The design and synthesis of functional models of the enediyne toxins is the primary aim of this proposal. The long range goal is the development of selective cytotoxic drugs based on novel mechanisms of antibiotic activity. The calicheamicins, esperimicins, and dynemicins are highly toxic species wit elegant mechanisms of activation which were not anticipated before the natural structures were determined. With the mechanisms now largely supported by in vitro studies, it is possible to design analogs which have all of the activation features, but will be readily available in quantity, can be tailored to avoid peripheral toxicity, and can have convenient tethers for conjugation with delivery agents such as antibodies. The models for calicheamicin are based on a geometry change provoked by enol to keto tautomerization of a bridgehead enol double bond, a change similar to the addition to a bridgehead double bond in the natural series. However, the key triggering step will be de-blocking of the enol, and that operation can be designed to involve several different biological mechanisms, including disulfide exchange as proposed for the natural products. The result will be a simplified model for the """"""""warhead"""""""" or aglycone which can be triggered to produce high energy diradicals under physiological conditions. A collaborative effort will evaluate the new compounds as selective DNA cleavage agents and arrange attachment of the more active derivatives to delivery systems, including the monocionnal antibody conjugate now under study with the natural products. Dynemicin is less well studied, and offers similar opportunities for modeling by synthesis. In this case, much will be revealed about the mechanism from simplified models and the models themselves can serve as new drug candidates. The triggering mechanism is different, apparently involving bioreductive alkylation, and the functional models may have a very different profile of activity oompared to the calicheamicins. Connected with the synthesis of functional models is a useful type of intermediate which should facilitate synthesis in the ene-diyne series generally. The geometry of the precursor should make key ring-forming reactions more efficient, and keep the intermediates stabilized until
| Semmelhack, M F; Wu, Lingyun; Pascal Jr, Robert A et al. (2003) Conformational control in activation of an enediyne. J Am Chem Soc 125:10496-7 |
| Semmelhack, M F; Jiang, Y; Ho, D (2001) Synthesis of the amino sugar from C-1027. Org Lett 3:2403-6 |
