Molecular conformations in a phospholipid bilayer extracted from dipolar couplings: a computer simulation study

This paper describes an analysis of NMR dipolar couplings in a bilayer formed by dimyristoylphosphatidylcholine (DMPC). The couplings are calculated from a trajectory generated in a molecular dynamics (MD) simulation based on a realistic atom-atom interaction potential. The analysis is carried out employing a recently developed approach that focuses on the construction of the conformational distribution function. This approach is a combination of two models, the additive potential (AP) model and the maximum entropy (ME) method, and is therefore called APME. In contrast to the AP model, the APME procedure does not require an intuition-based choice of the functional form of the torsional potential and is, unlike the ME method, applicable to weakly ordered systems. The conformational distribution function for the glycerol moiety of the DMPC molecule derived from the APME analysis of the dipolar couplings is in reasonable agreement with the "true" distributions calculated from the trajectory. Analyses of dipolar couplings derived from MD trajectories can, in general, serve as guidelines for experimental investigations of bilayers and other complex biological systems.     

NMR studies of molecular conformations in alpha-cyclodextrin

A new approach for analysis of NMR parameters is proposed. The experimental data set includes scalar couplings, NOEs, and residual dipolar couplings. The method, which aims at construction of the conformational distribution function, is applied to alpha-cyclodextrin in isotropic solution and dissolved in a dilute liquid crystal. An attempt to analyze the experimental data using an average molecular conformation resulted in unacceptable errors. Our approach rests on the maximum entropy method (ME), which gives the flattest possible distribution, consistent with the experimental data. Very good agreement between experimental and calculated NMR parameters was observed. In fact, two conformational states were required in order to obtain a satisfactory agreement between calculated and experimental data. In addition, good agreement with Langevin dynamics computer simulations was obtained.     

Molecular structure extracted from residual dipolar couplings: diphenylmethane dissolved in a nematic liquid crystal

This paper describes an analysis of 1H-1H residual dipolar couplings (RDCs) in diphenylmethane (DPM) dissolved in a nematic liquid crystal, reported by Celebre et al. [J. Chem. Phys. 118, 6417 (2003)]. In that article, the conformational distribution function for DPM was extracted from the RDCs, using the additive potential (AP) model which is based on the molecular-field theory. The AP approach is a powerful, and frequently used, tool for analysis of the nuclear-magnetic-resonance (NMR) parameters in liquid crystals. It requires, however, a priori knowledge of the functional form of the torsional potential, which may even for a simple molecule, such as DPM, be complicated to determine. Here, we analyze the same set of the RDCs using our APME procedure, which is a hybrid model based on the AP approach and maximum entropy (ME) theory. The APME procedure does not require any assumptions about the functional form of the torsional potential and, in contrast with the ME method, is applicable to weakly ordered systems. In the investigation reported in the present study, the results from the APME analysis are in good agreement with the AP interpretation, whereas the ME approach essentially fails in the extraction of the conformational distribution function for DPM.     

Interpolation by fast Wigner transform for rapid calculations of magnetic resonance spectra from powders

We introduce a novel interpolation strategy, based on nonequispaced fast transforms involving spherical harmonics or Wigner functions, for efficient calculations of powder spectra in (nuclear) magnetic resonance spectroscopy. The fast Wigner transform (FWT) interpolation operates by minimizing the time-consuming calculation stages, by sampling over a small number of Gaussian spherical quadrature (GSQ) orientations that are exploited to determine the spectral frequencies and amplitudes from a 10-70 times larger GSQ set. This results in almost the same orientational averaging accuracy as if the expanded grid was utilized explicitly in an order of magnitude slower computation. FWT interpolation is applicable to spectral simulations involving any time-independent or time-dependent and noncommuting spin Hamiltonian. We further show that the merging of FWT interpolation with the well-established ASG procedure of Alderman, Solum and Grant [J. Chem. Phys. 134, 3717 (1986)] speeds up simulations by 2-7 times relative to using ASG alone (besides greatly extending its scope of application), and between 1-2 orders of magnitude compared to direct orientational averaging in the absence of interpolation. Demonstrations of efficient spectral simulations are given for several magic-angle spinning scenarios in NMR, encompassing half-integer quadrupolar spins and homonuclear dipolar-coupled (13)C systems.     

Efficient orientational averaging by the extension of Lebedev grids via regularized octahedral symmetry expansion

Gaussian spherical quadrature methods in the guise of the Lebedev sampling grids are highly efficient for some orientational ("powder") averaging problems in solid state NMR. However, their applicability is currently restricted, as the sets of orientations are derived analytically and because they are not well adapted to simulate the broad peakshapes encountered, for example, in the NMR on static powders or on half-integer quadrupolar spins subject to second order quadrupolar interactions under magic-angle spinning conditions. We remedy these problems by (i) introducing the recursive procedure regularized octahedral symmetry expansion (ROSE), to which any existing Lebedev set may be subjected. Each recursive step gives a 9-fold enlarged set of orientations. (ii) We demonstrate that ROSE-expanded grids, in conjunction with spectral interpolation, is well suited for calculating broad peakshapes. These advances combine into the apparently most efficient general-purpose two-angle orientational averaging technique proposed to date for solid state NMR applications.