Targeting the Activator Interaction Domain of Mediator Subunit Med25.

Abstract

The Activator Interaction Domain of Mediator subunit Med25 is a coactivator binding partner for a number of transcriptional activators implicated in diverse cellular processes including the hijacking of transcriptional machinery by viral activators such as VP16, endoplasmic reticulum oxidative stress response, and most interestingly the progression of malignant cells to a metastatic phenotype. Thus, selective inhibitors against this domain will be useful as mechanistic probes or as potential therapeutics against these processes. Towards the goal of identifying such inhibitors, studies designed to elucidate the underlying molecular features that define activator-AcID interactions were performed and identified the minimal interaction sequences of an activator binding partner and that electrostatic contacts are particularly important in these interactions. These findings, combined with data supporting the hypothesis that the AcID motif is relatively plastic and contains distinct activator binding surfaces that may be in allosteric communication provided several potential mechanisms by which AcID-dependent interactions might be disrupted. A subsequent screening campaign against the VP16-AcID interaction identified norstictic acid, which belongs to the depsidone class of natural products, as a potent inhibitor of the domain. Biophysical and biochemical studies suggest that norstictic acid perturbs AcID-dependent interactions by binding covalently to lysine residues within the putative activator binding sites. Additional evidence indicates that the molecule allosterically induces structural shifts within AcID that contribute to the inhibitory activity, suggesting that norstictic acid is a mixed orthosteric/allosteric inhibitor of the AcID motif. Building on these results, the ERM-AcID interaction, which has been linked to metastasis in a number of cancers, was screened against a library of natural product extracts in an effort to identify novel non-covalent inhibitors of the domain. Through a series of counter screens and stringent hit filtering steps, a promising extract with potent activity in vitro against the ERM•AcID interaction has been identified and is currently undergoing deconvolution to identify the structures of the active natural products within the extracts. These natural products will subsequently be used as molecular probes to explore the importance of this interaction in transcriptional programs related to cancer processes such as metastasis and tumorigenesis.