DNA topoisomerases alter DNA topology by transiently breaking either one (topoisomerase I) or both (topoisomerase II) strands of DNA (1). The essential role that these enzymes play in cellular processes such as DNA replication, transcription and recombination (2) suggests that they may constitute logical targets in the design of antineoplastic agents. In fact, several useful antitumor agents have now been shown to function, at least in part, via inhibition of DNA topoisomerase II (3,4).

Bioassay-guided fractionation of a methanol extract of Erodium stephanianum found to inhibit DNA topoisomerase I mediated DNA relaxation afforded corilagin (1) and chebulagic acid (2). The latter is the most potent inhibitor of mammalian topoisomerase I yet reported.

While the design of molecules that inhibit or antagonize the functions of specific macromolecules is now well precedented, in many cases the structural information requisite to the design process is lacking. The tools of molecular biology can now furnish the target macromolecules for use in mechanism-based exploration; highly defined assays can be devised based upon the known biochemistry of these macromolecules to permit the discovery of novel inhibitors or antagonists present in chemical collections. Presently, we describe a set of assays directed toward the discovery of novel inhibitors of eukaryotic topoisomerase I, an enzyme critical to maintenance of chromosomal DNA topology and therefore essential for normal replication and transcription. The identification of chebulagic acid as an extraordinarily potent and mechanistically novel inhibitor of topoisomerase I illustrates the potential of this approach.

A novel flavone-C-glycoside, aciculatin (1), has been isolated from the methylene chloride extract of Chrysopogon aciculatis (Poaceae) collected in the Philippines. The structure of 1 was determined by analysis of spectral data. Aciculatin exhibits cytotoxicity towards KB cells that is reduced by an order of magnitude in the presence of exogenous DNA indicating that 1 binds to DNA. DNA binding assays indicated an apparent Kd of 15 – 50 μM for binding of 1 to calf thymus DNA.

A novel bioassay was developed to permit the identification of cytotoxic natural principles that bind to DNA. A hexane extract of Schoepfia californica cytotoxic to cultured KB cells displayed much less cytotoxic potential when the culture medium contained exogenously added calf thymus DNA. Fractionation of the extract afforded a purified principle shown to be 9-octadecynoic acid, an 18-carbon, unbranched acetylenic fatty acid. 9-Octadecynoic acid had an apparent DNA dissociation constant of 1.8 mM; it inhibited topoisomerase I mediated DNA filter binding but did not inhibit the DNA topoisomerase I mediated relaxation of a supercoiled plasmid DNA. The fatty acid was weakly inhibitory to DNA polymerase alpha. 9-Octadecynoic acid possesses none of the structural characteristics of known DNA binding molecules and may bind to DNA by some novel mechanism.

Analogous to certain radiosensitizers which are too hydrophilic to enter tumor cells, certain radioprotectors, because of their hydrophilicity, may also be hindered from entering tumor cells and thus protect only normal tissues. In testing this hypothesis, we utilized thin layer chromatography as convenient means to measure radioprotector hydrophilicity. Dose reduction factors (DRF's) for hematopoietic radioprotection were determined in BALB/c mice given half maximum tolerated doses (MTD/2) of 11 radioprotectors 30 min prior to graded doses of gamma rays. DRF's for tumor protection were determined in MCa-11 tumor-bearing mice using a regrowth delay assay. Differential radioprotection was found to be significantly correlated (r = 0.86) with hydrophilicity. Thus, radioprotector hydrophilicity appears to be a significant factor in the differential radioprotection observed and should be useful in designing or selecting better differential radioprotectors.