(Principal Investigator's) We will investigate enantioselective 1,2-additions of lithium acetylides (RLi) in the presence of chiral amino alkoxides (RstarOLi) employed as the critical steps in the Merck syntheses of HIV reverse transcriptase inhibitors L-743,726 and L-738,372. The work will be carried out in collaboration with Merck, Dupont-Merck, and ASI Applied Systems. Our goal will be to improve and generalize the process to include a wider range of carbanions and ketone substrates. During the first phase of the collaboration, NMR spectroscopic, IR spectroscopic, and computational studies of observable RLi-RstarOLi mixed aggregates (RstarOLi = synthetic lithium ephedrate) led to a mechanistic model for a highly enantioselective lithium acetylide addition to an ArCOCF3 ketone employed in the L-743,726 synthesis. The model will be tested by carrying out additional structure-selectivity studies, NMR spectroscopic investigations, theory-experiment correlations, and rate studies. With guidance provided by the mechanistic model, computations will play prominently in the design of new chiral auxiliaries. Alternative strategies for optimization of the 1,2-addition protocol will be based upon: (1) improved enantioselectivities noted at elevated (3:1) RstarOLi/RLi proportions, and (2) new lithiated diamine auxiliaries that are isostructural to the lithiated amino alcohol derivatives. We will also initiate and carry out investigations of the highly enantioselective acetylide addition to imines employed in the synthesis of reverse transcriptase inhibitor L-738,372. While the highly selective additions to imines behave quite differently from the additions to ketones (requiring lithiated quinine or quinidine), the strategies and analytical methods used to study them will be similar.
