Improved methylation protection-based DNA footprinting to reveal structural distortion of DNA upon transcription factor binding.

Mapping the precise DNA sequence to which transcription factors bind contributes significantly to understanding the regulation of gene expression. In vitro footprinting techniques are typically employed to examine DNA-protein contacts (3). Methylation protection-based footprinting (MeP) is specifically used to identify the guanine contacts by a transcription factor because DMS will uniformly methylate guanines in the N7 position unless successfully protected by protein contact (5,8). Using this technique, guanines are protected from dimethyl sulfate (DMS) attack after the DNA-protein complex is formed. In addition, this method has unique benefits in that it can also detect local structural deformation of the DNA caused by allosteric interactions with the transcription factor, which are evident as guanine hypersensitivity to DMS after protein binding (2,6,7). Despite its powerful potential, MeP has been largely neglected as a practical tool for examining the interactions of transcription factors with DNA because, in this solution-based assay, the DNA probe must be saturated with protein or the footprint will be masked by the unbound background signal. Due to this consideration, methylation interference (MeI) is often employed in place of MeP to examine guanine contacts because it is a more reliable method, but it is unfortunately more limited in the information obtained (1). In MeI, the DNA is premethylated with DMS using one-hit kinetics, purified, then bound to the protein of interest, and the free and bound DNA are segregated using an electrophoretic mobility shift assay (EMSA). If a particular guanine is methylated that is necessary for the DNA-protein interaction, then that DNA molecule is segregated to the free fraction of DNA and is correspondingly missing from the bound fraction of DNA. Therefore, this method is useful for identifying guanine contacts that are necessary for protein binding to DNA. However, unlike MeP, MeI does not provide information about the DNA structural distortion that results from allosteric interactions with transcription factors because the DNA was methylated before its introduction to protein. In addition, MeI cannot always be used for cooperative DNA binding interactions because DNA-protein interactions can be compensated for by protein-protein interactions and the methylation of a singular guanine is not enough to dislodge the protein from the DNA. Thus, for multiple DNA-binding elements that exhibit cooperative binding, we have found that MeI will not provide DNA-protein contact information, whereas MeP is very effective at investigating cooperative DNA binding complexes (7) (Figure 1). In an MeI assay, guanines are Benchmarks