Analysis: Probability Distributions of Molecular Torsion Angles in Solution from NMR Spectroscopy.

ConspectusMonosaccharides adopt multiple conformations in solution, and this structural complexity increases significantly when they are assembled into oligosaccharides and polysaccharides. Characterization of the conformational properties of saccharides in solution by NMR spectroscopy has been hampered by several complicating factors, including difficulty interpreting spectra because of significant signal overlap, population averaging of NMR parameters, and unique properties of the spectra that make accurate measurements of NMR parameters prone to error (e.g., non-first-order effects on -couplings). Current conformational assignments rely heavily on theoretical calculations, especially molecular dynamics (MD) simulations, to interpret the experimental NMR parameters. While these studies assert that the available experimental data fit the calculated models well, a lack of independent experimental validation of the force fields from which MD models are derived and an inability to test all possible models that might be compatible with the experimental data in an unbiased manner make the approach less than ideal.NMR spin couplings or -couplings have been used as structure constraints in organic and other types of molecules for more than six decades. The dihedral angle dependence of vicinal (three-bond) H-H spin couplings () first described by Karplus led to an explosion of applications for a wide range of conformational problems. Other vicinal -couplings (e.g., , , and ) have been found to exhibit similar dihedral angle dependencies. values have been used to assign the preferred conformation in molecules that are conformationally homogeneous. However, many molecules, particularly those in biological systems, are conformationally flexible, which complicates structural interpretations of values in solution. Three-state staggered models are often assumed in order to deconvolute the conformationally averaged values into conformer populations. While widely applied, this approach assumes highly idealized models of molecular torsion angles that are likely to be poor representations of those found in solution. In addition, this treatment often gives negative populations and neglects the presence of librational averaging of molecular torsion angles.Recent work in this research group has focused on the development of a hybrid experimental-computational method, analysis, that provides probability distributions of molecular torsion angles in solution that can be superimposed on those obtained by MD. Ensembles of redundant NMR spin couplings, including (vicinal), (geminal), and sometimes (direct) values, are used in conjunction with circular statistics to provide single- and multistate models of these angles. analysis provides accurate mean torsion angles and circular standard deviations (CSDs) of each mean angle that describe the librational motion about the angle. Both conformational equilibria and dynamics are revealed by the method. In this Account, the salient features of analysis are discussed, including some applications to conformational problems involving saccharides and peptides.