MPE•Fe(II) Footprinting: Drug Binding Sites on Native DNA

Citation

Van Dyke, Michael W. (1984) MPE•Fe(II) Footprinting: Drug Binding Sites on Native DNA. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/vy36-rq29. https://resolver.caltech.edu/CaltechTHESIS:12052018-113900386

Abstract

Many small molecules, such as antibiotics used in cancer chemotherapy, are believed to execute their therapeutic action through binding to the DNA template and impeding the progress of transcription and replication. While it is possible by spectroscopic means to determine the overall affinities and stoichiometries of these small molecule/DNA interactions, the exact locations and sizes of these sites of interaction are not known. The analogous question of protein: DNA binding specificities has been answered through the use of DNase I footprinting. This technique combines DNase I cleavage of protein protected DNA fragments and Maxam-Gilbert sequence determination methods, relying on the relatively low specificity of DNase I in a partial digestion and the ability of DNA-bound proteins to prevent phosphodiester bond hydrolysis between the base pairs they cover.

Reported within is a direct technique, MPE•Fe(II) footprinting, which allows the determination of the preferred binding sites of several small molecules on heterogeneous double helical DNAs. Methidiumpropyl EDTA [MPE], in the presence of ferrous ion and oxygen efficiently creates single-strand breaks in double helical DNA and with significantly lower sequence specificity than DNase I. Utilizing MPE•Fe(II) as a small synthetic scissor one is capable of footprinting the preferred locations and binding site sizes of small molecules bound on native DNA.

The small molecules actinomycin D, chromomycin A₃, distamycin A, echinomycin, mithramycin, netropsin, and olivomycin have been shown to demonstrate sequence specific MPE•Fe(II) cleavage inhibition, thus allowing a determination of their preferred binding sites and site sizes. Actinomycin D was found to have a minimum site size of 3 base pairs and an absolute guanine requirement in its binding site. Distamycin A and netropsin demonstrated equivalent binding specificities, preferring contiguous A+T rich regions of 5 base pairs in length. Chromomycin A₃, mithramycin, and olivomycin all shared similar binding specificities, demonstrating typically 3 base pair binding site sizes and preferring a 5'-GC-3' sequence within. Echinomycin protected a minimum of 4 base pairs; nearly all sites containing a central 5'-CG-3' sequence with 5'-CCGG-3' being favored. All footprints demonstrated an opposite strand asymmetry with overprotection on the 3' end. From a collection of their preferred binding sites, binding models for these small molecules have been derived.

MPE•Fe(II) footprinting has been compared with DNase I footprinting in their ability to determine the binding specificities of both small molecules and proteins. While DNase I exhibits a slightly greater sensitivity, MPE•Fe(II) footprinting provided greater resolving capacity and importantly more precisely defined the binding locations and sizes of small molecule: DNA complexes. Both methods provided similar information in the case of a sequence-specific DNA binding protein, lac repressor. Investigations were also undertaken to study the effects of altered DNA conformation (Z-form DNA; binding cooperativity of different small molecules) on the interactions of small molecules with native DNA.

Thesis Files