This Title All WIREs
How to cite this WIREs title:
WIREs Comput Mol Sci
Impact Factor: 8.127

Synergistic use of NMR and MD simulations to study the structural heterogeneity of proteins

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

Abstract Nuclear magnetic resonance spectroscopy (NMR) and molecular dynamics (MD) simulations are powerful techniques for the structural characterization of macromolecules. NMR is unique in its ability to provide experimental information at atomic level on the structure as well as on the amplitude and rate of structural fluctuations. MD provides physically sound models and potential mechanisms that connect conformations in time. Nevertheless, none of these techniques allow yet obtaining experimentally validated movies of protein motions at atomic resolution. Instead, it is their complementarity and synergy which offer a unique opportunity toward this end. Here, we overview recent examples that illustrate how much these two techniques benefit from each other, both passively and actively, for the characterization of the structural heterogeneity in proteins. © 2012 John Wiley & Sons, Ltd. This article is categorized under: Molecular and Statistical Mechanics > Molecular Dynamics and Monte-Carlo Methods

The three‐dimensional Gaussian axial fluctuation (3D‐GAF motional) model that has been used with success to interpret heteronuclear relaxation rates47 as well residual dipolar couplings48 in terms of fluctuations of peptide bonds abound three orthogonal axes α, β, and γ. (Reprinted with permission from Ref 47. Copyright 2001 American Chemical Society.)

[ Normal View | Magnified View ]

Quantitative evaluation of CHARMM2731 molecular dynamics force field (FF) by nuclear magnetic resonance spectroscopy. The NH S2‐order parameters19,20 determined for a simulation (100 ns) of ubiquitin performed with CHARMM2731 FF are compared to experimental data.32 S2 were calculated using method 2 in Ref 33.

[ Normal View | Magnified View ]

Comparison of some structural models available for the protein ubiquitin. (A) Crystallographic structure obtained by X‐ray diffraction (pdb code 1UBQ). (B) Static average(s) structure obtained by solution nuclear magnetic resonance spectroscopy (pdb code 1D3Z). (C) Molecular dynamics simulation (1 µs with CHARMM force field) that collectively fits the NH residual dipolar couplings as shown in Figure 1 (100 structures).

[ Normal View | Magnified View ]

The histogram shows the agreement between the experimental14 and calculated residual dipolar couplings (RDCs) for each ubiquitin conformation of a 1 µs simulation carried out with the CHARMM27 FF in GROMACS. The agreement of the average RDCs with the experimental RDCs is shown as a black line. The values of Q for the X‐ray (pdb code 1UBQ), the static nuclear magnetic resonance spectroscopy (NMR) (pdb code 1D3Z) structure, and the recently determined ERNST15 NMR ensemble (pdb code 2KOX) are also provided. This figure is similar to Figure 1 of Ref 18, which was obtained with the AMBER99SB FF.

[ Normal View | Magnified View ]

Contact maps for (A) native ubiquitin (pdb code 1UBQ), (B) the statistical coil models reported by Jha et al.64 (16,000 members), and (C) the ERIDU63 ensemble. The ERIDU ensemble clearly indicates that there are contacts between native strands 1 and 2 as previously shown by Meier et al.67 through 3hJNC' couplings. The maps are color coded according to the fractional number of contacts between residues (Cα − Cα distance lower than 10 Å) among the ensemble. (Reprinted with permission from Ref 63. Copyright 2009 American Chemical Society.)

[ Normal View | Magnified View ]

Comparison of the ability of the 1 µs molecular dynamics (MD) simulation of ubiquitin presented in Figures 1 and 2C with that of the ensemble obtained by using ensemble MD simulations restrained by NH residual dipolar couplings (RDCs) to agree with independent nuclear magnetic resonance spectroscopy parameters. (A) RNH,NH cross‐correlated relaxation rates. (B) Transhydrogen scalar couplings (h3JNC′). The root‐mean‐squared deviation (rmsd) between calculated and experimental values is shown.

[ Normal View | Magnified View ]

Related Articles

Imaging: An Interdisciplinary View

Browse by Topic

Molecular and Statistical Mechanics > Molecular Dynamics and Monte-Carlo Methods

Access to this WIREs title is by subscription only.

Recommend to Your
Librarian Now!

The latest WIREs articles in your inbox

Sign Up for Article Alerts