Identification of novel antitumor agents from mixture-based synthetic combinatorial libraries using cell-based assays

Jon R. Appel, Richard A. Houghten
Torrey Pines Institute for Molecular Studies, 3550 General Atomics Court, San Diego, CA 92121, U.S.A.

Jill Johnson, Ven L. Narayanan
Drug Synthesis and Chemistry Branch, Developmental Therapeutics Program, National Cancer Institute, Executive Plaza North, Room 831, 6130 Executive Boulevard, Bethesda, MD 20892, U.S.A.

Abstract
A new strategy is presented here which integrates combinatorial library technology with the antitumor in vitro screening system at the National Cancer Institute in the search for novel antitumor agents. Mixture-based synthetic combinatorial libraries (SCLs) representing hundreds of thousands to millions of individual compounds were screened against the cell-based assay, which evaluates compounds for their ability to inhibit the growth of 60 different human tumor cell lines. Five different SCLs, composed of peptides, peptidomimetics, polyamines or small molecules were first tested against three cell lines to identify the most active SCLs. Two SCLs, namely the N-perbenzylated pentamine and the N-acylated permethylated triamine, were deconvoluted to yield individual compounds having significant activities against the 60 tumor cell lines. Active compounds were tested in mice to determine the maximum tolerated dose, followed by in vivo testing in a hollow fiber assay. Using this strategy, three different compounds identified directly from SCLs are currently being evaluated in human tumor xenografts. This study demonstrates for the first time the use of in vitro cell-based assays to identify antitumor lead compounds from mixture-based combinatorial libraries.

Keywords
antitumor agents, cancer drug discovery, combinatorial chemistry, high-throughput screening, mixtures, positional scanning, synthetic combinatorial libraries