Motions of the SecA protein motor bound to signal peptide: Insights from molecular dynamics simulations

Abstract

SecA is an essential part of the Sec pathway for protein secretion in bacteria. In this pathway, SecA interacts with the N-terminal fragment of the secretory protein - the signal peptide, and couples binding and hydrolysis of adenosine triphosphate with movement of the secretory protein across the SecY protein translocon. How interactions with the signal peptide alter the conformational dynamics and long-distance conformational couplings of SecA is a key open question that we address here with molecular dynamics techniques. Analyses of protein motions indicate that the signal peptide alters SecA dynamics not only at the site where this peptide binds, but also at a nucleotide-binding domain. Hydrogen bond clusters contribute to the long-distance propagation of changes in SecA dynamics.

Bridge: A Graph-Based Algorithm to Analyze Dynamic H-Bond Networks in Membrane Proteins

“…In order to visualize graphs of H-bond networks from MD simulations, we first performed Principal Component (PCA) Analyses to project onto a two-dimensional (2D) plane the 3D Cartesian coordinates of the amino acid residues participating in the network. We thus generated the covariance matrix of the atomic coordinates relative to their average values; by diagonalizing this matrix we obtained eigenvectors and eigenvalues that give, respectively, the principal components and the magnitude of the variance. ,, We used the first two principal components (i.e., the two eigenvectors with the highest variance) to span the 2D plane that the atomic coordinates are projected on, producing the 2D graph of the H-bond network.…”

“…We thus generated the covariance matrix of the atomic coordinates relative to their average values; by diagonalizing this matrix we obtained eigenvectors and eigenvalues that give, respectively, the principal components and the magnitude of the variance. 30,44,45 We used the first two principal components (i.e., the two eigenvectors with the highest variance) to span the 2D plane that the atomic coordinates are projected on, producing the 2D graph of the H-bond network.…”

“…The more recent MDAnalysis , defines H bonds according to the default H-bond donors and acceptors of the CHARMM27 force field − and allows analyses of dynamical H bonds . Two-dimensional H-bond maps representing interactions between H-bond donor and acceptor groups were used to dissect long-distance conformational coupling in allosteric proteins. ,, …”

“…28 Two-dimensional H-bond maps representing interactions between H-bond donor and acceptor groups were used to dissect long-distance conformational coupling in allosteric proteins. 9,29,30 Here, we present Bridge, an algorithm tailored to analyses of dynamic H bonds in membrane proteins (Scheme 1), including long-distance protein−water H-bond bridges that could conduct protons or H-bond networks that couple remote regions of the protein. The algorithm relies on an efficient implementation of graph-based searches of H bonds.…”

Using Graphs of Dynamic Hydrogen-Bond Networks To Dissect Conformational Coupling in a Protein Motor

“…SecA, a well-studied model system for the DEAD-box proteins, functions as the protein motor of the Sec protein secretion pathway in bacteria, where it uses the energy from binding and hydrolysis of ATP to push secretory proteins across the membrane-embedded SecYEG protein translocon . Hydrogen (H) bonds between functional domains of the protein are thought to participate in long-distance allosteric coupling. − Here, we present an algorithm to derive simple graphical representations of a protein’s H-bond networks and used these graphs to explore how SecA responds to mutations and changes in the nucleotide-binding state.…”