Order statistics inference for describing topological coupling and mechanical symmetry breaking in multidomain proteins

TL;DR

This paper introduces a new order statistics-based theory to analyze how mechanical forces induce coupling and symmetry breaking between protein domains during unfolding, revealing force-dependent topological interactions.

Contribution

The study develops a novel theoretical framework inspired by order statistics to characterize force-induced domain coupling in multi-domain proteins, supported by GPU-accelerated simulations.

Findings

Mechanical symmetry breaks under increased tension.

Unfolding transitions become more correlated at higher forces.

Force induces topological coupling between protein domains.

Abstract

Cooperativity is a hallmark of proteins, many of which show a modular architecture comprising discrete structural domains. Detecting and describing dynamic couplings between structural regions is difficult in view of the many-body nature of protein-protein interactions. By utilizing the GPU-based computational acceleration, we carried out simulations of the protein forced unfolding for the dimer WW-WW of the all-beta-sheet WW domains used as a model multidomain protein. We found that while the physically non-interacting identical protein domains (WW) show nearly symmetric mechanical properties at low tension, reflected, e.g., in the similarity of their distributions of unfolding times, these properties become distinctly different when tension is increased. Moreover, the uncorrelated unfolding transitions at a low pulling force become increasingly more correlated (dependent) at higher forces. Hence, the applied force not only breaks "the mechanical symmetry" but also couples the physically non-interacting protein domains forming a multi-domain protein. We call this effect "the topological coupling". We developed a new theory, inspired by Order statistics, to characterize protein-protein interactions in multi-domain proteins. The method utilizes the squared-Gaussian model, but it can also be used in conjunction with other parametric models for the distribution of unfolding times. The formalism can be taken to the single-molecule experimental lab to probe mechanical cooperativity and domain communication in multi-domain proteins.