Lipopeptidomimetics are Selective and Modifiable Coactivator Protein-Protein Interaction Inhibitors

Abstract

The protein-protein interactions (PPIs) of transcriptional coactivators are key to the synergistic activation of gene expression. The dysregulation of these PPI networks, particularly in the interactions between coactivator and activator proteins, is present in several forms of disease. Inhibition of coactivator PPIs is thereby a strategy to dissect the functional role of the interactions between transcriptional components in dysregulated contexts. Coactivator PPIs occur through intricate mechanisms of recognition, which involve dynamic complex formation, an undefined surface topology, and multiple binding partners. The functional disruption of these interactions with synthetic molecules has historically been challenging, considering that these factors limit the structural information available for developing inhibitors using rational-design or structure-based approaches. Here we propose that short peptides derived from the sequences within the interaction surfaces of coactivator-activator PPIs, have the potential to be developed into potent and selective inhibitors of these complexes. We demonstrate that peptide lipidation is a powerful form of modification to enhance the inhibitory activity of short activator-like peptides against coactivator complexes. This dissertation presents the development and evaluation of lipopeptidomimetics (LPPMs) as inhibitors of the PPIs of coactivators. In our initial assessment of this strategy, we used a peptide with an amino acid sequence that shares characteristics with the composition of transcriptional activation domains (TAD) of activators, against the PPIs of coactivator Med25. This protein, a subunit of Mediator, regulates the expression of genes implicated in various types of cancer. We demonstrate that the incorporation of a medium-chain, branched fatty acid to a heptameric peptide, LPPM-8, increases the compound’s inhibitory activity by over 20-fold, rendering it a selective inhibitor of Med25 PPIs. Structure-activity relationship studies, combined with biophysical analyses, revealed that the lipid structure, specific amino acid residues, and the C-terminal moiety of the molecule each contribute to LPPM-8’s effectiveness and the structural propensity as an inhibitor. We determined that this molecule acts primarily as an orthosteric inhibitor of Med25 PPIs, and we observed its biological activity in a cellular context. Next, aiming to determine whether this strategy could be applied to multiple coactivator targets, we tested it against the PPIs of the KIX domain of coactivator CBP. We found that specific sequence modifications in LPPMs lead to altered selectivity for different coactivator targets. In particular, changing a single amino acid from aspartic acid to alanine (LPPM-8-D2A) resulted in a 10-fold selectivity switch towards the inhibition of CBP KIX compared to Med25 PPIs. This selectivity switch was validated by evaluating the LPPM-8-D2A multiple contexts, revealing its allosteric inhibition of KIX PPIs. These findings suggest that LPPMs are tunable scaffolds with potential as a generalizable strategy for inhibiting coactivator PPIs. Chapter 4 outlines the potential steps necessary to refine LPPM design into a high-throughput approach for the development of inhibitors and explores the application of this method to other coactivator and intrinsically disordered protein systems.