A-Site Residues Move Independently from P-Site Residues in all-Atom Molecular Dynamics Simulations of the 70S Bacterial Ribosome

Pande VS (2012) A-Site Residues Move Independently from P-Site Residues in all-Atom Molecular Dynamics Simulations of the 70S Bacterial Ribosome. PLoS ONE 7(1): e29377. doi:10.1371/journal.pone.0029377 A-Site Residues Move Independently from P-Site Residues in all-Atom Molecular Dynamics Simulations of the 70S Bacterial Ribosome Relly Brandman 0 Yigal Brandman 0 Vijay S. Pande 0 Jie Zheng, University of Akron, United States of America 0 1 Chemical and Systems Biology, Stanford University, Stanford, California, United States of America, 2 Chemistry, Stanford University , Stanford, California , United States of America The ribosome is a large macromolecular machine, and correlated motion between residues is necessary for coordinating function across multiple protein and RNA chains. We ran two all-atom, explicit solvent molecular dynamics simulations of the bacterial ribosome and calculated correlated motion between residue pairs by using mutual information. Because of the short timescales of our simulation (ns), we expect that dynamics are largely local fluctuations around the crystal structure. We hypothesize that residues that show coupled dynamics are functionally related, even on longer timescales. We validate our model by showing that crystallographic B-factors correlate well with the entropy calculated as part of our mutual information calculations. We reveal that A-site residues move relatively independently from P-site residues, effectively insulating A-site functions from P-site functions during translation. - Funding: RB was funded by a Bio-X Graduate Student Fellowship. Computer resources on the Stanford Bio-X2 cluster were provided by National Science Foundation award Division of Computer and Network Systems CNS-0619926. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. The dynamic motion of macromolecular machines such as the ribosome is coordinated across multiple chains to accomplish complex, multi-step functions and elucidation of these motions is fundamental to understanding how these large machines work [1,2,3,4,5]. The ribosome translates mRNA into protein by coordinated dynamics of its hundreds of thousands of atoms in more than fifty RNA and protein chains. These chains are assembled into two subunits: the large subunit catalyzes peptide bond formation, and the small subunit decodes mRNA by facilitating the binding of the mRNA codon to the corresponding tRNA anticodon. Co-factors and GTP hydrolysis catalyze events in protein synthesis, however the ability of the ribosome to translate without GTP hydrolysis [6] and even without any cofactors [7,8] suggests that functionally important dynamics are intrinsic to the ribosome itself. The molecular details of ribosomal translation are still an active area of research, and understanding ribosome dynamics is central to understanding how this robust molecular machine orchestrates protein synthesis. Our current understanding of ribosome dynamics began largely by comparing EM and x-ray crystallography structures of the ribosome in various states (for example bound to co-factors or tRNA) [1,2,3,4]. The ribosome translates proteins on the order of seconds (e.g. ,20 amino acids per second in Escherichia coli). Largescale dynamics such as motion in the GTPase association center [9] and L1 stalk [10] (large subunit protuberances at the tRNA entrance and exit sites, respectively) have been measured on similar timescales using methods such as single molecule fluorescence resonance energy transfer (smFRET) [10,11,12]. Thus enzymatic timescales are on the seconds timescales, and there is likely a broad range of timescales characterizing ribosome dynamics. Elucidating motions in the ribosome continues to be an active area of research by both experimental and computational methods. Computational models of ribosome dynamics are challenging because of the ribosomes large size. Methods for modeling enzyme dynamics range from reduced complexity coarse-grained models to the all-atom detail of explicit solvent molecular dynamics simulations (MD) [13,14,15,16,17]. The more detail included in a computational model, the more computationally demanding the calculations are. Coarse-grained models are able to access biologically relevant timescales (e.g. peptide bond formation occurs on the seconds timescale), while more detailed models such as MD require decades of CPU years to reach tens of nanoseconds for the millions of atoms in a system as large as the ribosome. Multiple, independent models of ribosome dynamics on various timescales complement each other and fill in molecular-level spatial and temporal timescales that can be difficult to access experimentally. Our goal is to find features in nanosecond length all-atom MD of ribosome dynamics that are biologically relevant. Correlated motions have been previously linked to enzyme function [18], including coarse-grained simulations of the ribosome on the seconds timescale [15]. Correlated motions in short timescale (relative to biological timescales) MD have been used to gain insight in other systems (e.g. [19]). We expect that the ns dynamics in our simulations are largely from thermal motions and local conformational changes around the crystal structure rather than large-scale conformational changes. We calculate correlated motion in order to group atoms with coupled dynamics together. We hypothesize that atoms with coupled dynamics are functionally related, and that these relationships are also relevant on longer timescales. Here we present data from two all-atom MD trajectories of the complete (both subunits, 70S) bacterial Thermus Thermophilus bacterial ribosome [20] and calculate correlations in residue dynamics using mutual information. One trajectory is of the ribosome alone (53 ns), and the other trajectory contains the ribosome with mRNA and tRNA (32 ns). Because of the short timescale (compared to the timescale for translation) and the fact that the initial starting configuration reflects a stalled ribosome in the pre-translocation state, we do not expect to see motions corresponding to translation. We expect that motions in our simulation are largely thermal motions on the ns timescale and local conformational changes around the crystal structure, thus our conclusions can be likened to characterizing a structure based on how it vibrates. We describe correlated motions between residues in order to characterize groups of atoms that may function together as parts of the larger macromolecular machine, thus providing insight into ribosome dynamics. Correlated motions corroborate previously established parts such as the two symmetry-related regions in the active site [21] and the peptide exit tunnel constriction site formed by two proteins [22]. Residues in the large subunit along the path of tRNA transloc (...truncated)