Development of a grid search molecular mechanics modeling strategy to study elution behavior in cyclodextrin modified capillary electrophoresis

Novel methods to use molecular mechanics (MM) in modeling solute-receptor interactions are developed that are pertinent to computational studies of dynamic separation or sensing processes. Specifically, a grid search MM modeling approach is used to study separation behavior in cyclodextrin (CD)-modified capillary electrophoresis (CE). Laboratory CE separations of di-substituted naphthalene compounds are accomplished employing an anionic, single isomer carboxymethyl-CD as a running buffer additive. Computational work involved the docking of these solutes into the cavity of this CD using a systematic grid search to probe the interaction space. It was discovered that the size of incremental changes in position and dimensional freedom in defining the grid are significant parameters. At each grid position the configuration of the atoms in the complex is altered in a process to minimize energy. It was discovered that comprehensive minimization involving many iterations of the process is extremely important in locating true low energy conformers. Interaction energy contours, involving translation and rotation of the solute into the CD cavity, are produced that graphically reveals the strength of the CD-solute interaction. Increasing the number of minimization iterations from 100s to 1000s resulted in remarkable changes in those contours, and improved the correlation between computationally determined and experimental distribution coefficients. The correct ordering of distribution coefficients was achieved with a few approaches, but with the comprehensive minimization approach experimental and computational coefficients were generally within a factor of 4 of each other. Moreover, the comprehensive minimization approach resulted in many points in the grid search, micro-positional states, that minimized to a single conformer. Consideration of the degeneracy of these states is also shown to be significant.