A MOLECULAR-DYNAMICS INVESTIGATION OF CHIRAL DISCRIMINATION COMPLEXES AS CHIRAL STATIONARY-PHASE MODELS - METHYL N-(2-NAPHTHYL)ALANINATE WITH N-(3,5-DINITROBENZOYL)LEUCINE N-PROPYLAMIDE

Molecular dynamics simulations were performed on complexes of (S)-methyl N-(2-naphthyl)alaninate (NAP) with the enantiomers of N-(3,5-dinitrobenzoyl)leucine n-propylamide (DNB), which are used as models for chiral stationary-phase systems developed by Pirkle and co-workers. These studies were undertaken to qualitatively examine (pictorially) the role of entropic effects in these systems. The results of the dynamics calculations were used to refine the search for low-energy conformers. The structures were refined by the use of BioDesign's molecular mechanics method implemented in Biograf. The results of the structural refinements support our previous observation that the SR complex can achieve the same three primary interactions which are observed in the SS structure (i.e., two intermolecular hydrogen bonds and pi stacking) without a significant increase in energy. In addition, these primary interactions are conserved during molecular dynamics simulations with the occurrence of conformations which differ only in the rotational states of the alkyl side chains and ester group (which bears two potential hydrogen bond acceptors utilized in both the homo- and heterochiral complexes). The major difference in the two complexes is the relative position of the sec-butyl group and hydrogen atom on DNB's chiral center, both of which are outside the primary interaction region. All other local minima which have different relative pi orientations ("front-back," "back-back", and "back-front" as defined herein) are not sufficiently populated to make more than a negligible contribution to the statistical (time- or energy-averaged) analysis of the (SS)- and (SR)-NAP-DNB complexes. Thus the entropic effects observed in this study (e.g., alkyl side chain or ester group rotations) do not show evidence of qualitative differential effects on the maintenance of the same three primary interactions by both the homo- and heterochiral complexes. The reliability of the present study, which provides pictorial representations of the entropic effects, is not sufficient to determine whether the entropic effects observed herein are sufficient to achieve enantiomeric discrimination alone or in conjunction with other factors (e.g., conformational strain energy). Thus, all of the computational studies we have performed to date (i.e., our previous studies, which include strain energy and through-space field effects, and the present study, which includes entropic effects) show no evidence of any qualitative difference in the homo- and heterochiral complexes in terms of maintaining the same three "contact points."