New promise in combinatorial chemistry: synthesis, characterization, and screening of small-molecule libraries in solution.

Abstract

The increasing interest in combinatorial chemistry as a tool for the development of therapeutics has led to many new methods of creating molecular libraries of potential lead compounds. Current methods have made it possible to develop libraries of several million compounds. As a result, the limiting factor in the screening of libraries has become the identification and characterization of active species. We have recently described a method for generating libraries of water-soluble compounds containing mixtures of 10(4) to 10(5) different small organic molecules by using generally applicable solution phase chemistry. We set out to develop new methods to characterize and decode these libraries. Libraries were generated by condensing a multi-acid-chloride core molecule with various amines, producing molecules with functional groups about a rigid backbone. Composition and complexity of the libraries was evaluated using electrospray mass spectrometry to analyze model libraries containing up to 55 different molecules. The number of peaks obtained in mass spectrometry is directly correlated with the complexity of the library, and we were therefore able to deduce which of the expected compounds had in fact been formed in the library, and which of the building blocks in the library were not efficiently used. An iterative selection procedure was developed using this information, which allowed the screening of libraries of up to 50,000 chemical species to produce a competitive inhibitor of the enzyme trypsin. Our strategy for the identification of active species should be broadly applicable to other methods of generating complex libraries of small molecules. The selection from the library of a compound with desired biological properties augurs well for the potential value of generating and screening complex mixtures of small molecules in solution.