ATP regulates the function of many proteins in the cell by transducing its binding and hydrolysis energies into protein conformational changes by mechanisms which are challenging to identify at the atomic scale. Based on molecular dynamics (MD) simulations, a method is proposed to analyze the structural changes induced by ATP binding to a protein by computing the effective free-energy landscape (FEL) of a subset of its coordinates along its amino-acid sequence. The method is applied to characterize the mechanism by which the binding of ATP to the nucleotide-binding domain (NBD) of Hsp70 propagates a signal to its substrate-binding domain (SBD). Unbiased MD simulations were performed for Hsp70-DnaK chaperone in nucleotide-free, ADP-bound and ATP-bound states. The simulations revealed that the SBD does not interact with the NBD for DnaK in its nucleotide-free and ADP-bound states whereas the docking of the SBD was found in the ATP-bound state. The docked state induced by ATP binding found in MD is an intermediate state between the initial nucleotide-free and final ATP-bound states of Hsp70. The analysis of the FEL projected along the amino-acid sequence permitted to identify a subset of 27 protein internal coordinates corresponding to a network of 91 key residues involved in the conformational change induced by ATP binding. Among the 91 residues, 26 are identified for the first time, whereas the others were shown relevant for the allosteric communication of Hsp70 s in several experiments and bioinformatics analysis. The FEL analysis revealed also the origin of the ATP-induced structural modifications of the SBD recently measured by Electron Paramagnetic Resonance. The pathway between the nucleotide-free and the intermediate state of DnaK was extracted by applying principal component analysis to the subset of internal coordinates describing the transition. The methodology proposed is general and could be applied to analyze allosteric communication in other proteins. The precise biophysical characterization of the mechanisms of the protein conformational changes controlled by a nucleotide remains a challenge in biology. Molecular dynamics simulations of proteins in different nucleotide-binding states contain information on the nucleotide-dependent conformational dynamics. However, it is difficult to extract relevant information about the conformation-induced mechanism from the raw molecular dynamics data. Herein, we addressed this issue for the major ATP-dependent molecular chaperones Hsp70 s, which contribute to crucial cellular processes and are involved in several neurodegenerative diseases and in cancer. To function, Hsp70 undergoes several conformational changes controlled by the state of its nucleotide-binding domain. We demonstrated that the analysis of the effective free-energy landscape of the protein projected along the amino-acid sequence and computed from the molecular dynamics simulations of Hsp70 in different nucleotide-binding states, holds the key to identify the key residues of the conformational induced pathway. Identification of the key residues involved in the propagation of the structural changes induced by ATP binding offer alternative druggable specific sites other than the ligand binding clefts. The methodology developed for Hsp70 is general and can be adapted to any ligand induced conformational change in proteins.

Structural fluctuations of a protein are essential for the function of native proteins and for protein folding. To understand how the main chain in the native state of a protein fluctuates on different time scales, we examined the rotational correlation functions (RCFs), C(t), of the backbone N-H bonds and of the dihedral angles γ built on four consecutive C(α) atoms. Using molecular dynamics simulations of a model α/β protein (VA3) in its native state, we demonstrate that these RCFs decay as stretched exponentials, ln[C(t)] ≈ D(α)t(α) with a constant D(α) and an exponent α (0 < α < 0.35) varying with the free-energy profiles (FEPs) along the amino acid sequence. The probability distributions of the fluctuations of the main chain computed at short time scale (1 ps) were identical to those computed at large time scale (1 ns) if the time is rescaled by a factor depending on α < 1. This self-similar property and the nonexponential decays (α ≠ 1) of the RCFs are described by a rotational diffusion equation with a time-dependent diffusion coefficient D(t) = αD(α)t(α-1). The present findings agree with observations of subdiffusion (α < 1) of fluorescent probes within a protein molecule. The subdiffusion of (15)N-H bonds did not affect the value of the order parameter S(2) extracted from the NMR relaxation data by assuming normal diffusion (α = 1) of (15)N-H bonds on a nanosecond time scale. However, we found that the RCF does not converge to S(2) on the nanosecond time scale for residues with multiple-minima FEPs.

Structural fluctuations of a protein are essential for a protein to function and fold. By using molecular dynamics (MD) simulations of the model alpha/beta protein VA3 in its native state, the coupling between the main-chain (MC) motions [represented by coarse-grained dihedral angles (CGDAs) gamma(n) based on four successive C-alpha atoms (n - 1, n, n + 1, n + 2) along the amino acid sequence] and its side-chain (SC) motions [represented by CGDAs delta(n) formed by the virtual bond joining two consecutive C-alpha atoms (n, n + 1) and the bonds joining these C-alpha atoms to their respective C-beta atoms] was analyzed. The motions of SCs (delta(n)) and MC (gamma(n)) over time occur on similar free-energy profiles and were found to be subdiffusive. The fluctuations of the SCs (delta(n)) and those of the MC (gamma(n)) are generally poorly correlated on a ps time-scale with a correlation increasing with time to reach a maximum value at about 10 ns. This maximum value is close to the correlation between the (delta(n))t and gamma(n)(t) time-series extracted from the entire duration of the MD runs (400 ns) and varies significantly along the amino acid sequence. High correlations between the SC and MC motions [delta(t) and gamma(t) time-series] were found only in flexible regions of the protein for a few residues which contribute the most to the slowest collective modes of the molecule. These results are a possible indication of the role of the flexible regions of proteins for the biological function and folding.