The long-term objective of the proposed research is to obtain valid representations for the binding of mitomycins and certain anthracenes to DNA, in order to design new, improved agents more rationally. We have experience in the chemistry and spectroscopy of these agents and many analogues with interesting structural features and proven antitumor activity are on hand. The DNA binding of these compounds will be studied on a theoretical level by molecular mechanics and computer graphics. Experimental studies on binding will be made in order to test the theoretical models and to help simplify the computer searches of random drug-DNA interactions. Although the mode of mitomycin binding to DNA has been studied by numerous investigators, much of it is uncertain. It is known that mitomycins are activated by bioreduction and that the resulting reactive intermediates bind with and alkylate DNA. Recently, some of the sites of alkylation have been defined by analysis of DNA hydrolysates. However, the reactivity of the putative intermediates precludes their isolation. We wish to consider the initial non-covalent binding of two structures proposed as intermediates, examining how their calculated electrostatic potential surfaces might have complementarity with certain regions of DNA, and if such complementarity correlates with the known binding sites. Also to be investigated is the approach of these intermediates to DNA using a """"""""docking"""""""" algorithm, with the known sites of covalent binding invoked to simplify the selection of """"""""best"""""""" answers from many possible ones. Anthracene derivatives to be investigated include structures related to bisantrene, mitoxantrone, and a promising new type that we have discovered. These compounds are all on hand and they have been tested for comparative potencies in human tumor clonogenic assays. Anthracenes are thought to be intercalating agents. Theoretical representations will be derived for the intercalation of anthracenes and they will be compared with 1H-nmr evidence obtained on anthracene-dinucleotide complexes. Binding energies will be calculated and compared with those determined experimentally. Quantitative structure-activity relationships will be derived using these binding energies and potencies in antitumor assays.
