Calicheamicin gamma1I (CLM), the neocarzinostatin chromophore, esperamicin A1, dynemicin, kedarcidin, c-1027 and maduropeptin are members of the structurally unprecedented and growing family of diynene antitumor antibiotics. Reductive activation by thiols and/or simple thermal rearrangement generates diradical species by an electrocyclization process that is characteristic of this class. When bound in the minor groove of DNA these radicals initiate helix cleavage by hydrogen abstraction from one or both strands. The resulting DNA radicals react with molecular oxygen and fragmentation ensues. While single-strand breaks are typically the major cleavage event observed, CLM is both notably sequence-selective and very largely a double-strand cutter. Using a diverse array of experiments outlined in this renewal proposal we seek to understand the origins of these properties and how CLM can be used as a probe of protein/nucleic acid structure and in the design of useful tools in molecular biology. The presumed lethality of DNA damage, the induction of apoptosis and the demonstrated cytotoxicity and cell cycle effects of these compounds have animated hope that, through preferential uptake or selective delivery to cancer cells, effective therapies might be possible. Clinical trials testing the latter principle are underway. Investigation of the low kinetic isotope effects associated with CLM- induced hydrogen abstraction from DNA and the apparent absence of isotope-induced branching will be completed. Whether these two events occur in a simultaneous or stepwise manner will be distinguished. The unusual behavior of A-tracts will be examined for kinetic acceleration as a function of minor groove narrowing. The efficiencies of hydrogen abstraction will be monitored and the distribution of cuts in the cleavage cascade will be evaluated as a function of temperature. The interaction of CLM with supercoiled plasmid DNA will be investigated as a function of inserted A/T-rich segments causing varying degrees of curvature. Extensive experiments are planned with nucleosomes to study the occurrence of """"""""hot spots"""""""" to understand whether CLM cleavage is associated with physical location in the nucleosome or features of particular DNA sequences. Automated methods will be used to evaluate large amounts of data generated in mixed sequence nucleosomes. A degradation product of CLM having essentially no sequence selectivity but retaining double-strand cutting properties will be linked to homeodomains to achieve specific binding for high efficiency cleavage of DNA. Success in this effort will be evaluated in tests with DNA sequences already in hand and will lay the groundwork for eventual experiments to be undertaken for gene identification e.g. in Drosophila. Finally, CLM reaction with transfer-, ribosomal- and messenger-RNA will be investigated. In eukaryotic cells, RNA is a more accessible target than DNA. These experiments will monitor the susceptibility of RNA to reaction with CLM and particularly with rRNA in the ribosome for comparison to nucleosomes and chromatin.
| Kuduvalli, P N; Townsend, C A; Tullius, T D (1995) Cleavage by calicheamicin gamma 1I of DNA in a nucleosome formed on the 5S RNA gene of Xenopus borealis. Biochemistry 34:3899-906 |
| Mah, S C; Townsend, C A; Tullius, T D (1994) Hydroxyl radical footprinting of calicheamicin. Relationship of DNA binding to cleavage. Biochemistry 33:614-21 |
