Search

Functional protein motions covering a wide range of timescales can be studied by NMR and molecular dynamics (MD) computer simulations. Nuclear magnetic resonance (NMR) spectroscopy of uniformly 15N-labeled proteins provides unique insights into protein backbone motions on the pico- to nanosecond timescale. Nuclear spin relaxation parameters (R1, R2, and heteronuclear {1H}-15N NOE) are routinely measured by well-established multidimensional relaxation experiments, which can be time-consuming typically taking on the order of a week. Recently, NMR chemical exchange saturation transfer (CEST) has emerged as a useful method to probe slow millisecond motions complementing spin relaxation in the rotating frame (R1ρ) and Carr-Purcell-Meiboom-Gill (CPMG) type experiments. CEST also provides site-specific R1 and R2 relaxation parameters requiring only short measurement times of the order of a few hours. We demonstrate that the CEST-derived relaxation parameters accurately reflect relaxation parameters obtained using the traditional method. A “lean” version of the model-free analysis (MFA) is introduced for the interpretation of R1 and R2 resulting in S2 order parameters that closely match those obtained using a standard MFA. The new methodology is demonstrated with experimental data of ubiquitin and arginine kinase and backed up with simulated data derived from microsecond MD simulations of ten different proteins. MD simulations of proteins now routinely extend into the hundreds of nanoseconds time scale range exceeding the overall tumbling correlation times (τc) of proteins in solution. However, presently there is no consensus on how to rigorously validate these simulations by quantitative comparison with model-free order parameters derived from NMR relaxation experiments. For this purpose we conducted MFA of NMR relaxation parameters computed from 500-ns MD trajectories of ten proteins. The resulting model-free S2 order parameters are then used as targets for S2 values computed directly from the trajectories by the isotropic Reorientational Eigenmode Dynamics (iRED) method by either averaging over blocks of variable lengths or by using exponentially weighted snapshots (wiRED). We find that the S2iRED values are capable of reproducing the target S2 with high accuracy provided that the averaging window is chosen as five times the length of τc. These results provide useful guidelines for the derivation of NMR order parameters from MD for a meaningful comparison with their experimental counter parts. Protein loops with their flexible nature on a wide range of timescales are critical for many biologically important events at the molecular level, such as protein interaction and recognition processes. To obtain a predictive understanding of the dynamic properties of loops, we performed long MD simulations of 38 proteins and validated the simulations using NMR chemical shifts. A total of 169 loops were analyzed and classified into three types (fast, slow, static) according to the correlation times computed from the trajectory. Chemical and biophysical loop descriptors, such as amino-acid sequence, average 3D structure, charge distribution, hydrophobicity, and local contacts were used to develop and parameterize a novel algorithm (ToeLoop) for the prediction of the flexibility and motional timescale of every protein loop, which is also implemented as a public web server. The results demonstrate that loop dynamics with their timescales can be predicted rapidly with reasonable accuracy, which will allow the screening of average protein structures to understand the dynamic properties of loops in protein-protein interactions (PPI) and allostery. We reported the loop motions and their dominant timescales for a database of 230 proteins that form protein-protein complexes using the ToeLoop predictor of loop dynamics. We observe a clear tendency of loops that move on relatively slow timescales of tens of ns to sub-µs to be directly involved in binding and recognition processes. Complex formation leads to a significant reduction in loop flexibility at the binding interface, but in a number of cases it can also trigger increased loop flexibility at distal sites in response to allosteric conformational changes. We explored the relationship between loop dynamics and experimental binding affinities and found that a prevalence of high loop rigidity at the binding interface is an indicator of increased binding strength. The importance of loop dynamics and allostery is highlighted by case studies of antibody-antigen complex, K-Ras GTPases, adenylate kinase, and ubiquitin-conjugating E2 enzyme.