Analysis: Unbiased Multi-State Conformational Modeling of Exocyclic Hydroxymethyl Group Conformation in Methyl Aldohexopyranosides.

analysis (. , , 3135-3141) has been applied to model exocyclic hydroxymethyl group conformation in methyl β-D-glucopyranoside (βGlcOMe), methyl β-D-galactopyranoside (βGalOMe), and methyl β-D-mannopyranoside (βManOMe) in an unbiased manner. Using up to eight NMR -couplings sensitive to rotation about the C5-C6 bond (torsion angle ω), two-state models of ω were obtained that are qualitatively consistent with the relative populations of the , and rotamers reported previously. analysis gave consistent unbiased ⇌ models of ω in βGalOMe, with more populated than and mean values of ω for each population similar to those obtained from aqueous 1-μs MD simulation. Using different combinations of -couplings had little effect on the βGalOMe model in terms of the mean values of ω and circular standard deviations (CSDs). In contrast, analysis of ω in βGlcOMe and βManOMe produced more than one two-state model independent of the ensemble of -values used in the analyses. These models were characterized by ⇌ conformer exchange as expected, but the mean values of ω in both conformers varied significantly in the different fits, especially for the rotamer. Constrained (biased) analyses in which only staggered geometries about ω were allowed gave RMSDs slightly larger than those obtained from the unbiased fits, precluding an assignment of an unbiased model. It is unclear why analysis gives consistent and predictable unbiased models of ω in βGalOMe but not in βGlcOMe and βManOMe. One possibility is that the distribution of ω in one or both of the and conformers in the latter does not conform to a von Mises function (i.e., is not Gaussian-like), but rather to a broad and/or flat distribution that cannot be fit by the current version of . Nevertheless, the results of this study provide new evidence of the ability of analysis to treat multi-state conformational exchange using only experimental NMR data, extending recent applications to furanosyl ring pseudorotation ( , , 239-251).