, instead existing as conformational ensembles of states that may include pre-formed structural nuclei (Dyson, 2016). Other counterexamples are provided by proteins which interconvert among multiple folded states, ranging from a “fragile fold”, in which substitution of a few amino acids – even one – results in the domain co-existing in two states (Evans et al, 2009; Evans and Gardner, 2009; Ha and Loh, 2012) to the general concept. Xu et al.: Effects of ligands and other factors on a fragile PAS fold of a “metamorphic protein”, which is one that can reversibly adopt different stable folds in different environmental conditions (Murzin, 2008)

Methods

Results

Discussion

Conclusion