(675b) Investigation of the Chromatographic Separation of Chiral Drugs By Molecular Dynamics Simulation

Many drugs have a chiral center as a key constituent. Chiral molecules have two enantiomers, which have the same molecular composition but different three-dimensional structures. The two enantiomers could have different therapeutic results; one enantiomer could be beneficial while the other could be toxic. Consequently, over the past decades pharmaceutical companies have made significant investments in purifying drugs to create an enantio-pure product; however, the design of these purification systems is challenging. One of the most effective methods for chiral separation is chromatography; but method development is still slow owing to a large number of combinations of chiral stationary phases (CSPs), mobile phases, pH, and temperatures, in the screening process to purify the racemate.

To help speed up the process of method development and to elucidate the chiral recognition mechanism, we have used molecular dynamics (MD) simulations to investigate the interaction between the CSP [in this case Amylose tris-(3,5-dimethyl-phenylcarbamate)] and various drug molecules. Our aim is to successfully relate the calculations from MD to the experimental observations. We first constructed and validated the model of the CSP by comparing the configuration generated from simulations with that observed in the NMR study (left-handed helical backbone structure). We then identified properties such as the lifetime of the hydrogen bonds between the CSP and the drug, which correlated well with the elution order and the selectivity of the drug racemates observed in the lab. Finally, the effect of the mobile phases on the chiral separation and on the conformation of the CSPs was investigated. The mobile phases were found to have a significant effect on the conformation of the CSP, thus affecting the chiral separation.

Our ultimate goal is to develop a model that will successfully predict the chiral stationary phase and mobile phase combination required for enantiomeric separations. The successful development of our predictive model will accelerate drug discovery in the pharmaceutical industry by reducing R&D time and costs.