Characterization of the Dynamic Interactions of Transcriptional Activators

Abstract

Transcription is initiated through a series of coupled binding equilibria between transcriptional activators and their array of protein targets within the transcriptional machinery. However, previous efforts to kinetically characterize these interactions have produced conflicting models for the mechanism of complex formation, which is hampering the discovery of non-natural mimics of their transcriptional activation domains (TADs). Using fluorescence stopped-flow techniques, we determined that the activators Gal4, Gcn4, and VP16 interact with the same coactivator, Med15, via a two-step binding mechanism comprised of a bimolecular association step and a conformational change. We further hypothesized that the life-times of these interactions should be more revealing of differences in activator potency (i.e., transcriptional output); thus, we analyzed the microscopic rate and equilibrium constants defining the individual steps within our mechanism, in order to identify key trends that can differentiate the activators from one another in terms of their ability to recruit the transcriptional machinery to a gene promoter. We determined that it is the favorability of the conformational change step and its partition ratio that correlates with the ability of an activator to stimulate transcription. Future studies will focus on determining how the different structural propensities of the TAD sequences contribute to the stability of the intermediate that they form. Furthermore, another significant challenge in the development of artificial transcription factors (ATFs) is a lack of small molecules that can be used to localize them to a gene promoter in a cellular context. We propose a novel approach to accomplish this task which relies on the interaction of a ligand with an endogenous DNA-bound protein. To this end, we have successfully used an in vitro phage display selection with a random 12 amino acid peptide library to isolate ligands that are capable of interacting with the DNA-binding proteins Gal4 (residues 1-100) and LexA (residues 1-202). Future studies will entail the implementation of a selection with a conformation-constrained peptide library to obtain ligands that may possess increased stability and specificity within the cellular milieu. In particular, protein scaffolds that promote helix stabilization would aid in the future identification of peptidomimetic or small molecule replacements.