| dc.description.abstract | ABSTRACT RNA-binding proteins (RBPs) are key regulators of post-transcriptional gene expression, and underlie many disease-relevant processes. However, they have historically been challenging to target with drug-like compounds. Inspired by the “anchor residues” of protein-protein interactions, we developed a computational approach for rationally designing small-molecule inhibitors of RBPs. In this dissertation, we first selected Musashi-1 and Musashi-2 to apply our “RNA mimicry” approach. Both Musashi proteins are well-studied RBPs, known principally as stem‑cell markers that are upregulated in many cancers. In the future, we hope our “RNA mimicry” approach can be generally applied to inhibitor design of diverse target RBPs. To design inhibitors of Musashi proteins, we applied our strategy by mimicking the three-dimensional interactions in the protein-RNA complex. As described in Chapter II, by using pharmacophoric screening, we searched for drug-like compounds that can present the same geometric arrangement of functional groups as the RNA in the complex. We hypothesized that such ligands would engage Musashi in a similar manner as the RNA binds to Musashi. Since the interaction geometries can be quite distinct from one another for different RBPs, we anticipated that this strategy would lead to inhibitors that were selective for Musashi. To facilitate characterization of these candidate Musashi inhibitors, I developed the “isothermal analysis” approach. As described in Chapter III, this method allows us to calculate quantitative binding constants by using differential scanning fluorimetry (DSF) data. The method requires only the protein unfolding information at a given temperature as a function of ligand concentration, and thus no thermodynamic parameters are included in the calculation. Finally, I describe the use of computational docking to better understand the basis for PROTAC-mediated degradation of target proteins. PROteolysis TArgeting Chimeras (PROTACs) are heterobifunctional small molecules which can induce target protein degradation through cell ubiquitination process. Rational design of PROTACs is still challenging, however, because of the limited structural understanding of their mechanism. In Chapter IV, I seek to predict the formation of the ternary structure complex by including both effects of the protein-protein interaction and effects of the chemical linker. Looking ahead, I hope to use these ternary structure models to explain the activity and selectivity of the given PROTAC molecules, and ultimately to use our designed Musashi inhibitors as a starting point for building new PROTACs to degrade Musashi. The text of Chapter II is a manuscript that is in preparation for publication as: Bai N‡, Adeshina Y‡, Lan L, Makhov PB, Xia Y, Gowthaman R, Miller SA, Johnson DK, Boumber Y, Xu L, Karanicolas J. Rationally designing inhibitors of the Musashi protein-RNA interaction by hotspot mimicry. ‡equally contributing co-authors The supporting information for this chapter is included as Appendix A.1. The text of Chapter III is a reprint of the material from: Bai N, Roder H, Dickson A, Karanicolas J. Isothermal analysis of ThermoFluor data can readily provide quantitative binding affinities. Sci. Rep. 9, p. 2650 (2019). Note: the software disseminated with this paper has accumulated 1000 downloads in the 9 months since publication The supporting information for this chapter is included as Appendix A.2. The text of Chapter IV is a manuscript that is in preparation for publication as: Bai N, Karanicolas J. Predicting PROTAC-mediated ternary complex formation using Rosetta. The supporting information for this chapter is included as Appendix A.3. | |
