Predicted dynamical couplings of protein residues characterize catalysis, transport and allostery

Abstract

Motivation Protein function is intrinsically linked to native dynamics, but the systematic characterization of functionally relevant dynamics remains elusive besides specific examples. Here we exhaustively characterize three types of dynamical couplings between protein residues: co-directionality (moving along collinear directions), coordination (small fluctuations of the interatomic distance) and deformation (the extent by which perturbations applied at one residue modify the local structure of the other one), which we analytically compute through the torsional network model. Results We find that ligand binding sites are characterized by large within-site coordination and co-directionality, much larger than expected for generic sets of residues with equivalent sequence distances. In addition, catalytic sites are characterized by high coordination couplings with other residues in the protein, supporting the view that the overall protein structure facilitates the catalytic dynamics. The binding sites of allosteric effectors are characterized by comparably smaller coordination and higher within-site deformation than other ligands, which supports their dynamic nature. Allosteric inhibitors are coupled to the active site more frequently through deformation than through coordination, while the contrary holds for activators. We characterize the dynamical couplings of the sodium-dependent Leucine transporter protein (LeuT). The couplings between and within sites progress consistently along the transport cycle, providing a mechanistic description of the coupling between the uptake and release of ions and substrate, and they highlight qualitative differences between the wild-type and a mutant for which chloride is necessary for transport. Availability and implementation The program tnm is freely available at https://github.com/ugobas/tnm Supplementary information Supplementary data are available at Bioinformatics online.

Structural basis for allosteric transitions of a multidomain pentameric ligand-gated ion channel

“…Normal Mode Analysis. Dynamical couplings between atoms in four functional sites were computed as previously described (30), adopting the normal modes of the torsional network model (61). Briefly, 1) directionality coupling was defined as the Boltzmann average of the scalar product between the directions of motion of two atoms; 2) coordination coupling was a constant minus the root mean square of the fluctuation of the distance between the two atoms; 3) deformation coupling was defined as the deformation in one atom produced by a perturbation applied to the other with constant amplitude and direction, such that the perturbation is maximized.…”

“…Given the evident relevance of both TMD and LBD sites to allosteric gating transitions in this family, we qualitatively assessed the coupling between DeCLIC domain motions using normal mode analysis. A new method by Bastolla and colleagues (30) has been developed specifically to characterize coupling of ligand-binding sites based on similarities in the direction of motions (codirectionality), fluctuations in interatomic distances (coordination), and how much perturbations in one site modify the structure of another (deformation). Torsional normal mode analysis predicted atomic fluctuations that correlated well with B-factors in both DeCLIC structures (correlation factors of 0.94 and 0.88 for the apparent closed and open states, respectively) (SI Appendix, Fig.…”

Analyzing In Silico the Relationship Between the Activation of the Edema Factor and Its Interaction With Calmodulin

“…As already quoted in the introduction, several approaches for prediction of allosteric sites are based on a measure of the deformation of the protein described by an elastic network (Panjkovich and Daura, 2014;Guarnera and Berezovsky, 2016) or described by a normal mode perturbation (Greener and Sternberg, 2015). Also, a recent analysis of a large set of protein structures containing ligands showed (Alfayate et al, 2019) that the binding sites of allosteric ligands display larger deformations. Similarly, several bioinformatics approaches predict allosteric pockets as the ones on which ligand binding induces the largest variations in protein structures (Panjkovich and Daura, 2014;Guarnera and Berezovsky, 2019).…”

Dynamical Correlations Reveal Allosteric Sites in G Protein-Coupled Receptors

“…For instance, the PARS [17] and SPACER [18] programs calculate the effect of the perturbation of pockets on the vibrational normal modes of a protein, identifying as potential allosteric sites those leading to significant changes in the collective dynamics when perturbed. Other methods measure correlations between coordinates [19,20] or volumes [21] of distinct sites, assuming that allosteric sites are dynamically coupled to the regions they regulate. Although these approaches may provide insights into mechanisms of allosteric modulation, and in spite of the importance of collective motions for GPCR function, a detailed account of the dynamical couplings of binding sites in GPCRs is still lacking.…”